2023 CAMBODIA PUBLIC FINANCE REVIEW FROM SPENDING MORE TO SPENDING BETTER 2023 Report No: AUS0003416 © 2023 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 work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “World Bank. 2023. Cambodia Public Finance Review: From Spending More to Spending Better. © World Bank.” All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. CONTENTS Glossary of acronyms..............................................................................................................................viii Acknowledgment....................................................................................................................................... xi Executive summary.................................................................................................................................... 1 Chapter 1. Achieving fiscal resilience and maximizing fiscal space................................................... 18 1.1. Introduction................................................................................................................................ 18 1.2. Recent macroeconomic trends................................................................................................... 19 1.3. Revenue..................................................................................................................................... 20 1.4. Expenditure ............................................................................................................................... 22 1.5. Fiscal balance............................................................................................................................ 24 1.6. Public debt................................................................................................................................. 26 1.7. Maintaining fiscal space and promoting progressivity.................................................................. 27 Chapter 2. Enhancing quality of public spending ................................................................................. 31 2.1. Public expenditure trends in aggregate ...................................................................................... 31 2.2. Strategic public expenditure prioritization.................................................................................... 33 2.3. Distributional impacts of fiscal policies........................................................................................ 64 2.4. Strengthening public sector management for improved quality of public expenditure.................. 65 2.5. Cambodia performance on anti-corruption................................................................................. 72 2.6. Conclusion and recommendations............................................................................................. 73 Chapter 3. Maximizing effectiveness of public health policy............................................................... 79 3.1. Introduction................................................................................................................................ 79 3.2. Sectoral context......................................................................................................................... 79 3.3. Health sector financing............................................................................................................... 94 3.4. Summary of key findings.......................................................................................................... 106 3.5. Recommendations................................................................................................................... 107 Chapter 4. Strengthening performance orientation of public spending to boost productivity in agriculture ........................................................................................................................... 111 4.1. Introduction.............................................................................................................................. 111 4.2. Sectoral context....................................................................................................................... 111 4.3. Agriculture public expenditure.................................................................................................. 113 4.4. Public spending in agriculture by MAFF.................................................................................... 115 4.5. Public financial management ................................................................................................... 119 4.6. Recommendations................................................................................................................... 126 Chapter 5. Improving capital budget management for improved irrigation and water resources management........................................................................................................................ 130 5.1. Introduction.............................................................................................................................. 130 5.2. Sectoral context....................................................................................................................... 131 5.3. Public expenditure in irrigation and water resource .................................................................. 134 5.4. Budget management ............................................................................................................... 142 5.5. Summary of key findings.......................................................................................................... 146 5.6. Recommendations................................................................................................................... 147 Chapter 6. Key lessons on subnational PFM based on case study of Preah Sihanouk province..... 150 6.1. Introduction.............................................................................................................................. 150 6.2. The provincial administration ................................................................................................... 152 6.3. The district and municipality administrations............................................................................. 166 6.4. The commune/sangkat administrations.................................................................................... 170 6.5. Recommendations................................................................................................................... 171 References.............................................................................................................................................. 175 Annex....................................................................................................................................................... 177 Annex to Chapter 1......................................................................................................................... 177 Annex to Chapter 3......................................................................................................................... 184 Annex to Chapter 4......................................................................................................................... 194 Annex to Chapter 5......................................................................................................................... 196 Cambodia Public Finance Review - From Spending More to Spending Better i LIST OF TABLES Table ES.1. Summary of recommendations on PIM from the 2019 PER.................................................... 8 Table ES.2. Summary of recommendations for improving education outcomes from the 2019 PER ....... 11 Table ES.3. Summary of key selected policy options .............................................................................. 13 Table 1.1. Government fiscal intervention (% of GDP)............................................................................ 23 Table B.1.2.1. Share of relief granted through QIP or SEZ........................................................................... 28 Table 2.1. Schools in rural areas have the highest level of teacher shortage in 2021.............................. 40 Table 2.2. Irrigation outputs reported from MOWRAM spending............................................................ 55 Table 2.3. Quality of infrastructure (2018).............................................................................................. 60 Table 2.4. Summary of recommendations for improving education outcome from the 2019 PER.......... 75 Table 2.5. Summary of recommendations on PIM from 2019 PER ....................................................... 76 Table 2.6. Summary of key selected policy options .............................................................................. 78 Table 3.1. Enrollment in different social health protection schemes........................................................ 84 Table 3.2. Access to social health protection........................................................................................ 85 Table 3.3. Comparison of benefit packages and payment rate of HEF and NSSF ................................. 87 Table 3.4. Options to improve health outcomes and increase health equitable access ....................... 110 Table 4.1. Functional composition of MAFF recurrent spending........................................................... 118 Table 4.2. Budget strategic plan, annual budget and actual spending of MAFF (KHR, million)............. 122 Table 4.3. BSP performance reporting structure and examples........................................................... 123 Table 4.4. BSP performance reporting for program-level objectives.................................................... 124 Table 4.5. Options to strengthen budget management in agriculture................................................... 128 Table 5.1. Key indicators and targets for 2019-2023 of MOWRAM..................................................... 133 Table 5.2. Budget by key subprograms............................................................................................... 136 Table 5.3. Budget allocation for irrigation rehabilitation (KHR, million).................................................. 139 Table 5.4. Budget allocation for irrigation O&M, by level of administration (KHR, million)...................... 140 Table 5.5. Budget allocation for establishment and support to FWUC (KHR, million)........................... 140 Table 5.6. Activities and budget under establishment of FWUC (SP1.3) and managing FWUC (SP1.4) (KHR, million)............................................................................................... 140 Table 5.7. Budget allocation for emergency responses (KHR, million).................................................. 141 Table 5.8. Reported outputs relating to rice versus other crops........................................................... 141 Table 5.9. Budget execution rate for the central MOWRAM................................................................ 143 Table 5.10. Budget execution rate for PDRAM...................................................................................... 143 Table 5.11. Number of investment projects by types of procurement methods used............................. 145 Table 5.12. Outputs reported from investment expenditure (Chapter 21)............................................... 146 Table 5.13. Recommendations for irrigation sector............................................................................... 149 Table 6.1. State asset information compiled by PDEF (2021)............................................................... 160 Table 6.2. Percentage and timing of budget execution (2018-2020).................................................... 161 Table 6.3. Procurement thresholds for SNA........................................................................................ 164 Table 6.4. Number of days between payment orders and payments (non-procurement)..................... 164 Table 6.5. Recommendations by timeframe........................................................................................ 173 Table A.1.1. Fiscal balance..................................................................................................................... 177 Table A.1.2. Selected indicators............................................................................................................. 178 Table A.1.3. Loan disbursements........................................................................................................... 179 Table A.1.4. Fiscal projections................................................................................................................ 179 Table A.1.5. VAT foregone as a share of value added tax collected......................................................... 181 Table A.1.6. Customs and special taxes relief as a share of potential collections.................................... 181 Table A.1.7. Share of relief granted through QIP or SEZ......................................................................... 183 Table A.3.1. Distribution and characteristics of public health care providers ........................................... 188 Table A.3.2. List of private health facilities and pharmaceutical and medical supply institutions in Cambodia (September 2023).......................................................................................... 190 Table A.3.3. Policy and key health care reform priorities, implementation progress, and challenges ahead...... 191 Table A.3.4. Distribution of public health facilities by province and OPD cases (2022)............................. 193 Table A.3.5. MOH budget outturn.......................................................................................................... 193 Table A.4.1. MAFF program structure across administrative units........................................................... 195 Table A.5.1. Numbers of irrigation schemes and irrigated areas.............................................................. 196 ii Cambodia Public Finance Review - From Spending More to Spending Better LIST OF FIGURES Figure ES.1. Government expenditure, by sector grouping (% of GDP) (2011-2021)................................... 2 Figure ES.2. Cambodia’s public sector wage premium over time................................................................ 2 Figure ES.3. Government expenditure on education (% of GDP vs GNI per capita, 2021)........................... 3 Figure ES.4. Learning outcomes for Grade 6 students deteriorated between 2013 and 2021..................... 3 Figure ES.5. Type of health provider consulted first..................................................................................... 4 Figure 1.1. Cambodia has substantially improved domestic revenue mobilization (% of GDP)................. 20 Figure 1.2. Expenditure increased significantly during the COVID-19 pandemic (% of GDP)................... 22 Figure 1.3. Cambodia has pursued a conservative fiscal policy resulting in low debt and high reserves........ 24 Figure 1.4. Cambodia’s fiscal discipline has enabled countercyclical spending (% of GDP)..................... 25 Figure B.1.2.1. Tax relief on imports as a share of tax imposed at customs................................................ 28 Figure B.1.2.2. Share of taxes relieved on imports for QIPs and SEZs........................................................ 28 Figure 2.1. Government overall spending increased between 2011 and 2021, driven by rise in wage bill and non-wage recurrent planned expenditure ...................................................................... 31 Figure 2.2. Government expenditure, by sector grouping (% of GDP) (2011-2021)................................. 31 Figure 2.3. Compensation of government employee (% of total expenses) (2011-2019)......................... 32 Figure 2.4. Trend of capital spending, by sector (% of GDP) (2011-2021) .............................................. 32 Figure 2.5. Net investment in government nonfinancial assets (% of GDP) (2011-2019).......................... 33 Figure 2.6. Deviation of actual spending against NSDP planning (2014-2021)........................................ 33 Figure 2.7. Expenditure trend for social sectors, by line ministry (% of GDP) (2011-2021)....................... 34 Figure 2.8. MOEYS spending, by category (% of total MOEYS domestic spending) (2011-2021)........... 35 Figure 2.9. MOEYS spending, by category (% of GDP) (2011-2021)....................................................... 35 Figure 2.10. Planned capital spending and budget outturn....................................................................... 36 Figure 2.11. Government expenditures on education percent of GDP vs. GNI per capita (2021)............... 36 Figure 2.12. Learning outcomes for Grade 6 students deteriorated between 2013 and 2021 .................. 36 Figure 2.13. Percentage of Grade 8 students proficiency in Khmer reading and Mathematics (2014-2022)..... 37 Figure 2.14. Loss in learning adjusted years of schooling and increase in learning poverty....................... 38 Figure 2.15. Net enrollment rate, by urban and rural areas (2009-2019/20) ............................................. 38 Figure 2.16. Cambodia improved literacy rate, but still lags behind its peers in ASEAN............................. 39 Figure 2.17. Net school teacher shortage, by province (2018-2021)......................................................... 40 Figure 2.18. MOH expenditure (2017-2022) ............................................................................................ 43 Figure 2.19. MOH spending, by category (% of total) (2017-2022)........................................................... 43 Figure 2.20. MOH spending, by economic classification (KHR, million) (2017-2021)................................. 43 Figure 2.21. MOH spending on social benefits expenses, by economic classification (KHR, million) (2017-2021)......................................................................................................................... 43 Figure 2.22. Cambodia’s health spending as compared to its peer countries in ASEAN, by health care functions, (% of government health expenditure) (2019)........................................................ 44 Figure 2.23. Health expenditure (% of GDP)............................................................................................. 45 Figure 2.24. General government health expenditure per capita, PPP (current intl. US$)........................... 45 Figure 2.25. Domestic health expenditure (% of general government expenditure).................................... 45 Figure 2.26. Cambodia health financing relied mainly on OOP over the last two decades and OOP expenditure tends to increase further.................................................................................... 46 Figure 2.27. Trends in immunization coverage.......................................................................................... 47 Figure 2.28. Trend of childhood mortality.................................................................................................. 47 Figure 2.29. Life expectancy as compared to ASEAN peer countries....................................................... 47 Figure 2.30. Infant and under-5 mortality for both urban and rural significantly declined over the last two decades; child mortality gap between urban and rural areas and gap between poorest and richest persists..................................................................................................................... 48 Figure 2.31. Recurrent spending 2021 (% of total) and poverty rate (2021), by province........................... 48 Cambodia Public Finance Review - From Spending More to Spending Better iii Figure 2.32. Recurrent spending 2021 (% of total) and infant mortality 2021 (deaths per 1,000 live births), by province........................................................................................................................... 48 Figure 2.33. Expenditure trend for economic sector, by line ministry (% of GDP) (2011-2021)................... 49 Figure 2.34. Agriculture allocation as % of total government budget: An international comparison........... 50 Figure 2.35. MAFF expenditure, by category (% of total)........................................................................... 50 Figure 2.36. Composition of MAFF expenditure (% of total domestic budgeted spending)........................ 50 Figure 2.37. MAFF budget outturns ......................................................................................................... 51 Figure 2.38. Actual capital expenditure..................................................................................................... 51 Figure 2.39. GDP growth, constant prices (% year-on-year change)......................................................... 52 Figure 2.40. Growth in Agriculture subsectors (2017-2020)...................................................................... 52 Figure 2.41. Recurrent and external financing for MOWRAM (KHR, million).............................................. 53 Figure 2.42. Composition of MOWRAM spending (2017-2022)................................................................ 53 Figure 2.43. MOWRAM recurrent spending, by economic classification (% of recurrent) (2017-2021)....... 53 Figure 2.44. MOWRAM budget outturn (2017-2021)................................................................................ 54 Figure 2.45. Expenditure trend for infrastructure sectors, by line ministry (% of GDP) (2011-2021)........... 56 Figure 2.46. Number and investment costs of PPPs in Cambodia and selected countries (2017 – 2022)....... 58 Figure 2.47. Cambodia’s PPP Performance (2020)................................................................................... 58 Figure 2.48. Trend of MPWT budget allocation (% of GDP and % of total government budget) (2017-2021)...... 59 Figure 2.49. MPWT budget, by category (% of total)................................................................................ 59 Figure 2.50. Spending by budget classification (% of recurrent)................................................................ 59 Figure 2.51. MPWT budget outturns (% of approved budget, by budget type)......................................... 60 Figure 2.52. Logistics performance index: quality of trade and transport related infrastructure (2007-2018)..... 61 Figure 2.53. Paved roads as percentage of total road network (2015-2020)............................................. 61 Figure 2.54. Expenditure trend for services and cross sectors, by line ministry (% of GDP) (2011-2021) ...... 62 Figure 2.55. Civil servants employed at different levels within selected line ministries; number and share of the total, 2020............................................................................................................................ 63 Figure 2.56. Central and provincial shares of line ministry recurrent spending, 2021................................. 63 Figure 2.57. Monthly average pay of Cambodian civilian civil servants, 2021............................................ 63 Figure 2.58. The footprint of the public wage bill...................................................................................... 64 Figure 2.59. Gini Index before and after fiscal interventions in Cambodia.................................................. 64 Figure 2.60. Change in Gini after fiscal policy (Including in-kind transfers)................................................. 65 Figure 2.61. Overall budget outturn at national level (2011-2021) ............................................................ 66 Figure 2.62. Overall budget outturn (2019-2021) ..................................................................................... 67 Figure 2.63. Public sector wage premium (compared to all private sector workers) (2019)........................ 69 Figure 2.64. Cambodia’s public sector wage premium over time (2003-2019).......................................... 69 Figure 2.65. Wage premium by education level (compared to private formal workers) (2019).................... 70 Figure 2.66. Public sector wage premium by occupational groups (compared to private formal workers) (2019)................................................................................................................................... 70 Figure 2.67. Gender wage gap in Cambodia............................................................................................ 71 Figure 2.68. Transparency in Cambodia compared to its peer countries (2021)........................................ 72 Figure 2.69. Transparency score for Cambodia over time (2010-2021)..................................................... 72 Figure 2.70. Budget oversight in Cambodia (2021)................................................................................... 73 Figure 2.71. Index of public integrity for Cambodia (2020)........................................................................ 73 Figure 2.72. Percentage of public services users paid a bribe (2020)........................................................ 74 Figure 2.73. Percentage of people who think that people in these institutions involved in corruption (2020)..... 74 Figure 3.1. Maternal mortality ratio......................................................................................................... 81 Figure 3.2. Access to maternal health services, gap between poorest and richest (2005-2021/22)........ 82 Figure 3.3. Life expectancy relative to income (2020).............................................................................. 82 Figure 3.4. Cambodia improved life expectancy and mortality rates for both infants and children under-5..... 82 Figure 3.5. Density of doctors, nurses, and midwives in ASEAN countries.............................................. 83 Figure 3.6. Deaths from non-communicable diseases are relatively high and on the rise......................... 83 iv Cambodia Public Finance Review - From Spending More to Spending Better Figure 3.7. Total burden of disease in Cambodia in DALYs Rate 2000 and 2019.................................... 84 Figure 3.8. Cambodia’s health spending and out-of-pocket expenditures as compared to other ASEAN countries (2019).................................................................................................................... 88 Figure 3.9. Expenditure on health and catastrophic health spending, by quintile (2019).......................... 89 Figure 3.10. Financial hardship composition, by quintile (left); by family structure (right)............................ 89 Figure 3.11. Impoverishment from health expenditures, ID Poor card holders and non-ID Poor holders (2019)................................................................................................................................... 90 Figure 3.12. Existence of insurance cards by income quintile................................................................... 90 Figure 3.13. Percent distribution of household members who were ill or injured in the past 30 days and sought health care for the first treatment, by place of treatment, by urban and rural residents (2021)................................................................................................................................... 91 Figure 3.14. Health facilities and doctors at facilities by province (2022)................................................... 92 Figure 3.15. Use of private pharmacies for wealthier and poorer quintiles................................................. 93 Figure 3.16. Type of health care provider consulted first........................................................................... 94 Figure 3.17. Share of total external funding by health focus area (2008-2018).......................................... 95 Figure 3.18. Total external funding percent of GDP and percent of government health expenditure (2008-2018)......................................................................................................................... 95 Figure 3.19. Largest share of total health expenditure was spent on private healthcare providers ............ 96 Figure 3.20. Current health expenditure by diseases and conditions, in current US$ per capita, 2013-2019......96 Figure 3.21. MOH spending, by category (KHR, million) (2017-2021)....................................................... 97 Figure 3.22. MOH budget out turns ......................................................................................................... 97 Figure 3.23. MOH spending on health administration gradually reduced, but its share of total government expenditure remained higher than its peer countries in ASEAN............................................. 98 Figure 3.24. MOH spending, by economic function (2017-2021) ............................................................. 99 Figure 3.25. Starting salary for doctors (KHR/year) (2017-2023)............................................................... 99 Figure 3.26. Program 4 (Strengthen Health System) dominated MOH program budget allocation........... 100 Figure 3.27. Program budgets are centralized........................................................................................ 101 Figure 3.28. MOH spending at provincial level (2017-2021), by economic classification ........................ 101 Figure 3.29. Recurrent spending (percent of total) 2021 vs distribution of staff at health facilities (2022) vs distribution of population (2020) by province........................................................ 102 Figure 3.30. Medical doctors in provinces vs population in provinces (2019).......................................... 102 Figure 3.31. Recurrent spending in provinces (% of total) vs population in provinces (2019)................... 102 Figure 3.32. Spending by level of health facilities (2017-2021)................................................................ 103 Figure 3.33. Spending in provinces, by economic classification, by level of health facilities (2021).......... 103 Figure 3.34. Sources of financing by facility level, 2019.......................................................................... 104 Figure 3.35. Breakdown sources of income by types of facilities (2019) ................................................. 104 Figure 3.36. Unit costs for health centers (in US$ 2019 values).............................................................. 106 Figure 3.37. Tools of government for private sector engagement for UHC in Cambodia......................... 109 Figure 4.1. Employment in agriculture (% of total employment, 2019)................................................... 112 Figure 4.2. Evolution of agriculture’s share of GDP and employment in selected countries (2000-2019)....... 112 Figure 4.3. Public spending in the agriculture sector, (US$ million and % of GDP).................................. 113 Figure 4.4. ODA in Agriculture sector (US$ million actuals)................................................................... 113 Figure 4.5. Agriculture sector spending (domestically financed and DP-financed), % of sector by ministry.... 113 Figure 4.6. Spending by source in 2021 (% of total spending for Agriculture sector)............................. 114 Figure 4.7. MAFF spending, economic composition by chapter (KHR, million)...................................... 115 Figure 4.8. Economic composition of MAFF spending (% of total domestic spending).......................... 115 Figure 4.9. Program-based budget, by program (KHR, million)............................................................. 116 Figure 4.10. Program-based budget, by program (% of total, excluding externally financed projects)..... 116 Figure 4.11. Program budget outturns................................................................................................... 116 Figure 4.12. MAFF recurrent spending, by level of administration (% of total) (2017-2020)......................... 117 Figure 4.13. Program budget, by program across level of administration, (% of total) 2020........................ 117 Cambodia Public Finance Review - From Spending More to Spending Better v Figure 4.14. Civil servants in selected LMs............................................................................................. 117 Figure 4.15. Wage allocations in selected LMs....................................................................................... 117 Figure 4.16. Recurrent budgets and poverty levels in provinces............................................................. 118 Figure 4.17. Recurrent non-wage budgets and poverty levels in provinces............................................. 118 Figure 4.18. BSP and budget planning................................................................................................... 121 Figure 5.1. Trends of MOWRAM’s spending ........................................................................................ 134 Figure 5.2. Economic composition MOWRAM’s spending ..........................................................................134 Figure 5.3. Expenditure outturns compared to original investment expenditure budget (Chapter 21) (2017-2021) ................................................................................................................................135 Figure 5.4. MOWRAM budget by programs (2017-2022) .................................................................... 135 Figure 5.5. MOWRAM budget by functions (KHR, million)..................................................................... 136 Figure 5.6. MOWRAM spending by economic classification (% of total) (2017-2021)............................ 137 Figure 5.7. Central vs PDWRAM spending (KHR, million) (2017-2021).................................................. 137 Figure 5.8. Central vs PDWRAM spending (% of total) (2017-2021)...................................................... 137 Figure 5.9. Central vs PDWRAM personnel number (2017-2021)......................................................... 138 Figure 5.10. Central vs PDARAM spending – Wage expenditure (% of total) (2017-2021)....................... 138 Figure 5.11. Central vs PDRAM spending – Non-wage recurrent spending (% of total) (2017-2021)....... 139 Figure 5.12. Irrigation investment budget 2019 and poverty level in 2019/20.......................................... 142 Figure 5.13. The key steps in the PIM process....................................................................................... 144 Figure 5.14. Agricultural land, by irrigation facilities (000 hectare) (2014-2021)........................................ 146 Figure 5.15. Agricultural land, by irrigation facilities (% of total) (2014-2021)........................................... 146 Figure 6.1. SNA total budget in 2019 (KHR, million).............................................................................. 151 Figure 6.2. SNA budget per capita in 2019 (KHR)................................................................................ 151 Figure 6.3. Proportion of executed budget of SNAs and LDs in Preah Sihanouk province......................... 151 Figure 6.4. Preah Sihanouk province revenue (including transfer), planned and actual (2018-2020)...... 152 Figure 6.5. Preah Sihanouk province expenditure, planned and actual (2018-2020)............................. 152 Figure 6.6. Budget allocation by programs for 2020 (KHR, million)....................................................... 153 Figure 6.7. Tax revenue by composition – plan and actual (2018-2020) ............................................... 156 Figure 6.8. NTR of Preah Sihanouk province by composition, plan vs actual (KHR, million) (2018-2020)...... 156 Figure 6.9. Revenue of Preah Sihanouk province OWSU as % of national total (KHR, million) (2018-2020).........................157 Figure 6.10. Other NTR items besides OWSU fee (KHR, million) plan and actual (2018-220).................. 157 Figure 6.11. Provincial NTR versus DMK NTR (2018-2020).................................................................... 159 Figure 6.12. Monthly execution rate of repair and maintenance (Chapter 6105) and investment expenditure (Chapter 21) in 2020........................................................................................ 163 Figure 6.13. Recorded revenue and expenditure difference, plan and actual (2018-2020)...................... 166 Figure 6.14. Revenue of the Preah Sihanouk province DMK (including transfer), plan and actual (2018-2020)....................................................................................................................... 167 Figure 6.15. Expenditure of the Preah Sihanouk province DMK by economic classification (2018-2020)....... 167 Figure 6.16. Personnel transfer to DMK by ministry and province (2020)................................................ 169 Figure 6.17. Budget allocation for CS budget (2018-2020)..................................................................... 170 Figure 6.18. CS budget execution by months for road construction in 2020........................................... 171 Figure A.1.1. Forgone revenue as a share of collections paid (exc VAT)................................................... 180 Figure A.1.2. Foregone revenue vs. amounts (KHR, billion)...................................................................... 180 Figure A.1.3. Foregone revenue by relief mechanism (KHR, billion).......................................................... 181 Figure A.1.4. Average relief claimed per transaction................................................................................. 181 Figure A.1.5. Top five imported commodities........................................................................................... 182 Figure A.1.6. Importations of construction equipment by type of investor................................................ 182 Figure A.1.7. Share of total QIP or SEZ relief ........................................................................................... 183 vi Cambodia Public Finance Review - From Spending More to Spending Better LIST OF BOXES Box 1.1. Cambodia has a low risk of debt distress according to the LIC-DSF.......................................... 27 Box 1.2. Tax expenditure analysis summary............................................................................................. 28 Box 1.3. The Law on Investment.............................................................................................................. 