38703 Africa Region Working Paper Series No. 98 MEASURING INTERGOVERNMENTAL FISCAL PERFORMANCE IN SOUTH AFRICA Issues in Municipal Grant Monitoring Navin Girishankar David DeGroot T.V. Pillay MAY 2006 Africa Region Working Paper Series No. 98 May 2006 Abstract This paper argues that the design of monitoring systems in decentralized settings is inextricably linked to the design of intergovernmental fiscal instruments. In South Africa, the consolidation of sectoral municipal grants is the predominant feature of its fiscal strategy to empower the local government sphere. Accordingly, the country's approach to measuring intergovernmental and municipal performance has started with a focused initiative on municipal grant monitoring. The grant monitoring agenda has been characterized by efforts (i) to balance both the service delivery and empowerment objectives of decentralization; (ii) to build monitoring systems--both in terms of what they measure and how they measure it--in a sequenced manner; and (iii) to complement vertical accountability relationships (from municipalities to central government transferring departments) with horizontal ones (from municipalities to their citizens). In the decade ahead, municipal grant monitoring will play an increasingly important role in helping South Africa chart its governance frontier. This can be achieved in part by measuring the development impact of large consolidated grants; using information to enrich local democratic processes, and benchmarking municipal performance to promote inter-jurisdictional competition. In reviewing the South African experience, the paper offers a point of reference for reforming governments that are concerned with evaluation capacity development, management of fiduciary risks, and results-based management in the context of decentralization. The Africa Region Working Paper Series expedites dissemination of applied research and policy studies with potential for improving economic performance and social conditions in Sub-Saharan Africa. The series publishes papers at preliminary stages to stimulate timely discussions within the Region and among client countries, donors, and the policy research community. The editorial board for the series consists of representatives from professional families appointed by the Region's Sector Directors. For additional information, please contact Momar Gueye, (82220), Email: mgueye@worldbank.org or visit the Web Site: http://www.worldbank.org/afr/wps/index.htm. The findings, interpretations, and conclusions in this paper are those of the authors. They do not necessarily represent the views of the World Bank, its Executive Directors, or the countries that they represent and should not be attributed to them Authors'Affiliation and Sponsorship Navin Girishankar Senior Public Sector Specialist, AFTPR, Africa Region, World Bank Email: Ngirishankar@worldbank.org David DeGroot Senior Urban Specialist, AFTU1, Africa Region, World Bank Email: Ddegroot@worldbank.org T.V. Pillay Chief Director, Local Government, South African National Treasury Email: TV.Pillay@treasury.gov.za ACKNOWLEDGEMENTS This paper is based on the findings of a multi-year Analytical and Advisory Activity (AAA) program undertaken jointly by the Intergovernmental Relations Branch of the South African National Treasury (NT) and the Africa Region of the World Bank between early 2002 and late 2004. The National Treasury team was led by T.V. Pillay and supported by Malcolm Booysen. The World Bank team was co-led by Navin Girishankar and Dave DeGroot, and included the following expert consultants: David Savage, Mike Frazer, Rajen Govender, Jeremy Druker, Chris Heymans, and Phillip van Ryneveld. The paper and the joint AAA work were prepared under the overall guidance of Ismail Momoniat of the South African National Treasury and Jaime Biderman, Fayez Omar, and Ritva Reinikka of the World Bank. The authors would also like to thank Mike Stevens, Harry Garnett, and Jim Hicks for their peer review comments, and Somik Lall for his review of issues relating to impact assessment methodology. CURRENCY EQUIVALENTS (Exchange Rate Effective July 2002) Currency Unit = South African Rands US$ = R11 FISCAL YEAR National and Provincial Government April 1 - March 31 Local Government July 1-June 30 ACRONYMS AND ABBREVIATIONS AAA Analytical And Advisory Activity ANC African National Congress CFO Municipal Chief Financial Officer CMIP Consolidated Municipal Infrastructure Programme DPLG Department Of Provincial And Local Government DBSA Development Bank of Southern Africa DORA Division Of Revenue Act DWAF Department of Water Affairs And Forestry FFC The Financial and Fiscal Commission FMG Financial Management Grant GAMAP Generally Accepted Municipal Accounting Practices HIAS Household Impact Assessment Survey IDP Integrated Development Plan MBFR Municipal Budget And Financial Reform Program MFMA Municipal Financial Management Act (2003) MFMTAP Municipal Financial Management And TA Project MIG Municipal Infrastructure Grant MIIF Municipal Infrastructure Investment Framework MSA Municipal Systems Act (2000) MTBPS Medium Term Budget Policy Statement NT South African National Treasury PFMA Public Financial Management Act PIMSS Planning And Implementation Management Support System NSDP National Spatial Development Perspective MTEF Medium Term Expenditure Framework KPI Key Performance Indicators R South African Rands StatSA Statistics South Africa SA South Africa SACN South African Cities Network TA Technical Assistance USD United States Dollars WB World Bank TABLE OF CONTENTS EXECUTIVE SUMMARY........................................................................................................i-iii I. Context.........................................................................................................................................1 Intergovernmental Reform In Post-Apartheid South Africa...................................................1 The Monitoring Challenge......................................................................................................2 National Treasury-World Bank Collaboration--Goals And Outputs.....................................2 Organization Of Paper............................................................................................................4 II. Analytical Framework ..............................................................................................................4 Institutional Transformation--Special Conditions For Managing Decentralization..............4 Implications For Analyis Of Monitoring Issues .....................................................................6 Methdology And Approach....................................................................................................7 III. South Africa's Municipal Finance System--From Transition To Consolidation..............8 Fiscal Transition--Resourcing The Local Government Sphere.............................................8 Grant Consolidation--Promoting Autonomy With Performance ..........................................9 Implications For A Municipal Monitoring Strategy.............................................................14 IV. The Grant Monitoring Challenge--Prioritizing Financial Accountability ......................16 DORA And The Local Government Fiduciary Framework .................................................16 Data Capture Requirements And Municipal Capacity .........................................................17 Fiscal Monitoring As An Entry Point For Grant Monitoring...............................................20 Pursuing Option B--Key Components Of A Phased Approach ..........................................21 Implications For Phasing In The System..............................................................................25 V. Grant Monitoring And Beyond..............................................................................................27 Moving Beyond Inputs .........................................................................................................27 The Pilot Household Impact Assessment Survey.................................................................30 Implications For Service Delivery--Performance And Perception......................................31 Implications For Household Income And Expenditure Patterns ..........................................34 Measuring Municipal Processes And Institutions ................................................................37 The Way Forward.................................................................................................................38 VI. Municipal Monitoring And South Africa's Governance Frontier....................................39 Seeking The Governance Frontier........................................................................................40 Annexes 1. A Brief History Of South Africa's Intergovernmental System Since 1992 ...............................42 2. Other Monitoring Undertaken At National Level ......................................................................48 3. Web Based Report Form ............................................................................................................53 4. Database Design.........................................................................................................................54 5. Impact Assessment Indicators....................................................................................................55 6. Pilot Questionnaire.....................................................................................................................57 Boxes 1. The Hidden Cost Of Unpredictable And "Source-less" Funds Flows .......................................19 2. Municipal Management And Household Level Impact Evaluation--Ongoing Efforts .............29 3. Do Client Scorecards And Ratings Convey Meaningful Information........................................33 4. Intergovernmental Principles Underpinning Municipal Financial and Fiscal Reforms .............46 Figures 1. Elements Of Intergovernmental Design During Transformation.................................................5 2. From Grant Monitoring To Municipal Monitoring....................................................................15 3. Long Term Monitoring Flow Within The Municipal Grant System..........................................21 4. First Phase Of A System Roll Out..............................................................................................22 5. Summary Of Plans For First Phase Of System Roll Out............................................................25 Tables 1. Intergovernmental Monitoring--Salient Design Features And Institutional Factors...................7 2. Real Vertical Division Of Revenues, 1995-2005.......................................................................10 3. Fiscal Transition And Consolidated As Evidenced By Real Grant Allocations,1995-2005 ......11 4. Projected National-Local Transfers, 2006/07.............................................................................13 5. Existing Reporting Systems And Requirements For Municipalities..........................................18 6. Indicative Overview Of The Pilot Sample .................................................................................31 7. Rated Importance Of Individual Services In Urban And Rural Households..............................32 8. Mean Household Income Across Districts.................................................................................34 9. Sources Of Monthly Income As Percentage Of Total Income...................................................35 10. Uncorrected And Corrected Mean Differences In Income And Expenditure ..........................35 11. Grouped Costs For Urban And Rural Households ...................................................................36 12. Costs For Households Receiving Services...............................................................................37 13. Municipal Grants and Grant Monitoring By Department. .......................................................49 EXECUTIVE SUMMARY 1. Following nearly a decade of intergovernmental reform and decentralization, South Africa has established a constitutionally-mandated, independent local government sphere, which is responsible for a broad range of the country's economic and social objectives. Significant progress has been made in developing the legal, administrative and fiscal framework for empowering municipalities including the amalgamation of racially fragmented authorities, resourcing of local mandates through strengthening of tax bases and establishment of an array of predictable and equitable municipal grants. In parallel, a far-reaching program of planning, budgeting, and accounting reforms are designed to strengthen the municipal fiduciary framework. 2. These "first ten years" have led to considerable transformation of public institutions in South Africa; they have also thrown up an array of implementation challenges related to the ability of municipalities to effectively deliver services and empower citizens. Meeting the challenge of implementation ultimately requires inter alia a stronger focus on municipal monitoring--and specifically, the monitoring of municipal grants. Equally important is the need to provide feedback to national, provincial, and local officials on "what works" in order to promote learning by doing and to enable the three cooperative spheres to more effectively carry out their respective functions. Over the 2002-2005 period, the South African National Treasury (NT) has focused considerable efforts on development of the systems necessary to rigorously and transparently monitor the fiscal and operational performance of municipal grants. 3. This report summarizes the work and findings of an NT-World Bank Analytical and Advisory Activity (AAA), launched in September 2002, to review the municipal grant monitoring system; identify options for improving the flow of public resources to municipalities; and to develop monitoring tools to more effectively track expenditure utilization and other indicators for service delivery improvement at the local level. In the course of this AAA, the following outputs were prepared: (i) a review of the intergovernmental fiscal framework, (ii) strategic reviews of financial and fiscal monitoring of municipal grants, (iii) the design and testing of web-based reporting formats for municipal grants, and the implementation of a relational data base on NT's existing server, (iv) a comprehensive monitoring and evaluation review, and (v) the design and testing of pilot household surveys within municipalities covering issues of household income and expenditure patterns, service levels, and client satisfaction. Key findings and lessons from the AAA are summarized below. Municipal Finance--From Transition To Consolidation 4. By 1999, South Africa had completed the structural reform process at the subnational level, yielding nine provinces and 284 local governments. Soon after in 2000, the first municipal elections paved the way for the transition to an intergovernmental fiscal framework based on principles of adequacy, predictability, and transparency. Two features of South Africa's fiscal transition illustrated the Government's early commitment to the creation of a viable local government sphere. First, yearly increases have been implemented in the overall vertical division of revenues between national and subnational levels. A second feature of the fiscal transition was the successful phasing out of the ad-hoc and opaque transfers that were a vestige of the apartheid era. An expansion and increased predictability of the new municipal grants were supported by tighter regulation over fiscal flows to local governments under DORA and accompanying public financial regulations, notably the Municipal Finance Management Act (2003). 5. Even as the new transfers sought to reduce the infrastructure backlog, the Government was aware of the need to balance these sector specific goals with the underlying objective of i decentralization--to empower communities to have greater voice in public decision-making at the local level. The capacity requirements of managing several different sectoral transfers were also becoming apparent. Therefore, in 1998, the Government embarked on a major initiative to consolidate municipal grants into a limited number of multisectoral "pipes." By FY 06/07 the great majority of SA's municipalities will have to manage, monitor and evaluate a maximum of six--and in most cases only five--grants through DORA: (i) an unconditional Equitable Share and related transfers that meet recurrent needs; (ii) conditional transfers to meet municipal investment needs including water, power, sports and recreation, and integrated rural development, now being consolidated in the Municipal Infrastructure Grant (MIG); (iii) a set of conditional transfers designed to meet recurrent municipal expenditures (including grants for capacity building and restructuring). 1 The total value of projected national transfers to local governments in 06/07, at current exchange rates, will be about USD 3 billion. Implications For A Medium Term Municipal Monitoring Strategy 6. The evolution of intergovernmental fiscal relations in South Africa has far-reaching implications for how municipal monitoring systems are developed and established. Specifically, the AAA highlighted three priorities that should shape a medium term municipal monitoring strategy for NT. First, there is need to strike a balance between the service delivery and local autonomy objectives of decentralization including progressive re-design of fiscal transfers (municipal grants) as well as grant monitoring systems. Second, the monitoring strategy will phase in elements of a full-fledged monitoring system in step with local and national data carrying capacity, the requirements of sound fiscal and financial management, as well as program priorities. Third, feedback mechanisms should balance vertical or top-down accountability (of municipalities to transferring departments) with horizontal accountability (of municipalities to their citizens and their elected representatives). 7. In line with the above-mentioned monitoring strategy, NT has adopted a sequenced approach that builds on ongoing financial management capacity building initiatives--that is, the Municipal Budget and Financial Reform Program. Specifically, the introduction of the standardized web-based reporting format and of the relational data base in NT are important first steps to achieving accountability upward of all municipalities for receipt and expenditure of funds, and clarifying the responsibilities and required capacities of all three spheres involved in the SA intergovernmental fiscal system. The phased approach also allows the development of individual components but under an overall monitoring system development framework. The principle of progressive enhancement of systems or even the creation of a new database (Treasury A-style returns) can be placed within the context of a complete municipal monitoring system, with high level users accessing via a front end query tool to extract information from all of the databases. Beyond Grant Monitoring 8. Based on a comprehensive review of impact analysis in South Africa, the NT and Bank teams concluded that, over time, monitoring instruments will need to measure more than just the inputs or resource flows within the municipal grant system. Investment in tools that captured data on local governance processes, service levels, and household behavior patterns would be necessary over the medium term. To test the feasibility of such instruments, a household impact assessment survey was piloted to generate measures of the objective needs as well as the subjective expectations of local communities that receive municipal services, and to analyze 1The Restructuring Grant is available only to the 15 largest municipalities. ii trends in how these needs and expectations are changing as a result of municipal performance. Lessons from the pilot include (i) the need to carefully select the master sample for the roll-out of the pilot to ensure that it is representative; (ii) the value of both objective data on service levels as well as subjective rankings of citizen preferences and satisfaction levels; (iii) the importance of assessing the impact of transfer programs on household behavior including income and expenditure patterns; and (iv) the benefits of evaluating public processes such as planning and budgeting, or accountability processes at the municipal and household levels, and determining their impact on service delivery and empowerment. 9. In broadening the current M&E focus beyond inputs, NT and other transferring departments such as DPLG will need to make additional investments in household impact surveys, local government assessments, and facility level surveys. NT's decision to roll out the HIAS pilot after further refinement of the survey instrument is an important step towards moving grant monitoring beyond inputs, and moving beyond grant monitoring to outputs and outcomes. 10. Another important conclusion of this AAA is the need to think about how data is used within the intergovernmental system. More analysis and testing are required to understand how information on grant performance (and municipal performance more broadly) can be best used by citizens, local administrators, and elected officials to improve local municipal performance. Public reporting or cross-jurisdictional benchmarking of results will need to be more fully exploited to ensure that data can generate the necessary feedback from both higher spheres of government as well as local citizens. This information will improve municipal in year internal and external reports, including annual reports and performance evaluation. Helping Chart The Governance Frontier 11. As South Africa enters its second decade of reform, municipal monitoring will play an increasingly important role in helping South Africa chart its governance frontier. The AAA work suggested three potential frontier issues. First, there is need to ensure that over the short to medium run, investments in grant monitoring focus on the issues of efficiency and efficacy of large consolidated grants such as the Equitable Shares, the MIG, and possibly, the anticipated consolidated capacity building transfer. Second, provided that performance data can be made more accessible to citizens on a regular basis, municipal monitoring can enrich local democratic and political processes and possibly encourage evidenced-based decision-making. Finally, municipal monitoring and benchmarking could also help promote healthy competition between municipalities that seek to attract investment, while signaling to municipal leaders those areas of local governance that require improvement. iii SECTION I. CONTEXT Intergovernmental Reforms In Post-Apartheid South Africa 1. In May 2004, the Republic of South Africa (SA) completed its first decade of democracy. Over this period, the primary focus of policymakers, citizens, and civil society has been on wholesale restructuring of basic institutions responsible for promoting economic development and poverty reduction. Integral to this process has been the design and implementation of intergovernmental reforms across the national, provincial and municipal levels. South Africa's 284 local governments or "municipalities" make up a constitutionally-mandated, independent third sphere, responsible for a broad range of the country's economic and social objectives.2 In both urban and rural areas, local governments play a central role in providing basic infrastructure and social services, and promoting local economic development. 2. Significant progress has been made in developing the legal, administrative and fiscal framework for empowering the local government sphere. Racially fragmented local authorities from the apartheid era were amalgamated and reconstituted as democratic and unified municipal governments across the country. In 1999, the jurisdictions of these amalgamated municipalities were finally demarcated, and enabling legislation such as the Municipal Systems and Municipal Finance Management Acts (MSA and MFMA) is in place.3 Six metropolitan governments serve nearly 15 million citizens--about a third of the country's total population--through uni-city structures. The roles and functions of district councils (outside metropolitan nodes) are being redefined especially in relation to smaller municipalities. Local governments are also empowered to restructure their systems to encourage community and private sector participation in service provision. 3. South Africa's intergovernmental fiscal framework has evolved considerably in order to ensure that municipalities are adequately resourced to carry out their mandated functions. The Government has sought to preserve and strengthen municipal tax bases (rates and levies), while clarifying municipal authority to change tax rates and implement user-charges. Also, the ad-hoc intergovernmental grants of the apartheid era have been replaced with a more efficient, equitable, and predictable system comprising three sets of transfers defined under the Division of Revenue Act (DORA): unconditional grants to meet recurrent needs, conditional grants to meet recurrent needs, and conditional grants to service delivery investment needs. Through these transfers, on the 2003-2004 financial year, approximately 61% of national revenue or R200 billion was channeled to subnational authorities. The share of revenues going to provinces and municipalities has grown systematically, including a 10% increase in national-provincial transfers, and a nearly 30% increase in municipal transfers over the 2002/3 and 2003/4 period. 4. In parallel, the Government has undertaken far-reaching public financial management reforms including the implementation of a Medium Term Expenditure 2The number of municipalities may change as the system develops. 3See www.treasury.gov.za for all relevant legislation. 