2021 Village Public Expenditure Management in Indonesia: Towards Better Budgeting and Spending Findings from a Village Public Expenditure Review ACKNOWLEDGEMENTS This assessment was led by Anna O'Donnell (Senior Social Development Specialist, World Bank), with guidance and support from the Ministry of Home Affairs. Analysis for the public expenditure review was conducted by Ihsan Haerudin and Andrew Laing, with support from Findi Firmanliansyah. The report was written by Ihsan Haerudin, Tara Moayed, Alexander B Setiadji, and Shah Zaman Farahi. The authors would like to thank the team from MoHA’s Directorate of Facilitation for Village Finance and Asset Management and the Siskeudes team, without whom this report would not be possible. The report has benefited from useful comments by peer reviewers: Bert Hofman, Kathleen Whimp, Jurgen Rene Blum, and Ahmad Zaki Fahmi. This report was made possible through funding from the Australian Government Department of Foreign Affairs and Trade (DFAT). 1 CONTENTS Abbreviations ........................................................................................................................................................ 4 Executive Summary ............................................................................................................................................... 5 Summary of Findings .................................................................................................................................................6 Future Directions .......................................................................................................................................................8 1. Introduction .................................................................................................................................................. 9 2. Village Financial Management ..................................................................................................................... 10 Village Resources Since Introduction of Village Law ..................................................................................10 Village Revenue Sources ............................................................................................................................12 Village Financial Management Regulations ...............................................................................................13 Village Financial Reporting .........................................................................................................................13 3. Methodology ............................................................................................................................................... 15 4. Part 1: Village Reporting Analysis ................................................................................................................ 17 Finding 1. Nearly 95% of villages across Indonesia use Siskeudes, while data from nearly 60% of villages is submitted into the National Consolidation Application. ....................................................................................17 Finding 2. Most village reports included both original budget and actual expenditure data, and the majority passed basic data validity tests; however, poor district guidance may impact reliability of reports. .......................................18 5. Part 2: Village Expenditure Analysis ............................................................................................................ 19 Finding 3. The largest share of village spending is for public works (36%), followed by salary and operations (26%), housing/sanitation (7%), health (4%), and education (4%). ....................................................................19 Finding 4. Across all regions, the majority of village expenditures (80%) go toward capital and goods and services. ..............................................................................................................................................................21 Finding 5. Village priorities vary significantly across provinces. .........................................................................21 Finding 6. Since 2016, villages have increased spending on housing/sanitation, education, and health, while reducing their share of spending on public works and salary and operations. ..................................................23 Finding 7. 97% of villages invest in stunting-related activities, at an average rate of 12%. Papua has the highest spending share with nearly 30%, followed by NTT and Gorontalo at nearly 20% .................................24 6. Part 3: Village Revenue Analysis .................................................................................................................. 27 Finding 8. Dana Desa makes up just over half of total village resources. ...........................................................27 Finding 9. Share of village revenues from Dana Desa Ranges from 23% in Bali to 81% in Papua. .....................27 Finding 10. Village Own-Source Revenue contributes 3% to overall village revenues, with most of this revenue coming from villages in Java and Bali..................................................................................................................29 Finding 11. Spending priorities shift significantly between different fund sources. ..........................................31 7. Part 4: Village Budget Credibility Analysis ................................................................................................... 32 Finding 12. 75% of villages receive an A or B grade in a modified PEFA assessment of budget reliability. ................33 Finding 13. There is a large variation in budget reliability between regions, with Sulawesi showing the best performance and Java the poorest. ....................................................................................................................34 Finding 14. Predictable resourcing leads to better budgets. ..............................................................................36 8. Future Directions ......................................................................................................................................... 37 Key Lessons from 2019 ViPER .................................................................................................................................37 Annex A: Village Reporting .................................................................................................................................. 39 Annex B: Village Expenditures ............................................................................................................................. 40 Annex C: Village Revenues ................................................................................................................................... 42 Annex D: Village Budget Credibility ..................................................................................................................... 45 Annex E: Dana Desa Financial Flow Mechanism (Fiscal Year 2020 as per PMK 205/2019) ...................................................... 47 2 List of Figures Figure 1: Village revenues 2013 to 2019 (in trillion IDR, nominal) ..............................................................................11 Figure 2: Village government expenditure as a share of national public spending (in trillion IDR).............................11 Figure 3: Villages Using Siskeudes and Report Availability in Siskeudes National Consolidation Application .................................16 Figure 4: Reporting completeness ...............................................................................................................................18 Figure 5: Cut and paste test on revenue and expenditure reports .............................................................................18 Figure 6: Frequency of unreliable (Rev) reports by district ..............................................................................................18 Figure 7: Village Expenditure by Sub-Bidang (2019) ....................................................................................................20 Figure 8: Village spending by economic classification .................................................................................................21 Figure 9: Top five spending priorities by province.......................................................................................................22 Figure 10: Village expenditure priority shift across major sub-Bidang (2016 and 2019) ............................................23 Figure 11: Percent of Villages With Stunting-Related Expenditures and Frequency Distribution by Spending...........................25 Figure 12: Village Stunting Expenditures by Province .................................................................................................26 Figure 13: Village Revenue by Source ..........................................................................................................................27 Figure 14: Village Revenue Composition by Province .................................................................................................28 Figure 15: Village Own-Source Revenue Breakdown (2019) .......................................................................................29 Figure 16: PADes trend in nominal terms and as a share of total village revenue (2013-2019) .................................29 Figure 17: Village Own-Source Revenues by Province ................................................................................................30 Figure 18: Top Village Expenditure Categories by Source of Fund ..............................................................................31 Figure 19: Modified PEFA Grades in Key Budget Reliability Indicators for 25,430 Villages .........................................33 Figure 20: Average Village PEFA Scores .......................................................................................................................33 Figure 21: PEFA Grades by Region Across 25,430 Villages - Count and Rate ..............................................................34 Figure 22: Average Village PEFA Grade by Province for 25,430 Villages .....................................................................35 Figure 23: Distribution of Aggregate Revenue and Expenditure Scores as Percentage of Budget to Actuals.............................36 List of Tables Table 1: Village revenue trends before and after the Village Law (in trillion IDR, nominal) .......................................10 Table 2: Description of Village Revenue Sources .........................................................................................................12 Table 3: Village Chart of Accounts Summary ...............................................................................................................14 Table 4: Village-Level Stunting Priority Interventions .................................................................................................24 Table 5: Average and median of village spending on stunting-related interventions .................................................25 Table 6: Villages Using Siskeudes and Report Availability in Siskeudes National Consolidation Application ..........................39 Table 7: Village Spending by Functional Classification (GFS) .......................................................................................40 Table 8: Village Spending by Economic Classification .................................................................................................40 Table 9: Village Spending by Economic Classification by Region .................................................................................40 Table 10: National estimates of Village Expenditure by Sub-Bidang & Infrastructure Content, 2019 ........................41 Table 11: Village Revenues by Fund Source by Province .............................................................................................42 Table 12: Top Village Expenditure Categories by Source of Fund ...............................................................................43 Table 13: Financing contribution of multiple funds across sub-bidang .......................................................................44 Table 14: Average PEFA Scores by Province ................................................................................................................45 Table 15: PEFA Scoring Tables .....................................................................................................................................46 3 ABBREVIATIONS ADD Village Fund Allocation (Alokasi Dana Desa) APBDes Village Budget (Anggaran Pendapatan dan Belanja Desa) BH-PDRD Shared Tax and Levies from District Government BPD Village Council (Badan Permusyawaratan Desa) BPK Audit Board of Indonesia (Badan Pemeriksa Keuangan) BPS Statistics Indonesia (Badan Pusat Statistik) BumDes Village-Owned Enterprise (Badan Usaha Milik Desa) DAK Special Allocation Fund (Dana Alokasi Khusus) DD Dana Desa (Village Fund) DIY Daerah Istimewa Yogyakarta ECD Early Childhood Development FY Fiscal Year GAKIN Poor Households (Keluarga Miskin) GFS Government Finance Statistics IDR Indonesian Rupiah