29 Box 1.4. Global experience - Health excise taxes..................................................................................... 30 Box 2.1. MOEYS’s effort to improve student learning............................................................................... 37 Box 2.2. MOEYS efforts to address the issue of surplus and shortage of teacher.................................... 41 Box 2.3. Current status of Public Investment Management ..................................................................... 54 Box 2.4. Legal and regulatory framework for Public Private Partnership................................................... 57 Box 2.5. Program budgeting (PB)............................................................................................................ 68 Box 3.1. Decentralization and Deconcentration reforms in the healthcare sector...................................... 80 Box 3.2. Health Equity Fund ................................................................................................................... 86 Box 3.3. Strengthening financial management at health facilities under HEQIP....................................... 105 Box 4.1. Agriculture in the 2019 Public Expenditure Review................................................................... 127 Box 6.1. Key rules and regulations for investment projects at provincial levels........................................ 162 Box 6.2. Roles and engagements of private sector in DM service delivery.............................................. 168 Box A.3.1. Roles and responsibilities of public health administrations and facilities .................................. 184 Box A.3.2. Summary of CDHS 2021 and government achievements........................................................ 185 Box A.3.2. Summary of CDHS 2021 and government achievements (Cont.)............................................. 186 Box A.3.3. Level of public health care services as regulated by MOH........................................................ 187 Cambodia Public Finance Review - From Spending More to Spending Better vii GLOSSARY OF ACRONYMS ABEs Authorized Budget Entities GDNT General Department of National ADB Asian Development Bank Treasury ADP Agriculture Development Policy GDP Gross Domestic Product AP Account Payable GDSNAF General Department of Subnational API Advocacy and Policy Institute Finance AR Account Receivable GDPP General Department of Public ASEAN Association of Southeast Asian Nations Procurement ASMP Agriculture Sector Master Plan GDT General Department of Taxation AOI Agriculture Orientation Index GIZ German Agency for International BOG Board of Governors Cooperation GmbH BSP Budget Strategic Plan HEF Health Equity Fund BSRS Budget System Reform Strategy HC Health Center CBHI Community-Based Health Insurance HIC High-income Country CCC Cooperation Committee for Cambodia HCwB Health Center with Bed CCWC Commune Committee on Women and H-EQIP Health Equity and Quality Improvement Children Project CDHS Cambodia Demographic and Health HP Health Post Survey HSP Health Sector Policy CHE Current Health Expenditure ICT Information Communication CISIS Cambodia Irrigation Scheme Technology Information System IP Investment Plan COM Council of Ministers IPI Index of public Integrity COVID-19 Coronavirus Disease of 2019 IRF International Road Federation CP Capital and Provincial KHR Khmer Riel CPA Complementary Package of Activities LIC Low-income Country CPS Center for Policy Studies LPI Logistic Performance Index CRDB Cambodian Rehabilitation and LM Line Ministry Development Board LMIC Lower Middle-income Country CS Commune and Sangkat LJR Legal and Judicial Reform CSES Cambodia Socio-economic Survey MAFF Ministry of Agriculture, Forestry and CSOs Civil Society Organizations Fisheries DBST Double Bituminous Surface Treatment MCFA Ministry of Culture and Fine Arts D&D Decentralization and Deconcentration MCS Ministry of Civil Service DFAT Department of Foreign Affairs and M&E Monitoring & Evaluation Trade MEF Ministry of Economy and Finance DAH Development Assistance for Health MISTI Ministry of Industry, Science, DMK District, Municipality and Khan Technology & Innovation DI Department of Investment MME Ministry of Mines and Energy DP Development Partner MOC Ministry of Commerce EAP East Asia Pacific MOCAR Ministry of Cults and Religion EMIS Education Management Information MOE Ministry of Environment System MOH Ministry of Health FAO Food and Agriculture Organization MOEYS Ministry of Education, Youth and FMIS Financial Management Information Sports System MOIn Ministry of Information FMWG Financial Management Working Group MOLVT Ministry of Labor and Vocational GA General Administration Training GDA General Directorate of Agriculture MOSVY Ministry of Social Affairs, Veterans and GDB General Department of Budget Youth Rehabilitation viii Cambodia Public Finance Review - From Spending More to Spending Better MOT Ministry of Tourism PISA Program for International Student MOWA Ministry of Women’s Affairs Assessment MOWRAM Ministry of Water Resources and PEFA Public Expenditure and Financial Meteorology Accountability MPA Minimum Package of Activities PER Public Expenditure Rview MRD Ministry of Rural Development PFM Public Financial Management MTFF Medium-term Fiscal Framework PFR Public Finance Review MTBF Medium-term Budgetary Framework PHD Provincial Health Department NASLA National School for Local PIM Public Investment Management Administration PIMA Public Investment Management NCDD National Committee on Democratic Assessment Development Pos Producer Organizations NP-SNDD National Program on Subnational PPP Public Priavte Partnerships Democratic Development PPPs Purchasing Power Parity NGO Non-Governmental Organization RMS Revenue Mobilization Strategy NPAR National Public Administration Reform RGC Royal Government of Cambodia NSDP National Strategic Development Plan RS Rectangular Strategy NSSF National Social Security Fund SDGs Sustainable Development Goals ODA Official Development Assistance SDG Service Delivery Grant OD Operational District SOA Special Operating Agencies OPD Out Patient Department SWiM Sector-Wide Management OOP Out-of-pocket SEZ Special Economic Zone OBS Open Budget Survey SOF School Operations Fund OWSU One Window Service Unit SNA Subnational Administration O&M Operation and Maintenance SSCA State Secretariat of Civil Aviation PAEC Policy on Agriculture Extension in TYDA Samdech Techo Voluntary Youth Cambodia Doctor Association PB Program Budget TWG-AW Technical Working Group on PDAFF Provincial Departments of Agriculture Agriculture and Water Forestry and Fishery UHC Universal Health Coverage PDEF Provincial Department of Economy UMIC Upper Middle-income Country and Finance UNFCCC United Nations Framework Convention PFM Public Finance Management on Climate Change PFMRP Public Financial Management Reform WB World Bank Program WDI World Development Indicators PFR Public Finance Review WFP World Food Programme PIB Performance-informed Budgeting WHO World Health Organization PIP Public Investment Program Cambodia Public Finance Review - From Spending More to Spending Better ix ACKNOWLEDGMENT This Public Finance Review (PFR) is the work of a • Maximizing effectiveness of public health policy World Bank Group team co-led by Sokbunthoeun So, by Ildrim Valley, Khy Touk, and Paul Jacob Robyn Senior Public Sector Specialist (Task Team Leader), with contribution from Nareth Ly, Tong Kimsun, Serdar Yilmaz, Lead Public Sector Specialist (co- and Chanrith Prom. Task Team Leader), and Sodeth Ly, Senior Economist • Strengthening performance orientation of public (co-Task Team Leader). Khy Touk provided research spending to boost productivity in Agriculture by assistance and support for the preparation of this Ildrim Valley, Sophal Chan, Mudita Chamroeun, consolidated report. Lindley Higgins and Caroline Kimsun Tong with contribution from Mio Takada, Hughes provided editorial support. Guo Li, and Pike Pike Aye. • Improving capital budget management for This report has been prepared through a joint improved irrigation and water resources exercise with the General Secretariat of PFM Reform management by Kimchoeun Pak, Virak Chan, Steering Committee, Cambodia Ministry of Economy Pike Pike Aye, and Thu Rein Maung Maung with and Finance, following their request for technical contribution from Sokbunthoeun So, Serdar assistance on budget management effectiveness Yilmaz, and Kimsun Tong. issues. It was delivered under the overall guidance of • Key lessons on subnational PFM based on case Alma Kanani (Practice Manager), Sebastian Eckardt study of Preah Sihanouk province by Kimchoeun (Practice Manager); Maryam Salim (Country Manager, Pak, Serdar, Yilmaz, Sokbunthoeun So, and Cambodia); Kim Alan Edwards (Senior Economist Kimsun Tong. and Program Leader); Lalita M. Moorty (Regional These working papers have been discussed with Director), Hassan Zaman (Regional Director); and the relevant government counterparts in a series of Mariam Sherman (Country Director, Cambodia, Lao consultation workshops and have also been peer PDR and Myanmar). The team benefits from the reviewed by relevant sectoral experts. The authors advice of Christian Eigen-Zucchi (Lead Economist) are grateful to peer reviewers for their constructive on the guideline for the preparation of public finance/ feedbacks for refining each working paper including: expenditure review. Miguel Eduardo Sanchez Martin (Senior Economist, The report draws on 6 discussion papers, covering: EECDR); Bernard Myers (Senior Public Sector Specialist, ELCG2); Rama Krishnan Venkateswaran • Achieving fiscal resilience and maximizing (Lead Public Sector Specialist, EAEG1); Farah Zahir fiscal space by Sodeth Ly, Bradley Robert (Senior Economist ESAC2); David Cal MacWilliam Larson, Runsinarith Phim, and Ashima Neb (Senior Economist, ELCMU); Elena Georgieva- with contribution from Sidong Kim and William Andonovska (Senior Public Sector Management Chandler. Specialist, EEAG1); Amal Talbi (Lead Water • Enhancing quality of public spending by Specialist, SWADR); Viengmala Phomsengsavanh Sokbunthoeun So, Serdar Yilmaz, and Khy Touk (Public Sector Specialist, EEAG1); Ajay Tandon (Lead with contribution from Lindley Higgins, Kimsun Economist, HSAHN); Reem Hafez (Senior Economist, Tong, Flavia Giannina Sacco apurro, Zahid Hasnain, HMNHN); Zelalem Yilma Debebe (Senior Economist, Muhammad Faisal Ali Baig, Fata No, Paul Jacob HEAHN); Wendy Karamba (Economist, EEAPV); Robyn, Kimchoeun Pak, and Sadig Aliyev. Anuja Kar (Senior Agriculture Economist, SEAAG); Cambodia Public Finance Review - From Spending More to Spending Better xi Irina Schuman (Senior Agriculture Economist Secretary of State and Secretary General of the SMNAG); and Helene Grandvoinnet (Lead Public General Secretariat of Public Financial Management Sector Specialist, EGVPA). The team would also like Reform Steering Committee (GSC), MEF; H.E. to thank peer reviewers of this consolidated report: Youk Bunna, Permanent Secretary of State, MCS; James Anderson (Lead Governance Specialist, H.E. Yeth Vinel, Under Secretary of State and EGVPA); Christian Eigen-Zucchi (Lead Economist, Deputy Secretary General of GSC, MEF; and H.E. EMFTX); Tuan Minh Le (Lead Economist, EMFTX). Dr. Dy Sovann, Deputy Director General of General Department of Subnational Finance, MEF. The team would like to thank the Ministry of Economy and Finance (MEF), Ministry of Health Coordination support by Seakheang Heng (Program (MoH), Ministry of Agriculture, Forestry, and Fisheries Assistant), Markara Nuon (Consultant), Jolyda (MAFF), Ministry of Water Resources Management Sou (Consultant) during field missions for the and Irrigation (MOWRAM), Preah Sihanouk Provincial preparation of different working papers is gratefully Administration, Ministry of Education, Youth, and acknowledged. Sport, Ministry of Civil Service (MCS), National Committee for Subnational Democratic Development The preparation of this report is co-financed by (NCDD), and all other relevant stakeholders for their the World Bank and a Multi-Donor Trust Fund on support and contributions to the study. Specifically, Public Finance Management and Service Delivery the team would like to thank H.E. Ros Seilava, contributed by Australia and the European Union. xii Cambodia Public Finance Review - From Spending More to Spending Better EXECUTIVE SUMMARY This Public Finance Review (PFR) is delivered fiscal buffers—during the same period government following the request of the Ministry of savings increased ninefold, from 2.1 percent of GDP Economy and Finance for technical assistance in 2011 to 17.8 percent of GDP in 2022. Government to support more effective budget management. savings peaked at 23.6 percent in 2020. All of these The objective of the study is to inform the next phase improvements enabled the government to strengthen of the government’s public financial management its overall fiscal position, providing Cambodia with (PFM) reform strategy (achieving performance fiscal space to respond to Covid-19. However, accountability) and to provide analytical basis despite all of these efforts, Cambodia’s public for the government’s fiscal reform efforts in the debt increased during the COVID-19 pandemic, post-pandemic period. The PFR covers six main from US$8.8 billion (28.1 percent of GDP) in 2019 chapters. The first chapter (achieving fiscal resilience to US$9.9 billion (33.7 percent of GDP) as of end- and building fiscal space) examines macro fiscal 2022. Nevertheless, Cambodia remains at low risk performance. The second chapter (enhancing quality of external and overall debt distress under the Low- of public spending) reviews public expenditure Income Countries Debt Sustainability Framework covering selected key ministries in social, economic according to the joint World Bank/International and infrastructure sectors as well as administrative Monetary Fund Debt Sustainability Analysis in 2022. services and cross sectoral programs. The remaining Expanded fiscal space enabled higher four chapters provided a deep dive exploration into spending, especially in social sectors expenditure and public finance management in: Health, Agriculture, Irrigation and Water Resources, Increased fiscal space allowed the government and a case study of Preah Sihanouk province. The to boost overall spending, which reached 27.9 analysis at the subnational level complements an percent of GDP in 2021, up from 20.2 percent earlier study on intergovernmental fiscal architecture in 2011 (Figure ES.1). In line with the government’s with a case study of Preah Sihanouk province. The medium-term development vision, which prioritized selection of ministries and subnational administration human capital accumulation, spending in the for these deep dive explorations follows the request healthcare and education sectors increased more of the Royal Government of Cambodia. Together than others. The share of social sectors spending these four deep dive chapters cover 21.7 percent more than doubled as a share of GDP to 10.5 of total sectoral spending or 5.6 percent of GDP in percent in 2021 from 4.6 in 2011. The increase in 2021. social sectors spending was driven by the education and health sectors. In contrast, the overall spending Cambodia sustained average annual GDP allocation in economic and infrastructure sectors growth of 7.6 percent between 1995 and 2019, declined between 2011 and 2021. becoming one of the fastest-growing economies in the world. The strong economic growth However, this boost in spending was driven performance together with the implementation of a partly by across-the-board salary increase Revenue Mobilization Strategy (RMS)—RMS 2014- with the average public sector wage now 2018 and RMS 2019-2023—helped to boost tax exceeding the average wage in the private revenues from 12.8 percent of GDP in 2011 to 22.5 sector for the same job percent in 2022. Overall domestic revenues peaked at 25.1 percent of GDP in 2019. In addition, prudent The public sector is a major source of management of public finances helped to increase employment in Cambodia and improvements in Cambodia Public Finance Review - From Spending More to Spending Better 1 Figure ES.1. Government expenditure, by sector grouping (% of GDP) (2011-2021) General administration Defense and security 27.9 30 30 26.8 Social sectors Economic sectors 24.0 25 Multisectors Total expenditure (RHS) 21.8 25 20.2 21.0 20.9 21.7 6.3 20.6 19.5 20.3 6.4 20 20 % of GDP % of GDP 5.4 8.4 6.1 5.4 15 7.8 7.3 5.6 10.5 15 8.7 6.3 8.9 6.8 10 4.5 6.3 6.9 10 5.6 4.6 5.4 5.3 4.5 4.6 3.1 4.1 4.4 5 3.7 3.9 4.1 5 3.1 2.9 3.3 3.3 4.7 4.8 4.8 4.7 2.8 2.5 2.6 2.5 2.9 3.4 2.8 0 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Sources: Budget Settlement Law (2011-2021); World Bank staff estimates Note: RHS = Right-hand scale public sector wages over the years helped to for observable characteristics such as experience, address negative public sector wage premium. educational qualification, gender, and location) was Average monthly public sector pay reached KHR 1.5 18.8 percent in 2019. million in 2019, compared to KHR 1.25 million for the private sector. Before 2019, the public sector workers Higher public sector wages are the main in Cambodia incurred a wage penalty for the same source of growth in the public sector wage bill. job in the private sector (Figure ES.2). Between 2003 Cambodia’s wage bill has increased over time, from and 2012, the wage differentials between public and 3 percent of the country’s GDP and 26 percent of private sector workers have shown a big income public expenditure in 2005 to more than 8 percent of gap in favor of private employment, with a wage GDP and 33 percent of public expenditure in 2020. penalty for public sector workers ranging from 38 to Even though some regional peers show a similar 25 percent. With the introduction of a compensation trend between 2005 and 2020, the rate at which the reform in 2013, the public sector salaries started to public expenditure on the wage bill has increased is improve gradually, leading to positive wage premium. unique to Cambodia. Cambodia’s public wage bill has Cambodia’s public sector wage premium relative to a relatively high footprint on its general government private sector employees (even after accounting revenues compared to its regional peers. Figure ES.2. Cambodia’s public sector wage premium over time 30% 20% premium relative to private sector 18.8% 10% % of public sector wage 0% -10% -20% -24.8% -30% -40% -38.4% -43.4% -50% -48.7% -60% 2003 2006 2008 2012 2019 Source: World Bank’s Worldwide Bureaucracy Indicators 2 Cambodia Public Finance Review - From Spending More to Spending Better Figure ES.3. Government expenditure on education (% of GDP vs GNI per capita, 2021) 8,000 4.1% 5% 3.9% 3.7% 4% 7,000 3.5% 6,000 3.1% 3.1% 4% 3% % of GDP US$/year 5,000 2.3% 3% 4,000 2% 3,000 2% 2,000 1% 1,000 1% 0 0% Cambodia Lao PDR Vietnam Indonesia Philippines Thailand Malaysia GNI per capita Government expenditure on education (% of GDP) (RHS) Source: WDI (extracted 2023) Note: RHS = Right-hand scale Improvements in public sector wages and percent of GDP), and Malaysia (3.9 percent of GDP) higher spending in social sectors have yet (see Figure ES.3). to translate into better outcomes Cambodia’s learning outcomes remain low. Cambodia’s spending on education is the Between 2013 and 2021, an average student in second lowest in the region in terms of percent Grade 6 could answer less than 52 percent of Khmer of GDP and slightly lower than regional average questions correctly. Student learning outcomes for of 3.4 percent of GDP. 1 Cambodia’s gross national Mathematics were even poorer. On average, Grade income per capita is the lowest compared to regional 6 students answered less than half of the questions neighbors and its education spending of 3.1 percent correctly—43 percent in 2013, 48 percent in 2016, of GDP is higher than Lao PDR, on par with Thailand and 38 percent in 2021 (Figure ES.4). (3.1 percent of GDP), and slightly lower than Indonesia (3.5 percent of GDP), the Philippines (3.7 In the health sector, although overall expenditure Figure ES.4. Learning outcomes for Grade 6 students deteriorated between 2013 and 2021 (Percentage of questions correctly answered) 60% Khmer Mathematics 52% 46% 48% 47% 43% 38% 40% 20% 0% 2013 2016 2021 Source: MOEYS 1 Cambodia’s education spending as a percent of GDP is based on current GDP value and is expected to be lower with the upcoming GDP rebasing. Cambodia Public Finance Review - From Spending More to Spending Better 3 Figure ES.5. Type of health provider consulted first 90% 75.3% 83.9% 80% 76.5% 69.1% 70% 60% 50% 2009 2014 2019/20 2021 40% 23.5% 30% 20.1% 22.7% 20% 15.3% 10% 7.4% 0.7% 3.4% 0.6% 0.5% 0.6% 0% 0% Public Private Non-medical Overseas Sources: CSES (2009-2021) Note: Public care includes national hospital, provincial hospital, district hospital, health center, health post, provincial or community based rehabilitation center, and other public service. Private care includes private hospital, private clinic, private pharmacy, visit to home/office of trained health worker/nurses, visit to trained health worker/nurses, other private medical service, shop selling drugs/market. Non-medical care includes Kru Khmer/magician, monk/religious leader, traditional birth attendant. were increased over the years, there is room for Competitiveness Index. The quality of roads, further improving efficiency. In 2021 public health railroads, ports and air transport ranked respectively expenditure was 2.2 percent of GDP , up from 1.8 2 99, 94, 81 and 106. In addition, paved roads as a percent in 2017. The spending increase was driven share of total road networks in Cambodia remained mainly by a rise in health sector wages, in-patient the lowest in ASEAN between 2015 and 2020. services and operational expenses for health centers, While 100 percent of 2,254 Km 1-digit national road whereas spending on drugs remained stable. network is paved, only 72 percent of 2-digit national However, the high levels of expenditure on overhead roads (5,007 km) and 30 percent of 3- and 4-digit (health administration), a smaller percentage of provincial roads (9,031 km) are paved.3 Cambodia expenditure for front line service delivery, and low also ranked lower than most ASEAN neighbors, utilization of public health facilities suggest that there except Myanmar, on the logistics performance index. is a lot of room for improving the efficiency of public More efficient public investment management and health expenditure. The low level of usage of public effective public private partnership (PPP) initiatives health facilities at 15.3 percent (compared to 83.9 could help to improve quality infrastructure. percent for private sector) suggests that public health policy has not been effective (Figure ES.5). In terms of the agriculture sector, Cambodia’s public spending—combined spending of the Similar to social sectors, spending on other Ministry of Agriculture, Forestry and Fisheries sectors has yet to translate into better (MAFF) and Ministry of Water Resources and outcomes Meteorology (MOWRAM)—is relatively high compared to other countries in the region. Cambodia spent 2.8 percent of GDP (combining Cambodia spends around 8 percent of total both domestically and externally funded government budget on agriculture, while the average resources) on infrastructure in 2021; however, spending by the EAP and ASEAN countries was the quality of infrastructure remains poor. around 7 percent and 4 percent respectively between Cambodia scored 3.4 out of 7 and ranked 99 out 2016 and 2020. MAFF recurrent budget allocation of 137 countries on overall infrastructure quality as a share of total budget allocation increased from in the World Economic Forum’s 2019 Global 81 percent in 2017 to 94 percent in 2021, driven 2 MOH expenditure excludes Covid-19 spending in 2021. 3 ADB, transport sector assessment (2019) 4 Cambodia Public Finance Review - From Spending More to Spending Better mainly by a rise in the wage bill. This increase in the Budgeted amounts approved by the National wage bill crowded out public investment resources, Assembly are generally closer to the actual amounts which was reduced three-fold, from 19 percent of spent for the total budget. The most recent PEFA total domestic budget in 2017 to 6 percent in 2021. assessment (2021) suggested a high level of budget This reduction in available resources for agricultural credibility since expenditure outturns compared to infrastructure investment was partly offset by approved budget for 2017, 2018, and 2019 was mobilizing external financing. MOWRAM is especially within +/- 5 percent: 97.9 percent in 2017, 99.4 dependent on externally sourced funds. percent in 2018, and 99.3 percent in 2019 (PEFA 2021). Expenditure outturns based on comparison Cambodia needs to do a better job in between Budget Law and Budget Settlement Law addressing inequalities: fiscal policies was 91 percent in 2020 and 92 percent in 2021.4 reduced inequality but only marginally However, the difference between budgeted and actual spending amounts varies substantially among Fiscal policies—revenue and expenditure— ministries. For example, the Ministry of Agriculture, reduce inequality as measured by a reduction Forestry and Fisheries (MAFF) spent around three in Gini coefficient. While Cambodia does reduce folds its planned investment budget in 2021, inequality through taxes and transfers, the degree of while the Ministry of Education, Youth and Sport redistribution is small by international comparison. (MOEYS) spent only 63 percent of domestic capital The redistributive effect of fiscal policies (including budget for the same year. The Ministry of Post and in-kind transfers) is lower in Cambodia than in Telecommunication’s recurrent budget outturn some other lower middle-income countries. Further fluctuated from 120 percent to 151 percent of total improvements in targeting poorer provinces, such budget between 2017 and 2021. as for the healthcare sector, can help to improve the equity of spending. Strengthening budget discipline can help to improve operational efficiency, and hence the Limitation of planning impedes achieving quality of expenditure. A close examination of allocative efficiency and quality of expenditure suggests a need for more attention to expenditure operational efficiency. In general, recurrent budget There is a significant deviation between amounts outturn is much closer to the approved budget; planned in the National Strategic Development however, capital expenditure fluctuated significantly. Plan (NSDP) and overall government spending, Historically, there has been a huge fluctuation in reflecting weak linkages between government budget outturn for both government and externally strategic planning and budget execution. Annual funded public investment. For government-funded investment spending did not reflect targets in the public investment programs, the budget execution NSDP capital planning. NSDP planned investment rate has been 90 percent for all years, except in 2019 expenditure were over-executed between 2014 when the execution rate reached 130 percent. For and 2019 and under-executed in 2020 and 2021. externally funded investments, the budget execution While the NDSP is expected to be superseded by rate was above 100 percent between 2012 and annual budgets as circumstances evolve, significant 2016, however, the rate went below 100 percent financing gaps between the NSDP and the annual between 2017 and 2021. budgets that Cambodia has experienced so far indicate that planning needs further improvements. Despite steps to enhance the result orientation of the budgeting process, However, annual budget credibility is high. linkages to outcomes remains weak 4 Budget Law and Budget Settlement Law 2020 and 2021 Cambodia Public Finance Review - From Spending More to Spending Better 5 Budget Strategic Plan (BSP) and Program Achieving fiscal resilience and maximizing Budgeting (PB) have been introduced to fiscal space to address structural spending all ministries under the Public Financial needs Management Reform Program (PFMRP). BSP and PB attempt to connect line ministry expenditure Cambodia needs fiscal space to address allocations to policy priorities and outcomes across remaining and emerging development a three-year horizon. The preparation of the BSP challenges. Poverty reduction gains are continually precedes the annual budget preparation and provides threatened by households’ exposure to shocks and the background information and parameters to low human capital formation, particularly among poor shape the basis for budget allocations. However, the and vulnerable households.5 Additionally, Cambodia contents of the BSP currently have weak linkages to is ranked in the top 40 countries most at risk from annual PB. While planned budgets in BSP increasingly the impacts of climate change, particularly to floods, improved over the year with the medium-term budget droughts, windstorms, and seawater intrusion. This framework introduced in the BSP budget circular, is likely to impact Cambodia’s economy, especially BSP could be further improved in terms of realism the agriculture sector. Addressing these development of its planned budget and performance indicators. needs will require sustained public investment while Further, expenditure by a program currently do not maintaining adequate buffers to respond to crises if accurately capture all of the program costs, including and when they occur. personnel costs and capital expenditure costs. This is because PB has been designed to accommodate It will be necessary to further strengthen the the existing administrative organigram of ministries. tax administration system in the next phase Salary expenses are registered in one program of Revenue Mobilization Strategy using the (usually the last program) rather than reflecting the Tax Administration Diagnostic Assessment actual personnel costs of each program. Tool (TADAT). In addition, potential areas for expanding revenue could be explored. This Summary of recommendations may include rationalizing tax expenditure, improving health excise taxes, and introducing climate smart During the last decade Cambodia succeeded fiscal policy. Rationalizing tax expenditure supported in achieving significant domestic resource by robust tax expenditure analysis can help to ensure mobilization which in turn enabled a rapid that exemptions granted contribute to economic expansion of social spending. During the development needs. Cambodia forfeits nearly a third next decade it should shift from spending of its potential revenues from imports as a result more to spending better. The large expansion of of tax exemptions. While free trade agreements public spending on critical social sectors demands will naturally result in the decline of revenues from commensurate efforts to improve the quality of public customs, ineffective incentives can result in further services and ensure that higher spending translates decline of revenues without substantial gain in into better human development outcomes. The economic growth.6 The implementation of the 2021 following policy options could be helpful in achieving Law on Investment, which is more generous than the efficiency and equity to ensure that increased old law, will likely have impact on profit tax collection. spending translates into improved outcomes and Similarly, health excise taxes can contribute to equitable service provision. improved public health while generating revenues 5 Cambodia has recently redefined the poverty line, using the most recent Cambodia Socio-Economic Survey for 2019-20, cost-of- basic need, and common basket approach. The national poverty line is now KHR 10,951 per person per day. Under the new poverty line, about 17.8 percent of the population is identified as poor. See World Bank (2018). Realizing the Promise of Social Assistance. Washington, DC: World Bank. See https://openknowledge.worldbank.org/handle/10986/30925 6 An assessment has not been undertaken about the effectiveness of the the country’s new Law on Investment (2021) increasing the risk that tax incentives erode the revenue base. 6 Cambodia Public Finance Review - From Spending More to Spending Better to help boost health spending needs. Cambodia program budgeting architecture and shift has implemented excise taxes since the mid-1980s from budget negotiation practice based on and has made gradual reforms to the excise tax line items to outcomes to make the best use structures. Increases in tax rates have been few. of performance-informed budgeting reform. There is no comprehensive legal instrument governing The current program architecture of ministries excise tax policy and administration. Bringing excise does not support this performance orientation. tax structure in line with good practice and trends More importantly, Cambodia’s central systems in other countries can help to boost revenue. At (human resource and financial management) the same time, the introduction of climate smart are disconnected and fragmented to support fiscal policy could provide Cambodia with buffers to performance-orientation of budgets. Last but not mitigate climate effects and improve resilience for the least, BSP of sector ministries requires a very economy. Introducing carbon taxes help Cambodia close examination to improve realism and ensure to increase fiscal space to mitigate climate change achievability of targets. impacts on its economy. The foundations of performance-orientation for Cambodia needs to improve the quality of budgets can be strengthened by the PFM reform public expenditure through a number of program. Currently, the Financial Management policy and institutional measures Information System (FMIS) connects the Ministry of Economy and Finance (MEF) to all provincial Linking public sector pay increase to treasuries, all provincial departments of economy performance and finance, and all line ministries; however, the full benefits of the system have not materialized. First, Continuing across-the-board pay increase despite the installation of a treasury single account, without linking them to performance is fiscally there is still an extensive use of cash advance in unsustainable. Salary increase across the board line ministries. Due to inadequate business process has not contributed to improved performance. streamlining, cash advance has been used to The increase in the public sector wage bill places provide quicker access to resources at line ministry Cambodia among those countries that spend a large and service delivery spending unit levels.7 Second, share of government revenues on public wages. All currently FMIS is not functioning at the spending regional peers except Vietnam and Palau (both of unit levels. Hence, all of the spending units of the which have higher per capita income levels) spend sector ministries are not connected to the system. less than Cambodia on their wage bill. A higher Therefore, the transactions for these units are wage bill has the potential to crowd out necessary conducted offline. The recent use of FMIS portal at resources for critical public investment projects. the commune/sangkat level brings FMIS closer to spending units—a step in the right direction. Third, The full scale implementation of there are still limitations in the budget classification, reforms in budget execution system thereby restricting the potential for use of information (rolling out of Financial Management for management purpose. For example, program Information System-FMIS) as well as classification usefulness is limited by program improvements in program budgeting architecture, which does not capture all expenditures including personnel costs and capital investment. Cambodia needs to refine the Budget Strategic Therefore, we would not know for sure how much Plan (BSP) of ministries together with their has been spent on a particular program. This severely 7 FMIS budgetary coverage is quite high, but a more granular assessment of the expenditure transactions ecosystem reveals an interesting profile: although the national level captures 62.9 percent of the total expenditure, it only covers 8.7 percent of all payment transactions. Since only 15 percent of central expenditure’s financial transactions were below US$2,500, the FMIS was mainly used for drawing out advances from Treasury which were then processed off-line. Cambodia Public Finance Review - From Spending More to Spending Better 7 limits the ongoing effort to use performance informed more); whereas the headquarters of the MAFF and budgeting. It is important to fully implement all seven MOWRAM have excess staff. Similarly, within the segments of budget classification, which captures healthcare sector, there are sharp disparities in the information about economic, administrative, program, availability of resources in the richest and the poorest function, geography, source of fund, and project provinces, as well as between urban and rural areas. classification. A comprehensive classification system About 60 percent of the health budget, for example, is the fundamental building block for a performance- is spent at the provincial level; however, the average informed budget management system, facilitating distance to a health center in rural areas is very long, decision-making and contributing to improvements and few doctors are available in rural public health in financial and performance accountability. centers. Improving the adequacy and Strengthening quality of public effectiveness in the distribution of investment management for improved responsibilities and resources across quality of infrastructure investment levels of government Improving quality of infrastructure will require Having the right amount of public resource strengthening public investment management and personnel at the right place is critical to (PIM). Key recommendations in the 2019 Public improving public sector performance, and Expenditure Review (PER) pointed to the need ultimately improving the quality and quantity for upgrading PIM upstream project appraisal; of public services. Cambodia is one of the most strengthening PIM downstream project budget centralized countries in the region, with less than execution; and upgrading PIM database and build 10 percent of total budgetary resources going capacity. While some progress has been made with to subnational governments. Concurrently, the regard to upstream process (as seen in the adoption Cambodian public administration remains highly of PIM Sub-decree and PIM reform strategy) and centralized, even 20 years after the initiation of the upgrading of PIM database with monthly report being government’s deconcentration and decentralization transformed into shared database that automates program. The lack of qualified technical and frontline production of consolidated monitoring report, personnel has been noted in rural Cambodia, while required capacity for downstream implementation central offices of ministries are over-staffed. In the appeared limited. Required PIM guidelines have irrigation sector, for example, each province has only been adopted and it is necessary to build required about 25 staff, less than half what would be required capacity for implementation in line with the guidelines to run just one large irrigation scheme (40,000 ha or (see Table ES.1). Table ES.1. Summary of recommendations on PIM from the 2019 PER Areas for PIM Key What has been achieved and remaining challenges reform recommendations Strengthen Upgrade PIM • Improved PIM legislation: PIM Sub-decree adopted; however, PIM upstream strategy and this sub-decree does not include climate change to support project planning with a PIM resilient infrastructures. appraisal Sub-decree • Public Investment Management System Reform Strategy (PIMSRS) 2019-2025 adopted • Fiscal risk statement developed as part of the MTFF document; however, MTFF is not yet formally adopted by the Council of Ministers. 