1 Framework featuring a transparent division of revenues over a medium term horizon; roll out of medium-term planning and budgeting in the context of Integrated Development Plans (IDPs) at the local level; reform of the municipal accounting system as per the 2003 Municipal Finance Management Act4, and development of a policy and regulatory framework on municipal borrowing that inter alia would structure access to debt finance. In addition, specific purpose capacity building grants to municipalities such as the Financial Management Grant (FMG) support capacity building, including an internship program, in accounting, financial reporting, cash management, and other aspects of budget execution at the local level. The simultaneous implementation of fiscal decentralization and financial management reforms was designed to enable provinces and municipalities to manage the growth of transfers and own-source revenues in an efficient, effective, and transparent manner. The Monitoring Challenge 5. There is broad recognition among students and observers of South Africa's intergovernmental reform experience that the country has systematically established the key prerequisites for promoting local self-governance and decentralized service delivery. Yet, the success of the "first ten years" of reforms has thrown up an array of implementation challenges related to the ability of municipalities to effectively deliver services and empower citizens. For instance, significant growth in grant revenue to rural municipalities has often outstripped their spending capacity, resulting in large cash balances being reported in some areas. Municipalities have also struggled with reporting on their use of the various transfers they receive from the national and provincial levels. The ability of municipalities to manage service provision (including attracting private sector participation) and empower citizens in local decision-making is increasingly on the minds of South Africa's policymakers and public servants as they enter their "second ten years." Yet, the timeliness and quality of data on these processes within municipalities, or data on the relationship between municipalities and higher spheres of government remains poor. 6. Meeting the implementation challenge requires a stronger focus on the monitoring of intergovernmental and municipal performance--and specifically, on the monitoring of municipal grant performance. In tracking the utilization and impact of these transfers, a key task is to assess whether the intergovernmental system is effectively managing fiduciary, service delivery, and governance risks. Equally important is the need to provide feedback to national, provincial, and local officials on "what works" in order to promote learning by doing and to enable the three cooperative spheres to more effectively carry out their respective functions. National Treasury-World Bank Collaboration--Programmatic AAA 7. Rationale. In anticipation of this challenge, the South African National Treasury (NT) has focused on developing systems to rigorously and transparently monitor the 4The MFMA came into effect on July 1, 2004 and a phased implementation program will bring all 284 local governments into compliance over three years. 2 fiscal and operational performance of municipal grants. Specifically, NT sought to systematically review and progressively address issues affecting the efficiency and effectiveness of municipal grant monitoring including the design of fiscal instruments and funds flows, the choice of indicators, the establishment of reporting and coordination arrangements across spheres of government, the elaboration of data verification measures, and the demand for feedback to promote accountability. 8. A joint NT-World Bank Analytical and Advisory Activity (AAA) was launched in September 2002 to support these efforts. In keeping with the Government's desire for a highly focused, client-driven product, this activity was led by NT's Intergovernmental Relations Branch and supported by a cross-unit World Bank team drawn from the Africa Region's Urban Technical and Public Sector Reform-Capacity Building Units. This AAA work complements ongoing capacity building and policy reform work managed by NT, with support through the Municipal Financial Management Technical Assistance Project (MFMTAP, which is the sole IBRD loan to South Africa). Bank-supported work on urban management indicators, through successive Cities Alliance grants to the South African Cities Network, also contributes directly to NT's continuing work program. Progress in improving municipal financial management is periodically reviewed through NT's publication of the "Intergovernmental Fiscal Review" (see www.treasury.gov.za), the next edition of which is scheduled for mid-2006. 9. Objectives. This work aimed to review the municipal grant monitoring system; to identify options for improving the flow of public resources to municipalities; and to develop monitoring tools to more effectively track expenditure utilization and other indicators for service delivery improvement at the local level. Specifically, these objectives were pursued by (i) strengthening the strategic framework for Division of Revenue Act (DORA) monitoring; (ii) improving the quality of financial and non- financial data in municipal grant reporting; and (iii) supporting municipal surveys and database development on a pilot basis. 10. Outputs. A key feature of this work was the timely provision of advisory support for the development and implementation of monitoring options for municipal grants under DORA. A range of activities were carried out--from traditional expert assessments and reviews to consultants to support the design of web-based and related monitoring tools or pilot surveys. Several outputs were produced as part of this AAA including the following: a review of the intergovernmental fiscal framework including recent efforts at grant consolidation and their implications for monitoring and evaluation; strategic reviews of financial and fiscal monitoring of municipal grants in line with DORA requirements; the design and testing of web-based reporting formats for municipal grants, and the implementation of a relational data base on NT's existing server; a comprehensive monitoring and evaluation review with a focus on issues of strategic coordination across departments and spheres of government; 3 the design and testing of pilot household surveys within municipalities covering issues of household income and expenditure patterns, service levels, and client satisfaction Organization Of Paper 11. This paper provides a synthesis of several background papers described in the concept note. Accordingly, Section II discusses a framework for reviewing the objectives, architecture, and outputs of intergovernmental monitoring systems, as well as the methodology, and approach used by the team. Section III reviews the evolution of South Africa's intergovernmental fiscal system including the revenue sharing arrangements over time, and discusses implications for the Government's municipal monitoring strategy. Section IV provides an overview of issues related to the monitoring of fiduciary risks. Section V discusses prospects for progressively introducing monitoring of processes, outputs, and impact. Finally, Section VI concludes with a brief discussion of the role of municipal grant monitoring (and intergovernmental monitoring in general) as a lever for charting and pushing the public sector governance frontier in South Africa. Recommendations for furthering this agenda are also made. SECTION II. ANALYTICAL FRAMEWORK Institutional Transformation--Special Conditions For Managing Decentralization 12. As developing countries across Africa, Asia, and Latin America embarked on decentralization efforts over the past two decades, both practitioners and academics have recognized the importance of studying the design of intergovernmental systems, their constituent parts, and their implications for the development and performance of local governments. A considerable literature on decentralization already exists on a wide range of issues including the macro-fiscal implications of intergovernmental design; the incentive impact of various types of grant instruments on [direct and indirect] revenue mobilization and service delivery performance; the allocative, distributional, and targeting implications of transfers; staff redeployment and administrative decentralization; prospects for improved expenditure tracking; as well as the institutional impact of grant systems on local government budget processes. Yet, this established inventory of professional "good practice" on intergovernmental design has generally-- with some noteworthy exceptions--emerged in country settings that have been territorially and constitutionally stable for extended periods of time (for e.g., OECD countries, some Latin American countries, China, and India). Practitioners and scholars are only now beginning to focus on settings where the decentralization process is intimately connected with constitutional and institutional reforms following civil conflict, introduction of representative governance, or other exigencies that lead to rapid change. 13. Decentralization in these transformational settings therefore offers a set of exceptional conditions under which conventional approaches to "good practice"-- specifically, in the design of intergovernmental transfer systems and their interdependencies with other aspects of public management--should be reviewed and 4 Figure 1: Elements Of Intergovernmental Design During Transformation Fiscal design Fiscal design (own sources) (transfers) Accountability (downward) Cross-cutting Enabling legislation expenditure (expend and revenue management systems assignments) Accountability (fiscal, sectoral monitoring and feedback) Cross-cutting HRM systems Jurisdictional and administrative structures Consolidation, implementation Source: N. Girishankar and D. DeGroot possibly reformulated. In particular, three cases of transformation in Africa stand out not only in terms of scope and scale, but also the centrality of devolution as a strategy for reordering the fundamental rules of the game by which power is sought and authority projected. In Ethiopia, Uganda, and South Africa, reforming governments have sought to answer fundamental questions of political legitimacy, public management, and state formation in part through the establishment of new local government structures, the transfer of major decentralized delivery responsibilities, the initiation of far-reaching restructuring and re-demarcation efforts, and the design of radically new intergovernmental fiscal systems--within a hard budget constraint--intended to empower local communities with increased expenditure autonomy. 5 14. As reforming governments in these three countries embarked on implementing devolution, they have faced similar challenges in terms of pursuing the multiple objectives of transformation, and managing interdependencies between elements of intergovernmental design. These conditions place a premium on identifying entry points for launching and managing the decentralization process in a manner that is sustainable. Pursuing Multiple Objectives: In all three countries, decentralization has sought to achieve multiple, sometimes, competing objectives. On the one hand, devolution- led transformation in these African countries is intended to fashion democratic 5N. Girishankar and D. DeGroot. Rapid Transformation and Intergovernmental Design in Africa-- Durability, Efficiency, and Accountability. Background Note and Presentations. World Bank. 2002. 5 institutions that allow for greater decision-making authority over how public resources are spent. On the other hand, these newly created democratic local governments are to effectively and efficiently deliver a range of services to the poor (for example, primary education, primary health care, local infrastructure, water and sanitation), often at a level that was unprecedented. On the margin, balancing empowerment and service delivery objectives has meant that governments have had to make careful decisions about the phasing of administrative decentralization, the design of fiscal transfers, as well as the implementation of monitoring and evaluation systems. Managing Institutional Inter-linkages And Interdependencies: Decentralization in these settings have also required major support for institutional and capacity building. Reforming governments have had to build basic public management systems (such as those governing planning, budgeting, accounting, and human resource management) in parallel with efforts to transfer responsibilities and resources to lower tiers of government. As a result, they have been challenged to manage the inter-linkages between key elements of intergovernmental fiscal design such as managing fiduciary risks associated with fiscal decentralization, and also exploit interdependencies such as the providing incentives for efficient personnel management enforcing a hard budget constraint as personnel are redeployed (Figure 1). Identifying Entry Points To Maintain Momentum For Reform: In order to ensure that the decentralization process can be sustained over time, experience demonstrates that rapidly reforming countries have typically chosen different elements of intergovernmental design as entry points for responding to the political demand for change (Figure 1). For instance, in South Africa, jurisdictional reform and the amalgamation of formerly racially segregated municipalities provided a key entry point for guiding a range of other initiatives including the assignment of own-revenue authority, the roll-out of expenditure management reforms, etc. In Ethiopia, the overriding political imperative of empowerment through resource transfer led regional governments to choose the creation of a consolidated block grant as the lever for radically devolving resources in line with newly defined expenditure responsibilities. In Uganda, a highly consultative process around constitutional and legal reforms provided a strong basis for beginning the process of fiscal and administrative decentralization.6 Implications For Analysis Of Monitoring Issues 15. Given the special conditions that inform the decentralization process in South Africa, the above-mentioned criteria should be applied to reviews of intergovernmental reforms including monitoring of fiscal transfers. Based on the international experience in both developed and developing countries, Table 1 identifies three building blocks of an intergovernmental monitoring system and key institutional quality factors that should be 6Ibid. 6 considered when elaborating monitoring objectives and systems. These issues are considered in Sections III, IV, and V, and have provided a foundation for the ongoing NT-World Bank dialogue on municipal monitoring. As such, they also inform the discussion of frontier issues in Section VI. Table 1. Intergovernmental Monitoring--Salient Design Features And Institutional Factors Salient Design Features Institutional Quality Factors 1. Intergovernmental Predictability, adequacy, and simplicity of fiscal flows transfer system Balance between conditional and unconditional transfers Coordination between national departments 2. Data carrying Coverage, reliability, and quality, timeliness, and accuracy of capacity of system financial, input, process, and output data Safeguards against noncompliance Clarity of roles and responsibilities in line with mandates 3. Vertical and local Commitment to public reporting at the local level accountability Feedback to and learning by transferring departments, provinces, and local governments Methodology and Approach 16. The task team adopted a three-pronged methodological approach to the review. For the overview of the intergovernmental fiscal framework and the financial management system, structured interviews with officials in NT directorates as well as other transferring departments at the central level were undertaken. These interviews were supplemented by detailed reviews of the organic finance act and related legislation, Government gazettes, and other reports. This review work included an interrogation of the existing reporting formats, applicable software systems and relevant hardware compliance issues as well as a review of prevailing protocols and available data on current legacy systems. 17. Detailed diagnostic assessments of fiscal and financial management policies, procedures and systems at the local level were also conducted based on site visits to selected municipalities. Finally, pilot household level survey data was used to capture preliminary outcome data based on a clarification and definition of the full set of impact assessment indicators with subsets measuring the constituent elements of impact of government services on households; development of a survey instrument; development of a pilot sampling frame representing the key features of a master sample of municipalities; and, analyses of obtained data according to the analytic framework defined by the indicators. 7 SECTION III. SOUTH AFRICA'S MUNICIPAL FINANCE SYSTEM--FROM TRANSITION TO CONSOLIDATION7 18. By 1999, South Africa had completed a structural reform process at the subnational level, which resulted in the creation of nine provinces and 284 local governments. Soon after in 2000, the first municipal elections paved the way for the implementation of an intergovernmental framework based on principles of adequacy, predictability, and transparency. In recent years, the Government has sought to further consolidate intergovernmental financing in order to strike a more robust balance between fiscal autonomy (designed to encourage empowerment) and performance in particular service delivery sectors (designed to reduce the apartheid era service backlogs).8 Described below is a discussion of key issues underpinning two critical moments in the evolution of South Africa's municipal finance system--(i) the resourcing of local governments through increasingly predictable, less discretionary transfers, followed by (ii) the move to consolidate multiple sectoralized conditional transfers into multisectoral instruments. This section also discusses implications of grant consolidation for a municipal monitoring strategy. Fiscal Transition--Resourcing The Local Government Sphere 19. Two features of South Africa's fiscal transition illustrated the Government's early commitment to the creation of a viable local government sphere. First, efforts were made to effect yearly increases in the overall vertical division of revenues between national and subnational levels. As illustrated in Table 2, municipalities enjoyed considerable growth in the global allocation from national government from 2000 onwards, rising by a nominal annual average of 25% from R5,5 billion in 2000/01 to a projected R10,8 billion in 2004/05. Local governments' share of nationally raised revenue has risen from 3% to 4% over the same period, an average annual growth of 18.3%. By 2005/06, local government's share of national revenue will have risen by 91% in real terms since 1995/96, from 3.2% to 4.4% budgeted spending (less debt repayments)9 In addition to the increase in the scale of nationally raised revenue transferred to provinces and municipalities, the Government took systematic steps towards improving the transparency and predictably of this vertical division of revenue through a system of multiyear allocations as part of the overall Medium Term Expenditure Framework. 20. A second feature of the fiscal transition was the successful phasing out of the ad- hoc and opaque transfers that were a vestige of the apartheid era. Table 3 illustrates a systematic phasing out of ad-hoc, earmarked recurrent transfers and the scaling up of a more equitable and redistributive set of unconditional recurrent and conditional capital transfers. The expansion and increased predictability of the new municipal grants were supported by tighter regulation over fiscal flows to local governments under DORA and 7This section draws upon D. Savage, C. Heymans, and P. van Ryneveld. "Review of the Intergovernmental Fiscal Framework for Local Government in South Africa", 2003. 8Annex 1 contains a brief history of the debates in SA that led to the current structure. 9Nominal longitudinal data on the vertical division of revenue and grant allocations are readily available in the annual Intergovernmental Fiscal Review published by the National Treasury (see www.treasury.gov.za) and not repeated here. 8 accompanying public financial regulations. It is worth noting that, during early period of fiscal decentralization, the Government's intergovernmental strategy emphasized the need to reduce and eliminate the "infrastructure backlog"--or the unequal distribution of access to quality services, which was one of the legacies of the apartheid era. Concerns about disparities in access and quality in specific priority sectors justified design of transfers which--while predictable--were conditioned or earmarked around specific sectoral priorities (for example, those related to the Consolidated Municipal Infrastructure Programme or the Department of Water Affairs And Forestry). Accordingly, for each transfer program, a set of sector specific financial and physical reporting arrangements were established between municipalities (or provincial governments) and the corresponding transferring department at the national level. Grant Consolidation--Promoting Autonomy With Performance 21. Even as the sectoral objectives of these transfers were pursued, NT was cognizant of the other objectives that underpinned the country's local government and decentralization strategy, and therefore the fiscal stance it would take towards local governments. Put simply, the transfer system was not only intended to promote responsive service delivery in priority sectors, but also promote local autonomy, or the ability of local communities within racially amalgamated jurisdictions to have a greater voice in how municipal resources would be spent across sectors (for example, for water, sanitation, public amenities, roads and local economic development). Also, greater recognition of capacity constraints on newly demarcated local governments suggested that the sector-specific transfers along with their independent, stove-piped reporting arrangements needed to be reviewed. In seeking to re-balance sector-specific priorities with those related to local autonomy, the Government opted to consolidate sectoral transfers into multisectoral instruments. By rationalizing the number of transfers to municipalities, consolidation would also reduce the transaction costs on local authorities of multiple reporting requirements imposed by various transferring departments. 22. Consolidation Of Unconditional Grants. In 1998, NT embarked on a major initiative in grant consolidation with the introduction of the equitable share of nationally raised revenue for local government, an unconditional constitutional entitlement from national tax revenues distributed by formula to municipalities. The Equitable Share grant was formed through a consolidation of over 20 operating grants paid by provinces and national departments to municipalities. An ongoing program to consolidate other operating grants into the equitable share has been pursued, with R293 grants, which facilitated the transfer of municipalities, being fully incorporated in the 2001/02 financial year. The water operating grant was also ring-fenced and prepared for consolidation into the equitable share in this year. The latter was a grant-in-kind, funded by an augmentation to the Water Trading Account on the DWAF vote, which pays for the operation of water services schemes inherited by the department. These schemes are scheduled for transfer to municipalities, with a phased consolidation occurring after these transfers have occurred. 9 Table 2: Real Vertical Division Of Revenues, 1995-2005 (in 2000/01 Rands) Deflators 68.0 75.0 82.0 88.3 94.4 101.8 108.4 115.6 122.0 127.7 133.4 R' m Actual Actual Actual Actual Actual Actual Actual Revised Budget Budget Budget 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 National 69300 71956 71391 68860 70317 71890 80902 85490 89297 92042 94709 Provincial 105096 110951 105534 105614 105356 106982 111700 118410 130276 137393 143641 Local 5743 6867 7377 5308 4883 5439 6010 7611 9833 10409 10964 Total 180138 189773 184302 179781 180556 184311 198613 211512 229406 239843 249314 Total Growth 91% Percentage Shares (%) National 38.5 37.9 38.7 38.3 38.9 39.0 40.7 40.4 38.9 38.4 38.0 Provincial 58.3 58.5 57.3 58.7 58.4 58.0 56.2 56.0 56.8 57.3 57.6 Local 3.2 3.6 4.0 3.0 2.7 3.0 3.0 3.6 4.3 4.3 4.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Growth in Shares (%) National 3.8 -0.8 -3.5 2.1 2.2 12.5 5.7 4.5 3.1 2.9 Provincial 5.6 -4.9 0.1 -0.2 1.5 4.4 6.0 10.0 5.5 4.5 Local 19.6 7.4 -28.0 -8.0 11.4 10.5 26.6 29.2 5.9 5.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 10 Table 3: Fiscal Transition And Consolidation As Evidenced By Real Grant Allocations, 1995-2005 (in 2000/01 Rands) Actual Projected Budget Deflators 68.0 75.0 82.0 88.3 94.4 101.8 108.4 115.6 122.0 127.7 133.4 R'm 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 Equitable share (incl IGGs) 1221 1075 1101 27 1772 1834 2415 3428 5197 5542 5771 R293 personnel 824 1152 1077 1077 490 455 ­ ­ ­ ­ ­ R293 transfer 126 107 83 ­ 42 ­ ­ ­ ­ ­ ­ Transition grant ­ ­ ­ ­ ­ 98 231 193 ­ ­ ­ CWSS (Operating) 0 663 601 678 752 733 638 605 685 672 700 Equitable share & related 2171 2996 2862 1783 3057 3120 3284 4226 5882 6214 6472 % increase ­ 38% -4% -38% 71% 2% 5% 29% 39% 6% 4% % of total allocation 40% 46% 39% 43% 62% 56% 54% 54% 59% 59% 59% Agency transfers 78 82 95 ­ ­ ­ ­ ­ ­ ­ ­ Traditional subsidies 23 26 59 ­ ­ ­ ­ ­ ­ ­ ­ Other 48 162 55 ­ ­ ­ ­ ­ ­ ­ ­ CMIP 126 540 1032 795 737 867 917 1445 1840 2133 2261 Water Services 496 321 1121 1151 663 598 758 864 903 742 777 Com-based pub works ­ ­ ­ 172 290 367 345 225 213 ­ ­ Sports & Rec. prog. ­ ­ ­ ­ ­ ­ 37 66 101 ­ ­ LED Fund ­ ­ ­ 5 102 70 96 96 ­ ­ Urban Transport ­ ­ 246 42 32 22 75 35 7 ­ ­ National Elec Prog ­ ­ ­ ­ ­ ­ ­ 197 197 192 193 Integ Rural Devt ­ ­ ­ ­ ­ ­ 30 28 ­ ­ ­ Municipal Infra Grant ­ ­ ­ ­ ­ ­ ­ ­ 39 92 73 Unallocated ­ ­ ­ ­ ­ ­ ­ ­ ­ 435 441 Infrastructure transfers 771 1131 2607 2160 1728 1957 2232 2955 3395 3593 3746 % increase ­ 47% 131% -17% -20% 13% 14% 32% 15% 6% 4% % of total allocation 14% 17% 35% 52% 35% 35% 37% 38% 34% 34% 34% Current transfers Agency transfers 1406 1432 1309 ­ ­ ­ ­ ­ ­ ­ ­ Traditional subsidies 87 131 104 ­ ­ ­ ­ ­ ­ ­ ­ Other 946 800 315 ­ ­ ­ ­ ­ ­ ­ ­ Land dev objectives ­ ­ 4 27 15 43 ­ ­ ­ ­ ­ Municipal systems improve ­ ­ ­ ­ ­ ­ 40 81 123 143 317 Fin management ­ ­ ­ ­ ­ 49 55 133 174 156 156 LG support grant ­ ­ 177 205 145 147 148 208 190 172 ­ Restructuring grant ­ ­ ­ ­ ­ 221 323 216 258 269 272 Current transfers 2438 2363 1907 232 160 461 565 639 745 739 745 % increase ­ -3% -19% -88% -31% 188% 23% 13% 17% -1% 1% % of total allocation 45% 36% 26% 6% 3% 8% 9% 8% 7% 7% 7% Total 5380 6490 7377 4174 4944 5538 6081 7821 10023 10546 10963 11 23. Taken together these are unconditional grants that are formula-driven primarily on the basis of poverty and own resource indicators (the formula is updated periodically and is explained in each year's Budget Review). As entitlements, these transfers are in keeping with SA's decentralized and cooperative governance system, and are programmed and utilized in accordance with local priorities. The rules governing these transfers are simple and codified in the MFMA and MSA. Local governments are obliged to consult with and report back to their constituents when prioritizing and using these grants (like all budgetary resources including own revenues). Upward accountability for unconditional grant utilization is more limited, primarily focused on compliance with the organic finance legislation. 24. Notably, consolidation has also provided an opportunity to improve the allocation rules around unconditional transfers to finance local government recurrent expenditures. Specifically, two alterations have been made to the allocation formula of the equitable share. Firstly, a funding bias towards municipalities in the urban renewal and rural development programs has been introduced. In theory, this bias affected all 14 national transfer programs from 2002/03, although this bias has not been uniformly introduced. Simultaneously, a grant to support the development of institutional capacity in these "nodal" municipalities was also created. Secondly, from the 2002/03 financial year the equitable share has been allocated to district municipalities that perform basic service delivery functions. These allocations were made uniformly to qualifying district municipalities for the first time in the 2003/04 national financial year. Further refinements of the formulae, reflecting adjustments in priorities, are made periodically. However, the core principles of transparency and predictability are maintained. 