MoF Ministry of Finance MoHA Ministry of Home Affairs NGO Non-governmental Organization NTB West Nusa Tenggara (Nusa Tenggara Barat) NTT East Nusa Tenggara (Nusa Tenggara Timur) OMSPAN Online Monitoring Application for the State Treasury and Budget System OSR Village Own-Source Revenue (PADes) PADes Village Own Source Revenue PAUD Early Childhood Education Centre PEFA Public Expenditure and Financial Accountability PER Public Expenditure Review Permendagri Minister of Home Affairs Regulation (Peraturan Menteri Dalam Negeri) Polindes Village Maternity Hut (Pondok Bersalin Desa) Posyandu Integrated Health Post (Pos Pelayanan Terpadu) PP Government Regulation (Peraturan Pemerintah) Puskesmas Community Health Clinics (Pusat Kesehatan Masyarakat) RT/RW Neighborhood and Hamlet Heads RTLH Unhealthy Households (Rumah Tidak Layak Huni) Siskeudes Village Financial System (Sistem Keuangan Desa) StraNas Stunting National Strategy to Accelerate Stunting Prevention 4 EXECUTIVE SUMMARY Since the Government of Indonesia began implementation of the Village Law, more than IDR 547 trillion (US$ 38 billion) has been transferred to Indonesia's nearly 75,000 villages. In 2020, intergovernmental transfers to villages constituted 10% of subnational fiscal transfers, playing an important role in supporting complementary activities for frontline services. The Village Law reaches 176 million people in Indonesia, including 117 million people in rural areas, who constitute most of the country's poor. Village funds can reach Indonesia's rural poor with improved access to infrastructure and basic services; help to reduce stunting and improve village health, nutrition, and early childhood education; empower women, youth, and vulnerable groups; provide "last-mile" rural connectivity; and strengthen local natural resource management, disaster response, and climate resilience. In 2019, for the first time, the Government of Indonesia began collecting financial data on how village governments across the country were using their village funds. This data allows the government to analyze village priorities, expenditures, revenues, and budget credibility. Prior to the implementation of a unified village chart of accounts in 2018 and the roll-out of the village financial management system (Siskeudes) National Consolidation Application, the government was not able to systematically access village financial data. The data collected before 2019 only covered village activities under the Dana Desa grant, transferred by the central government, accounting for only half of total village resources. This report represents the largest assessment of village spending since the enactment of the Village Law. The findings presented here are based on a public expenditure analysis of financial data from 45,625 rural villages across 248 districts and 33 provinces.1 This represents over 58% of all villages, although coverage of villages by province varies between 5% in Papua Barat to 100% in Bali.2 The objective of this report is to provide an initial assessment of Indonesia's village spending and budget credibility, which will in turn strengthen the ability of the Government of Indonesia and village administrations to solve the pressing problems facing rural communities. The report is structured into four parts: analysis of village reporting, expenditures, revenues, and performance. Each section aims to answer the following questions: • Reporting: Are villages submitting financial reports, and are reports credible? • Expenditures: What are village spending priorities, how much do they spend, and how much do priorities vary across villages? • Revenues: What is the composition of village revenues, how does it vary across villages, and how are villages spending revenues from different sources? • Budget Credibility: Are village budgets credible, and are villages spending according to their plans? (Based on an adapted methodology from the PEFA framework on Budget Reliability) 133 provinces have rural villages in Indonesia. 2Villages for the PER were not randomly selected, by include all villages for which financial data had been submitted into MOHA’s Siskeudes National Consolidation Application. 5 Summary of Findings Part 1: Village Reporting Analysis Finding 1. Nearly 95% of villages across Indonesia use Siskeudes, while data from nearly 60% of villages is submitted into the National Consolidation Application by district governments. The likely reasons for the gap in reporting are the lack of sanctions for districts that do not submit consolidated reports, limited capacity of district staff to use Siskeudes, and internet access and reliability. Finding 2. The majority of village reports included both the original budget and actual expenditure data. Most village reports also pass basic data validity tests; however, poor district guidance may impact reliability of reports. While 92% of reporting village in FY 2019 submitted both original budgets and actuals, gaps remain in other data points, including revised budgets, outputs data, assets, and liabilities. Distribution analysis shows that villages with data validity errors are clustered in the same districts, suggesting poor policies or guidance from the district may impact village performance. Part 2: Village Expenditure Analysis Finding 3. The largest share of village spending is for public works (36%), followed by salary and operations (26%), housing/sanitation (7%), health (4%), and education (4%). The majority of public works expenditures were for construction, rehabilitation, and maintenance of roads and bridges. Village spending on salary and operations was 4% below the 30% limit set by national regulation. While housing and sanitation composed the third-largest category of spending, within this category, improved access to water only amounted to 1.7% of village spending, despite the major gap in access to potable water in Indonesia. Finding 4. Across all regions, the majority of village expenditures (80%) go toward capital and goods and services. This pattern is consistent across all regions. Expenditure on personnel averages at 19% nationally, ranging from 16% in Papua to 20% in Java. Finding 5. Village priorities vary significantly across provinces. Spending on public works ranges between 50% of total spending in Bengkulu and Banten to less than 7% in Papua Barat. Spending on housing and sanitation varies from a high of over 20% in Gorontalo and Papua Barat to less than 5% in Banten, Riau, and Yogyakarta. Finding 6. Since 2016, villages have increased spending on housing/sanitation, education, and health, while reducing their share of spending on public works and salary and operations. Extrapolating spending shares to nominal spending for all villages suggests this represents a 130% increase in education spending and an 85% increase in housing/Sanitation, health, and Agriculture. Finding 7. 97% of villages invest in stunting-related activities, at an average rate of 12%. Villages in most provinces with high stunting prevalence spent a relatively large share of their budget for stunting-related interventions. Papua has the highest share with nearly 30%, followed by NTT and Gorontalo at nearly 20%. 6 Part 3: Village Revenue Analysis Finding 8. Dana Desa makes up just over half of total village resources. Despite the fact that most consolidated analysis of village revenues focused only on Dana Desa until 2019, central government transfers account for only half of total village revenues. However, inter- government transfers (including district and provincial transfers) account for nearly 97% of all village revenues. Finding 9. Share of village revenues from Dana Desa Ranges from 23% in Bali to 81% in Papua. The Dana Desa is not a per capita allocation. Less populated areas with higher poverty rates receive higher per capita allocations. However, it is important to note that even in poor and lagging regions, villages still have significant revenues from other sources – highlighting the need for a holistic analysis of village spending. Finding 10. Village Own-Source Revenue (PADes) contributes only 3% to overall village revenues, with most of this revenue coming from villages in Java and Bali. Even though most villages projected to raise own-source revenue in their original budget, 61% of all villages could not generate PADes. Between 2013 and 2019, PADes has declined both as a share of total revenues and nominally. The increasing inter-governmental transfers may disincentivize village governments from improving their PADes generation. Finding 11. Spending priorities shift significantly between different fund sources. Currently, the central government and Ministry of Finance exclusively analyze village spending under Dana Desa. However, Dana Desa expenditures alone do not provide a complete picture of village spending and prioritization. Part 4: Village Budget Credibility Analysis Finding 12. 75% of villages receive an A or B grade in a modified PEFA assessment of village budget reliability, indicating high budget reliability and financial management capacity. Less than one-fifth of villages scored a C, and only 6% received a D score. Finding 13. There is a large variation in budget reliability between regions. Over 70% of villages in Sulawesi received A and B grades, compared to less than 40% in Java. The five provinces with the highest share of villages with A and B grades are located in Sulawesi, Sumatra, and Papua. Finding 14. Predictable resourcing leads to better budgets. Analysis of the distribution of average village expenditure and revenue scores shows that villages can develop credible budgets when village revenues are predictable. The revealed correlation between the budget revenue and expenditure execution performance aligns with Public Finance Theory, highlighting the importance of predictable resourcing. 7 Future Directions The availability of village financial data provides the government with a unique opportunity for evidence-based policymaking, targeted support to villages, and performance-based incentives to villages and districts. This report serves as a first assessment of data from the Siskeudes National Consolidation Application, but it also aims to demonstrate the opportunity to use village financial data to measure the impact of village funds and inform future policy direction. The 2019 ViPER findings provide some relevant policy indications for the government and its partners that support Village Law implementation. 1. Improving the availability of complete consolidated village financial data will require better incentives to districts and clear sanctions for non-compliance with reporting requirements. 2. Strengthening the financial performance of villages requires better support and monitoring of both villages and districts. 3. Building a local-level interface to share information back to districts and villages can improve data quality and support local planning. 4. Leveraging financial data for evidence-based policymaking requires a holistic analysis of all village resources, not only Dana Desa. 5. Provincial spending patterns can inform sectoral technical assistance and allow for more strategic targeting of village-level support by line ministries and directorates. 6. Institutionalizing village expenditure analysis, such as the ViPER methodology in this report, will support policymaking for the Ministry of Home Affairs, the Ministry of Finance, and other government institutions that focus on poverty reduction and frontline service delivery. 