8 Cambodia Public Finance Review - From Spending More to Spending Better Areas for PIM Key What has been achieved and remaining challenges reform recommendations Strengthen project • Standard criteria adopted in Sub-decree# 41 ANKr.BK provided pre-appraisal guidelines on project rationality, economic and financial viability, and appraisal project options, affordability principles, deliverability principle, and procedures and project implementation. guidelines • Public investment committee (PIC) has been formed to act as gatekeeper and harmonize all public investment sources. • Ensuring the quality of project appraisal remains a challenge. Upgrade project • Projections of project cost for major investment included in prioritization Public Investment Plan attached to budget document and budgeting • Mechanisms to give funding priority to ongoing domestically procedures funded projects exist. • Routine maintenance included in budget, but capital maintenance (infrastructure renovation, reconstructions and enlargements) is not included in any budget documents. Frame the PPP • Law on Public Private Partnership adopted in 2021 and various upstream processes guidelines under a unified • Policy Paper on the Development of PPP Mechanism for Public framework with PIM Investment Project Management 2016-2020 issued in 2016 • Limited PPP information in budget documents Strengthen Strengthen project • Projects are required by Law to adopt competitive and open PIM implementation and procurement. downstream procurement • Required PIM guidelines including for domestic fund have been project/ adopted. budget • No requirement for monitoring or for the independent execution review of complaints; lack of database to monitor key steps in the tender process Develop clear • Several Line Ministries (LMs) have multi-year projections of project adjustment capital spending in Budget Strategic Plan (BSP). procedures and • Capital funding are annually appropriated for domestically and guidance externally financed project. • There may be shifting in budget prioritization resulting in reduced capital investment. Introduce • Department of Investment (DI) verified domestically financed completion review project completion and reported to MEF on monthly and quarterly and ex-post basis. evaluation • Some portfolio oversight has been conducted while ex post audits of individual projects not systematically required. Upgrade PIM Build a PIM • IT systems are being developed to support macro-fiscal database and database (CAPEX) statistics management, performance management, and asset build capacity to register public registry; FMIS supports budget and treasury functions. investment projects • Asset monitoring system is primarily used for controlling and processes asset sales rather than as an input for decision making throughout the public investment management cycle. Sources: PER (2019), PIMA (2019), PEFA (2022) and World Bank staff Cambodia Public Finance Review - From Spending More to Spending Better 9 Cambodia can also leverage upon effective Other than BSP and program budgeting public private partnership (PPP) initiative to challenges discussed earlier, Cambodia’s address infrastructure gap. However, a PPP health sector also faces some key technical project is not ‘free’ as the end users must pay for challenges. Spending at provincial level is not the project through escalating tolls or tariffs over the reducing regional disparities since there is no targeting of poorer areas during the budget term of the concession. It is, therefore, critical to preparation exercise. With more spending at central carefully assess the technical and financial proposals level, there is a shortage of health workers, medicine, (particularly in terms of the tariffs/tolls being charged and equipment at the facility level. Last but not and the tax/royalty exemptions being sought) to least, there are inadequate resources for emerging ensure that projects are providing ‘value for money’ challenges, like non-communicable diseases, at the to the public. health center level—which provide primary health care service to people in rural areas. Maximizing effectiveness of public health policy Improving human resource management and public finances for Cambodia’s health sector is challenged by high better quality education services out of pocket expenditure, inequalities in health outcome, and low utilization rate for public The 2019 PER made three recommendations for health facilities. Out-of-pocket (OOP) spending education sector (i) improving program budgeting in Cambodia at around 60 percent of total health by delegating authority to budget managers, (ii) spending is high compared to around 30 percent for addressing uneven distribution of teachers, and (iii) similar income level countries, exposing households enhancing student learning outcomes. Efforts have been made but there has been limited achievement to financial risks associated with major illnesses in relations to these earlier recommendations (See or accidents. While health outcomes generally Table ES.2). improved, disparity between the richest and the poorest and between urban and rural areas remain Strongly salary-driven budget and ineffective high. Between 2005 and 2021, the rates of mortality human resource management continue for infants and children under five significantly to impact the performance of Cambodia’s declined in both urban and rural areas; however, education sector. With more than 78 percent of rural areas and the poorest populations continue budget spent on wages, human resource is the to experience higher mortality rates for infants and largest expenditure item in the education sector. The children under five. Low utilization of public health Ministry of Education, Youth and Sports (MOEYS) faces challenges in addressing the distribution services reflects the limited effectiveness of public of teachers—there are surplus and shortage of health expenditures. Over the last decade, there was teachers in geographic areas as well in schools. The an increase in households seeking first treatment in distribution of teachers is skewed toward urban areas private sector facilities from 76.5 percent in 2009 to and not responsive to learning needs of students. In 83.4 percent in 2021, while utilization of public health 2021, urban areas had teacher shortage of 490 while services declined from 20 percent to 15 percent rural areas had a net teacher shortage of 19,867. during the same period. This trend is seen even for At provincial level, the teacher distribution is tilted households with access to HEF and NSSF. Higher toward Phnom Penh where surplus of teachers private health care utilization means higher chance of concentrated between 2018-2021. In addition, there increased OOP spending, thus less financial health are surpluses and shortages of teachers in certain protection. subjects at the school level. 10 Cambodia Public Finance Review - From Spending More to Spending Better Table ES.2. Summary of recommendations for improving education outcomes from the 2019 PER Areas of What has been achieved and remaining Recommendations reform challenges Flexibility of • Delegation of authority to MOEYS prepares BSP based on guidance from MEF. school using budget managers to facilitate Three-year rolling PB is prepared. DP funded projects funds expenditure execution are included as part of BSP. • Assign personnel costs to Challenge: (1) School has limited flexibility in specific programs expenditure execution; (2) personnel cost is • Include DP-funded projects capture under only Program 5 and not allocated in in the BSPs respective program. Progress towards school-based School-based management has been introduced and management gradually rolled out Challenge: Delegation of authority to school in managing resources has been limited. Addressing Expand the coverage and Effort has been made to improve enrollment rate in enrollment funding of the current student primary school. gap between grant programs to reduce the Challenge: There is still a gap between urban and rich and opportunity cost of schooling rural areas in education enrollment, especially for poor secondary school between 2014 and 2021. Addressing Reallocate teachers to nearby Sub-decree # 102 dated 2002 set out one time uneven schools through monetary allowance for redeployment; MOEYS’ teacher norm distribution incentives or allowance for was introduced in 2018. of teachers travels Challenge: Shortages and surpluses of teachers continue to be a key challenge for schools in Enforce student-teacher ratios delivering education services. Improving Increase parents’ engagement Effort has been made to strengthen teachers’ student in improving student learning accountability by engaging parents in improving student learning outcomes learning outcomes under school based management outcomes pilots. Enhancing quality of teaching Challenge: many public schools continue to and reducing teacher absence experience limited accountability from teachers. Sources: PER (2019) and World Bank staff The Ministry of Education, Youth and Sports year, leaving half of SOF advance payment idle in (MOEYS) and its provincial departments school commercial bank accounts. Enabling schools also face challenges in budget management to have direct access to common pool of resources practices and systems. At the school level, the in the Treasury Single Account using something management of school operating fund (SOF) is not similar to FMIS portal (successfully piloted at the responsive to efficient public service delivery. SOF is subnational administration level) but tailored to wired to schools as an advance payment via their accommodate the need and business processes of bank account. Transfer of SOF takes place twice a MOEYS could help to make SOF more responsive to year (January and June) to school bank account; spending needs of schools. however, schools withdraw the funds four times a Cambodia Public Finance Review - From Spending More to Spending Better 11 Strengthening performance unit-cost analysis, combining the information with orientation of public spending to boost that in the Cambodia Irrigation Scheme Information agricultural productivity System (CISIS) and use it to support informed decision-making (not yet implemented). Agriculture remains an important sector of Cambodia’s economy. Cambodia’s public Similar to MAFF, improving expenditure spending on agriculture-including MAFF and management is at the heart of MOWRAM MOWRAM- is relatively higher than other countries spending efficiency. MOWRAM’s expenditure in the region. is dominated by capital investments. Therefore, improving the realism of budget planning and the MAFF planned expenditure were characterized execution of capital investment are necessary by recurrent spending and dominated by conditions to increase spending efficiency in the wage bill. The wage bill is increasingly dominating irrigation and water resources management sector. spending in the agriculture sector, reducing fiscal MOWRAM’s investment spending was significantly space for development spending and provision higher than plan, ranging between 100 percent to of key agricultural public goods. Although MAFF 200 percent of approved budget between 2017 and piloted program budget (PB), the implementation 2021. Limited planning practices, poor quality of of PB does not link expenditure to outcomes. Like medium-term investment estimates, and the limited the health and education sectors, the structure practice of open bidding in public procurement of PB in MAFF is not responsive to performance- remained key challenges for improving efficiency in informed budgeting. In addition, there is a limited MOWRAM’s expenditure execution. expenditure planning practice in the sector. MAFF’s BSP amounts significantly deviated from actual The examination of irrigation and water resources spending. Last but not least, MAFF’s performance management’s capital expenditure further monitoring indicators lack realism. This was reflected emphasizes the needs to improve public investment by weak performance matrix and performance management (PIM). Specific to irrigation and water information which are not comparable across resources, key areas that require improvement budget years. Some of these budget management include 1) project pre-appraisal and selection and 2) challenges were also noted in the 2019 PER for increasing the number of open bidding procurement. agriculture, which recommended a number of key Experiences from other countries suggested that actions including (1) improving BSP by consolidating strategic sourcing methodology for procurement has the number of BSP subprograms and improving yielded substantial saving and improved efficiency alignment with national priorities, while systematically on goods and service spending. Another important collecting and monitoring performance data (partially aspect of investment budget management is to implemented); (2) enhancing strategic engagement improve the forecast of ‘flood damage repair’ needs and coordination among sector actors by revamping in order to reduce the ad-hoc /in-year budget from the Technical Working Group for Agriculture the ‘unallocated budget.’ and Water (TWGAW) (partially implemented); (3) undertaking an institutional assessment of current Key lessons on sub-national PFM capabilities and future resource needs in inspection based on experiences from Preah services, plant and animal health, and water usage Sihanouk province (not yet implemented); (4) elaborating on a national agricultural research strategy addressing research In Preah Sihanouk province, overall budget priorities as well as institutional and financing matters management at provincial level encountered a (not implemented); and (5) undertaking regular project couple of challenges. First, there is limited capacity 12 Cambodia Public Finance Review - From Spending More to Spending Better for revenue forecasting and expenditure planning District, Municipality, and Khan (DMK) in the in the province. Lack of data and methodology for province face personnel shortage. Despite the estimation limits the accuracy of forecasting for both transfer of some personnel as part of the transfer of tax and non-tax revenues. For example, there was service delivery responsibilities, personnel shortage significant variation between planned and actual at DMK level remains a key challenge and has an impact on services delivery. In addition, there is a revenue collection for property taxes between 2018 lack of coordination between DMK administration and 2019. The revenue collection from stamp taxes and line ministries to ensure effective local service was five times higher than the plan during the same delivery. At the commune and sangkat (CS) level, period. Second, the province suffers from ineffective the capacity for executing investment projects is budget execution. Although the execution of staff very low. Therefore, delay in project implementation salaries is smooth, the execution of repair and is a major challenge. There is a need to build the maintenance expenditures varies—from 70 percent capacity of commune officials to plan and implement to 195 percent of approved budget. In addition, investment projects. investment expenditure tended to be over executed 1.3 to 4.3 times of the approved budget and there Table ES.3 below presents a summary of policy is a high level of expenditure bunching during Q4 of options considered most important to improving fiscal year. public expenditure efficiency. Table ES.3. Summary of key selected policy options Medium-term Short-term recommendations Sector Challenges recommendations (1- 2 years) (3-5 years) Improving Weak link between • Formally adopting a medium-term Ensure full budget PFM across planning and fiscal and expenditure framework discipline across all line all sectors budgeting to serve as the basis for multi-year ministries development planning • Strengthening coordination among sectors Improving linkages • Revisit and refine Budget Strategic Fully implement between spending Plan to define realistic outcome performance informed and performance indicators and performance budgeting; provide outcomes information that is comparable more discretion to across the reporting period budget managers, and • Provide program directors greater ensure accountability authority over resources allocation for result and hold them accountable for performance outcomes • Review and restructure program budgets across the sector to support links between expenditure and performance outcomes • Develop effective systems to collect non-financial performance data and direct it to program managers and budget officials Cambodia Public Finance Review - From Spending More to Spending Better 13 Medium-term Short-term recommendations Sector Challenges recommendations (1- 2 years) (3-5 years) Addressing foundational elements of PFM: • Strengthen treasury single account and devising mechanism to allow spending unit to have direct access to common pool of resources • Expand FMIS coverage including the use of FMIS portal or something similar for small budget spenders, such as at the school levels • Further improve budget classification and move beyond level one to include all spending units Improving • Introduce performance-based Introduce performance- accountability management for mid-management based management for framework for level (general department and all civil servants performance in public department level) across all sectors sector • Link pay increase to performance Challenges in capital • Further strengthening project pre- Build capacity for all budget management appraisal and selection ministries to implement and public investment • Build capacity for key ministries PIM in line with the PIM management (PIM) involving large public investment legislation Achieving Strengthening fiscal • Maintaining macro fiscal prudence Broadening the fiscal resilience and meet and maintain debt at low risk level income tax base that resilience longer-term spending • Further improve tax administration the informal sector and needs system can otherwise erode maximizing by delivering social • Diversifying revenue sources: fiscal space transfers through Considering more progressive and refundable tax credits stable revenue alternatives such as when there is a fully taxes on personal income, capital operational personal gains, property transfers, carbon, income tax health excise, and digital services • Boosting job creation and social protection programs Build resilience Adopt climate smart fiscal policy and Incentivize green private against climate introduce carbon and coal taxes as a investment projects catastrophe part of broader tax reform 14 Cambodia Public Finance Review - From Spending More to Spending Better Medium-term Short-term recommendations Sector Challenges recommendations (1- 2 years) (3-5 years) Maximizing Shifting from a heavy • Improve efficiency of public health • Revisit the benefits effectiveness reliance on out-of- spending with more resources packages included of public pocket spending allocated for preventive and in current social health policy to increasingly and curative care and less on health health protection efficiently using public administration through shifting line- schemes to improve spending on health in item budgeting elements of provider the protection of line with the Universal payment mechanisms to capitation these schemes Health Coverage • Increase government health to avoid financial Roadmap expenditure for Health Equity Fund hardship (HEF) service payments • Increase health coverage by building on existing social health protection schemes Increasing equitable Expand social protection coverage for Integrate all existing access to health the poor under existing schemes social protection services for the poor schemes, which includes (i) establishing an administration and governance body to oversee a unified health protection scheme, (ii) harmonizing health benefits packages of all health protection schemes, and (iii) ensuring sufficient budget to finance healthcare under the unified health scheme Cambodia Public Finance Review - From Spending More to Spending Better 15 Medium-term Short-term recommendations Sector Challenges recommendations (1- 2 years) (3-5 years) Achieving Improving HR • Improve information on teacher • Delegate authority better quality governance surpluses and shortages for to schools that meet from public management purposes school effectiveness spending in • Improve teacher recruitment and standards to hire education deployment to ensure that newly contractual teachers recruited teachers will stay at their and allocate assigned workplaces resources for it • Introduce redundancy (termination) • Introduce teacher for teachers in subject matters not licensing that in demand and early retirement for requires renewal those not teaching the minimum every specific required number of hours number of years, and that requires • Introduce performance-based salary mandatory service increase and career promotion for in remote locations teachers and education staff • Continue to build capacity Making SOF more Make use of FMIS portal (or something Fully implement school- responsive to school similar tailored to the need of education based management spending needs sector) for school spending needs and strengthen accountability; Enforce teacher standards Agriculture Strengthening • Further strengthening of BSP • Explicit costing and irrigation performance content in terms of realism of of new proposed and water orientation of budget planned budget (following available subprograms, resources to improve agriculture budget ceiling in the formally activities, policy productivity in line adopted circular for BSP) and initiatives, capital with earlier PER performance indicators investments, and recommendations, • Integration of BSP and program required funding which had been logic into negotiations (shifting from to continue with partially implemented. line item based budget negotiation existing activities. to outcome based negotiation), This could include appropriation, monitoring, and expenditures related evaluation, and the legal framework to risks such as climate events. • Map out resources needed for government at national and subnational level 16 Cambodia Public Finance Review - From Spending More to Spending Better Medium-term Short-term recommendations Sector Challenges recommendations (1- 2 years) (3-5 years) Enhancing capital In addition to improvement to PIM in Application of strategic budget management general, there is a need to: sourcing methodology for improved irrigation • Increase the number of open bidding for procurement to investment and procurement improve saving and management efficiency on goods and • Improve the forecast of ‘flood service spending for damage repair’ needs in order to irrigation work reduce the ad-hoc /in-year budget from the ‘unallocated budget.’ Key lessons Improving allocative • Strengthen planning capacity of • Consider increasing from sub- and operational provincial administration the percentage of national PFM efficiency at provincial • Improve tax and non-tax revenue tax revenue sharing in Preah administration level forecasting from CP to DMK and Sihanouk CS levels in order to • Strengthen state asset management province reduce the current • Improve budget execution process vertical imbalance among the three tiers of SNAs • Ensure check and balance between the C/P governors and C/P councils Improving HR • Address personnel shortages for key Move toward governance at DMK services such as water supply, solid decentralized finance administration waste management, and mining and HR management • Clarify how technical staff at the with centralized DMK administrations should perform information supported their daily operations under the new by modern shared working arrangements platform GovTech system Building capacity of Support CS officials to implement Monitor the CS administration the circular associated with the CS implementation and on project development project implementation to collect feedback for implementation prevent delays to project development future adjustments and and preparation strengthening Cambodia Public Finance Review - From Spending More to Spending Better 17 1 ACHIEVING FISCAL RESILIENCE AND MAXIMIZING FISCAL SPACE 1.1. Introduction growth culminated in a fiscal surplus in 2018–2019. However, slower revenue collection and increased Prior to the COVID-19 pandemic, Cambodia was fiscal support to offset economic and social impacts one of the fastest growing economies in the of the pandemic resulted in a fiscal deficit of 4.5 world, with a sustained average real growth rate percent of GDP in 2020 and 6.1 percent of GDP in of 7.6 percent over the period 1998–2019. This 2021. To finance these deficits, the government drew growth was driven largely by tourism, manufacturing upon its fiscal reserves and borrowed from official exports, and real estate and construction. The creditors. As result, Cambodia’s outstanding public country reached lower middle-income status in debt increased from 28.1 percent of GDP in 2019 to 2015 and is aspiring to attain upper-middle-income 33.7 percent of GDP in 2022 while its fiscal reserves status by 2030 and high-income status by 2050. (government deposits) declined from 21.7 percent of However, the global epidemiological and economic GDP to 17.8 percent of GDP over the same period.8 crises unleashed by COVID-19 significantly hurt the Rising demand for public expenditures will continue to economy. After shifting to “living with COVID-19” in put upward pressure on the government’s budget for late 2021, the economy is on a path to recovery, but years to come. Cambodia needs to adopt measures challenges remain for public finances. under the new Revenue Mobilization Strategy (RMS) to further strengthen tax administration system Cambodia has pursued a conservative fiscal and increase revenue collection to meet future policy that created sufficient fiscal buffers expenditure requirements and restore fiscal buffers. for the government to respond effectively to economic fallout from the COVID-19 pandemic. This chapter assesses Cambodia’s fiscal policy Decades of fiscal reforms and solid economic from a fiscal sustainability perspective. It starts 8 Despite the increase in public debt, Cambodia remains at low risk for debt distress. 18 Cambodia Public Finance Review - From Spending More to Spending Better with an overview of recent economic trends. It then revenues in 2020, while total expenditures increased describes the drivers of revenue and expenditure because of higher spending commitments under trends over the last decade, including during the the public health, social, and economic intervention COVID-19 pandemic period. It then assesses the programs, which amounted to 3 percent and 4.9 impact of economic trends and fiscal reforms on percent of GDP in 2020 and 2021 respectively. The Cambodia’s fiscal balance, debt stock, and debt COVID-19 pandemic thereby turned modest overall service requirements. It concludes with policy surpluses into widening fiscal deficits. Nevertheless, recommendations to improve fiscal management government deposits stand at around 17.8 percent and reestablish fiscal buffers. of GDP as of end-2022 and Cambodia is assessed to be at a low risk of debt distress under the 2022 1.2. Recent macroeconomic Joint World Bank/International Monetary Fund Debt trends Sustainability Analysis.10 But a program of fiscal consolidation, including increased attention to fiscal The economic shock caused by COVID-19 sustainability and resilience, is necessary to rebuild was severe, pushing Cambodia into its first fiscal buffers and guard against future shocks. recession in 25 years. The pandemic’s impact was propagated through falling global demand, supply After shifting to “living with COVID-19” in chain disruptions, and nationwide lockdowns. late 2021, the economy is firmly on a path to GDP growth dropped sharply from 7.1 percent in recovery and has now returned to its pre- 2019 to -3.1 percent in 2020, before recovering pandemic growth trajectory. Cambodia’s to 3 percent and 5.2 percent in 2021 and 2022, economic recovery solidified in 2022 with real growth respectively. Cambodia’s key merchandise exports accelerating to 5.2 percent, up from 3 percent in of garment, footwear, and travel goods contracted 2021. Initially led by the strong performance of by 8.1 percent in 2020 as a result of multiple shocks, export-oriented manufacturing, growth drivers are including the temporary shortage of raw materials rotating to the services and agriculture sectors. from China, plunging export orders from the United The overall contribution of the services sector to States and Europe, and the partial withdrawal of the economic growth is returning to 2019 levels, and EU’s “Everything But Arms” (EBA) trade preferential the complete removal of COVID-19-related mobility treatment. Simulations indicate that the economic restrictions and China’s recent reopening have contraction reversed two years of Cambodia’s caused a rebound in international tourist arrivals. poverty reduction progress, increasing the poverty Cambodia’s real GDP growth is projected to rate by 2.8 percentage points and pushing around accelerate to 5.5 percent in 2023, despite a decline in 460,000 individuals into poverty.9 external demand, which has weakened the country’s export-oriented manufacturing. Over the medium The government took advantage of the fiscal term, the economy is expected to trend back to space afforded by its prudent fiscal policy to potential, growing at 6 percent. However, a return respond to these shocks, but those buffers are to pre-pandemic growth rates of above 7 percent now depleting. After years of running fiscal deficits, is not expected, largely reflecting the challenging Cambodia achieved small fiscal surpluses in 2018 external environment, including synchronous policy and 2019. Subdued domestic economic activity tightening around the world aimed at containing very and pandemic-related fiscal relief reduced total high inflation, worsening financial conditions, and 9 Poverty simulations are based on actual GDP, sectoral growth rates, and estimated employment figures for 2020. Simulations were implemented using the ADePT macro-simulation module and using the new national poverty line of KHR10,951/person/day in 2019/20 Phnom Penh prices. 10 See https://www.imf.org/en/Publications/CR/Issues/2022/12/15/Cambodia-2022-Article-IV-Consultation-Press-Release-Staff- Report-and-Statement-by-the-526993 Cambodia Public Finance Review - From Spending More to Spending Better 19 continued disruptions stemming from Russia’s war reserves. Finally, increasingly leveraged corporate in Ukraine.11 balance sheets create the potential for a debt overhang, which could weigh on future investment Key economic indicators are expected to and hiring decisions. Fast credit growth and the normalize in the coming years. Inflation has eased concentration of domestic credit in the construction significantly, with high levels of dollarization and a and real estate sector may also lead to a rise in broadly stable KHR versus U.S. dollar exchange Non-Performing Loans (NPLs), which could pose a rate keeping expectations relatively well-anchored. threat to financial stability, with potential spillovers Goods and services export growth is projected to to the real economy. The NPL ratios for the banking improve, outpacing goods and services import and microfinance sectors were 4 percent and 3.1 growth, due largely to a rapid recovery in tourism and percent, respectively in the second quarter of 2023, remittances. according to the mid-year 2023 report of the National Bank of Cambodia. Nevertheless, the broader macroeconomic outlook is subject to downside risks emanating 1.3. Revenue from a deteriorating global environment. The ongoing impacts of higher energy prices are Prior to the COVID-19 pandemic, Cambodia more expected to cause the current account deficit to than doubled its domestic revenue collection remain at relatively high levels. Cambodia’s economy under the Public Finance Management Reform remains vulnerable to downturns in external demand Program (PFMRP), from 10.4 percent of GDP and disruptions in global supply chains, given its in 2005 to 25.1 percent of in 2019. Under the reliance on a relatively undiversified set of exports. PFMRP, the government improved customs and tax Cambodia has also been experiencing a deceleration administration, eliminated off-budget expenditures, of foreign currency deposit growth, reflecting the and established a cash management system that easing of capital inflows, and depreciation pressures sweeps all revenue account balances held in the on the KHR/U.S. dollar exchange rate that required banking system into a Treasury Single Account interventions in the foreign exchange market and (TSA) at the end of the business day, enabling the led to the decline in Cambodia’s gross international government to move from a situation of chronic cash Figure 1.1. Cambodia has substantially improved domestic revenue mobilization (% of GDP) a. Cambodia, 2011-2022 b. Peers, average (2017-2019) 25 25 20 20 % of GDP 15 15 10 10 5 5 0 0 a 2015 2020 M DR pp a La ia Th es C and di i Ph ays s in bo ne P l ai o al am do Taxes Grants ili In of which: VAT Other revenues of which: Pro t Sources: Staff calculations based on RGC Budget Settlement Laws, World Bank staff estimates, and GFS (2023) 11 World Bank (2023). East Asia Pacific Economic Update: Reviving growth. Washington, D.C. See https://openknowledge.worldbank. org/collections/892c8eb3-e78d-5012-875b-98af1149d8a8. 20 Cambodia Public Finance Review - From Spending More to Spending Better shortages to a rapid accumulation of cash surplus of GDP, representing a 21.9 percent year-on-year (fiscal reserves). Taxes on goods and services increase in 2022. In 2022, direct revenue, consisting accounted for the largest source of growth, while mainly of profit tax, rose quickly to 6.1 percent of non-tax revenue remained largely unchanged (Figure GDP, exceeding VAT collection of 5.8 percent of GDP 1.1, panel a). On the eve of the COVID-19 pandemic, for the first time in the recent history of domestic Cambodia’s revenue as a share of GDP was higher revenue collection. Non-tax revenues are expected than its regional peers (Figure 1.1, panel b).12 to remain well below pre-pandemic levels, as they are heavily dependent on tourism receipts. These reforms helped redress fiscal problems and established the buffers necessary to Cambodia’s current tax mix is such that it respond to the COVID-19 pandemic. Payment raises most of its revenues from indirect taxes, arrears that amounted to about one quarter of the contributing to a relatively efficient tax system. annual budget in 2005 were eliminated in 2007, It is worth noting that Cambodia’s tax structure ensuring the predictability and credibility of the consists of both direct tax (profit tax), indirect taxes government budget. After years of running fiscal (VAT and excise tax) and international trade tax. deficits, Cambodia achieved a fiscal surplus in 2018 The largest component of Cambodia’s tax revenue and 2019 (of about 0.4 percent and 1.5 percent of continued to be taxes on goods and services GDP, respectively). The authorities accumulated (indirect taxes), comprising mainly the value-added government deposits (fiscal reserves), which peaked tax (VAT) and excise taxes, accounting for 56.2 in 2020 at 23.7 percent of GDP and were used as percent of tax revenue (10.6 percent of GDP) in fiscal buffers during the pandemic. Accumulation of 2022. However, collection of direct tax revenues government deposits have been underpinned by the improved steadily, thanks to better tax administration success in establishing the Treasury Single Account, and modernization. The share of direct tax revenue built upon a centralized cash management system collection to total tax revenue rose to 32.3 percent after eliminating off-budget expenditures under (6.1 percent of GDP) in 2022, up from 19.8 percent the Public Financial Management Reform Program (2.3 percent of GDP) in 2012. During the same period, (PFMRP). Revenue collection improvements together taxes on international trade declined to 11.5 percent with an efficient cash management system enabled of tax revenue (2.2 percent of GDP), down from both more allocation of spending and accumulation 20.7 percent (2.4 percent of GDP) due to increased of fiscal reserves. commitments under bilateral and regional free trade agreements. During the pre-COVID-19 period, rising The impact of the COVID-19 pandemic on the tax revenues from goods and services were driven in economy, coupled with tax relief to affected part by strong consumption and investment demand firms, significantly impacted Cambodia’s for durable goods during the construction boom revenue base. Total revenues, including grants, fell period covering 2015–2019. Continued strong from a peak of 27 percent of GDP in 2019 to 22 performance of domestic tax collection is attributed percent of GDP in 2021.13 to better tax administration and modernization, with the introduction of e-filing, e-payment, and better Domestic revenue collection quickly recovered taxpayer services. In addition, efforts have been following the COVID-19 pandemic, reflecting made to eliminate loopholes in the Law on Taxation the resiliency of the government’s fiscal reforms and to harmonize the legal framework to prevent and the strength of the country’s broad-based fiscal evasion, transfer pricing abuses, and double economic recovery. General government domestic taxation. A taxpayers’ survey revealed that 88 revenue (excluding grants) reached 22.5 percent percent of respondents agreed that modernization 12 Budget Settlement Laws and World Bank staff estimates; WDI 2023. 