25. Consolidation Of Conditional Transfers--Capital Financing and Capacity Building. The Government's grant consolidation strategy has also been applied to capital funding mechanisms (for example, CMIP). In March 2003 Government took a policy decision to consolidate all municipal infrastructure grants into a single, decentralized and formula driven capital allocation to municipalities, known as the Municipal Infrastructure Grant (MIG). Seven existing sectoral grant programs are being phased into the MIG over three years, from 2003/04. This initiative has built on the limited consolidation first achieved in the formation of the Consolidated Municipal Infrastructure Programme(CMIP) in 1998, and will provide non-sector based, multi-year funding to municipalities for the installation of basic municipal infrastructure for poor households. A similar approach is planned for capacity building. 26. The MIG would serve as the key mechanism through which national resources are distributed to local governments for the specific purposes of "incorporat(ing) national priorities in municipal budgets" and "address(ing) backlogs and regional disparities in municipal infrastructure" (see 2004 Budget Review, pg. 276-279). As such, local governments are justifiably required to meet more extensive reporting standards for upward accountability in the utilization of conditional MIG funding. MIG allocations are authorized through DORA and are administered by the Department of Provincial and Local Government (DPLG). The MFMA addresses key issues, which have constrained the effective utilization of conditional infrastructure grants to local governments in the past, particularly with regard to multi-year funding of larger infrastructure projects. Roll 12 over of funds, including MIG, across financial years is now authorized through the MFMA, and DPLG is finalizing detailed MIG processing and reporting guidelines consistent with monitoring reforms focused on outputs. Indicative of the Government's commitment to grant consolidation, the MIG is currently the only conditional grant projected to be available through DORA to local governments in FY06/07.10 27. The creation of multisectoral "pipes"--conditioned on strong budget management disciplines rather than sectoral allocations-- to finance recurrent and investment needs has emerged as a distinctive South African innovation that other countries undertaking rapid decentralization have increasingly sought to emulate (for example, Uganda and Ethiopia). As Table 4 indicates that by FY 06/07 the great majority of SA's municipalities will have to manage, monitor and evaluate a maximum of six--and in most cases only five--grants through DORA: (i) an unconditional Equitable Share and related transfers (including the R239 Personnel Grant) that meet the recurrent needs of local governments; (ii) conditional transfers to meet municipal investment needs including water, power, sports and recreation, and integrated rural development, now being consolidated in the Municipal Infrastructure Grant (MIG); (iii) a set of conditional transfers designed to meet recurrent municipal expenditures (including grants for capacity building and restructuring).11 The total value of projected national transfers to local governments in 06/07, at current exchange rates, will be on the order of US$ 3 billion. The extent and methods by which these transfers are monitored must reflect the nature and purpose of the transfers themselves. Table 4: Projected National-Local Transfers, 2006/07 Type of Transfer Amount in R million Equitable Share 9,365 Water/Sanitation Operating 991 Unconditional 10,355 Municipal Infrastructure Grant Conditional 5,987 Restructuring Grant 350 Financial Management Grant 199 Municipal Systems Improvement 200 Current Transfers 749 Total Transfers To Local Government 17,091 10Current transfers will, by FY 06/07, be limited to three specific transitional grant programs, all of which focus on capacity building and restructuring. Each of these programs has specific requirements for access and reporting with which participating local governments must comply. As specific purpose, transitional grants with finite life spans, it is not necessary to make provisions for ongoing monitoring of these transfers within the overall intergovernmental system. 11The Restructuring Grant is available only to the 25 largest municipalities. 13 Implications For A Municipal Monitoring Strategy 28. The evolution of intergovernmental fiscal relations in South Africa (and other decentralized settings) has far-reaching implications for how municipal monitoring systems are developed and established. Monitoring concerns such as the data carrying capacity of various reporting arrangements, mechanisms for data verification, and the importance of bottom-up feedback should inform the design of the fiscal system over time. Based on the framework laid out in Section II, the NT and Bank teams jointly noted that the following issues as informing a robust municipal monitoring strategy for South African municipalities over the medium-run: Striking a balance between the service delivery and local autonomy objectives of decentralization requires re-design of fiscal transfers (municipal grants) as well as grant monitoring systems. Specifically, the creation of a limited number of consolidated, multisectoral conditional transfers implies a reduction in parallel reporting relationships between municipalities and higher levels of government, and potentially less complicated reporting requirements on capacity constrained local governments. It also places a premium on clearly selecting financial and physical performance indicators, creating efficient verification mechanisms, and improving coordination between national departments that previously managed separate programs. Consolidated multisectoral grants are also designed to encourage local empowerment in making resource allocation decisions; monitoring systems should therefore seek to provide feedback to improve local decision-making. A robust strategy is one that "phases in" elements of a full-fledged monitoring system in step with local and national data carrying capacity, the requirements of sound fiscal and financial management, as well as program priorities. The Government has taken a risk aware approach by first prioritizing fiscal monitoring (including grant utilization, appropriate use of funds, quality of fiscal data) within municipal grant programs before taxing limited local capacity to track public management processes, institutional quality, service outputs, and impact. The fiscal and financial focus ensures that the basic fiduciary requirements of MFMA are met first. As the decentralization process matures in SA, monitoring should go beyond inputs provided by grants, and begin to capture data on local government processes that inform grant management, the service outputs they deliver, and their impact on households. What is now a grant monitoring system--narrowly defined--should be progressively broadened to address a larger set of public accountability processes in municipalities. Feedback mechanisms should balance vertical or top-down accountability (of municipalities to transferring departments) with horizontal accountability (of municipalities to their citizens and their elected representatives). SA's municipal monitoring strategy should employ tools that serve the needs of financial and physical reporting to higher levels of government, as well as the needs of local decision makers. The latter implies more information and a greater emphasis on public reporting, benchmarking, and knowledge sharing in order to encourage local level learning. 14 Figure 2. From Grant Monitoring To Municipal Monitoring Accountabilities Vertical Horizontal (totransferringdepartments) (tocitizensandtheirrepresentatives) Inputs Grant Monitoring(Phase 1) - focus onfinancial monitoring (suchas utilization, Withpublic reporting quality of fiscal data) Processes Grant Monitoring (Phase 2) Withpublic reporting rage - focus onprocesses, outputs, andimpact veoC Outputs Data Impact Benchmarking - Fiscal andfinancial performance, all key publicprocesses, service levels, andimpact 29. Taken together, SA's municipal monitoring strategy aims to progressively improve the quality and coverage of data on the inputs, processes, outputs, and impact in grant programs; and to report these data in a manner that is increasingly accountable to both transferring departments and local decision-makers (through public reporting and benchmarking arrangements). The vision of a full-fledged monitoring system is one that ultimately benchmarks the performance of the municipality as a whole--its expenditure and revenue (fiscal) performance, key processes (such as planning, budgeting, personnel management, and consultative processes), service levels and other outputs, and impact on communities and households. The MFMA requires local performance reporting on a quarterly basis. In other words, the performance of grants is but one element of this larger system. 30. The above-mentioned phased approach provide the basis for the discussion of fiscal data quality and financial accountability within grant programs in Section IV, and the development of more comprehensive municipal monitoring systems in Section V. 15 SECTION IV. THE GRANT MONITORING CHALLENGE--PRIORITIZING FINANCIAL ACCOUNTABILITY12 31. As noted in Section III, NT's monitoring strategy first seeks to ensure that financial resources under municipal grant programs are utilized in an efficient, accountable, and transparent manner, in line with the obligations laid out in South Africa's public finance legislation. This section therefore summarizes the findings of an assessment, undertaken in late 2003, of the core fiduciary framework underpinning DORA. The assessment examined the flow of funds to municipalities and its implications for the design and data carrying capacity of fiscal reporting arrangements; the capacity of various municipalities to manage reporting under existing and future fiscal flows; and options for rolling out a municipal financial reporting system (including the design of new formats) that build on existing capacities.13 DORA And The Local Government Fiduciary Framework 32. South Africa's Division of Revenue Act (DORA) requires that NT gazette the three-year allocations for transfers and grants in-kind to municipalities from the national government. Publication of this information enhances the predictability and transparency of transfers to municipalities. Provincial governments are also required to gazette their own transfers to municipalities. Taken together the national and provincial government gazettes allow municipalities to identify the full extent of transfers allocated to them for the current and two subsequent national fiscal years. Section 71 of the MFMA also requires that every municipality receiving assistance under a grant program submit monthly reports to the relevant transferring national or provincial department. Monthly reports from municipalities are submitted to provincial treasuries and onward to national level in consolidated form. Consolidated reports for each grant are also submitted by transferring national departments on a monthly basis to NT using a basic spreadsheet format. Efforts are underway to streamline the inter-departmental arrangements that govern monthly reporting--under Section III2 of the Public Financial Management Act (PFMA), Section 71 of the MFMA, and DORA--from transferring departments to NT, and feedback mechanisms to national departments. 33. As NT collects and analyzes data on transfers and expenditures for grant programs, it reviews the coverage of these data on a quarterly basis. Specifically, there is need to expand the types of fiscal management data generated on municipal grants-- based on clear criteria and existing data collection capacities--to more thoroughly analyze grant performance on a routine basis. For instance, data on receipts of cash transfers into municipalities would help track the flow of funds and also monitor whether (and how large) a proportion of transfers are absorbed by departments or district municipalities before reaching local municipalities. Information on the timeliness of receipts from various grant programs, as well as cash or liquidity management within municipalities would also provide valuable indicators, which could then be integrated 12This section draws upon M. Frazer. "Strategic Review of DORA Monitoring," 2003; and J. Druker. "DORA Grant Monitoring--A System for Benchmarking Municipal Grant Performance," 2004. 13A concept underlying the review of the fiduciary framework is the potential of dual reporting arrangements in improving the quality of fiscal data in part through data validation and reconciliation. 16 into the proposed grant-monitoring module of the Local Government Monitoring Database--which is currently under development by NT. 34. In addition to coverage, reliability of existing fiscal and financial data collected by transferring departments and compiled by NT is central to the integrity of DORA monitoring. As noted above, the absence of dual reporting on municipal grants means that NT lacks a parallel flow of information against which grant data from transferring departments can be reconciled and verified. NT is moving towards a dual reporting system for municipalities and is considering the provincial grant monitoring system as a model. Specifically, the dual reporting model for provincial grants could be instructive for the design of municipal grant reporting and monitoring. Under the provincial grant system, fiscal reports are sent by provincial departments to their national counterparts, and are then consolidated and submitted to NT. These data in turn are reconciled against reports provided by provincial treasury departments to NT. Monthly reports, once reconciled, are consolidated and analyzed for expenditure trends on a quarterly basis. These analyses are used inter alia to inform budgeting of allocations under provincial grants in following years. An automated Early Warning System is to be implemented for capture and reconciliation of expenditure on provincial grants. Options for setting up a similar dual reporting system for municipalities are being explored as a way of ensuring a streamlined and more systematic approach for data reconciliation and analysis within NT. Data Capture Requirements And Municipal Capacity 35. As the local government system matures, municipalities require the capacity to respond to a range of demands including financial management and reporting, human resource management, implementation, procurement and contracting, monitoring and evaluation, as well as a range of other local level processes. From a grant monitoring perspective, it is important to design reporting and monitoring systems in a manner that is "fitted" to the capacity constraints in municipalities. 36. Accordingly, the NT and Bank team undertook a systematic review of reporting systems across grant programs. The reporting systems that already exist or are being developed under various initiatives to address these data needs are illustrated in Table 5. It is apparent that numerous parallel reporting arrangements are currently aiming to capture financial, physical, and output data from municipalities on a weekly, monthly, quarterly, or annual basis. The assessment noted opportunities for developing a full fledged municipal monitoring system by (i) building on and streamlining existing reporting relationships, (ii) tailoring reporting formats and arrangements to municipal capacities, and (iii) guiding efforts to develop the system on the overriding short-term priority of ensuring financial accountability. 37. Municipal Reporting Capacity. A critical concern would be to identify least cost methods (for example, through a system of coding specific grants by name, source and fiscal year to each municipality: "FMGNT 06" ­ Financial Management Grant/National Treasury/FY) to progressively elaborate such as system based on a realistic assessment of data generation, collection, and verification capacities at the local level. The NT and Bank teams therefore undertook field visits to the following municipalities: Amatole, 17 Table 5: Existing Reporting Systems And Requirements For Municipalities System Tools used Frequency Financial Data Outcome Data MBFRP1 Phase Financial Reporting to 1 & 2: 39 pilot Treasury database in Monthly* Yes Limited Municipalities Treasury B of Budget Returns format Yes, but at Debt Returns on Debt level Quarterly present only from debt No suppliers Grant Returns from Yes, but monitoring Departments on Monthly Yes varies by spreadsheets dep't Financial returns via spreadsheets entered on Monthly Provinces database, validation to Yes No national financial data Weekly using Vulindlela system Submission of Budget Annual Budget Returns returns in Treasury A of Yes No Budget Returns format Buffalo City, Nelson Mandela, Mangaung, Eastvaal and Polokwane. During these visits, the first point of contact in each case was with the Municipal Treasurer's Department and the issues addressed included responsibility for and frequency of reporting at the local level; the full range of reporting required by provincial and national departments; capacity constraints (shortage of staff, lack of training/skills, infrastructure, IT etc); other difficulties experienced; view of legal obligations with regard to grants and Equitable Share (DORA, MFMA); as well as relationships with and reporting to provincial departments. 38. A key finding of the assessment was that municipal capacity to handle existing reporting requirements is highly varied. Metros or large districts have in many cases adequate resources, but capacity tails off dramatically outside this top grouping. While the MBFR initiatives targeted 80% of the total local government spend, the capacity gap appears to be widening between the 39 municipalities participating in the MBFR and the remaining 245 municipalities. Even MBFR pilot municipalities have experienced difficulties, particularly in coping with the transformation process, the complexity of returns, impact of new systems and generally accepted municipal accounting practices (GAMAP). There is significant scope for identifying which reporting relationships work better than others (or are likely to work), streamlining those relationships, and using them as an entry point for building a workable system. 39. In addition to complex reporting relationships, the assessment noted that municipal capacities are taxed by unpredictable and "source-less" tranches of grant resources to municipalities within any given fiscal year (Box 1). It is envisaged that grant consolidation efforts (for example, the creation of the MIG) will increase predictability, 18 Box 1: The Hidden Costs Of Unpredictable And "Source-less" Funds Flows A pressing concern with regards to data quality is the impact of existing transfer and flow of funds arrangements on the ability of municipalities to appropriately identify and report on cash transfers by source (for example, the specific grant program, the level of transferring authority, and predictability of receipts vis-à-vis cash projections). Anecdotal evidence based on field visits suggests that municipalities--even those with robust cash management systems--had been hampered by unpredictable, and unidentifiable cash releases into their bank accounts through the budget year. The lack of predictability was caused in part by the failure of transferring departments (whether national, provincial or district municipal) to provide prior notification of cash releases to municipalities. The transfers therefore appeared to be arbitrary and inconsistent with cash projections prepared for each grant program. Moreover, the flow of funds varied across grants programs. For instance, some grants transfer cash to provincial departments or district municipalities before sending them on to municipalities. Anecdotal evidence indicated that a portion of the transfer may be absorbed--often arbitrarily--by these spheres of government en route to municipalities. As a result of the constraints imposed by these transfer arrangements, accounting personnel in municipalities spend considerable time and effort trying to "reconstruct" the source of cash transfers (for example, the specific grant and level of government) as a basis for their monthly reports. The implications for fiscal data quality within municipal grant programs are significant and show cause for concern. limit problems with identification of funding sources, and reduce the reporting burdens on municipalities. Yet, the assessment of the reporting framework suggested that transfers like the MIG would demand more inter-departmental coordination to effectively monitor fiscal and operational data across various sectors. By FY 06/07 SA's 284 local governments will be enabled to access a single source of national transfers to address infrastructure backlogs that still face previously disadvantaged citizens, the Municipal Infrastructure Grant. MIG is administered by DPLG and by FY 06/07 will, supplemented by the Municipal Systems Improvement transfer, be the main conditional instrument by which that Department can directly support local government investment. The primary beneficiaries of MIG are smaller municipalities with relatively weak own resource bases, and it is specifically these municipalities that generally face the most critical capacity constraints across the board, and certainly in terms of ability to meet complex conditions for draw down of funds and reporting on expenditures, outputs and outcomes. 40. Even as reporting arrangements are rationalized and transfer mechanisms are strengthened, there is need to build the capacity of local jurisdictions. Accordingly, the Government has instituted useful measures such as the MFMA capacity building program. The core of financial reporting is the same for each of the existing monitoring returns to various sources. This financial information is cash based, and is therefore already partially disassociated with physical outcome/progress information. There is clearly an opportunity to consolidate and streamline the financial component of monitoring to create a common flow, preferably deliverable directly from a standardized report format feeding into a single database. 19 Fiscal Monitoring As An Entry Point For Grant Monitoring 41. As part of its "Strategic Review of DORA Monitoring," NT considered various approaches in formulating its strategic stance towards grant monitoring in particular. NT concluded that the following principles could usefully inform options for phasing in a full fledged monitoring system for the local government sphere: Since the major constraining factor is the capacity of all municipalities to co- operate, it is also essential that any new initiative not aggravate this situation, at least in the short term, by adding additional reporting requirements. NT and other transferring departments should ensure that fiduciary risks are managed first by creating a graduated approach by which the remaining 239 municipalities can implement the MBFRP. To ensure data quality, NT should build on the provincial grant experience and create a minimum number of parallel information flows--for example, through dual reporting arrangements--in order to facilitate regular data validation and verification. As it is developed, the system and its constituent elements (such as reporting formats, relationships, and verification procedures) should be fitted to the capacity of municipalities to comply in compliance with MFMA requirements. The system should progressively but cautiously generate process, outputs, and outcomes data, initially from Program Managers' information at Departments, but ultimately from a flow of management accounting information direct from municipalities. 42. Based on these principles, the strategic review identified two set of options that would allow the Government to a develop a monitoring flow that allows NT, over the long run, to capture--and share with all key Government departments--all financial and out-turn information on municipalities required for effective intergovernmental fiscal management (Figure 3). 43. Option A, considered during the review, was to design and implement a new system at national level that replaces existing systems in a fully integrated manner. The approach had some advantages in terms of information technology, because variations of software and of spreadsheet models in particular have created difficulties in consolidating data. Even though the existing systems have been created independently of each other, with different objectives, they seek to capture the same core financial information from municipalities. The development of an integrated system with standard financial data is an attractive option in technical terms. 44. Notwithstanding these potential benefits, Option A does pose several problems. First, the creation of a new system would require NT, national departments, and municipalities to manage what might be an extended transition to a new system, during which the quality and coverage of fiscal and financial data may suffer. Second, the 20 Figure 3: Long Term Monitoring Flows Within The Municipal Grant System National Treasury Management Accounting Information Monitoring System (Appendix B plus output monitoring) Grant Reporting Data Municipalities Output Information Departments Share (assuming that all In MBFR roll-out) Validation report Provinces Monitoring Key Sector Primary Data Flow Data Validation Flow mixed track record of grant programs in capturing data from municipalities suggests that a significant investment in local capacity would be required; the review concluded that local jurisdictions would very likely have difficulties absorbing such a scale-up. Third, the adoption of yet another system would create redundancies with existing initiatives already underway and heavily funded through the pilots in the Municipal Budget and Financial Reform Program (MBFR). Fourth and finally, creating an integrated system from scratch raises questions of ownership across national level departments including those concerned with specific sectors such as water, sports and recreation, etc. 45. The review also considered a more incremental approach, Option B, which would build on existing reporting systems--enhancing, updating and replacing as necessary to realize the long term monitoring flow illustrated in Figure 3. While Option B does not immediately and directly address issues of compatibility (noted above), it possesses several advantages over Option A. First, Option B would build on capacity investments already made through MBFR and other such initiatives. Second, an incremental approach does not preclude the development of a strategy that allows development of various elements of an integrated system. Third, the risk of a "dip" in monitoring systems below the existing level is reduced. Fourth, this approach builds on existing mandates and ownership of various initiatives by national departments. 