8 1. INTRODUCTION Since 1999, Indonesia has undertaken an administrative and fiscal decentralization from the national to district, and more recently, to the village level. This has led district and village governments to have substantial financial resources and authority to oversee and deliver basic services, including health, education, and infrastructure development. In 2014, Indonesia's Village Law (No. 6/2014) ushered in a new chapter in the country's decentralization and rural development. One of the largest village decentralization programs globally, the Village Law substantially increases the autonomy, responsibilities, and funding transferred to Indonesia's nearly 75,000 villages, building on Indonesia's rich history of community-driven development. The Law states that the objective of village development is to improve villagers' welfare and quality of life and to alleviate poverty through fulfillment of basic needs, provision of village infrastructure, development of local economic potential, and sustainable utilization of natural resources and environment. The Village Law is a key instrument of the Government's National Medium-Term Development Plan (RPJMN) 2020 to 2024 to deliver basic services, increase accessibility, and contribute to national economic growth. Since the Government of Indonesia began implementation of the Village Law, more than IDR 547 trillion (US$ 38 billion) has been transferred to Indonesia's nearly 75,000 villages.3 Despite the challenges, in the first year of Village Law implementation, the government successfully disbursed nearly 95% of village funds through inter-governmental transfer mechanisms.4 Over the next four years, transfers to villages more than doubled. During this period, the government introduced a new system for village public financial management, including a village-specific chart of accounts. The new classification system was integrated into a digital village financial management system (Siskeudes), which has helped improve and simplify the monitoring of village budgets and expenditures across Indonesia's nearly 75,000 villages. In a relatively short time, the government has established the legal and regulatory framework for the Village Law, significantly increased transfers to the village level, mobilized large-scale capacity support, and upgraded policies and systems to monitor village performance. The government has allocated Dana Desa across villages based on a pro-poor formula, required villages to report on outputs in addition to expenditures, and is paying increasing attention to improve the quality of village spending. More than 25,000 facilitators are deployed by the Ministry of Village to support implementation, including technical infrastructure support. There has been significant and increasing high-level government attention on villages and Village Law implementation, including multiple cabinet and inter-ministerial meetings, presidential visits, and focus on Village Law implementation and its impact. The Village Law reaches 176 million people in Indonesia, including 117 million people in rural areas, constituting most of the country's poor.5 Village funds can reach Indonesia's rural poor with improved access to infrastructure and basic services; help to reduce stunting and improve village health, nutrition, and early childhood education; empower women, youth, and vulnerable groups; provide "last-mile" rural connectivity; and strengthen local natural resource management, disaster management, and climate resilience. As of 2019, with the adoption of the Siskeudes National Consolidation Application, the Government of Indonesia has access to an unprecedented level of village financial information, including the activities villages are prioritizing in their plans, village expenditures, and village revenues. With improved data and 3 World Bank estimation 4 Hans Antlöv, Anna Wetterberg, Leni Dharmawan, Village Governance, Community Life, and the 2014 Village Law in Indonesia, August 2016 5 World Bank, No One Left Behind, 2020 9 analysis of financial and non-financial data, the government will be able to better monitor and assess the impact of village funds on national development objectives. The objective of this report is two-fold: first, to provide a broad analysis of village budgets, and second to provide a proof-of-concept for how village financial data can be used to assess village budget reliability. This report summarizes findings from a public expenditure analysis based on 2019 village financial data from 45,625 rural villages across 248 districts and 33 provinces.6 This represents the largest assessment of village spending in Indonesia since the enactment of the Village Law. 2. VILLAGE FINANCIAL MANAGEMENT Village Resources Since Introduction of Village Law Between 2014 and 2019, village governments saw their budgets increase by 360 percent . The increase in funding was driven, in large part, by the introduction of the Village Law and the Dana Desa inter- governmental transfer in 2015. Since then, transfers to villages have been steadily increasing year-over- year (Table 1 and Figure 1). In 2019, the Government of Indonesia transferred an average of nearly IDR 1.6 billion (over US$107,000) to each village. The total value of fiscal transfers to villages was US$8.1 billion in 2019, 4.3% of the national budget, and 0.7% of the GDP (Figure 2). Table 1: Village revenue trends before and after the Village Law (in trillion IDR, nominal) 7 2013 2014 2015 2016 2017 2018 2019 Village Own Source Revenue (PADes) 4.1 4.2 4.2 3.5 3.1 3.5 2.9 Transfer Revenue 17.5 21.3 47.2 78.3 96.7 98.0 113.4 Dana Desa (DD) - - 19.5 45.6 57.6 56.9 67.3 Shared Tax and Levies from District Govt (BH-PDRD) 0.6 0.9 1.7 2.0 2.5 3.0 3.6 Alokasi Dana Desa (ADD) 8.1 10.2 22.8 26.4 30.5 31.8 35.2 Financial Assistance (from Central/Province/District) 8.8 10.1 3.2 4.3 6.1 6.2 7.3 Other Revenue 1.0 1.1 0.6 0.5 0.5 0.7 1.2 Total Village Revenue IDR (trillions) 22.6 26.7 52.1 82.3 100.2 102.1 117.4 Total Village Revenue USD (billions) 1.6 1.8 3.6 5.7 6.9 7.0 8.1 6 33 provinces have rural villages in Indonesia. 7 The data for 2015-2019 is based on realized/actual revenue (BPS). The BPS does not record realized DD as separate item from realized bantuan keuangan in 2015. The DD data of 2015 in the table use Ministry of Finance (MoF) estimates of 93.8% of disbursed IDR 20.8 trillion of DD allocation, which in turn adjusts BPS data of bantuan keuangan of respected year from IDR 22.7 trillion to become IDR 3.2 trillion. The DD amount of 2016 , 2017, 2019 are BPS data which show smaller figures compared to MoF data on DD allocation at IDR 47 trillion (2016), IDR 60 trillion (2017), 60 trillion (2018), and 70 trillion (2019). The BPS estimates a smaller amount of ADD and BH-PDRD compared to MoF estimates. For ADD, MoF estimates at IDR 33.8 trillion, IDR 35,5 trillion, and IDR 34.1 trillion in 2015, 2016, and 2017 respectively, while for BH-PDRD the MoF estimates at 2.7 trillion, 2.8 trillion, and 3.2 trillion. The USD figure is based on exchange rate of 1 USD = 14.500 IDR 10 Figure 1: Village revenues 2013 to 2019 (in trillion IDR, nominal) 120 100 80 60 40 20 0 2013 2014 2015 2016 2017 2018 2019 DD ADD Financial Assistance BH-PDRD PADes Other Revenue Figure 2: Village government expenditure as a share of national public spending (in trillion IDR)8 3,000.0 5.0% 4.3% 4.5% 2,500.0 3.8% 4.3% 4.0% 4.0% 3.5% 2,000.0 3.0% 2.4% 1,500.0 2.5% 2.0% 1,000.0 1.3% 1.2% 1.5% 1.0% 500.0 0.5% - 0.0% 2013 2014 2015 2016 2017 2018 2019* Village Govt Expenditure Subnational Govt Expenditure (excluding transfer to other SNGs and villages) Central Govt Expenditure (excluding transfer to SNGs and villages) Share of Village Govt Expenditure (%, right axis) 8World Bank calculation based on LKPP (CG expenditure), DG Fiscal Balance Database (SNG expenditure), and BPS’s Financial Statistics of Village Government, multiple years. Village govt expenditure data of FY-2019 is provisional data. 11 Village Revenue Sources Reporting on village resources to date has almost exclusively focused on Dana Desa spending, which only makes up about half of total village revenues. However, village budgets consist of three revenue categories: inter-governmental transfers, village own-source revenue, and other revenue. Table 2 describes these revenue sources. The breakdown of village revenues is further analyzed in Part 3. Table 2: Description of Village Revenue Sources Village Revenue Source Description Transfer Revenue Revenue from inter-governmental transfers as listed below. Dana Desa ("village fund") is first mentioned in government regulation (PP) No. 60/2014 on Dana Desa to refer to the mandate of the Village Law on the transfer to villages from the central government budget. The Village Law states that this transfer should be 10% of, and in addition to, transfers to the regions (see Article 72 of the Village Law). Currently, the Dana Desa allocation Dana Desa (DD) formula includes a. Basic allocation (65%) distributed equally to all villages; b. Affirmative allocation (1%) given to lagging and extremely lagging villages; c. Performance allocation (3%) for top performing 10% of villages in each district; d. Formula allocation (31%) based on: population (10%), poverty rate (40%), village area (20%), and accessibility (30%). Alokasi Dana Desa ("village allocation fund") is a block grant transfer from the district government to village governments. ADD was established before the Alokasi Dana Desa Village Law (Permendagri 37/2007). However, under the Village Law, a new (ADD) provision requires districts to allocate 10% of balancing funds, excluding earmarked funds (such as Dana Alokasi Khusus/DAK), to villages. BH-PDRD is local taxes and levies shared between districts and villages. Regional Tax Revenue Similar to ADD, BH-PDRD was established before the Village Law was Sharing and Regional enacted. However, districts must transfer at least 10% of district taxes and Retribution (BH-PDRD) levies to villages under the Village Law. Financial Assistance constitutes transfers from central, provincial, and/or Financial Assistance district governments that may be ad-hoc or based on local regulations and is (Bankeu) additional to DD, ADD, and BH-PDRD. Village own-source revenue is generated directly by the village, including Village Own-Source returns from village assets, investments, and community contributions. Revenue (PADes) Village assets revenue includes rent from village land and revenue from village-owned enterprises (BumDes). Other revenues include grants and contributions given to villages by third Other Revenue parties, including NGOs, the private sector, and other contributors. 12 Village Financial Management Regulations In 2018, the Government of Indonesia updated the village public financial management system through Permendagri 20/2018, which significantly improved village budget systems. The regulation provides three functional classification levels for village expenditures. These are shown in Table 3. a. Bidang: Based on village authorities as described under the Village Law. b. Sub-Bidang: This classification was introduced in the new regulation. The Sub-Bidang is harmonized with sectoral categories (Urusan) used by other levels of government. The village expenditure can be classified and aggregated sectorally for education, health, public works, and other functional sectors. c. Activity: Activity-level classification was also introduced in the new regulation. This new classification level allows village expenditures to be aggregated at a much more detailed level, which was not previously possible. Village Financial Reporting Since 2019, the Ministry of Home Affairs has begun the full roll-out of the Siskeudes National Consolidation Application to unify different reporting requirements and ensure the availability of complete village financial data. Nearly all villages in Indonesia are now using Siskeudes to manage their budget and expenditures. The new village chart of accounts has been integrated into Siskeudes and has helped standardize budget classification and enable automated analysis of village financial data from 2019 onward. Previously, consolidated village financial reports were only available for Dana Desa. Villages report all budgets and expenditures to the district government, who then submit Dana Desa reports to the Ministry of Finance (MoF) through OMSPAN (Online Monitoring for State Treasury and Budget System). OMSPAN is a system used by MoF to monitor the budget and disbursement execution by all government entities, including districts. This is the data used for verification of Dana Desa tranche transfers. The government has developed a mechanism to allow districts and villages to use Siskeudes to produce the Dana Desa reports that the district can then upload into OMSPAN. Until 2019, the only consolidated village financial data available were Dana Desa reports through this system. Villages must submit complete financial reports to the district government; however, until the Siskeudes National Consolidation Application roll-out, these reports could not be consolidated and used for meaningful analytics. Districts required all villages to submit their financial reports to be annexed to district financial reports audited by Supreme Audit Board (BPK). Audited District Financial reports are submitted to the District Parliament (DPRD) for accountability evaluation. These reports are still submitted for each village and not consolidated. These reports may be attached as hard copies or in PDF format and cannot be analyzed and consolidated. Some districts have already replaced this process by using Siskeudes, but there is an opportunity to support more districts to adopt this practice, reduce reporting requirements on villages, and streamline the overall data collection process. 