13 Revenue collection remained relatively strong in 2020, when ballooned profit taxes for 2019 were filed. Cambodia Public Finance Review - From Spending More to Spending Better 21 of tax administration, especially automation, helped sectors through targeted relief (see Annex 1.2). improve tax payments.14 1.4. Expenditure Cambodia could substantially increase revenue from direct taxes by introducing a personal Prior to the pandemic, Cambodia’s rapidly income tax and rationalizing tax incentives. expanding fiscal revenue enabled it to There is currently no personal income tax (PIT), substantially boost general government per se, in Cambodia. Instead, a monthly salary tax expenditure, from 22.5 percent of GDP in 2011 to is imposed on individuals who derive income from 25.5 percent of GDP in 2019 (Figure 1.2). Across- employment. Introducing a PIT would also mitigate the-board salary increases comprised the largest the regressivity of Cambodia’s current consumption- share. Over the period 2013–2020, Cambodian based tax structure. Furthermore, while the country authorities tripled civil servants’ minimum wage to has a competitive corporate income tax rate, more than 1 million KHR (US$250) a month, in turn collections underperform because of the country’s increasing wages and compensation as a share of generous tax incentive system. Cambodia currently GDP from 5 percent to 8.1 percent. Over the same forfeits nearly a third of its potential revenues from period, spending on social sectors—especially imports. Over the period 2016–2020, the amount education—increased significantly, broadly in line of revenue foregone on imports into Cambodia with the government’s priorities (see chapter 2). rose about 45 percent from around KHR 3.3 trillion Meanwhile, Cambodia’s rapidly expanding physical to nearly KHR 4.7 trillion. Cambodia is expected infrastructure was increasingly financed by domestic to forego even more revenue in the years to come capital,15 underpinning a structural transformation as under the 2021 Law on Investment, which is more the economy grew, with demand for operations and generous than its predecessor. Any future tax maintenance increasing commensurately. expenditure reforms will need to balance revenue performance and the necessity to stimulate specific The government responded to the economic Figure 1.2. Expenditure increased significantly during the COVID-19 pandemic (% of GDP) a. Cambodia, 2011-2022 b. Peers, average (2017-2019) 30 25 25 20 20 % of GDP 15 15 10 10 5 5 0 0 2015 2020 a am DR ia es a nd di si es in bo ay la P Capital expenditure Expenditure on goods and services on pp ai o al Th La d ili M Other expenditure Wages and compensation In Ph C Interest payment Sources: Staff calculations based on RGC Budget Settlement Laws, World Bank staff estimates, and GFS (2023) 14 “Tax Bulletin - Quarter Four, 2022”; https://www.tax.gov.kh/en/tax-bulletin. 15 The authorities have been boosting their domestically financed capital investment projects, underpinning manufacturing and tourism activities, two main growth drivers to facilitate the on-going structural transformation. A US$300 million public investment project to build 34 roads with a total length of 82.5 kilometers in the Preah Sihanouk province where Cambodia’s largest deep-sea port and special economic zone are located, was approved in 2019 and completed in 2020. A similar US$150 million public investment project to build 38 roads in Siem Reap province, the country’s largest tourist attraction site, was approved in 2020 and completed in 2021. 22 Cambodia Public Finance Review - From Spending More to Spending Better crisis caused by the COVID-19 pandemic 2022, the program had disbursed US$714 million through timely and targeted support to (2.4 percent of GDP) and was widely successful in households, workers, and firms (Table 1.1). reaching IDPoor households and alleviating financial Among the most significant interventions was the hardship resulting from the COVID-19 pandemic. A government’s expansion of the country’s social recent study found significant positive impacts of the protection systems. In June 2020, the government cash transfer program across human development leveraged the existing IDPoor registry to deliver dimensions and socioeconomic indicators including cash transfers to vulnerable households. It has food security, nutrition, children’s education, savings, since extended the program multiple times to reach debt repayments, productivity, healthcare, gender approximately 706,060 households (2.8 million empowerment, and perception of local and national individuals) or 17 percent of the population. As of July governments.16 Table 1.1. Government fiscal intervention (% of GDP) 2020 2021 2022 Intervention Plan Disbursed Plan Estimate Plan Health Masterplan Scaling up health response by increasing 0.4 0.1 0.1 2.5 0.9 & Outbreak prevention and detection facilities, clinical Prevention and management, and treatment as well as Treatment coordination and support system Wage subsidy and Providing partial wage subsidies of US$40 0.2 0.2 0.2 0.2 0.2 skill training per month and technical/soft skills training for furloughed workers in the tourism and garment industries Cash for work Providing jobs in rural areas through 0.4 0.4 0.5 0.5 0.3 construction, upgrade, and maintenance of rural roads, drainage and small-scale irrigation Cash Transfer Providing monthly cash grants to poor and 1.2 1.1 0.7 1.1 0.6 vulnerable individuals that are registered in the government’s IDPoor database Food support Providing food support to local people during 0 During Lockdown lockdown period Financing through Providing low-interest loans for working 0.2 0.2 ARDB capital and investment in agricultural sector through capital injections to Agricultural and Rural Development Bank Co-Financing Co-financing with commercial banks by 0.2 0.2 through SME Bank providing low-interest loans for working capital and investment in 6 targeted SME sectors through newly established SME bank Credit Guarantee Providing capital to establish the Credit 0.8 0.7 Fund Guarantee Corporation of Cambodia which helps bear risk-sharing with businesses SME Financing Reserving contingent funds to provide any 1.2 0.1 0.9 0.5 0.8 Facility necessary financing to SME sector Total 4.6 3.0 2.4 4.8 2.8 Source: RGC (2023) 16 See “Socioeconomic Impacts of the COVID-19 Cash Transfer Program in Cambodia.” Cambodia Public Finance Review - From Spending More to Spending Better 23 The cash transfer program appears to have administration). Wage bill spending has also been ushered in a new era of more robust social contained, in part due to a freeze in public-sector protection systems in Cambodia. Additional salaries since 2021. As a result, the wage bill social assistance measures were introduced from declined to 7.2 percent of GDP in 2021,18 although December 2022 to March 2023 to address the the country’s public-sector workers still earn a higher food and oil price shocks and fallout from major average salary than their private sector counterparts floods.17 The government also recognized the need because of the government’s reforms in 2013–2018. for medium-term reforms to support the country’s economic recovery, including measures to improve 1.5. Fiscal balance Cambodia’s competitiveness for medium to long- term growth, particularly by improving productivity, For decades, Cambodia has pursued a trade facilitation, and ease of doing business. conservative fiscal policy resulting in low debt and high government deposits. Revenue and Driven mainly by these interventions, public expenditure increased in tandem until the COVID-19 expenditure reached 26.8 percent of GDP pandemic, resulting in small fiscal deficits in most in 2020, up from 24 percent of GDP in 2019. years (Figure 1.3, panel a). To finance these deficits, Spending on subsidies and transfers as part of the government drew upon its fiscal reserves the fiscal response program rose from 1.6 percent and borrowed from official creditors (Figure 1.3, in 2019 to 2.9 percent in 2020 and 3.2 percent of panel b). Despite economic downturns during the GDP in 2021. Development (capital) expenditures— Asian Financial Crisis and COVID-19 pandemic, much of which are financed by development Cambodia’s conservative fiscal policy enabled it to partners—also increased, from 7.7 percent of GDP preserve fiscal buffers. in 2019 to 8.7 percent in 2020. To offset these increases, the government reduced spending on Cambodia’s fiscal buffers have enabled it to goods and services and “other expenditures” (which implement a countercyclical fiscal policy. Given include administrative expenses and subnational the high level of dollarization of the Cambodian Figure 1.3. Cambodia has pursued a conservative fiscal policy resulting in low debt and high reserves a. Revenue, expenditure, and scal balance b. Debt and scal reserves 40 40 30 30 20 % of GDP % of GDP 20 10 0 10 -10 2008 2011 2014 2017 2020 2023f 0 2008 2011 2014 2017 2020 2023f Revenue Expenditure Overall balance Primary balance General government debt Fiscal reserves Sources: Staff calculations based on RGC Budget Settlement Laws and World Bank staff estimates Note: Revenue and balance figures include grants. 17 https://www.khmertimeskh.com/501253915/cash-transfer-programme-for-flood-affected-families-completed/ 18 In part due to a reclassification of social security and social assistance were. 24 Cambodia Public Finance Review - From Spending More to Spending Better economy, this was the only viable policy instrument from 3.1 percent of GDP in 2019 to 4.5 percent for maintaining macroeconomic stability. In 2009, of GDP in 2022 (see Annex 1.1, Table A.1.3). The the government provided fiscal stimulus equal to authorities effectively used the availability of fiscal approximately 4.2 percent of GDP to mitigate the space (capacity to borrow) to help mitigate the negative impacts of the 2008–2009 global financial detrimental effects of the pandemic by introducing a crisis. Subsequent fiscal consolidation enabled it to number of social protection programs, especially the rebuild the fiscal buffers it then employed in response COVID-19 cash transfer. While global trends in cash to the COVID-19 pandemic (Figure 1.4, panel a). transfer responses indicated that cash transfers As a result, Cambodia’s fiscal intervention in 2019 were of short duration, averaging 4.5 months,19 the was relatively large compared to other lower-middle Cambodian authorities extended the COVID-19 income countries, including those in the region cash transfer program in 2023. As of March 2023, (Figure 1.4, panel b). it had disbursed US$994 million since the launch in June 2020. The funds have been well spent for the Due to the scale of the government fiscal programs which will be phased out over time. intervention during the COVID-19 pandemic, the overall fiscal position has deteriorated As the economy recovers, authorities remain quickly, from modest fiscal surpluses in 2018– committed to restoring fiscal discipline, 2019 to historically large fiscal deficits in 2020– rebuilding fiscal buffers, and maintaining 2022. A drawdown of government deposits (fiscal prudent fiscal management and stable reserves) covered 20.5 percent of the deficit in 2020 macroeconomic performance. Due mainly and nearly two-thirds of the deficit in 2021, reducing to improved revenue collection, the fiscal deficit the stock of government deposits from 23.6 percent is expected to narrow to 4.7 percent of GDP in of GDP at the end of 2020 to 17.8 percent of GDP 2022. Plans to improve revenue collection in the in 2022. The rest of the deficit was financed by coming years include modernizing the tax system external borrowing, with Cambodia’s official creditors and customs administration by embracing further accelerating their loan disbursements, which rose reforms. A new revenue mobilization strategy to cover Figure 1.4. Cambodia’s fiscal discipline has enabled countercyclical spending (% of GDP) a. Output gap vs. scal balance b. Fiscal response to COVID-19 50 LOW MIDDLE HIGH 6 INCOME INCOME INCOME 3 40 0 % of GDP % of GDP 30 -3 -6 20 -9 10 -12 0 2000 2005 2010 2015 2020 500 5000 50000 GDP per capita Primary scal balance Output gap Cambodia ASEAN Other Sources: Staff calculations using data from IMF 2022; World Bank 2023; and Kose, M. Ayhan, Sergio Kurlat, Franziska Ohnsorge, and Naotaka Sugawara. 2022. “A Cross-Country Database of Fiscal Space.” Journal of International Money and Finance 128 (November): 102682. 19 Ugo Gentilini. 2022. Cash transfers in Pandemic Times: The World Bank. See https://documents1.worldbank.org/curated/ en/099800007112236655/pdf/P17658505ca3820930a254018e229a30bf8.pdf Cambodia Public Finance Review - From Spending More to Spending Better 25 the period 2024–2028 is being prepared that may of loans in 2022 financed the country’s public introduce a personal income tax and excise policy infrastructure sector, including transport, irrigation, reforms. Authorities also continue to work to improve energy, and water supply, while the remaining 31 resource allocation, linking the budget to medium percent funded other prioritized (non-infrastructure) term planning and development objectives under sectors, including health, education, and agriculture. the public financial management reform program as a part of Platform 3 (Policy Budget Linkage) and The joint World Bank/International Monetary the current Platform 4 (Performance-Accountability). Fund Debt Sustainability Analysis conducted in However, continued upward demand for public 2022 indicated that Cambodia remained at low expenditures to finance social protection and the risk of external and overall debt distress under general election in 2023 will remain, and total public the Low-Income Countries Debt Sustainability expenditure is projected to hover around 27 percent Framework (Box 1.1). The total public and publicly of GDP in 2023 and 2024. As a result, the deficit guaranteed debt-to-GDP ratio is projected to rise by is projected to deteriorate to 5.9 percent of GDP in around 4 percentage points during the next decade. 2023 before again improving to 4.3 percent of GDP The present value of the external debt-to-GDP ratio in 2024. breaches its threshold in the exports stress test, which would imply a moderate risk rating. However, 1.6. Public debt the breaches are small, short-lived, and driven by the exceptional volatility of exports in 2020 that has Cambodia’s public debt increased during the increased the sample volatility for the standardized COVID-19 pandemic, from US$8.8 billion (28.1 shock. Moreover, other debt burden indicators are percent of GDP) in 2019 to US$9.9 billion (33.7 projected to remain well below their thresholds under percent of GDP) as of end-2022. Approximately the baseline and the shock scenarios. two-thirds of Cambodia’s public debt is owed to bilateral donors, with China accounting for about A gradual fiscal consolidation is expected to 40 percent of the total debt stock.20 Approximately result in public debt stabilizing at its current one-third of public debt is owed to multilateral level. As growth converges to potential, revenue donors, mainly the Asian Development Bank. Only performance is expected to improve, while spending a tiny fraction (0.2 percent) of public debt is owed to is expected to remain higher than pre-COVID levels domestic lenders, but Cambodia is in the process of due to a continuation of support measures for establishing a domestic debt market to help diversify vulnerable households, increased public investment financing sources, promote domestic savings, and to support the economic recovery, and larger de-dollarize the economy. Public debt is mainly transfers to subnational administrations to support denominated in U.S. dollars. decentralization reforms. Taken together, these trends are consistent with a gradual narrowing of the External borrowing remained highly fiscal deficit over the next few years, with public debt concessional during the pandemic, with a expected to stabilize at the current level of just above weighted grant element of 44.8 percent in 35 percent of GDP. Around 80 percent of gross 2022—high by international standards. The financing needs will continue to be met by external weighted average interest rate of contracted loans borrowing, with the remainder to be financed by a was 1.1 percent per year, while weighted average drawdown of fiscal reserves and domestic bond maturity was 26.7 years. Approximately 69 percent issuance. 20 There is no loan classification by creditor and by sector; therefore, it is not possible to elaborate what sectors the proceeds from Chinese loans have been invested in. However, based on the latest biannual public debt statistical bulletin publication (volume 18, Q1-2023), 70 percent of all loan proceeds have been invested in physical infrastructure (transport, energy, water supply, irrigation and other infrastructure) while the remaining 30 percent in other priority sectors (health, education, agriculture and other priorities). 26 Cambodia Public Finance Review - From Spending More to Spending Better Box 1.1. Cambodia has a low risk of debt distress according to the LIC-DSF Cambodia Joint Bank-Fund Debt Sustainability Analysis Risk of external debt distress Low Overall risk of debt distress Low Granularity in the risk rating Not applicable Application of judgment Yes: the breach of the export shock stress test is small and temporary. Cambodia remains at low risk of external and overall debt distress under the Low-Income Countries Debt Sustainability Framework (LIC-DSF). The current debt-carrying capacity remains consistent with a medium classification. The baseline scenario assumes steady economic recovery, largely driven by exports of tourism services and, to a lesser extent, manufacturing goods, notwithstanding pressures on external demand in the near term. The total public and publicly guaranteed (PPG) debt-to-GDP ratio is projected to rise by around 4 percentage points during the next decade. The present value of the external debt-to-GDP ratio breaches its threshold in the exports stress test, which would imply a moderate risk rating. However, the breaches are small, short-lived (only four periods, the last barely), and driven by the exceptional volatility of exports in 2020 that has increased the sample volatility for the standardized shock. Moreover, other debt burden indicators are projected to remain well below their thresholds under the baseline and the shock scenarios. Given the expectation of a steady and solid recovery, and because the breaches for one of the tests are small, temporary, and driven by the exceptional volatility of exports in 2020, judgment has been applied that the external risk of debt distress remains low. Nonetheless, the analysis shows that debt sustainability remains vulnerable to shocks in exports and growth. These findings reinforce the importance of implementing reforms to increase the economy’s resilience to external shocks and to facilitate export and economic diversification. Efforts to mobilize fiscal revenue, strengthen the Public Investment Management Framework, and further enhance monitoring of Public-Private Partnerships (PPP) and financial sector risks are needed to ensure debt sustainability over the medium term. Source: The 2022 WB/IMF DSA Note: The 2022 WB/IMF DSA follows the Guidance Note of the Guidance Note of Joint Bank-Fund Debt Sustainability Framework for Low-Income Countries, issued in February 2018. Cambodia’s Composite Indicator (CI) index, based on the April 2022 WEO update and the World Bank’s 2020 CPIA, indicates that the country’s debt-carrying capacity remains medium (2.968), the same as in the 2021 DSA (in which the CI index was 2.966). 1.7. Maintaining fiscal space balance fiscal tightening against economic growth. In the short term, Cambodia needs to unwind the and promoting progressivity countercyclical fiscal policy deployed during the COVID-19 pandemic to shore up its fiscal buffers Meeting Cambodia’s longer-term public while maintaining the important progress it has made spending needs while maintaining fiscal in establishing a social safety net. sustainability and restoring fiscal buffers will require difficult tradeoffs. Cambodia’s ability A key source of savings could be the elimination to respond effectively to the COVID-19 pandemic of ineffective tax incentives. Currently available relied on fiscal buffers established over years of data show that the country forfeits nearly a third strong economic growth. Those buffers are now in of its potential revenues from imports through tax decline, and concerted action is necessary to restore expenditures under the 2003 Investment Law (see them. However, Cambodia’s tax burden is high by Box 1.2). While free trade agreements will naturally regional standards, and the government needs to result in the decline of revenues from customs, the Cambodia Public Finance Review - From Spending More to Spending Better 27 Box 1.2. Tax expenditure analysis summary Customs, which is often responsible for collecting taxes on international trade, may not be the best way to generate revenue from an equity, efficiency, and growth perspective. However, Cambodia’s customs which, in addition to custom duties, collects value-added tax and excise on imports, contributed as much as 40 percent of total domestic revenue in 2022. And the country forfeited 28 percent of its revenue from imports to tax incentives in 2020 (Figure B.1.2.1). Between 2016 and 2020 the amount of revenue foregone on imports into Cambodia rose about 45 percent from around KHR 3.3 trillion to nearly KHR 4.7 trillion. Figure B.1.2.1. Tax relief on imports as a Figure B.1.2.2. Share of taxes relieved share of tax imposed at customs on imports for QIPs and SEZs COP SOP VOP CPP Special Taxes VAT Customs SPP VPP APT Total 40% 40% 35% 28% 30% 30% 23% 23% 22% 25% 22% 20% 20% 15% 10% 10% 5% 0% 0% 2016 2017 2018 2019 2020 2016 2017 2018 2019 2020 Note: COP: Customs other than petroleum, SOP: Note: QIPs: Qualified investment projects; SEZs: Special Special taxes other than petroleum, VOP: VAT on other Economic Zones. than petroleum, CPP: Customs on petroleum products, SPP: Special taxes on petroleum products, VPP: VAT on petroleum products, APT: Additional petroleum taxes Relief for special taxes (excise), value added taxes, and customs duties on imports is attributable mainly to Qualified Investment Project (QIPs) and Special Economic Zones (SEZs) beneficiaries (Figure B.1.2.2). These two categories of beneficiaries receive a disproportionate amount of customs tax relief and their share rose consistently between 2016 and 2020 (Table B.1.2.1). Table B.1.2.1. Share of relief granted through QIP or SEZ 2016 2017 2018 2019 2020 QIP 63% 63% 59% 59% 51% SEZ 29% 29% 32% 30% 34% Non-QIPSEZ 8% 8% 9% 11% 15% Total 100% 100% 100% 100% 100% Note: Non-QIPSEZ refers to relief granted to investments that are not qualified investment within special economic zones. The tax incentive targets and instruments under the 2021 Law on Investment (see Box 1.3) which are more generous than those under the 2003 Law on Investment should be revisited to prevent abuses. Given the new sub-decree to implement the 2021 Law on Investment has not been finalized, the decision making process remains to be seen. However, under the previous Investment Law (2003), the Council for the Development of Cambodia decided what was qualified and on what basis.2 Source: World Bank (2023) 1 Analyzing tax expenditures is constrained by the lack of the data from the General Department of Taxation. The data used for this analysis are those from the General Department of Customs and Excise alone. 2 For more detail, please see https://cdc.gov.kh/incentives-and-schemes/ 28 Cambodia Public Finance Review - From Spending More to Spending Better ineffective incentive can result in further decline of the pandemic on poor and vulnerable households. revenues without substantial gain in economic growth. However, its coverage remains relatively narrow and The country’s new Law on Investment (2021) (Box 1.3) formal social protection in Cambodia is incipient. is more generous than its predecessor, increasing The government should continue strengthening the risk that tax incentives erode the revenue base. the system, particularly in light of the country’s At the same time, the COVID-19 relief cash transfer vulnerability to climate risks that disproportionately program mitigated some of the worst impacts of affect poor and vulnerable households.21 Box 1.3. The Law on Investment The 2021 Law on Investment (LOI) attempts to: (i) create transparent and conducive investment climate to attract FDI, (ii) promote sustainable investments through incentives for investment schemes that are environmentally friendly, (iii) enhance productivity by attracting high value added FDI. While a quantitative assessment has not been undertaken, the LOI has the potential to erode revenue base. First, the LOI provides more generous incentives to investors which will erode the base for income taxes. These include a tax holiday from 3 to 9 years with an additional 6 years of reduced CIT rates, and the tax holiday can be extended beyond this period, if investments are qualified as expanded QIP; and accelerated depreciation at the rate of 40 percent of the value of tangible property purchased by QIPs. Profit-based tax incentives such as tax holidays are less effective for attracting and retaining capital investment than cost-based incentives such as investment tax credits or accelerated depreciation (Source: PCT, 2015). Retaining both options – a profit tax holiday and an accelerated depreciation – adds additional costs and would lead to revenue losses without additional economic or social benefits. Second, by expanding import duties and special tax exemptions on a wide range of imported construction materials and equipment, and VAT exemptions on locally sourced construction materials and equipment, and production material and inputs for export, foregone tax revenue is projected to increase substantially. Between 2016 and 2020, forgone revenue as a result of taxes exemption accounted for around half of the overall tax revenue collected or approximately 9 percent of GDP as compared to 6.2 percent in 2001.22 In 2020, QIP has taken around 51 percent of total tax exemption, followed by SEZ (34 percent) and non-SEZQIP (15 percent). Broader scope of tax exemptions by the LOI will lead to higher foregone tax revenue. Third, unclear definition of incentive schemes could potentially lead to a greater risk of interpretation to increase exemptions. For example, generous additional incentives in the form of a 150 percent deduction of expenditure incurred on some activities such as R&D and innovation, staff training and welfare, construction of workers’ accommodation and canteens, will add to the period of the tax holiday causing additional tax expenditures. Another case in point includes special incentives for priority industries such as the motor vehicle assembly industry. Clarifying the definition of incentives schemes is important to reduce the risk of violating the essence of incentives. Based on international practices, a sunset clause is recommended to set an end date to incentives provision. These sunset clauses mean that provisions are more likely to be periodically re-evaluated and can make ineffective relief easier to repeal. Clearly defining sunset provisions and ensuring transparency of its implementation to investors are vital for effective implementation of incentives provision.23 Source: World Bank (2023) 21 https://www.imf.org/en/Publications/CR/Issues/2022/12/15/Cambodia-2022-Article-IV-Consultation-Press-Release-Staff-Report- and-Statement-by-the-526993 22 IMF, Cambodia Rebuilding for a Challenging Future 2006 23 South Korea offers a best practice model applying a sunset clause to 60 percent of the tax expenditure items (157 out of 276 in 2019) Cambodia Public Finance Review - From Spending More to Spending Better 29 In the long term, Cambodia needs to make management and fiscal policy.24 Health excise tax is space for emerging spending needs through another potential area that can contribute to improved a more balanced, efficient, and equitable public health while generating revenues to help tax system. To this end, preparation for the new boost health spending needs (Box 1.4). Finally, the revenue mobilization strategy for 2024–2028 is government could introduce carbon taxes, charged underway. Under tax system reform activities, on the carbon content of fossil fuels, to raise revenue customs and tax administration diagnostic analyses and help mitigate Cambodia’s contribution to global (DIAMOND and TADAT) have been conducted, climate change in the context of broader tax reform while potential introduction of personal income tax, without earmarking. This can facilitate the transition excise tax reforms, and tax expenditure analysis are to a greener, low-carbon economy by investing being discussed. The medium-term fiscal framework in climate-smart infrastructure such as renewable (MTFF) and medium-term budgetary framework power generation and supporting research and (MTBF) are being prepared to ensure sound financial development in climate-smart technologies. Box 1.4. Global experience - Health excise taxes In the current global context, health taxes are also being explored to simultaneously support targeted revenue use for specific sectors, programs or populations and as a way to meet particular policy objectives. Health excise taxes, or excise taxes imposed on products that have a negative public health impact (e.g., tobacco, alcohol, sugar-sweetened beverages (SSBs)) can be used as a tool to target negative internalities and externalities and through changes in price, reduce consumption. Through this pathway, health excise taxes can generate health benefits, while still accruing revenue. For instance, tobacco generates on average 0.6 percent, and alcohol 0.3 percent of GDP in most countries with excise taxes in place on these products. SSB excise taxes generally generate very low revenues (less than 0.1 percent of GDP) due to less inelastic demand. Health taxes can provide additional revenue for health without earmarking by way of increasing allocations to the health sector—either by increasing the size of the government budget and keeping current sector allocations constant, or by active reprioritization of health through health tax reform and adaptive social assistance programs, spurring growth and increasing progressivity while protecting households against crisis. This includes complementary measures like targeted transfer mechanisms that can support vulnerable populations, as was used in Cambodia through the IDPoor program to deliver cash transfers to registered households. However, in part due to the special nature of health tax revenue being linked so closely to health-related internalities and externalities, earmarking of health tax revenue has often, although not exclusively been proposed as a mechanism to direct resources towards the health sector. Earmarking for the health sector has also been employed as an option globally and within the region (notably Philippines, Thailand, Vietnam). However, applying earmarks as a policy option needs to be done with a consideration of the current fiscal system in a country, as well as in alignment to public financial management rules and capacities and an understanding of the resulting tradeoffs in terms of budget flexibility. 24 World Bank Technical Assistance Mission - Cambodia Tax System Reform March 2-3, 2023. 30 Cambodia Public Finance Review - From Spending More to Spending Better 2 ENHANCING QUALITY OF PUBLIC SPENDING 2.1. Public expenditure trends 2011 to 27.9 percent in 2021. However, Cambodia’s expenditure boost was driven partly by across-the- in aggregate board salary increase for its civil servants. Between The improvements in the fiscal space allowed 2011 and 2020, the public sector wage bill as the government to boost total spending. a share of GDP almost doubled, peaking at 8.1 Expenditure increased from 20.2 percent of GDP in percent of GDP in 2020, up from 4.3 percent of GDP Figure 2.1. Government overall spending Figure 2.2. Government expenditure, by increased between 2011 and 2021, driven sector grouping (% of GDP) (2011-2021) by rise in wage bill and non-wage recurrent planned expenditure 30% General administration 27.9 Wages 30 Defense and security 30 Social sectors 26.8 Non-wage recurrent Economic sectors 24.0 25% Capital, domestically nanced 4.8% 4.9% Multisectors 25 21.8 25 Capital, externally nanced Total expenditure (RHS) 4.6% 20.2 21.0 20.9 20.6 21.7 6.3 19.5 20.3 20% 5.4% 6.4 4.7% 4.3% 5.6% 20 20 % of GDP % of GDP % of GDP 4.7% 3.2% 5.4 6.5% 7.1% 7.0% 6.3% 5.1% 2.2% 2.5% 8.4 6.1 5.4 15% 2.2% 15 5.6 15 7.8 8.7 7.3 10.5 2.1% 1.9% 2.1% 6.3 8.9 2.6% 2.1% 10.4% 8.9% 8.2% 6.8 7.1% 7.0% 10% 6.2% 6.6% 10 4.5 6.3 6.9 10 6.9% 6.4% 5.4 5.3 5.6 6.8% 6.8% 4.6 4.5 4.1 4.6 4.4 3.1 5% 5 3.7 3.9 4.1 5 7.6% 8.0% 7.4% 8.1% 7.2% 3.1 2.9 3.3 3.3 5.7% 6.1% 6.8% 4.3% 4.0% 5.0% 4.7 4.8 4.8 4.7 2.8 2.5 2.6 2.5 2.9 3.4 2.8 0% 0 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Sources: Budget Settlement Law (2011-2021); World Bank staff estimates Cambodia Public Finance Review - From Spending More to Spending Better 31 in 2011 (Figure 2.1). This represented a tripling of civil In general, capital investments relied heavily on servants’ minimum wage, compared to the level in development-partner (DP) financing. However, 2013, having reached the target of KHR 1 million a there has been a significant increase in the domestic month by 2018. Non-wage expenditure increased financing of capital investments starting in 2020 at a slower pace, from 6.8 percent of GDP in 2011 (Figure 2.4.a). The increase in domestically financed to 8.2 percent in 2020, and jumped to 10.4 percent investments is particularly pronounced in the social in 2021. The increase in the social sector spending, sectors, which reached 1.8 percent of GDP in which almost doubled from 4.6 percent of GDP in 2021. In contrast, DPs’ capital spending declined 2011 to 10.5 percent in 2021, is the main driver from 6.5 percent of GDP in 2011 to 4.8 percent in of expanded recurrent budget allocation (Figure 2021, due mainly to the decline in capital spending 2.2). The increase in the public sector wage bill put in the economic sector (Figure 2.4.b). Government Cambodia ahead of other ASEAN countries in terms net investment as a share of GDP fluctuated but of share of wage expenses as a percentage of total remained the highest in ASEAN between 2011 and government operating expenditure (Figure 2.3). 2019 (Figure 2.5). Figure 2.3. Compensation of government employee (% of total expenses) (2011-2019) Cambodia Indonesia Malaysia Thailand 60 Singapore Philippines Myanmar 50 40 30 20 10 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: WDI Note: Expense is cash payments for operating activities of the government in providing goods and services. It includes compensation of employees (such as wages and salaries), interest and subsidies, grants, social benefits, and other expenses such as rent and dividends. Figure 2.4. Trend of capital spending, by sector (% of GDP) (2011-2021) a. Capital spending, domestically financed, by sectors b. Capital spending, externally financed, by sectors grouping (% of GDP) (2011-2021) grouping (% of GDP) (2011-2021) 6 General administration 8 General administration Defense and security 5 7 Defense and security Social sectors Social sectors 6 Economic sectors 4 Economic sectors Multisectors 5 % of GDP % of GDP 3 Multisectors 4 3 2 2 1 1 0 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Sources: Budget Settlement Law (2011-2021); World Bank staff estimates 32 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.5. Net investment in government nonfinancial assets (% of GDP) (2011-2019) Cambodia Indonesia Malaysia Thailand 12 Singapore Philippines Myanmar 10 8 6 4 2 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 Sources: WDI, IMF Article IV and Cambodian authorities Note: Net investment in government nonfinancial assets includes fixed assets, inventories, valuables, and non-productive assets. 