49691 PRACTICAL GUIDELINES for Analyzing Public Expenditure at the Sub-National Level Public Disclosure Authorized Public Disclosure Authorized Workshop edition The WORLD BANK OFFICE JAKARTA Indonesia Stock Exchange Building, Tower II/12-13th Fl. Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 The WORLD BANK The World Bank 1818 H Street N.W. Washington, D.C. 20433 USA Tel: (202) 458-1876 Fax: (202) 522-1557/1560 Email : feedback@worldbank.org Website : www.worldbank.org Printed in May 2010 Table of Contents Glossary of Terms ii Chapter 1. Introducing Public Expenditure Analysis 1 1.1 What is Public Expenditure Analysis (PEA)? 1 1.2 Why is the PEA useful? 1 1.3 Who will use the PEA? 2 1.4 What is included in the PEA? 2 Chapter 2. Doing a Public Expenditure Analysis 5 2.1 Writing the research proposal 5 2.2 Collecting the data 5 2.3 Entering the data into a formal that is useful for analysis 6 2.4 Analyzing the data, formulating conclusions and recommendations 6 2.5 Writing the report 6 Chapter 3. Research Proposal 7 Chapter 4. Data Collection and Preparation for Analysis 9 4.1 Collecting data 9 4.2 Preparing the data for analysis 10 Chapter 5. Data Analysis, Conclusions and Recommendations 13 5.1 Overview chapter 13 5.2 Planning and budgeting chapter 20 5.3 Revenue chapter 21 5.4 Expenditure chapter 32 5.5 Strategic sectors 39 Chapter 6. Writing the PEA Report 43 6.1 Know your audience 43 6.2 Common mistakes in report writing 43 6.3 Consistent format, clear structure 44 6.4 Writing the PEA report as a team 45 6.5 Further references 45 6.6 Extra tips : some lessons learned 45 APPENDICES Appendix 1: PEA Standard Outline 47 Appendix 2: Data Requirements 65 Glossary APBD Regional Government Budget (Anggaran Pendapatan dan Belanja Daerah) APBN State Budget (Anggaran Pendapatan Belanja Nasional) Bappeda Regional Development Planning Agency (Badan Perencanaan Pembangunan Daerah) Bappenas National Development Planning Agency (Badan Perencanaan Pembangunan Nasional) Bawasda Regional Monitoring Agency (Badan Pengawasan Daerah) BKD Regional Civil Service Agency (Badan Kepegawaian Daerah) BKN State Civil Service Agency (Badan Kepegawaian Negara) BMT Budget Master Table BPHTB Land and building transfer fee (Bea Perolehan Hak atas Tanah dan Bangunan) BPS Central Bureau of Statistics (Badan Pusat Statistik) BPS-SK Financial statistics from Central Bureau of Statistics (Statistik Keuangan Badan Pusat Statistik) Bupati District Head CPI Consumer Price Index CSO Civil Society Organization DAK Special Allocation Fund (Dana Alokasi Khusus) D&L Damage and Loss DAU General Allocation Fund (Dana Alokasi Umum) Desa Village Dinas Local Technical Agency Office DPRD Provincial House of Representatives (regional parliament) (Dewan Perwakilan Rakyat Daerah) GDP Gross Domestic Product GER Gross Enrollment Rate GoI Government of Indonesia GRDP Gross Regional Domestic Product HDI Human Development Index Kabupaten District (regency) Kecamatan Subdistrict Kelurahan Urban village Kepmen Ministerial Decree (Keputusan Menteri) Keppres Presidential Decision (Keputusan Presiden) Km Kilometer Kota City (urban district) LG Local Government MDG Millennium Development Goal MoF Ministry of Finance MoHA Ministry of Home Affairs MoNE Ministry of National Education MSS Minimum Service Standard NGO Non-Governmental Organization NR Natural Resources O&M Operations and Maintenance PAD Own-Source Revenue (Pendapatan Asli Daerah) PBB Land and Building Tax (Pajak Bumi dan Bangunan) PDAM Local Water Supply Utility (Perusahaan Daerah Air Minum) Perpu Regulation in Lieu of Law (Peraturan Pemerintah Penggati Undang-Undang) Perda Regional Regulation (Peraturan Daerah) ii Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level PFM Public Financial Management Podes BPS Village Potential Survey (Potensi Desa) Polindes Village Maternity Center (Pos Persalinan Desa) Puskesmas Community Health Center at Sub-district Level (Pusat Kesehatan Masyarakat) Posyandu Integrated Health Service Unit (Pusat Pelayanan Terpadu) Pustu Sub-community Health Center (Puskesmas Pembantu) Regional Budget Consolidated Budget consisting of Central Government Budget (Deconcentrated), Provincial Budget and District Budget. Renstra Ministry/Agency Medium-Term Strategic Plan (Rencana Strategis) RGDP Regional GDP RKPD Regional Government Work Plan (Rencana Kerja Pemerintah Daerah) RPJMD Regional Medium-Term Development Plan (rencana Pembangunan Jangka Menengah Daerah) Sakernas Labor Force Survey (Survei Tenaga Kerja Nasional) SDO Subsidy for Autonomous Region (Subsidi untuk Daerah Otonom) SIKD Regional Finance Information System (Sistem Informasi Keuangan Daerah) SKPD Regional Government's Working Unit (Satuan Kerja Pemerintah Daerah) SME Small/Medium Enterprise STR Student Teacher Ratio Sub-National Budget Consolidated Budget consisting of Provincial and District Budgets, but excluding Central Government. Susenas BPS National Socio-Economic Survey (Survei Sosial Ekonomi Nasional) TKD Regional Performance Bonus (Tunjangan Kinerja Daerah) WB World Bank Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level iii Chapter 1 Introducing Public Expenditure Analysis Chapter 1 Introducing Public Expenditure Analysis 1.1 What is Public Expenditure Analysis (PEA) Public expenditure analysis (or PEA) is a method of analyzing how governments allocate and manage their financial resources. The purpose of doing the analysis is to provide recommendations on how governments can manage public finances more efficiently and effectively in the future. 1.2 Why is the PEA useful? All governments have limited resources. This means there is a need to carefully decide how those financial resources will be allocated in order to achieve the maximum public benefits. However, in order to make these decisions, governments need accurate and timely information and analysis in order to answer key questions, such as: 1. How much money does the government have to spend? Where does the revenue come from? What is the potential, if any, for increasing the government's financing envelope? 2. What has the government spent its resources on previously? 3. What sort of public services have been provided with the current budget? Which sectors have good service provision and which sectors need improvement? 4. Who are the main beneficiaries of government spending? For example, is it the rich or the poor? Women or men? Rural or urban areas? Are the benefits spread evenly? Do the beneficiaries have equal access to services? Are there disadvantaged strata of the population who need special attention? 5. Have the services provided resulted in better human development outcomes for the population? For instance, have literacy rates, education levels, morbidity rates, life expectancy etc. improved? 6. How effective is the current planning and budgeting framework and process? Is the budget approved on time? Are the funds available in a timely manner? Are the planning priorities reflected in the budget? 7. What is the capacity of the civil service in public financial management? Are there areas that could be improved? If so, through what means could improvements be made? The answers to questions such as these will help governments to identify priorities that need to be tackled through government spending and help them make informed decisions on how best to allocate funds. 1.3 Who will use the PEA? Many widely differing groups of people will find PEA of use: Local Government (executive branch) The executive branch of the government, as the party responsible for initially setting the development priorities and allocating budget to meet development objectives, can use the analysis to help it in the allocating process. If Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 1 Chapter 1 Introducing Public Expenditure Analysis the PEA is conducted on a regular basis (e.g. annually or every two years), it can also be used as a monitoring tool to evaluate whether development targets have been met. The PEA report also identifies capacity development needs, providing a platform for various programs to support local governments. Local Government (legislative branch) The legislature is responsible for approving both the medium-term expenditure plan (RPJMD) and the budget (APBD). Therefore, it needs analysis to help in the evaluation of the submitted budget, especially to establish whether it is in accordance with the development objectives and priorities identified in the RPJMD. The PEA can also serve as a monitoring tool for the legislature in evaluating the performance of the executive. . Central Government The PEA report provides a foundation for the central government (GoI) to observe how local governments have spent their money and the extent to which local governments have implemented regulations and followed guidelines laid down in the regional finance and autonomy sectors. Lobbyists Individuals and organizations (for example, CBOs and CSOs) with specific interests and advocacy aims can use the PEA to help them in their advocacy and lobbying work. Advocacy policies are developed based on sound analytical work and the PEA serve as one source for this analytical work. Researchers and Academics Researchers can also use the PEA methodology as a framework for their own research. It is a method for analyzing how governments allocate their resources and can supplement their existing research. PEA reports can also be used as an information source for researchers and students interested in sub-national public finance issues. Other Donors and Local Government Programs Other donors and local government programs can use the capacity development needs identified in the PEA report as a basis to assist them in developing their own programs. Private Sector Entities from the private sector already with investments, or interested in investing in a certain province or kabupaten/kota, can use the PEA to increase their knowledge of the locality. The PEA is a source of general information on economic structure, sources of revenue, the local government's spending priorities, and public service delivery successes and challenges. The PEA also reflects local government's capacity, as well as the level of transparency in its activities. Public Publishing a PEA is one important way for governments to be transparent with regard to their use of public resources. A PEA presents financial information compiled from long and often complex budget documents in a format that is far easy to understand. Furthermore, a PEA contains a clear analysis of spending trends that the public can then use to hold politicians and governments accountable for their promises and responsibilities. 1.4 What is included in the PEA? PEA is a dynamic methodology that should be regularly revised to meet the needs of the users (section 1.3). However, as a minimum, any PEA should include all of the following: Chapter 1: Overview This chapter should introduce the reader to the geographic area researched. It should include information about its history, geography, population, government structure, economic structure and trends on important macroeconomic indicators, such as poverty level, growth rate, employment levels, human development index (HDI) and minimum service standards (MSS). 2 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 1 Introducing Public Expenditure Analysis Chapter 2: Planning and Budgeting This chapter should provide the reader with an overview of the planning and budgeting process at the national level and how the regional level contributes to the national framework. The analysis should include analysis on whether the sub-national government is following the national framework. Furthermore, the chapter should establish the level of consistency between planning documents (RPJMD, Renstra, RKPD etc) and whether the priorities set in those planning documents are reflected in the budget. Chapter 3: Revenue The purpose of this chapter is to provide a comprehensive picture of revenue at the sub-national level. This includes calculating the total revenue envelope, including an analysis of revenue-source trends. For example, does the sub- national government derive its revenue predominantly from own-sourced revenue (PAD) or transfers from central government? This chapter also includes calculations of the deficits and surpluses of the sub-national government, and its financing policy. Finally, this chapter includes analysis of how the sub-national government records its revenue and financing, and whether there are discrepancies. Chapter 4: Expenditure The purpose of this chapter is to provide a comprehensive picture of expenditure at the sub-national level. This includes calculating total expenditure and analyzing spending trends across time, sector, economic classification and by which level of government (province, kabupaten/kota or central governments). This chapter also looks at ability of the sub-government to absorb its budget by analyzing realization rates. The purpose of this chapter is to provide recommendations on how governments can improve the effectiveness and efficiency of public spending. Chapter 5: Strategic Sectors: Education, Health and Infrastructure The purpose of this chapter is to provide sharper analysis of those sectors most important for public service delivery: health, education and infrastructure. For each of these three sectors, this chapter analyzes spending and compares it with what the spending has achieved, both in terms of outputs such as staff, buildings and services, and also whether the spending has achieved an improvement in outcomes. For example, has health spending increased? If so, has this resulted in improved services (access, quality etc) and, if better services have been delivered, has this resulted in lower illness or mortality rates. Recommendations from this analysis may include identifying priority areas of health or education that the government should target, or how the government can improve its spending within a specific sector (e.g. more funds should be allocated to maintaining existing infrastructure rather than building new infrastructure). Attached in Appendix 1 is the PEA Minimum Standard Outline that provides the key chapters, sections and questions that all PEAs must include. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 3 Chapter 2 Doing a Public Expenditure Analysis Chapter 2 Doing a Public Expenditure Analysis The following briefly describes the main steps in conducting a PEA: 2.1 Writing the Research Proposal Before starting any research activities, it is important to have a comprehensive research proposal. The purpose of the research proposal is to set out what the research will be about, why it is important, who will do it, who are the target audiences, and the timing and scope of the research. This preparatory step is important in avoiding undertaking irrelevant work and, if working in a team, ensuring that there is a common understanding between team members. This step is important because defining the scope of research also defines the scope of data that will need to be collected. Chapter 3 provides further guidance on writing a research proposal. 2.2 Collecting the Data Once the proposal is completed, the researcher (or research team) can start the data collection. This is one of the most crucial stages of the research, as the quality of research is directly correlated to data quality. There are four important characteristics of data quality: data should be detailed, comparable, accurate and timely. The more detailed the data the better the analysis. If data are very broad and general, the researcher will experience difficulties in finding explanations for trends and drawing conclusions. In order to make comparisons across time, or with different kabupaten/kota, the data used must be comparable. For example, if the expenditure in the agriculture sector for kabupaten 1 includes fisheries but does not include fisheries in kabupaten 2, this means the agriculture information is not comparable unless the fisheries information is removed from kabupaten 1. It is important that the researcher examine the data carefully to ensure that they are comparable. Accuracy of data is also important, although this may not always be within the control of the researcher and may depend on the methodology used by those who responsible for primary data collection (e.g. BPS). For example, there are often many different versions of the APBD (pre-audit, post-audit, etc) and it is important that the data used are the most accurate available. There are two aspects to whether data are timely: first, conclusions on public finance can only be based on data from multiple years. A general rule of thumb is that data must be collected over a minimum of five years. The longer timeframe the data are available the better the analysis and its conclusions. Furthermore, in order for the research to be relevant, the most recent data are required. Gaining access to financial information is not always a smooth process and the researcher will need to be creative in finding different ways of ensuring that data collected are complete. Furthermore, the researcher may find inconsistent data from difference sources and will need to make judgments on which source is more appropriate (based on the principles of detailed, comparable, accurate and timely). Chapter 4, Section 4.1 will provide further guidance on the process of data collection. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 5 Chapter 2 Doing a Public Expenditure Analysis 2.3 Entering the data into a format that is useful for analysis Raw data (for example, found in the APBD or RPJMD) are often in a format unsuitable for analysis. Some common problems include: a) the data are only available in hardcopy; b) the data are not comparable because the budget format has changed; and c) the sectoral and economic classifications vary between governments. Therefore, in order to ensure consistency in the usage of data across the analysis, it is essential that data are entered into a format that allows the information to be compared and analyzed. This may include entering the information into a common table. This manual includes a suggested master table that all budget information should be entered into prior to analysis. Chapter 4, Section 4.2 provides further guidance on entering data into a format that is useful for analysis. 