13 Table 3: Village Chart of Accounts Summary Bidang Sub-Bidang Summary of Select Activities • Wages and salaries and social security for Village Heads and Apparat Provision of Fixed Income & Allowances, and GOVERNMENT ADMINISTRATION • Village government operations (equipment, utilities, allowances, etc.) General Village Government Operations • Allowances and operational cost for Village Council (BPD) and RT/RW heads • Provision of fixed assets for village office Village Administration, Facilities, and Infrastructure • Construction, rehabilitation, maintenance of village office infrastructure • Provision and community mobilization for civil registration Civil Registration, Statistics, and Archives • Data collection and management, participatory mapping, and analysis of poverty • Village deliberations, such as MusDes and planning, including RPJMDes, APBDes Village Planning, Budgeting, Accounting, Reporting • Village Information System development • Land certification, registration, and conflict mediation Land Administration • Property tax administration (PBB) and village boundary development/determination • ECD: operations, teacher allowances, equipment, construction, maintenance Education • Support and scholarships for students from poor households • Operation, construction, maintenance of village health posts and Posyandu services Health • Community health education and training VILLAGE DEVELOPMENT • Construction and maintenance of village roads, bridges, community centers Public Works and Spatial Planning • Village social mapping and spatial planning • Construction and maintenance of village water and sanitation infrastructure Housing/Sanitation • Construction and maintenance of parks and playgrounds Forestry and the Environment • Environmental management and community awareness-raising Transportation, Communication, and Information • Village signs, posters, public information material Energy and Mineral Resources • Development and maintenance of alternative energy facilities and infrastructure Tourism • Village-level tourism development Social Order and Community Protection • Village security posts, local disaster preparedness, village Satlinmas capacity building DEVELOPMENT COMMUNITY Culture and Religion • Arts, cultural, and religious festivals, and management of cultural infrastructure Youth and Sports • Support for infrastructure and operations of youth organization, sports, and clubs Community Organization Empowerment • Development and operation of customary village organizations, such as PKK • Support for village fisheries, including construction of village-owned Karamba / fisheries ponds / Marine and Fisheries small river fishing ports. • Training and technical assistance on fisheries and introduction of new technology COMMUNITY EMPOWERMENT • Strengthening village-level food securities, including development of village granaries Agriculture and Livestock • Maintenance of tertiary irrigation infrastructure • Support for agriculture and livestock programs, including input provision and training Capacity Building for Village Apparatus • Training and capacity building for village head, officials, and BPD Women's Empowerment, Family Planning, and • Training and community mobilization on women's empowerment, child protection, and Child Protection support for people with disabilities Coops, Micro, Small and Medium Enterprises • Development of micro, small, and medium enterprise infrastructure and cooperatives Investment • Establishment and support for BumDes Trade and Industry • Development and maintenance of village markets Disaster Response • Disaster response RESPONSE DISASTER Emergency • Emergency Contingency • Contingency 14 3. METHODOLOGY The village public expenditure analysis, on which this report is based, is the largest assessment of village spending in Indonesia since the enactment of the Village Law. The data utilized to undertake this analysis covers 43,675 out of 74,957 villages in Indonesia, 58% of all villages. It presents data from villages across all 33 provinces and 287 districts out of 434 districts with rural villages. The analysis uses financial data from FY 2019 of all 43,675 villages reported in the system. The selection is not based on any random sampling method. The specific coverage and breakdown of 2019 Siskeudes data is further provided in Part 1 and shown in Figure 3. A complete breakdown of data coverage can be found in Annex A. This analysis aims to assess the achievements and challenges of Indonesia's village fiscal performance in different dimensions and resolutions, which will, in turn, strengthen the ability of the Government of Indonesia and village administrations to solve the pressing problems facing the government and citizens. The report is structured into four parts: analysis of village reporting, expenditures, revenues, and budget credibility. Each section aims to answer the following questions: 1. Reporting: Are villages submitting financial reports, and are reports credible? 1. Expenditures: What are village spending priorities, how much do they spend, and how do priorities vary across villages? 2. Revenues: What is the composition of village revenues, how does it vary across villages, and how are villages spending revenues from different sources? 3. Budget Credibility: Are village budgets credible, and are villages spending according to their plans? Part 1 of the report covers the current status of village financial reporting, both coverage and data credibility. The coverage compares the percentage of villages using Siskeudes and the number of districts reporting consolidated village data. All 43,675 village reports for FY 2019 are analyzed in this section. In this section, a "Cut and Paste" test was performed on data to identify whether villages had copied original budget data in place of actual expenditure data. The exercise was a simple test to assess the validity of actuals. Part 2 summarizes village spending, compares 2019 spending with a public expenditure analysis of village funds conducted in 2016, and specifically analyses village spending related to stunting prevention. Findings are based on FY 2019 village expenditure realization reports from 36,242 villages. In Part 2 of the report, expenditure categories are based on village revenue realization, broken down by sub- Bidang classification used by villages are per village financial management regulation. However, expenditures were also cross-coded at the activity level with GFS standards (see Annex B) to conduct the modified PEFA analysis calculations. Expenditure analysis based on economic classification is also provided under finding 6, which is already embedded as part of the village chart of accounts. Part 3 summarizes trends related to village revenues. A description of each revenue source and different inter-governmental transfers to villages is provided in Table 2. The analysis in Part 3 of this report is based on an analysis of 36,929 villages that had submitted complete revenue data for FY 2019. Part 4 presents results from a modified PEFA assessment to review village budget reliability by comparing original budgets and expenditures. The Modified PEFA budget reliability assessment undertaken for villages is adapted from the 2016 PEFA Framework, using five standard PEFA indicators on budget reliability. The analysis has adjusted the time horizon of the analysis to one year, as at this time, village data is only available for FY 2019. The analysis in this section covers 33,220 villages that had 15 provided original and actual data for revenues and expenditures. Methodological details on this component are provided in Part 4, and the breakdown of scoring for each dimension is in Annex D. The analysis is limited in some areas due to the lack of data availability. In terms of coverage, Figure 3 shows the coverage of villages by provinces (yellow bars). Findings from five provinces represent less than 50% of villages, including Papua Bara, Papua, Sulawesi Tenggara, Kamilantan Utara, and NTT. Therefore, future analysis will be required once data coverage for these provinces increases to ensure the accuracy of patterns identified in the current analysis. There are also limitations in the scope of data availability, including district budgets and output data. This prevents meaningful analysis on economic efficiency, complementarity of village and district spending, and reliability of district transfers. Figure 3: Villages Using Siskeudes and Report Availability in Siskeudes National Consolidation Application 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Bali Kep. Riau Bengkulu Kalimantan Selatan Jambi Riau Kalimantan Barat Gorontalo Kalimantan Timur Sumatera Barat DIY Sulawesi Tengah Kep Babel Sulawesi Barat Jawa Barat Jawa Tengah Jawa Timur Maluku Sulawesi Utara NTB Kalimantan Tengah National Average Sulawesi Selatan Maluku Utara Banten Lampung Sumatera Selatan Sumatera Utara Aceh NTT Kalimantan Utara Sulawesi Tenggara Papua Papua Barat % of Villages Using Siskeudes % of Consolidated Data Submission 16 4. PART 1: VILLAGE REPORTING ANALYSIS Finding 1. Nearly 95% of villages across Indonesia use Siskeudes, while data from nearly 60% of villages is submitted into the National Consolidation Application. With the development of Siskeudes, the Government of Indonesia now has access to an unprecedented level of village financial information. Critically, because of a unified village chart of accounts, budgets and expenditures can be aggregated and compared across villages. Siskeudes is the only means by which the complete village budget (APBDes) is reported. Across Indonesia, nearly 95% of villages are using Siskeudes as their financial management system. Most provinces have nearly full coverage of Siskeudes, with only Papua, Papua Barat, Sumatera Barat, and Kalimantan Utara falling below the national average of 95% coverage. Many villages use an offline version of Siskeudes due to limitations with internet access and reliability. For villages using Siskeudes offline, they submit financial reports to districts, who then submit them to the National Consolidation Application. For the 2019 fiscal year, village financial reports from 58% of villages were available in the Siskeudes National Consolidation Application. The data covers 100% of reporting provinces, 66% of reporting districts, and 61% of reporting kecamatan. The coverage by province varies between 5 percent for Papua Barat to nearly 100% in more than half of provinces (Figure 10). 20 out of 33 provinces have village coverage above the national average of 58%. Four provinces remain below 25% village coverage (Papua, Papua Barat, Sultra, and Kalut). Figure 3 shows the gap between the adoption of Siskeudes by villages and the availability of financial reports in the National Consolidation Application. While Siskeudes can support village financial management, there is also an opportunity to improve district capacity to consolidate and report comprehensive village financial data to MOHA. The likely reasons for the gap in reporting are the lack of sanctions for districts that do not submit consolidated reports, limited capacity of district staff to use Siskeudes, and internet access and reliability. As discussed in section 2.4, Dana Desa tranche transfers are based entirely on financial data on Dana Desa itself. Therefore, while many districts submit the data into the Siskeudes National Consolidation Application, the regulatory requirement (see Permendagri 20/2018) is still weak and lacks clear sanctions for districts that do not meet reporting requirements. In some cases, districts also have difficulties managing the Siskeudes consolidated database using SQL due to lack of staff capacity; however, as MOHA scales up district training for Siskeudes, the consolidated reporting should increase. Finally, limited access to the online Siskeudes platform can create challenges for both villages and districts, as the report sharing cannot be automated for villages using Siskeudes offline, which creates challenges for districts with poor internet coverage and reliability. 17 Finding 2. Most village reports included both original budget and actual expenditure data, and the majority passed basic data validity tests; however, poor district guidance may impact reliability of reports. Most reporting villages (91.5%) submitted both original Figure 4: Reporting completeness budgets and actual expenditures for FY 2019 into Siskeudes. This number is particularly impressive as 2019 was the first year that most villages were reporting based on the new financial management regulation and against the updated village chart of accounts. 91.5% of analyzed reports A "Cut and Paste" test of village revenues and expenditures included budget and was performed to identify whether villages had copied original expenditure data budget data in place of actual expenditure data. The exercise was a simple test to assess the validity of actuals. Less than 2% of villages had identical budgets to actual expenditures, and under 4% had identical budgets to actual revenues. Budget and Expenditures No Expenditure Data The distribution of invalid revenues by frequency (Figure 6) shows that district policies may play a factor in the No Original Budget reliability of village budgeting. This breakdown of findings down to the district level shows that villages across each Figure 5: Cut and paste test on district generally have similar budget reliability. Figure 6 revenue and expenditure reports shows that most districts had zero villages with unreliable reports, while 110 districts had over 90% of villages with unreliable reports. However, there were gaps in village reports, particularly for submission of revised budgets, output data, village assets, 96% of villages passed and liabilities. Villages may need more support and revenue data test incentives to complete and submit these reports. Figure 6: Frequency of unreliable (Rev) reports by district 200 Number of Districts 150 100 98% of villages passed 50 expenditure data test 0 Percentage of Unreliable Reports Reliable Unreliable 18 5. PART 2: VILLAGE EXPENDITURE ANALYSIS Finding 3. The largest share of village spending is for public works (36%), followed by salary and operations (26%), housing/sanitation (7%), health (4%), and education (4%). New nomenclature for village financial management has allowed the Government of Indonesia to analyze village budgets based on priority investment areas. Expenditure analysis based on sub-Bidang describes the purpose of spending, rather than economic classifications. Table 3 provides more detail on the activities that fall under each sub-Bidang. The largest share of village spending was on public works and spatial planning category, which constituted over one-third of all village expenditures (36%). The majority of these expenditures were for construction, rehabilitation, and maintenance