46. Based on the recommendations of the review, NT has sought to pursue Option B as a basis for developing an integrated grant monitoring, and eventually a full-fledged municipal monitoring system. What follows is a further elaboration of a phased approach to implementing Option B, which was initiated inter alia during this AAA work. Pursuing Option B--Key Components Of A Phased Approach 47. As a starting point, one of the priority issues from the Treasury perspective, but also a need shared by DPLG, StatSA and Program Managers for Grants in Departments is "financial" monitoring. The original source data for all the financial information, flowing from the municipality should be the municipal finance system. Some of these flows go to provinces, which act primarily as consolidation/validation nodes. Phased 21 implementation of Option B would use existing systems to build four components into the monitoring system. These components, described below, are the SQL System, Validation Reports, Grant Funds, and Reconciliation. 48. SQL System. This system developed, as part of MBFR is relatively new and at an early stage in rollout, but it is being developed on SQL software that can handle the scale of data that will eventually originate from municipalities. Provided that it develops with some flexibility in approach and design, the MBFR system can serve as the backbone for the financial aspects of municipal monitoring overall. 49. Validation Report Flows. With regards to the information flow referred to as the Validation Report, Section 71 of the MFMA requires in-year reporting as follows: Monthly. By the tenth working day of each month, municipal managers must submit reports on budget progress to mayors and provincial treasuries, and provincial treasuries must submit to NT consolidated reports for all municipalities no later that the 22nd working day of each month. Quarterly. By the 30th day following the last month of each quarter, the mayor must submit to council a report on budget implementation and the municipality's financial position. Once adopted by council, the municipal manager must post the quarterly report on the municipality's website within 5 days. Semi-annually. By January 25 each year, the manager must submit to the mayor, NT, and the provincial treasury a mid-year budget report and performance assessment (based on that year's service delivery and budget implementation plan, as required under Sec. 69(a) of the MFMA) Figure 4. First Phase Of A System Roll Out 22 Annual audited reports, to be tabled in council January 31, and to include an annual performance report as prescribed in Section IV6(1) of the Municipal Systems Act. 50. Given these reporting requirements, the intervening consolidation by provinces on a monthly basis, and the mid-year and annual performance reports, there is ample opportunity for parallel data validation. 51. As noted earlier, a key concern is the lack of capacity at municipal level, and how to motivate completion of the report. One of the lessons learned in current municipal financial reporting systems is that quarterly reporting is difficult to achieve on schedule; consequently, some incentive mechanism--in terms of lowered transaction costs and nominal or fewer demands on municipal capacities--would need to be incorporated into the process for municipalities. 52. Introduction Of A Treasury A-Style Report. Accordingly, an approach identified during the strategic review was to create a report that partially mirrors the format of the Treasury A Annual Budget report, yet makes sure the report format is short and easy to complete. Such a reporting arrangement would work as follows: All municipalities have to complete the Treasury A Budget form annually. A 4 or 5 page report designed in a format similar to Treasury A could be derived directly from the Municipality Accounting Package with an expansion of the grants section. The report would be familiar in layout to Municipal Chief Finance Officers (CFO). It would also be of benefit to the CFO in preparation of Treasury A budget forms and would create a parallel financial reporting flow, and fill the gap until MBFR is universal in all municipalities. The routing/flow of this return to NT via provinces will build on relationships that already exist in national/province/municipality transfers, and would "encourage" the capture of this information. It will also leverage relationships between provinces and their constituent municipalities. The treasury function rather than local government function would be responsible for Annex A-style reports at the provincial level for two reasons. First, the content of the reports is financial in format. Second, there already exists a return and compliance culture between the provincial treasuries and NT (building on the provincial grant monitoring experience and MFMA delegations). 53. The introduction of a Treasury A style monitoring report is also based on the premise that it can be derived from the more detailed Treasury B report, which is generated from the systems being developed by the MBFR municipalities. In other words, the Treasury A style report is a summarized version of Treasury B. This makes it possible for those municipalities in transition under MBFR to submit reports via the Treasury A-style system. In addition to ensuring complete municipal coverage, the report would serve as a stepping-stone in to assist the rollout of MBFR. 23 54. Grant Monitoring. The systems in place for grant monitoring are all Excel spreadsheet based. Each grant has a spreadsheet model based around a generic template that has been developed in NT. Beneath this level, with the individual program managers in departments the collection of data varies between grants, although in most cases the information from provinces and municipalities is stored in spreadsheet format. 55. The NT spreadsheet monitoring system is currently a `stand-alone' tool and the spreadsheets themselves are large flat files containing the entire financial year. Each municipality is covered and significant amounts of analysis undertaken. In their present format they are not easy to integrate into a database. The format of collection by grant in each spreadsheet model also makes cross analysis by municipality more difficult. 56. The present approach does have significant advantages for the collection of the data. Specifically, it allows the flexibility of completion from their own grant system for Program Managers in the departments, also as a model developed in­house the users in NT have a high degree of familiarity. However, given the longer-term objective of a coordinated system that holds all of the monitoring data in compatible format, there is need to enhance the grant monitoring system with an integrated SQL Server software solution. 57. Various criteria were applied when opting for this spreadsheet-based approach. First, the transition process did not exacerbate the risk of a "dip" in reporting. The current monthly returns level being stipulated from all municipalities is the ultimate reporting target in terms of speed, for all forms of monitoring. Second, the flexibility that spreadsheets allow in terms of interpretive analysis and graphical presentation should be maintained. Third, there is little or no training required for Program Managers and their staff in completion of the spreadsheets, with which they are familiar. 58. An appropriate short-term solution therefore is to restructure the existing spreadsheet models to facilitate capture of the grant data in limited database format. To this end, support was provided through this AAA to develop a web-based reporting format, feeding into a relational database implemented on NT's existing server. The web based reporting form developed during this AAA work has been included in Annex 3 for illustrative purposes. This standardized form has been tested and provides NT and provinces with important capabilities. NT and provinces now have uniform information that can be monitored on funds received and spent by municipalities and explanatory notes on non-compliance with conditions. This form of reporting also allows reconciliation between grant administrator reports on transfers and municipal reports on receipts. It also allows identification of slow spending on grants leading to analysis of blockages. The spreadsheet report allows analysis of reasons for non-compliance providing indicators for targeted interventions. 59. It is worth nothing that municipalities require only a client terminal that runs Windows and Internet Explorer in order to utilize the web-based reporting form. No additional IT training or equipment is required in the great majority of municipalities to utilize this form. Annex 4 provides a snapshot of the design of the database in which reported data will be managed. The database is implemented on an SQL 2000 server 24 already owned and operated by NT. Again, operation of this database will not require additional training or equipment for NT. 60. Reconciliation Tool. The final component to add to the Phase 1 model is the tool that allows reconciliation reporting between the now technically compatible data sources. As indicated above all data is in SQL-compatible format and, therefore, the reconciliation tool will be a series of reports which track and identify across common data. The creation of reconciliation reports is scheduled from the start of year 2 onwards. Implications For Phasing In the System 61. Overall, the introduction of the standardized web based reporting format and of the relational data base in NT are important first steps to achieving accountability upward of all municipalities for receipt and expenditure of funds, and clarifying the responsibilities and required capacities of all three spheres involved in the SA intergovernmental fiscal system. The phased approach also allows the development of individual components but under an overall monitoring system development framework. The principle of progressive enhancement of systems or even the creation of a new database (Treasury A-style returns) can be placed within the overall context of a complete municipal monitoring system, with high level users accessing via a front end query tool to extract information from all of the databases. 62. The issue of skills and capacity in municipalities has been noted, but it is also important to highlight that the activities being considered will require an improved level of capacity at NT, DPLG and provinces. These skills will include: Project Management, Systems Development Management, Systems Maintenance, and Data Analysis. This is based on the assumption that the programming skills requirement will be met from either an internal specialist unit or contracted out. Even if work is outsourced, internal Figure 5. Summary Of Plans For First Phase Of System Roll Out Year 1 Year 2 Year 3 Activity Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Component 1 - MBFR Roll-out MBFR Roll-out Phase 1 MFBR Roll-Out Phase 2 Further Roll-Out MFBR Component 2 - Validation Enhanced MFMA Partial Shift to in-house operation In-house Validation System Component 3 - Grant Monitoring Development of Grant Monitoring Database Testing and Operation of Database Component 4 - Reconciliation Development and Implementation 25 management and analytical capacity building will be required. Consequently, it will be necessary, in parallel with capacity building measures at municipal level, to ensure that data is actually generated and also invest in the skills of staff at provincial and national level to interpret and use the systems. 63. Additional work in three key areas will be required as the monitoring system matures--debt monitoring, budget returns, and out-turn monitoring. Debt monitoring is not included in the Phase 1 work discussed here because, in terms of interim financial year debt, some information is already generated in the returns from the MBFR initiative. For a debt monitoring system to be effective the phasing of debt during the financial year needs to be established, usually by historical analysis. It is possible that the increasing monitoring by other systems will affect this phasing. It is also important to recognize that the reasons for fluctuations in debt may be affected by the funding flows. Given these factors, the incorporation of debt monitoring into the other reporting flows should be undertaken after the other systems have settled down and some history can be incorporated into the monitoring process. In the short-term, monitoring via debt suppliers will continue and give NT an early warning signal in terms of a municipality changing its debt levels significantly. Budget returns are not treated as a separate flow because when the Treasury B style returns are fully operational they will cover the budget position of the Municipality. Another significant area not specifically addressed in the first phase of activities is that of out-turn and other related types of measurement. The reasons for this are that (i) in some areas such as grant monitoring there is already limited out-turn data received, and (ii) there is a general question of responsibility for evaluating out-turns/outcomes that needs to be resolved, and which is better left until the success of the financial monitoring initiatives is evaluated. 64. If deemed feasible over the medium-term, these areas can be added successfully without damaging the flow of information, the following can be added to the first phase of work (i) increased out-turn reporting in MBFP pilots and enhancements of returns, and (ii) increased data collection fields specifying out-turn reporting for program managers. The introduction of these initiatives, however, presupposes the successful rollout and implementation of the basic financial reporting systems. To ensure there is not a "dip" from the current level of monitoring on physical performance, existing levels of reporting in grant monitoring should be retained. Pilot work on improved impact monitoring is discussed in Section V of this report. 65. The financial information in all of the reporting requirements comes from the same source--the municipal finance system. Therefore, duplication of the same cash based financial information needs to be eliminated through consolidated reporting arrangements. In addition to capacity investments at the local level, provincial and national departments also require increased capacity to sustainably manage and 26 effectively utilize improved information flows from local governments. In addition to the MBFR, support is being provided through the MFMA implementation program to ensure that all 284 local governments acquire needed capacity and systems within the three-year roll out period. These efforts are building a sound basis for building future integrated monitoring systems. SECTION V. GRANT MONITORING AND BEYOND 66. This section draws upon an assessment of monitoring and evaluation methods relevant to municipalities and municipal level service delivery, as well as pilot research carried out with National Treasury in order to develop a practical approach to enhancing impact evaluation capacity to inform intergovernmental policies. Significant value-added from this AAA derives from the pilot activity, which was primarily designed to demonstrate the kinds of information that can be collected and analyzed through household surveys that can be carried out routinely as part of NT's monitoring responsibilities. The research did not attempt to deal with issues of sample selection, which will be addressed as the system matures and is rolled out nationally.14 What follows is a discussion of the framework within which impact assessment of municipal programs can be carried out, an overview of existing efforts, and more detailed explanation of how the pilot was designed and implemented to push the municipal monitoring frontier. Moving Beyond Inputs 67. There are myriad ways in which South Africa's local government policy framework--particularly as it applies to service delivery and purpose-built poverty alleviation programs--impact or will likely impact wellbeing. These factors include the profile of the poor themselves, their general characteristics and specific behaviors, as well as the quality of local government institutions and processes, and how they impact the interface between public servants and citizens. To ensure effective municipal management, it is important to regularly monitor and evaluate the entire chain of public action--that is, revenue mobilization, public spending, resource management, implementation, outcomes, and impact--and use those findings to refine future government policies and behaviors. Put simply, monitoring of municipal grant resources and their utilization is necessary but not sufficient to promote effective municipal governance and the achievement of various policy objectives. In order to evolve the grant monitoring system beyond its current focus on fiscal and financial concerns, there is a need to progressively improve public information on the performance of municipalities and to assess the quality of targeting of grant programs within these jurisdictions and specifically how municipalities are dealing with the spatial, distributive, and economic legacies of apartheid. As such, municipal monitoring needs to go beyond inputs, and begin to benchmark local government processes as well as household and economic development impact. 14This section draws on a background paper by R. Govender. "Improving Municipal Monitoring and Impact Assessment," 2004. 27 68. Notably, South Africa's MFMA and MSA legislation provide an enabling framework for local governments to prepare and regularly evaluate performance and service delivery plans. In addition, its Medium Term Budget Policy Statement (MTBPS) sets out the country's strategic and fiscal frameworks for the realization of national economic and social goals, with poverty reduction as the overarching policy objective. The 2003 MTBPS clearly defines the ongoing need to design and install mechanisms for the monitoring and evaluation of the progress achieved by government structures and programs, thereby providing a knowledge feedback loop that informs policy and strategy development further downstream. Moreover, the Government Gazette of municipal transfers currently delineates the processes and output or intermediate outcome measures which are supposed to be collected for each grant. Transferring national departments are responsible to collect such data, however, it is widely acknowledged that detailed monitoring of outputs and impact will require significant investments over time. Various institutions in South Africa (such as the Development Bank of South Africa) have also developed methodologies for rating financial management capacity, outputs, and other indicators of municipal performance. In many cases, these have been used to build various databases that rate or benchmark local government performance in different ways. The reviews undertaken as part of this AAA concluded that the DORA grant monitoring system would need to more systematically exploit its potential for monitoring municipal institutional capacity, service delivery outputs, and impact. 69. Several exercises are underway across the country that can provide useful lessons on the type of measurement instruments that would be required to measure the above- mentioned performance factors. Most of these exercises have been designed to survey and benchmark public management processes and service level performance in their various dimensions at the municipal level (for example, administration, finance, human resources, responsiveness, infrastructure and social service provision). These include surveys and rating systems established variously under the Planning and Implementation Management Support System (PIMSS), the Financial Management Grant (FMG) database exercise (supported through this AAA and the Municipal Financial Management Technical Assistance Project), DBSA's municipal project appraisal system, and other departmental monitoring efforts. 70. The growing importance of household surveys as a tool for measuring the impact of municipal grant and other government programs is noteworthy. Household surveys have grown in importance globally and are now firmly placed in many countries as the backbone around which national statistical systems are constructed and the empirical framework upon which policy formulation is achieved. The growth in the utility of household surveys has also ensured that many developing and transition countries have made appropriate budgetary allocations and program space for the design and execution of such surveys and the analysis and application of obtained results. South Africa itself has an established tradition of undertaking household surveys. Box 2 provides a more thorough stocktaking of the household impact assessment landscape in South Africa. 28 Box 2. Municipal Management And Household Level Impact Evaluation--Ongoing Efforts In preparing for the establishment of new municipalities, the Department of Provincial and Local Government (DPLG) commissioned an independent investigation into the status of monitoring and evaluation mechanisms of service delivery and anti-poverty programs in 2000. The research spanned 15 departments in 14 ministries, and focused on those departments responsible for delivery of basic services. It highlighted government-wide deficiencies in M&E and stressed the need to promote the status and role of local government in M&E, the urgency for capacity development, and the need for well-defined impact evaluation indicators as well as their possible inclusion of the indicator set in the October Household Survey (now replaced by the SA Community Survey). South African Cities Network (SACN) has been providing support to the nine member cities on the development of indicators for impact evaluation. To this end, an Urban Indicators Reference group (now called the Well Governed Cities Reference Group) was tasked with developing a common set of household level indicators for all nine major cities. This rationalization was required to ensure a proper informational base for comparative analysis across the cities. The current work of the group includes a completed detailed situational analysis of the use and application of information and indicators by the nine cities, the rationalization of the surveys currently undertaken by individual cities (five conduct regular surveys which range from quality of life surveys to customer satisfaction surveys), and the establishment of common standards for quality management of data collection, storage, and use. The Financial and Fiscal Commission (FFC) is a constitutional Chapter 9 institution established to provide oversight over government financial and fiscal processes and procedures. In the first term of democracy, the FFC focused principally on the alignment of the national and provincial budgets to the goals of the constitution, specifically with regard to the delivery of basic services. With the establishment of the new municipalities, the FFC has added emphasis on investigating and overseeing fiscal transfers between National Treasury and the local government sphere. The FFC has a register of constitutionally mandated basic services (CMBS) and is currently developing a set of indicators--to be assessed at the household level in local jurisdictions--to evaluate the progressive realization of CMBS targets by municipalities. The FFC intends to employ these indicators to make recommendations to National Treasury via DORA on the formulae employed for allocation of the Equitable Share and other conditional grants. The Development Bank of Southern Africa (DBSA) has an established M&E system that is focused on its projects. The DBSA recently expanded its focus with the establishment of a dedicated Monitoring and Evaluation Unit that expand its scope to include impact evaluation. The unit intends developing a set of impact evaluation indicators that will provide the Bank with an improved informational base for decision making in respect of future grant funding and loan provision to municipalities. To this end, it is interacting with the SACN and has made a budgetary allocation to assist municipalities develop the capacity and resources required for the successful application of these indicators in local planning and management. StatisticsSA has engaged with a range of agencies and institutions in developing a set of impact evaluation indicators that will form the backbone of the imminent SA Community Survey. Prior to 2004, StatisticsSA was charged with the task of undertaking the census, the Labour Force Survey and the October Household Survey (OHS). Following cabinet acceptance of the SA Council of Statistics recommendation in 2004, the census will only be undertaken every ten years in accordance with global practice, and the OHS was replaced by the SA Community Survey, a household survey which will provide data representation at municipal level for program planning and policy formulation. The SA Community Survey, to be piloted in 2006 and launched in 2007, aims to correct the previous problem of an abundance of national and provincial data will little resolution at the local government sphere. The proposed scope of the survey is considerably larger than any existing household survey in the country, in the order of 170 000 dwelling units nationally drawn from 17 000 primary sampling units, with a probability-proportionate-to-size (PPS) sampling frame to capture the spread of the various category municipalities. In conjunction with the Well Governed Cities Reference group of the SACN, StatsSA is developing a proposal for the provision of capacity to the nine cities for the rollout of the household impact evaluation survey. Various models for the rationalization of the city surveys with the Community Survey are available. 29 The Pilot Household Impact Assessment Survey 71. An array of measurement instruments are required to appropriately measure factors that affect the performance of the municipal grant system and the municipal system as a whole. As part of this AAA, one such instrument--a household impact assessment survey (HIAS)--was piloted to test various methods for measuring the impact of transfers on service levels as well as income and expenditure patterns. Specifically, the pilot sought to establish baselines against which the impact of social safety net and municipal grants could be measured. The pilot attempted to provide measures of the objective needs as well as the subjective expectations of the local communities who are the recipients of municipal service delivery, and over time, trend analyses of how these needs and expectations are changing as a result of municipal performance. The use of such indicators to assess the impact of transfers is already evident in the country (for example, in the measurement of various service and infrastructure backlogs by the three spheres of government). However, backlog measures, which are based on the census methodology and data, are both limited in application and proscribed by other variables typically not addressed (for example, household capacity to pay for services). Hence the development of indicators to measure the actual impact of government services on alleviating poverty at household level forms an integral aspect of the survey. The indicators employed for the household impact assessment study necessarily encompass both objective needs and subjective assessments and expectations. The full listing of indicators and questionnaire employed in the pilot HIAS appear as Annex 5 and Annex 6 to this report. 