2.4 Analyzing the data, formulating conclusions and recommendations Once data collection has been completed and the data entered into one location and format, the researcher(s) are ready to conduct the analysis. While Chapter 5 sets out a variety of methodologies needed to analyze the data in order to write a PEA report, below are some broad suggested steps for conducting the data analysis: (a) Based on the available data, what trends can be seen? Has the indicator remained steady, increased or decreased? (b) Why has the trend happened? Is there a reason for the trend? Components (a) and (b) form the conclusion of the analysis? (c) Based on the reason explaining the trend, is there something that the government can do to improve the situation? This becomes the recommendations. Overall, the hallmark of good analysis is that the trends, conclusions and recommendations are linked to each other, and that these in turn stem from the available data. Further guidance on analyzing data is found in Chapter 5 2.5 Writing the report Based on the analysis and conclusions drawn in Step 4, the researcher(s) will consolidate the most important research and messages into a final written report. While strong analysis based on quality data is essential for the integrity and validity of the conclusions and recommendations, a well-written concise report is the medium by which the main messages are delivered to its audience. There are several characteristics to a well-written report: a) It has a clear and consistent structure. This manual has provided a general structure (see Section 1.4) that which most PEA reports to date have followed. While researchers should be free to add additional relevant material in order to make comparisons with other PEA research, this manual recommends that researchers remain faithful to the overall structure; b) all content included in the PEA report is relevant to its overall message; and c) sentences should be concise and free of jargon. Chapter 6 provides further guidance on writing a PEA report. 6 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 3 Research Proposal Chapter 3 Research Proposal Before starting data collection or conducting any analysis, it is important to have a comprehensive research proposal. The purpose of the research proposal is to set out what the research will be about, why it is important, who will do it, who are the target audiences, and the timing and scope of the research. This preparatory step is important in avoiding undertaking irrelevant work and, if working in a team, ensuring that there is a common understanding between team members. As a minimum, a PEA research proposal needs to cover the following: Background: What circumstances have led to the need for a PEA? Has anyone asked for the PEA? If so, who? Purpose of research: Outline what the PEA is trying to achieve. As mentioned in section 1, the main purpose of the PEA is to provide recommendations to government on how budgets can be allocated more effectively. Are there supplementary goals that the PEA is trying to achieve? For example, is it trying to build the capacity of team members to work in a team? Is it trying to collect budget and planning documents for university libraries or a database? Key audience: Identify who this research is primarily intended to reach and how the audience(s) is expected to use the research. Scope of research: This should include identifying the key research questions that need to be answered; an agreed time frame that this research will cover (e.g. years and data needed); geographic scope of the research (e.g. how many kabupaten/ kota will the PEA cover?); content scope of research (e.g. type of data required); an overview of the report structure, although this may change over time depending on the data collected and analysis conducted so the structure should remain flexible; and discuss generally the type of conclusions and recommendations expected. Timing: Provide a Gantt chart on the timing of the research, including key milestones (data collection completed, first draft completed etc). Research team ­ roles and responsibilities: This section will describe the research team composition and expected roles of team members. This section is not relevant if there will only be one researcher. Some suggested roles include: Team Leader: Responsible for ensuring that all team members understand their respective roles and responsibilities and are working as a team towards reaching a common goal. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 7 Chapter 3 Research Proposal Some tasks include: - Ensuring that the research proposal is complete and agreed upon by the team and related parties. - Overseeing the timely completion of tasks by each member of the research team. - Requesting (in reality, chasing) team members to submit their analysis and written material. - Compiling and collating each chapter written by different team members. - Editing all written materials. This task includes editing for consistency, style of language, structure/flow and logical progression between chapters, sections and sub-sections, and between sentences throughout the entire report. Researcher: This person will be responsible for conducting the portion of the research assigned to him/her by the team leader. This includes collecting data, analyzing the data and writing the relevant sections or chapters of the report. The researcher should also ensure that her/his work is consistent with the rest of the team, so that when the team puts together the final draft, the analysis is of similar structure, content and quality. The proposal should assign specific responsibilities to each researcher. Research Assistant: the researcher may choose to take on a research assistant to help her/him with any aspects of the research. The number of senior researchers and research assistants needed for a PEA team depends on the scope of research (time frame, geography etc). 8 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 4 Data Collection and Preparation for Analysis Chapter 4 Data Collection and Preparation for Analysis 4.1 Collecting data The purpose of this section is to explain to the researcher the data collection steps (what needs to be collected and from which sources), what issues the researcher may encounter and how these issues may be overcome. 4.1.1 Data requirements In order to complete a PEA, both quantitative and qualitative data are required. As discussed in Chapter 2, the more detailed and complete the data the better the analysis will be. For a full list of minimum data requirements, refer to Appendix 2. This section will provide a general overview of the data needed and their sources. Quantitative data: · Fiscal data: Data include APBN, APBD, DAU, DAK, borrowing, local taxes, revenue-sharing, and PAD. · Non-fiscal data: Data include population data; poverty rates; employment rates; education indicators such as the number of schools, teachers, students; enrollment rates; health indicators such as the number of staff and health centers; infrastructure indicators such as the length of roads, water and sanitation facilities available; the composition of DPRD; and the composition of civil servants. Qualitative data: · Planning process: Are the pre-requisite plans in place? Is planning participatory? How are the plans monitored? · Budgeting process: What is the level of public participation in the budgeting process? Is quantitative information used to make budget decisions? How are budget decisions made in specific sectors? Is the budget made available to the public? Which unit is responsible for disbursement? Has a local treasury been established? What are the payment mechanisms? Has performance budgeting been introduced? How is budget performance monitored? · Transfers: How many installment transfers does the local government receive from the central government and how timely are they? Data sources: Data may be collected from a variety of sources (and these may be inconsistent), but below is a suggested list of sources: · Ministry of Finance · Central Bureau of Statistics (BPS) · Ministries of Public Works, Health and Education · State Civil Servant Agency (Badan Kepegawaian Negara, or BKN) · Provincial governments (Finance Agency, Planning Agency, Health Agency, Education Agency, Public Works Agency) · Relevant agencies of the kabupaten/kota governments · World Bank and United Nations For a full list of minimum data requirements and sources, refer to Appendix 2. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 9 Chapter 4 Data Collection and Preparation for Analysis 4.1.2 Data collection issues During the process of collecting data from government agencies and other sources, the researcher may encounter many obstacles. This section outlines some common issues faced by researchers and suggests some solutions in overcoming those issues. This is not an exhaustive list of the potential issues and it will ultimately be up to researchers to find their own solutions. 1. No access to data. Researchers will often encounter obstacles in gaining access to data. Even where the provincial government has provided the requisite authorization papers, researchers will eventually run into sub-national government or agencies that are reluctant to make documents available. Suggested solution 1: Discuss the issue with the researchers' contacts in the government or donor agencies (if they are financing your research). For example, if the research is part of a larger donor-funded program, the program may have a program management committee or similar with representatives of the government sitting on this committee. This committee or related donor agencies (such as the World Bank) may be able to provide additional papers directly addressed to the particular government/agency reluctant to provide data access. Suggested solution 2: If solution 1 does not work, ask the researchers' contacts in the government or donors to intervene directly. For example, the program management committee or related donor may be able to contact the government agency directly themselves. They may be able to gain access to higher authorities at that particular region or, if possible, a representative of the committee may be able to accompany the researcher to the government agency themselves. Suggested solution 3: Find other sources for that particular data. Options include using close substitutes as proxies or using similar data from central (Jakarta) sources. 2. Data from different sources Data may be collected from sources that are inconsistent (i.e. showing different numbers) and it is possible that the data may differ significantly. For instance, the population data from sub-national BPS and Bappeda often contain large variations. Suggested solution: Gather more information on the different data, especially on the definitions, objectives, how the data are collected and the data availability. Based on these considerations the researcher can then decide which data are more accurate and relevant. 4.2 Preparing the data for analysis This section will discuss important aspects of preparing the data for analysis. First, this involves deciding on data characteristics prior to commencing analysis. In particular, it will discuss why it is important to reach agreement on the data and the analysis. Second, this chapter will explain the data table, what it aims to do and how to fill it out. 4.2.1 Agreeing on data sources, scope and analysis The aim of agreeing on data sources is to use a consistent data set throughout the analysis. This is especially important in cases where analysis in different sections is carried out by different researchers. The following issues need to be agreed: 10 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 4 Data Collection and Preparation for Analysis Source of data: central government versus sub-national governments Data can come from two sources: central government or sub-national governments. For example, the Ministry of Finance keeps sub-national government fiscal data in a database called the Sistem Informasi Keuangan Daerah (SIKD) and is more easily accessible through the SIKD web site. If the researchers are from the regions, they may have direct access to the fiscal data from sub-national governments. There are advantages and disadvantages from both sources. The advantage of central government sourced data is that the data are available for all regions and therefore comparable across provinces (and across kabupaten/kota). However, since the data are manually entered into the SIKD, they may be subject to human input error. Alternatively, data collected directly from sub-national governments may be more accurate. However, if the researcher has access to post-audited reports these may also contain more detailed information. If this information is available, it can be used for intra-regional comparisons (e.g. between provincial and kabupaten/kota governments within the same province) but should not be used for inter-regional analysis (e.g. between provinces or between kabupaten/kota in different provinces). This difference between central government and sub-national government sources applies to fiscal data, as well as some other quantitative output and outcome indicators. The research team needs to agree which set of data will be used in the entire PEA analysis. Source of population data: Population data are available from many sources ranging from the central BPS, regional BPS (provincial capitals) and different agencies within sub-national governments. Agreement on the data used is important because these data will generate per capita figures. As discussed in 3.1.3, where inconsistencies exist between different data sources it is for the researcher to decide which data set to use based on information concerning methodology, data availability, the definition of the data set etc. Time span: As discussed in the proposal stage, the time span for analysis needs to be agreed upon from the outset. As discussed in Chapter 1, the longer the time span, the better the analysis. However, there are significant limitations in data availability. For example, provinces established after decentralization, will not have data prior to 2000. Ideally, the minimum time span is five years. Also, where possible, the same years should be used for all data. One limitation is that the most recent years for outcome indicators (Susenas, Sakernas, health and education indicators) are often not available, so they may be of different time spans to the fiscal data (when more recent data are readily available). This is a limitation, but it remains important that the PEA analysis uses at least five years of data. Consumer Price Index (CPI): The research team needs to agree on the CPI because all nominal fiscal data must be converted to real figures (i.e. constant price figures) to take into account inflation. If the regional CPI is unavailable, agreement must be reached on which CPI figure to use as a proxy. Furthermore, researchers must agree on the base year for the real figures. Usually, the base year is the earliest (or oldest) year in the dataset. However, in order for the analysis to be forward looking, the authors recommend that researchers use the latest (i.e. most recent) year as the base year. The advantage of doing so is that when comparing trend figures and cross-section figures (which are the latest) the numbers will be consistent and comparable. This has proved to be most effective when the report is written by a team of researchers who work independently most of the time. 4.2.2 The Budget Master Table (BMT) Once agreement has been reached on data sources, the CPI and the time span, and the data collection has been completed, the next step is to input the data into a format that makes it accessible for analysis. As discussed in the research proposal section, the data must be comparable. Therefore, the raw data must be re-formatted so that the figures are comparable across years, geographic areas, sectors and economic classifications. There are two significant limitations in comparability. First, budget formats have changed several times during the past decade. Second, different governments may structure their government departments in different ways. For example, the scope of responsibilities for the Public Works Dinas may differ between districts. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 11 Chapter 4 Data Collection and Preparation for Analysis To help researchers re-structure the fiscal data into a comparable format, this manual is accompanied by a CD containing an Excel file called the "Budget Master Table" (hereafter referred to as the BMT). The BMT uses a format into which all fiscal data should be entered prior to any analysis. The objective of the BMT is to consolidate all sub-national governments' budget information into one location in a clear and consistent format, irrespective of different economic or sectoral classifications. Having all fiscal data in one source is particularly crucial if research is conducted by a team, as it will ensure that different researchers will use consistent figures in their respective analyses. The BMT Excel spreadsheet consists of four sections: 1. Read me: This section provides all the information regarding the data used in the BMT. This section also provides population and regional CPI data that is used to generate real and per capita figures. 