of roads, bridges, and road-related infrastructure (such as culverts, road drainage, and ditches). This category also includes other infrastructure spending, such as community centers, cemeteries, and village-owned historic sites. Infrastructure content makes up 90 percent of the share in this spending category. Non-infrastructure spending (which makes up 10% of this component), such as spatial planning and social maps, is also categorized here. Villages spent less than 26% of their total spending on Village Apparatus Salary and Operations, 4% below the 30% cap set by the central government in 2019. This component consists of the wages and salaries for village government officials, including the village head; allowances for village council representatives (BPD); allowances for neighborhood and hamlet heads (RT/RW); operation costs for the village government and BPD – including office supplies, office equipment, uniforms, and utilities such as electricity, internet, and telephone. In 2019, the central government mandated that villages only spend a maximum of 30% of total spending to finance salaries and allowances for village government officials and allowances and operational cost of the village council (BPD).9 However, the average village spending on Village Apparatus Salary and Operations was more than 4% below the limit set by government regulation. The third-largest spending category was housing and sanitation at 7% of total village spending. One- quarter of this spending (25%) was for poor households (GAKIN) under the Rumah Tidak Layak Huni (RTLH) program, providing beneficiaries with in-kind assistance to improve their housing conditions, such as roofs, floors, and walls. Excluding RTLH spending, the total portion of funds villages spent on improving housing, sanitation, and access to clean water comes to just over 5 percent of total expenditures. Total spending on clean water projects amounted to 1.7% of village funds and 2.23% on sanitation. Nearly half of the sanitation component consisted of the construction or maintenance of public toilets. The low levels of spending on water and sanitation contrast with village prioritization in past community-driven programs, where water supply and sanitation investments consisted of more than double the current share of expenditures.10 Additional qualitative analysis may provide insight on lower village spending in these sectors, particularly 9 Based on PP 11/2019, Second Revision on PP No. 43/2014 on Implementation of Village Law. Note: calculation of village spending under the regulation excludes income from village-owned land. 10 PNPM Rural Annual Report 2007 19 considering the low rural access to drinking water and improved sanitation in Indonesia (nearly 20% of households lack access to drinking water, and 36% lack access to basic sanitation).11 The next two largest expenditures were for health and education, accounting for 8.3% of all village expenditures. Under health, the largest share of spending (nearly 41%) was for Posyandu services, including supplementary feeding programs, incentive for Posyandu Cadre, and classes for pregnant women. The next largest category is construction or rehabilitation of village health infrastructure (26% of health spending), such as Posyandu and Polindes. This is followed by the implementation of public health campaigns and messaging (11%). Under the education sub-Bidang, the largest categories of investments are construction and rehabilitation of Early Childhood Development (ECD) facilities, including PAUD (34%) nearly equal to expenditure on the operation of ECD facilitates (34%), including teacher honorarium, supplies, and utilities. Overall, infrastructure spending makes up 47% of village education spending and 29% of village health spending. Figure 7: Village Expenditure by Sub-Bidang (2019) Public Works and Spatial Planning 36.01% Village Apparatus salary and operations 25.93% Housing/Sanitation 7.06% Health 4.28% Education 4.00% Culture and Religion 3.30% Village Office Facilities and Infrastructure 3.06% Village Planning, Budgeting, and Reporting 2.49% Agriculture and Livestock 2.14% Community Organization Empowerment 2.04% Youth and Sports 2.03% Capacity Building for Village Apparatus 1.25% Social Order and Community Protection 0.86% Trade and Industry 0.80% Tourism 0.56% Women's Empowerment & Child Protection 0.55% Energy and Mineral Resources 0.54% Marine and Fisheries 0.54% Transportation and ICT 0.53% Investment 0.44% Land Administration 0.42% Coops and MSME 0.42% Civil Registration, Statistics, and Archives 0.34% Forestry and the Environment 0.27% Disaster Response 0.06% Contingency 0.03% Emergency 0.02% Others 0.02% 11 WHO/UNICEF Joint Monitoring Programme (JMP) for Water Supply, Sanitation and Hygiene 20 Finding 4. Across all regions, the majority of village expenditures (80%) go toward capital and goods and services. 80% of village expenditures go toward capital investments and goods and services, a pattern that is consistent across all regions. There is, however, greater variation between the share of capital compared investments to goods and services expenditures between regions. Capital investments make up more than 55% of expenditures in Java compared to less than 35% in Bali and Nusa Tenggara. In contrast, spending on Goods and Services is lowest in Java at under 25%, and highest in Bali and Nusa Tenggara at over 46%. Expenditure on personnel is similar across regions, averaging at 19%. The exception is in Papua, where spending on personnel makes up a smaller share of the total budget, at 16%, than in other regions. Between all other regions, the share ranges from 20% in Java to 17% in Sumatra. Less than 0.2% of village budgets go toward contingency spending. The highest percentage is in Papua, where contingency spending accounts for 0.17% of total spending. Bali and Nusa Tenggara have the lowest share of contingency spending, at only 0.04%. Figure 8: Village spending by economic classification 0% 20% 40% 60% 80% 100% Java 55% 25% 20% Sumatra 54% 28% 17% Kalimantan 53% 29% 18% Sulawesi 49% 31% 19% Papua 44% 41% 15% Bali & Nusa Tenggara 35% 46% 19% Capital Goods and Services Personnel Contingency Finding 5. Village priorities vary significantly across provinces. Village priorities vary across provinces, but in most provinces, public works has the largest share, followed by spending on village apparatus salary and office operation. There is significant variation for the third-largest priority, ranging from housing/sanitation, education, health, or agriculture. In Figure 8, provinces listed between Sumatera Barat to Gorontalo spend a larger share of their budget on salary and operations than public works. In Papua, the largest share of the spending is devoted to agriculture (36%), followed by housing and sanitation (15%). In Papua Barat, the largest share is devoted to housing and sanitation (25%). Villages in Papua Barat and Maluku also invested a significant portion of their budget on marine and fisheries (19%). Villages in Bali spent the largest share of their budget on Culture and Religion (25%), much of which may be from provincial allocation designated specifically for such activities. 21 Figure 9: Top five spending priorities by province Bengkulu 49.9% 24.0% 6.5% 4.5% 3.0% 12.2% Banten 48.6% 25.3% 4.3% 7.8% 3.1% 10.9% Jawa Tengah 46.2% 22.4% 7.3% 3.3%3.3% 17.5% Lampung 44.2% 23.9% 6.5% 5.7% 5.5% 14.2% Sumatera Utara 44.1% 21.9% 7.6% 3.6% 4.7% 18.2% Sumatera Selatan 40.4% 23.9% 6.2% 3.9% 4.0% 21.6% Jawa Barat 39.9% 28.1% 6.0% 4.6% 3.4% 18.0% Sulawesi Barat 39.5% 23.8% 9.0% 4.5% 4.3% 18.9% Jawa Timur 39.2% 29.5% 5.2% 3.9% 4.4% 17.8% Sulawesi Selatan 37.4% 26.0% 8.1% 4.4% 4.0% 20.1% Kalimantan Selatan 36.6% 26.6% 6.0% 6.0% 6.7% 18.1% Kalimantan Barat 36.1% 27.8% 6.0% 4.3% 4.7% 21.2% Jambi 34.9% 24.2% 5.6% 7.8% 4.6% 22.8% Kalimantan Tengah 34.7% 27.9% 5.1% 5.2% 5.5% 21.6% Riau 34.0% 29.7% 4.9% 5.6% 4.9% 21.0% Sulawesi Utara 32.3% 27.3% 10.2% 4.7% 3.8% 21.6% NTT 31.8% 23.2% 16.6% 6.8% 6.0% 15.6% DIY 30.5% 29.0% 4.8% 3.7% 6.7% 25.2% Sulawesi Tenggara 29.0% 24.4% 9.2% 5.3% 4.9% 27.1% Maluku Utara 28.0% 23.9% 10.3% 5.3% 6.5% 26.1% Aceh 27.7% 18.5% 13.2% 4.2% 9.4% 26.9% Sumatera Barat 27.4% 28.7% 8.0% 5.0% 4.8% 26.0% Kalimantan Timur 27.4% 29.0% 5.6% 4.6% 4.3% 29.1% Kalimantan Utara 25.1% 28.5% 6.9% 4.8% 4.6% 30.2% Kep. Riau 24.7% 29.4% 4.6% 4.7% 6.5% 30.0% NTB 23.8% 28.4% 9.5% 7.8% 4.5% 26.1% Sulawesi Tengah 20.7% 27.5% 9.8% 6.7% 7.2% 28.1% Bali 18.2% 23.4% 4.6% 24.5% 5.4% 23.8% Kep. Babel 15.7% 28.7% 6.2% 6.6% 5.2% 37.6% Maluku 15.2% 26.2% 14.2% 5.3% 7.1% 32.0% Gorontalo 15.0% 23.5% 20.4% 7.4% 5.9% 27.8% Papua 9.3% 8.1% 15.3% 10.9% 36.0% 20.5% Papua Barat 6.5% 13.2% 24.7% 7.7% 18.9% 29.1% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Public Works and Spatial Planning Village Apparatus salary and operations Housing/Sanitation Education Health Agriculture and Livestock Village Office Facilities and Infrastructure Culture and Religion Marine and Fisheries Youth and Sports Village Planning, Budgeting, and Reporting Land Administration Energy and Mineral Resources Community Organization Empowerment Capacity Building for Village Apparatus Others 22 Finding 6. Since 2016, villages have increased spending on housing/sanitation, education, and health, while reducing their share of spending on public works and salary and operations. In 2016, the World Bank conducted a Village Expenditure analysis, using data from 1,868 villages across Indonesia, translating activities into functional classification, which now align with the sub-Bidang categories. The re-classification was used to analyze thousands of spending line items contained in village budgets into functional categories. Beyond looking at how villages used resources in FY2016, the study offered recommendations to improve classification and reporting to better analyze future spending. Based on these recommendations and the government's own analysis, the village chart of accounts was updated in the new financial management regulation. This analysis is used here to draw some comparisons on how the composition of village budgets has shifted between 2016 (the second year of Village Law implementation) to 2019. However, it is important to note that the 2016 data covers a much smaller number of villages. Between 2016 and 2019, the share of spending on village apparatus salary and operations, and public works has been reduced, whereas spending on housing/sanitation, health, education, agriculture, and youth and sports has increased. Although declining as a share of spending, in 2019, public works replaced village apparatus spending as the largest expenditure category compared to 2016. Extrapolating spending shares to total village revenues across the country suggests that the nominal spending on village apparatus salary and operations, and culture and religion, both reduced, while nominal spending for all other sub-Bidang increased. This represents a 130% increase in nominal spending on education and an 85% increase in housing/Sanitation, health, and Agriculture. Nominal spending for public works and spatial planning increased by 30%, well youth and sports increased by 180%.12 Figure 10: Village expenditure priority shift across major sub-Bidang (2016 and 2019) Public Works and Spatial Planning 38.1% 36.0% Village Apparatus Salary and Operations 39.0% 25.9% Housing/Sanitation 5.2% 7.1% Health 3.2% 4.3% Education 2.4% 4.0% Culture and Religion 5.8% 3.3% Agriculture 1.6% 2.1% Youth and Sports 1% 2.0% VIPER World Bank 2016 Others 3.7% 15.2% SISKEUDES 2019 12Estimated nominal spending is based on total national village spending of DR 117.4 trillion (BPS’s finance statistics of vill age, 2020). Calculation is based on available data from 36,242 villages, rather than on statistical sampling methods. 23 Finding 7. 97% of villages invest in stunting-related activities, at an average rate of 12%. Papua has the highest spending share with nearly 30%, followed by NTT and Gorontalo at nearly 20% In August 2017, the Government of Indonesia launched its 2018-2024 National Strategy to Accelerate Stunting Prevention (StraNas Stunting). The StraNas Stunting is centered on the understanding that reduction in stunting requires a coordinated effort across different government agencies and levels of government, including villages. As most services are delivered to citizens at the local level, villages have a critical role in delivering last-mile services to help reduce stunting in Indonesia. Activities toward reducing stunting can be divided into nutrition-specific and nutrition-sensitive interventions. Nutrition-specific interventions improve the quality of nutrition for mothers and children, including micronutrient supplementation programs and operation of Posyandu. Nutrition-sensitive interventions create an enabling environment that can impact nutrition, including access to health care for mothers and infants, food security and agriculture activities, water and sanitation, and early childhood education (see Table 4).13 Table 4: Village-Level Stunting Priority Interventions Function Activities • Operational cost of village health posts (Posyandu) • Specific Posyandu services (supplementary feeding, classes for pregnant Health mothers, incentive for Posyandu Cadre) • Health education and training for Posyandu Cadre • Joint-Caring for Toddler Family Development (BKB) • Operational cost of early childhood development (ECD) facilities (including teacher honorariums) Education • Equipment and supplies for ECD facilities • Construction and maintenance of ECD Facilities • Implementation of Village Health Alert Health • Guidance and supervision for traditional health efforts • Construction and maintenance of village health facilities • Construction and maintenance of village drinking water infrastructure Sensitive • Construction and maintenance of village sanitation infrastructure, Housing/Sanitation including drainage and wastewater disposal • Construction and maintenance of public toilets Forestry and • Training, socialization, and awareness-raising about environment and Environment natural resources Community Organization • Support and service of Family Welfare Movement (PKK) Empowerment Agriculture and Livestock • Strengthening food security, including development of village granaries Women's Empowerment, • Child protection training and counseling Child Protection 13 Lancet, Maternal and child nutrition series, 2013. 