2.2. Strategic public development objectives. RS4 pursues four strategic goals: (i) maintaining a sustainable and resilient real expenditure prioritization economic growth rate of around 7 percent per year; (ii) creating more jobs; (iii) achieving the poverty The two documents that guide the government’s reduction target of below 10 percent; and (iv) further public expenditure are the 2019-23 Rectangular strengthening the capacity and governance of Strategy Phase IV (RS4) and the National public institutions to ensure an effective and efficient Strategic Development Plan (NSDP) 2019-2023. delivery of public services. NSDP provides the The RS4 is a blueprint that guides the government macro-economic framework, and outlines policies policies, plans and strategies in achieving sustainable and priority actions that the relevant ministries will development. It is prepared as a dynamic policy carry out, estimating the resources and expenditures document to address challenges as they emerge needed and framework for monitoring and evaluation. within a 5-year planning cycle during the period of 2019-23. Unlike other Rectangular Strategies, the Cambodia could benefit from improved 2019-23 RS4 targets human capital as the top priority linkages between NSDP planning and overall to underpin the authorities’ inclusive and sustainable government spending toward improvement of Figure 2.6. Deviation of actual spending against NSDP planning (2014-2021) 150% % NSDP budget planning 140% 130% 120% 110% 100% 100% 90% 80% 70% 60% 2014 2015 2016 2017 2018 2019 2020 2021 Capital, domestically nanced Capital, externally nanced Recurrent Sources: NSDP (2014-2018) and NSDP (2019-2023), Budget Settlement Law (2014-2021) and World Bank staff estimates Note: Over 100 percent means actual spending exceeds planned expenditure. Lower than 100 percent means under execution of planned expenditure. Cambodia Public Finance Review - From Spending More to Spending Better 33 expenditure prioritization in line with the national spending on social sectors increased from 4.3 development strategy. Recurrent spending peaked percent of GDP in 2011 to 10.4 percent in 2021, at 125 percent of NSDP in 2018 from 100 percent in driven by recurrent spending. Recurrent spending 2014; the spending then dropped to 71 percent of in social sectors doubled between 2011 and 2021, NSDP planned expenditures. Similarly, NSDP capital reaching 7.3 percent of GDP in 2021 from 3.5 planning did not reflect investment spending. NSDP percent. Among social sector ministries, recurrent planned investment expenditures over-executed spending in the MOEYS increased more than others, between 2014 and 2019 and under-executed in except that there was a significant increase in MOH’s 2020 and 2021. However, DP funded capital was spending in 2021 due to COVID-19 (Figure 2.7.a). In mostly under-executed for all years except in 2014 terms of capital spending, Cambodia mainly relied (Figure 2.6). While the NDSP is expected to be on DPs for financing until 2020. The government superseded by annual budgets as circumstances significantly increased investment spending to 1.8 evolve, significant financing gaps between the percent of GDP in 2021, from 0.2 percent of GDP in NSDP and the annual budgets that Cambodia has 2019. The domestic capital expenditure of MOEYS experienced so far indicate that planning needs peaked in 2018 reaching 0.4 of GDP, then declining further improvements. to 0.1 percent of GDP in 2021. Driven by Covid-19 spending, MOH capital spending spiked in 2021 2.2.1. Public expenditure and (0.4 percent of GDP). For the MOSVY, government outcomes: Social sectors investment spending jumped to 1.1 percent of GDP in 2021 from a very low base in 2011 (Figure 2.7.b). With an emphasis on human capital development DP-funded spending fluctuated significantly over the in Rectangular Strategy 4, the government same period (Figure 2.7.c). The significant fluctuation Figure 2.7. Expenditure trend for social sectors, by line ministry (% of GDP) (2011-2021) a. Reccurent spending for social sectors, by line b. Government capital spending for social sectors, ministry (% of GDP) (2011-2021) by line ministry (% of GDP) (2011-2021) 8 2.0 7 MOH 1.8 MOEYS 1.2 1.6 6 MOSVY 1.4 MOH % of GDP 5 MOLVT 1.2 % of GDP 2.6 1.2 MOEYS 0.9 0.9 0.9 4 0.8 1.0 MOSVY 0.8 0.7 0.8 3 0.5 0.6 2.9 0.8 MOLVT 2.5 2.7 2.7 1.6 1.7 1.9 2.0 2.3 0.6 2 1.5 3.3 0.4 1 1.7 0.2 1.3 1.4 1.4 1.2 1.3 1.3 1.3 1.4 1.4 0 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 c. DP capital spending for social sectors, by line ministry (% of GDP) (2011-2021) 1.2 MOH MOEYS 1.0 MOSVY MOLVT 0.8 % of GDP 0.6 0.4 0.2 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Sources: Budget Settlement Law (2011-2021); World Bank staff estimates 34 Cambodia Public Finance Review - From Spending More to Spending Better of available resources for capital investment MOEYS investment. From 2017 to 2021, there is an projects—for both government and DP funded—is a increasing importance of government investment major concern for capital budget planning. and fluctuating DP-funded capital investment in the education sector. MOEYS capital investment Trends in the education sector peaked in 2018 at 0.4 percent of GDP, later declined to 0.1 percent in 2021. Overall, DP-funded capital The government increased education investment and MOEYS investment fluctuated over expenditure between 2011 and 2021, driven the last decade (Figure 2.9). mainly by a rise in salary for the large number of education staff. Starting salary for upper Budget outturns for capital expenditures— secondary teachers doubled between 2014 and both domestically and externally financed— 2020 due to the increase in both base salary and significantly fluctuated (Figure 2.10). This functional allowance. In addition, the total number of suggests weaknesses in the planning and execution educational staff increased by 6 percent during the of capital expenditures. same period, from 107,395 in 2014 to 114,170 in 2020. Wage expenditure as a share of total MOEYS Compared with regional peers, Cambodia’s expenditure reached nearly 80 percent in 2021. At gross national income per capita is the lowest. the same time, the MOEYS capital expenditure has Cambodia’s education spending of 3.1 percent dropped significantly, declining from 27 percent in of GDP is the second lowest in the region and 2011 to only 17 percent of total MOEYS expenditure slightly lower than regional average of 3.4 in 2021 (Figure 2.8). It is necessary to ensure that percent of GDP 25 (Figure 2.11). adequate budget is allocated for recurrent non- salary and capital expenditures to support education Despite the persistent effort of the MOEYS development and reforms. to improve student’s learning (Box 2.1), the quality of spending remains low as reflected Education capital expenditures relied mainly on in low level of students’ learning outcomes for development partners (DP). Between 2011 and Grade 6 and Grade 8. Between 2013 and 2021, an 2016, DP-funded capital investment dominated total average student in Grade 6 could answer less than Figure 2.8. MOEYS spending, by category Figure 2.9. MOEYS spending, by category (% of total MOEYS domestic spending) (% of GDP) (2011-2021) (2011-2021) 100% 0% 0% 0% 1% 2% 2% 4% Capital, externally nanced 4% 7% 14% 7% 7% 5% 90% Capital, domestically nanced 3.3% 3.3% 27% 27% 26% 21% 19% 19% 17% 19% 15% 17% 4% Recurrent 3.4% 4% 80% 17% 2.8% 3.0% Total (RHS) 70% 3% 2.6% 3% 60% 2.3% 2.2% 3% 3% 50% 2.0% 1.8% 40% 1.8% 73% 73% 74% 78% 79% 79% 75% 69% 74% 77% 78% 2% 2% 30% 20% 2% 2% 10% 1% 1% 0% 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 1% 1% Wages 0% 0% 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Non-wage recurrent Capital,domestically nanced Sources: Budget Settlement Law (2011-2021); World Bank staff estimates Note: RHS = Right-hand scale 25 Cambodia’s education spending as a percent of GDP is based on current GDP value and is expected to be lower with the upcoming GDP rebasing. Cambodia Public Finance Review - From Spending More to Spending Better 35 Figure 2.10. Planned capital spending and Figure 2.11. Government expenditures on budget outturn education percent of GDP vs. GNI per capita (2021) 250% 8,000 4.1% 5% 3.9% 3.7% 4% 200% 7,000 3.5% 6,000 3.1% 3.1% 4% 150% 3% 5,000 2.3% US$/year 3% % of GDP 100% 4,000 2% 3,000 2% 50% 2,000 1% 1,000 1% 0% 2017 2018 2019 2020 2021 0 0% Cambodia Lao PDR Vietnam Indonesia Philippines Thailand Malaysia Recurrent Capital, domestically nanced GNI per capita Capital, externally nanced Government expenditure on education (% of GDP) (RHS) Sources: Budget planning figures sourced from Budget Law Source: WDI (extracted 2023) and actual spending sourced from Budget Settlement Law Note: RHS = Right-hand scale (2017-2021) and World Bank staff estimates Note: Budget outturns over 100 percent of the approved budget reflect actual spending exceeding the planned budget (overspent); under 100 percent means actual spending was below the planned budget (underspent). 52 percent of Khmer questions correctly. Student of Grade 8 performed below basic for math, while learning outcomes for mathematics were even students with basic, proficient and advanced level poorer. On average, Grade 6 students answered for math declined. In physics, student performance less than half the questions correctly—43 percent in deteriorated during the same period. Students with 2013, 48 percent in 2016, and 38 percent in 2021 below basic level increased around one third, while (Figure 2.12). Grade 6 student learning outcomes students with basic, proficient and advanced level showed slight improvement between 2013 and 2016 declined. There was an improvement in Khmer in both Khmer and Math. However, the deterioration reading for Grade 8 students. The proportion of of the 2021 test result was due to the negative impact students with basic and advanced level for Khmer of the COVID-19 pandemic. Similarly, 36 percent, reading increased. This was offset by the decline 19 percent, and 52 percent of students in Grade 8 in the proportion of students with below basic and performed below basic proficiency level for math, proficient level (Figure 2.13). Khmer reading, and physics respectively in 2017. Between 2017 and 2022, higher numbers of students The Covid-19 pandemic has had impact on Figure 2.12. Learning outcomes for Grade 6 students deteriorated between 2013 and 2021 Percentage of questions correctly answered 60% 52% 48% 46% 43% 47% 40% 38% 20% 0% 2013 2016 2021 Khmer Math Source: MOEYS 36 Cambodia Public Finance Review - From Spending More to Spending Better Box 2.1. MOEYS’s effort to improve student learning Improving early age reading and math: MOEYS introduced a new approach to improve early age reading and math for pupils. A pilot program was successfully conducted in Kampong Thom province and Siem Reap province. The approach is being rolled out across the country and in teacher training centers. Response to Covid-19: In response to Covid-19 pandemic, schools were fully or partially closed between February 2020 and February 2022. Since students in rural areas lack access to digital instruments, MOEYS has introduced digital education content such as video, self-study or virtual lessons and other digital materials broadcasted on TV and other online platforms and provided those digital materials to schools. Learning curriculum and textbooks: MOEYS has successfully improved national learning curriculum and it is working on improving and updating textbooks. Textbooks were distributed to schools and school management mechanism including school information system was strengthened to allow for Provincial Office of Education (POE) to effectively monitor and manage text books, including timely distribution of text books to schools in need. Teaching approaches: new teaching approaches such as self-regulated learning, project-based learning, research-based learning is strongly considered by integrating the teaching approaches at teacher training centers, new generation schools, and other schools across the country. Source: MOEYS Figure 2.13. Percentage of Grade 8 students proficiency in Khmer reading and Mathematics (2014-2022) a. Proficiency of Grade 8 students in Math b. Proficiency of Grade 8 students in Khmer reading 50% 50% 45% 45% 40% 40% 2014 2017 2022 35% 35% % of student 2014 2017 2022 30% % of student 30% 25% 25% 20% 20% 15% 15% 10% 10% 5% 5% 0% 0% Below basic Basic Pro cient Advanced Below basic Basic Pro cient Advanced c. Proficiency of Grade 8 students in Physics 80% 70% 60% % of student 50% 2014 2017 2022 40% 30% 20% 10% 0% Below basic Basic Pro cient Advanced Source: MOEYS Cambodia Public Finance Review - From Spending More to Spending Better 37 student learning loss and learning poverty More importantly, there is a gap between rate in Cambodia. Learning loss means quality urban and rural areas in education outcomes of education deteriorated and children learn less, (Figure 2.15). The net enrollment rate for primary, reflected by learning adjusted years of schooling lower secondary and upper secondary increased (LAYS). The simulation model of Azevedo et al. (2022) respectively, from 81.1 percent to 90.1 percent, from revealed that the Covid-19 has resulted in substantial 31.5 percent to 47.5 percent, and from 17.9 percent reductions in LAYS, an increase in learning poverty to 30.9 percent between 2009 to 2019. The literacy in Cambodia. The estimated average LAYS loss was rate for the population aged 15 and above increased between 1.46 years and 1.76 years, higher than EAP from 73.9 percent to 81.9 percent for the same LMIC, which range from 0.86 to 1.03 years. Learning period. However, there is a big difference between poverty rate (percentage of 10-year-olds who can urban and rural areas in terms of net enrollment and not read and understand a short passage of age- literacy rate. Although Cambodia has reduced the appropriate material) is expected to reach 99.7 percentage of people with no schooling, completion percent as compared to EAP LMIC at 79 percent rates for primary and secondary remained flat. (Figure 2.14). Figure 2.14. Loss in learning adjusted years of schooling and increase in learning poverty a. Loss in LAYS under different scenarios b. Learning poverty in optimistic scenarios 2.18 99.7 100.0 100.0 100 2.3 Learning poverty rate (%) 1.76 2.10 97.7 79.0 80 90.0 89.5 1.8 1.91 Years lost 1.65 1.30 60 70.9 1.3 1.46 1.03 1.16 0.96 40 0.8 0.98 0.86 20 0.3 Cambodia Lao PDR Myanmar EAP LMIC 0 Average Cambodia Lao PDR Myanmar EAP LMIC Average Intermediate Optimistic Pessimistic Optimistic Pre-COVID learning poverty Source: Simulation model from Azevedo et al. (2022) Note: Assumptions for mitigation effectiveness are based on the assumptions used by Azevedo et al. (2022) for Lower Middle-Income Countries (LMIC). Three scenarios include: (i) Intermediate scenario means 85 percent of the schools are closed and mitigation effectiveness is 7 percent; (ii) Optimistic scenario means 75 percent of the schools are closed and mitigation effectiveness is 14 percent, and (iii) Pessimistic scenario means 100 percent of schools are closed and mitigation effectiveness is 7 percent. Figure 2.15. Net enrollment rate, by urban and rural areas (2009-2019/20) 100% 2009 2014 2019-20 80% 60% 40% 20% 0% Primary Lower Upper Primary Lower Upper Primary Lower Upper Primary Lower Upper secondary secondary secondary secondary secondary secondary secondary secondary Total Phnom Penh Other urban Rural Cambodia Areas Sources: CSES (2014) and CSES (2019/20) 38 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.16. Cambodia improved literacy rate, but still lags behind its peers in ASEAN 100% 94.9% 95.7% 94.0% 95% 89.0% 90% 87.0% 85% 82.0% 80% 75% Cambodia (2020) Lao PDR (2021) Myanmar (2019) Thailand (2021) Malaysia (2019) Vietnam (2019) Sources: Cambodia extracted from CSES (2019/20) and other countries sourced from World Development Indicators (WDI) (extracted in March 2023) In comparison to its ASEAN peers Cambodia affects class sizes. At the national level, the total lags behind in education outcomes, such as number of teachers is generally inadequate. Class literacy rates (Figure 2.16). Completion rates are size remains relatively larger than its ASEAN peers. also lower than most ASEAN countries, suggesting Cambodia had the highest student-teacher ratio for that more work and investment need to be done to primary school from 2013 to 2018. For secondary retain students in secondary education. In addition, school, the student-teacher ratio remains high, but drop-out rates at primary school remained the just lower than Thailand and Myanmar for the same highest in the ASEAN region. While those with no period. The larger class size in Cambodia is mainly schooling dropped from 28 percent in 2009 to 18 attributable to the inefficient distribution of teachers. percent for the 2019-20 school year, students in At national level, based on the 2018 MOEYS norm Cambodia performed below the average score for student and teacher proportion, Cambodia ran for other participating ASEAN countries in three an overall shortage of teachers of 20,356 in 2021. domains: reading, mathematics, and science. The Teachers were surplus in lower secondary and 2022 Program for International Student Assessment shortage at upper secondary level for the same year (PISA) showed slight improvements in performance (Table 2.1). scores for Cambodian students in reading, mathematics and science. Average performance The shortage of teachers has a greater effect scores for reading increased to 329 in 2022 from in rural areas, as the distribution of teachers 321 in 2017, while the scores for mathematics appears skewed toward urban areas. In 2021, reached 336, an increase from 325 and the score urban areas have a low level of teacher shortage for science increased from 330 to 347 for the same (490) while rural areas have a net teacher shortage period. However, the scores for Cambodian students of 19,867 (Table 2.1). This is based on calculation are lower than its peer countries such as Philippines of actual demand for teachers by dividing total and Indonesia in reading, mathematics and science. number of student enrollment by students per class Around 8 percent of students in Cambodia reached as required by 2018 MOEYS norms. Pre-schools level 2 or higher in reading (student comprehends the and primary schools in rural areas were particularly main idea in a text and can reflect it), 12 percent in starved of teachers, reaching a net teacher shortage mathematics and 10 percent in science, compared of 4,833 and 13,403 respectively. At provincial level, to 74 percent in reading, 69 percent in mathematics the teacher distribution is tilted toward Phnom Penh and 76 percent in science in OECD countries. 26 where the surplus of teachers was concentrated between 2018-2021 (Figure 2.17). Most provinces The poor deployment of teachers negatively had shortages of teachers for the same period. 26 PISA (2018-2022) Cambodia Public Finance Review - From Spending More to Spending Better 39 While public sector pay is higher than private in delivering education services. A number of reasons sector, the education sector continues to face help explain the surplus and shortage of teachers: shortage of teachers in certain subjects. Despite MOEYS’s effort to address teacher surplus and • The Education Management Information System shortages (Box 2.2), Cambodia still has a teacher (EMIS) was introduced by MOEYS to provide deployment challenge. Shortages and surpluses of information on student enrollment. However, teachers appeared to be a key challenge for schools teacher surplus and shortage at school level was Table 2.1. Schools in rural areas have the highest level of teacher shortage in 2021 Lower Upper Pre-school Primary Total secondary secondary National Current teacher 5,621 45,148 14,704 29,245 94,718 Actual demand for teacher 10,454 58,551 13,928 32,141 115,074 Net teacher shortage/surplus -4,833 -13,403 776 -2,896 -20,356 Urban Current teacher 1,021 8,850 2,635 10,119 22,625 Actual demand for teacher 1,770 9,999 1,816 9,530 23,115 Net teacher shortage/surplus -749 -1,149 819 589 -490 Rural Current teacher 4,600 36,298 12,069 19,126 72,093 Actual demand for teacher 8,684 48,552 12,111 22,612 91,960 Net teacher shortage/surplus -4,084 -12,254 -42 -3,486 -19,867 Source: World Bank’s staff calculation based on MOEYS’ data in 2021 Note: Needs for teachers per the 2018 MOEYS norms: Pre-school: 1.2 teacher per class and 25 pupils per class. Primary: 1.15 teachers per class; for 1-3rd grade, 35 to 45 pupils per class and for 4-6th grade, 40 to 50 pupils per class. Lower secondary: 2.062 teachers per class and 40 to 50 pupils per class. Upper secondary: for 10th grade, 2.062 teacher per class and 40 to 50 pupils per class; for 11-12th grade, 2.187 teachers per class and 40 to 50 pupils per class. Figure 2.17. Net school teacher shortage, by province (2018-2021) 2000 Shortage/surplus of teacher 1000 2018 2019 2020 2021 0 -1000 -2000 -3000 -4000 -5000 Banteay Meanchey Battambang Kampong Cham Kampong Chhnang Kampong Speu Kampong Thom Kampot Kandal Kep Koh Kong Kratie Mondul Kiri Oddar Meanchey Pailin Phnom Penh Preah Sihanouk Preah Vihear Prey Veng Pursat Ratanak Kiri Siem Reap Stung Treng Svay Rieng Takeo Tbong Khmum Source: World Bank’s staff calculation based on MOEYS’ data (2018-2021) 40 Cambodia Public Finance Review - From Spending More to Spending Better Box 2.2. MOEYS efforts to address the issue of surplus and shortage of teacher MOEYS has introduced quality upgrading program for upgrading teachers from lower secondary to upper secondary level, within which 2,684 teachers were trained. This has contributed to addressing teacher shortage by subject at upper secondary level. Principles for distribution of new recruited teachers: • Planning for distribution of newly recruited teachers is prepared based on teacher shortage status by subjects at school level. • POE is required to prepare list of schools that have demand for teachers by subject through projecting teacher demand over the next two to four years based on teacher training program. • Candidates applying to be newly recruited teachers are required to choose schools that are in need for the new teachers and singed teaching contract upon deployment. Principles for addressing surplus teachers: • MOEYS issued an instruction # 28 dated May 5, 2014 not allowing for transfer of teacher to schools with teacher surplus. • MOEYS continues to implement Sub-decree # 102 dated 2002 which sets out one time allowance for redeployment: redeploy within commune, approx. US$ 74; redeploy between commune, approx. US$120; redeployment between districts, approx. US$200; redeployment from low land provinces to low land provinces, approx. US$ 250; and redeployment from low land from province to isolated province, approx. US$ 365. MOEYS introduced the following measures: • Conduct study at schools with teacher surplus, starting from the schools in Phnom Penh. • Set out conditions and criteria for teachers to be redeployed from teacher-surplus schools to teacher- shortage ones. • Set out principled incentives for redeployed teachers. • Review and reclassify hardship schools (type 1 and type 2) and request for principal to provide further incentives to teachers working at those schools. • Prepare a mechanism in redeploying surplus teachers and enforce the current teaching norm. • Set out conditions for redeploying teachers who teach less than 9 hours per week to teacher- shortage schools. • Review and increase teaching hour norm. Strengthen effectiveness of the current data management systems including Duraseksa website and Education Management Platform (EMP) so that MOEYS can manage data in a thorough and concrete way for the purpose of analysis and decision making. Teacher training is prioritized measures taken by MOEYS. The achievement includes: • Conduct review of teacher training programs at all level. • Prepare mechanism to attract high performance students to select teaching careers. • Convert teacher training centers in Battambang province and Kampong Cham province to be teacher training institutes which implement 12+4 years of training. • Implement full Bachelor + 2 program at National Institute of Education (NIE). • Start implementing teacher license through payment-based teacher training to respond to the future planning which delegate autonomy to school principal to recruit teachers based on school demand as well as avoiding employment of public teachers in the private schools. Implement Sub-decree on the transfer of education, youth and sport functions to district, municipality and khan administration across the countries. Review and monitoring on the implementation of administration, staff, finance and learning and teaching will be conducted. Source: MOEYS Cambodia Public Finance Review - From Spending More to Spending Better 41 not articulated in EMIS reports. Lack of credible is a lack of enforcement. For instance, teachers and transparent information on teacher surplus at surplus upper secondary schools taught and shortage at school level tended to exacerbate approximately 5 hours per week as opposed to the issue by encouraging further teacher transfer, the required 16 hours per week29 and engaged in hence increasing uneven distribution of teachers teaching at private schools or family businesses between rural and urban areas. in their spare time. Furthermore, since there • The shortage of teachers in the rural areas is no teacher performance assessment and tended to increase for a number of reasons. management in place, the salary for those First, there is a limited incentives to move to rural surplus teachers continued to rise as a part of areas, and insufficient housing for those who do. salary across the board increase. Second, there is a need for teachers to move • While teacher recruitment and deployment are closer to their spouses who live outside of their conducted at national level, teacher shortage duty station. Third, informal payment facilitated remains a key concern for schools to deliver the transfer process. Teacher recruitment was educational services. Within schools where there done at the central level based on demand of is a shortage of teachers, teachers are tasked to teacher by region. In general, only candidates teach long additional hours and teach subjects who are resident or high school graduate (Bac beyond their specialized areas. Schools have II level) from a particular province are allowed to limited resources to address the challenge. Aside apply for teaching in that province. New teachers from the School Operation Fund (SOF)—an were normally allocated to their own resident average annual SOF per school in 2023 was as province where they graduated high school low as KHR 10.8 million (equivalent to US$2,677 (Bac II level). Despite the requirement set out by per year)—resources for schools to address the MOEYS for teachers newly deployed to serve for teacher shortage are missing. a period specified in a contract, teachers were Trends in the health sector then transferred from rural to urban areas within the same province where there are opportunities Public health (Ministry of Health) expenditure for additional income earnings through private constitutes a small portion of total health tutoring.27 expenditure in Cambodia. A larger chunk of health • Redeployment faced tough challenges. Teachers expenditure comes from patients’ out-of-pocket in urban areas tended not to be redeployed spending. Prior to the COVID-19 outbreak, Ministry since that means moving away from their of Health spending slightly fluctuated, reaching 7.9 families and loss of opportunity for additional percent of government spending in 2021 from 8.4 income from additional teaching in urban private percent in 2017. (Figure 2.18). schools. Although the government introduced redeployment policy and provide some monetary Capital expenditures at MOH are mainly financed incentives for voluntary redeployment to schools by externally funded capital allocation. Between that have a shortage of teachers—especially to 2017 and 2021, MOH capital investment as a share the rural areas—the redeployment allowance was of total spending accounted for less than 1 percent, limited.28 except in 2021 when planned government capital • In 2018, the MOEYS introduced norm for teaching expenditures reached 18 percent (Figure 2.19), as a hours at school level. However, in practice, there part of broader COVID-19 response. 27 MOEYS, Teacher management and redeployment: issues and practical way forward (December 2018) 28 Sub-decree # 102 dated 2002 set out one-time allowance for redeployment: redeploy within commune, approx. US$74; redeploy between commune, approx. US$120; redeployment between districts, approx. US$200; redeployment from low land provinces to low land provinces, approx. US$250; and redeployment from low land from province to isolated province, approx. US$365. 29 MOEYS’ teacher norm (2018) 42 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.18. MOH expenditure (2017-2022) Figure 2.19. MOH spending, by category (% of total) (2017-2022) 9% 8.4% 100% 7.5% 7.9% 90% 15% 8% 7.3% 7.4% 28% 22% 22% 28% 7% 80% 70% 6% 60% 5% 50% 4% 40% 84% 78% 77% 72% 2.2% 30% 54% 3% 2.0% 1.8% 1.6% 1.7% 20% 2% 10% 1% 0% 0% 2017 2018 2019 2020 2021 2017 2018 2019 2020 2021 Capital, externally nanced MOH expenditure (% of total government expenditure) Capital, domestically nanced MOH expenditure (% of GDP) Recurent Sources: Budget Settlement Law (2017-2021); World Bank staff estimates Note: Recurrent expenditure includes expenditures other than infrastructure such as wages, drugs, medical supplies, and other operating expenses; it however, excludes Covid-19 spending in 2020 and 2021. Capital expenditures budget is used to acquire infrastructure fixed assets such as hospitals or health center buildings. Increase in the MOH recurrent spending in supplies (Figure 2.21). Between 2017 and 2021, nominal terms between 2017 and 2021 was the spending for nominal social benefits expenses driven mainly by social benefits expense and (Chapter 62) excluding Covid-19 spending slightly wages 30 (Figure 2.20). Social benefits expenses increased by 12 percent. As a share of GDP, the (Chapter 62) 31 support health service delivery and spending on social benefit expense increased include in-patient and operational expenses at from 0.5 percent in 2017 to 0.6 percent in 2020 the health facilities, drug purchases, and medical and slightly declined to 0.5 percent in 2021 due to Figure 2.20. MOH spending, by economic Figure 2.21. MOH spending on social benefits classification (KHR, million) (2017-2021) expenses, by economic classification (KHR, million) (2017-2021) 800,000 800,000 700,000 700,000 KHR, million 600,000 600,000 500,000 KHR, million 500,000 400,000 400,000 300,000 300,000 200,000 200,000 100,000 100,000 0 2017 2018 2019 2020 2021 0 Others Chapter 60: Chapter 61: Chapter 64: Chapter 62: Chapter 65: Chapter 63: Chapter 21: Purchases Services Payroll Social Susidy Taxes Investment 62028: Other social bene ts expenses expenses bene ts expenses and tarrifs expenditure 62023: Food and materials expenses 62022: Drugs 2017 2018 2019 2020 2021 62021: In-patient and operational expenses for health center Source: MEF Source: MEF Note: MOH spending excludes Covid-19 expenditure Note: MOH spending on social benefit expenses excludes Covid-19 expenditure 30 Social benefits expenses (Chapter 62) include in-patient travel expenses at health center, drug purchases expenses, and medical supplies 31 Chapter refers to group of account and sub-account of public revenues and expenditures for respective economic classification. Under Cambodia’s public expenditures, expenditure chapter by economic classification includes: Chapter 60=Purchases, Chapter 61=Services, Chapter 62=Social benefits expenses, Chapter 63=Taxes and tariffs, Chapter 64=Payroll expenses, Chapter 65=Subsidy expenses, and Chapter 21=Investment expenditure. Cambodia Public Finance Review - From Spending More to Spending Better 43 Covid-19. This was driven by a rise in subsidies to on preventive care is measured as a share of informal sector Health Equity Fund-HEF32 (Chapter government health expenditure, Cambodia was 62028) and drug expenditures (Chapter 62022). the lowest (Figure 2.22). The share of government Domestically financed capital spending (Chapter 21) spending on preventative care is the lowest among was modest in 2017 (KHR 1,199 million or 0.001 comparator countries. There should be a transition percent of GDP) with a modest increase in 2020 towards higher prioritization for preventive care, (KHR 84,738 million or 0.08 percent of GDP) and a including introducing policies to increase targeted substantial increase in 2021 (KHR 496,839 million or prevention measures where health spending is high 0.45 percent of GDP) (Figure 2.20). and institutionalizing incentives for people to seek preventive services. Spending on health system administration remains high compared to ASEAN countries, It is important to note that overall healthcare while spending on health service delivery expenditure in Cambodia is higher than all of (particularly preventive care) remains very low its neighbors despite the low level of public as a total share of public expenditures. In 2019, spending. On average, healthcare financing from Cambodia’s spending on health system administration all sources in Cambodia has been greater than 5.7 as a share of government health expenditures (35 percent of its GDP from 2009 to 2019, consistently percent), was twice as high as in Myanmar, more than four times higher than in Vietnam, and nearly higher than any of its ASEAN neighbors (Figure 2.23). three times higher than in Thailand and Malaysia. This level of spending persists despite relatively low On the other hand, as a share of government levels of government spending on health in both health expenditures, Cambodia’s public spending per capita terms (Figure 2.24) and as a share of the on curative care (62 percent) is only second lowest government budget (Figure 2.25) – both lower than to Lao PDR, and lower than Myanmar, Vietnam, nearly all in the region. This means the remaining Thailand, and Malaysia (66 percent, 86 percent, 75 is financed by external (DP) financing and out-of- percent, 76 percent respectively). When expenditure pocket spending. Figure 2.22. Cambodia’s health spending as compared to its peer countries in ASEAN, by health care functions, (% of government health expenditure) (2019) 100% 1% 2% 1% 4% 7% 3% 7% 10% 11% 90% 7% 15% Government expenditure on other health % of government expenditure 80% care 70% 62% Government expenditure on preventive 60% 51% care 50% 66% 84% 75% 76% 40% Government expenditure on curative care 30% 20% 35% 35% Government expenditure on governance, 10% 17% health system & nancing administration 8% 13% 12% 0% Cambodia Lao PDR Myanmar Vietnam Thailand Malaysia Sources: WHO-Global Health Expenditure (2022) and World Bank staff estimates Note: MOH=Ministry of Health Government health expenditure from 2016 to 2019 (graph a) are PPP-adjusted nominal values. Other healthcare includes rehabilitative care, long-term healthcare, ancillary services, medical goods, and other healthcare services 32 HEF is social health protection fund for the poor. HEF purchases healthcare for the poor from public health facilities. The detail of HEF is discussed in Chapter 2. 44 Cambodia Public Finance Review - From Spending More to Spending Better The volume of financing for health in Cambodia universal health coverage (UHC). Total health is consistent with what other countries are expenditure per capita increased in real terms from spending at similar income levels, but the about US$35 per annum in 2000 to about US$115 composition is skewed towards private out- in 2020 (Figure 2.26.a). Countries transitioning from of-pocket (OOP) expenditures. OOP financing low- to middle-income status typically also achieve is neither an efficient nor equitable way to achieve a health financing transition whereby the volume Figure 2.23. Health expenditure (% of GDP) 8 Brunei Darussalam 7 Cambodia EAP (excl. high income) 6 Indonesia Lao PDR 5 Low & middle income 4 Malaysia Myanmar 3 Philippines 2 Singapore Thailand 1 Vietnam 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: WDI Figure 2.24. General government health expenditure per capita, PPP (current intl. US$) 700 Singapore and Brunei D. off the chart Cambodia 600 EAP (excl. high income) 500 Indonesia Lao PDR 400 Low & middle income 300 Malaysia Myanmar 200 Philippines 100 Thailand 0 Vietnam 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: WDI Figure 2.25. Domestic health expenditure (% of general government expenditure) 16 Brunei Darussalam 14 Cambodia 12 EAP (excl. high income) 10 Indonesia 8 Lao PDR Malaysia 6 Myanmar 4 Philippines 2 Singapore 0 Thailand 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: WDI Cambodia Public Finance Review - From Spending More to Spending Better 45 Figure 2.26. Cambodia health financing relied mainly on OOP over the last two decades and OOP expenditure tends to increase further a. Volume of health financing per capita per year b. Over-reliance on OOP 120 Low income Lower middie income Upper middie income High income 110 80 AFG 100 MGA Share of OOP over CHE (%()) BOD MMR 70 CMR 90 Constant 2020 US$ TJK 80 60 Cambodia SLE GTM 70 50 CAF GIN HTI AK MRT 60 LBR NER SEN 50 40 COD UGA VNM BFA CIV 40 MDG ETH IDN 30 MLI GHA 30 COG KEN MWI TZA 20 20 10 MOZ 10 RWA ZWE 0 ZMB 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 0 General taxation SHI External (via government) External 250 500 1,000 2,500 10,000 35,000 100,000 OOP VHI Other (non-government) GNI per capita, US$ Sources: Author’s calculation based on various data sources Note: SHI=Social Health Insurance, VHI=Voluntary Health Insurance of financing invested in health increases and its perception of public health quality among citizens. Low composition transitions away from external and OOP utilization of public health services means that public dependence to predominant pre-paid mechanisms, expenditure has not achieved its fullest potential. including increased public health financing or some form of health insurance mechanisms (e.g., social Despite low utilization of public health facilities health insurance). In Cambodia however, while the and preference for the private sector, which share of external financing decreased since 2015 contributes to high out of pocket expenditure, (Figure 2.26.a), OOP expenditure remain the main Cambodia has significantly improved health financing channel for the sector, and the share of outcomes over the last two decades. Cambodia OOP financing (60 percent in 2020) is well above has seen substantial increases in immunization what is observed on average for countries at similar coverage and declines in the unvaccinated (Figure income levels (about 30 percent) (Figure 2.26.b). 2.27). It has also seen large reductions in childhood mortality, with under-5 mortality dropping from 124 Utilization of health care services provided by per thousand births in 2000 to 16 in the 2021-22 public health facilities is low. In 2021, more than health survey (Figure 2.28). three-fourths of sampled household members (83.9 percent of households) sought care for illness and Despite the improvement in health outcomes, injury from a private healthcare provider in the last Cambodia lags behind most peers in ASEAN in 30 days, while only 15.3 percent sought care from a life expectancy (Figure 2.29.a). Over the last two public healthcare provider. 33 Low utilization of public decades, life expectancy increased from 67.7 years health facilities is caused by distance of travels to in 2010 to peak at 70.4 years in 2019 and slightly the nearest health center, omnipresence of private decreased to 70.4 years and 69.6 year respectively service providers, which allows quick access, 34 and in 2020 and 2021.35 However, Cambodia’s life 33 Draft report: Analysis of Financial Risk Protection in Cambodia’s Health System using CSES (Cambodia Socioeconomic Survey) Data: 2009–2021. GIZ (2023, forthcoming). 34 In 2022 there was a total of 1,408 public sector health facilities, which accounts for less than 10 percent of the total health service providers across Cambodia with the majority of health care providers being private sector. Between 2017 and 2022 the number of Health Centers (primary care level) increased from 1,190 to 1,288. 