2. Regional budget: This section covers the budgets from the provincial government and all the kabupaten/kota within the province. It covers the planned and realization figures for the time span agreed upon. 3. Central budget/deconcentrated spending: This section covers the central spending (deconcentrated funds) allocated to the province. Due to data limitations, the available data are usually the realization figures and broken down by sector. 4. Consolidated figures: This section provides the consolidated budget figures in nominal, real, and per capita forms. The BMT is created by manually entering the data from the collected budget reports from the sub-national government. Each cell consists of relevant items from the budget reports. At some point, assumptions/estimates can be made to any information that does not have a direct relation to the budget report. For example, if we convert prices into real prices, there should be a clear reference to which inflation factor has been used and which year is the base year. The important point is to document any assumptions or estimates so that this information is kept for future reference. Documents in CD: Example of Master table (see master_gPEA_final.xls), Exercise (see master_gPEA_exercise.xls) 12 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Chapter 5 Data Analysis, Conclusions and Recommendations This chapter discusses what types of analysis are required to write the PEA report. This includes identifying which charts are useful for each chapter and explaining which data are required to conduct the analysis and create the graphs. It also considers the types of conclusion that can be drawn from the analysis. This chapter is broken down into sections based on the minimum chapters expected in a PEA report, which include: 1. Overview · Regional economic analysis · Regional demographic analysis · Poverty analysis 2. Planning and budgeting · Development priorities analysis · Planning and budgeting linkages · Good practice for planning and budgeting · Participative development planning and budgeting · Regulatory framework analysis 3. Revenue · Overall revenue picture · Detailed revenue analysis · Financing analysis 4. Expenditure · Overall spending picture · Sectoral spending analysis · Economic Classification Analysis · Local spending variation analysis · Central government spending analysis 5. Strategic Sectors (education, health and infrastructure) · Analysis of spending by strategic sectors · Performance analysis for each strategic sector Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 13 Chapter 5 Data Analysis, Conclusions and Recommendations 5.1 Overview chapter 5.1.1 Objective The objective of this chapter in the PEA is to provide a general demographic background of the province, with a focus on social and economic analyses. If the researchers and/or sub-national government raise issues that are specific to the province (considered "local issues") then these should also be introduced in this chapter. 5.1.2 Type of analysis In this section, the following topics would usually be discussed: 1. Regional economic analysis a. Economic growth b. Structure of the economy c. Inflation 2. Regional demographic analysis a. Population and employment structure b. Unemployment 3. Poverty analysis a. Poverty rates b. Human Development Index 4. Regionaleconomicanalysis The regional economic analysis should look at the general macroeconomic conditions of the province, particularly in comparison with other provinces and the overall Indonesian position. There are three main elements to the economic analysis: economic growth, structure of the economy and inflation. a. Economic growth (regional gross domestic product (RGDP) growth, RGDP per capita). The objective is to understand the size of the provincial/district economy as well as the rate at which the sub national economy has grown, especially in comparison to other provinces, districts in Indonesia and with the national average. Source of data: Regional Statistical Office and the Regional Planning Agency. Example 1: making provincial comparison The next bar graph is an example of how the size of the regional economy can be presented. It shows per capita RGDP at the Gorontalo province, in comparison with the other provinces in Sulawesi, in eastern Indonesia and across Indonesia. The graph suggests two trends: 1) Gorontalo has a very small economy, as measured by the per capita RGDP, compared to other provinces in Indonesia; and 2) While the Gorontalo economy has grown between 2000-2006, this follows a general trend across Indonesia. Per capita real RGDP is the preferred indicator because when using total RGDP the population is excluded, yet it contributes to the making of RGDP. 14 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Figure 1. Economic growth in Gorontalo, comparison with provincial and national level, 2000-2005 10 Real GRDP per capita (Rp million) 8 6 4 2 0 2000 2001 2002 2003 2004 2005 National Eastern Indonesia Sulawesi Gorontalo 14 12 10 Real GRDP growth (%) 8 6 4 2 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 -2 -4 -6 National Eastern Indonesia Sulawesi Gorontalo The above line graph shows the real economic growth, as measured by change in RGDP of Gorontalo between 1994 and 2005, compared to other provinces. The graph suggests that until 2005 Gorontalo grew faster than the average of other provinces in Sulawesi, eastern Indonesia and nationally. In 2005, however, eastern Indonesia's growth rate suddenly rose to over nine percent (up from less than one percent in 2004), and overtook Gorontalo province's growth rate. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 15 Chapter 5 Data Analysis, Conclusions and Recommendations Further issues to investigate: The key next step after identifying trends is to investigate why these trends may have happened. Some typical follow-up questions might include: · What has been causing the per capita RGDP to increase in Gorontalo and generally across all provinces in Indonesia? · What are drivers of Gorontalo's economic growth? · Why did eastern Indonesian provinces suddenly experience such significant growth in 2005? b. Structure of the economy (sectoral RGDP, sectoral per capita RGDP) The purpose is to analyze which sectors drive the sub-national (provincial or distric / city) economy. This analysis can compare one province/distric/city across time or compare against different sub-national governments. Source of data: Regional Statistics Agency and Planning Agency. Example 2: sectoral contributions The example in Figure 2 shows which sectors contribute to the sub-national economy. The pie graph below shows that agriculture represents the largest sector for Nias Islands (consisting of two districs), followed by trade, restaurants and hotels. The Islands have very little activity in the manufacturing, mining and electricity, gas and water sectors (combined 4.5 percent). Figure 2. Sectoral composition for GRDP in Nias islands, 2005 Financial Services Services 5.5% 10.1% Agriculture Transport & 43.0% Communication 6.9% Trade, Mining and Restaurant, & Quarrying Hotel 2.3% Manufacturing 22.1% Construction Industry 7.9% Electricity, 1.8% Gas, & Water 0.4% Figure 3 shows the per capita RGDP of Gorontalo between 2000 and 2005, in comparison to real wages. The graph shows several trends: 1) (i) Agriculture forms a significant portion of Gorontalo's per capita RGDP throughout all years. (ii) The overall sectoral composition of RGDP has remained fairly consistent although the percentage of RGDP from electricity, water and gas has increased. (iii) Although per capita RGDP has increased, real wages have decreased in 2005 and 2006. 16 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Figure 3. Sectoral composition for GRDP per capita in Gorontalo, 2000 -2006 2,500 1,000 855 879 Rp bn (constant 2000 prices) 783 Agricultural Work force (%) 754 2,000 689 750 GRDP per capita, 639 1,500 500 1,000 250 500 - - 2000 2001 2002 2003 2004 2005 2006 Agriculture Mining and Quarrying Manufacturing Electricity, Gas & Water Supply Construction Trade, Restaurant & Hotel Transportation and Communication Financial Service Services real wage Further questions for consideration: Some typical follow-up questions might include: · Why have real wages decreased? · Why have the electricity, gas and water sector increased? d. Inflation The objective is to provide an overall picture on inflation levels in the sub-national region. Inflation levels can be compared across years, between districts / cities, between provinces or with regional / national figures. Figure 4 Inflation rate for Gorontalo, comparison with provincial and national level, 2004-2007 25 National Manado 20 Makassar Gorontalo 15 10 5 0 4 4 4 4 4 4 5 5 5 5 5 5 6 6 6 6 6 6 7 7 7 7 -0 r-0 -0 l-0 -0 -0 -0 r-0 -0 l-0 -0 -0 -0 r-0 -0 l-0 -0 -0 -0 r-0 -0 l-0 Jan Ma ay Ju Sep Nov Jan Ma ay Ju Sep Nov Jan Ma ay Ju Sep Nov Jan Ma ay Ju M M M M Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 17 Chapter 5 Data Analysis, Conclusions and Recommendations Figure 4 provides one example of how inflation analysis can be presented. The graph suggests several trends: (i) Overall, inflation in Gorontalo has followed the trends for the rest of the country. (ii) Although consistent with the rest of Indonesia, Gorontalo seems to experience more extreme highs and lows, as shown by the sharp spikes in the line graph. (iii) Since mid-2007, inflation in Gorontalo has been lower than the rest of Indonesia. Further questions for consideration: Some typical follow-up questions might include: · Why does Gorontalo experience more extreme highs and lows in inflation rate? · Is there a reason why inflation changes in Indonesia more strongly affects Gorontalo? · What drives the price movement in Gorontalo (food, service sector, construction material, etc)? 5. Regionaldemographicanalysis The purpose of the regional demographic analysis is to look at the population and employment conditions, with a focus on the following issues: a. Population and employment structure The purpose of this analysis is to determine which are the key sectors that absorb employment and analyze the education level of those employed. This analysis can compare districts / cities, as well as with provincial and national figures. Source of data: Regional Statistical Office and the Regional Planning Agency. Example 3: employment analysis Figure 5 shows which sectors contribute the most jobs in Gorontalo between 2001 and 2006. The graph shows two notable trends: 1) the majority of jobs in Gorontalo are in the agricultural sector; and 2) although still by far the largest driver of employment, the proportion of jobs in the agricultural sector has decreased from 63 percent in 2001 to 55 percent in 2006. Figure 5. Employment at the sectoral level in Gorontalo, 2001-2006 400 100 Workforce ('000) 300 75 Agricultural Workforce (%) Others (L) 63 56 62 55 Manufacturing (L) 48 51 Public/social/personal service (L) 200 50 Trade (L) Agriculture (L) Agriculture workforce (%) (R) 100 25 0 0 2001 2002 2003 2004 2005 2006 18 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Further questions for consideration: Some typical follow-up questions might include: · Is the decrease in the proportion of agricultural workforce significant? · If so, what caused this decrease? · What sectors have grown in terms of size of workforce? · Why have those sectors increased? b. Unemployment rates The purpose of this analysis is to evaluate the sub-national unemployment rate with other comparisons such as the national average, rates in other similar provinces and/or to other districts / cities. Sources of data include the Regional Statistical Office and the Regional Planning Agency. Example 4: unemployment rates in Gorontalo Figure 6 compares Gorontalo's unemployment rate to the national level and the average for the provinces in eastern Indonesia. The graph suggests two trends: (i) Between 2001 and 2006, unemployment in Gorontalo decreased although unemployment in both eastern Indonesian and Indonesia-wide increased. (ii) Unemployment in Gorontalo experienced much larger fluctuations between years than compared to eastern Indonesian and Indonesia. Figure 6 . Unemployment level in Gorontalo, comparison with national level, 2001-2006 16 National Gorontalo Eastern Unemployment rate (%) 14 Indonesia 12 10 8 6 4 2 0 2001 2002 2003 2004 2005 2006 Further questions for consideration: Some typical follow-up questions might include: · Why has unemployment decreased in Gorontalo while the overall trend in both eastern Indonesia and Indonesia has increased? · Why does the Gorontalo's unemployment rate fluctuate so much between years? 6. Povertyanalysis a. Poverty Rate This analysis should investigate poverty trends over time, and compared to other provinces / districts / cities and the national poverty levels. The purpose of this analysis is to determine whether poverty levels in the province / district / city has increased or decreased during the years analyzed. Furthermore, it is important to determine whether trends identified in the province / district / city are consistent with trends in other provinces or in comparison with the poverty level in Indonesia. Source of data: Regional Statistical Office. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 19 Chapter 5 Data Analysis, Conclusions and Recommendations b. Human Development Index This analysis looks at poverty trends over time, and compared to other provinces / districts / cities and the national poverty levels. The purpose of this analysis is to determine whether Human Development Indicators (HDI) in the province / district / city has increased or decreased during the years analyzed. Furthermore, it is important to determine whether the trend in the province / district / city is consistent with trends in other provinces or in comparison with the HDI trends in Indonesia. Analysis in HDI could also incorporate the drivers of HDI (life expectancy, literacy rate, mean years in schooling, or per capita income) in the respective provinces; therefore a breakdown of data could add more insight in this analysis. Source of data: UNDP Human Development Report. Example 5: poverty rates and the human development index Table 1 shows how both poverty and HDI levels can be presented. The poverty rate in the table suggests several trends: 1) the poverty rate in Gorontalo province is 28.9 percent, which is much higher than the national poverty rate of 16.7 percent; 2) HDI level in Gorontalo is similar to the national HDI. Table 1 Poverty rate and HDI in Gorontalo, comparison with national level Indicator Gorontalo National Highest Lowest Average Highest Lowest Average GRDP per capita (Rp million, 2005) 5.2 2.8 3.7 407.3 0.9 12.6 Poverty rate (2004, %) 32.5 10.8 28.9 51.0 2.9 16.7 Adult literacy rate (2006, %) 83.1 60.0 70.8 99.3 7.3 71.5 Secondary net enrollment rate (2006, %) 67.9 39.3 46.2 90.4 8.2 57.8 HDI (2005) 70.4 65.9 67.5 77.4 46.9 69.6 Further questions for consideration: Some typical follow-up questions might include: · Has the poverty and HDI levels in Gorontalo changed over the years? · Has it increased or decreased? · If changed, what are some factors that caused that change? · Has the change in poverty and HDI levels changed over years? · Is the trend consistent with the national, or eastern Indonesia trends? If there is a difference, it is important to explain why there is a difference. 5.2 Planning and budgeting chapter 5.2.1 Objective The aim of this chapter in the PEA is to better understand regional development planning and the relationship to the budgeting process. The basic documents required to conduct the analysis are mainly budget documents (APBD) that can be obtained from budget offices and planning documents from relevant Bappeda offices. 5.2.2 Types of analysis required In this section, the following topics would usually be discussed: 1. Developmentprioritiesanalysis 2. Planningandbudgetinglinkages 3. Goodpracticeforplanningandbudgeting 4. Participativedevelopmentplanningandbudgeting 5. Regulatoryframeworkanalysis 20 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations 1. Developmentprioritiesanalysis In order to describe provincial development priorities, this section requires a discussion of the priority sectors mentioned in the provincial RPJMD (Regional Medium-Term Development Plan), a brief description of main function (urusan wajib) and optional function (urusan pilihan) of sub-national governments, as well as an assessment of the level of linkages and synchronization between various planning documents. 2. Planningandbudgetinglinkages This section looks at the function of the budget as a tool of development planning and how it is used in establishing development priorities. It assesses whether the existing budget (APBD) is really linked to the planning documents and whether the budget contains the development goals as stated in the planning documents. 3. Goodpracticeforplanningandbudgeting This section reviews whether the existing planning and budgeting documents have taken into account Medium- Term Expenditure Framework (MTEF) and Performance-Based Budgeting (PBB) principles. It aims to establish how well the concepts of MTEF and PBB are applied in the current budget system. 4. Participativedevelopmentplanningandbudgeting Here, we look at how well the region has applied a participative process in development planning, the benefits and drawbacks of using such process. 5. Regulatoryframeworkanalysis This section examines whether the process of planning and budgeting has been conducted according to existing regulations and has occurred within the required timeframe. 