24 In total, 97.4 percent of villages spend on stunting-related interventions, with average investments of 12%. The vast majority of villages (86%) spend on both nutrition-specific and nutrition-sensitive interventions. Breaking down the levels of investment shows a wide variance between villages (Figure 11). 10% of villages spent 4% or less on stunting-related interventions, while at the same time, the highest decile spent over 26% of their total budget. The median expenditure was 8.6%, with the share of stunting- related spending clustered at the left tail of the distribution. Despite this, nearly 1% of villages spent 54% or more of their total budget on stunting-related interventions. Table 5: Average and median of village spending on stunting-related interventions Average Median SENSITIVE Intervention Spending Share 10.0% 6.3% SPECIFIC Intervention Spending Share 2.6% 2.0% Overall Stunting Spending Share 12.0% 8.6% Figure 11: Percent of Villages With Stunting-Related Expenditures and Frequency Distribution by Spending None 2.6% 8.9% 2% - <4% 14.2% 13.0% 6% - <8% 10.2% 8.1% 10% - <12% 6.5% 5.8% 14% - <16% 5.0% 4.1% 18% - <20% 3.7% 3.0% Sensitive 22% - <24% 2.6% Comprehensive Only 2.3% (Both Sensitive 26% - <28% 1.7% Specific Only 1.5% and Specific) None 30% - <32% 1.2% 1.1% 34% - <36% 0.7% 0.6% 38% - <40% 0.6% 0.5% 42% - <44% 0.4% 0.4% 46% - <48% 0.3% 0.2% 50% - <52% 0.2% 0.2% 54% or more 0.8% 25 Figure 12: Village Stunting Expenditures by Province 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% Papua NTT Gorontalo Sulawesi Barat NTB Lampung Kalimantan Selatan Sulawesi Tengah Maluku Utara Aceh Kep. Babel Jambi Sulawesi Selatan Sulawesi Utara Kalimantan Tengah Kalimantan Barat Sumatera Selatan Sumatera Barat Kalimantan Utara Bengkulu Maluku Riau Kep. Riau Sumatera Utara Jawa Barat Sulawesi Tenggara Kalimantan Timur Jawa Timur Papua Barat Bali Jawa Tengah Banten DIY Sensitive Intervention Specific Intervention In most provinces with high stunting prevalence (Papua, NTT, Gorontalo, Sulbar, NTB, and Lampung), villages spent a relatively large share of their budget for stunting-related interventions (Figure 12). On average, villages in Papua spent almost 9% of their budget for nutrition-specific interventions and nearly 30% on all nutrition-related activities. Villages in Java and Bali generally spent less on such interventions. Excluding villages in Java and Bali, the average village spending on stunting interventions increases to 14% (compared to 12%), and the mean increases to 13% (compared to 9%). 26 6. PART 3: VILLAGE REVENUE ANALYSIS Finding 8. Dana Desa makes up just over half of total village resources. Nearly half of village revenues are from sources other than Dana Desa, although inter-governmental transfers account for 97% of all revenues (Figure 13). Dana Desa transfers contribute more than half of overall village revenues (53%), followed by ADD (31%), financial assistance from district and provincial governments (9%), and shared revenue from district taxes and levies (4%). Financial Assistance (Bankeu) consists of assistance from districts and provinces, with a slightly larger share from districts (4.4%) compared to provinces (4.1%). Figure 13: Village Revenue by Source Dana Desa (DD) 52.6% Government transfers to villages Alokasi Dana Desa (ADD) 30.8% 96.7% Financial Assistance (Bankeu) 8.6% Shared Revenue (BH-PDRD) 4.1% Village OSR (PADes) 3.1% Other Revenue 0.7% Finding 9. Share of village revenues from Dana Desa Ranges from 23% in Bali to 81% in Papua. There is a large variation in the share of village revenue from Dana Desa between provinces, ranging from 23% in Bali to 81% in Papua (Figure 14). In 23 provinces, Dana Desa accounts for more than 50% of total village revenues. In Papua, Dana Desa accounts for 81% of village revenues, followed by Aceh (79%) and NTT (74%). By contrast, in Bali, Dana Desa is only the third-largest revenue source at 23%. In 10 provinces, Dana Desa accounts for less than 50% of total village revenues. The revenue trend from Dana Desa is in line with the allocation formula, which provides a fixed base allocation for all villages. It includes affirmative allocations for lagging villages and areas with higher poverty rates (See Table 2 for a detailed breakdown of the Dana Desa allocation formula). However, it is important to note that even in poor and lagging regions, villages still have significant revenues from other sources – highlighting the need for a holistic analysis of village resources. The share of ADD from districts ranges from 59% of total revenues in Kalimantan Timur to 19% in Papua. In two additional provinces, ADD accounts for more than 50% of total revenues. In Kalimantan Utara and Kepulauan Riau, ADD contributes 56% and 54% of revenues, respectively. In six provinces, ADD accounts for less than one-quarter of revenues, with the lowest in Papua, followed by Aceh (20%), Jawa Tengah (21%), Bali (23%), and Banten (24%). 27 Financial Transfers (Bankeu) from provinces and districts make up a large share of revenues in select provinces. In Bali, Bankeu from the provincial government accounts for 19% of total revenues. Provincial transfers also make significant contributions to total revenues in Riau (11%), Jawa Tengah (10%), and Jawa Barat (7%). Six provinces do not provide Benkeu, Kalimantan Utara, Papua Barat, Maluku, Maluku Utara, NTT, and Papua. District Bankeu makes up 14% of total revenues in DI. Yogyakarta , followed by 11% in Jawa Tengah, and 8% in Jawa Timur. In 20 provinces, District Bankeu contributes less than 1% of total revenues. Figure 14: Village Revenue Composition by Province 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Papua Aceh NTT Gorontalo Lampung Sumatera Utara Bengkulu Sumatera Selatan Maluku Kalimantan Tengah Sulawesi Utara Sulawesi Barat Sulawesi Tengah Kalimantan Barat Maluku Utara Sulawesi Tenggara Kalimantan Selatan NTB Banten Papua Barat Sulawesi Selatan Jambi Sumatera Barat Jawa Timur Jawa Barat Jawa Tengah Kalimantan Utara DIY Kep. Riau Kep. Babel Riau Kalimantan Timur Bali DD ADD Bankeu of Province Bankeu of District BH-PDRD PADes Other Village Revenue 28 Finding 10. Village Own-Source Revenue contributes 3% to overall village revenues, with most of this revenue coming from villages in Java and Bali. Village Own-Source Revenue (PADes) contributes only 3% to overall village revenue. The majority of PADes, 2.4% of total revenues, comes from the rent of village land (Figure 15). Even though most villages projected to raise own-source revenue in their original budget, 61% of all villages could not generate PADes (Figure 17). About 15% of villages generated PADes of less than 1% as a share of total revenue. In nominal terms, 27% of villages generate PADes less than IDR 100 million a year. While most villages in Java-Bali (excluding Banten) could generate their own revenue, villages in other regions could not do so. Across all villages, the BUMDes return on investment contributes less than 0.1% to village revenue. Figure 15: Village Own-Source Revenue Breakdown (2019) Returns on Village Asset 2.6% Returns on Village 0.2% Investment Rent from village land asset Community contribution 0.1% Return from other asset utilization Other ROI Other OSR 0.2% BUMDes ROI Figure 16: PADes trend in nominal terms and as a share of total village revenue (2013-2019) 4.2 4.2 20% 4.1 4 16% 3.5 18% 3.5 3.1 15% 2.9 3 Trillion IDR 10% 2 8% 4% 3% 5% 1 3% 2% 0 0% 2013 2014 2015 2016 2017 2018 2019 PADes Nominal (left axis) PADes as Share of Total Village Revenue (right axis) 29 Between 2013 and 2019, PADes has declined both as a share of revenue and in nominal terms (see Figure 16). The decrease in share of revenue is expected with the significant increase of inter-governmental transfers, decreasing from a high of 18% of total revenues to 2% in 2019. Perhaps more surprisingly, PADes is also on a declining trend in nominal terms, going from a high of IDR 4.2 trillion in 2014 and 2015 to IDR 2.9 trillion in 2019. The increasing size of inter-governmental transfers may disincentivize village governments from improving their PADes generation. The government may want to consider options for incentivizing PADes generation for the long-term sustainability of village-level infrastructure projects. Figure 17: Village Own-Source Revenues by Province 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Papua Barat Papua Maluku Utara Sulawesi Tenggara Maluku Kalimantan Utara Kalimantan Barat Sulawesi Barat Sulawesi Tengah Sumatera Utara NTT Bengkulu Sulawesi Utara Kalimantan Timur Aceh Banten Kalimantan Selatan Kalimantan Tengah Lampung Kep. Riau Jambi Gorontalo Sumatera Selatan Sumatera Barat Kep. Babel Sulawesi Selatan Riau NTB Jawa Barat DIY Jawa Tengah Bali Jawa Timur No OSR Less than 1% OSR 1% - <10% OSR 10% or more OSR 30 Finding 11. Spending priorities shift significantly between different fund sources. Spending priorities shift significantly between different fund sources. Currently, the central government and Ministry of Finance exclusively analyze village spending under Dana Desa. However, Dana Desa expenditures alone do not provide a complete picture of village spending and prioritization. With improved data availability for all revenue sources, the government can now conduct a much deeper analysis of spending patterns. Alokasi Dana Desa (ADD) contributes the most toward Salary and Office Operations14, while it is also a major contributor for financing village office facilities and infrastructure, social order/security, and civil registration. The shared district tax and levies (BH-PDRD) and own-source revenue (PADes) contribute the most to the financing of land administration. The financial assistance from districts complements the ADD to finance the majority of village planning, budgeting, and reporting. Financial assistance from provinces contributes to activities in culture and religion. Figure 18: Top Village Expenditure Categories by Source of Fund Public Works and Spatial Planning Housing / Sanitation Health DD Education Agriculture and Livestock Others Apparatus Salary and Office Operations Village Office Facilities and Infrastructure Culture and Religion ADD Community Organization Empowerment Public Works and Spatial Planning Others Public Works and Spatial Planning Village Planning, Budgeting, Reporting District Housing / Sanitation Financial Village Office Facilities and Infrastructure Assistance Apparatus Salary and Office Operations Others Public Works and Spatial Planning Culture and Religion Provincial Housing / Sanitation Financial Village Office Facilities and Infrastructure Assistance Community Organization Empowerment Others 14 Note that villages are not allowed to use Dana Desa to cover salary and office operations costs. 31 7. PART 4: VILLAGE BUDGET CREDIBILITY ANALYSIS Assessment of the village level data is vital to improve public financial management at the lower tier of the governance and improve service delivery. The extension of PEFA to the subnational level can help understand PFM capacity at the village level and assess budget credibility. However, the PEFA assessment (2016 framework) does not cover the assessment of the subnational or village-level public financial management operations. The assessment is undertaken at the central level to assess the public finance management operations of the government using Government Finance Statistics (GFS) definitions, which covers ministries and extra-budgetary units at the central level. The PEFA framework has been adapted to assess the performance at the village level based on the Budget Reliability pillar. The budget reliability measures the accuracy of the budget by comparing the composition of expenditures and revenues to the year-end outrun. The scores measure the strength of the budget forecast, the establishment of a consistent budgeting framework, and revenue mobilization. Basing the budget on the previous year's actual expenditure enhances the consistency and provides a realistic baseline for next year's budget formulation. The analysis for budget reliability uses five standard PEFA indicators and applies two additional indicators that are non-standard PEFA indicators. 