72. In conducting the pilot, priority was also attached to the nature of the sample employed for the pilot, as the efficacy of the measurement instrument is related to, and dependent on, the nature of the sample. For this reason, it was decided to employ a pilot sample that would reflect characteristics of a future master sample. These features included the urban/rural split in human settlement patterns, the lower echelons of household income and asset ownership, and linguistic and race considerations.15 Given the logistical, resource, and time constraints, the above-mentioned sampling criteria provided the most appropriate method for ensuring structural equivalence between the pilot and the future master sample. This equivalence was pivotal to ensuring proper validation of the survey. 73. Caveat About Sampling. In adopting this sampling approach is it important to stress the following caveat: the districts sampled are not representative of the country as a whole. The purpose of the survey was to demonstrate and test various methods rather than to draw broad conclusions that could inform policy decisions. For the latter purpose, a large-scale future master sample would be required, and it should be selected in a manner that is representative of the country's diverse jurisdictions and households. 15Verification of incomes for the poorest of the poor is replete with problems. In the South African context, residential location and similar factors appear to be strong predictors of poverty. 30 74. Accordingly, the two municipalities surveyed were selected in a manner that would capture the key features of a master sample. One was an urban metropolitan municipality intended to be indicative of the nine major urban municipalities in the country, which collectively account for almost half of the country's population. The second was a district council intended to be indicative of the majority of municipalities in the peri-urban and rural areas of the country. The district councils constitute the majority of the non-metropolitan municipalities in the country. Within each of these municipalities, residential areas were selected on the basis of income distribution, employment, history of service delivery, etc. Language and racial group membership was also employed as stratification criteria so as to capture the enduring effects of apartheid urban planning. The sub-samples chosen for the pilot were thus urban and rural African and Coloured households with household income levels at an average of R2,500. In addition to these factors, the nature of the settlement was also considered. Hence, the sample featured households from established residential suburbs to informal settlements. Table 6 summarizes features of the households sampled in the pilot. Table 6: Indicative Overview Of The Pilot Sample Features Urban Rural HOUSEHOLD HEAD Males 60% 72% Females 40% 28% Average age 55.5 yrs 47 yrs Employed at time of survey 35% 55% HOUSEHOLD MEMBERS Number of household members 5.6 4.0 Number of adults 3.7 2.8 · Number of male adults 1.7 1.3 · Number of female adults 2.0 1.5 Number of children 1.9 1.2 · Number of male children 1.0 0.6 · Number of female children 0.9 0.6 Sleeping density per room per household 2.4 2.7 HOUSEHOLD INCOME AND EXPENDITURE Monthly household income R2 658-91 R1 643-42 Monthly household expenses R1 960-05 R1 100-27 % of income from salaries and earnings 78.2% 79.0% % of income from government grants and pensions 17% 14.8% % monthly spend on food and groceries 41.6% 55.6% % monthly spend on municipal services 13.1% 14.3% Implications For Service Delivery--Performance And Perception 75. The pilot sought to collect data on the availability, accessibility and affordability of the full range of services provided to households by municipalities as well those provided by provincial and national governments. The survey placed considerable emphasis on detailing the status of service connection and accounts for basic services such as water and electricity, primarily to investigate the impact of the recently 31 implemented policy of provision of a specified quantum of free water (6Kl) and electricity (50Kwh) to each household. As it is rolled out, a much larger sample of households is required to generate necessary data on actual service levels including access, efficiency, and satisfaction. 76. Particularly worthy of note is the pilot's attempt to also collect subjective data on household preferences and perceptions in addition to basic data on actual service levels at the municipal and household levels. Subjective household assessments of various services were compiled to provide a measure of the net gap between household expectations and satisfaction with received services both at household level and within the local community. Annex 7 reports on the service delivery assessment of the survey. 77. In addition to identifying the basket of services households receive, the pilot also sought to determine how people rate the importance of individual services to their household. The rationale for this approach is that such rankings may be useful in assisting municipalities in sequencing the roll-out of specific services within its jurisdiction. Table 7 details the rated importance of fourteen services by urban and rural households. Not unexpectedly, the basic services such as utilities, social services and safety and security services feature prominently at the top of the ranking, while less essential services such as recreational facilities and services feature at the bottom end. While there are no statistically significant differences in terms of how rural and urban householders rate and rank the importance of the fourteen services, it is important to consider these rankings carefully. Specifically, the rating and ranking of individual services must be viewed in light of the services that are already available to households; they may also be driven by price differentials and require more detailed analysis. 78. There are some preliminary issues worth highlighting based on these pilot household assessments. Pilot households do prioritize certain services over others. Highest priority is assigned to basic municipal services followed closely by social Table 7: Rated Importance Of Individual Services In Urban And Rural Households Services Urban Rural Rating Ranking Rating Ranking Water 4.74 1 4.86 1 Health 4.73 2 4.79 4 Sanitation 4.68 4 4.83 2 Education 4.69 3 4.76 6 Electricity 4.64 6 4.80 3 Safety And Security (Police) 4.65 5 4.76 7 Refuse Removal 4.61 8 4.78 5 Social Welfare 4.62 7 4.72 8 Street Lighting 4.55 10 4.70 9 Library Services 4.60 9 4.54 10 Roads 4.45 11 4.53 11 Community Facilities 4.36 12 4.43 12 Parks And Gardens 4.35 13 4.42 13 Telephone (Landline) 4.12 14 4.25 14 32 services, then safety and security and finally recreational facilities and services. Certain services are clearly the highest priority for all households, regardless of location. These households are also generally in agreement on the de-prioritization of landline telephone services. This can only be explained by the advent of much cheaper and more accessible cellular telephony options, though it is likely that poor service and high tariffs by Telkom also played a significant part in the assignment of the lowest rating to this service. Households appear to want the rollout delays in basic services resolved first before additional (non-essential) services are addressed. 79. The differential ratings for basic and social services across urban and rural households can be explained by the existing rate of rollout of these services in these areas. Rural areas lag behind in terms of basic services such as piped water and refuse removal; hence they tended to assign these a higher priority than urban households. Similarly, rural households assigned higher ranking to electricity, while urban households, who have on average had the service considerably longer than rural households, do not provide a similar rating and ranking to this service. Water is consistently rated the most important service by all households, indicating that households are well aware of the health and other risks associated with the use and consumption of contaminated water. Health services are rated highly by both rural and urban households, while urban households rate education much higher than rural households. The diminished emphasis on education in rural areas was borne out by analysis, which revealed considerably fewer years of school education in rural adults than in urban adults. 80. Notwithstanding the potential value-added of this approach to ranking services, Box 3 provides a cautionary note that suggests the need for objective measures. Box 3: Do Client Scorecards And Ratings Convey Meaningful Information? Indicators such as citizen's scorecards and satisfaction ratings are useful to monitor performance of public service delivery. However, the utility of these ratings depends on the extent to which they reflect conditions on the ground. They are of limited use if the scores or ratings only reflect household idiosyncratic characteristics. To illustrate this point, the team sought to examine the correlates of satisfaction with water supply. Water supply was chosen since all households consume water (from various sources) and have ranked it as the most important public service. Correlates of being satisfied (defined as being very satisfied or satisfied) were estimated with water services. The team found that the probability of being satisfied decreased with the frequency of service disruption and whether household members had to fetch and carry water to the house each day. This suggests that accessibility and service quality have a strong bearing on satisfaction scores. Having established this link between objective service quality measures and satisfaction scores, the team can use the subjective scores to discuss performance of service provision. Given the location specific attributes and household characteristics also influence satisfaction scores, the team controlled for these characteristics in the estimations. However, no statistically significant effect of these characteristics on satisfaction scores could be established. 33 Implications For Household Income And Expenditure Patterns 81. Equally important is the need to assess household income and expenditure patterns, and the impact of various grant programs on the same. As discussed in the introduction of this report, this pilot did not attempt to resolve the conflicting definitions between government agencies of the income level utilized to determine poverty at household level. Rather, and given its principal focus on government services and programs, it addressed the issue of poverty by examining the impact of these services and programs on the income and expenditure of households.16 Specifically, the analysis of household income and expenditure focused on the following salient characteristics of public policy in South Africa: Impact of government grants and services on household income, that is, the net proportion of household income constituted by such grants and services; Proportion of household expenditure used to access basic services (for example, utilities); Proportion of household expenditure used to access government services such as education and health; Level of household debt, and proportion of expenditure used to service this debt; Level of household savings, and proportion of income invested in such savings; Expenditure on consumables and transportation. 82. Household Income. The monthly household income across the six residential areas sampled in the pilot survey is displayed in Table 8. Table 8: Mean Household Income Across Districts (in Rands) Urban Districts Rural Districts Urban 1 Urban 2 Urban 3 Urban 4 Rural 1 Rural 2 R 2 422 R 3 581 R 2 054 R 2 540 R 1 152 R 2 160 83. The differences in household income across all six residential areas is statistically significant at p=0.00. Dividing the sample into urban and rural households reveals an average for urban households of R2,658 and an average for rural households of R1,643, a difference of R1,015 which is also statistically significant at the p=0.00 level. 84. Rural household incomes are not only significantly lower than those of urban households, but the range of incomes for urban households is consistently greater than that of rural households. This implies that the spread of income is considerably greater in urban households, whereas with rural households, and barring a few outliers, the majority of households are tightly clustered towards the lower end of the income distribution. In other words, while urban incomes are on average higher than those of rural households, the level of disparity is also higher amongst urban households than it is amongst rural households, and this may indicate a need for urban local authorities to increasingly refine targeting of subsidies and benefits. 16Concerns about estimating household incomes, noted earlier, remain valid. The pilot, however, sought to demonstrate how such impact analysis could be undertaken given the quality of incomes data. 34 85. The total monthly household income for the entire sample is constituted from the following sources as indicated in the Table 9 below. Overall, 78.5% of all income is constituted by earnings and salaries for temporary and permanent employment in the formal or informal sector with an additional 4.5% derived from private pensions. Government grants and assistance contribute 16.3% to average household income. Table 9: Sources Of Monthly Income As Percentage Of Total Income State Work Child Foster Salary/Earnings Pension Pension Disability Support Care Transfers 78.5% 10.9% 4.5% 3.1% 1.8% 0.5% 0.7% 86. The income sources data raises a number of interesting considerations. At a policy level, the mix of own income and government benefits targeted to the poor can be interrogated, e.g. is the average 16% composition of total household income deriving from government benefits in line with policy objectives? In other words, are poor households being targeted efficiently, and is the aggregate amount of grant transfers reaching the poor households at a level that conforms to policy objectives and is adequate to meet the basic needs of these households? As can be seen, there is little to differentiate urban from rural households in terms of the sources of monthly household income. Interestingly, despite recording high levels of unemployment amongst adults (33% for the total sample), none of the households reported receiving unemployment benefits. Likewise, no households report receiving care dependency grants. A second level of interrogation can be pursued relative to the accessibility of these grants, and the cognizance of poor households of the benefits available to them. Finally, the role of municipal governments in connecting poor households to the available resources could also be the subject of further investigation. 87. To understand how households from different districts differ in terms of their success at reconciling the monthly books, the mean difference between monthly household income and expenditure for each district was assessed. However, it may be argued that the mean difference between income and expenditure is not particularly meaningful if the household income and expenditure differs significantly across different households, which it does across the residential districts in this sample. To compensate for the differential incomes and expenditure across the six areas, the mean difference may be corrected and expressed as a percentage of the income itself. The uncorrected and corrected mean differences are displayed in the Table 10. Table 10: Uncorrected And Corrected Mean Differences In Income And Expenditure Urban Rural U1 U2 U3 U4 R1 R2 Uncorrected R 279 R 1 003 R 594 R 926 R 560 R 526 Corrected 2.5% 26.9% 23.7% 23.1% 39.5% 20.8% 88. As Table 10 indicates, when compared to the absolute savings, the corrected mean difference changes the picture quite considerably. Households in R1 have the highest savings-to-income ratio, followed by a close grouping of households from U2, U3, R2 and U4, with households in U1 trailing far behind. In essence, R1 households are able to 35 realize an average monthly surplus equivalent to two fifths of their income, while in U1 households are only able to manage a surplus equivalent to one-fortieth of their income. Again, interesting considerations arise. Residents of the R1 informal settlement have a relatively high income surplus but more limited access to services, given the informal nature of the settlement. The relatively high surplus in R1 and, to a lesser degree, all of the other communities except U1, provide indications of the ability and potential willingness to pay for more or better quality public services and/or capacity to service debt for housing loans (in the pilot sample, an average of only 4.5% of household expenditure went to bond repayments). 89. Composition of Household Expenditure. Total household expenditure is examined by deconstructing total expenditure into its various constituent components. To simplify the representation, recurrent and once-off costs were grouped as follows: Food and Groceries, as measured in the survey Homestead Costs, comprising payments for rentals, bond repayments and rates. Municipal Services, comprising charges for electricity, water and refuse removal. Telephony Services, comprising costs for landline and cellular telephones Transport, comprising expenses for public and private transport Social Services, comprising expenses for education and health Home and Personal Expenses, comprising expenses on clothing and furniture Sundry, comprising all other expenses 90. This results in the urban and rural household cost breakdown in Table 11. Rural and urban households do differ considerably in terms of the allocation of total household expenditure to specific grouped costs. The most pronounced differences are evident in the fact that rural households spend significantly more than urban households on food and groceries, home and personal expenses and sundry items, while the reverse is true for homestead costs, transport and telephone. Furthermore, not only are municipal services the second biggest grouped cost item after food and groceries, but of interest is the fact that the average percentage of expenditures devoted to municipal services is only slightly smaller than the average proportion of income deriving from government benefits, 13.5% versus about 16% respectively. Table 11: Grouped Costs For Urban And Rural Households Grouped Costs Urban Rural Food and Groceries 41.6% 55.6% Homestead 12.6% 2.8% Transport 9.7% 3.0% Telephone 7.0% 4.6% Municipal Services 13.2% 14.3% Social Services 7.8% 6.3% Home/Personal 4.2% 7.9% Sundry 3.8% 5.7% 36 91. Turning to households already receiving basic services (see Table12), we find that while urban and rural households differ in payments for these three services in real terms, in terms of these costs as a portion of household expenditure the only difference is that rural households pay significantly more for water and refuse removal, while the cost for electricity is about equivalent. Table 12: Costs For Households Receiving Services Services Urban Rural Electricity 8.23% 7.82% Water 4.25% 7.42% Refuse Removal 0.62% 1.28% 92. This result is significant in the context of how water and electricity prices are structured in different municipalities. For example, the electricity prices are regulated by the National Electricity Regulator for municipalities (who then purchase in bulk and sell to households). The MFMA contains provisions that regulate the process and consultation on bulk pricing more generally, attempting to strengthen the oversight role of government. There are, however, implicit hazard in that poorly functioning municipalities may hide the costs of broader mismanagement in inflated prices. Measuring Municipal Processes And Institutions 93. As noted earlier, there is growing recognition among policy and academic researchers that the impact of government programs depends in part on the quality of public management or institutional processes that influence program implementation. Also worthy of note are the implications of the actual institutional structures and processes as well as perceptions by citizens of these processes for service quality as well as household behavior. 94. To determine how householders view their municipality, the pilot sought to assess municipal operations and processes, and then establish performance evaluations provided by households of various aspects of these operations and processes. The assessments are based on the subjective evaluations of the householders. While such epiphenomenal data is not by itself an objective indication of how well a municipality is performing, such surveys do provide valuable information on how households perceive their municipality's performance. As such , the assessment captures in part how effectively municipalities and constituent households are able to manage their relationship to mutual benefit. 95. Accordingly, the pilot survey assessed five aspects of the municipality-household relationship. These are the household's (i) assessed level of trust or confidence in the municipality's ability to manage and govern; (ii) perceptions of the municipality's responsiveness to their needs; (iii) assessed level of satisfaction with the quality and efficiency of communication processes or information sharing by the municipality; (iv) assessed level of satisfaction with the inclusiveness of municipal consultations with citizens; and (v) perceptions of the level of corruption involving municipal administrators and elected officials. These variables were empirically assessed using five indexes or scales, each comprising a cluster of individual questionnaire items. 37 96. Various attempts at correlation and multiple regression analyses were undertaken between these above mentioned institutional quality factors and service levels and other household level outcomes. The team concluded that composite perceptual indicators such as those ones noted above were highly correlated with each other. In addition, the team concluded that the indicators were not specific to the actual public processes (such as planning, inspection, accountability, co-production) through which households and their municipal officials influence output performance. Such surveys of households (as well as possible surveys of public officials and service providers) should query respondents on their actual involvement or knowledge of specific processes such as the IDP and budget preparation process, school management committees, the publication of municipal budgets, or the last time they were asked to respond to a client survey when they visited a health clinic. 97. Ratings of these processes along with objective process indicators can provide the basis for more robust assessments of institutional quality, and therefore more effective correlation and even regression analysis to better understand the institutional determinants, if any, of changes in service level performance, or household income and expenditure patterns. As such, more intensive development of process benchmarks would enrich the perception-based assessments. The past decade has seen a considerable development in methods for assessing the quality of institutional processes. Instruments include public officials' surveys, facility level surveys, public expenditure tracking surveys, and client scorecards. These will need to be more fully explored as the M&E architecture underpinning municipal grants in South Africa evolves. The Way Forward 98. In implementing a medium term municipal monitoring strategy, NT and partnering departments should consider investing more systematically in three instruments discussed in this section (and summarized below). Household level surveys of trends in incomes and expenditures, responsiveness of service delivery, and the overall credibility of public processes. When rolled out, these surveys should build on the lessons learned from the NT pilot. The scale and sampling of these surveys should ensure that it is representative. Data analysis should consider the price effects associated with transfer programs the finance social assistance and basic services. Facility level surveys or assessments of providers including objective and subjective assessments of institutional constraints on provision. While these were not tested or piloted during the AAA work, the team concluded that considerably more information is required not just about households and local governments, but also the range of providers involved in the delivery of basic services, as well as their relationship with citizens. Based on existing efforts within SA as well as the international experience, NT and other national departments should consider piloting facility or provider surveys, particularly of services financed under MIG. Surveys would assess financial and personnel, management, and M&E processes undertaken by providers and factors affecting their relationship with clients. 38 Municipality level surveys or assessments including objective assessments of public management processes such as planning and budgeting and M&E has increasingly been acknowledged as important in the implementation process, as are the perceptions of public officials themselves. When measured and benchmarked, policymakers can determine whether (and which aspects of) municipal management practices influence service quality as well as behavior at the household level, and perhaps, wellbeing. 99. Taken together, this array of instruments provides the necessary tools for broadening the current focus on financial monitoring within municipal grants to municipal monitoring more broadly. NT's decision to roll out the HIAS pilot after further refinement of the survey instrument is an important step towards moving grant monitoring beyond inputs, and moving beyond grant monitoring. 