5.3 Revenue chapter 5.3.1 Objective The aim of this chapter is to understand the total financing envelope that is available to the region and the sources of that revenue, including the extent to which this comes from national or local governments. As such, the required figures are consolidated from all relevant districts / cities, the provincial government and Central government (deconcentrated funds and other assistance (tugas pembantuan)). The basic documents required to conduct the analysis are mainly budget documents (APBD) obtained from the budget offices or Bappeda. 5.3.2 Types of analysis required In this section, the following topics would usually be discussed in order to provide a comprehensive analysis of sub-national revenue: 1. Overallrevenuepicture The purpose of this analysis is to provide a consolidated picture of the general revenue trends in the province. This analysis must include revenue data from both the provincial and district / city governments. 2. Detailedrevenueanalysis The second section provides a more detailed analysis of the main sources of revenue (DAU, DAK, Shared Revenue and own-source revenue (PAD)). This includes for each type of revenue: trends over time, whether derived by provincial or district / city governments; and what factors cause each type of revenue to increase or decrease. 3. Financinganalysis This section aims to understand the level of surplus/deficit that the sub-national government has; and how the provincial / district / city government(s) finances the deficit or allocates the surplus; as well as how the accounts are recorded in the APBD. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 21 Chapter 5 Data Analysis, Conclusions and Recommendations 5.3.3 Overall revenue picture The purpose of this analysis is to provide an overall picture of how much consolidated revenue the region receives. Part of this general analysis is to compare the province's consolidated revenue with other comparable provinces in Indonesia, and with the Indonesian average. Key aspects of this analysis should include: · Trends over time: has total revenue increased or decreased over time? Why? How does the trend com- pare to other provinces in Indonesia or the Indonesian average? To compare with other provinces, the analysis must use per capita figures. · Level of government: which level of government receives the sub-national revenue? Provincial govern- ment or district / city governments? Has this changed over time? · Distribution across districts / cities: which district / city receives the highest levels of revenue? Which receives the lowest level of revenue? Why? To compare between districts / cities, the analysis must use per capita figures. · Composition of revenue: what are the main sources of revenue ­ DAU, DAK, Shared Revenue, PAD? Has the composition of revenue changed over time? Example 6: trends across time Figure 7 is an example of a time series graph that shows the revenue trends across time in Gorontalo. Figure 7. Fiscal resources per capita in Gorontalo, 1998-2006 1,800 1,600 Rp '000 1,400 1,200 1,000 800 600 400 200 0 1998 1999 2000 2001 2002 2003 2004 2005 2006* Own Source Revenue Shared Taxes Shared Nat Resources Autonomous Region Subsidy/SDO (Development) Presidential Grants/INPRES DAU DAK Other Balancing Fund Other Revenue The graph suggests several observable trends: (i) Since 2001, Gorontalo's per capita revenue has increased steadily. (ii) Since 2001, the sources of revenue have changed significantly. The majority of the revenue is from fiscal balancing funds, which comprises of DAU, DAK and Shared Revenue. (iii) The majority of the increase in revenue is from the balancing funds, which was the main tool used by the central government to transfer fiscal management to local governments, as a part of Indonesia's decentralization reforms. 22 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Further questions for consideration: Some typical follow-up questions might include: · Why has per capita revenue increased? · Is the growth in Gorontalo's revenue higher or lower than revenue growth in other provinces in Indone- sia? Why might this be? · What recommendations might help sub-national governments to increase revenue? Example 7: composition of revenue Table 2. Fiscal revenue composition in Gorontalo and consolidated districts, 2002-2006 2002 2003 2004 2005 2006* Province Rp bn % Rp bn % Rp bn % Rp bn % Rp bn % DAU 157.05 81 210.55 79 231.71 82 211.01 78 340.80 89 Own-Source Revenue 31.08 16 41.94 16 41.59 15 46.11 17 40.70 11 Shared Tax Revenue 6.04 3 7.23 3 7.58 3 12.99 5 3.57 1 DAK 0.00 0 5.61 2 0.00 0 0.00 0 0.00 0 Shared Natural Resource Revenue 0.46 0 0.32 0 0.72 0 0.08 0 0.00 0 Other Revenue 0.00 0 0.25 0 0.00 0 0.04 0 0.00 0 Total 194.63 100 265.90 100 281.60 100 270.24 100 385.07 100 2002 2003 2004 2005 2006* Districts/Cities Rp bn % Rp bn % Rp bn % Rp bn % Rp bn % DAU 478.46 82 558.80 78 559.96 74 569.46 72 825.53 77 DAK 13.66 2 26.94 4 48.60 6 56.74 7 103.00 10 Own-Source Revenue 31.13 5 48.70 7 49.32 6 45.91 6 52.80 5 Shared Tax Revenue 37.53 6 29.76 4 53.55 7 57.29 7 57.92 5 Shared Tax and Assistance from Province 0.00 0 10.40 1 11.88 2 12.17 2 11.01 1 Other Revenue 11.37 2 24.05 3 23.14 3 34.48 4 14.58 1 Shared Natural Resource Revenue 3.57 1 1.85 0 3.26 0 3.59 0 3.19 0 Adjustment Fund 4.26 1 20.31 3 12.08 2 10.90 1 0.00 0 Total 579.98 100 720.80 100 761.80 100 790.53 100 1,068.04 100 Table 2 presents revenue in Gorontalo by year and by composition, and separated between the provincial Government and its district / city governments. The table shows several observable trends: (i) Between 2002 and 2006, provincial revenue and district / city revenue almost doubled. (ii) The vast majority of the revenue at both the provincial and district / city level, is from DAU. (iii) At the district / city level, DAK has increased by almost 600 percent from 2002 to 2006. (iv) Very little revenue is from PAD, which means the province predominantly relies on transfers from central government. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 23 Chapter 5 Data Analysis, Conclusions and Recommendations Further questions for consideration: Some typical follow-up questions might include: · Why has the above trends happened? · How does revenue growth in Gorontalo compare with other provincial / district / city governments in Indonesia? · What factors affect the increase in DAK and DAU? · Are these factors different in Gorontalo compared to other local governments in Indonesia? Example 8: revenue composition by district / city Figure 8. Fiscal revenue composition in Gorontalo and individual districts, 2005 1,400 1,200 1,000 800 Rp'000 600 400 200 0 Gorontalo Bone Bolango Boalemo Pohuwato Kota Gorontalo Own Source Revenue Shared Tax Revenue Shared Natural Resource Revenue DAU DAK Shared Tax Revenue and Assistance from Prov. Adjustment Fund Other Revenue The above graph compares the revenue composition in the district / city governments in Gorontalo. The graph suggests several trends: (i) All district / city governments in Gorontalo derive most of their revenue from DAU. (ii) Several districts, including Bone Bolango, Baolemo and Pohuwato, received significant DAK transfers compared with the remaining two districts. (iii) Kota Gorontalo has the highest level of PAD. Further questions for consideration: Some typical follow-up questions might include: · Why did the three districts receive higher levels of DAK compared to Kota Gorontalo or Kabupaten Gorontalo? · Why does Kota Gorontalo has such a high level of PAD compared to the districts in Gorontalo? Example 9: comparison with other sub-national governments Figure 9 compares the per capita consolidated sub-national revenue in Gorontalo with other provinces in Indonesia. The other provinces selected are all provinces that were established in 1999, as part of Indonesia's decentralization reforms. The graph shows that Gorontalo's fiscal capacity is one of the weakest, amongst the newly-established provinces. 24 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Figure 9. Comparison of fiscal revenue resources of Gorontalo sub-national governments with other regions, 2005 Transfer per capita (L) Others per capita (L) 4,500 PAD per capita (L) 6,000 Sub-national per capita revenues (Rp `000) Total Sub national revenue (R) 4,000 5,000 Total Revenue (Rp. bn) 3,500 3,000 4,000 2,500 3,000 2,000 1,500 2,000 1,000 1,000 500 - n o at at ng ra u al te ia ta ar rBa u nt .R n U iB lit Ba o p ua u Be es or Ke uk p w G a Pa al la gk M Su an .B p Ke Further questions for consideration: A typical follow-up question might be: · Why is Gorontalo's per capita revenue lower than other newly-established provinces? 5.3.4 Detailed revenue analysis The purpose of this section is to analyze in more detail each source of revenue and how each type impacts on the overall revenue levels in the province. Regional governments derive revenue from three main sources: a. Own-source revenue (Pendapatan Asli Daerah) (PAD) This is revenue collected by sub-national governments and may include local taxes, fees and investment income. b. Balancing funds These funds are used by the central government to balance the fiscal capacities of sub-national governments: (i) Dana Alokasi Umum (DAU) or general allocation is a discretionary block grant transferred from the National government to all sub-national governments to achieve fiscal equality. DAU is allocated using a national formula based on factors such as population, regional area, per capita gross regional domestic product (RGDP), Human Development Index, civil servant salary budget and spending, the level of own- source revenue (PAD) and shared revenue (bagi hasil). (ii) DanaAlokasiKhusus(DAK) or special allocation is an earmarked cash grant transferred from the central government to sub-national governments to finance specific needs of the region. (iii) BagiHasilor shared revenue (from tax and natural resources) is revenue generated from taxes (national level) and natural sources that are shared between central government and the sub-national governments based on an agreed ratio found in legislation. d. Other revenues These include grant revenue, contingency funds, special autonomy funds (for example in Aceh and Papua), and other sharing revenue/financial assistance from other provincial or district governments. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 25 Chapter 5 Data Analysis, Conclusions and Recommendations For each type of revenue, this section should also examine the following issues: · Trends over time: has total revenue increased or decreased over time? Why might this be? How does the trend compare to other provinces in Indonesia or the Indonesian average? To compare with other prov- inces, the analysis must use per capita figures. · Level of government: which level of government receives the sub-national revenue? Provincial govern- ment or district / city governments? Has this changed over time? · Distribution across district / city: which district / city receives the highest levels of revenue? Which receives the lowest level of revenue? Why is there a difference? To compare between district / city, the analysis must use per capita figures. · Planned v realized revenues: did the sub-national government realized the revenue it had planned to collect? · Composition of revenue: for each type of revenue, what are the main factors that cause the increase or decrease of that revenue? How does the trend compare to other provinces in Indonesia? Example 10: revenue trends over time Figure 10. Trend of DAU per capita in Gorontalo, 2006 1000 900 888 800 DAU per capita (Rp'000) 700 600 626 633 624 500 561 400 367 300 200 239 258 232 184 100 0 2002 20032 004 2005 2006 Province District / City The above is a simple way of showing the growth in DAU in Gorontalo between 2002 and 2006. The graph suggests that although the DAU grew slightly between 2002 and 2005, the most significant increase occurred in 2006. Further questions for consideration: Some typical follow-up questions might include: · What factors caused the significant increase in 2006? · How does this increase compare to other provinces in Indonesia? Example 11: revenue trends across districts / cities Figure 11 shows how the DAU was distributed across districts in Gorontalo relative to their fiscal gaps-per-capita and civil service wage bills. 26 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Figure 11. Distribution of DAU per capita in Gorontalo sub-national governments, relative to fiscal gap and wage bill 0 3. 0 2. 8 1. 5 2. 6 1. 4 1. 0 2. Rp million Rp million 2 1. 1. 5 0 1. 8 0. 0 1. 6 0. 0. 5 4 0. 2 0. 0 0. 0 Gorontalo Boalemo Bone Bolango Kota Gorontalo Pohuwato DAU 2005 PC DAU 2006 PC Wage Bill PC Fiscal Gap PC This figure suggests that the DAU allocation seems to be correlated with the size of a district's wage bill. However, in some cases (for example, Kabupaten Pohuwato) the fiscal gap component compensates for a low wage bill. Fiscal gap is the difference between the expenditure needs of the district and the fiscal capacity of the district. Further questions for consideration: A typical follow-up question might be: · What are the factors that influence the size of the fiscal gap? For example, in Gorontalo, the fiscal gap in Kabupten Pohuwato is high because the district government provides services to a large land area; and because the cost of construction is high. Example 12: DAK allocation by sector Figure 12 shows the sectoral focus on DAK allocations. It suggests that in 2006, the majority of DAK was allocated to education, health and roads. Figure 12. DAK allocations by sector in Gorontalo, 2006 Government Environment 4% 1% Agriculture 10% Education Fisheries 23% 5% Water supply and sanitation 5% Irrigation 5% Health 23% Roads 24% Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 27 Chapter 5 Data Analysis, Conclusions and Recommendations Further questions for consideration: A typical follow-up question might be: · What was the DAK used to finance (eg classrooms, health facilities, training, road infrastructure)? One or two examples of DAK activities from district or provincial government should be provided. Example 13: planned vs realized revenue Table 3. Realization rate of shared revenue, Gorontalo, 2002-05 2002 2003 2004 2005 Province Shared tax revenue 132.24 111.76 109.88 249.36 Shared natural-resource (non-tax) revenue 33.34 23.94 82.69 n.a. Districts/Cities Shared tax revenue 121.94 116.03 103.98 89.61 Shared natural-resource (non-tax) revenue 84.75 60.96 151.06 247.78 The above table shows that Gorontalo exceeded its expectations and collected more shared tax revenue than expected, but generally did not meet expectations on non-tax shared revenue (natural resources). However, in 2004 and 2005, the district / city governments collected significantly more non-tax shared revenue than planned. Further questions for consideration: A typical follow-up question might be: · Why were some estimates of planned revenue under-estimated and others over-estimated? Example 14: composition of own-source revenue Table 4. Own-source revenue composition, Gorontalo, 2002-2006 2002 2003 2004 2005 2006* Province Rp bn % Rp bn % Rp bn % Rp bn % Rp bn % Tax 23.84 77 29.29 70 33.45 80 37.41 81 37.48 92 User charges 1.04 3 5.46 13 1.50 4 2.80 6 0.00 0 Profit from reg-owned 0.00 0 0.18 0 0.28 1 0.29 1 1.00 2 enterprises Other own-source revenue 6.20 20 7.01 17 6.37 15 5.61 12 2.22 5 Total own-source revenue 31.08 100 41.94 100 41.59 100 46.11 100 40.70 100 2002 2003 2004 2005 2006* Districts/Cities Rp bn % Rp bn % Rp bn % Rp bn % Rp bn % Tax 4.82 15 6.36 13 7.88 16 7.31 16 7.34 15 User charges 15.24 49 17.67 36 29.75 60 26.44 58 26.86 55 Profit from reg-owned 0.35 1 1.14 2 2.12 4 3.48 8 3.02 6 enterprises Other own-source revenue 10.71 34 23.53 48 9.57 19 8.67 19 11.23 23 Total own-source revenue 31.13 100 48.70 100 49.32 100 45.91 100 48.45 100 28 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations The above table, showing the composition of own-source revenue, suggests several trends: (i) Between 2002 and 2006, own-source revenue increased slightly, but still only represents a very small percentage of total sub-national revenue. (ii) The provincial government collects a similar amount of own-source revenue as the district / city governments. (iii) At the province level, the majority of own-source revenue is derived from taxes. In contrast, at the district / city level, over half of the own source revenue is from user charges. Further questions for consideration: Some typical follow-up questions might include: · What has caused the slight increase in PAD? How does the growth in PAD in this province compare to other provinces in Indonesia? · Why are the majority of taxes collected by the province, not the district / city governments? What is the legal basis for collecting those taxes? Can the tax collection be improved by becoming more efficient or effective? · What taxes/user charges are most commonly found in the province? · What types of services are the district / city governments providing in which they are collecting revenue? How can the district / city government increase its own-source revenue? Example 15: comparison of PAD across districts and cities Figure 13. Own-source revenue per capita, district comparison, Gorontalo, 2005 160 140 120 100 Rp'000 80 60 40 20 0 Bone Bolango Gorontalo Boalemo Pohuwato Province Kota Gorontalo Own-source revenue per capita The above graph shows the level of own-source revenue collected by the sub-national governments in Gorontalo and suggests that Kota Gorontalo has a significantly higher level of PAD than other districts and the provincial government. Further questions for consideration: A typical follow-up question might be: · Why is PAD highest in Kota Gorontalo? An explanation should be provided, such as: This may be because Kota Gorontalo is a city with a larger population and more economic activity than its rural districts, from which the government can impose user charges for a variety of services such as health services, rubbish collection, registration of motor vehicles, renting space in markets etc. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 29 Chapter 5 Data Analysis, Conclusions and Recommendations Example 16: comparison of taxes and user charges between districts Table 5. Taxes and user charges, district level, Gorontalo, 2005 Kota Gorontalo Boalemo Taxes Rp million % Taxes Rp million % Street lighting tax 2,526.77 69.31 Street lighting tax 291.30 50.62 Hotel and restaurant tax 500.67 13.73 Tax on mining of c-class materials 175.33 30.47 Billboard tax 341.17 9.36 Billboard tax 51.60 8.97 Tax on mining of c-class materials 216.23 5.93 Hotel and restaurant tax 37.99 6.60 Total Tax 3,645.79 Total Tax 575.43 User Charges Rp million % User Charges Rp million % Health 12,085.41 73.17 Administrative services 257.93 29.9 Building permits 1,154.23 6.99 Use of region's assets 257.07 29.8 Use of region's assets 572.24 3.46 Health 149.59 17.3 Jasa Pemberi Pekerjaan 401.77 2.43 Market 102.61 11.9 Market 354.03 2.14 Building permits 95.76 11.1 Total User Charges 16,516.43 Total User Charges 1,670.73 Table 5 provides an example of the different components of taxes and user charges in Kota Gorontalo and Kabupaten Baolemo. It suggests several conclusions: (i) Kota Gorontalo and Kabupaten Baolemo's taxes and user charges are consistent with the taxes and user charges listed in Law No. 34/2000. (ii) Entertainment taxes and parking fees are generally among the highest taxes and user charges, particularly in urban areas. However, this is not the case in Kota Gorontalo. Further questions for consideration: A typical follow-up question might be: · Why are entertainment taxes and parking fees not the main form of taxes and user charges collected by the district / city governments? A possible explanation: such taxes and user charges are less viable given that businesses and tourism remain relatively under-developed in Gorontalo and the number of cars is still limited. Only as the economy grows will opportunities increase for Kota Gorontalo to mobilize more significant own-source revenue. 30 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Example 17: planned vs realized revenue Table 6. Own-source revenue comparisons, provincial and consolidated districts comparison, Gorontalo, 2002- 2005 Percent realization rate Province 2002 2003 2004 2005 Tax 116.30 86.42 96.79 99.40 User charges 115.46 69.45 87.35 103.58 Profit from reg-owned enterprises 0.00 131.99 85.21 98.53 Other own-source revenue 89.19 148.05 67.47 128.53 Total own-source revenue 109.63 89.95 90.35 102.47 Districts/Cities 2002 2003 2004 2005 Tax 105.14 100.55 105.58 105.00 User charges 99.16 92.80 95.38 102.78 Profit from reg-owned enterprises 56.38 165.53 58.25 35.19 Other own-source revenue 102.33 126.21 73.99 58.63 Total own-source revenue 100.25 108.95 89.30 81.97 Table 6 shows whether planned own-source revenue is similar to the revenue levels actually realized by the sub- national governments. It suggests two trends: (i) There was no clear pattern in own-source revenue realization over the period between 2002 and 2005. An over-realization in one year is often followed by an under-realization in the following year. For example, the profits from locally owned enterprises and "other" own-source revenue in 2003 and 2004, both at the provincial and district / city levels fell significantly in the years after. (ii) Tax collection, however, was generally carried out as planned. With the exception of provincial tax collection in 2003, tax collection by the province and district / city diverged from planned amounts by less than 10 percent. Districts recorded a better realization rate than the province for user charges. Further questions for consideration: A typical follow-up question might be: · What is the method for estimating own-source revenue? Can this methodology be improved to ensure more accurate planning? 5.3.5 Financing Analysis This section aims to understand the level of surplus/deficit that the sub-national government has; and how the province / district / city government(s) finances the deficit or allocates the surplus; as well as how the accounts are recorded in the APBD. Example 18: provincial financial performance Figure 14 shows whether the provincial / district / city governments in Gorontalo posted deficits or surpluses, and we can see that the governments consistently posted surpluses between 2002 and 2005. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 31 Chapter 5 Data Analysis, Conclusions and Recommendations Figure 14. Surpluses and deficits, comparison among sub-national governments, Gorontalo, 2002-2005 15 10 % of total expenditure 5 0 2002 2003 2004 2005 -5 -10 Province Kota Gorontalo Bone Bolango Kab. Gorontalo Pohuwato Boalemo Further questions for consideration: A typical follow-up question might be: · What does the district / city generally do with the surplus? For example, does it invest the surplus or ac- cumulate it in the reserves? Example 19: reporting of carryovers Table 7 shows the data reported in the financing section of the APBD as stipulated in Kepmendagri No. 29/2002. Table 7 Carry-overs, comparison among sub-national governments, Gorontalo, 2006 Kota Kab. Province Bone Bolango Pohuwato Gorontalo Gorontalo Carry-over to 2006 10.16 12.25 0.49 4.30 14.17 Carry-over budgeted in 2006 6.49 12.25 - 4.30 14.17 Difference 3.67 - 0.49 - - The table shows that in general, financing reports are poorly documented and carry-overs are inconsistently reported. For example, carry-overs reported in one budget year are often inconsistent with carry-overs budgeted in the following year. This poor reporting not only creates errors in the budget plans for the following year, but also increases the risk of budget malfeasance. 5.4 Expenditure chapter 5.4.1 Objective The aim of this chapter is to understand how sub-national governments spend their budgets, how expenditure trends across time, how much money each level of government (central, provincial, district / city) spends in the province, as well as how money is spent across different economic classifications and sectors. The required figures are consolidated from all relevant district / city, as well as the provincial figures. 32 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations The basic documents required to conduct the analysis are mainly budget documents (APBD) obtained from the budget offices or Bappeda. 5.4.2 Types of analysis required In this section, the following topics would usually be discussed: 1. Overall spending picture 2. Sectoral spending analysis 3. Economic Classification Analysis 4. Local spending variation analysis 5. Central government spending analysis 1. Overallspendingpicture This section aims to illustrate trends in total regional expenditure over time and how per capita regional expenditure varies across district / city, as well as comparing them with other provinces in Indonesia. An understanding of which level of government spends most money (central government, provincial or district / city) is also important. Example 20: trends across time and between levels of government. Figure 15 shows the growth in expenditure in Gorontalo and to which level of government the funds are allocated. Figure 15. Total spending, provincial level and consolidated districts, Gorontalo, 2002-2006 Regional penditu (Rp bn) 4,000 700 National expenditure (Rp trillion) 3,500 600 e 3,000 r 500 2,500 400 2,000 ex 300 1,500 200 1,000 500 100 - 0 2002 2003 2004 2005 2006* District / City (L) Province (L) Deconcentrated (L) National Total (R) This figure suggests two trends: (i) Since 2002, expenditure in Gorontalo has increased significantly. Gorontalo's sub-national spending, which includes provincial and district level spending, has increased by more than 100 percent since 2002. (ii) The increase is more significant when central government deconcentrated spending is included in consolidated regional spending. In the period between 2002 and 2006, deconcentrated spending grew almost tenfold and its contribution towards Gorontalo's consolidated regional spending increased from 24 percent in 2002 to an estimated 60 percent in 2006. Further questions for consideration: Some typical follow-up questions might include: · What has caused the growth in deconcentrated spending? Why is it preferred to other forms of inter- governmental transfers? · Is the growth and dominance in deconcentrated spending in Gorontalo consistent with the trend in other provinces in Indonesia? Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 33 Chapter 5 Data Analysis, Conclusions and Recommendations Example 21: provincial per capita comparisons Figure 16 shows the per capita spending in Gorontalo compared to other provinces in Indonesia. It shows that expenditure in Gorontalo is below the national average, but is similar to other provinces in Sulawesi. Figure 16. Consolidated total spending per capita by province, 2005 5 Per capita regional spending (Rp '000) 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 West Kalimantan Central Java Banten NTB DI Yogyakarta South Sumatera Lampung East Java North Sumatera West Java South East Sulawesi Central Kalimantan South Kalimantan East Kalimantan Bangka Belitung Central Sulawesi North Sulawesi West Sumatera North Maluku Gorontalo South Sulawesi NAD (Aceh) DKI Jakarta Bengkulu Papua Jambi Maluku Bali NTT Riau Total Expenditure PC National Average Further questions for consideration: A typical follow-up question might be: · What factors causes certain provinces to have much higher per capita expenditure and other provinces much lower per capita expenditure, in comparison to Gorontalo? 2. Sectoralspendinganalysis This section looks at consolidated local spending categorized by sectoral classifications. The section should explain the current budget format, changes in the budget format and how spending is classified across a 5 to 6 year time span. Overall, the section should analyze the types of expenditure that dominate regional spending over time at the provincial and district levels. The focus of this analysis usually looks at four key sectors: (i) government apparatus (ii) education (iii) health (iv) infrastructure For the analysis of trends over time, constant price values are preferable. 34 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Figure 17. Consolidated sectoral spending, Gorontalo, 2002-2006 1600 1400 8% Sub-national spending (Rp bn) 1200 22% 1000 10% 8% 17% 17% 21% 800 19% 16% 20% 6% 26% 600 27% 5% 18% 28% 82% 400 7% 5% 27% 6% 5% 4% 37% 5% 200 6% 32% 36% 29% 24% 0 2002 2003 2004 2005 2006* Others Health Infrastructure Agriculture Education Administration and apparatus Figure 17 shows consolidated sectoral spending (province, district and city) according to sectors. It shows several trends: (i) The proportion spent on government administration and apparatus increased significantly from 24 percent in 2002 to 37 percent in 2006. (ii) The proportion of total expenditure spent on education has decreased since 2002. (iii) The proportion of total expenditure spent on infrastructure has increased, while the proportion spent on health has remained relatively constant. Further questions for consideration: Some typical follow-up questions might include: · What are the factors that have caused the increases and decreased in spending in the respective sectors? For example, why does the Gorontalo government spend so much more on administration and appara- tus in 2006 compared to 2002? · Why has the proportion of spending on education decreased? · How does Gorontalo's sectoral spending composition compare to other provinces in Indonesia? For example, do other provinces also spend more than one third of their expenditure on administration and apparatus? · How can the sectoral spending mix be improved? 3. EconomicClassificationAnalysis This section looks at consolidated local spending categorized by economic classifications. The section should explain the current budget format, changes in the budget format and how spending is classified across a 5- to 6-year time span. This section should also explain ways to address different budget format and what should be compromised in order to make time series comparison possible. Overall, the section should analyze the type of expenditure that dominates regional spending over time at the provincial and district levels. For trend analysis, constant price values are preferable. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 35 Chapter 5 Data Analysis, Conclusions and Recommendations Example 22: breakdowns of consolidated spending Table 8 shows the spending mix according to economic classification. Table 8. Provincial and districts consolidated spending based on economic classification, Gorontalo, 2002-2006 2002 2003 2004 2005 2006* Province Rp bn % Rp bn % Rp bn % Rp bn % Rp bn % Personnel 15 8 51 20 81 30 66 26 91 23 Goods and services 14 7 36 14 43 16 51 20 130 33 Travel 4 2 17 7 22 8 27 11 0 0 Maintenance 2 1 4 2 6 2 8 3 0 0 Capital 121 64 111 44 74 28 70 28 115 29 Others 34 18 32 13 41 15 31 12 55 14 Total 190 100 251 100 267 100 252 100 391 100 2002 2003 2004 2005 2006* Districts/Cities Rp bn % Rp bn % Rp bn % Rp bn % Rp bn % Personnel 282 51 356 49 409 54 405 52 510 46 Goods and services 19 3 78 11 81 11 135 17 186 17 Travel 6 1 15 2 18 2 18 2 12 1 Maintenance 4 1 22 3 30 4 12 1 6 1 Capital 206 37 230 32 197 26 175 22 314 28 Others 36 7 22 3 25 3 40 5 88 8 Total 553 100 722 100 760 100 784 100 1,115 100 The table shows several observable trends: (i) At the province level, approximately one-third of total expenditure is spent on goods and services, with another 29 percent on capital spending. Personnel spending consumes 23 percent of total provincial expenditure. The amount spent on capital spending has varied significantly between years, from nearly two-thirds in 2002, to below one-third in 2006, whilst spending on personnel has tripled over the same period. (ii) At the district / city level, almost half of total expenditure is spent on personnel, twice that of the provincial government. The district / city governments spent a similar proportion on capital works, but only 17 percent of total expenditure on goods and services: half the proportional amount spent at the Province. Further questions for consideration: Some typical follow-up questions might include: · Why has spending on personnel and goods and services increased so significantly at the provincial level? Is this trend due to an increase in the number civil service staff, salaries, or other associated costs? · Why is such a large proportion of district / city expenditure spent on personnel expenditure? How does this level compare to other provinces in Indonesia? · What sort of infrastructure has built from the capital expenditure? For example, have the funds been spent on government buildings, hospitals, schools, roads or bridges? 36 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations 4. Localspendingvariationanalysis In this section comparisons across districts are analyzed. District-level comparison provides useful information on overall spending levels and the spending mix. For instance, this may cover the level of per capita spending, the level of capital expenditure, and the level of sectoral or economic classification spending. Some key questions to investigate further would include: · which districts have the highest level of per capita spending? · which districts have the highest level of per capita education / health / infrastructure / personnel spend- ing? Where possible, use graphs to illustrate where there are differences, such as Figure 18 below. Example 23: district expenditure comparisons Figure 18 compares the per capita spending of the different districts and cities in Maluku province. It shows that the per capita expenditure in Kabupaten Maluku Tenggara Barat is twice the size of that in Kota Ambon. Figure 18. Central government deconcentrated spending per capita by province, 2005 4,500 4,000 3,500 3,000 2,500 Rp `000 2,000 1,500 1,000 500 0 Kab. M. Kab. Kep. Kab. SBT Kab. SBB Kab. M. Kab. Buru Kab. M. Kota Ten Bar Aru Tenggara Tengah Ambon Per Capita Expenditure Per Capita Revenue Source: SIKD, MoF (2009) Note: Figures are World Bank estimates Further questions for consideration: A typical follow-up question might be: · Why do certain districts have much higher per capita revenue than other districts and Kota Ambon? A possible explanation: newly established district generally have higher levels of per capita revenue. This may be due to the costs required in establishing the government structure (personnel and infrastructure). Comparisons with other newly established district would then be useful. 5. Centralgovernmentspendinganalysis In this section an analysis of spending by the central government at the provincial / district / city level is conducted with a focus on three key sectors, namely health, education and infrastructure. Such spending includes deconcentrated and Tugas Pembantuan (TP) funds. To some degree, central government spending also requires some cost sharing or additional financing from the APBD, which should be explained. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 37 Chapter 5 Data Analysis, Conclusions and Recommendations Example 24: comparison with other Indonesian provinces Figure 19 shows that in comparison to other provinces, Gorontalo has received a relatively high level of deconcentrated spending from the central government. Figure 19: Central government deconcentrated spending per capita by province, 2005 5 3, 0 Deconcentrated spending 0 3, 0 5 2, 0 per capita (Rp'000) 0 2, 0 5 1, 0 0 1, 0 50 - st a l K (A o ak a Go law n DI Sum ntan Ea ral S gku i W uth at a es en st ul lu va M an ra a rth ant ) nt NA ron si rth ma si ka Ja rta ul TT m u s Pa ku a ali B n rth um TB Ea l Jav lit i M ku No lim ceh ve g es ma ali g ra D tal t K la a m si gy er m r No h S N e Su nta nt Jav Be mb nt en e e pu tio La Ria No li e So Su ate ta es Su er l a un un g Ja ali we W ant Ce B aw tS N alu Ka aw Yo at a alu ra an na p Ce t B a as hK t ut hE ng W ut So Ba ut So Na Ce So Further questions for consideration: A typical follow-up question might be: · Why did Gorontalo receive such a high level of deconcentrated spending? Example 25: central government spending Figure 20 Central government spending on decentralized sectors, province and districts, 2005 700 600 Rp bn 500 400 300 200 100 0 Agriculture Health Administration InfrastructurO thersE ducation and apparatus Province District / City Deconcentrated Spending 38 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Figure 20 shows the sectoral focus on deconcentrated funds, which suggests several trends: (i) Deconcentrated funds are rarely used for government administration and apparatus. (ii) Apart from administration and apparatus, deconcentrated funds are a dominant source of funds for all other sectors, especially health and agriculture. Further questions for consideration: Some typical follow-up questions might include: · How have deconcentrated funds been used? Have central government funds been used to build ad- ditional infrastructure, or to run new programs? What changes have there been as a result of central government spending? · How does the central government decide on how to allocate deconcentrated funds? · When are general allocations funds (Dana Alokasi Umum (DAU)) used, rather than special allocation funds (Dana Alokasi Khusus (DAK)) · How does the central government identify initiatives suitable for deconcentrated funding? 5.5 Strategic sectors 5.5.1 Objective The aim of this chapter is to look at more detailed spending in the three strategic sectors of education, health and infrastructure, as well as to gauge the available fiscal resources for each strategic sector. A discussion of performance indicators available at the local level is also conducted by linking them to existing trends in spending in the past 5 to 6 years. A comparison between districts and across provinces, as well as with national figures, will be useful in providing insights. The required figures are consolidated from all districts/ cities, as well as the provincial figures. The basic documents required to conduct the analysis are mainly budget and planning documents at the Dinas level. 5.5.2 Types of analysis required The type of analysis required to write this chapter includes: 1. Breakdownanalysisforeachstrategicsectorsspending This section contains a detailed analysis of the strategic sectoral spending. This is achieved by looking at how much is spent within each sector according to economic classification, as well as according to major programs within the sector/Dinas. Figure 21. Trends in sub-national education spending, Gorontalo, 2002-2005 700 35 29 Rp bn (constant 2005 prices) 600 26 30 25 25 500 25 % 400 20 300 15 200 382 368 10 266 654 100 5 0 0 2002 2003 2004 2005 Regional Expenditure (L) share of total (R) Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 39 Chapter 5 Data Analysis, Conclusions and Recommendations Figure 21 shows that although total education spending has increased significantly, the proportion of total expenditure spent on education has actually decreased from 29 percent in 2002 to 25 percent in 2005. Further questions for consideration: Some typical follow-up questions might include: · What types of programs did the increased budget focus on? For example, did the increased funding go towards improving the quality of teaching, or improving access to education, or improving literacy rates? · How does the level of education spending in Gorontalo compare to other provinces in Indonesia? Example 27: comparisons with other provinces Figure 22 shows Gorontalo's sub-national education spending (district and provincial) in comparison with other provinces in Indonesia. Overall, education spending in Gorontalo is higher than average. This applies whether comparing education expenditure as a share of total spending or per capita spending. Figure 22. Sub-national education spending per student enrolled in primary and secondary school by province, 2005 3,000 40 35 2,500 30 2,000 Rp `000 25 1,500 20 % 15 1,000 10 500 5 0 Sulawesi Barat Lampung Bangka Belitung Sulawesi Selatan Kalimantan Tengah National Maluku Utara Jawa Tengah DKI Jakarta Jawa Barat Jawa Timur Bengkulu Kalimantan Selatan Banten NAD (Aceh) Maluku Papua Jambi Kalimantan Barat Riau Kep. Riau Sumatera Selatan Sumatera Barat SumateraUtara Sulawesi Tenggara Kalimantan timur Papua Barat Bali Sulawesi Tenggara Gorontalo DI Yogyakarta Sulawesi Utara NTT NTB Sub-national education expenditure (per student enrolled in primary and secondary school (L) Share of sub-national education expenditure (R) In 2005, Gorontalo's sub-national spending per school-aged student enrolled in primary and secondary schools was Rp 1.3 million per capita, Rp 311,000 higher than the national average of Rp 990,000. Meanwhile, Gorontalo's district and provincial governments spent 27 percent of their total expenditure on education, 2 percentage points higher than the national average. Example 28: availability of resources for education Table 9 shows the total fiscal resources made available to the education sector by the different levels of government (district / city, province and central governments). 40 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 5 Data Analysis, Conclusions and Recommendations Table 9. Available fiscal resources for education, province + districts + deconcentrated, 2002-2005 2002 2003 2004 2005 Rp bn % Rp bn % Rp bn % Rp bn % Province expenditure 6.3 2 14.1 4 11.7 3 13.9 2 District expenditure 191.2 72 257.1 67 264.9 72 252.9 39 Deconcentrated expenditure 68.9 26 110.8 29 91.0 25 387.3 59 Total regional education expenditure 266.4 100 382.1 100 367.6 100 654.1 100 (2005 prices) Nominal total regional education 218.9 326.8 334.3 654.1 expenditures Growth real total regional education 30.3 -3.9 43.8 expenditures Education expenditure (% of total 28.80 26.43 25.22 24.64 regional) This table suggested several trends: (i) Between 2002 and 2004, the majority of the education budget was spent by the district / city government. (i) In 2005, however, deconcentrated expenditure became the largest proportion of total education spending. (ii) Although education expenditure increased in 2003 (in real terms), it actually fell slightly in 2004 (3.9 percent) but increased significantly again in 2005 (43.8 percent). Further questions for consideration: Some typical follow-up questions might include: · Why did deconcentrated expenditure increase so significantly in 2005? What was the change in govern- ment policy that precipitated this growth? · Why did real education spending decrease in 2004 but increased in 2003 and 2005? Which factors caused this fluctuation in spending allocations? What new programs were added in 2003 and 2005 to merit the increase in expenditure? Figure 23. Economic classification of education spending, Gorontalo, 2005 Districts/cities Province 1% 7% 11 % 2% Travel, maintenance, 37 % 25 % Travel, maintenance, and other and other Capital Capital Goods and services Goods and services Personnel Personnel 86 % 31 % Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 41 Chapter 5 Data Analysis, Conclusions and Recommendations Figure 23 shows that the spending mix at district level differs to the education spending mix at the provincial level: (i) At the district level, all but 3 percent of total education expenditure is spent on either personnel or capital. This suggests that the majority of funds are spent on teachers or infrastructure (such as schools). It also highlights that very little funding is spent on the maintenance of schools and classrooms. (ii) In contrast, the provincial government spends a significant proportion of its education budget on goods and services and also a small portion on travel and maintenance. Further questions for consideration: After analyzing APBD documents further, some typical follow-up questions might include: · What types of personnel did the district government hire, and for what purpose? · How was capital works spending used, and how much of this was used on maintenance? · What are the identified priorities of the education sector in the province and does the budget allocation reflect a focus on these priorities? 2. Performanceanalysisforeachstrategicsector This section contains comparisons between spending on strategic sectors over time (5 to 6 years ideally) with changes in performance indicators specifically identified for each sector. These can be based on inputs, outputs or outcomes. The following indicators are examples of those used in the education, health and infrastructure sectors: Suggested indicators for education: · Educational attainment by year of birth · Access to public primary education by income group · Trends in net and gross enrollment rates for primary and secondary education · Student/teacher ratio, student/classroom ratio, average year of schooling Suggested indicators for health: · Health services utilization · Public health facility utilization by income groups · Access to healthcare services by income group · Per capita monthly out-of-pocket health spending · Morbidity rate · Doctor/population ratio Suggested indicators for infrastructure: · Irrigation coverage and access to road infrastructure · Access to safe water and electricity · Quality of sanitation, clean water and sanitation · Quality and quantity of road infrastructure Each performance indicator may be analyzed according to improvements (or deteriorations) over time, comparisons between districts / cities / provinces, and comparisons with national averages. Chapter 6 42 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 6 Writing the PEA Report Chapter 6 Writing the PEA Report Based on experience, there are some common errors that many writers make. This chapter aims to provide some guidance on how writers can avoid those common mistakes. 6.1 Know your audience In writing a PEA report, not only is it important to remember that the report is intended to be read by a diverse audience, but also that the core audience consists of key policy-makers at the central and local levels. Policy-makers typically receive a great many reports, some of which may be very thick and filled with lengthy explanations. It is important to remember that the reader a) may not have time to read because of other time demands; b) may have many other tasks to complete; and c) may not have in-depth knowledge of the topic of the report or of technical jargon. We observe that many reports remain unread because they are too thick, contain overly complicated explanations, or because the material is not presented in an interesting format. A PEA report should avoid these shortcomings. 6.2 Common mistakes in report writing As mentioned above, PEA reports should avoid common shortcomings in government reports. To do so, here are some ways to avoid common mistakes: Keep it short! PEA reports must be concise and to the point. Avoid repetition and complicated or roundabout language. Use precise, unambiguous terms. Include a glossary: This is necessary to help those with a non-technical background to understand the acronyms and abbreviations that apppear in the text. Avoid jargon and long sentences: Use simple, specific, precise sentences. Avoid particularly lengthy sentences. If the sentence is longer than three lines then it is too long. Use words that people without a finance or economics background will understand. Only include relevant material: Analysis should be objective, relevant and specific. Link paragraphs: Paragraphs should be logically ordered, and maintain a clear flow. Keep to one idea per paragraph: Sentences in each paragraph should also be logically ordered and clearly inter- connected. A controlling idea for each paragraph will help to guide effective use of words and maintain a clear flow between sentences. Use graphs and tables: A picture speaks a thousand words. Where possible, explain your analysis with a graph or a table Reference clearly : If the materials from other sources are used, reference clearly. In addition to data from government department, these include other academics, reports, ets. Do not plagiarize. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 43 Chapter 6 Writing the PEA Report A final checklist : Once the PEA is finished, every section should be reread and the following questions asked: - Are all important findings covered? - Do the contents follow logically from the team's findings? Are there contradictory findings that need to be explained specifically? - Are there overlapping explanations that need to be tidied up? - Are there points of substance that have not been clearly articulated? 6.3 Consistent format, clear structure One of the simplest errors researchers make is using an inconsistent report format. The minimum PEA outline found in Section 1.4 and Appendix 1 refer to the content of the body of the report. This section, however, refers to how the report is formatted. For example, are headings consistent? Are graphs and tables labelled in an easy to read manner? Is there a table of contents page? Is there a list of tables and graphs? Is there a list of references? Below are some simple tips with formatting your report. · Every chapter needs a title typed in the same font with the same size of letters. · To assist the reader, structure each chapter into sections and sub-sections as required. Each section and sub- section must include a heading that uses a consistent font. The hierarchy should be presented clearly, for instance by using capital letters, italics, underlining or bold for section and sub-section titles. This hierarchy should be reproduced in the table of contents. · Tables, graphs and images must be enumerated clearly and consistently. Each table, graph and image must also have a title typed in a font distinct from the body text. · Tables, graphs and images must cite the relevant data source. · As each chapter focuses on a specific thematic area, report recommendations are best written per chapter as the final section of each chapter Checklist Below is a checklist that all PEA reports should include: 1. Cover 2. Title page Includes report title; names of members of research team; the organizations that commissioned the report; the draft status (e.g. first draft, second draft, final draft), and the date of completion/submission. 3. Table of Contents 4. Lists of tables 5. List of graphs 6. List of images 7. List of annexes [Table of contents, list of tables, lists of graphs, list of images and list of annexes should be accurate with clear layout (including indentation) and include page numbers] 8. Glossary 9. Contents of report (body) 44 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Chapter 6 Writing the PEA Report As suggested in this Manual, a report may contain the following chapters: i. Overview 1. Section 1 a. Sub-section 1 b. Sub-section 2 2. Section 2 ii. Planning and Budgeting iii. Revenue iv. Spending v. Sectoral Chapter 10. References 11. Annexes 6.4 Writing the PEA report as a team PEA reports are often written by a team with different members of the team responsible for writing different sections of the report. Although this sharing of work means more data can be collected and analyzed, it also runs the risk that each researcher will write with different writing styles and use different formats. From the outset, the team should agree upon the font, font size, line spacing, spacing between sections and sub-sections, and other details pertaining to report formatting. While each member of the team is responsible for ensuring that each person's work is consistent with the overall report, the team leader is responsible for editing the report to ensure consistency across the work of the different researchers. 