15 1.1 Aggregate expenditure outturn (and by Fund) 2.1 Expenditure composition outturn by function 2.2 Expenditure composition outturn by economic type 2.3 Expenditure composition outturn by Fund - Non-Standard PEFA 3.1 Aggregate revenue outturn (and by Fund) 3.2 Revenue composition outturn by economic type 3.3 Revenue composition outturn by Fund – Non-Standard PEFA The analysis has also adjusted the time horizon of the analysis to one year, as at this time, village data is only available for FY 2019. PEFA for most indicators uses three years of data to score an indicator and two years where necessary to eliminate a one-off effect of an external shock. As data becomes available for additional fiscal years, MOHA will update the analysis. The analysis presented in this section cover 33,220 villages across all 33 provinces with rural villages for which original and actual budgets were available, including data on revenues and expenditures. While the assessment covers 45% of all villages, it is not a drawn sample from all villages and therefore may not represent budget credibility of all villages across the country. The process serves as a proof-of-concept to demonstrate the possibility of leveraging the newly available data to develop performance-based metrics on village financial management. It is important to note that these grades are not representative of the national average but only of the assessed villages. It is likely that in provinces with lower levels of reporting (see Part 1) – such as Papua (6%) and Sulawesi Tenggara (22%) – the average grades may shift. On the other hand, 98% of villages from Bengkulu were included in the modified PEFA assessment, and 92% of those villages received A and B grades. 15 Please refer to PEFA Guidelines for details on the dimensions, scoring scale, and calculation of averages. 32 Finding 12. 75% of villages receive an A or B grade in a modified PEFA assessment of budget reliability. Across the country, 75% of villages scored A or B grades on the overall PEFA average score, indicating high budget reliability and financial management capacity. Less than one-fifth of villages scored a C, and only 6% received a D score. As Figure 19 shows, more than half of villages received an A when analyzing Expenditure and Revenue grades separately. The same pattern holds for villages that received a D, with a higher percentage of villages receiving a D in the Expenditure Grade and Revenue Grade. This shows that while some villages perform poorly on analyzing their revenues or expenditures, they usually make up for the poor performance in the other category. This may suggest that, in general, budget reliability is high across villages, and external factors – such as district regulations or predictability of funds may contribute to lower performance in some areas. However, additional analysis is required to test this hypothesis. Figure 19: Modified PEFA Grades in Key Budget Reliability Indicators for 25,430 Villages 6% 19% 15% 34% 24% Average PEFA Expenditure 14% 52% Revenue Grade Grade 10% Grade 61% 10% 14% 41% A B C D Figure 20: Average Village PEFA Scores 33 Finding 13. There is a large variation in budget reliability between regions, with Sulawesi showing the best performance and Java the poorest. Regional analysis of grades shows a large variation between regions (Figure 21). Over 70% of villages in Sulawesi received A and B grades, compared to less than 40% in Java. Papua and Maluku had the highest share of villages scoring D – at over 30%. However, looking at provincial analysis (Figure 22) shows that most lower grades are from villages in Papua Barat and Maluku. Villages in Sulawesi and Sumatra were the highest performing, followed by Bali & Nusa Tenggara and Kalimantan, followed by Papua & Maluku and Java. Figure 21: PEFA Grades by Region Across 25,430 Villages - Count and Rate A Java B+ Sumatra B Kalimantan C+ Sulawesi C Bali & Nusa Tenggara D+ Papua & Maluku 0 2000 4000 6000 8000 10000 12000 14000 D Java Sumatra Kalimantan Sulawesi Bali & Nusa Tenggara Papua & Maluku 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 34 Figure 22: Average Village PEFA Grade by Province for 25,430 Villages Bengkulu Southeast Sulawesi Lampung West Sulawesi Papua Jambi North Sumatra Riau South Sumatra North Kalimantan North Sulawesi West Sumatra Central Sulawesi South Sulawesi Bali West Nusa Tenggara Central Kalimantan West Java Gorontalo Riau islands South Borneo Banten East Kalimantan West Kalimantan East Nusa Tenggara Bangka Belitung Islands North Maluku West Papua East Java Central Java Aceh Maluku D.I. Yogyakarta A B+ B C+ C D+ D 35 The five provinces with the highest share of villages with A and B grades are located in Sulawesi, Sumatra, and Papua. Nearly 92% of villages in Bengkulu, Sumatra with received A or B grades; Sulawesi Tenggara had 90%; Lampung, 89%; Sulawesi Barat, 84%; and Papua, 81%. In contrast, DI Yogyakarta had the lowest share of villages with A and B grades at just over 12%, followed by Maluku with 24%; Aceh, 25%; Jawa Tengah, 30%; and Jawa Timur, 33%. Finding 14. Predictable resourcing leads to better budgets. Analysis of the distribution of average village expenditure and revenue scores shows that villages can develop credible budgets when village revenues are predictable. Figure 23 shows the correlation between village aggregate revenue scores (y-axis) and aggregate expenditure scores (x-axis), with colors representing average PEFA budget reliability scores (where A is dark green, and D is bright red). The x- and y-axes represent the percentage of budget to actuals, with 100% significating a full match at the middle of the graph. The revealed correlation between the budget revenue and expenditure execution performance aligns with Public Finance Theory and highlights the importance of predictable resourcing. This means that when villages receive what they think they will get, and put that in their budget, they will spend their budget as planned – delivering a reliable budget. Figure 23: Distribution of Aggregate Revenue and Expenditure Scores as Percentage of Budget to Actuals 36 8. FUTURE DIRECTIONS Through Siskeudes, the Government of Indonesia has unprecedented access to comprehensive village financial data that can help measure the impact of village transfers and inform future policy direction. This data can help the government measure results, ensure value for money, monitor village and district performance, flag corruption risk, and make better policy decisions. In the process of conducting this expenditure analysis, the World Bank has worked with MOHA to develop a proof-of-concept application that has automated many of the key processes for multi-year trend analyses. MOHA has already begun work to take critical next steps to strengthen both the application of Siskeudes and the processes that can improve village reporting. These next steps include increasing uptake of villages and districts to report through the Siskeudes National Consolidation Application; sharing information and analysis back down to local governments so they can compare their performance with others; leverage Siskeudes to support District Inspectorate with village oversight and audit; and use findings from village performance to better target capacity building and technical assistance to villages; and link Siskeudes with other data systems, including SIPADES (Village Asset Management System). Complete village financial data availability provides the government with a unique opportunity for evidence-based policymaking, targeted support to villages, and performance-based incentives to villages and districts. As the quality of input data improves, the government can undertake economic efficiency analysis of village investments. By pairing this data with district budgets, the government can analyze frontline service delivery and identify critical gaps. In the area of stunting, the government can link data from BPS with village and district spending to better understand the impact of investments. This data can also improve financial management and oversight of supra-village governments, such as implementing risk-based audits. Key Lessons from 2019 ViPER 1. Improving the availability of complete consolidated village financial data will require better incentives to districts and clear sanctions for non-compliance with reporting requirements. The government can expand disbursement requirements for Dana Desa to require districts to submit complete village financial reports (APBDes), not only for a single revenue source. There is also an opportunity to include reporting requirements as a component of the performance-based Dana Desa allocation, holding back a certain portion of allocations for non-compliance to reporting requirements. A key consideration for this should be that the reporting bottleneck is often at the district level, not at the village level. Therefore, incentives and sanctions should also target district governments, such as conditionality for part of DAU. 2. Strengthening the financial performance of villages requires better support and monitoring of both villages and districts. The modified PEFA assessment and "cut and paste" test show that village performance depends on district performance. Findings from these analyses can be used to directly work with districts where regulations and guidelines lead to poor financial management. This includes requirements for expenditure reports to match original budgets – even where this means numbers do not report actual spending. It also includes delays in informing villages about annual transfer 37 ceilings and delays in disbursements. Analysis of village financial data can help the government identify regions with suboptimal performance and provide needs-based training and refresher courses to the relevant village or district government officials. 3. Building a local-level interface to share information back to districts and villages can improve data quality and support local planning. Village financial data presented here can play an important role in evidence-based policymaking for the central government. However, this data can also be of equal importance for district and village governments to improve their performance. Currently, districts and villages do not have the capacity to analyze the financial data they submit to the central government. As with most other data in Indonesia, financial data collection is an extractive process where local governments do not see the results and findings of their reports. Building an interface that allows local governments to see this data in a meaningful format will both help with local planning, particularly in targeting capacity building support, and can also help to improve the quality of data as villages and districts will be able to see where incorrect inputs have impacted their performance on key indicators. 4. Leveraging financial data for evidence-based policymaking requires a holistic analysis of all village resources, not only Dana Desa. Currently, the MoF is only analyzing reports from Dana Desa. As articulated in this report, this hides some important trends. A more holistic analysis of resources can help ensure better complementarities between different revenue sources; ensure fiscal fairness in the distribution of resources as a whole; and improve oversight and monitoring for the nearly US $4 billion annually spent by villages from revenues outside of Dana Desa. As the availability of consolidated data increases across the country, the Ministry of Finance will also be able to use wholistic financial data from the National Consolidation Application to update and refine the current performance-based incentive methodology for the Dana Desa allocation formula, to make transfers contingent on performance on all village revenues. 5. Spending patterns can inform sectoral technical assistance and allow for more strategic targeting of village-level support by line ministries and directorates. The variation in village spending patterns between provinces shows how communities prioritize activities based on the local context. This data can help inform sectors about opportunities for better collaboration between village governments and local service delivery systems. For example, data from 2019 shows that there is potential for the Ministry of Fisheries to work closely with village governments in Papua Barat and Maluku, where villages are spending a significant portion of their budget on activities in that sector. Similarly, the Ministry of Agriculture has an opportunity to work closely with villages in Papua, which spend 36% of their total village revenue on agriculture activities. To improve access to water and sanitation, the Ministries of Health and Public Works have the opportunity to coordinate more closely with villages in Papua Barat and Gorontalo, where villages are spending over 20% of their funds toward WASH activities. Better targeting of technical support and village-district coordination can help ensure complementarity of spending between different levels of government and help progress toward sectoral targets. 6. Institutionalizing village expenditure analysis, such as the ViPER methodology in this report, will support policymaking for MOHA, the Ministry of Finance, and other government institutions that focus on poverty reduction and frontline service delivery. As a part of this PER, an automated interface has been developed that can help government routinely monitor the use of resources under the Village Law. Regularly providing such analysis can help institutions across government better understand the capacity and opportunities available through village-level delivery mechanisms. 