100. Equally important is the need to ensure that information generated from these surveys is used at the local level in a manner that is accessible to local citizens in order to irrigate the process of democratic accountability in municipalities themselves. In addition to developing robust technical capacity to conduct process, output, and impact assessment, it will be important for SA's municipalities and relevant national departments to develop feedback mechanisms (for example, through local government benchmarking or public reporting) that enable local officials, providers, and citizens, as well as interested parties at the national level to learn-by-doing. SECTION VI. MUNICIPAL MONITORING AND SOUTH AFRICA'S GOVERNANCE FRONTIER 101. This report summarizes the issues identified and progress made over the course of this AAA work on municipal grant monitoring in South Africa. Several propositions related to grant and municipal monitoring were highlighted over the course of this AAA. 102. DORA monitoring is evolving in step with the ongoing development and innovation of SA's intergovernmental fiscal system. The development of SA's grant monitoring is being progressively phased in order to "fit" reporting requirements and data demands to varied municipal capacities. Near term improvements in the quality of fiscal and financial data will provide the basis for greater investment in data on service levels and household income and expenditure patterns. Yet, monitoring must focus on more than just the data. A range of new methods and processes should be tested to making information on municipal performance more readily available to citizens, administrators, and elected officials. Such processes should also encourage regular feedback to foster learning by doing. 103. Taken together, these propositions were confirmed by the AAA work and form the basis for NT's municipal monitoring strategy. As South Africa enters its second decade of reform, municipal monitoring will likely weigh more heavily in searching for and expanding the country's "governance frontier." 39 Seeking The Governance Frontier 104. The AAA work suggested three potential frontier issues--(i) ensuring efficiency and development impact of large consolidated grant programs; (ii) enriching local democratic and political processes with evidence-based information on municipal performance; and (iii) encouraging local economic and spatial development of newly amalgamated municipalities. 105. Ensuring The Efficiency And Development Impact Of Consolidated Grant Programs. South Africa' s Medium Term Expenditure Framework and related financial regulations have been widely acknowledged as good practice examples of measures that promote the transparency and accountability of public resource management. Another set of principles underpinning the MTEF is the need to focus on improving the efficiency and effectiveness of expenditure policy across all three spheres of government, including the municipal grant programs. In light of the ongoing efforts to consolidate municipal grants and rationalize the overall number of transfers to local government, the AAA team anticipates increased pressure to evaluate the allocative and operational efficiency of public spending financing through the grant programs such as MIG and Equitable Share, as well as demand for more evidence-based analysis of their effectiveness in achieving stated objectives and desired outcomes at the household level. 106. From a governance perspective, municipal monitoring serves as an important tool in helping national, provincial, and local governments, civil society, and citizens at large to evaluate whether grant consolidation is working over the medium term. Moreover, the monitoring system may identify various critical success factors such as municipal capacities that would improve the likely impact of different mixes of fiscal instruments (for example, conditional and unconditional grants). Finally, the monitoring system, as described in this report, should serve as an important tool for assessing the effectiveness of different types of conditionality within the fiscal systems including the balance between sectoral specific and multi-sectoral or budget management conditions, equity oriented and performance oriented rules of access, etc. 107. Deepening Democratic Accountability. Democratic accountability in a decentralized setting such as South Africa will be fully realized when key participants in the system play their roles effectively. Three key stakeholder groups in South Africa-- citizen voters, appointed officials, and elected officials in all three spheres--have collective experience of only one decade of democratic governance, and only five years of local self-government. Thus, democratic accountability at the local level remains a work-in-progress. A critical success factor is the ability of stakeholders to have access to performance information, process it, and then utilize it to improve their local conditions within the framework of local democratic decision-making. 108. Given the limited experience of all stakeholders, it is unlikely that the availability of basic performance monitoring information will play a decisive role in the 2005/6 local government elections. It is reasonable, however, to target the 2010/11 local government elections as the date by which all three key stakeholder groups will have acquired ability to use performance information as a basis for evaluation and informed decision-making. This will require at least three types of capacity development initiatives. First, there is a 40 need to develop the necessary mechanisms for information sharing and feedback such as public reporting, benchmarking, focus groups, client score cards, and electronic fora to translate M&E data into actionable information that citizens can use to inform themselves. Second, local officials and citizens will need to be better educated about the types of work they will need to do. Third, experience in other countries suggests that media and civil society will also need to be prepared, and suitably trained, to interpret and translate the evaluations of major programs and cross-jurisdictional benchmarks in a more accessible manner. 109. Promoting A Favorable Investment Climate And Spatial Development. As the municipal monitoring system develops, it is envisaged that it would go beyond simply grant monitoring and begin to benchmark performance in areas relevant not only to publicly financed programs but also to the local investment climate. If measured correctly and credibly, such benchmarks could produce healthy competition between jurisdictions to improve performance in key areas. It could also provide the basis for rewarding performance either by directing additional public resources or attracting private investment. 110. In this regard, SA's National Spatial Development Perspective (NSDP) enunciates a policy that not only provides for the achievement of minimum standards for all jurisdictions, but also targets additional public investment to accelerate sustainable economic growth. Robust, credible benchmarks of municipal management and performance could be vital in ensuring that NSDP-inspired investments by parastatals as well as provincial and national agencies (for example, in the form of industrial development zones, ports and airports, development corridors, etc.) are channeled to jurisdictions that perform and to purposes that provide the greatest return. In other words, municipal monitoring has the potential of resolving the tensions between public investment by non-local entities and municipal management in fast-growing jurisdictions. 41 ANNEX 1: A BRIEF HISTORY OF SOUTH AFRICA'S INTERGOVERNMENTAL SYSTEM SINCE 1992 1. Significant debate as to the eventual structure of governance in the "new" South Africa occurred in the early `90s, and in 1992 a clear position emerged on provincial17 government to be taken into negotiations. This was preceded by the compilation of a discussion document used for consultation amongst African National Congress (ANC) structures prior to the conference. The document, entitled ANC Regional Policy,18 was compiled under the auspices of the ANC's Constitutional Committee and Department of Local and Regional Government and Housing, and based on a number of meetings and workshops. Some of the team involved in the drafting were linked to local civic struggles and, subsequently, the provision of technical support to the `non-statutory' side in LGNF discussions. Thus the debate was influenced by a strong local perspective. Much of the substance of the discussion document became the basis of the ANC negotiating position, and subsequently found its way into the interim and final constitutions. A number of points had particularly important effect on the system that emerged after 1994: 2. The Regional Policy document argued that there was general consensus that a new constitution should provide for three tiers of government, but it sought to establish a focus on co-operation and linkages between the tiers, and an end to the ethnic balkanization of the country. These concepts eventually came to form a key feature of Chapter 3 of the final constitution, entitled Co-operative Government, following immediately after the Founding Provisions and the Chapter on the Bill of Rights. It also placed the notion of metropolitan governments firmly on the agenda, stating that: "The ANC views the creation of metropolitan governments in certain parts of the country as essential to the cause of unifying, de-racializing and democratizing cities, in addition to the more efficient and effective provision of affordable services."19 This was a reflection of calls for "One City One Tax Base" as well as the desire to exercise coherent authority across functional urban areas. 3. There was significant debate as to whether metropolitan governments should be regions (or provinces), in their own right, having a direct relationship with the national level, or whether they should fall under the jurisdiction of the second tier. The view of the localists was that even if there were to be compromise on the creation of provinces, the formation of metropolitan governments as part of the second tier would enable a direct relationship with the national level. This was linked to another view amongst the localists, that it was preferable to have a larger number of smaller regions, with the existing Regional Services Councils forming the basis of the second tier rather than the old provinces. This position had insufficient support to be included in the discussion 17The federalists used the term `states' to describe the proposed second tier governments. The ANC used the term `region' since it saw this tier as much weaker. Eventually the term `provinces' was used as a compromise. Many of the documents of the time referred to SPR's which was short-hand for States/Provinces/Regions! 18ANC Constitutional Committee and Department of Local and Regional Government and Housing, ANC Regional Policy; draft discussion document published by the Centre for Development Studies, University of the Western Cape, October 1992. 19Ibid., p. 7. 42 document but did subsequently re-emerge some years later in the debate around district municipalities. 4. A section on Finance and Resources laid out significant elements of the basis for the inter-governmental fiscal system that eventually emerged through the negotiating process. It was written against the backdrop of attempts by federalists to argue for the devolution of taxes to the provincial tier, seeking to counter this, while at the same time creating the basis for fiscal decentralization to the local level. It argued that a new system "must be appropriate to modern economic conditions, seeking to enhance democratic accountability while ensuring that the public resources of the country are shared fairly amongst the whole population....The starting point should be a strong emphasis upon the need to strengthen local control over the use of public resources. This helps to ensure that usage is efficiently and appropriately tailored to local conditions. The link between paying taxes and receiving public services must be recognized as an important element in the strengthening of democratic accountability, and is most direct at local level."20 5. Notwithstanding the benefits of fiscal decentralization, the policy also was cognizant of the constraints. It was careful to highlight the importance of not compromising the capacity for the central authorities to exercise sound management over the economy as a whole, or, in a country of profound regional inequalities, the importance of not compromising the capacity for redistribution. Furthermore, there were significant technical constraints on the devolution of taxes, most of which were difficult to administer at sub-national level. The conclusion was that while some taxes could be retained at local level, especially property taxes, the bulk of taxes should be collected nationally and, to the extent that expenditure took place at lower levels, should be distributed by fair and transparent means through fiscal transfers. 6. The ANC document also delineated the need for transfers to be formula driven, and to be comprised of both conditional and unconditional tranches. "Where regional and local government is given responsibility for implementation of national policies, transfers will have conditions attached to ensure that national policies are indeed adhered to in implementation. Thus, within clearly defined nationally determined parameters, decentralized governments would be able to fine-tune the actual pattern of expenditure to suit local needs. The majority of transfers are likely to fall into this category. Other transfers, however, would have far less stringent conditions attached, and would be aimed at enabling lower tiers of government to implement policies in areas where the constitution gives them power to act autonomously. These grants would have to take into account the capacity of various lower level governments to raise their own resources in order for inequalities amongst regions and localities could be counteracted."21 7. The document argued it would be unwise to leave control over the transfers entirely to central government: "On the other hand, trying to fix in the constitution the detail of how transfers are to be made would tend to be either too vague or too rigid, or both." Instead it argued that the constitution contain a set of guidelines to govern 20Ibid., p. 9. 21Ibid., p. 11. 43 transfers while at the same time, a "permanent Advisory Fiscal Commission structured on a non party-political basis' be created in which certain powers for advising on the structure and mechanism of fiscal decentralization would be vested". The guidelines on transfers in the constitution were to be "consistent with the extent of political autonomy decentralized government (was) to have, and with the Bill of Rights ... Such guidelines should ensure that transfers are made in such a way that lower levels of government are able to plan properly; that they are structured so as to enhance efficiency and local accountability and that they are open to clear and effective monitoring. The guidelines must seek to redress inequalities between regions."22 It also argued that the Fiscal Commission should have a say in granting powers of taxation to lower levels of government within this overall framework, thereby indicating the need to adapt to the system of tax devolution over time. 8. Many of these concepts became embedded four years later in the final constitution, and therefore, provided the basis for the current system of inter- governmental fiscal relations. As part of its election campaign and process of preparing to govern, the African National Congress drew up the Reconstruction and Development Program, popularly known as the RDP. It was a detailed document,23 setting out a number of targets, including in the area of housing and municipal service provision. With regard to housing it sought to construct a million low-cost houses over five years. With respect to municipal services there was a short term aim of "establishing a national water and sanitation program which aim(ed) to provide all households with a clean, safe supply of 20-30 liters per capita per day within 200 meters, an adequate/safe sanitation facility per site, and a refuse removal system to all urban households."24 Regarding electrification it stated that "An accelerated and sustainable electrification program must provide access to electricity for an additional 2,5 million households by the year 2000, thereby increasing the level of access to electricity to about 72 percent of all households (double the present number)."25 9. After coming to power the new government created a Ministry in the Office of the President for the implementation of the Reconstruction and Development Program, and issued a RDP White Paper that intimated an important role for local government in development. The RDP Office was abandoned in 1996 and its programs integrated into other mainstream line functions. In 1995, the Housing Department together with the RDP ministry drew up a Municipal Infrastructure Investment Framework (MIIF), which laid out financial mechanisms and an institutional approach for overcoming service delivery backlogs. The MIIF argued for a ten year--rather than five year--program to overcome backlogs. The Framework spelled out clearly that, in terms of the Interim Constitution, `local government (was) responsible for ensuring the delivery of services; the preparation of infrastructure investment programs; and the selection, preparation and implementation of infrastructure projects in consultation with communities...The choice of service levels (was) a local decision, subject to local affordability and national and provincial 22Ibid., p. 11. 23ANC. The Reconstruction and Development Programme: A Policy Framework. 1994. 24Ibid., p 29. 25Ibid., p. 33. 44 guidelines.' Furthermore, it stressed that `responsibility for project selection and capital investment must not be separated from responsibility for recurrent costs' (emphasis in the original text). 10. The MIIF stated that the program would require significant capital grants from central government `in the order of R30-R35 billion'. Municipalities were to finance capital expenditure through a combination of the redirection of existing local resources together with recurrent and capital grants and loans for capital development. The MIIF set out six key elements of a new financial framework for the provision of service infrastructure, including that: Capital expenditure should be financed by borrowing; and current expenditure financed by current income; Redistributive mechanisms should be transparent and addressed through fiscal flows in the context of the emerging system of devolved taxes and grants, rather than by manipulating the financial sector; Capital finance should be accessed by integration into the mainstream private financial system, rather than through special public sector lending institutions; where public sector lending institutions do exist they should support this process. 11. Guided by the MIIF, one of the early programs was the Municipal Infrastructure Program, subsequently combined with other programs and called the Consolidated Municipal Infrastructure Programme (CMIP). CMIP provided funding for bulk and connector infrastructure, while the national housing subsidy provided funds to assist in the provision of service reticulation within new low cost housing developments together with land and top structure. 12. But for a variety of reasons, many to do with remaining uncertainty in the system, the municipal sector has generally not been as successful as was hoped in raising capital from the private financial sector, with capital grant funding from national government playing a larger role than was envisaged. Nevertheless, the above three principles have remained prominent ideals, with redistribution through manipulating the financial system ­ especially through guarantee mechanisms--largely being avoided, and the use of borrowings to finance current expenditure outlawed in terms of the pending Municipal Financial Management Act. Local Government And The `Final' Constitution (1997) 13. The `final' constitution was adopted in 1996 after two years of discussion and consultation, and came into effect in 1997. It largely continued the key thrusts of the interim constitution, although there were some significant changes. Important amongst these was the vision of co-operative government between three spheres ­ rather than the more hierarchical `tiers'. Section 40(1) states that "...government is constituted as national, provincial and local spheres of government, which are distinctive, interdependent and interrelated." 45 14. This meant that municipalities were not seen merely as agencies for, or a subservient extension of higher tiers, but they were constitutionally recognized with a distinct identity and leadership role, functioning in a relationship of "co-operative governance" with the other spheres of government. They were no longer a `provincial competence', although provinces were given a monitoring and supporting role in respect of municipalities. Box 4: Intergovernmental Principles Underpinning Municipal Financial And Fiscal Reforms The new system of government in South Africa implemented after 1994 differs fundamentally from the past system. Given the complexity of the three sphere system, and that these spheres are "distinct, interdependent and interrelated", it was critical to develop some clear principles to guide the relationship between governments in the three spheres. This set of principles is intended to guide the intergovernmental system. Intergovernmental Principles. The intergovernmental system as outlined in the Constitution rests on a coherent set of seven high-level principles. It is important that policy, legislation and practice remain consistent with these principles in order to protect the coherence of the intergovernmental system. Distinctiveness And Accountability Of Governments In Each Sphere: All governments in the three spheres are distinctive and accountable in their own right, with specific powers and responsibilities. This makes each government accountable to its legislature, allows each government to set its own priorities, but such priorities must take into account constitutional and other nationally-legislated obligations and ensure that, being a government instead of an administration, each must take full and complete responsibility for its own actions or inactions. In addition, the system of intergovernmental relations must be fair, and promote good governance. The design of the system must avoid perverse incentives so limiting moral hazard. National (or provincial) government must not provide guarantees to governments in other spheres, nor provide bale-outs to governments in trouble. Co-operative Governance: The constitution entrenches `cooperative governance' which implies that a sphere should not assume any power or function except those conferred by the Constitution, is obliged to cooperate with other spheres and avoid legal proceedings against one another, should respect the geographical, functional and institutional integrity of governments in other spheres, should consult over any legislation affecting other spheres, and should assist in building capacity of other spheres. Good Governance: Accountability is inextricably tied to good governance. At the heart of the principle of good governance is the relationship between councilors and officials, their accountability to the electorate and consumers, and the reporting arrangements within and between spheres aimed at transparency and accountability. The political executive (Minister, MEC, Executive Mayor) should be responsible for outcomes and policy, whilst the municipal manager is responsible for implementation and outputs. To hold the head official responsible for performance on outputs, apart from the top management cadre, the head official should have the power to make appointments (within a framework or targets determined by the political executive). The head official should also be part of the process to appoint the top management cadre, together with the political executive. Increased Access To Services: The Constitution and current government policy prioritize service delivery to all South Africans. This principle is about the challenge to broaden access to services at affordable costs to consumers, design appropriate levels of service to meet customer needs affordably, explore innovative alternative modes of service delivery and leverage in additional resources, notably from the private sector, to acquire capital for investment and, where appropriate, enhance efficiency. (continued): 46 Box 4 (continued): Equitable Vertical And Horizontal Sharing Of Resources: The Constitution puts in place a revenue- sharing system, which takes into account fiscal capacity and the functions assigned to each sphere. This principle revolves around the system's intention that each government has the power to determine its own budgets, with the power to collect revenue and spend funds; that each government is expected to maximize its fiscal effort and collect all revenue due; that each sub-national government must avoid running an operating deficit, and limit its expenditure to the sum of its own revenue collected, together with any equitable share and conditional and unconditional grants it has received; and that the system of intergovernmental grants must be simple and comprehensive. It should consist of the equitable share, and conditional grants for infrastructure, capacity building and restructuring. Since 1994, Government has strongly prioritized redressing socio-economic inequalities, and alleviating poverty. Hence all three spheres have important roles to play in this regard. National government is the sphere primarily responsible for redistribution, because it holds the bulk of the tax base and can best address the unequal distribution of wealth. Thus redistribution is largely effected by means of the equitable share and some conditional grants which jointly assist provinces and local government. Redistribution in any particular sphere must be managed so as to not undermine the fiscal viability of the relevant institution and the economic viability of its area of jurisdiction. A Single Process For Vertical Division: This principle acknowledges that, in determining allocations to each sphere (the vertical division), trade-offs are involved between spheres. For this reason, the budget allocation process (to determine the vertical division) should be a comprehensive process, driven by political priorities, and covering all aspects of local governance and service delivery. The danger of allowing separate and ad hoc requests for funds is that it not only fragments the budget allocation process, but undermines the political process of prioritization. Delivery Efficiency And Effectiveness: This principle ties service delivery through the spheres of government to the Government's macro-economic concerns. Services delivered by the public sector (either directly or through a system of public regulation) are targeted at the delivery of public goods and quasi- monopolistic services, as well as addressing poverty and stimulating economic development. This means that these services and expenditures are directly linked to Government's macro-economic policy objectives. It is therefore critical that functions are performed, services delivered and expenditures managed with the greatest possible level of productive efficiency and effectiveness. In addition, available resources should be geographically and sectorally targeted to meet service delivery priorities, as identified by both government and the users of services themselves. Finally, innovative alternatives, like private sector participation, must be explored to enhance productive efficiencies, and thereby to release resources for investment in core priorities and areas of great need. 47 ANNEX 2: OTHER MONITORING UNDERTAKEN AT NATIONAL LEVEL 1. In 2003-04, there were seven departments involved in monitoring of grants to municipalities. An analysis of these grants by department and value of grant allocation are provided in Table 13. Since many of the key issues are similar, rather than examine each of these grants in detail, the investigative work undertaken focused on the role of the Department of Provincial and Local Government; as well as dealing with the largest Grant (CMIP), this Department also monitors the Equitable Share and operates a municipality monitoring system called the Project Viability System. DEPARTMENT OF PROVINCIAL AND LOCAL GOVERNMENT (DPLG) The Equitable Share 2. As with many of the areas in this strategic study the nature and distribution of the Equitable Share is in a state of flux. There is, however, an upward trend in the size of the Equitable Share and, for the first time in 2003/04, it is detailed in DORA down to the level of each municipality and will be transferred directly to the municipality in quarterly installments from Treasury. DORA in 2003/04 also sets out specific guidelines with regard to monitoring (para. 5 (7)) and also defines what constitutes financial misconduct and how it was to be monitored. 