6.5 Further references The websites below provide further guidance on report writing: University of South Australia, Report Writing (cited June 17, 2008), http://www.roma.unisa.edu.au/07118/language/ reports.htm The International Development Research Centre, Writing A Research Report (cited June 17, 2008), http://www.idrc. ca/en/ev-56466-201-1-DO_TOPIC.html 6.6 Extra tips: some lessons learned Based on the experience of researchers conducting PEA, we have compiled some additional tips and tricks that will help the researcher write a better PEA report. The following illustrate some of the common problems that arise in PEA reports: Always ask "why"? The most common trap that researchers fall into when writing a PEA report is that while they describe the trend very well they forget to follow up that description with an explanation of why the trend occurred. For example, if the trend shows that per capita revenue increased between 1999 and 2005, then the question is why has this happened? Is this caused by an increase in DAU, DAK or PAD? If caused by DAU, why has the DAU increased? Has the fiscal capacity increased? Or has the PAD increased? If so, then why? Has the tax collection been more efficient? Or are more businesses registered and therefore there is more registration tax? The above example shows that simply explaining that the PAD has increased is insufficient. When writing, always try to think from the readers' perspective. In the above example, the audience will automatically ask: why has the Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 45 Chapter 6 Writing the PEA Report PAD increased? The audience will expect to find a policy or behavioral reason that explains the trend. For example, was there a new type of PAD allowed by regulation? Or has the sub-national government put more attention on collecting local taxes? If there is a trend, there is generally a reason explaining why this trend occurred and a good researcher will always ask "why". Go beyond the minimum: This manual has discussed at length how to fulfill the minimum requirements for a PEA. In addition to meeting the minimum standard, researchers are encouraged to include research ideas of their own that will contribute to the overall quality of the research and the recommendations. The PEA structure is dynamic and researchers should not feel restricted to the methodologies and indicators proscribed by this manual. Some ways of going beyond the minimum include: · Collecting more data: The researcher should feel free to collect more data than that set out in Appendix 2. This may include adding new outcome indicators in health, education and infrastructure, increasing the data time span (i.e. 10 years is better than five), or obtaining more detailed APBD information by sourcing more detailed expenditure reports from the relevant Dinas. All of these will help to improve the depth of the PEA analysis considerably. · Using other methodologies: Chapter 4 outlines some suggested methodologies for analyzing data. Researchers should not be constrained to using only those methodologies found in Chapter 4. · Analyzing other sectors: The PEA has a focus on public service delivery so the sectoral focus is inevitably on the education, health and infrastructure sectors. However, researchers should feel free to identify other sectors of particular interest to the government or the researcher him/herself. For example, agriculture and fisheries are often important sectors to some provinces, and this may be another sector of interest in the PEA. In all instances, researchers may face data limitations and must be mindful that to do meaningful analysis and draw sensible conclusions the data still need to be detailed, accurate, comparable and timely. Rounding: One final minor although important tip is that when presenting analysis try to round numbers to a level that the reader will find easy to read while detailed enough to remain meaningful. For example, the per capita revenue for a province is Rp 95,698 in kabupaten 1, but and Rp 98,374 for kabupaten 2. In this instance, rounding is not required because all numbers are needed to show the difference between kabupaten 1 and 2. However, if the comparison is the total revenue for kabupaten 1 is Rp 128,476,890,346.00, while the total revenue for kabupaten 2 is Rp 130,900,087,756.00, then the numbers should be rounded to the nearest billion rupiah to become Rp 128 billion and Rp 131 billion, respectively. Another rounding method is to use decimals. In the second example, the rounded numbers become Rp 128.48 billion and Rp 130.9 billion, respectively. In both methods, the numbers are much easier to read and comprehend. It is up to the researcher how she or he chooses to round, but the researcher should be vigilant about consistency in rounding. 46 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Appendices Appendices Appendix 1: PEA Standard Outline Chapter Key Sections Purpose Key Questions Key Indicators Key Processes Key Policy E xe c u t i ve Summary To provide brief Numbers of kabupaten, kota, and Administrative description of context Introduction Administrative context the size of each administrative context of the regional region administrative areas To describe location Geography & and population of the Location Map of province population province Pie graph showing landmass and contribution of kabupaten/kota landmass to total province area Pie graph showing total number of Population population per kabupaten/kota Graph showing trend of population growth Brief explanation on ethnic configuration, history of governance, etc (if possible) To describe local economic What is the economic structure development: economic Economy of the province? What sectors Graph showing structure of economy structure, growth, dominate? employment, local issues. Has the growth rate increased or Graph showing RGDP growth by decreased over time? Why? sector Is the provincial growth rate Line graph showing provincial and higher or lower than national the national provincial growth rate of growth rate? RGDP What sectors contribute most to Graph showing structure of employment? employment at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure 47 48 Which employment sectors are Graph showing employment growth growing the fastest? Slowest? by sector Why? Is the provincial employment Line graph showing provincial and growth rate higher or lower than the national provincial growth rate of national employment growth employment rate? Why? What are some specific local at the Sub-National Level economic problems? E.g. Specific issues on local economy malnutrition? Conflict? What problem and chalange caused these specific problems? To describe poverty Is the provincial poverty rate Graph showing poverty headcount of Poverty conditions compared with higher than the national poverty the province and national provincial national condition rate? Or lower? Has poverty rate increased or Graph showing trend of proverty decreased over time? Describe the current level of Human Practical Guidelines for Analyzing Public Expenditure HDI and SPM Development Index and the minimum service standards 1. To describe the national planning and budgeting framework. 2. To assess whether local planning and budgeting processes follow the Planning and national framework. Budgeting 3. To assess the consistency (Budgeting of planning documents Cycle) and whether consistent with the national plan. 4. To assess whether the budget reflects the development priorities found in the planning documents. What planning documents are To assess whether required by national guidelines? P l a n n i n g provincial and district Have the sub-national Number of planning documents Processes planning follow the governments produced those national guidelines documents Are the planning documents Date when planning documents submitted according to the submitted national timeline? Appendices Has the planning process followed the process set out by Compare the planning processes the national guidellines? If not follow guidelines, how to To assess whether Is each year's budget improve? B u d g e t i n g provincial and district document following the T h r o u g h Processes budgeting follows the format required by the capacity national guidelines national regulations? building? Te c h n i c a l assistance? Are the budgeting documents submitted Date when budgeting documents according to the national submitted. timeline? Has the budgeting process followed the process set out Compare budgeting processes. in the national guidelines? Development What are the priority sectors To describe provincial Priorities and mentioned in the provincial RJPMD development priorities Budgeting RPJMD? Brief description on "urusan wajib" and "urusan pilihan" Are the development priorities in the provincial To assess linkages planning document also between planning RJPMD, Renstra etc reflected in the Renstra documents and kabupaten planning documents? Are the provincial development priorities consistent with the national plan? Understanding Does the public budget linkages between reveal the development RJPMD, Renstra etc planning and goals stated in the planning budgeting documents? If not, what are Does the budget reflect the ways to link development priorities in the planning and planning documents? budgeting? at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure 49 50 Summary & Recomen- dation 1. To calculate total revenue envelope 2. To understand revenue sources and at the Sub-National Level constraints 3. To calculate deficit Revenue and and surplus (identify Financing policy) 4. To understand how the local governments record their revenue and financing Practical Guidelines for Analyzing Public Expenditure How does the province s h o u l d Overall Revenue 1. How much revenue does Fiscal revenue per capita (for each Benchmark a l l o c ate Picture the region have overtime? revenue item) analysis their money to equalize the fiscal capacity? 2. How is the revenue distributed across Table 1 kabupaten/kota 3. What is the composition of Table 2 the revenue (DAU, DAK, etc)? How effective is the DAU DAU (General 1. What is the description DAU Per Capita (Province & Kab/ Benchmark formula? Allocation Grant) and constraint Kota) analysis From an inequalization prespective? 2. What is the trend over Correlation time? Why? analysis 3. What is the distribution within province? Relative Poverty rates, civil servants and to poverty rate? Relative to expenditure needs (from DAU civil servants? Relative to formula) expenditure needs? Appendices 1. What is the description DAK (Special and constraint? How does it DAK by sector and by province/ Benchmark Allocation Grant) work in practice? DAK is used kab/kota (allocation vs realization) analysis to finance what? Appendices DAK per capita (province & kab/ 2. How much DAK over time? kota) 3. What is the sectoral composition of DAK? 4. How is the DAK distributed across levels of govt (prov & kab/kota) and between kab/ kota Shared Tax Revenue / Shared tax revenue / Shared 1. What is the regulation and Benchmark Shared Natural natural resource revenue per the formula? analysis R e s o u r c e capita (province & kab/kota) Revenue 2. How much over time? Are they consistent over time? (province and kab/kota) 3. What is the composition? 4. How are they distributed within province? Any significant inequality? Why? Does the 1. How much over time? Are region need to PAD (Own PAD per capita (province & kab/ Benchmark they consistent over time? increase PAD? Source Revenue) kota) analysis (province & kab/kota) If yes, which ones? 2. What is the composition? Analyze the Perda of main taxes and user 3. What are the main taxes charges. What is and user charges (province & the rate? Relative kab/kota)? to other kab/ kota or other provinces? at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure 51 52 4. How efficient is the collection mechanism? Do Revenue collected vs cost of the provinces and kab/kota collection have one roof system? Does the region have 1. How much surplus / Financing in flow and out flow, Benchmark Financing fiscal space? deficit? carry over. analysis How does it at the Sub-National Level create it? 2. How do the province/kab/ kota finance the deficit? 3. How do the province/kab/ kota allocate the surplus 4. How do they record the surplus/deficit? Is it consistent with the real Practical Guidelines for Analyzing Public Expenditure surplus/deficit calculated by total revenue minus total expenditure? Summary & Recommen- dation To calculate total Expenditure regional expenditure. To understand expenditure trends across time. To understand who (central, provincial, district) spends the money. To understand how money is spent - economic classification, sectors. To understand variations between districts. Line graph s h o w i n g What is the total regional Overview See Table 1 sub-national expenditure over time? expenditure over time. Appendices Bar graph How does per capita regional showing per expenditure compare to Total Expenditure; population capita spending other provinces in Indonesia? compared to Appendices other provinces. Is central gover n ment Bar graph Who spends the money? dominating s h o w i n g Central government? spending? See Table 1 who incurred Provincial government? Or P o l i c y expenditure over district governments? implications? time. E c o n o m i c Consider and Sectoral Explain budget format, implications Classification changes in budget format of budget (Routine v and how spending is f o r m a t D evelopment; classified. changed too Capital v Non- often. capital) Assuming that Bar/line graph Overall, what type of one of type showing mix expenditure dominates expenditure See Table 2 between types regional spending over time? is more of expenditure Why? effective than over time. others (e.g. Bar/line graph showing mix What type of expenditure between types of dominates provincial See Table 3 expenditure over spending over time? Why? time at provincial level. Bar/line graph showing mix What type of expenditure between types of dominates district spending See Table 4 expenditure over over time? Why? time at district level. at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure 53 54 Has a sufficient portion of the budget been allocated to key service sectors such Graph showing as education, Which sectors have s e c t o r a l health and at the Sub-National Level Sectoral dominated spending over See Table 5 distribution infrastructure? time? Why? over comparing [Assumes several years. that service delivery has a strong correlation with amount of budget allocated to Practical Guidelines for Analyzing Public Expenditure those sectors] Graph showing At provincial level, what are s e c t o r a l the most important sectors? See Table 6 distribution Why? over comparing several years. Graph showing At district level, what are s e c t o r a l the most important sectors? See Table 7 distribution Why? over comparing several years. Is central In the most important gover n ment sectors - health, education dominating and infrastructure - which See Table 8 Table 8 spending? dominates spending (central, P o l i c y provincial or districts)? Why? implications? Is there strong inequality Bar graph Which districts have the bet ween Spending by comparing highest level of per capita See Table 9 districts? If so, District expenditure by spending? Why? why? Can this different districts be improved? How? Bar graph Which districts have the comparing highest level of per capita See Table 9 education education spending? Why? expenditure by Appendices different districts Bar graph Which districts have highest comparing level of health spending? See Table 9 h e a l t h Why? expenditure by Appendices different districts Bar graph Which districts have highest comparing level of infrastructure See Table 9 infrastructure spending? Why? expenditure by different districts Has the gover n ment What is the overall realization managed to Planned versus rate? Has government been Total realization rate. spend their Realization able to spend its budget? If allocated not, why? funds? If not, why? What is the realization trend for economic classification? Table showing Explain reasons for under- See Table 10 realization realization and over- percentage. spending. What is the realization trend Table showing for sectors? Explain reasons See Table 11 realization for under-realization and percentage. over-spending. Who does the bulk of Findings the spending? Policy implications? What types of spending dominate expenditure? Policy implications? What sectors dominate spending? Policy implications? Are there high inequalities between districts? Policy implications? Have governments generally met budget targets (realization rates)? Policy implications? Summary & recommen- at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure dation 55 56 1. Overview of the sectoral allocations and - Agree on the current sectoral s e c t o r a l performance (the link definition between expenditures - Agree on data as inputs and sources sectoral performance - Analyze data or indicators as from BPS and/or outcomes). Hopefully, Dinas at the Sub-National Level Sectoral the outcome and - Compare output justify how spending- a region spends its output-outcome budget. with other 2. To see how the regions outcome reflects - Benefit the spending, and incidence (very spending reflects the basic) priorities. Practical Guidelines for Analyzing Public Expenditure Health -Is health a priority in the Agree on To analyze the health province's planning? p l a n n i n g Planning Planning documents policy for the province. - If so, what are the specific documents that health priorities? will be analyzed. - Are there different health priorities between districts? - Are the health priorities consistent with the national level health priorities? - What percentage of total sub-national spending is on health? - Has the share of total To analyze the health spending increased or Spending (Input) spending patterns in APBD, Dekon. HH decreased over time? Why? health spending - Has the health sector spending increased or decreased over time? Why? Appendices - What is the per capita spending on health? `- Is it higher than the national per capita average? Appendices Compared with other similar provinces? - What is the per capita health spending of each district? Are there big variations between districts? - What proportion of the health sector spending is by district government, provincial government and central government? - What is the trend over time? Why? - What is the household expenditure on health? - Is this higher or lower than national average? Other similar provinces? - What has the health budget been allocated to? According to program classification? According to economic classification? - How many Puskesmas are To analyze what there in the province? outputs have been - How does this compare achieved through S p e n d i n g with the national average? health spending (e.g. Number of Puskesmas (Output) - What is the number of infrastructure, facilities, Puskesmas in each district? human resources, Variations between districts? programs etc). at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure 57 58 - How many doctors and midwive are there per 10,000 people? Per km2? - How does this compare with the national ratio? Number of health workers - Compare between districts in the province? Are the big variations beween districts? Between urban and rural? at the Sub-National Level - Ratio of births attended by healthcare worker Some general indicators: Morbidity rates/child Analyze the current nutrition/infant mortality Morbidity rate, child nutrition, Outcomes condition of health for the province? Between infant mortality. districts? Compare with Practical Guidelines for Analyzing Public Expenditure national rate - Immunization rate for province? For each district? Immunization rate Compare with national rate? Health