38 ANNEX A: VILLAGE REPORTING Table 6: Villages Using Siskeudes and Report Availability in Siskeudes National Consolidation Application # of Villages % of Villages # of Villages % of Villages Province Total Villages Reporting in Reporting in Using Siskeudes Using Siskeudes Siskeudes Siskeudes Aceh 6496 6461 99% 2741 42% Bali 636 636 100% 635 100% Banten 1238 1238 100% 666 54% Bengkulu 1341 1341 100% 1316 98% DIY 392 392 100% 315 80% Gorontalo 657 657 100% 556 85% Jambi 1399 1399 100% 1265 90% Jawa Barat 5312 5312 100% 3920 74% Jawa Tengah 7809 7573 97% 5740 74% Jawa Timur 7724 7535 98% 5453 71% Kalimantan Barat 2031 2031 100% 1812 89% Kalimantan 1864 1862 100% 1737 93% Selatan Kalimantan 1433 1432 100% 917 64% Tengah Kalimantan Timur 841 841 100% 698 83% Kalimantan Utara 447 406 91% 109 24% Kep Babel 309 309 100% 246 80% Kep. Riau 275 275 100% 274 100% Lampung 2435 2435 100% 1232 51% Maluku 1198 1198 100% 795 66% Maluku Utara 1063 1063 100% 603 57% NTB 1005 995 99% 652 65% NTT 3026 3026 100% 871 29% Papua 5411 2679 50% 346 6% Papua Barat 1742 1317 76% 84 5% Riau 1591 1591 100% 1422 89% Sulawesi Barat 575 575 100% 457 80% Sulawesi Selatan 2255 2255 100% 1306 58% Sulawesi Tengah 1842 1842 100% 1479 80% Sulawesi Tenggara 1908 1876 98% 416 22% Sulawesi Utara 1507 1507 100% 980 65% Sumatera Barat 928 806 87% 757 82% Sumatera Selatan 2853 2852 100% 1392 49% Sumatera Utara 5417 5410 100% 2492 46% 39 ANNEX B: VILLAGE EXPENDITURES Table 7: Village Spending by Functional Classification (GFS) GFS Classification Actuals Final (%) Original Budget (%) Housing and community amenities 37.78% 37.26% Economic affairs 35.66% 34.96% Recreation, culture, and religion 10.03% 10.29% General public services 4.76% 5.41% Health 4.13% 4.12% Education 3.96% 3.97% Environmental protection 2.37% 2.43% Public order and safety 0.83% 1.02% Social protection 0.48% 0.53% Table 8: Village Spending by Economic Classification Type Average Capital 52.33% Goods and Services 28.70% Personnel 18.89% Contingency 0.08% Table 9: Village Spending by Economic Classification by Region Region Capital Goods and Services Personnel Contingency Java 55.05% 24.57% 20.31% 0.08% Sumatra 54.32% 28.23% 17.34% 0.10% Kalimantan 53.11% 29.06% 17.77% 0.06% Sulawesi 49.20% 31.46% 19.27% 0.08% Papua 43.53% 40.80% 15.50% 0.17% Bali & Nusa Tenggara 34.94% 46.43% 18.59% 0.04% 40 Table 10: National estimates of Village Expenditure by Sub-Bidang & Infrastructure Content, 2019 Total Spending Infrastructure Content Share Sub-Bidang coefficient billion billion trillion IDR % trillion IDR USD* USD* Public Works and Spatial Planning 36.0% 42.30 2.92 90.6% 38.31 2.64 Village Apparatus salary and office 25.9% 30.46 2.10 operations Housing/Sanitation 7.1% 8.29 0.57 90.2% 7.48 0.52 Health 4.3% 5.03 0.35 28.5% 1.43 0.10 Education 4.0% 4.70 0.32 47.0% 2.21 0.15 Culture and Religion 3.3% 3.88 0.27 22.2% 0.86 0.06 Village Office Facilities and 3.1% 3.59 0.25 42.6% 1.53 0.11 Infrastructure Village Planning, Budgeting, and 2.5% 2.92 0.20 1.4% 0.04 0.00 Reporting Agriculture and Livestock 2.1% 2.51 0.17 10.8% 0.27 0.02 Community Organization 2.0% 2.40 0.17 Empowerment Youth and Sports 2.0% 2.39 0.16 70.7% 1.69 0.12 Capacity Building for Village 1.3% 1.47 0.10 Apparatus Social Order and Community 0.9% 1.01 0.07 Protection Trade and Industry 0.8% 0.94 0.06 69.9% 0.65 0.05 Tourism 0.6% 0.66 0.05 70.5% 0.47 0.03 Women's Empowerment & Child 0.6% 0.65 0.04 Protection Energy and Mineral Resources 0.5% 0.63 0.04 74.3% 0.47 0.03 Marine and Fisheries 0.5% 0.63 0.04 12.4% 0.08 0.01 Transportation and ICT 0.5% 0.62 0.04 Investment 0.4% 0.52 0.04 Land Administration 0.4% 0.50 0.03 Coops and MSME 0.4% 0.49 0.03 Civil Registration, Statistics, and 0.3% 0.40 0.03 Archives Forestry and the Environment 0.3% 0.32 0.02 Disaster Response 0.1% 0.07 0.00 Contingency 0.0% 0.03 0.00 Emergency 0.0% 0.02 0.00 Others 0.0% 0.02 0.00 41 ANNEX C: VILLAGE REVENUES Table 11: Village Revenues by Fund Source by Province Bankeu of Bankeu of DD ADD BH-PDRD PADes Other Province District Aceh 78.8% 19.7% 0.1% 0.0% 0.7% 0.2% 0.4% Bali 22.6% 23.4% 19.1% 6.1% 27.1% 0.9% 0.7% Banten 57.1% 23.5% 4.1% 6.7% 6.7% 0.2% 1.6% Bengkulu 67.6% 31.5% 0.0% 0.1% 0.3% 0.1% 0.4% DIY 40.7% 28.3% 0.9% 13.8% 7.8% 7.2% 1.3% Gorontalo 70.2% 27.8% 0.0% 0.5% 0.8% 0.2% 0.5% Jambi 56.2% 34.3% 4.8% 2.6% 1.1% 0.5% 0.5% Jawa Barat 46.1% 28.8% 6.8% 6.1% 10.1% 1.3% 0.8% Jawa Tengah 45.6% 20.6% 9.9% 11.2% 2.2% 9.5% 0.9% Jawa Timur 47.3% 30.5% 0.9% 7.9% 2.9% 9.5% 1.0% Kalimantan Barat 62.2% 34.0% 0.1% 0.7% 2.2% 0.5% 0.3% Kalimantan Selatan 58.6% 38.7% 0.0% 0.0% 1.6% 0.3% 0.7% Kalimantan Tengah 64.1% 34.1% 0.1% 0.1% 1.1% 0.2% 0.4% Kalimantan Timur 37.4% 59.1% 0.0% 1.9% 1.0% 0.2% 0.4% Kalimantan Utara 40.9% 56.1% 0.0% 0.0% 1.7% 0.3% 1.2% Kep. Babel 39.0% 47.8% 2.7% 1.2% 5.4% 1.2% 2.7% Kep. Riau 39.1% 53.8% 0.0% 0.3% 6.3% 0.1% 0.4% Lampung 69.1% 26.9% 0.5% 1.4% 1.4% 0.4% 0.4% Maluku 64.2% 35.1% 0.0% 0.0% 0.3% 0.1% 0.3% Maluku Utara 61.0% 33.9% 0.0% 0.0% 0.2% 0.1% 4.8% NTB 58.4% 35.9% 0.3% 1.8% 2.6% 0.4% 0.6% NTT 73.8% 24.0% 0.0% 0.0% 1.0% 0.8% 0.4% Papua 80.9% 19.1% 0.0% 0.0% 0.0% 0.0% 0.0% Papua Barat 56.6% 42.7% 0.0% 0.0% 0.7% 0.0% 0.0% Riau 38.9% 43.3% 10.9% 3.3% 2.6% 0.5% 0.5% Sulawesi Barat 62.6% 34.9% 1.7% 0.0% 0.4% 0.0% 0.3% Sulawesi Selatan 56.4% 40.1% 0.0% 0.5% 1.8% 0.6% 0.6% Sulawesi Tengah 62.6% 33.4% 0.0% 0.1% 2.4% 0.1% 1.4% Sulawesi Tenggara 60.0% 37.1% 0.0% 2.3% 0.3% 0.1% 0.3% Sulawesi Utara 63.7% 34.9% 0.1% 0.2% 0.6% 0.2% 0.4% Sumatera Barat 52.1% 44.5% 0.0% 0.3% 2.0% 0.4% 0.6% Sumatera Selatan 66.7% 30.4% 2.0% 0.0% 0.4% 0.2% 0.3% Sumatera Utara 68.2% 30.4% 0.0% 0.2% 0.8% 0.1% 0.3% 42 Table 12: Top Village Expenditure Categories by Source of Fund Dana Desa 1 Public Works and Spatial Planning 57.51% 2 Housing / Sanitation 10.92% 3 Health 6.75% 4 Education 6.37% 5 Agriculture and Livestock 3.58% 6 Others 14.87% ADD 1 Apparatus Salary and Office Operations 71.30% 2 Village Office Facilities and Infrastructure 5.60% 3 Culture and Religion 3.90% 4 Community Organization Empowerment 3.60% 5 Public Works and Spatial Planning 3.50% 6 Others 12.20% District Financial Assistance 1 Public Works and Spatial Planning 48.62% 2 Village Planning, Budgeting, Reporting 15.72% 3 Housing / Sanitation 10.56% 4 Village Office Facilities and Infrastructure 4.62% 5 Apparatus Salary and Office Operations 3.70% 6 Others 16.77% Provincial Finance Assistance 1 Public Works and Spatial Planning 46.47% 2 Culture and Religion 17.05% 3 Housing / Sanitation 9.98% 4 Village Office Facilities and Infrastructure 6.35% 5 Community Organization Empowerment 5.40% 6 Others 14.76% 43 Table 13: Financing contribution of multiple funds across sub-bidang Sources of Fund Share Total Sub-Bidang Spending Bankeu Bankeu DD ADD BH-PDRD PADes Others Total Districts Provinces Public Works and Spatial 36.0% 85.5% 3.1% 0.8% 5.5% 4.1% 0.4% 0.7% 100% Planning Village Apparatus salary 25.9% 0.6% 88.0% 3.6% 0.6% 0.6% 6.1% 0.5% 100% and operations Housing/Sanitation 7.1% 82.9% 3.7% 1.3% 6.1% 4.5% 0.4% 1.1% 100% Health 4.3% 84.4% 7.9% 3.2% 1.2% 2.0% 0.5% 0.8% 100% Education 4.0% 85.2% 9.2% 1.8% 1.8% 0.5% 0.8% 0.8% 100% Culture and Religion 3.3% 19.3% 37.7% 14.4% 4.5% 16.5% 5.3% 2.3% 100% Village Office Facilities 3.1% 6.4% 58.6% 15.5% 6.2% 6.7% 3.0% 3.6% 100% and Infrastructure Village Planning, 2.5% 10.3% 43.4% 9.0% 25.7% 0.7% 7.7% 3.2% 100% Budgeting, and Reporting Agriculture and 2.1% 89.8% 2.5% 0.6% 3.4% 1.5% 1.0% 1.3% 100% Livestock Community 2.0% 16.6% 56.3% 11.4% 4.0% 8.5% 2.3% 1.0% 100% Organization Empowerment Youth and Sports 2.0% 70.1% 17.9% 6.3% 2.6% 0.9% 1.1% 1.1% 100% Capacity Building for 1.3% 52.1% 35.5% 8.8% 0.4% 0.3% 1.3% 1.6% 100% Village Apparatus Social Order and 0.9% 21.2% 54.6% 14.3% 2.5% 2.8% 3.0% 1.6% 100% Community Protection Trade and Industry 0.8% 87.6% 2.3% 1.0% 3.4% 2.2% 1.9% 1.6% 100% Tourism 0.6% 85.0% 3.0% 1.2% 5.0% 0.8% 1.7% 3.2% 100% Women's 0.6% 82.2% 7.4% 2.9% 2.5% 2.8% 1.1% 1.1% 100% Empowerment & Child Protection Energy and Mineral 0.5% 88.5% 5.8% 0.7% 2.3% 1.4% 0.1% 1.0% 100% Resources Marine and Fisheries 0.5% 92.4% 4.2% 0.5% 0.7% 0.5% 0.2% 1.4% 100% Transportation and ICT 0.5% 82.4% 9.0% 2.6% 3.0% 1.0% 0.8% 1.2% 100% Investment 0.4% 82.5% 3.1% 1.0% 1.5% 8.3% 0.6% 3.0% 100% Land Administration 0.4% 4.0% 14.5% 33.2% 3.2% 0.4% 30.4% 14.3% 100% Coops and MSME 0.4% 76.4% 5.6% 1.4% 8.7% 5.8% 0.4% 1.7% 100% Civil Registration, 0.3% 25.5% 55.5% 13.4% 0.9% 1.3% 2.1% 1.2% 100% Statistics, and Archives Forestry and the 0.3% 74.8% 8.4% 9.9% 3.8% 0.3% 1.2% 1.6% 100% Environment Disaster Response 0.1% 52.9% 18.8% 10.0% 6.0% 1.9% 6.2% 4.1% 100% Contingency 0.0% 65.6% 12.6% 5.7% 0.3% 0.5% 12.5% 2.9% 100% Emergency 0.0% 44.6% 23.6% 9.5% 0.7% 0.8% 15.4% 5.4% 100% Others 0.0% 86.0% 14.0% 0.0% 0.0% 0.0% 0.0% 0.0% 100% 44 ANNEX D: VILLAGE BUDGET CREDIBILITY Table 14: Average PEFA Scores by Province Province A B+ B C+ C D+ D Total Aceh 129 41 366 327 468 794 10 2135 Bali 160 55 124 56 35 50 11 491 Bangka Belitung Islands 21 9 70 63 27 14 3 207 Banten 116 24 45 61 65 28 6 345 Bengkulu 496 39 140 22 15 22 2 736 Central Java 368 216 816 1027 966 880 397 4670 Central Kalimantan 227 40 245 136 100 105 5 858 Central Sulawesi 342 42 321 104 110 74 2 995 D.I. Yogyakarta 7 5 23 36 32 112 66 281 East Java 196 95 1113 962 682 707 490 4245 East Kalimantan 70 18 184 85 71 93 5 526 East Nusa Tenggara 29 7 227 73 68 117 9 530 Gorontalo 10 1 218 105 42 31 407 Jambi 342 43 496 140 40 33 2 1096 Lampung 129 6 439 40 13 18 2 647 Maluku 35 5 92 58 121 246 2 559 North Kalimantan 1 1 51 15 3 3 74 North Maluku 95 20 143 54 79 134 10 535 North Sulawesi 288 44 225 79 50 99 785 North Sumatra 804 98 244 140 54 90 1 1431 Papua 126 1 20 6 3 26 182 Riau 80 288 461 161 78 39 3 1110 Riau islands 19 5 103 64 28 12 231 South Borneo 268 88 498 363 172 174 11 1574 South Sulawesi 222 46 546 119 64 69 87 1153 South Sumatra 453 35 114 141 36 42 2 823 Southeast Sulawesi 216 13 88 18 5 13 353 West Java 971 142 655 430 285 358 148 2989 West Kalimantan 382 67 386 134 268 383 13 1633 West Nusa Tenggara 169 17 108 53 31 63 8 449 West Papua 10 3 18 45 2 6 84 West Sulawesi 264 11 108 27 27 17 454 West Sumatra 223 90 135 126 44 10 4 632 Grand Total 7268 1615 8822 5270 4084 4862 1299 33220 45 Table 15: PEFA Scoring Tables Dimension 1.1. Aggregate Expenditure Outturn Score Scoring Scale Aggregate expenditure outturn was between 95% and 105% of the approved aggregate budgeted A expenditure in at least two of the last three years. Aggregate expenditure outturn was between 90% and 110% of the approved aggregate budgeted B expenditure in at least two of the last three years. Aggregate expenditure outturn was between 85% and 115% of the approved aggregate budgeted C expenditure in at least two of the last three years. D The performance is less than required for a C score Dimension 2.1. Expenditure composition outturn by function16 Score Scoring Scale Variance in expenditure composition by program, administrative or functional classification was less than A 5 percent in at least two of the last three years. Variance in expenditure composition by program, administrative or functional classification was less than B 10 percent in at least two of the last three years. Variance in expenditure composition by program, administrative or functional classification was less than C 15 percent in at least two of the last three years. D The performance is less than required for a C score Dimension 2.2. Expenditure composition outturn by economic type 17 Score Scoring Scale Variance in expenditure composition by economic classification was less than 5% at least two of the last A three years. Variance in expenditure composition by economic classification was less than 10% at least two of the last B three years. Variance in expenditure composition by economic classification was less than 15% at least two of the last C three years. D The performance is less than required for a C score Dimension 3.1. Aggregate revenue outturn Score Scoring Scale Actual revenue was between 97 percent and 106 percent of budgeted revenue in at least two of the last A three years. Actual revenue was between 94 percent and 112 percent of budgeted revenue in at least two of the last B three years. Actual revenue was between 92 percent and 116 percent of budgeted revenue in at least two of the last C three years. D The performance is less than required for a C score Dimension 3.2. Revenue composition outturn18 Score Scoring Scale A Variance in revenue composition was less than 5 percent in two of the last three years. B Variance in revenue composition was less than 10 percent in two of the last three years. C Variance in revenue composition was less than 15 percent in two of the last three years. D The performance is less than required for a C score 16 2.1 composition is done at the high (10 functions) level of COFOG. 17 2.2 is at GFS 2-digit level (economic classification). The same formula is used for the composition according to Fund DD, ADD, etc) 18 3.2 composition is at GFS 2-digit level (economic classification). The same formula is used for the composition according to Fund 46 ANNEX E: DANA DESA FINANCIAL FLOW MECHANISM (FISCAL YEAR 2020 AS PER PMK 205/2019) 47 2021 © 2021 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this event do not necessarily reflect the views of the World Bank, the Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work.