3. The nature of this reporting at the time of this Strategic Study had not yet been defined, but there are likely to be three sets of reports (for SANT, DPLG and Provincial). In reality, as the municipal year starts on 1 July, "free basic services" which is a component of the Equitable Share will only commence on that date and the reporting structure as far as DPLG are concerned would commence for the quarter ending September. 4. The Equitable Share is monitored by DPLG on a spreadsheet system, which has been developed by consultants. Two consultant staff are retained to maintain what is described as a `complicated' spreadsheet model. As with other forms of distribution of funds it would appear that enforcement in terms of reporting is not taken to the ultimate stage. There is conflict between ensuring that municipalities get the funds regularly and on schedule and the discipline of not releasing new funds until reports on previous allocations are received. Despite the legal authority for withholding funds or roll-over, this is seldom used since priority is given to distribution in the cash year, not enforcement of the conditions for allocation. Grant Monitoring 5. With regard to grants, three specific grants in DPLG were investigated. Each illustrated the complexities of the problems in grant monitoring. 6. Transition Fund. This grant demonstrates the flexibility that must be inherent in any monitoring activity as it existed in the financial years 2001/02 and 2002/03 but in 2003/04 it is not part of DORA. Thus, monitoring systems in the grant area have a defined or limited lifespan. 48 Table 13. Municipal Grants and Grant Monitoring By Department Department Grant 2003/04 R'000 Provincial and Local Government Municipal Systems Improvement 150,418 Schedule 6 National Treasury Local Government Financial Management 151,000 National Treasury Local Government Restructuring 315,000 Provincial and Consolidated Municipal Infrastructure Local Government Programme (CMIP) 2,246,253 Provincial and Local Economic Development and Local Government Social Plan Measures 117,000 Provincial and Local Government Municipal Infrastructure (MIG) 47,000 Schedule 6A Sport and Building for Sport and Recreation Recreation Programme 123,095 Public Works Community Based Public Works Programme 249,820 Transport Urban Transport Fund 9,100 Mineral and Energy National Electrification Programme 240,000 Public Works Community Based Public Works Programme 10,000 National Treasury Financial Management Grant 60,915 Schedule 7A Water Affairs and Forestry Water Services Operating Subsidy 836,436 Water Affairs and Implementation of Water Services Forestry Project Grants 1,101,812 TOTAL 5,657,849 7. Municipal Systems Improvement Programme. Data is collected/collated by a contracted-out consultancy consortium on a monthly basis. DPLG have used this service for monitoring some but not all municipal activities for several years. The compliance with reporting requirements is sporadic.26 It is also clear that DPLG considers that the quality of some of the information that is returned is questionable. The reasons for this probably range from capacity difficulties to an indifference to the request for monitoring information at municipality level. 8. Consolidated Municipal Infrastructure Programme (CMIP). The CMIP is aimed at providing at least basic levels of services to low income households and it is aimed at contributing to other government strategic and intervention policy objectives including SMME Development Utilization and Empowerment of Affirmative Business Enterprises. CMIP is the biggest value grant in DORA (nearly 40%); its 26Fifty municipalities, which failed to reply in February 2004, were told that the grant would be withheld unless reports were received. This policy was to some extent an idle threat since the grant was ultimately disbursed. 49 purpose is to fund municipalities to provide internal bulk, connector infrastructure and community facilities to poor households. For infrastructure projects the source documentation originates from the contractors/consultants undertaking the work and there are approximately 50 of these nationwide. Some municipalities are capable of verifying the information while others are not, with reasons ranging from capacity difficulties to a lack of interest. The reporting flow is from contractors/consultants to municipality to province to DPLG. The province also undertakes a monitoring/verification role as well as preparing the data for forwarding on to national level. The use of the province as a coordinator/monitor means that the level of information that is handled nationally is regulated at a reasonable level 9. This particular grant illustrates many of the difficulties and lack of standardization that makes the completion of reporting from the municipal level problematic. For example, the forms for reporting by contractors/consultants are not consistent between municipalities although they contain essentially the same information. 10. An issue of concern with the operation of this grant and an item, which again demonstrates the variability of grant structure and allocation, is the perceived lack of flexibility by Program Managers for re-allocation/virement within year. Partly this was blamed on the "gazetting" process for approving such transfers in December each year. This timing means that there is not on occasion enough time remaining in the financial year to deliver fully the impact a reallocation would allow. This has implications both directly and indirectly by: Creating inflexibility and thus increasing the chance of an `underspend' Creating a pressure to spend the budget and an artificial funds flow pressure at the end of the year. (While roll-over is possible, the pressure for in-year cash budget compliance is a very strong influence) Creating opportunity for the money to be absorbed for other uses and, in an effort to get the job done, some of the Key Performance Indicators (KPI's) may be compromised. 11. The virement level necessary to alleviate some of this pressure is said to be possibly around 3%, and does not need to be inter-region/province. However, to establish a threshold would further complicate monitoring and eats into the completeness of accountability and transparency via the gazetting process. Part of the difficulty is the constraint that delivery of large infrastructure projects within a fixed cash financial year brings. Funding within a three-year Medium-Term Framework-- with flexibility/variance within the overall MTEF period--could be an appropriate tool to substitute for or complement virement. However, the ability to do this would require inter-year facilities, something which adds to the complexity of monitoring. 12. The following reports are submitted from municipalities for CMIP: Financial Monthly Report Quarterly Key Performance Indicators (KPIs) Monthly Report From Nodal Areas Quarterly Reports On Capacity Building Monthly Site Visit Reports 50 13. The key performance indicators include targets for training, employment, urban/rural utilization, female employment, youth employment, integration with Development Plan etc. With regards to disbursement, the model SANT imply in DORA, without intermediaries between the grant allocation and the municipalities, is not used for CMIP. Disbursement is to the top 20 (in size) municipalities, district municipalities and metros. The allocation to smaller (B) municipalities is done via the district (C) municipalities. While this is allowed technically under Para 26 of DORA and is justified on the basis that splitting the allocation up into small proportions makes the allocation to a small municipality not viable for any significant infrastructure exercise, it does not follow the principle of every municipality being seen to get a share. A cut-off level of R1.3m is used as the minimum scale. 14. Another issue noted in the operation of infrastructure grants was the minimal or zero O&M budgeting that is taking place in the municipalities. It would appear that maintenance is not treated as an important item, and that a proportion of CMIP will eventually start to cover an element of infrastructure enhancement that hides under- investment in maintenance. This issue is an example of the incompleteness in monitoring arrangements, which focus on the grant usage not the interrelationships that cover wider prudent financial management in municipalities. The Project Viability System 15. DPLG currently operate a system known as the Project Viability System. This originally started as a system called Project Liquidity in 1996/97. It was reviewed and suggestions for refinement given in 1998 and was renamed the Project Viability. It has now evolved into a bigger system with additional data collected. 16. The data is collected by out-sourced consultancy contracts. Provincial Project Managers (PPM) oversee the work of the teams in each province. The method of collection is via questionnaires, which are issued each quarter and are available also in a spreadsheet format. Some support for weaker municipalities is given. A sample of the financial schedules of these questionnaires is attached as Annex C. 17. The data is collected manually and put on a spreadsheet which is then collated and stored on an Access database. Until 2002 this was solely under the control of the out-sourced provider but there is now a copy on a DPLG server, although as yet no DPLG staff has been trained on the use of the system. 18. The DPLG staff did not profess a great deal of confidence in the reliability of the data on the Project Viability system, and the process was said to be "consultant driven". DPLG are reviewing the overall Project Viability process as part of an overall restructuring that is taking place in the Department. 19. In terms of relevance, reliability and speed of data on the system, the latest reports on the Project Viability System were at the time of this study for the quarter ended six months previously and there was no certainty when the next quarter would be available on the system. Consequently, at best, the system is running between 3 to 6 months behind real time. In terms of quality/validation of the data, the quality measure appears to be " a return is received". In this context the overall percentage nationwide is about 80%, with provinces like Western Cape achieving 100%. Reasons for this variation in performance include the fact that some provinces took ownership 51 themselves rather than the appointed consortium driving the system, although it is perhaps also pertinent that larger historical investment/capacity in the municipalities in some provinces may be a factor. 20. The Consortium providers have identified municipalities where additional support is necessary because of capacity difficulties and this is being given in 2003. However there must be concern that what was originally a "temporary" collection method and the actual data storage for Project Viability has been out-sourced to a consultancy consortium on an annual contract for the last 7 years. As a result the institutional capacity/memory is with the out-sourced consortium, and to a large extent they own the Project Viability system. Use of the Viability System in any overall initiative to improve monitoring must recognize this issue. STATISTICS SA AND OFFICE OF THE AUDITOR GENERAL 21. In a wider sense, Statistics SA and the Auditor General are part of the national monitoring mechanism for municipalities. Both of these organizations undertake an "after the event" approach to monitoring, and therefore, are not able to contribute to an "early warning" approach. Rather, such institutions place a premium on accuracy rather than timeliness of monitoring reports. 22. While their ex post emphasis should not preclude their involvement in national monitoring of municipalities, both Stats SA and the Office of the Auditor General, their analysis is, by its very nature, based on historical/previous year information. In fact, while Statistics SA is vital in establishing allocations to the municipal level, it does not have an impact on the in-year monitoring of DORA. 52 ANNEX 3: WEB-BASED REPORTING FORM 53 ANNEX 4: DATABASE DESIGN 54 ANNEX 5: IMPACT ASSESSMENT INDICATORS OBJECTIVE INDICATORS Household Indicators Unemployment amongst adult men and women in household Density per household of retired/pensioned/disabled persons Highest education attained by adult men and women Household density (persons per room) Household sleeping density (persons per sleeping room) Chronic disease profile amongst adults and children in household Homestead Indicators Nature of dwelling structure Ownership of dwelling Mortgage/rent burden Receipt of government subsidy Nature of defects in structure Income and Expenditure Indicators · Mean household income and expenditure · Net portion of household income derived from government grants and benefits · Net portion of household income derived from salaries and wages · Net portion of household income derived from transfers · Net portion of family income derived from retirement packages · Net portion of household expenditure allocated to food and groceries · Net portion of household expenditure allocated to transport costs · Net portion of household expenditure allocated to payment for municipal services. · Net portion of household expenditure allocated to payment for health and education services · Net portion of household expenditure allocated to servicing debt · Net portion of household expenditure allocated to purchasing fuel · Net portion of household expenditure allocated to property/homestead costs · Net portion of household income available after expenditure Municipal Services Indicators · Receipt/non-receipt of basic municipal services such as water, electricity, sanitation. · Type of electricity connection to household and nature of account · Receipt/non-receipt of municipal services such as roads, recreational and community facilities. · Duration of receipt of all assessed services at household · Duration and receipt of free services at household · Account status for basic services such as water and electricity · Current and past suspensions of electricity and water services due to non-payment. · Primary sources of fuel for cooking, heating and lighting · Convenience cost to household of alternatives to basic services (e.g.; collecting wood, water, etc.) Telephony Services · Nature of telephony services at household (landline and cellular) · Household ownership of landline and cellular telephones · Nature of landline and cellular telephony accounts National/Provincial Service Indicators · Receipt of government grants (pension, disability, child foster care, etc.) · Rated accessibility of various health services (mobile and stationery clinics and hospitals) · Rated affordability of various health services (mobile and stationery clinics and hospitals) · Rated accessibility of education services (crèches, primary and secondary schools) · Rated affordability of education services (crèches, primary and secondary schools) · Assessment of police services (service-at-station, foot/motor patrols, response to distress calls, special ops) · Receipt of other government services per household (skills development, job creation, etc.) 55 ANNEX 5: IMPACT ASSESSMENT INDICATORS (Cont'd) SUBJECTIVE INDICATORS Personal Safety · Assessment of personal safety in house during day and at night · Assessment of personal safety in street during day and at night · Assessment of personal safety in public commercial and worship spaces during day and at night · Assessment of personal safety in public facilities such as taxi ranks and stations during day and at night. Assessment of Service Delivery · Satisfaction with receipt of all services · Expectations with respect to all services · Differential between expectation and satisfaction. · Modal recommendations for improvement in services Ranked Importance of Different Services · Assessed importance to the household of a range of municipal, provincial, national government services. Assessment of Municipality Performance · Trust­the household's assessed level of trust in the municipality · Responsiveness­household's assessed responsiveness of municipality to their needs · Communication­household's assessed satisfaction with the communication channels of the municipality · Consultation­household's assessed satisfaction with the consultative channels of the municipality · Corruption­the household's perception of corruption by public officers and officials 56 ANNEX 6: PILOT HOUSEHOLD IMPACT ASSESSMENT QUESTIONNAIRE HOUSEHOLD SERVICE DELIVERY IMPACT SURVEY Checked by Interviewer Interviewer sign Date of back check Household back-checked? Edited by Sign Supervisor signature Yes 1 Coded by Sign No 2 Household Selection Procedure Select a random household. A household is a group of people who presently eat together from the same pot. Start your walk pattern selecting every 10th dwelling. Keep to the right side of the road. If a call is unsuccessful, use the table below to record your progress until you make a successful call. Answer a code number for unsuccessful calls only. Reasons for Unsuccessful Calls Household Household Household Household Household Household Household 1 2 3 4 5 6 7 Refused to be interviewed 1 1 1 1 1 1 1 Person selected was never at home 2 2 2 2 2 2 2 Household/Premises empty for the survey period 3 3 3 3 3 3 3 Spoke only a foreign language/ deaf 4 4 4 4 4 4 4 Other (specify)_________________________________ 6 6 6 6 6 6 6 Not Applicable 997 997 997 997 997 997 997 Fieldworker Name: NB NB NB Record starting time of interview HH MM A: Interview Record Good Day. My name is ...............................and I am from Citizen Surveys, a research organization. We are conducting this survey to learn about the services you receive from government and to understand how satisfied you are with these services. We are collecting this information to identify general problems and gaps in local service delivery and to assist government to improve the delivery of services to household such as yours. We ask that you assist us by completing this short questionnaire. The information you give us is confidential. We ask for your name and address and contact information only in case we need to contact you to check that the information was recorded correctly. We shall contact some of our respondents in order to verify the information. Names and addresses will then be discarded; they will not be recorded on our computers. Personal information will never be made public. Department of Finance Pretoria NOTE FOR FIELDWORKER a. Please ensure that this survey is completed either with the head of the household or an adult with access to the necessary information (income, expenses, etc.). b. Please read all questions to respondents and indicate the responses available by slowly reading them aloud. Repeat response options if necessary. c. Show cards where indicated on questionnaire d. Please ensure that ALL APPLICABLE questions are completed with the necessary information. 57 B: RESPONDENT DETAILS SECTION 1: HOUSEHOLD DETAILS YES NO 1 Is the respondent the head of the household? 1 2 2 Please provide the following details for the head of this household (even if different from person being interviewed): Age YEARS MALE FEMALE Gender 1 2 YES NO Employed at present? 1 2 Retired/Disabled/Pensioned 1 2 Highest education level attained (please specify) 3 How many people live in this house for more than 15 days in each month? ADULTS (Over 18 Years) CHILDREN (Under 18 Years) Male Female Male Female 4 Please indicate how many there are of the following in the household: Adults presently Adults presently Children (under 18 employed/ employed/ Adults unemployed Adults unemployed years) attending self employed in self employed in for less than six for more than six Retirees, pensioners, school formal sector informal sector months in past year months in past year disabled, etc. 5 Please indicate the highest level of education attained by all adults in the household. Number of Males Number of Females No formal school education Less than Grade 7 (Standard 5) completed Completed Grade 7 (Standard 5) Completed Grade 10 (Standard 8) Completed Grade 12 (Matric) Post matric three year qualification (including diplomas) Degree/s 58 SECTION 2: HOMESTEAD DETAILS 6 OBSERVATION BY INTERVIEWER What type of dwelling is the main homestead? (Please answer only one type of dwelling and indicate the number thereof) Type Of Dwelling Number Dwelling/house or brick structure on separate stand or yard 1 Flat in a block of flats 2 Town/cluster/semi-detached house (simplex, duplex or triplex) 3 Unit in retirement village 4 Dwelling/house/flat/room in backyard 5 Traditional dwelling/hut/structure of traditional materials 6 Informal dwelling/shack in backyard 7 Informal dwelling/shack NOT in backyard (in informal/squatter settlement) 8 Other (specify) 9 7 How long has the household lived at this dwelling? Years Months 8 Does the household own this homestead? YES 1 Go to Question 9 NO 2 Go to Question 13 9 Did anyone in this family receive a government-housing subsidy to obtain this homestead? YES 1 Go to Question 10 NO 2 Go to Question 11 10 How much is the subsidy and how long Amount of Subsidy Time since subsidy was received since it was first received? R Yrs Mths 11 Does the household have a bond or loan on this homestead? YES 1 Go to Question 12 NO 2 Go to Question 13 12 How much is still owed on the bond or loan? Amount Owed on Bond or Loan (Rands) R 13 Please indicate the following for the household dwelling: Total Number of Rooms (including kitchen, lounge etc. but excluding bathroom/toilet) Number of Rooms Used for Sleeping 59 SECTION 3: WATER SERVICES 14 What is the most often used source of drinking water in this household SINGLE MENTION? Piped water ­ tap/s inside dwelling 1 Go to Question 15 Piped water - tap/s in yard 2 Go to Question 15 Water carrier/tanker 3 Go to Question 15 Public tap/kiosk 4 Go to Question 15 Borehole/Rainwater tank/well 5 Go to Question 19 Flowing river/stream/spring 6 Go to Question 19 Dam/stagnant water 7 Go to Question 19 Other (specify) 8 Go to Question 19 15 How long has it been since this household started receiving water services Years Months from government? 16 Do you at present receive free water services from government? NO 1 Go to Question 17 YES 2 IF YES, how long have you received free service? Months IF YES What quantity is received each month? KiloLitres Don't Know -99 17 For households paying for water: Looking back over the past year only, please indicate which of the following is true for this Amount Owed/ Service suspended Owing for? (Months) household: SINGLE MENTION The water service is current and the account is not in arrears 1 The water service is current but the account is in arrears 2 R The water service is current but was previously suspended for non-payment 3 R The water service is at present suspended for non-payment 4 R 18 For all households receiving water services, how Very Some- Don't often in the past year have you experienced Often Often times Rarely Never Know or N/A disruptions in services due to factors other than non- payment(poor maintenance, storms, etc.)? 1 2 3 4 5 -99 19 Do resident household members have to fetch and carry water to the house each day? YES 1 Go to Question 20 NO 2 Go to Question 22 20 What is the average number of trips made to fetch water every day? 21 How long does each round trip take on average (including time spent Min waiting in queue)? 60 SECTION 4: ENERGY SERVICES 22 Is the dwelling connected to an electricity supply? YES ­ Paid by monthly account 1 Go to Question 23 YES ­ Pre-paid card 2 Go to Question 23 YES ­ Free electricity service 3 Go to Question 23 NO ­ No electricity connection to house 4 Go to Question 29 23 How long has this household been connected to an electricity supply? Years Months Don't Know -99 24 FOR MONTHLY ACCOUNTS: Looking back over the past year Service only, please indicate which of the following is true for this household Amount Owed/ Owing suspended SINGLE MENTION for?(mnths) The electricity service is current and the account is not in arrears 1 The electricity service is current but the account is in arrears 2 R The electricity service is current but was previously suspended for non-payment 3 R The electricity service is at present suspended for non-payment 4 R 25 FOR PRE-PAID CARDS: Looking back over the past year only, please indicate which of the following is true for this household SINGLE MENTION : There was always enough on the card for a continuous supply of electricity 1 The electricity supply was sometimes interrupted for less than one week because the card was empty 2 The electricity supply was sometimes interrupted for more than one week because the card was empty 3 The electricity supply was frequently interrupted because the card was empty 4 26 For households receiving free services: how long has it been since Years Months Don't you started receiving free electricity services? Know -99 27 How often does the household use electricity Never/ Not Some of for: at All Rarely the Time Most of the Time Always N/A a) Cooking 1 2 3 4 5 99 b) Lighting 1 2 3 4 5 99 c) Heating 1 2 3 4 5 99 d) Using appliances (TV, kettle, etc.) 1 2 3 4 5 99 28 For all households connected to electricity, how often Very Don't in the past year have you experienced disruptions in Often Often Sometimes Rarely Never Know or N/A services due to factors other than non-payment of accounts (equipment failure, storms, etc.)? 