35 Cambodia’s 2019 Population Census reported that life expectancy at birth was 75.5 which was approximately 5 years higher than Worldwide Development Indicators (WDI). WDI used the data from United Nations Population Division’s World Population Prospects which interpolated data from 5-year period. Therefore, WDI data may not reflect observed data in specific survey in Cambodia. However, for comparative purpose, WDI data is used. 46 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.27. Trends in immunization coverage Figure 2.28. Trend of childhood mortality 90% 140 % of children agen 12-23 months 80% 79% 76% 73% 120 124 deaths per 1,000 live births 70% 67% Under-5 mortality 100 Infant mortality 60% 95 No vaccinations 83 Neonatal mortality 50% 40% 80 40% Fully vacinated (Basic antigens) 66 60 54 30% 45 22% 40 37 20% 35 28 27 28 16 7% 20 18 10% 4% 2% 3% 12 0% 0 8 2000 2005 2010 2014 2021-22 2000 2005 2010 2014 2021-22 Sources: Cambodia Demographic and Health Survey (CDHS) (2021/22) Key Indicators Report (2022) Figure 2.29. Life expectancy as compared to ASEAN peer countries a. Life expectancy vs. regional peers (2010-2021) b. Child mortality rate and life expectancy in Cambodia (2021) 85 45 41.8 42.5 40 80 deaths per 1,000 live births 35 Mortality rate, infant 33.7 34.2 Mortality rate, under-5 75 30 22.2 25.7 24.8 25 70 20.6 20.5 21.3 20 18.9 11.5 16.4 65 15 7.6 8.3 10 9.6 2.1 6.5 7.1 60 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 5 1.7 Cambodia Indonesia Philippines Vietnam 0 s lip a ai a B nd V iD M ia In am R o r M ore Thailand Malaysia La a am e hi i Th ysi D P nes d nm C pin la ne bo tn P p a Lao PDR Singapore ga ie ya al ru do in Myanmar Brunei D S Source: WDI (2023) expectancy was only higher than Lao PDR, Myanmar, under-5 mortality between the highest and lowest and Philippines but remained lower than the rest income quintiles declined as well, during the same ASEAN peer countries. Cambodia had improved period. However, the gap between the poorest child mortality rate over the last two decades. and richest in infant mortality and under-5 mortality However, Cambodia’s infant and under-five mortality persists. In 2021, the poorest infant mortality and rate was around three times higher than in Thailand under-5 mortality were respectively 16 deaths per and Malaysia (Figure 2.29.b). 1,000 live births and 21 deaths per 1,000 live births higher than those of the richest (Figure 2.30). The Cambodia has disparities in health outcomes evaluation of targeting quality is hindered by data between rural and urban areas and among availability constraints, which prevent the inclusion wealth levels that are likely to remain as a of more up-to-date health outcome data, as well as challenge. Between 2005 and 2021, infant mortality data at the district and commune levels to evaluate and under-5 mortality for urban and rural places the equity of access within provinces. However, significantly declined; however, the urban and rural correlations across 25 provinces between the size gap remained. Infant mortality and under-5 mortality of the recurrent budget against poverty levels and in rural areas were respectively 7 and 9 deaths health outcomes suggest that government resources per 1,000 live births higher than mortality rates in may not be effectively targeting some of the poorer urban areas in 2021. The gap in infant mortality and provinces (Figure 2.31 and Figure 2.32). Cambodia Public Finance Review - From Spending More to Spending Better 47 Figure 2.30. Infant and under-5 mortality for both urban and rural significantly declined over the last two decades; child mortality gap between urban and rural areas and gap between poorest and richest persists a. Child mortality gap, infant and under-5, between b. Child mortality gap, infant and under-5, between poorest and richest (deaths per 1,000 live births) urban and rural (deaths per 1,000 live births) (2005- (2005-2021/22) 2021/22) 90 Infant mortality gap Under-5 mortality gap 50 Infant mortality gap Under-5 mortality gap 45 80 deaths per 1,000 live births death per 1,000 live births 40 70 35 60 30 50 25 40 20 30 15 20 10 10 5 0 0 2005 2010 2014 2021-22 2005 2010 2014 2021-22 Sources: CDHS (2005-2021/22) Note: CDHS=Cambodia Demographic and Health Survey Infant mortality: the probability of dying between birth and the first birthday. Under-5 mortality: the probability of dying between birth and the fifth birthday. Child mortality gap=rural mortality rate-urban mortality rate. The higher the figure, the wider the gap. Figure 2.31. Recurrent spending 2021 (% of Figure 2.32. Recurrent spending 2021 (% of total) and poverty rate (2021), by province total) and infant mortality 2021 (deaths per 1,000 live births), by province 8% 8% 7% 7% 6% 6% % of spending % of spending 5% 5% 4% r² = 0.0545 4% 3% 3% r² = 0.0971 2% 2% 1% 1% 0% 0% 0% 10% 20% 30% 40% 0 10 20 30 40 50 % of poverty Deaths per 1,000 live births Sources: MEF, MOH (2022), CDHS (2021/22), Cambodia Poverty Assessment (2022), General Population Census (2019) and World Bank staff estimates Note: MEF=Ministry of Economy and Finance, CDHS=Cambodia Demographic and Health Survey Infant mortality: the probability of dying between birth and the first birthday. Under-5 mortality: the probability of dying between birth and the fifth birthday. The budget allocation at the provincial level covers PHD, OD and health facilities in the provinces; that excludes budget allocation for regional health training centers in the provinces 2.2.2. Public expenditure and 2.8 percent in 2021, due to a decline in both outcomes: Economic sectors government investment spending and DP- funded projects. DP-funded investment spending Overall expenditure for economic sectors has the largest share of total spending, accounting declined from 3.4 percent of GDP in 2011 to for around two-thirds of total spending for the sector. 48 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.33. Expenditure trend for economic sector, by line ministry (% of GDP) (2011-2021) a. Reccurent spending, by line ministry (% of GDP) b. Government public investment spending, by line (2011-2021) ministry (% of GDP) (2011-2021) 0.8 1.0 0.7 0.9 0.8 0.6 0.7 0.5 % of GDP % of GDP 0.6 0.4 0.5 0.3 0.4 0.3 0.2 0.2 0.1 0.1 0 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 MOC MAFF MRD MOWRAM MISTI MOC MAFF MRD MOWRAM MISTI c. DP-funded public investment spending, by line ministry (% of GDP) (2011-2021) 2.0 1.5 % of GDP 1.0 0.5 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 MOC MAFF MRD MOWRAM MISTI Sources: Budget Settlement Law (2011-2021); World Bank staff estimates Recurrent spending in economic sectors slightly 2021, driven mainly by a rise in the wage bill (Figure increased from 0.5 percent of GDP in 2011 to 0.7 2.36). Wage bill costs reached 46 percent of the total percent in 2021, driven by a rise in recurrent spending budget, an increase from 37 percent in 2017, which allocated to MAFF and MOWRAM, while MRD, MOC has the potential to crowd out capital spending. and MISTI recurrent spending remain stable (Figure 2.33.a). Government investment spending in the Externally financed project spending economic sectors declined from 0.9 percent of GDP accounted for nearly half (46 percent) of total in 2011 to 0.5 percent in 2021, driven by a decline MAFF expenditure, up from 30 percent in 2016. in MOWRAM and MRD capital expenditures (Figure Externally financed capital allocations have increased 2.33.b). Similarly, MOWRAM’s and MAFF’s external from KHR 114,410 million in 2017 to KHR 246,136 funded capital spending also declined, resulted in million in 2021. This is a continuation of a long- the decline for the economic sectors (Figure 2.33.c). term trend, following a doubling of donor-funded spending in agriculture between 2007 and 2015 Trends in the agriculture sector (World Bank, 2019). In contrast, the domestically- funded capital spending have decreased from 14 Public budget allocation on agriculture sector percent of total MAFF expenditure to 2 percent in is spread between MAFF and MOWRAM. 2019, before reaching 11 percent in 2021 (Figure Cambodia allocated more on agriculture than 2.35). From 2017 and 2020, a rise in wage bill and others in the region (Figure 2.34). MAFF recurrent non-wage expenditure appeared to have crowded budget allocation as a share of total budget allocation out government public investment. In 2021, wage increased from 81 percent in 2017 to 94 percent in bill and non-wage expenditure declined while MAFF Cambodia Public Finance Review - From Spending More to Spending Better 49 Figure 2.34. Agriculture allocation as % of total government budget: An international comparison % of government budget 10% 9% 9% 8% 8% 8% 7% 7% 7% 7% 7% 6% 6% 4% 4% 4% 4% 3% 4% 2% 0% 2016 2017 2018 2019 2020 Cambodia South-east Asia East Asia Sources: MEF; FAO Note: Agriculture budget includes MAFF and MOWRAM budget. Figure 2.35. MAFF expenditure, by category Figure 2.36. Composition of MAFF (% of total) expenditure (% of total domestic budgeted spending) 100% 100% 2% 90% 10% 20% 18% 21% 33% 32% 80% 30% 36% 80% 46% 48% 70% 40% 2% 7% 60% 60% 14% 12% 41% 41% 36% 50% 11% 40% 40% 30% 56% 62% 61% 55% 49% 50% 43% 20% 39% 40% 43% 20% 10% 0% 0% 2017 2018 2019 2020 2021 2017 2018 2019 2020 2021 Recurrent Wages Capital, domestically nanced Non-wage recurrent Capital, externally nanced Capital, domestically nanced Sources: Budget Settlement Law (2017-2021); World Bank staff estimates public investment jumped to 21 percent of MAFF two of the years showing a variance of less than 5 domestic spending (Figure 2.36). percent) but have since deteriorated, reaching a 44 percent expenditure outturn in 2020. On average, Actual aggregate expenditure that deviates there is a gap in performance between domestically significantly from the original approved budget financed capital projects (that tend to overspend) undermines fiscal discipline and the ability to and externally financed projects (that tend to control the total budget and, subsequently, underspend). The pattern is observed in the sector manage it effectively. Service delivery may also as a whole (Figure 2.38). This could be partially be affected where large deviations from planned explained by the difference in budgetary practices expenditure result in the contraction of services, that guide the planning and implementation of limitations on essential expenditures for key inputs, domestically and externally financed projects. or the suspension of certain services. DP-funded budgets tend to have lower spending relative to The agriculture sector continues to play a planned amounts (Figure 2.37). The deviations from significant role in Cambodia’s economy, planned spending of externally funded resources contributing approximately 22 percent of the country’s have been modest between 2016 and 2018 (with GDP. Impacted by the COVID-19 pandemic, the 50 Cambodia Public Finance Review - From Spending More to Spending Better economy contracted by 3.1 percent in 2020. Against GDP, and has risen by 9.3 percent during the 2021- the backdrop of the pandemic-induced slowdown, 22 rice production year (World Bank, 2022). While agriculture was the only sector that showed real no value added for specific crops is presented in the annual positive growth – albeit at a slower pace of national accounts, export data reveals a substantive 0.6 percent (Figure 2.39). increase in exports of a few major commodities.37 In 2020, Cambodia exported over 13 million tons of For the agriculture sector, crops are the best- agricultural products worth approximately US$3.5 performing subsector, with livestock and billion, a 17 percent increase from a year earlier poultry and fisheries growing again in 2020 (MAFF, 2021). These account for 5.5 percent of after declining in 2019 (Figure 2.40). Agriculture total exports, which are predominantly comprised contribution to GDP decline in 2019 was due of garments and footwear products.38 Agriculture mainly to bad weather. Total rice production 2019- exports39 increased by 76 percent between 2010 and 2020 harvesting was marginally lower than 2018 2021, driven by the surge of exports of cashew nuts production, while rice yield declined by 0.5 percent, (+328 percent, at more than US$1,000 million), fresh reaching 3.1 metric tons per hectare. 36 Agriculture mango (+27 percent), dried mango (179 percent), is projected to grow at an annual pace of 1 to 1.5 and black pepper (+478 percent).40 percent between 2021 and 2023. As the best- performing subsector, crops account for 12.9 Strengthening the agricultural product market percent of GDP in 2020 or about half of the agriculture infrastructure could help increase production. sector’s output. Production of rice, Cambodia’s main Collection points, marketplaces, packing houses, crop, accounts for close to 60 percent of agricultural and cold chains are essential to meet greater Figure 2.37. MAFF budget outturns Figure 2.38. Actual capital expenditure (% of approved budget, by budget type) (% of approved budget, by ministry and source of financing) 450% 200% 400% Recurrent 383% 150% 350% Capital, domestically nanced Capital, externally nanced 300% 100% 250% 50% 200% 182% 150% 129% 0% 105% 2015 2016 2017 2018 2019 2020 95% 101% 95% 88% 100% 96% MAFF (domestically nanced) 99% MAFF (externally nanced) 50% 84% 63% 68% 43% MOWRAM (domesically nanced) 44% 0% MOWRAM (externally nanced) 2017 2018 2019 2020 2021 MRD (domestically nanced) Sources: Comparison of Budget Law and Budget Settlement Source: MEF; World Bank staff estimates Law (2017-2021); World Bank staff estimates Note: Budget outturns over 100 percent of the approved Note: Budget outturns over 100 percent of the approved budget reflect actual spending exceeding the planned budget budget reflect actual spending exceeding the planned budget (overspent); under 100 percent means actual spending was (overspent); under 100 percent means actual spending was below the planned budget (underspent). MOWRAM 2017 below the planned budget (underspent). budget outturn was 416 percent 36 World Bank, Cambodia Economic Update (May 2020) 37 Data reported by the General Department of Agriculture (GDA) of MAFF. 38 World Bank (2021). Cambodia Economic Update June 2021. Phnom Penh: World Bank. 39 Data from the Department of Plantation and Sanitary and Phytosanitary of the General Department of Agriculture, MAFF 40 Data obtained from the General Directorate of Agriculture, Department of Plant Projection and SPS, MAFF Cambodia Public Finance Review - From Spending More to Spending Better 51 Figure 2.39. GDP growth, constant prices Figure 2.40. Growth in Agriculture (% year-on-year change) subsectors (2017-2020) 15 3% 11.6 2.2% 9.7 1.6% 11.3 8.5 2% 10 7 7.5 1.7% 7.1 6.2 8.7 1.3% 0.7% 4.5 5 7 5.5 1% 0.6% 1.7 6.8 6.2 1.8 3.6 1.2% 0.6% 2.2 1.5 0% 1.1 0.6 0.3% 0 -0.5 1 1.3 -1.4 -1.5 -1.4% -1.6% -3.1 -1% -1.2% -5 -1.8% -6.5 -2% -1.5% -10 -2.6% 2017 2018 2019 2020 2021p 2022p 2023p -3% GDP at constant market prices (% change) 2017 2018 2019 2020 Agriculture Crops Livestock & Poultry Industry Services Fisheries Forestry & Logging Sources: National Institute of Statistics for data from 2017 to 2020 and World Bank (2021) Cambodia Economic Update (December: 2021) for projection data from 2021 to 2023 production and modernization. Although the road of externally funded capital allocation (Figure networks and grid lines have been expanded across 2.41). MOWRAM domestic spending fluctuated as Cambodia in the past decade, many dirt roads still well. It increased from 2017 and peaked in 2019, the need to be upgraded to accommodate large trucks dropped between 2020 and 2022 due to the fiscal to transport agricultural produce. Cambodia has not constraint resulting from the COVID-19 pandemic. benefited from the increase in demand (hence higher From 2021 and 2022, the domestic spending prices for crops) due to its weakness in infrastructure, remained flat (Figure 2.42). Figure 2.43 shows that as well as costly trade logistics and trade facilitation the wage bill as a share of recurrent spending was (World Bank, 2022). flat over the period 2017 and 2021. The wage bill did not seem to have implications for Operations and Trends in the irrigation sector Maintenance (O&M) (Chapter 61). O&M spending as a share of recurrent spending slightly increased from Irrigation is another key area for the economic 63 percent in 2017 to 66 percent in 2021. sector, particularly for agriculture. Irrigation contributes to productivity improvements in the Similar to other ministries, MOWRAM’s agricultural sector, especially the crop sub-sector. In recurrent budget was well executed between the wake of the COVID-19 crisis, the Government 2017 and 2021. Recurrent spending was within a gives more priority to agriculture (and agro-industry) range of 5 percent deviation from recurrent budget in order to maintain economic stability and ensure allocation. However, actual domestically financed food security.41 The Ministry of Water Resources capital spending deviated from the originally approved and Meteorology (MOWRAM) is responsible for the budget. The deviation from the approved budget irrigation water supply and the use of water resources grew, from 174 percent of approved budget in 2017 in general. Most of MOWRAM’s funding comes from to 231 percent in 2021 (Figure 2.44), suggesting external financing, averaging more than two-thirds of weak government budget planning in rehabilitation the agency’s budget in 2021, most of which comes and construction of irrigation. The budget of from China. investment expenditure (Chapter 21) was generally approved in the budget law at the beginning of the In terms of start-of-the-year allocation, the year. The government has introduced regulatory overall spending for MOWRAM has fluctuated framework governing the implementation of PIM (Box between 2017 and 2022, driven by fluctuation 2.3). However, within the implementation phase, the 41 RGC, 2021 52 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.41. Recurrent and external Figure 2.42. Composition of MOWRAM financing for MOWRAM (KHR, million) spending (2017-2022) 1,200,000 100% 90% 1,000,000 KHR, million 80% 800,000 70% 64% 65% 69% 71% 68% 60% 600,000 50% 400,000 40% 30% 200,000 28% 27% 20% 25% 22% 25% 0 10% 2017 2018 2019 2020 2021 0% 6% 6% 8% 7% 8% Central Ministry-recurrent Central Ministry-capital 2017 2018 2019 2020 2021 Provincial Dept-recurrent External ncing Wages Non-wage recurrent Capital, domestically nanced Sources: Budget Settlement Law (2017-2021); World Bank staff estimates budget of investment expenditure (Chapter 21) can be on O&M jobs for newly completed irrigation projects’ adjusted significantly, from 0 percent to 100 percent budgets for them under service expenses (Chapter of budget approved in the budget law. This tended to 61). At the same time, it has also allocated part of the reduce the operational efficiency of spending. There national budget for investment expenditure (Chapter should be proper planning for rehabilitation and 21) to rehabilitate irrigation schemes if the cost is less construction of irrigation (Chapter 21) in the budget than US$3 million. The O&M budget has increased law at the beginning of the year. Execution of DP- over time both in absolute amount and as compared funded investment fluctuated between 60 percent of to the rehabilitation budget. However, effective investment budget and 108 percent (Figure 2.44). O&M is limited by unclear assignment of roles and responsibilities for Provincial Departments of Water Effective operations and maintenance (O&M) Resources and Management (PDWRAM), especially is critical to ensuring the functioning of on regular tasks such as overall water management, irrigation schemes. Development partners who O&M of small and medium size irrigation schemes, have invested large sums rehabilitating irrigation and support to Farmer Water User Committee schemes suggested the importance of effective (FWUC). O&M to ensure maximum utilization. As a response, the government has ensured that MOWRAM takes Along with support to farmers, another Figure 2.43. MOWRAM recurrent spending, by economic classification (% of recurrent) (2017-2021) 100% 7% 7% 6% 5% 6% 80% 20% 23% 21% 23% 22% 60% 40% 63% 62% 65% 65% 66% 20% 0% 9% 8% 7% 6% 5% 2017 2018 2019 2020 2021 Chapter 60: Purchases Chapter 61: Services Chapter 64: Payroll expenses Chapter 62: Social bene ts expenses Chapter 65: Subsidy expenses Chapter 63: Taxes and tarrifs Sources: Budget Settlement Law (2017-2021) Cambodia Public Finance Review - From Spending More to Spending Better 53 Box 2.3. Current status of Public Investment Management Public Investment Management (PIM) is at its early stage of development. In 2020, RGC started piloting the integration of capital and recurrent budgeting within three line ministries: the Ministry of Public Work and Transport (MPWT), the Ministry of Rural Development (MRD), and the Ministry of Water Resources and Meteorology (MOWRAM). Built on experience working with the three ministries, MEF scaled up the implementation of full-fledged budget integration by combining capital and recurrent expenditure in BSP in nine line ministries, including MOH in 2022. PIM system reform strategy 2019-2025 has been adopted. To strengthen PIM, Sub-decree 41 on the management of public investment dated 2020 outlined (i) SOP for domestically financed public investment, (ii) public investment for sub-national administration, and (iii) project preparation and prioritization. Six main criteria for public investment project selection and budgeting: (1) rationality, (2) economic and financial viability, (3) best option, (4) affordability principle, (5) delivery principle, and (6) project readiness. MEF has adopted guidelines on Monitoring and Evaluation in the framework of performance informed budgeting in January 2020, setting out obligations and procedures for the preparations of quarterly and plans and reports. Department of Monitoring and Evaluation was established under GDB to manage the mechanism for performances monitoring and evaluation on the government financed programs. Trainings were provided to 23 LMs on annual performance plan and annual performance report. Fourteen additional LMs will be trained. Sources: MEF’s PEFA (2021) and Interview with MEF Note: MEF=Ministry of Economy and Finance; MRD= Ministry of Rural Development; MPWT= Ministry of Public Work and Transport; MOWRAM= Ministry of Water Resources and Meteorology, PIM= Public Investment Management; SOP=Standard Operating Procedure; PPP=Public Private Partnerships contextual relevance to the irrigation sector is dam construction have affected traditional irrigation climate change. Cambodia is one of the world’s systems, fishing, and transportation systems. The most flooding-exposed nations. Higher global share of climate change expenditure in GDP in temperatures are also expected to increase climate 2020 and 2021 were 2.2 percent and 2.3 percent risks and natural hazards for the country. Droughts respectively, mainly infrastructure development, and floods have affected agricultural sectors, while general health intervention and cash distribution in fluctuations in Mekong River flows partly due to social protection to the poor and the vulnerable. Figure 2.44. MOWRAM budget outturn (2017-2021) 250% 231% 230% 212% 206% 210% 179% 190% 174% 170% 150% 130% 98% 101% 108% 110% 93% 95% 90% 70% 99% 85% 83% 60% 50% 70% 2017 2018 2019 2020 2021 Recurrent Capital, domestically nanced Capital, externally nanced Sources: Comparison of Budget Law and Budget Settlement Law (2017-2021); World Bank staff estimates Note: Budget outturns over 100 percent of the approved budget reflect actual spending exceeding the planned budget (overspent); under 100 percent means actual spending was below the planned budget (underspent). 54 Cambodia Public Finance Review - From Spending More to Spending Better Table 2.2. Irrigation outputs reported from MOWRAM spending Irrigated areas Irrigated areas Reported outputs in irrigation coverage (ha) (% of total) 2019 2020 2021 2019 2020 2021 Total increase in irrigated areas for rice 28,489 27,962 22,588 98.5% 98.1% 94.7% Increase in irrigated areas for rainy season rice 19,432 17,902 15,601 67.2% 62.8% 65.4% Increase in irrigated areas for dry season rice 9,057 10,060 6,987 31.3% 35.3% 29.3% Other crops 412 535 1,261 1.4% 1.9% 5.3% Total 28,901 28,497 23,849 100% 100% 100% Sources: MOWRAM’s annual report (2019-2021) A related area for review is the Ministry’s focus While Cambodia continued to depend on on irrigation for rice. Based on MOWRAM data, external financing for its physical infrastructure its investment in irrigation is for use in rice cultivation development, rising domestic revenue and more than 98 percent in 2019 and 2020 and about expanding fiscal space allowed the country 95 percent in 2021 (Table 2.2). to boost its domestically-financed capital spending. In the past, externally-funded capital The main gap in the current system is the lack spending was the largest share of total spending on of information on outcomes from investment infrastructure.42 It dropped by nearly half, accounting and spending on irrigation. The government is for around 1.8 percent of GDP in 2021, a decline aware of the need for more information at the impact from 3.4 percent in 2011. Domestically funded level and has concrete plans for addressing this infrastructure spending increased from 0.8 percent of shortcoming. While much information on outputs GDP in 2011 to 1 percent in 2021. Ministry of Public has been produced, there is virtually no information Work and Transport (MPWT) spending allocated available on impact. for infrastructure has taken up the largest share of infrastructure expenditures. 2.2.3. Public expenditure and outcomes: Infrastructure sectors Recurrent spending in infrastructure sectors fell as low as 0.2 percent of GDP in 2011- While Cambodia’s growth was impressive 2013, spiked in 2014 corresponding with a in the last decade, it lacked a diversified spending rise in MME, then started to stabilize base. Economic diversification is necessary to at around 0.6 percent (Figure 2.45.a). The increase underpin sustained and robust growth, driven by in government staff salary between 2011 and 2021 diversifying exports and moving up to high value- appeared to crowd out other non-wage expenditures added products. However, economic diversification in infrastructure sectors. Over the period 2011 and requires investments in infrastructure. The NSDP 2021, government staff salary as a share of GDP 2019-2023 articulates four interconnected and tripled. However, this salary increase was offset by complementary priorities: (i) improving the logistics the decline in purchases expenses (Chapter 60), system, enhancing physical infrastructure, energy, maintenance services expenses (Chapter 61) and and digital connectivity; (ii) developing important and social benefits expenses (Chapter 62), resulting in novel sources of growth; (iii) being prepared for the stabilized total recurrent spending in infrastructure digital economy; and (iv) fostering the growth of the sectors. The spike of expenditure subsidy in 2004 financial and banking sector. was due to a rise of subsidy including grant to 42 Cambodia transport sector assessment, strategy and roadmap (2019) Cambodia Public Finance Review - From Spending More to Spending Better 55 establishment of MME in early 2014. The MME for large infrastructure projects. The new PPP Law subsidy spending in 2014 reached KHR 193,189 (2021), together with the related Standard Operating million (0.3 percent of GDP) (Figure 2.45.b). Procedures and Guidelines, is expected to help strengthen the institutional and legal framework Government investment spending from around PPPs. To help support the development of domestic fund on infrastructure sectors PPPs, the RGC has also upgraded the PPP Unit of increased from 0.8 percent of GDP in 2011 the Ministry of Economy and Finance to a General to 1 percent in 2021, driven by an increase in Department of PPPs. The new PPP law: clarifies government spending on investment for MPWT institutional responsibilities in the different stages of and MME. However, there were a lot of variations the PPP project cycle (from identification, appraisal, in government investment spending on SSCA, preparation, and project approval through to MPWT, MME, and MPTC between 2011 and 2021 implementation of the PPP contract), and sets out the (Figure 2.45.c). DP-funded capital investment in types of investment incentives and state support— infrastructure declined due to reduction in DP funded including direct fiscal payments, contingent liability capital investment in MPWT and MME overall (Figure guarantees, investment incentives, and in-kind 2.45.d). contributions—that can potentially be made available to support PPP projects (PPP regulatory framework In addition to government spending on is shown in Box 2.4). infrastructure, the efficiency of public private partnerships (PPP) could be explored. Cambodia Few PPP projects outside of the energy sector has been looking to increasingly utilize PPP models have been successfully implemented to date. Figure 2.45. Expenditure trend for infrastructure sectors, by line ministry (% of GDP) (2011-2021) a. Reccurent spending in infrastructure sectors, by b. Recurrent spending in infrastructure sectors, by line ministry (% of GDP) (2011-2021) economic classification (% of GDP) (2011-2021) 1.0 0.45% 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 0.9 0.40% 0.8 0.35% 0.7 % of GDP 0.30% 0.6 % of GDP 0.5 0.25% 0.4 0.20% 0.3 0.15% 0.2 0.10% 0.1 0 0.05% 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 0% Chapter 60: Chapter 61: Chapter 64: Chapter 62: Chapter 65: Chapter 63: Purchases Services Payroll Social bene ts Subsidy Taxes SSCA MME MPTC MPWT expenses expenses expenses and tariffs c. Government capital spending in infrasructure d. DP-funded capital spending in infrastructure sectors, by line ministry (% of GDP) (2011-2021) sectors, by line ministry (% of GDP) (2011-2021) 1.2 5.0 1.0 4.0 0.8 % of GDP % of GDP 3.0 0.6 2.0 0.4 1.0 0.2 0 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 SSCA MME MPTC MPWT SSCA MME MPTC MPWT Sources: Budget Settlement Law (2011-2021); World Bank staff estimates 56 Cambodia Public Finance Review - From Spending More to Spending Better Box 2.4. Legal and regulatory framework for Public Private Partnership Recognizing the importance of PPP in the development of infrastructure in Cambodia, the government has taken initiatives to develop law and regulations governing the PPP arrangements. The Law on concession, enacted in 2007, granted the Council for the Development of Cambodia (CDC) to review PPP investment proposal. However, the guiding process to review PPP projects were yet to be developed. In June 2016, the government adopted “Policy Paper on Public-Private Partnerships for Public Investment Project Management 2016 – 2020”. It laid out framework for implementing PPPs. In November 2021, Law on Public Private Partnership was enacted to replace the 2007 Law on Concession for the implementation of PPP projects. SOP for implementing PPP projects was adopted in August 2022. PPP is contractual arrangement whereby a public and private entities share risks, jointly allocate required capital and other resources to deliver an infrastructure service. In Cambodia PPP can be seen in the development of the infrastructure sector. PPP projects take five main forms of contractual arrangements: Services Contract, BOO/BOT, Concession and License. 1. Service Contract: It refers to a contractual arrangement by which the government places upfront investment in the infrastructure; the private sector contractor provides designs and construction, and operates and maintains the facility. 2. Build-Own-Operate (BOO): Under BOO agreement, the private sector contractor financed infrastructure up-front investment by using its own resources, own and manage the infrastructure facilities and it received fixed annual payments from the government under availability payments. The government does not take ownership of assets at the end contract period. 3. Build-Operate-Transfer (BOT): Under BOT, private sector up-front investment is required. The private sector contractor is responsible for initial construction, upgrade or rehabilitation of infrastructure. The assets will be handled to the government party at the end of operations period. 4. Concession: It is a form of contract under which the private sector contractor is responsible for operate and maintain an already-built facilities. The private sector contractor generates revenues from third parties under user fees. Designed Build Lease (DBL) for water operators is one example of concession in which the water supply facilities were built by the government and they were leased to the private contractors to operate to deliver water services. 5. License: Under license, the government does not involve in any investment in infrastructure. Instead, a license is granted to private operators, which make investment, operate and maintain the facility in return for revenue collected from customers. Licensed private water operators have to comply with services requirements set out by government agencies. Source: World Bank staff based on Laws and regulations governing PPP projects in Cambodia. The World Bank’s Private Participation in billion), Philippines (33 projects at US$8.85 billion) infrastructure (PPI) Database includes only 8 projects and Vietnam (65 projects at US$23.73 billion). One that had been closed in Cambodia between 2017 of the main reasons for this relatively small number and 2022 (Figure 2.46) with a total investment cost of of projects is that Cambodia has historically had a US$1.34 billion.43 The number and volume of PPPs in relatively weak enabling environment for PPPs in Cambodia is relatively low compared with benchmark terms of project preparation, procurement, and countries, including Thailand (19 projects at US$9.53 managing unsolicited proposals when compared 43 This data does not include the recently completed Phnom Penh - Sihanoukville Expressway project. Cambodia Public Finance Review - From Spending More to Spending Better 57 Figure 2.46. Number and investment Figure 2.47. Cambodia’s PPP Performance costs of PPPs in Cambodia and selected (2020) countries (2017 – 2022) 23.73 bn 25,000 65 projects 12 8 20,000 US$ millions 15,000 Preparation Procurement 8.85 bn 9.53 bn 33 projects19 projects 10,000 6.17 bn 19 projects 76 17 2.45 bn 5,000 7 projects 1.34 bn 8 projects Contract Management Unsolicited Proposals 0 Global Average Regional Average Income Group Average Bangladesh Cambodia Myanmar Philippines Thailand Vietnam Source: Private Participation in Infrastructure (PIP) Database, Source: Benchmarking Infrastructure Development, (World (World Bank, 2022) Bank, 2020) to regional and income-level peers (Figure 2.47). It Guidelines which will set the institutional framework is expected that Cambodia’s PPP benchmarking and processes for managing fiscal risks. It will be performance will improve with implementation of the critical that RCG ensures that the framework set out new PPP Law. in these guidelines is strictly applied to ensure that fiscal risks are properly managed. While the new PPP Law and subordinate regulations have strengthened the PPP PPPs are not ‘free’, even when the public sector framework, some constraints remain in does not provide any direct or contingent ensuring that PPPs in Cambodia are structured financial support to a project. Some of planned properly and provide value for money. Executing large infrastructure PPPs in pipeline are expected agencies have limited capacity and experience on to be financed through bilateral financing, without PPP projects, which are usually complex and require direct or contingent liabilities to the government. specialized technical skills. Another constraint is the While the government may not have provided any provision in the PPP Law that permits the widely direct or contingent financial support to the project, and even received a small equity stake at no cost, a used application of direct negotiation or solicitation of PPP project is not ‘free’ as the end users must pay projects, thereby circumventing the use of competitive for the project through escalating tolls or tariffs over bidding to select projects. The RGC should ensure the term of the concession. Therefore, it is critical to that the use of direct negotiation is strictly limited, carefully assess the technical and financial proposals and that competitive bidding is the primary method (particularly in terms of the tariffs/tolls being charged of procurement, which typically produces better and the tax/royalty exemptions being sought) to value for money. In addition, the RGC must also ensure that projects are providing ‘value for money’ develop and enforce a strong framework to manage to the public. PPP-related fiscal risks to ensure that fiscal risks are carefully assessed, approved, and managed Trends in the roads sector throughout the lifecycle of a PPP project. The PPP Law assigns the responsibility of managing PPP MPWT budget allocation as a share of total related fiscal risks to MEF. It is understood that the government budget has been on decline General Department of Public-Private Partnerships between 2017 and 2021. MPWT budget allocation (GDPPP) is currently drafting Fiscal Management reached 8 percent of total government budget in 58 Cambodia Public Finance Review - From Spending More to Spending Better 2021, a gradual decline from 12 percent in 2017. road maintenance expenditures declined from 79.1 However, its budget allocation as a share of GDP percent of total recurrent spending in 2017 to 71.4 was stable at around 2.7 percent between 2017 percent in 2021 as payroll expenses rose (Figure and 2020, but that declined to 2.2 percent in 2021 2.50). (Figure 2.48). Between 2017 and 2021, the MPWT recurrent MPWT overall budget allocation showed that budget was well executed, within 5 percent recurrent funding as a share of total budget deviation. Government infrastructure investment had stabilized, while government planned (capital expenditure) in transport sector exceeded investment expenditures were on the decline, budgeted expenditures during the same period. offset by increase in DP-funded capital Overspending in government investment in transport allocation between 2018 and 2021 (Figure 2.49). sector tended to increase, reaching 4.2 times of The recurrent spending was driven by a rise in wage planned spending in 2021 from 1.2 times in 2017. bill expenditure, which increased around 1.5 times This reflected weak expenditure planning. Routine between 2017 and 2021. With stable recurrent maintenance expenditures were planned in budget spending, this significant increase in wage bill has under service expenses (Chapter 61). However, implication on road maintenance expenditures. capital maintenance in transport sector was not Service expenses (Chapter 61) which covered reflected in budget planning. Within the execution Figure 2.48. Trend of MPWT budget allocation (% of GDP and % of total government budget) (2017-2021) 21% % of GDP/% of total government budget 12% 11% 12% 11% 10% 8% 1% 2.7% 2.5% 2.7% 3.0% 2.2% 2017 2018 2019 2020 2021 MPWT budget (% of GDP) MPWT budget (% of total government budget) Sources: Budget Law (2017-2021); World Bank staff estimates Note: Budget allocation includes recurrent and capital-both domestically financed and externally financed. Figure 2.49. MPWT budget, by category Figure 2.50. Spending by budget (% of total) classification (% of recurrent) 90% 2017 2018 2019 2020 2021 100% 80% 80% 70% % of recurrent 66.1% 62.0% 60% 70.2% 72.4% 72.5% 60% 50% 40% 40% 30% 21.8% 10.4% 20% 21.4% 16.1% 15.0% 20% 16.2% 13.7% 17.0% 10% 12.5% 12.6% 0% 0% 2017 2018 2019 2020 2021 Chapter 60: Chapter 61: Chapter 64: Chapter 62: Chapter 65: Chapter 63: Purchases Services Payroll Social Subsidy Taxes Recurrent expenses bene ts expenses and Capital, domestically nanced expenses tarrifs Capital, externally nanced Sources: Budget Law (2017-2021); World Bank staff estimates Sources: Budget Settlement Law (2017-2021; World Bank staff estimates Cambodia Public Finance Review - From Spending More to Spending Better 59 Figure 2.51. MPWT budget outturns (% of approved budget, by budget type) 450% 426% 400% 350% 300% 250% 200% 202% 150% 129% 112% 130% 102% 100% 100% 97% 99% 102% 71% 78% 72% 94% 50% 63% 2017 2018 2019 2020 2021 Recurrent Capital, domestically nanced Capital, externally nanced Sources: Budget Law (2017-2021) and Budget Settlement Law (2017-2021); World Bank staff estimates Note: Budget outturns over 100 percent of the approved budget reflect actual spending exceeding the planned budget (overspent); under 100 percent means actual spending was below the planned budget (underspent). process, this capital maintenance spending was 2019 Global Competitiveness Index. Quality of captured under specific project spending. It should roads, railroads and ports and air transport ranked be properly embedded in PIP and made available respectively 99, 94, 81 and 106 (Table 2.3). 