services utilization: self treatment, compared Health services utilization rate. with health clinics etc. What is the utilization rate of public health facilities? Compared with national B e n e f i t - average? Other similar Public health services utilization incidence provinces? rate. analysis Compare utilization rates between districts? Variation? Utilization rates of public B e n e f i t - Public health services utilization health facilities according to incidence rate. income levels. analysis Utilization rates of public health facilities according to gender? Any other outcomes e.g. HIV rates etc. Appendices If there is the link, what Analyze the link are possible Findings between spending, - Is there any link? areas for focus outputs and outcomes. within the health sector? Are there inequality in health between districts? Between girls and boys? Is this also reflected in the spending levels? Summary & Recommen- dation Education -Is education a priority in the Agree on To analyze the province's planning? p l a n n i n g Planning education policy for Planning documents - If so, what are the specific documents that the province. education priorities? will be analyzed. - Are there different education priorities between districts? - Are the education priorities consistent with the national level education priorities? - What percentage of total sub-national spending is on education? - Has the share of total To analyze the education spending Spending (Input) spending patterns in increased or decreased over APBD, Dekon. HH education spending. time? Why? - Has the education sector spending increased or decreased over time? Why? at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure 59 60 - What proportion of the education sector spending is by district government, provincial government and central government? - What is the trend over time? Why? - What is the per capita at the Sub-National Level spending on education? `- Is it higher than the national per capita average? Compared with other similar provinces? - What is the per capita education spending of each district? Are there big variations between districts? Practical Guidelines for Analyzing Public Expenditure - What is the household expenditure on education? - Is this higher or lower than national average? Other similar provinces? - What has been the highest spending within the education sector according to economic classification? - Program and economical classification - How many schools are there in the province? To analyze what - What is the average number outputs have been of students per school in the achieved through province? S p e n d i n g education spending - How does this compare Number of schools - GDS analysis (Output) (e.g. infrastructure, with the national average? facilities, human - Average number of resources, programs students per school in each etc). district? Variations between districts? Appendices - What is the student-teacher- ratio (STR) for the province? Compare with national STR. - What is the STR for each Number of teachers Appendices district? Variation between districts? Why? - What is the level of teachers' Teachers' qualification qualifications? - What is the level of teacher Teachers' attendance absenteeism? - What is the condition of classrooms in the province? - Compare with national Condition of classrooms average? Compare between districts in province? There may be other outputs. Text books, distance to schools E.g. text books, distance to etc. schools etc. - What is the gross enrollment rate (GER) at primary school level? Junior secondary school level? Senior secondary level? - How does it compare with the national average? - Compare the GER for each district. - Compare enrollment rates by gender? Is there a correlation between distance to school and enrollment rate? B e n e f i t- - Usage of public school incidence facilities by income level? analysis B e n e f i t- - Usage of public school incidence facilities by gender? analysis Any other outcomes. at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure 61 62 If there is the link, what Analyze the link are possible Findings between spending, - Is there any link? areas for focus outputs and outcomes. within the education sector Are there inequalities in education between districts? at the Sub-National Level Between girls and boys? Is this also reflected in the spending levels? Summary & recommen- dation Infrastructure Practical Guidelines for Analyzing Public Expenditure -Is infrastructure a priority in Agree on To analyze the the province's planning? p l a n n i n g Planning infrastructure policy - If so, what are the specific Planning documents documents that for the province. infrastructure priorities? will be analyzed. - Are there different infrastructure priorities between districts? - Are the infrastructure priorities consistent with the national level infrastructure priorities? - What percentage of total sub-national spending is on infrastructure? - Has the share of total To analyze the infrastructure spending spending patterns Spending (Input) increased or decreased over APBD, Dekon.HH in infrastructure time? Why? spending. - Has the infrastructure sector spending increased or decreased over time? Why? Appendices - What is the per capita spending on infrastructure? `- Is it higher than the national per capita average? Appendices Compared with other similar provinces? - What is the per capita infrastructure spending of each district? Are there big variations between districts? - What proportion of the infrastructure sector spending is by district government, provincial government and central government? - What is the trend over time? Why? - What is the household expenditure on infrastructure? - Is this higher or lower than national average? Other similar provinces? - What has been the highest spending within the infrastructure sector according to sub-sectors? E.g. transportation, electricity, telecommunications, government buildings etc. What has dominated infrastructure spending according to economic classification. - Length of roads To analyze what - Quality of roads outputs have been - Compare over time. achieved through Compare with other S p e n d i n g infrastructure provinces, compare with Length of roads; quality of roads. (Output) spending (e.g. national level, compare infrastructure, facilities, between districts within human resources, at the Sub-National Level Practical Guidelines for Analyzing Public Expenditure provinces. programs etc). 63 64 Analyse infrastructure indicators (public sanitation, no sanitation, piped water, Analyze the current improved water, sanitation, Outcomes condition of irrigation coverage, road infrastructure. coverage, access to electricity and any other relevant indicators) to evaluate the following: at the Sub-National Level - Have there been improvements over time? - Higher or lower than national rate? Compare within districts in province B e n e f i t - (is there inequality between incidence districts?) analysis, time- - Differentiate the access series analysis. according to groups such as Practical Guidelines for Analyzing Public Expenditure income level (or gender, if possible). If there is the link, what Analyze the link are possible Findings between spending, - Is there any link? areas for focus outputs and outcomes. within the infrastructure sector? Are there inequalities in infrastructure between districts? Between income levels? Is this also reflected in the spending levels? Summary & Recommen- dation Local Issues To be decided Summary & Recommen- dation Appendices Appendices Appendix 2: Data Requirements The objective of this section is to provide researchers with data requirements (both quantitative and qualitative) and some advice on where to obtain the data. The following also includes some advice on time-span of data, but this depends on the scope of the research. As discussed earlier, we recommend collecting data over at least five years, the general rule being that the more data the better the quality of analysis. Data can be collected at both the national and sub-national levels. Where there are two sources for the same data (for example, APBD), the researcher needs to decide, at the outset, which source will be used for analysis. Each data source has its own advantages and disadvantages. See Section 4.2.1 for guidance on the appropriateness of each source. 1. National Sources: Ministry of Finance (MoF); Statistics Bureau (BPS); No Type Name Description Source Indicators Deconcentrated National expenditure (provincial level) by 1 Fiscal APBN MoF and assistance routine and development. task Expenditure Regional budget (Kab/Kota & provincial level) consists of revenue (by items), routine exp (by bagian and econ. class), and development BPS ; Revenue, routine, APBD exp (by sectors).(Makuda format), Public and MoF development apparatus spending (Kepmendagri 29 format); or Direct and indirect spending (Permendagri 13 format) DAU allocation, population, DAU allocation and basic data used for DAU DAU MoF p o v e r t y calculation (Kab/Kota & provincial level) incidence, area, IHBG, etc. DAK allocation DAK Dana Reboisasi and DAK Non Dana (infrastruc ture, DAK MoF Reboisasi (Kab/Kota & provincial level) health, education sectors) Realized loan, Data on local government borrowing and outstanding Borrowing MoF arrears amount of loan, total arrears Local Taxes Revenue Data on local taxes and levies MoF Taxes, levies Sharing PBB, BPTHB, PPh, Revenue Tax and non-tax revenue sharing for local SDA allocations MoF Sharing governments by type and region 2 Non Fiscal Population Census National population census, conducted once Social or other BPS in ten years or other annual population data Population Indicators population from BPS data from BPS Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 65 References E d u c a t i o n attainment, Susenas consists of CORE (annual) and literacy rate, % of MODULES (once in three years). It covers Susenas BPS urban population, characteristics of household and members of household household within sampled household. income and expenditure. Labor force The Annual National Labor Force Survey (by sector), (Sakernas) covers national labor market Sakernas BPS employment rate, characteristics of all working age individuals unemployment within sampled households. rate, etc. no. of facilities ( s c h o o l s , Puskesmas, etc), Village Potential Survey (Podes) provides % of family with Podes information about villages/desa characteristics BPS phone, type of and infrastructures. road available, acres of paddy field, etc. Economic Regional Products (kab/Kota & provincial level) GRDP BPS Sectoral products Indicators by current & constant price Length of road, Road Type of road at the kabupaten level MoPW share of each type of road Governance indicators and Local Gov't WBOJ Governance Decentralization Survey 1, GDS decentralization Charac GDS & PSKK- 1+, GDS 2 ( t r a n s p a r e n c y, teristics UGM accountabilit y, service quality) Number of civil servants by structural Civil classification, Servant Number of civil servants by localities BKN functional Census classification, characteristics, golongan, etc 66 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Appendices 2. Data from Sub-National Source : Quantitative Type Name Years Source Indicators Mandatory Population Latest Regional BPS To create per capita value Indicators Regional CPI Latest Regional BPS To create real value Most detailed available breakdown of · Revenue · Expenditure Provincial - Routine & Development and local (Makuda format) governments - Public & Apparatus Fiscal APBD Latest financial reports (Kepmendagri 29 Format) (preferably - Direct & Indirect (Permendagri audited financial 13 Format) reports) - Detailed expenditures on health facilities and schools, by type · Financing (new budget format) · Level of debt Provincial and Fiscal Borrowing Latest · Loans by source (central local government government, bank etc). · Criteria used for allocation shared Shared natural revenue for oil, gas and forestry to Provincial Fiscal resource Latest kab/Kota government revenue · Amounts allocated to each kab/ kota · Local government's estimate of L o c a l total population Social Population Latest government · Local BPS office estimate of population · Local government's estimate of L o c a l poverty Social Poverty Latest government · Local BPS office estimate of poverty · Number of schools, by type · Number of students · Number of teachers (including regular and non-regular) L o c a l Social Education Latest · Amount of retribution collected government · Education indicators used by local government (literacy rate, enrollment rate, education attainment, etc) Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 67 Appendices · Number of health centers · Number of doctors and other health staff (including regular and non-regular) L o c a l Social Health Latest · Amount of retribution collected government · Health indicators used by local government (births attended by skilled workers, infant mortality rate) · Current status and condition of Water and Provincial and water infrastructure Infrastructure Latest Sanitation local government · Access to clean water · Length of roads, by type Provincial and Road Latest · Condition of roads local government · Composition of DPRD members (political parties represented) Provincial KPU Local Politics DPRD Latest · Characteristics of DPRD members office (educational background, gender) Provincial · Number of civil servants, by Governance Civil Servants Latest and local structural and functional category governments and by professional grade Qualitative Type Name Years Source Indicators · What plans are in place? · Does the public participate in the planning process? If yes, how? · Is there a plan/strategy for Planning Planning Latest Bappeda improvement of the economy? If Process yes, which sector(s) does it focus on? · How is the achievement of the plans monitored? · Is there a formal mechanism for the public to participate in the budgeting process? If yes, how effective is it? · How is quantitative information Biro Keuangan, Budget used to make budget decisions? Budget Latest Bappeda, head Preparation · How are decisions made regarding of each agency government intervention in a particular sector? · Which institution/body makes decisions on the final budget allocations? · Is the budget made available to Budget Latest Bappeda, Sekda the public? If yes, how (newspaper, Publication government gazette etc.)? 68 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level Appendices · Which unit is responsible for disbursement? · Has the local treasury (BUD) been established? Biro Keuangan, · What are the payment mechanisms Budget Latest Bappeda (SPP)? · Are there problems in with cash management? If yes, which? · Has the budget been revised during the fiscal year? · Has performance budgeting been Performance introduced yet? Latest Bappeda Budgeting · If yes, how is performance monitored? Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 69 References · When and in how many installments does the local government receive transfers from the central government and the Disbursement Transfer Latest Biro Keuangan province? of Transfer · DAU · DAK · Shared natural resource revenue · Shared tax revenue Format -- electronic wherever possible 70 Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level References References Aran, Meltem (2007). Note on "Pro-Poor Targeting and the Effectiveness of Indonesia's Fuel Subsidy Reallocation Programs". Jakarta. Indonesia Ghozali, Abbas. "Analisis Sejarah Kebijakan, Penyelenggaraan, dan Kondisi Pendidikan Dasar serta Implikasinya pada Pendidikan Dasar Gratis". Makalah individual untuk studi Pendidikan Gratis yang diselenggarakan oleh BAPPENAS. Jakarta. Indonesia. Pemerintah Daerah Kabupaten Belu (2004). Rencana Strategis Kabupaten Belu 2004-2008 Pemerintah Daerah Kabupaten Belu (2004). Rencana Strategis Dinas Pendidikan Kabupaten Belu Periode 2004- 2008 Pemerintah Daerah Kabupaten Minahasa (2003). Rencana Stratejik Dinas Pendidikan dan Kebudayaan Kabupaten Minahasa Tahun 2003-2007 Pemerintah Daerah Kabupaten Timor Tengah Selatan (2004). Rencana Strategis Kabupaten Timor Tengah Selatan 2004-2008 Pemerintah Daerah Kabupaten Timor Tengah Selatan (2005). Rencana Strategis Dinas Pendidikan dan Kebudayaan Kabupaten Timor Tengah Selatan Tahun 2005-2009 Pemerintah Daerah Kota Binjai. Peraturan Walikota Binjai nomor 050-6525 tahun 2005 tentang Rencana Pembangunan Jangka Menengah Daerah Kota Binjai 2006-2010. Pemerintah Daerah Kota Magelang (2005). Peraturan Daerah nomor 9 tahun 2005 tentang Rencana Pembangunan Jangka Menengah Kota Magelang Tahun 2005-2010. Pemerintah Daerah Kota Manado (2005). Peraturan Daerah nomor 04 tahun 2005 tentang Rencana Pembangunan Jangka Menengah Daerah Kota Manado 2005-2010. Pemerintah Daerah Kota Manado (2005). Matriks Program Lima Tahunan (RPJMD dan Renstra SKPD). Dinas Pendidikan Kota Manado Pemerintah Indonesia (2007). Peraturan Pemerintah Republik Indonesia nomor 38 tahun 2007 tentang Pembagian Urusan Pemerintahan antara Pemerintah, Pemerintahan Daerah Provinsi, dan Pemerintahan Daerah Kabupaten. Pemerintah Indonesia (2005). Peraturan Pemerintah Republik Indonesia nomor 19 tahun 2005 tentang Standar Nasional Pendidikan. Pemerintah Indonesia (2005). Undang-Undang Republik Indonesia nomor 14 tahun 2005 tentang Guru dan Dosen. Pemerintah Indonesia (2003). Undang-Undang Republik Indonesia nomor 20 tahun 2003 tentang Sistem Pendidikan Nasional. Pemerintah Indonesia (2002). Keputusan Menteri Dalam Negeri nomor 29 tahun 2002 tentang Pedoman Pengurusan, Pertanggungjawaban dan Pengawasan Keuangan dan Belanja Daerah, Pelaksanaan Tata Usaha Keuangan Daerah dan Penyusunan Perhitungan Anggaran Pendapatan dan Belanja Daerah. Practical Guidelines for Analyzing Public Expenditure at the Sub-National Level 71