1 2 3 4 5 99 61 29 a. Which of the following does the household use as its PRIMARY energy for cooking, lighting and Primary Primary Primary heating SINGLE MENTION PER COLUMN Energy for Energy for Energy for 29B. Average cost Cooking Heating Lighting per month (Rands) b. What are the average costs per month? Electricity 1 1 1 Paraffin 2 2 2 R Gas 3 3 3 R Wood 4 4 4 R Coal/Charcoal 5 5 5 R Candles 6 6 6 R Other (Please specify) 7 7 7 R 30 Do resident household members collect wood at least once a week? YES 1 Go to Question 31 NO 2 Go to SECTION 5 : SANITATION SERVICES 31 What is the average number of trips made to fetch wood per week? 32 How long does each round trip take on average (including time spent collecting Min wood)? SECTION 5: SANITATION SERVICES 33 What kind of toilet does the household primarily use? Flush toilet 1 Improved pit latrine with ventilation (VIP) 2 Other pit latrine 3 Bucket toilet 4 Chemical toilet 5 Other (Specify) 6 None (no toilet available) 7 34 If available, where is the toilet located? Inside dwelling 1 Outside dwelling - on stand/in the yard 2 Outside dwelling - off stand/outside the yard 3 35 If any sanitation services (for flush toilet, collection of buckets, etc.) are Years Months Don't know/ received from government, how long have you been receiving this? NA -99 62 SECTION 6: TELEPHONE SERVICES 36 Is this household connected to a telephone line (Telkom landline)? YES ­ Paid by monthly account 1 Go to Question 37 YES ­ Pre-paid card 2 Go to Question 37 NO ­ No Telkom line connected 3 Go to Question 40 37 If the household does have a Telkom landline, how long has it Years Months Don't know/ NA been since the telephone line was installed? -99 38 FOR MONTHLY ACCOUNTS: Looking back over the past year only, Amount Service please indicate which of the following is true for this household SINGLE Owed/ suspended MENTION: Owing for?(mnths) The telephone service is current and the account is not in arrears 1 The telephone service is current but the account is in arrears 2 R The telephone service is current but was previously suspended for non-payment 3 R The telephone service is suspended for non-payment 4 R 39 FOR PRE-PAID CARDS: Looking back over the past year only, please indicate which of the following is true for this household SINGLE MENTION: There was always enough on the card for a continuous supply of telephone services 1 The telephone service was sometimes interrupted for less than one week because the card was empty 2 The telephone service was sometimes interrupted for more than one week because the card was empty 3 The telephone service was frequently interrupted because the card was empty 4 40 How many cell phones are there at present in this household? (Including contract and pre-paid accounts) 41 If no phone at all, how long does it take to travel each round trip to the nearest Time Taken working telephone (by your usual means of transport and including time spent (minutes) waiting in queue)? SECTION 7: REFUSE REMOVAL 42 How is refuse removed from your premises? Removed by local authority 1 Go to Question 43 Removal by community team or members (e.g., Masekane) 2 Go to Question 43 Communal refuse dump 3 Go to Section 8 Own refuse dump 4 Go to Section 8 Other means of removal 5 Go to Section 8 No refuse removal 6 Go to Section 8 43 If local authority/Masekane team removes refuse, how has Years Months Don't know long this household received this service? -99 63 For households where the local authority /Masekane Once a Once a Once Less than Don't every 3 Once a once a Know or team collects refuse, how frequent is the service? week fortnight weeks month month N/A 5 4 3 2 1 9 44 For households where local authority/Masekane Very Very Don't team collects refuse, how reliable is the service? reliable Reliable Average Unreliable unreliable Know or N/A 1 2 3 4 5 9 SECTION 8: ROADS 45 Is the road to your homestead primarily? SINGLE MENTION Tarred and without potholes 1 Tarred but with potholes 2 Gravel road regularly maintained 3 Gravel road poorly maintained 4 Informal single vehicle lane 5 Informal pedestrian lane 6 No road to dwelling 7 46 Is the road to your homestead: Useable in all types of weather 1 Useable only in good and fair weather 2 Useable only in good weather 3 47 Does the road have proper provision for storm water run-off (drains, cutaways, etc.)? YES 1 NO 2 DON'T KNOW -99 48 Is there provision for safe passage of pedestrians (pavements, etc.) YES 1 NO 2 DON'T KNOW -99 49 Are there streetlights in your road? YES 1 Go to Question 50 NO 2 Go to Section 9 PARKS AND RECREATIONAL 50 Are the streetlights in All are working Most are working Few are working None are Working Don't Know/Unsure working order? 1 2 3 4 -99 Repaired Repaired after a few Repaired after Rarely/Not repaired Don't know/Unsure 51 If broken and reported to promptly complaints many complaints at all municipality, are they? 1 2 3 4 -99 64 SECTION 9: PARKS AND RECREATIONAL FACILITIES AND SERVICES 52 Is there a public park/public gardens within 2 kms of your homestead? YES 1 Go to Question 53 NO 2 Go to Question 57 Don't Know -99 Go to Question 57 53 How long has the park been open to the public? Years Months Don't Know -99 54 Is the park/garden well Most of the time Some of the time Rarely Not at all Don't Know/Unsure maintained? 1 2 3 4 -99 55 Does the park/garden have any of the following Yes No Don't Know/Unsure Proper benches for sitting 1 2 -99 Clean working toilets 1 2 -99 Drinking taps 1 2 -99 Child safe playground 1 2 -99 Walkways 1 2 -99 56 Is it safe for families to spend Most of the time Some of the time Rarely Not at all Don't time in this park/garden? Know/Unsure 1 2 3 4 99 57 Is there a community centre within 2 kms of your homestead? YES 1 Go to Question 58 NO 2 Go to Question 61 Don't Know 99 Go to Question 61 58 How long ago did the community centre open to Years Months Don't Know the public? 99 59 Is the community centre Most of the time Some of the time Rarely Not at all Don't Know well maintained? 1 2 3 4 99 60 Is the community centre Most of the time Some of the time Rarely Not at all Don't Know/Unsure safe to use? 1 2 3 4 99 61 Is there a library/library service within 2 kms of your house? YES 1 Go to Question 62 NO 2 Go to Section 10 HEALTH SERVICES Don't Know 9 Go to Section 10 HEALTH SERVICES 62 How long has the library service been running? Years Months Don't Know/Unsure 9 63 Is the library service regular Most of the time Some of the time Rarely Not at all Don't Know/Unsure and satisfactory? 1 2 3 4 9 65 SECTION 10: HEALTH SERVICES 64 In terms of time taken and distance travelled, how easy/difficult is it for a member of the Don't household to access health services at the Very Manage- Very Easy Easy able Difficult Difficult Know or following places: one answer per row N/A a. At a clinic (stationery) 1 2 3 4 5 99 b. At a clinic (Mobile) 1 2 3 4 5 99 c. At a hospital 1 2 3 4 5 99 65 In terms of the cost involved, how easy/difficult is it for a member of the household to access Very Manage- Very Don't health services at the following places: one Easy Easy able Difficult Difficult Know or answer per row N/A a. At a stationery clinic 1 2 3 4 5 99 b. At a mobile clinic 1 2 3 4 5 99 c. At a hospital 1 2 3 4 5 99 SECTION 11: SAFETY AND SECURITY SERVICES 66 How would you rate the following police activities Very Poor/ Good/ Very Don't in your area? one answer per row Poor Inadequate Ave Adequate Good Know or N/A a. Service at the police station 1 2 3 4 5 99 b. Police response to distress call from community 1 2 3 4 5 99 c. Police liaison/relationship with community 1 2 3 4 5 99 d. Police officers on foot patrols 1 2 3 4 5 99 e. Police officers in vehicle patrols 1 2 3 4 5 99 f. Police special operations (roadblocks, etc.) 1 2 3 4 5 99 g. Police special operations (e.g., crime sweeps) 1 2 3 4 5 99 67 How safe do you feel in the following places and at the Very Very Don't following times? (one answer per row ) Unsafe Unsafe Safe Safe Know or N/A a. At home during the day 1 2 3 4 99 b. At home during the night 1 2 3 4 99 c. On the street during the day 1 2 3 4 99 d. On the street at night 1 2 3 4 99 e. In public places (shopping centres, church, etc.) in the day 1 2 3 4 99 f. In public places (shopping centres, church, etc.) at night 1 2 3 4 99 g. In public places such as taxi ranks and train stations in the day 1 2 3 4 99 h. In public places such as taxi ranks and train stations at night 1 2 3 4 99 SECTION 12: EDUCATION SERVICES 68 In terms of time taken and distance travelled, how easy/difficult is it for a member of the Very Don't Easy Manageable Difficult Very Know or household to attend: one answer per row Easy Difficult N/A a. A crèche/pre-school 1 2 3 4 5 99 b. Primary school 1 2 3 4 5 99 c. Secondary School 1 2 3 4 5 99 66 69 In terms of the cost involved, how easy/difficult Don't is it for a member of the household to attend: Very Very Easy Easy Manageable Difficult Difficult Know or one answer per row N/A a. A crèche/pre-school 1 2 3 4 5 99 b. Primary school 1 2 3 4 5 99 c. Secondary School 1 2 3 4 5 99 SECTION 13: OTHER GOVERNMENT SERVICES 70 In the past year, have any members of the household participated in government sponsored programs or projects or If Yes, how many courses in the following areas (If yes please indicate the YES NO household members have participated? number who have done so): one answer per row a. Job creation programs (e.g.; public works programs) 1 2 b. Skills development programs (e.g., Dept of Labour) 1 2 c. Business development programs (e.g., DTI) 1 2 d. Other: Please Specify 1 2 e. Other: Please Specify 1 2 SECTION 14: IMPORTANCE OF SERVICES Please indicate to us how important the Some- following services are to your household: Extremely Very Not Don't Not Important Important Important what Important Know Applicable one answer per row Important 71 Electricity Services 5 4 3 2 1 99 6 72 Water Services 5 4 3 2 1 99 6 73 Telephone (Landline) services 5 4 3 2 1 99 6 74 Refuse removal services 5 4 3 2 1 99 6 75 Sanitation services 5 4 3 2 1 99 6 76 Road infrastructure 5 4 3 2 1 99 6 77 Street lighting 5 4 3 2 1 99 6 78 Parks and gardens 5 4 3 2 1 99 6 79 Community facilities 5 4 3 2 1 99 6 80 Library services 5 4 3 2 1 99 6 81 Safety and security (police) services 5 4 3 2 1 99 6 82 Health services 5 4 3 2 1 99 6 83 Education services 5 4 3 2 1 99 6 84 Social welfare services 5 4 3 2 1 99 6 67 SECTION 15: ASSESSMENT OF SERVICES Please indicate how satisfied you are at present with the following services (include in Neither your assessment the quality of the service, Very Satisfied Nor Very Don't Not Satisfied Satisfied Dissatisfied DissatisfiedDissatisfied Know Applicable cost, etc.): one answer per row 85 Electricity Services 5 4 3 2 1 99 6 86 Water Services 5 4 3 2 1 99 6 87 Telephone (Landline) services 5 4 3 2 1 99 6 88 Refuse removal services 5 4 3 2 1 99 6 89 Sanitation services 5 4 3 2 1 99 6 90 Road infrastructure 5 4 3 2 1 99 6 91 Street lighting 5 4 3 2 1 99 6 92 Parks and gardens 5 4 3 2 1 99 6 93 Community facilities 5 4 3 2 1 99 6 94 Library services 5 4 3 2 1 99 6 SECTION 16: ASSESSMENT OF MUNICIPALITY Please indicate how much you agree or disagree Neither with the following statements about your local Strongly Agree Strongly Don't Nor Agree Agree Nor Disagree Disagree Know Applicable municipality: one answer per row Disagree 95 The municipality is genuinely interested in the well 5 4 3 2 1 99 6 being of our community 96 Municipal officials are generally responsive to our 5 4 3 2 1 99 6 needs when we communicate these to them 97 The municipality is sympathetic to our grievances and 5 4 3 2 1 99 6 problems 98 The municipality responds promptly to our complaints 5 4 3 2 1 99 6 99 The municipality has staff who are well trained and 5 4 3 2 1 99 6 competent 100Municipal staff are not prone to corruption and 5 4 3 2 1 99 6 nepotism 101Municipal elected officials are not prone to corruption 5 4 3 2 1 99 6 and nepotism 102The municipality makes an effort to consult with 5 4 3 2 1 99 6 communities on matters affecting them 103The municipality ensures that it is transparent in its 5 4 3 2 1 99 6 decision making 104The municipality ensures that it is transparent in its 5 4 3 2 1 99 6 planning and implementation 105The municipality makes an effort to include us in 5 4 3 2 1 99 6 planning and development 106We can generally trust municipal officials to do what is 5 4 3 2 1 99 6 in the best interests of our community 107The municipality provides easy to understand billing 5 4 3 2 1 99 6 information on accounts 108The municipality informs us properly of new 5 4 3 2 1 99 6 developments in service delivery 109The municipality maintains good communication 5 4 3 2 1 99 6 68 channels with local communities 110What is the monthly household income from the following sources? (if possible, please use the monthly average over last six months, otherwise record amounts for month prior to survey) SOURCES OF MONTHLY INCOME AMOUNT a. Salary/earnings from formal & informal sector and self employment b. Pension from work/retirement package c. State Pension/Old Age grant d. Unemployment Insurance (UIF) e. Disability Grant f. War Veterans Grant g. Care dependency Grant h. Child Support Grant i. Child Foster Care Grant j. Regular transfers/allowances from family members living elsewhere k. Other (e.g., grant in aid, etc.) TOTAL MONTHLY INCOME 111What are the monthly household expenses for the following items? (If possible, use average over past six months, otherwise record for month prior to survey): MONTHLY HOUSEHOLD EXPENSES AMOUNT Food and groceries (excluding clothing, fuel, furniture, accounts, etc.) Public transport (bus, taxi, train) Private transport (including fuel, repairs, etc.) Rent Bond repayments Property rates Loans Telephone ( TELKOM) Cell phone Airtime Municipal services: Water Electricity Sanitation/refuse removal Hire purchase: Motor vehicle/s Furniture Clothing Other Other energy expenses (e.g., wood, paraffin, charcoal/coal, candles, gas, purchasing/charging batteries, diesel oil for generators, etc.) Education (including clothing, books, school fees, etc.) Health (including trips to doctor, clinic/hospital, medicines, medical aid contributions etc.) Lotto Other (All other expenses) TOTAL MONTHLY EXPENSES 69 112What language do you speak at home the most? RECORD BELOW. SINGLE MENTION Afrikaans 01 Sesotho 07 English 02 Setswana 08 Xhosa 03 SiSwati 09 Ndebe 04 Tshivenda 10 Zulu 05 Xitsonga 11 Sepedi 06 Other 12 113 Race: BY OBSERVATION ONLY African 01 Coloured 02 Indian 03 White 04 ----------THANK YOU VERY MUCH FOR YOUR CO-OPERATION-------- INTERVIEWER DECLARATION I hereby certify that this interview has been completed and checked in strict accordance with the instructions given to me by Citizen Surveys. INTERVIEWER NAME SIGNATURE DATE LENGTH OF INTERVIEW 70 Africa Region Working Paper Series Series # Title Date Author Progress in Public Expenditure ARWPS 1 Management in Africa: Evidence January 1999 C. Kostopoulos from World Bank Surveys Toward Inclusive and Sustainable ARWPS 2 Development in the Democratic March 1999 Markus Kostner Republic of the Congo Business Taxation in a Low- ARWPS 3 Revenue Economy: A Study on Ritva Reinikka Uganda in Comparison with June 1999 Duanjie Chen Neighboring Countries ARWPS 4 Pensions and Social Security in Sub- Luca Barbone Saharan Africa: Issues and Options October 1999 Luis-A. Sanchez B. Forest Taxes, Government Revenues ARWPS 5 and the Sustainable Exploitation of January 2000 Luca Barbone Tropical Forests Juan Zalduendo The Cost of Doing Business: Firms' ARWPS 6 Experience with Corruption in June 2000 Jacob Svensson Uganda On the Recent Trade Performance of ARWPS 7 Sub-Saharan African Countries: August 2000 Francis Ng and Cause for Hope or More of the Same Alexander J. Yeats ARWPS 8 Foreign Direct Investment in Africa: Old Tales and New Evidence November 2000 Miria Pigato The Macro Implications of ARWPS 9 HIV/AIDS in South Africa: A November 2000 Channing Arndt Preliminary Assessment Jeffrey D. Lewis Revisiting Growth and ARWPS 10 Convergence: Is Africa Catching December 2000 C. G. Tsangarides Up? Spending on Safety Nets for the ARWPS 11 Poor: How Much, for How Many? January 2001 William J. Smith The Case of Malawi ARWPS 12 Tourism in Africa February 2001 Iain T. Christie D. E. Crompton 71 Africa Region Working Paper Series Series # Title Date Author ARWPS 13 Conflict Diamonds February 2001 Louis Goreux Reform and Opportunity: The ARWPS 14 Changing Role and Patterns of March 2001 Jeffrey D. Lewis Trade in South Africa and SADC ARWPS 15 The Foreign Direct Investment Environment in Africa March 2001 Miria Pigato ARWPS 16 Choice of Exchange Rate Regimes for Developing Countries April 2001 Fahrettin Yagci ARWPS 18 Rural Infrastructure in Africa: Policy Directions June 2001 Robert Fishbein S. Paternostro ARWPS 19 Changes in Poverty in Madagascar: 1993-1999 July 2001 J. Razafindravonona David Stifel Information and Communication ARWPS 20 Technology, Poverty, and Development in sub-Saharan Africa August 2001 Miria Pigato and South Asia Handling Hierarchy in Decentralized Settings: Governance Underpinnings Navin Girishankar ARWPS 21 of School Performance in Tikur September 2001 A. Alemayehu Inchini, West Shewa Zone, Oromia Yusuf Ahmad Region Child Malnutrition in Ethiopia: Can ARWPS 22 Maternal Knowledge Augment The October 2001 Luc Christiaensen Role of Income? Harold Alderman ARWPS 23 Child Soldiers: Preventing, Demobilizing and Reintegrating November 2001 Beth Verhey ARWPS 24 The Budget and Medium-Term Expenditure Framework in Uganda December 2001 David L. Bevan Design and Implementation of Guenter Heidenhof ARWPS 25 Financial Management Systems: An January 2002 H. Grandvoinnet African Perspective Daryoush Kianpour B. Rezaian Francis Ng ARWPS 26 What Can Africa Expect From Its Traditional Exports? February 2002 Alexander Yeats 72 Africa Region Working Paper Series Series # Title Date Author Jeffrey D. Lewis ARWPS 27 Free Trade Agreements and the SADC Economies February 2002 Sherman Robinson Karen Thierfelder Medium Term Expenditure ARWPS 28 Frameworks: From Concept to P. Le Houerou Practice. Preliminary Lessons from February 2002 Robert Taliercio Africa ARWPS 29 The Changing Distribution of Public Samer Al-Samarrai Education Expenditure in Malawi February 2002 Hassan Zaman Serge Michailof ARWPS 30 Post-Conflict Recovery in Africa: An Agenda for the Africa Region April 2002 Markus Kostner Xavier Devictor Efficiency of Public Expenditure Distribution and Beyond: A report ARWPS 31 on Ghana's 2000 Public Expenditure May 2002 Xiao Ye Tracking Survey in the Sectors of S. Canagaraja Primary Health and Education Addressing Gender Issues in ARWPS 33 Demobilization and Reintegration August 2002 N. de Watteville Programs Johan A. Mistiaen Berk Soler ARWPS 34 Putting Welfare on the Map in Madagascar August 2002 T. Razafimanantena J. Razafindravonona Gerald Foley ARWPS 35 A Review of the Rural Firewood Paul Kerkhof Market Strategy in West Africa August 2002 Djibrilla Madougou ARWPS 36 Patterns of Governance in Africa September 2002 Brian D. Levy Obstacles and Opportunities for Senegal's International ARWPS 37 Competitiveness: Case Studies of September 2002 Stephen Golub the Peanut Oil, Fishing and Textile Ahmadou Aly Mbaye Industries A Macroeconomic Framework for S. Devarajan ARWPS 38 Poverty Reduction Strategy Papers : October 2002 Delfin S. Go With an Application to Zambia The Impact of Cash Budgets on Poverty Reduction in Zambia: A Hinh T. Dinh ARWPS 39 Case Study of the Conflict between Well Intentioned Macroeconomic November 2002 Abebe Adugna Policy and Service Delivery to the Bernard Myers Poor ARWPS 40 Decentralization in Africa: A Stephen N. Ndegwa Stocktaking Survey November 2002 ARWPS 41 An Industry Level Analysis of Manufacturing Productivity in December 2002 Professor A. Mbaye 73 Africa Region Working Paper Series Series # Title Date Author Senegal Tanzania's Cotton Sector: ARWPS 42 Constraints and Challenges in a December 2002 John Baffes Global Environment ARWPS 43 Analyzing Financial and Private Sector Linkages in Africa January 2003 Abayomi Alawode Modernizing Africa's Agro-Food Steven Jaffee ARWPS 44 System: Analytical Framework and February 2003 Ron Kopicki Implications for Operations Patrick Labaste Iain Christie Hippolyte Fofack ARWPS 45 Public Expenditure Performance in Rwanda March 2003 C. Obidegwu Robert Ngong ARWPS 46 Senegal Tourism Sector Study March 2003 Elizabeth Crompton Iain T. Christie ARWPS 47 Reforming the Cotton Sector in SSA March 2003 Louis Goreux John Macrae HIV/AIDS, Human Capital, and ARWPS 48 Economic Growth Prospects for April 2003 Channing Arndt Mozambique Rural and Micro Finance Regulation ARWPS 49 in Ghana: Implications for William F. Steel Development and Performance of June 2003 David O. Andah the Industry Microfinance Regulation in Benin: ARWPS 50 Implications of the PARMEC LAW for Development and Performance June 2003 K. Ouattara of the Industry Microfinance Regulation in ARWPS 51 Tanzania: Implications for Bikki Randhawa Development and Performance of June 2003 Joselito Gallardo the Industry ARWPS 52 Regional Integration in Central Ali Zafar Africa: Key Issues June 2003 Keiko Kubota ARWPS 53 Evaluating Banking Supervision in Africa June 2003 Abayomi Alawode Microfinance Institutions' Response in Conflict Environments: Eritrea- ARWPS 54 Savings and Micro Credit Program; West Bank and Gaza ­ Palestine for June 2003 Marilyn S. Manalo Credit and Development; Haiti ­ Micro Credit National, S.A. Malawi's Tobacco Sector: Standing AWPS 55 on One Strong leg is Better than on June 2003 Steven Jaffee None Tanzania's Coffee Sector: AWPS 56 Constraints and Challenges in a June 2003 John Baffes Global Environment AWPS 57 The New Southern AfricanCustoms June 2003 Robert Kirk 74 Africa Region Working Paper Series Series # Title Date Author Union Agreement Matthew Stern How Far Did Africa's First Generation Trade Reforms Go? An Lawrence Hinkle AWPS 58a Intermediate Methodology for June 2003 A. Herrou-Aragon Comparative Analysis of Trade Keiko Kubota Policies How Far Did Africa's First Generation Trade Reforms Go? An Lawrence Hinkle AWPS 58b Intermediate Methodology for June 2003 A. Herrou-Aragon Comparative Analysis of Trade Keiko Kubota Policies Rwanda: The Search for Post- AWPS 59 Conflict Socio-Economic Change, October 2003 C. Obidegwu 1995-2001 Linking Farmers to Markets: Morgane Danielou AWPS 60 Exporting Malian Mangoes to October 2003 Patrick Labaste Europe J-M. Voisard Evolution of Poverty and Welfare in AWPS 61 Ghana in the 1990s: Achievements October 2003 S. Canagarajah and Challenges Claus C. Pörtner Reforming The Cotton Sector in AWPS 62 Sub-Saharan Africa: SECOND November 2003 Louis Goreux EDITION AWPS 63 Republic of Madagascar: Tourism Iain T. Christie (E) Sector Study November 2003 D. E. Crompton AWPS 63 République de Madagascar: Etude Iain T. Christie (F) du Secteur Tourisme November 2003 D. E. Crompton Migrant Labor Remittances in AWPS 64 Africa: Reducing Obstacles to Novembre 2003 Cerstin Sander Development Contributions Samuel M. Maimbo Government Revenues and Francisco G. Carneiro AWPS 65 Expenditures in Guinea-Bissau: January 2004 Joao R. Faria Casualty and Cointegration Boubacar S. Barry How will we know Development Results when we see them? Building Jody Zall Kusek AWPS 66 a Results-Based Monitoring and June 2004 Ray C. Rist Evaluation System to Give us the Elizabeth M. White Answer An Analysis of the Trade Regime in AWPS 67 Senegal (2001) and UEMOA's June 2004 Alberto Herrou-Arago Common External Trade Policies Keiko Kubota Bottom-Up Administrative Reform: Talib Esmail AWPS 68 Designing Indicators for a Local June 2004 Nick Manning Governance Scorecard in Nigeria Jana Orac Galia Schechter AWPS 69 Tanzania's Tea Sector: Constraints and Challenges June 2004 John Baffes Tanzania's Cashew Sector: AWPS 70 Constraints and Challenges in a June 2004 Donald Mitchell Global Environment 75 Africa Region Working Paper Series Series # Title Date Author An Analysis of Chile's Trade Francesca Castellani AWPS 71 Regime in 1998 and 2001: A Good July 2004 A. Herrou-Arago Practice Trade Policy Benchmark Lawrence E. Hinkle Regional Trade Integration inEast Lucio Castro AWPS 72 Africa: Trade and Revenue Impacts of the Planned East African August 2004 Christiane Kraus Community Customs Union Manuel de la Rocha Post-Conflict Peace Building in AWPS 73 Africa: The Challenges of Socio- Economic Recovery and August 2004 Chukwuma Obidegwu Development An Analysis of the Trade Regime in Francesca Castellani AWPS 74 Bolivia in2001: A Trade Policy Benchmark for low Income August 2004 Alberto Herrou-Aragon Countries Lawrence E. Hinkle Vincent da Cruz AWPS 75 Remittances to Comoros- Volumes, Trends, Impact and Implications October 2004 Wolfgang Fendler Adam Schwartzman Salient Features of Trade AWPS 76 Performance in Eastern and October 2004 Fahrettin Yagci Southern Africa Enrique Aldaz-Carroll Alan Gelb AWPS 77 Implementing Performance-Based Aid in Africa November 2004 Brian Ngo Xiao Ye Poverty Reduction Strategy Papers: Rene Bonnel AWPS 78 Do they matter for children and Young people made vulnerable by December 2004 Miriam Temin HIV/AIDS? Faith Tempest Experience in Scaling up Support to Jean Delion AWPS 79 Local Response in Multi-Country December 2004 Pia Peeters Aids Programs (map) in Africa Ann Klofkorn Bloome What makes FDI work? A Panel AWPS 80 Analysis of the Growth Effect of February 2005 Kevin N. Lumbila FDI in Africa AWPS 81 Earnings Differences between Men Kene Ezemenari and Women in Rwanda February 2005 Rui Wu The Medium-Term Expenditure AWPS 82 Framework: The Challenge of April 2005 Chukwuma Obidegwu Budget Integration in SSA countries Rules of Origin and SADC: Paul Brenton AWPS 83 The Case for Change in the Mid June 2005 Frank Flatters Term Review of the Trade Protocol Paul Kalenga Chukwuemeka Anyamele AWPS 84 Sexual Minorities, Violence and AIDS in Africa July 2005 Ronald Lwabaayi Tuu-Van Nguyen, and Hans Binswanger 76 Africa Region Working Paper Series Series # Title Date Author Poverty Reducing Potential of AWPS 85 Smallholder Agriculture in Zambia: July 2005 Paul B. Siegel Opportunities and Constraints Jeffrey Alwang Infrastructure, Productivity and AWPS 86 Urban Dynamics in Côte d'Ivoire: Zeljko Bogetic An empirical analysis and policy July 2005 Issa Sanogo implications Poverty in Mozambique: Louise Fox AWPS 87 Unraveling Changes and August 2005 Elena Bardasi, Determinants Katleen Van den Broeck Operational Challenges: Community Home Based Care AWPS 88 (CHBC) for PLWHA in Multi- August 2005 Nadeem Mohammad Country HIV/AIDS Programs Juliet Gikonyo (MAP) for Sub-Saharan Africa AWPS 89 Framework for Forest Resource Management in Sub-Saharan Africa August 2005 Giuseppe Topa AWPS 90 Kenya: Exports Prospects and Francis Ng Problems September 2005 Alexander Yeats Lawrence E. Hinkle Albero Herrou Aragon AWPS 91 Uganda: How Good a Trade Policy Benchmark for Sub-Saharan-Africa September 2005 Ranga Rajan Krishnamani Elke Kreuzwieser AWPS 92 Community Driven Development in David Everatt Lulu South Africa, 1990-2004 October 2005 Gwagwa The Rise of Ghana''s Pineapple AWPS 93 Industry from Successful take off to November 2005 Morgane Danielou Sustainable Expansion Christophe Ravry South Africa: Sources and AWPS 94 Constraints of Long-Term Growth, December 2005 Johannes Fedderke 1970-2000 AWPS 95 South Africa''s Export Performance: Lawrence Edwards Determinants of Export supply December 2005 Phil Alves Industry Concentration in South AWPS 96 African Manufacturing: Trends and December 2005 Gábor Szalontai Consequences, 1972-96 Johannes Fedderke The Urban Transition in Sub- AWPS 97 Saharan Africa: Implications for Economic Growth and Poverty December 2005 Christine Kessides Reduction 77 Africa Region Working Paper Series Series # Title Date Author Measuring Intergovernmental Fiscal Navin Girishankar AWPS 98 Performance in South Africa Issues in Municipal Grant May 2006 David DeGroot Monitoring T.V. Pillay 78