100 in the budget documents. 44 Unlike government percent of the 2,254 Km national road network public investment, externally funded investment were (1-digit)-connecting the country and its borders were consistently under-executed (Figure 2.51). paved, while 72 percent of national roads (2-digit roads comprising 5,007 km) were paved, and only Cambodia spent 2.8 percent of GDP on 30 percent of provincial roads (3- and 4-digit roads infrastructure in 2021; however, the quality comprising 9,031 km) paved.45 of infrastructure remains limited compared to regional peers. Cambodia scored 3.4 out of Despite investments in road improvement, 7 and ranked 99 out of 137 countries on overall which dominated MPWT spending, Cambodia’s infrastructure quality in World Economic Forum’s logistics performance declined (Figure 2.52). Table 2.3. Quality of infrastructure (2018) Overall Electricity Roads Railroads Ports Air transport infrastructure supply Score Rank Score Rank Score Rank Score Rank Score Rank Score Rank Vietnam 3.6 89 3.4 92 3.0 59 3.7 82 3.7 103 4.3 90 China 4.5 47 4.6 42 4.8 17 4.6 49 4.9 45 5.0 65 India 4.6 46 4.3 55 4.4 28 4.6 47 4.6 61 4.7 80 Cambodia 3.4 99 3.2 99 1.6 94 3.7 81 3.7 106 3.5 106 Thailand 4.1 67 4.3 59 2.6 72 4.3 63 5.2 39 5.2 57 Philippines 3.0 113 3.1 104 1.9 91 2.9 114 2.9 124 4.2 92 Indonesia 4.1 68 4.1 64 4.2 30 4.0 72 4.8 51 4.4 86 Source: World Economic Forum Global Competitiveness Index (2019 report) Note: Score 1-7 Best; Rank: Out of 137 countries 44 Cambodia, Public Investment Management Assessment (PIMA), 2019 45 ADB, transport sector assessment (2019) 60 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.52. Logistics performance index: Figure 2.53. Paved roads as percentage of quality of trade and transport related total road network (2015-2020) infrastructure (2007-2018) 100% 94.4% 3.2 90% 80% 2.7 68.3% 69.2% 70% 59.4% 59.5% 2.2 60% 50% 1.7 40% 30% 20% 16.6% 18.7% 1.2 2007 2010 2012 2014 2016 2018 10% Cambodia Myanmar 0% Lao PDR Thailand es a a R nd ar m di si PD na in nm la bo ne Vietnam Philippines pp ai et o ya do am Th Vi ili La Indonesia M In Ph C Sources: World Bank, Logistic Performance Index (LPI) Surveys Source: International Road Federation (IRF) (extracted April (extracted April 2023) 2023) and World Bank staff Note: LPI measures Logistics professionals’ perception of country’s quality of trade and transport related infrastructure (e.g. ports, railroads, roads, information technology), on a rating ranging from 1 (very low) to 5 (very high). Scores are averaged across all respondents. Respondents evaluate eight markets on six core dimensions on a scale from 1 (worst) to 5 (best). Paved roads as a share of total road networks in for cross-sectoral programs increased from 2.3 Cambodia remained the lowest in ASEAN between percent of GDP in 2011, peaked 4.1 percent in 2013 2015 and 2020 (Figure 2.53). Similarly, Cambodia and reached 3 percent in 2021. This was driven by ranked lower than most ASEAN neighbors, except spending fluctuation in General Administration (GA) Myanmar, on the logistics performance index (Figure sector (Figure 2.54.a). Government public investment 2.52). spending was stable at around 0.6 percent of GDP between 2011 and 2018, and jumped to 1.8 percent, 2.2.4. Public expenditure and 2.2 percent and 2 percent respectively for 2019, outcomes: Administrative services & 2020 and 2021, mainly driven by government capital cross-sectoral programs spending for GA (Figure 2.54.b). DP-funded capital spending in cross sectors remarkably fluctuated. It The NSDP planned to spend about 21 percent peaked 0.5 percent of GDP in 2013, then dropped of the total budget for Services and Cross- to 0.19 percent in 2021. The spending fluctuation is sectoral Programs. The actual amount spent seen in spending key ministries including GA, MOE, by all the agencies in this sector family—which is MOT, and MOCAR etc. (Figure 2.54.c). responsible for promoting gender equity, tourism, environment and conservation, community and The Ministry of Civil Service (MCS) is social services, culture and the arts, and governance responsible for all matters related to more than and administration—was slightly higher, with about 200,000 government employees. The geographic 23 percent of the budget for the four sectors. distribution of civil servants among central line ministries varies by sector (Figure 2.55). Overall, the The cross-sectoral spending increased from line offices of the central ministries at the District/ 3.2 percent of GDP in 2011 to 5.3 percent in Municipality/Khan (DMK) level absorb around 60 2021. The spending on cross-sectoral programs percent of total wage spending, with the remaining significantly fluctuated, driven primarily by the share split evenly between the central and provincial fluctuation of recurrent budget. Recurrent spending levels. Much of this reflects the sectors with large Cambodia Public Finance Review - From Spending More to Spending Better 61 Figure 2.54. Expenditure trend for services and cross sectors, by line ministry (% of GDP) (2011-2021) a. Reccurent spending in services and cross sectors, b. Government capital spending in services and by line ministry (% of GDP) (2011-2021) cross sectors, by line ministry (% of GDP) (2011-2021) 4.5 2.5 GA MCS MOIn MCFA MOE MOCAR MOWA MOT 4.0 GA MCS MOIn MCFA MOE MOCAR MOWA MOT 2.0 3.5 3.0 % of GDP 1.5 % of GDP 2.5 2.0 1.0 1.5 1.0 0.5 0.5 0 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 c. DP-funded capital spending in services and cross sectors, by line ministry (% of GDP) (2011-2021) 0.6 0.5 0.4 GA MCS MOIn MCFA MOE MOCAR MOWA MOT % of GDP 0.3 0.2 0.1 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Sources: Budget Settlement Law (2011-2021); World Bank staff estimates cadres of front-line service providers, such as across-the-board for all civil servants, and education and health, in which the subnational share spending rose to about 7 percent of GDP by of workers, and thus of the wage bill, is considerably 2022, worker performance was not linked to higher (Figure 2.55). Education accounts for 69 pay increases. International experience points to percent of all subnational staff, and 81 percent of the importance of determining public sector wage staff working at the DMK level (and along with health policy as an integral part of an overall civil service these two sectors account for more than nine out reform, which includes human resource development of ten workers at the DMK level). Smaller ministries, and management. Annual across-the-board salary such as Land Management, Urban Planning and increases with limited reform in human resource Construction, Environment, and Mines and Energy, development and management did not appear to are also relatively deconcentrated. On the flip side, have improved public service delivery. While the the most centralized ministries are relatively large: government has increased the average monthly pay Public Works and Transportation; Social Affairs and per civil servant, administrative sectors benefited the Veterans; and Rural Development. 46 Overall, over most from the increase (Figure 2.57). three-quarters of overall expenditures are executed at the central level in 2021 (Figure 2.56). Regional comparison shows that Cambodia’s public sector wage spending as a share of While the government increased salaries public expenditures is on the high side (Figure 46 MoSVY at the central level controls almost half of wage spending, but 97 percent of non-wage. However, this should be treated with caution, as it includes social security benefit and not operational budget. 62 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.55. Civil servants employed at Figure 2.56. Central and provincial shares different levels within selected line ministries; of line ministry recurrent spending, 2021 number and share of the total, 2020 100% National 83% 17% 866 364 301 MPTC 87% 13% 90% MOE 53% 47% 80% MOT 72% 28% 70% 132,372 13,291 MCFA 84% 16% 2,806 838 MLMUP 62% 38% 60% 853 107,800 MOC 87% 13% 50% MLVT 77% 23% 40% MAFF 60% 40% MME 41% 59% 30% 40,730 5,467 MRD 74% 26% 20% 2,867 947 MOSVY 96% 4% 688 10% 45,619 MOH 99% 1% 11,955 4,383 MPWT 87% 13% 0% 3,696 MOEYS 10% 90% Total MOEYS MOH MAFF MOSVY MRD 0% 20% 40% 60% 80% 100% Cental Provincial District Central Provincial Sources: MCS and MEF Note: Recurrent spending includes wage and non-wage expenditures. Miscellaneous expenditures were integrated in Central spending at National level. MOH recurrent spending excludes Covid-19 expenditures. 2.58). However, due to the challenges facing public a share of GDP rose to 8 percent in 2018, up from administration reform, public sector wage increases 4.3 percent during the same period. However, after were not accompanied by an improvement in public the minimum wage target was met or exceeded, the service delivery. wage bill increase moderated. As a share of total total spending, the wage bill declined to 31 percent (7.4 The public sector wage bill started to quicky perent of GDP) in 2019 and 26 percent (7.2 percent increase in 2013 when the authorities targeted of GDP) in 2021. To save budgetary resources for civil servants’ minimum wage to reach at least financing necessary public health expenditures to KHR 1 million (US$250) a month by 2018. As a fight the spread of the corona virus, salary increase result, the total wage bill as a percentage of total was frozen during the period 2020-21. While the government spending peaked at 37 percent in increase has been reintroduced in 2023, rising 2018, up from 21.1 percent in 2013, declined to demand requires a substantial increase in public 26 percent in 2021. Wages and compensation as outlays, which are projected to reach almost 30 Figure 2.57. Monthly average pay of Cambodian civilian civil servants, 2021 2000 1800 1600 1400 US$/month 1200 1000 800 600 400 200 0 Royal Palace MOLMUC MOIn MME MOWRAM MOI MPTC MOC MOJ ACU MPWT COM MOE MISTI MEF MOEYS MAFF MOH MRD MOP MOT MCS MONSRI MOSVY MCR SSCA MOWA MFAIC MCFA MLVT Sources: Cambodian authorities (2021 Budget Settlement Law) Note: Actual spending under Chapter 64 for all ministries (except for MOH which is for 2020) and 2022 civilian civil servant statistics (MCS). Cambodia Public Finance Review - From Spending More to Spending Better 63 Figure 2.58. The footprint of the public wage bill a. Wage bill (as % of public expenditure) over time b. Wage bill and GDP pc in the EAP region 40% 45 Vietnam Percentage of public expenditure 40 Palau Percentage of public expenditure 35% Cambodia 35 Papua New Guinea 30 30% Indonesia 25 Micronesia 20 25% Philippines Thailand 15 10 20% Mongolia 5 Myanmar 0 15% Timor Leste 2005 2007 2009 2011 2013 2015 2017 2019 10% Thailand Cambodia Myanmar 7 7.5 8 8.5 9 9.5 10 Philippines Indonesia Vietnam GDP per capita (constant 2015 PPP), log Sources: World Bank’s Worldwide Bureaucracy Indicators based on IMF data percent of GDP in 2023, up from 25.5 percent of considering direct taxes and direct transfers, the Gini GDP in 2020, reducing authorities’ fiscal space for index reduces slightly to 32.2 percent for Disposable raising civil servants’ wages much further. Income. Indirect taxes and subsidies have limited effect, and their consideration leaves the Gini index 2.3. Distributional impacts of for Consumable Income at 32.2 percent. “In-kind” fiscal policies transfers from health and education, on the other hand, have the largest effect on inequality with the In Cambodia, the overall fiscal system Final Income Gini at 31.4 percent. This reflects the somewhat reduces inequality. Bearing in mind fact that in-kind transfers represent a significant the caveat that top income households are often share of pre-fiscal income and proportionally benefit underrepresented in household survey data—a lower-income households more than those at the common problem with most surveys, especially upper end of the income distribution. in developing countries—the fiscal system in Cambodia reduces the Gini coefficient by one While Cambodia does reduce inequality through percentage point. Inequality, as measured by the Gini taxes and transfers, the degree of redistribution coefficient, falls between Market Income and Final is small by international comparison. Figure 2.60 Income (Figure 2.59). Before any fiscal interventions, demonstrates that the redistributive effect of the Market Income Gini index is 32.4 percent. After fiscal policy (including in-kind transfers) is lower Figure 2.59. Gini Index before and after fiscal interventions in Cambodia 34 Gini cof cient (%) 33 32.4 32.2 32.2 32 31.4 31 30 Market Disposable Consumable Final income income income income Sources: Authors’ calculations based on CSES (2019/20) 64 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.60. Change in Gini after fiscal policy (Including in-kind transfers) 0 -5 Point Change -10 LIC -15 HIC UMIC LMIC -20 -25 Dominican Republic Costa Rica Burkina Faso El Salvador Sri Lanka South Africa Cambodia Comoros Nicaragua Colombia Colombia Russia Honduras Ecuador Guatemala Tanzania Uganda Panama Jordan Romania Indonesia Venezuela Belarus Uruguay Poland Kenya Egypt Croatia Ghana Tunisia Namibia Mexico Mexico Mexico Argentina Ethiopia Albania Bolivia Brazil Ukraine Paraguay Chile Armenia Peru Peru Iran High income Upper middle income Lower middle income Low income Sources: CEQ and World Bank databases and World Bank calculations (see Rodriguez and Wai-Poi 2020) Note: HIC = high-income country, UMIC = upper middle-income country, LMIC = lower middle-income country, LIC = low-income country. in Cambodia than in some lower-middle-income decentralization reform effort resulting in inadequate countries. Some lower-middle-income countries distribution of resources at front line public service achieve inequality reduction of up to 9 percentage providers. Each of the issue is explored in detail points from the pre-fiscal income. When in-kind below. transfers are excluded, the redistributive effect in Cambodia is even lower. This is because of the very 2.4.1. A need to strengthen strategic low impact of direct transfers on inequality. budget management 2.4. Strengthening public sector The aggregate annual domestic budget is generally credible. Budgeted amounts approved management for improved by the National Assembly are generally closer quality of public expenditure to the actual amounts spent for the total budget. The most recent PEFA assessment (2021) The limited outcomes from public spending suggested a high level of budget credibility since outlined in earlier sections are attributable to expenditure outturn compared to approved budget three main issues. First, there is limited budget for 2017, 2018, and 2019 was within +/- 5 percent: management effectiveness as reflected in a weak 97.9 percent in 2017, 99.4 percent in 2018, and 99.3 link between planning and budgeting, and between percent in 2019 (PEFA 2021). Expenditure outturns spending and performance outcomes; in addition, based on comparison between Budget Law and the lack of budget discipline at sector level, limited Budget Settlement Law was 91 percent in 2020 and usefulness of program structure for performance 92 percent in 2021.47 However, there is a need to informed budgeting, and weak public investment link national development planning with budgeting to management also point to its limited effectiveness. improve allocative efficiency. Second, human resource management remains weak as reflected in the absence of performance- Strengthening budget discipline can help to based management while public sector pay is on improve operational efficiency, and hence average higher than private sector’s salary. There is quality of expenditure. A closer look at the no human resource management information system public expenditure patterns suggests a need for to support performance-based management. Third, more attention to operational efficiency. In general, Cambodia remains highly centralized after 20 years of recurrent expenditure outturn is much closer to the 47 Budget Law and Budget Settlement Law 2020 and 2021 Cambodia Public Finance Review - From Spending More to Spending Better 65 Figure 2.61. Overall budget outturn at national level (2011-2021) 150% 141% 140% 137% 130% 123% 130% 120% 106% 110% 98% 95% 97% 101% 99% 101% 99% 84% 97% 98% 100% 101% 96% 95% 90% 97% 98% 91% 94% 80% 86% 85% 94% 91% 79% 91% 86% 86% 70% 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Recurrent Capital, domestically nanced Capital, externally nanced Sources: Budget Settlement Law (2012-2021); World Bank staff estimates Note: Recurrent spending excludes Covid-19 expenditure approved budget; however, capital expenditures Refining budget strategic planning linked with fluctuated significantly. There is a big difference budgeting processes and realistic outcome between expenditure outturn for government and targets can help to support performance- externally funded public investment projects. For informed budgeting implementation. Strategic government funded public investment programs, budget planning could be further improved to the budget execution rate has been 90 percent for strengthen the linkage between policy and budgeting. all years, except in 2019 when the execution rate Achieving performance goals in a Budget Strategic reached to 130 percent. For DP funded investments, Plan (BSP) appears to be challenging across the budget execution rate was above 100 percent ministries. We demonstrate this point in Chapter 4 between 2012 and 2016, however, the rate fell below by providing a detailed analysis of the weaknesses 100 percent between 2017 and 2021 (Figure 2.61). in the performance matrix design and achievements against the indicators in the MAFF. Chapter 4 The budget and actual spending amount compares the 13 program-level yearly indicators deviates substantially across ministries. In (excluding the “final result” indicators) between the 2021, while the Ministry of Agriculture, Forestry and two most recent BSPs. The analysis demonstrates Fisheries under-executed its recurrent budget by 22 that the target values are unreasonable. The percent compared to budget plan, the Ministry of achievement of many of the performance indicator Mines and Energy overspent by 16 percent. Without targets require substantially more budgetary taking into account Covid-19 expenditure, MOH resources than allocated to the MAFF. recurrent expenditure outturn reached 87 percent of its budgeted level in 2021. MPTC recurrent The program structure of ministries needs budget outturn reached around 151 percent of total substantial changes to support performance- recurrent budget in the same year (Figure 2.62.a). In informed budgeting. The current program structure the economic and infrastructure sector, government reflects the existing administrative structures and capital investment tended to be over executed: arrangements rather than capturing sector level 231 percent for MOWRAM, 166 percent for MRD, outcomes. More importantly, all of the salary and 426 percent for MPWT in 2021 (Figure 2.62.b). expenses for a ministry are often registered under a Budget execution for DP funded capital allocation separate program, rather than distributing personnel fluctuated significantly across sectors and line costs across all programs. Therefore, expenditure by ministries (Figure 2.62.c). a program currently does not accurately capture all program costs, including personnel cost and capital 66 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.62. Overall budget outturn (2019-2021) a. Reccurent budget outturn, by selected line ministry b. Government capital budget outturn in selected line (% of total) (2017-2021) ministry (% of total) (2017-2021) % of budget execution % of budget execution 160% 450% 150% 400% 140% 350% 130% 300% 120% 250% 110% 200% 100% 150% 90% 100% 80% 50% 70% MOWRAM 0% MOH MOWRAM MPTC MOEYS MRD MME MCS MOE MAFF MOH MOSVY MPWT MOWA MOEYS MPTC MME MCS MOE MAFF MRD 60% MOSVY MPWT MOWA Social Economic Infrastructure Cross Social Economic Infrastructure Cross sectors sectors sectors sectors sectors sectors sectors sectors 2017 2018 2019 2020 2021 2017 2018 2019 2020 2021 c. DP capital budget outturn in selected line ministry (% of total) (2017-2021) 350% % of budget execution 300% 250% 200% 150% 100% 50% 0% MOH MOEYS MOSVY MAFF MOWRAM MRD MPTC MME MPWT MCS MOE MOWA Social Economic Infrastructure Cross sectors sectors sectors sectors 2017 2018 2019 2020 2021 Sources: Budget Law (2017-2021) and Budget Settlement Law (2017-2021); World Bank staff estimates Note: MOH recurrent spending excludes Covid-19 spending expenditure. In addition, achieving performance lower than in low-middle income developing results from budget expenditure requires a result- countries. The capital budget is appropriated as based management system, where program lump sum and allocation to individual projects managers have significant control over resources, occurs only after the budget is approved, preventing results are accurately measured and reported, coordination with the recurrent budget. Identifying programs are evaluated, and managers are individual projects and incorporating those in the incentivized and held accountable by a performance- budget preparation will improve the quality of budget oriented human resource management system. planning and implementation.48 Thus, current program structure does not support performance orientation. Having the right foundation Implementation of DP investment projects for program-based budgeting is an important building remain a key challenge in sectors such as block toward achieving performance accountability education, health and agriculture. Budget outturn (see Box 2.5). for DP capital investment tended to fluctuate, with Cambodia’s investment efficiency is below the weaknesses in the planning and implementation of average of emerging and developing Asian DP investment projects. There is a need for a more countries. Access to infrastructure in education, proactive role in coordinating with DPs in identifying, electricity, water supply and public health remained selecting and implementing the investment projects. 48 Public Investment Management Assessment (PIMA) 2019 Cambodia Public Finance Review - From Spending More to Spending Better 67 Box 2.5. Program budgeting (PB) Program budgeting reforms around the world typically intend to shift the focus of budgetary processes from the control of inputs to producing measurable results. It aspires to enable governments to deploy resources to priority areas and assess whether the resources have been translated into intended results. Linking resources with results can support in holding managers accountable for the delivery of specific targets and how the resources are deployed and used. Successful implementation program budget requires that results are accurately measured and reported, programs are evaluated, all expenditures associated with the program including capital expenditure and personnel costs are accounted for, managers have discretion over resources they manage, and managers are incentivized and held accountable by a performance oriented human resource management system. However, in many countries, PB implementation has faltered mainly due to the difficulty of combining the objectives of planning, management, and control in one budget system. In some countries, instead of providing more information to the government to manage its resources, program budgeting, as implemented, may have resulted in the implementation of more, rather than less, controls and a fragmentation of the budget, which is exactly counter to the aspirations of program budgeting. Source: World Bank (2022) Brumby,James A.; Hashim,Ali; Piattifuenfkirchen,Moritz Otto Maria Alfons. 2022. Introducing the New PPB: Pragmatic Program Budgeting: Overcoming Design Obstacles to Planning, Management, and Control (English). Equitable Growth, Finance and Institutions Insight Washington, D.C.: World Bank Group. http://documents.worldbank.org/curated/en/601921643045774672/Introducing-the-New-PPB-Pragmatic-Program-Budgeting- Overcoming-Design-Obstacles-to-Planning-Management-and-Control 2.4.2. Linking performance to salary characteristics such as experience, educational increase toward maximization of qualification, gender, and location) was 19 percent public expenditure quality in 2019.50 This positive wage premium experienced by public sector workers is higher than the regional There is a need for introducing performance- average of 3 percent and is among the highest based management to improve the quality of compared to all its peers, with only the Philippines service from public spending. Salary increases higher (Figure 2.63). Furthermore, the results hold across the board have not contributed to improved in the education and health sector, where the wage premium for working in the public sector is also performance. Cambodia’s public sector workers positive and significant in the country (6 percent in currently earn a higher salary compared to their healthcare and 17 percent in education). private sector counterparts. Average monthly public sector pay is higher than private sector, reaching Before 2019, the public sector in Cambodia KHR 1.5 million in 2019, compared to KHR 1.25 used to pay a wage penalty over similar private million for private sector.49 Cambodia’s public sector sector workers. In the two decades up to 2019, wage premium relative to the private sector wage the wage differentials between public and private employees (even after accounting for observable sector workers have shown a big income gap in 49 While public sector wage levels are an important determinant of personnel motivation and of state capacity, they have important consequences for expenditure efficiency, fiscal sustainability, and equilibrium outcomes in the entire labor market. However, answering the question “does the public sector pay too much or too little” naturally requires an appropriate comparator. Comparing the wages of public and private sector workers is analytically useful given that the most likely outside option to employment in the public sector is the private sector. 50 These wage differentials do not take additional sources of compensation like allowances and in-kind transfers into account which are likely higher in the public sector. Therefore, if anything, these estimates may be underreporting the wage premia. However, the lack of such information on the household surveys affects the ability to perform this additional analysis. 68 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.63. Public sector wage premium (compared to all private sector workers) (2019) 50% 45% % of public wage premium relative 40% 30% 18.8% to private sector 20% 10% 6.9% 1.7% 2.8% 0% -10% -3.4% -20% -15.1% -30% -24.90 Cambodia Philippines Thailand Micronesia Timor Leste Mongolia Indonesia EAP average Source: World Bank’s Worldwide Bureaucracy Indicators favor of private sector employment, with a wage higher compensation of public sector workers gave penalty for public sector workers ranging from 38 to room for a more competitive wage for the public 25 percent in the 2003-2012 period (Figure 2.64). workforce, resulting in a positive wage premium During that period, because of low wages offered by compared to their private sector counterparts, the government (compared to those in the private though with negative implications in terms of fiscal sector), government jobs were unattractive to the sustainability. Moreover, this higher premium may newly graduated and other skilled workers. This led be due to the inability of the private sector to keep to problems among government employees, such pace with wage growth in the public sector. The fact as moonlighting, while some other young skilled that these jobs pay significantly higher wages and workers left the government for the more generous provide more stable working conditions with better private sector jobs, thereby adversely affecting benefits could add distortions to the labor market government efficiency. 51 In the subsequent period, by raising the reservation wages, which could then and despite fiscal constraints, the government has result in queuing and crowding-out effects in the attempted to improve matters by applying across- private sector labor market. the-board annual increases in basic salaries. This Figure 2.64. Cambodia’s public sector wage premium over time (2003-2019) 30% % of public wage premium relative to private sector 20% 18.8% 10% 0% -10% -20% -24.8% -30% -40% -38.4% -43.4% -50% -48.7% -60% 2003 2006 2008 2012 2019 Source: World Bank’s Worldwide Bureaucracy Indicators 51 Oudom (2012). Public-private wage differentials in Cambodia and its implication for Public Human Resource Management. Cambodia Public Finance Review - From Spending More to Spending Better 69 Figure 2.65. Wage premium by education Figure 2.66. Public sector wage premium level (compared to private formal workers) by occupational groups (compared to (2019) private formal workers) (2019) 20% 60% 14.7% 15% 40.4% % of public wage premium % of public wage premium relative to private sector 40% relative to private sector 10% 7.5% 20% 5% 1.3% 0% 0% -20% -5.0% -5% -20.8% -23.8% -10% -40% -15% -60% -20% -16.4% -80% -65.0% No Primary Secondary Tertiary Senior ProfessionalsTechnicians Clerks Elementary education education education education of cials occupation Source: World Bank’s Worldwide Bureaucracy Indicators The difference in pay between the public and higher for entry-level civil servants than senior civil private sector varies by level of education. servants, which may be particularly disruptive to the Public sector workers with tertiary education earn 7.5 overall labor market. Given these high premia among percent more than their private sector counterparts. technicians, it will not be surprising to see a massive However, public sector workers with no education demand for these public sector jobs, with hundreds make 16.4 percent less than their private sector of applicants for each vacancy. counterparts. 52 The wage gap between public and formal private sector workers is highest among Female workers in the public sector suffer a workers with secondary education, at 14.7 percent significant gender wage gap in many segments in favor of public sector workers (Figure 2.65). A high of the public sector workforce. Globally, women wage premium among highly educated workers, often earn significantly less than men for doing together with a larger representation of these the same work in both public and private sectors. workers in the public sector relative to the private In Cambodia, females earn 94 percent the wage sector, suggests potential skilled labor shortages for of males in the private sector and 89 percent the the private sector with large implications for public wage of their male counterparts in the public sector sector wage bill sustainability. and this wage gap observed in the public sector has increased over time (Figure 2.67.a). Moreover, Technical occupations enjoy a higher premium women in Cambodia perform most tasks in the in the public sector than senior managerial public sector education and healthcare sectors, occupations. In Cambodia, public sector but their participation is mostly confined to lower- technicians enjoy a wage premium of 40 percent paid occupations. However, women working in the compared to private formal workers, mainly driven education industry face a wage penalty of 7 percent by the high proportion of police officers and public compared to their male counterparts working sector administrative officers. All other occupations similar jobs with similar hours (Figure 2.67.b), even in the public sector pay a wage penalty and the after accounting for differences in age, educational highest penalty is observed among senior officials qualifications, and location. In public administration, (65 percent), as seen in Figure 2.66. This is widely women experience a wage penalty of more than 30 observed in other countries, where premiums are percent. However, women in healthcare earn almost 52 This heterogeneity in the wage gaps by education level could be caused by many factors, and it is impossible to identify these factors without data from household surveys. It may stem from different productivity gaps in the public sector that vary across occupations, such as stronger union bargaining power of lower-skilled employees. 70 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.67. Gender wage gap in Cambodia a. Female to male wage ratio by sector of b. Public sector gender wage premium by industry employment (2019) % of female wage relative 110% 20% relative to male counterparts % of female wage premium 107% 107% 9.8% to male counterparts 105% 10% 100% 94% 0% 91% 93% 90% 89% 87% 86% -10% -7.0% 80% 80% -20% -30% 70% -30.7% -40% 2003 2006 2009 2012 2019 Public Education Healthcare Private sector Public sector administration Source: World Bank’s Worldwide Bureaucracy Indicators 10 percent more than men in the same industry. The percentage of GDP than in neighboring countries; wage penalty experienced by women is observed however, 60 percent of this expenditure is funded globally, however in Cambodia, the size of these by out of pocket and the largest share of it goes to penalties is higher. The average global gender wage private providers (47 percent). The government’s penalty is 3 percent for education and 2 percent for budget for health facilities per capita amounts to public administration industries, respectively. US$2.5 a year – inadequate to achieve Cambodia’s targets for the sustainable development goals. 2.4.3. Adequacy in the distribution However, 35 percent of government spending goes of resources across levels of to administration costs – this is the highest in ASEAN, government equal with the Lao PDR, and about the same as the amount of spending overall on public hospitals. The Cambodian public administration remains Again, this may reflect a tendency to over-employ highly centralized, even 20 years after the administrative staff in Phnom Penh, at the expense initiation of the government’s decentralization and of severe shortages of trained technical or front-line deconcentration program. The lack of qualified staff in rural areas, where the need is greatest. technical and frontline personnel has been noted in rural Cambodia, while central offices are heavily The relationship between national and staffed. In the irrigation sector, for example, each subnational levels requires further clarification. province has only about 25 staff, less than half than For example, the central health ministry transfers would be required to run a large irrigation scheme budget to the provincial health ministry under the (40,000 ha or more). Inadequate staffing is used heading “support to provinces,” and the provincial to explain the slow pace of deconcentration in the level ministry is supposed to allocate the money to sector. different program headings. However, inconsistency in practices at provincial level undermines the Similarly, within the healthcare sector, there comparability of different provincial reports and the are sharp disparities in outcomes between the ability of the central ministry to determine where richest and the poorest, and between urban money is being spent across the country. and rural populations. The average distance to a health center in rural areas is high, and few doctors In the irrigation sector, decentralization of are available in rural public health centers. Overall, functions has been very limited, because of the spending on healthcare in Cambodia is higher as a retention of 92 percent of overall spending at Cambodia Public Finance Review - From Spending More to Spending Better 71 the central level. This contributes to an imbalance peers. Cambodia ranked 85 out of 120 countries on in personnel: 50 percent of the ministry’s staff are budget transparency score54 in 2021. Figure 2.68 located in Phnom Penh, with the other 50 percent and Figure 2.69 show that budget transparency spread across 25 provinces. Often, the provincial score for Cambodia has slightly improved to 33 out level is contracted to carry out work for the center, but of 100 in 2021, up from 15 score in 2010. Despite responsibility for planning the work and managing the the improvements, budget transparency score for finances is not devolved. The resources most often Cambodia remained below the required sufficient allocated directly to the provincial level are ad hoc transparency score (61). In 2021, Cambodia’s budget funding for flood response. Technical guidelines as to transparency score was the lowest compared to what tasks constitute regular and cyclical operations ASEAN peers, except Myanmar. Cambodia’s budget and maintenance are needed to standardize transparency score was about half of Indonesia’s (70) operations in a way that would allow delegation of and Philippines scores (68). Its score was lower than the full responsibility to provincial authorities. global average (45). Cambodia performed adequately on audit oversight (score 72 out of 100) but remains 2.5. Cambodia performance weak in terms of legislative oversight (score of 25 out on anti-corruption of 100). Figure 2.70 shows that budget oversight by legislative and audit institution was limited, averaging Corruption can have negative impact on public oversight scores of 41 out of 100. sector performance. It can reduce government revenue, create leakages in available resources that Cambodia scored low on index of public integrity could be utilized for delivery of public goods, and compared to world average, countries of reduce public expenditure efficiencies, resulting poor similar development level and regional average quality of infrastructure and public service delivery.53 (Figure 2.71). Incidence of corruption is reported It is therefore useful to examine how Cambodia among 37 percent of public service users (Figure perform on anti-corruption work compared to its 2.72). According to transparency international, 40 Figure 2.68.Transparency in Cambodia Figure 2.69. Transparency score for compared to its peer countries (2021) Cambodia over time (2010-2021) 100 Cambodia’s Global Average 45 ranking:85 of 120 Suf cient countries Indonesia 70 Philippines 68 61 Timor-Leste 52 Malaysia 47 Insuf cient Vietnam 44 32 33 Cambodia 33 20 15 15 Myanmar 30 8 0 0 100 0 Insuf cient 61 Suf cient 100 2010 2012 2015 2017 2019 2021 Source: International Budget Partnership, Open budget survey (2021) 53 International Monetary Fund, why worry about corruption? 1997 54 The Open Budget Survey (OBS) is the world’ s only independent, comparative and fact-based research instrument that uses internationally accepted criteria to assess public access to central government budget information. The OBS measures public access to information on how the central government raises and spends public resources. 109 equally weighted indicators were used to score each country on a scale of 0 to 100. A transparency score of 61 or above indicates a country is likely publishing enough material to support informed public debate on the budget. 72 Cambodia Public Finance Review - From Spending More to Spending Better Figure 2.70. Budget oversight in Cambodia (2021) Legislative oversight Audit oversight 0 25 100 0 72 100 weak adequate KEY 0-40: Few 41-60: Limited 61-100: Adequate Source: International Budget Partnership, Open budget survey (2021) Figure 2.71. Index of public integrity for Cambodia (2020) Cambodia IPI Components vs Averages Cambodia World avg Income avg Regional avg 7 6 5 4 3 2 1 0 IPI Budget Admin Online Judicial Freedom e-Citizenship transparency transparency services independence of the press Source: Index of Public Integrity (IPI) Note: The IPI score is the mean of the six components scores, which result from the standardization and normalization of original source data to range between 1 and 10 using a min-max-transformation, with higher values representing better performance. For Budget Transparency, last value available is for 2019. For Online Services, last value available is for 2020. For Judicial Independence, last value available is for 2019. For the E-citizenship sub-components, last values available are also for 2020, and missing values in any of the sub-indicators were replaced with the latest available data point. percent of public users paid bribes for IDs, followed 2.6. Conclusion and by 38 percent bribes to police, 29 percent to utilities, recommendations 24 percent for public clinics and health centers, The Government has been able to enlarge 18 percent in public schools. Fourteen percent of people surveyed perceived that police officials fiscal space, maintain fiscal sustainability, were involved in corruption, followed by judges and and increase spending across the sectors in magistrates (11 percent), local government officials Cambodia, from 20.2 percent of GDP in 2011 to (9 percent), central government officials (8 percent) 27.9 percent in 2021. Thanks to improved domestic and member of parliament (6 percent) (Figure 2.73). revenue, the government has been able to support Further strengthening Cambodia’s performance on civil service compensation reform and to finance anti-corruption can help to improve performance COVID-19 relief packages. However, the economic of key public policies toward better public service contraction in 2020 combined with large spending on delivery and equity from public spending. COVID-19-induced healthcare and social assistance Cambodia Public Finance Review - From Spending More to Spending Better 73 Figure 2.72. Percentage of public services Figure 2.73. Percentage of people who users paid a bribe (2020) think that people in these institutions involved in corruption (2020) 45% 16% 40% 14% 40% 37% 38% 14% 35% 29% 12% 11% 30% 25% 24% 10% 9% 8% 20% 18% 8% 6% 15% 6% 10% 4% 5% 2% 0% Overall Public Public IDs Utilities Police 0% bribe rate schools clinic and Police Central Local Members Judges health government government of and centers of cials of cials parliament magistrates Source: Transparency International (2020) led to a substantial increase in fiscal deficits and The overall fiscal system somewhat reduces rising debt levels reaching 33.7 percent of GDP as of inequality. While Cambodia does reduce inequality end-2022, from 28.1 percent of GDP in 2019. through taxes and transfers, the degree of redistribution is small in international comparison. Public spending quality, however, needs to be Further improvement in targeting poorer provinces, significantly improved. Increased spending was such as for the healthcare sector, can help to improve significantly driven by salary increases across effectiveness of spending. board for government officials. The wage bill as a share of GDP nearly doubled over the last decade, Although overall budget credibility is high, there an increase from 4.3 percent of GDP in 2011 to 7.2 is significant variation amongst ministries. percent in 2021. Without introduction of performance The annual government’s budget at aggregate measurement, salary increases cannot ensure level is credible with budgeted amounts approved improved performance and hence, the quality of by the National Assembly that are much closer to spending. Average monthly public sector pay is higher the actual amounts spent for the overall budget, than private sector, reaching KHR 1.50 million in 2019, typically coming within +/-5 percent except for compared to KHR 1.25 million for private sector. 2020 and 2021; however, a closer look at ministry level suggested a substantial difference between The social sector expenditure significantly budgeted recurrent and actual expenditure, for increased from 4.3 percent of GDP in 2011 to example 116 percent for MME and 151 percent for 10.1 percent 2021 while the spending in other MPTC for 2021. In the economic and infrastructure sectors were stable. Key social sectors including sector, government capital investment tended to be education and health saw substantial increases over executed over the same period, around 383 in budget allocations. The increase in spending is percent for MAFF, 231 percent for MOWRAM, 166 broadly in line with the government expenditure percent for MRD and 426 percent for MPWT. This policy, which prioritized the social sector. Across the implies a need to strengthen budget planning. sectors, capital expenditures relied mainly on external funding. More domestic financing for capital budget Moving toward performance-informed is necessary as external funds for infrastructure have budgeting requires further refinement of BSP tended to decline. Improvement in public investment and revisiting the current program budgeting management is necessary to achieve quality from structure. Expenditure by program currently does public spending. not capture all program costs including personnel 74 Cambodia Public Finance Review - From Spending More to Spending Better cost and capital expenditure, making it difficult to increase in teacher salaries. This reinforces the assess performance. BSP across ministries could be findings and recommendations of the 2019 PER, further strengthened to improve realism and ensure which made three recommendations for education achievability of target set. sector (i) improving program budgeting by delegating authority to budget managers, (ii) addressing uneven Improvements in human resource management distribution of teachers, and (iii) enhancing student and the introduction of performance-based learning outcomes. While efforts have been made, management and strengthening accountability there has been limited achievement in relations to mechanisms could help to improve the quality these earlier recommendations (See Table 2.4). of service delivery. For example, in the education sector, uneven distribution of teachers and inability Low utilization of public health facilities to address teacher surpluses and shortages and preference for the private sector, which continue to impact quality education. Without an contributes to high out-of-pocket expenditure, accountability framework in place, students learning reflects the general perception of quality of outcomes deteriorated despite the significant public health facilities and is impacting equity Table 2.4. Summary of recommendations for improving education outcome from the 2019 PER Areas of Recommendations What has been achieved and remaining challenges reform Flexibility of • Delegation of authority MOEYS prepares BSP based on guidance from MEF. school using to budget managers to Three-year rolling PB is prepared. DP funded projects funds facilitate expenditure are included as part of BSP. execution Challenge: (1) Budget managers have limited • Assign personnel costs to authority in expenditure execution; (2) personnel specific programs cost is capture under only Program 5 and not • Include DP-funded projects allocated in respective program. in the BSPs Progress towards school- School-based management has been introduced and based management gradually rolled out. Challenge: Delegation of authority to school in managing resources has been limited. Addressing Expand the coverage and Effort has been made to improve enrollment rate in enrollment funding of the current student primary school. gap between grant programs to reduce the Challenge: There is still a gap between urban and rich and poor opportunity cost of schooling rural areas in education enrollment, especially for secondary school between 2014 and 2021. Addressing Reallocate teachers to nearby • Sub-decree # 102 dated 2002 set out one time uneven schools through monetary allowance for redeployment; MOEYS’ teacher norm distribution of incentives and allowance for was introduced in 2018. teachers short distance travels Challenge: Shortages and surpluses of teachers continues to be a key challenge for schools in Enforce student-teacher ratios delivering education services. Improving Increase parents’ engagement Effort has been made to strengthen teachers’ student in improving student learning accountability by engaging parents in improving learning outcomes student learning outcomes under school based outcomes management pilots. Enhancing quality of teaching Challenge: many public schools continue to and reducing teacher absence experience limited accountability from teachers. Sources: PER (2019) and World Bank staff Cambodia Public Finance Review - From Spending More to Spending Better 75 of outcomes. Major issues that have impact on strengthening project pre-appraisal and appraisal improving technical efficiencies should be priorities procedures and guidelines; strengthening to be addressed: (1) health services quality remained independent review and oversight; upgrading project inadequate due to limited resources, specifically, prioritization and budgeting procedure; and framing in terms of the number of health professionals, PPP upstream processes under a unified framework diagnostics, and treatment; (2) inadequate capacity with PIM); strengthening PIM downstream project of the public health system to deal with diseases and budget execution (including project implementation illnesses; (3) investments on health infrastructure and procurement, establishing effective mechanism including technology and information system for monitoring project implementation, clear project remained limited; and (4) inappropriate healthcare adjustment procedure, and introduction of completion seeking behavior of the population, especially review and ex-post evaluation); and upgrading PIM those in the rural/remote areas who delay seeking database and build capacity. While progress has healthcare, and who engage in self-medication.55 been made with regard to upstream process (as seen in the adoption of PIM Sub-decree and PIM Quality of infrastructure remains poor and reform strategy) and upgrading of PIM database Cambodia’s logistics performance declined. with monthly report being transformed into shared Paved roads as a share of total road networks in Cambodia remained the lowest in ASEAN between database that automates production of consolidated 2015 and 2020. Similarly, Cambodia ranked lower monitoring report, required capacity for downstream than most ASEAN neighbors, except Myanmar, on implementation appeared limited. It is necessary to the logistics performance index. build required capacity for implementation in line with the guidelines (see Table 2.5). Strengthening PIM is necessary for improved quality of investment in infrastructure. Key This PFR recommends a number of key policy options recommendations in the 2019 PER pointed to considered the most important for improving the the need for upgrading PIM upstream project quality of public expenditures. Table 2.6 summarizes appraisal (including upgrading PIM legal framework; proposed policy options. Table 2.5. Summary of recommendations on PIM from 2019 PER Areas for PIM Key What has been achieved and remaining challenges reform recommendations Strengthen Upgrade PIM • Improved PIM legislation: PIM subdecree adopted; however, PIM upstream strategy and this sub-decree does not include climate change to support project appraisal planning with a PIM resilient infrastructures. sub-decree • Public Investment Management System Reform Strategy (PIMSRS) 2019-2025 adopted. • Fiscal risk statement developed as part of the MTFF document; however, MTFF is not yet formally adopted by Council of Ministers. Strengthen project • Standard criteria adopted in Sub-decree# 41 ANKr.BK pre-appraisal provided guidelines on project rationality, economic and and appraisal financial viability, project options, affordability principles, deliverability principle, and project implementation. procedures and • Public investment committee (PIC) has been formed to act as guidelines gatekeeper and harmonize all public investment sources. • Ensuring the quality of project appraisal remain a challenge. 55 National Voluntary Review 2019 76 Cambodia Public Finance Review - From Spending More to Spending Better Areas for PIM Key What has been achieved and remaining challenges reform recommendations Upgrade project • Projections of project cost for major investment included in prioritization Public Investment Plan attached to budget document. and budgeting • Mechanisms to give funding priority to ongoing domestically procedures funded projects exist. • Routine maintenance included in budget, but capital maintenance (infrastructure renovation, reconstructions and enlargements) is not included in any budget documents. Frame the • Law on Public Private Partnership adopted in 2021 and PPP upstream various guidelines. processes under a • Policy Paper on the Development of PPP Mechanism for unified framework Public Investment Project Management 2016-2020 issued in with PIM 2016. • Limited PPP information in budget documents. Strengthen PIM Strengthen project • Projects are required by Law to adopt competitive and open downstream implementation procurement. project/budget and procurement • Required PIM guideline including for domestic fund have execution been adopted. • No requirement for monitoring or for the independent review of complaints; lack of database to monitor key steps in the tender process. Develop clear • Several Line Ministries (LMs) have multi-year projections of project adjustment capital spending in Budget Strategic Plan (BSP). procedures and • Capital funding are annually appropriated for domestically guidance and externally financed project. • There may be shifting in budget prioritization resulting in reduced capital investment. Introduce • Department of Investment (DI) verified domestically financed completion review project completion and reported to MEF on monthly and and ex-post quarterly basis. evaluation • Some portfolio oversight has been conducted while ex post audits of individual projects not systematically required. Upgrade PIM Build a PIM • IT systems are being developed to support macro-fiscal database and database (CAPEX) statistics management, performance management, and asset build capacity to register public registry; FMIS supports budget and treasury functions. investment projects • Asset monitoring system is primarily used for and processes controlling asset sales rather than as an input for decision making throughout the public investment management cycle. Sources: World Bank’s 2019 PER, PIMA (2019), PEFA (2022) Cambodia Public Finance Review - From Spending More to Spending Better 77 Table 2.6. Summary of key selected policy options Medium-term Short-term recommendations Challenges recommendations (1-2 years) (3-5 years) Weak link between Formally adopting a medium term fiscal and Ensure full budget discipline planning and expenditure framework to serve as the basis for across all line ministries budgeting development planning Absence of realistic Setting realistic targets for program outcomes by Establishing robust M&E target for program establishing working groups in ministries to identify systems to monitor outcome realistic targets and costing them out achievement of targets Public expenditure Linking government spending with performance • Making sure that not linked to outcomes by: programs are adequately outcome • Revisiting and refining BSP to define outcomes financed reflecting their indicators and performance information that are costs realistic on the ground and comparable across • Fully implement reporting period performance-informed • Allowing for program directors to have authority budgeting over resource allocation and be accountable for the performance outcomes • Reviewing and restructuring program budget across the sector to support linkage between budget allocation and performance outcomes Challenges in Further strengthen project pre-appraisal and Build capacity for all capital budget selection ministries to implement PIM management and Build capacity for selected key ministries involving in line with the adopted PIM public investment legislation and guidelines large public investment management (PIM) Salary increase Adopt national program for public administration Introduce performance- not contributed reform and based management for all to improved • Improve HR management information across civil servants and link salary performance the government by upgrading the HRMIS increase to performance • Introduce performance-based management for mid-management level (general department and department level) across all sectors and link salary increase to performance Inadequate • Review and appropriately cost transferred • Adequate resources resources (finance function and development of a concrete (finance and human and human deconcentration plan for each service delivery resources) provided resources) at ministry to subnational subnational levels • Establishing an M&E framework to monitor administrations to perform progress of deconcentration plan toward their functions to performed transferred improving service delivery at the subnational • M&E framework functions and levels implemented deliver services 78 Cambodia Public Finance Review - From Spending More to Spending Better 3 MAXIMIZING EFFECTIVENESS OF PUBLIC HEALTH POLICY 3.1. Introduction In 2022, there was a total of 1,408 public sector health facilities, which account for less than Decentralization of health service delivery, social 10 percent of the total health service providers health protection schemes, and distribution of across Cambodia, with the majority of health resources at the facility level are among the key care providers being private sector. Between challenges for health service delivery in Cambodia. 2017 and 2022 the number of Health Centers High out-of-pocket expenditures, persistent health (primary care level) increased from 1,190 to 1,288. inequities, and low utilization rate suggest rather Utilization rates for public health centers are quite limited effectiveness of public health policy. low, with an average of 0.46 contacts per capita 3.2. Sectoral context annually. Private health care providers, which include approximately 16,000 registered providers and Cambodia Ministry of Health (MOH) implements 3,400 pharmacies, outnumber public health facilities its policies through three general directorates: (i) by over tenfold, accounting for 16,185 facilities in General Directorate for Health, (ii) General Directorate 2022. Most privately registered health care providers for Administration and Finance, and (iii) General are for-profit entities. Directorate for Inspection. In addition, since 2019, Decentralization and Deconcentration (D&D) reforms 3.2.1. Health outcomes in the health sector have transferred service delivery and management functions to the provincial/capital Alongside sustained economic growth, administrations. Subsequently, responsibilities for Cambodia has seen steady improvements in overseeing health service delivery also fall under the health outcomes in recent decades. Between jurisdiction of the provincial/capital administrations 2016 and 2020, Cambodia had achieved one (Box 3.1). out of 14 key health sector related Sustainable Cambodia Public Finance Review - From Spending More to Spending Better 79 Box 3.1. Decentralization and Deconcentration reforms in the healthcare sector The 2001 Law on Commune and Sangkat Administrative Management provided the basis for decentralization and local-level democracy for public services (including health) in Cambodia, with local elections held in February 2002. In 2005 the Royal Government of Cambodia adopted the Strategic Framework of Decentralization and Deconcentration, setting out the government’s vision and objectives for expanding the governance of public services to sub-national levels. In 2008, the MOH began adopting the D&D reform framework, with 38 ODs (for roles and responsibilities of operational districts (ODs) (see Annex to Chapter 3, Box A.3.1 for more detail) becoming Special Operating Agencies (SOAs) (see Annex to Chapter 3, Table A.3.3), with the aim of improving the quality of health services delivery at the subnational level. In December 2019 Sub-decree #193, the “Assignment of Health Management Function and Health Service Delivery to the Capital and Provincial Administrations” was signed. The aim of the Sub-decree # 193 was to improve health service delivery by bringing decision making and responsibility for service management closer to end-users. This included 1) the transfer of responsibility from the MOH to capital and provincial level and a change from vertical administration and leadership to an integrated approach, with 2) elected capital and provincial councils responsible for priority and agenda setting, and preparing and implementation of annual budgets. The health functions transferred included management and service delivery functions of provincial health departments (PHD) and OD (Annex to Chapter 3, Table A.3.3); complementary package activities (CPA) (Annex to Chapter 3, Box A.3.3 ) services of provincial referral hospitals’; and minimum package of activities (MPA) (Annex to Chapter 3, Box A.3.3) services of health centers/health posts. In practices, provincial health budget is recorded separately from the provincial administration budget allocation. Accountability framework: provincial/capital administrations supervise and are accountable for the performance of capital/provincial health care services and PHDs (under supervision of MOH). PHDs supervise and monitor the performance of provincial hospitals and ODs. ODs, in turn, supervise and monitor the performance of district referral hospitals and HC/HP. Source: Sub-decree # 193 dated December 4, 2019 on “The transfer of management functions and health services delivery to provincial/capital administration” Note: MOH=Ministry of Health; MOI=Ministry of Interior; PHD=Provincial Health Department; OD=Operational Districts; MPA= Minimum Package of Activities; CPA=Complementary Package of Activities; HC=Health Center; HP=Health Post. Health budgets allocated to provincial administrations are based on health function transfer as part of the decentralization process. Development Goal 3 (SDG3)56 indicators—preventing doctors and nurses) and other essential inputs for new HIV infections. Maternal, new-born and child health service delivery. At the same time, Cambodia’s mortality indicators are on track to be achieved. 57 economic transformation presents challenges for the Cambodia has improved health outcomes in some health system through the increased morbidity and areas but continues to lag behind most other ASEAN mortality from non-communicable disease and the countries. The country’s health sector suffers from dual burden of disease (communicable and non- a lack of infrastructure, human resources (such as communicable diseases) which is threatening to 56 SDG (Sustainable Development Goals) aims to end poverty, protect the planet and ensure that all people enjoy peace and prosperity. There are 17 SDGs. SDG3 aims to promote healthy life and well-being for all. Cambodia committed to achieve SDG 3 targets by 2030. In 2018, the government approved CSDG in which targeted indicators of SDG3 were set- including improving financial protection risks, maternal mortality, newborn and children mortality, ending epidemics, reducing mortality from non-communicable diseases and from traffic accidents, improving reproductive health etc. 57 Sustainable development report 2022 80 Cambodia Public Finance Review - From Spending More to Spending Better exacerbate poverty and exert an enormous financial Figure 3.1. Maternal mortality ratio cost to health care and society. Maternal mortality ratio (modeled estimate, per 1,200 There have been positive gains in maternal and child health outcomes over the past two 1,000 decades. Life expectancy from birth rose from 100,000 live births) 800 67.7 years in 2010 to peak at 70.7 years in 2019, before slightly declined to 69.6 years in 2021. 600 Maternal mortality has decreased from 351 deaths 400 per 100,000 live births in 2005 to 154 in 2021.58 Childhood immunization coverage has expanded to 200 cover 76 percent of children ages 12–23 months fully 0 vaccinated in 2021 compared to 40 percent in 2000. 2.5 3.5 4.5 5.5 The prevalence of contraceptive (any method) use GDP per capita, PPP (current intl. US$), log among married women has increased steadily, from Source: World Bank Development Indicators (2022) 24 percent in 2000 to 62 percent in 2021. Over this same period, the use of any modern method of family quintile (richest) of income and the lowest quintile planning has climbed from 19 percent to 45 percent. (poorest) reduced for both antenatal care received These trends have contributed to a steep decline from skilled providers and the number of births in infant and under-five mortality rates. Neonatal delivered by skilled providers between 2005 and mortality has significantly decreased from 37 deaths 2015. In 2005, women in the top quintile (richest) per 1,000 live births to 8 in 2021. Infant mortality had almost twice the chance to have antenatal dropped from 95 deaths per 1,000 live births in 2000 care received from a skilled provider compared to 12 in 2021; under-five mortality dropped from to those women in the lowest quintile (poorest). In 124 deaths per 1,000 live births in 2000 to 16 in 2015, approximately 9.9 out of 10 women in the 2021. Additional discussion on health outcomes are top quintile received antenatal care from a skilled presented in Annex to Chapter 3 Box A.3.2. provider as opposed to 9 out of 10 women in the lowest quintile. In 2005, only 20 percent of women in The maternal mortality rate is low, on par with the lowest quintile had a birth delivered by a skilled wealthier countries (Figure 3.1). Coverage of provider, as compared to 90 percent of women in antenatal care from skilled providers has significantly the highest quintile. In 2020, this gap has almost expanded over the last 15 years. The percentage disappeared, with 95 percent of women in the of women who received antenatal care from skilled lowest quintile receiving a birth delivered by a skilled providers reached 99 percent in 2020, a significant provider, relative to 100 percent of women in the increase from 69.3 percent in 2005. The percentage highest quantile (Figure 3.2). of births delivered by skilled providers doubled within this period, from 43.8 percent in 2005 to 90 percent Life expectancy has steadily improved over the in 2021, while the percentage of births delivered at a last two decades and reached relatively a high health facility increased from 21.5 percent in 2005 to level compared to peers with similar incomes 98 percent in 2020. (Figure 3.3 and Figure 3.4). However, access to quality health care remains constrained, particularly Cambodia remarkably narrowed the gap in for the poor and vulnerable groups, and this has access to maternal health care between the deteriorated further due to COVID-19. Vaccination richest and poorest. The gap between the highest coverage (for all basic antigens) in 2021 ranged from 58 Cambodia Demographic and Health Survey (CDHS) 2021-22 Key Indicators Report, June 2022. Cambodia Public Finance Review - From Spending More to Spending Better 81 Figure 3.2. Access to maternal health services, gap between poorest and richest (2005-2021/22) Antenatal care received from skilled provider Birth delivered by skilled provider Birth delivered in a health facility 60% 40% 20% 0% 2005 2010 2015 2021/22 Sources: CDHS (2005-2021/22) and World Bank staff estimates Note: CDHS=Cambodia Demographic and Health Survey Skilled provider includes doctor, nurse, and midwife. Data on antenatal care received from the skilled provider was not yet available in CDHS 2020 (selected indicators) Gap=utilization rate of poorest -utilization rate of richest. The higher the figure means the wider the gap is. 65.6 percent among the poorest quintile to 79.6 Cambodia lags behind other ASEAN countries in percent among the highest, and similar comparisons the number of qualified medical doctors per 10,000 are found for other key services such as treatment population. In 2021, Cambodia had the lowest for childhood respiratory and diarrheal illnesses . 59 number of doctors per 10,000 population, four times Communicable diseases combined with maternal, lower than Vietnam and almost five times lower than perinatal and nutritional deaths accounted for 26 Thailand. Cambodia has comparable rates for nurses percent of all deaths in the country in 2016. 60 and midwives per 10,000 population, with Lao PDR, Myanmar, and Vietnam but this remains much lower The low levels of qualified health professionals than other ASEAN countries. impact population health outcomes (Figure 3.5). Figure 3.3. Life expectancy relative to Figure 3.4. Cambodia improved life income (2020) expectancy and mortality rates for both infants and children under-5 90 140 80 Life expectancy at birth, total (years) deaths per 1,000 live births 120 70 80 60 100 Mortality rate, infant Mortality rate, under-5 50 70 80 years Average life expectancy (RHS) 40 60 30 60 40 20 20 10 50 2.5 3.0 3.5 4.0 4.5 5.0 5.5 0 0 GDP per capita, PPP (current intl. US$), log 2000 2005 2010 2014 2020 Source: World Bank Development Indicators (2022) Sources: CDHS and Cambodia population census Note: RHS = Right-hand scale 59 CDHS 2021-2022 60 World Health Organization (2018). Noncommunicable diseases country profiles 2018. World Health Organization. https://apps.who. int/iris/handle/10665/274512. License: CC BY-NC-SA 3.0 IGO 82 Cambodia Public Finance Review - From Spending More to Spending Better Figure 3.5. Density of doctors, nurses, and midwives in ASEAN countries 70 60 50 40 30 20 10 0 Cambodia Lao PDR Indonesia Myanmar Philippines Vietnam Thailand Brunei D Malaysia Singapore Medical doctor (per 10,000 population) Nurse and midwife (per 10,000 population) Sources: MOH, WHO Global Health Observatory data depository (2022), and World Bank staff estimates Note: MOH=Ministry of Health; WHO=World Health Organization Cambodia’s medical doctor per 10,000 population and nurse and midwife per 10,000 population were compiled from MOH data taken from the 2021 annual report. Deaths from non-communicable diseases are continue to experience in the near future (HFSA, on the rise (Figure 3.6 and Figure 3.7). In many 2020). countries, as the socio-economic status improves, there is also an increase in non-communicable 3.2.2. Health sector strategies and diseases due to lifestyle and environmental changes. social health protection schemes. Cambodia is no exception, with the pace of transformation presenting a challenge for the health The government implemented several policy system. However, the challenges associated with initiatives aimed at expanding access to health traditional infectious diseases and MCH problems services in the 1990s. Building on the experiences continue to be significant, generating the “dual in the 1990s, a Sector-Wide Management (SwiM) burden of disease” (Figure 3.7) that Cambodia will approach was introduced in 2000 and mobilized Figure 3.6. Deaths from non-communicable diseases are relatively high and on the rise a. Cause of death, by non-communicable diseases b. Cause of deaths (2010-2019) (2019) Cause of death, by non-communicable diseases 100 3,500 3,043 DALYs per 10,000 population 90 3,000 2,789 2,575 80 2,500 2010 2015 2019 70 2,000 (% of total) 60 1,500 1,267 1,106 50 955 882 1,000 812 785 40 571 534 561 500 30 0 Communicable Maternal, Non-communicable Injuries 20 diseases neonatal, diseases 2.5 3.5 4.5 5 nutritional conditions GDP per capita, PPP (current intl. US$), log Source: World Bank Development Indicators (2022) Source: Global Burden of Disease Study (2019) Note: DALYs estimates the potential years lost due to premature death and includes equivalent years of healthy life lost due to poor health or disability; combines mortality and morbidity into one metric. Cambodia Public Finance Review - From Spending More to Spending Better 83 Figure 3.7. Total burden of disease in Cambodia in DALYs Rate 2000 and 2019 a. Top ten cause of DALYs (2000) b. Top ten cause of DALYs (2019) Neonatal conditions 7,733.4 Neonatal conditions 2,626.4 Lower respiratory infections 7,382.5 Stroke 2,071.6 Diarrhoeal diseases 5,062.3 Lower respiratory infections 1,841.1 HIV/AIDS 3,935.6 Cirrhosis of the liver 1,755.3 Tuberculosis 2,278.8 Ischemic heart disease 1,544.5 Stroke 2,012.8 Road injury 1,138.7 Cirrhosis of the liver 1,818.7 Tuberculosis 1,004.7 Congenital anomalies 1,376.2 Diabetes mellitus 901.1 Ischemic heart disease 1,241.6 Congenital anomalies 789.5 Drowning 1,216.1 Diarrhoeal diseases 769.8 0 2,000 4,000 6,000 8,000 0 1,000 2,000 DALYs per 10,000 population DALYs per 10,000 population Source: Global Burden of Disease Study (2019) resources from the government and development covered all public health facilities in the country. The partners (DPs) to address priority areas in the sector. Health Strategic Plan (HSP) 2023-2030 prioritizes In 1997, user fees for health services were introduced (i) improving the health of the general population, (ii) at public health facilities. The current guideline for the strengthening resilience and modernizing the health use of revenue from user fees notes that 60 percent system, and (iii) strengthening governance and of the revenues are allocated for incentives for health regulatory mechanisms. facility’s staff, 39 percent for facility operating costs, and 1 percent going to the National Treasury. The While out-of-pocket spending remains high, Health Equity Fund (HEF) (Box 3.2) was introduced Cambodia has taken important steps to expand in 2000 to increase access to health services and social health protection (SHP) schemes (Table financial health protection for the poor. By 2015, it 3.1). Government programs to improve financial Table 3.1. Enrollment in different social health protection schemes Scheme Year Est. Contributory Coverage (August 2023) NSSF health insurance for private sector formal 2017 Yes 1,412,714 workers NSSF voluntary health insurance for self- 2023 Yes 40,000* employed NSSF health insurance for public sector 2018 Yes 439,243 employees Health Equity Fund 2012 No 3,214,920* Health Equity Fund for at-risk households 2023 No 1,096,000** HEF extension for informal workers and special 2018 Yes 96,259 beneficiary categories Source: Government data and authors’ analysis *Approximate number as of September 2023 ** Planned HEF expansion to “at-risk” households: approximately 311,000 “at-risk” households (1,096,000 individuals) were identified in late 2022 through targeting measures for Cash Transfer programs (Floods, High Inflation, and Newly Identified and will likely be enrolled in HEF in the near future. 84 Cambodia Public Finance Review - From Spending More to Spending Better protection for health, such as the Health Equity Fund approximately a third of the Cambodian population (HEF) and the National Social Security Fund (NSSF), is covered by a Social Health Insurance Program have shown notable growth in recent years. As of (SHP) program, health expenditures through such end of 2022, SHP schemes in Cambodia covered programs account for only 3 percent of total health around 38.7 percent of total population. The NSSF expenditure. provides coverage for formal private and public sector employees. As of 2022, a total number of 3,023,949 HEF has been vital to improving access to health people are covered under the NSSF schemes. care and financial protection for the poor (Box In August 2023, the government introduced a 3.2). However, expanding the scheme’s coverage health insurance scheme on a voluntary basis for to new groups (such as the near poor or children dependents of NSSF members and self-employed. under two) will require additional resources. HEF- The government is looking to extend NSSF health supported outpatient cases numbered 2,811,981 insurance to certain uncovered population groups, in 2021, with inpatient utilization reaching 206,817 such as farmers, university students, TVET trainees cases in 2021, a 13 percent increase from 183,795 and migrants, etc., while the HEF covers health care cases in 2018. However, expanding the scheme costs for the poorest quintile of the country, with a will demand resources. For example, the estimated coverage of approximately three million people with cost (using HEF payment rate) for expanding HEF IDPoor. By 2022, HEF covered a total of 3,214,920 to cover all targeted beneficiaries (approximately 3.5 people. The government is extending HEF to at-risk million people) between 2020 and 2025 is estimated households/near poor61 of 1,834,801 in September at around US$66.4 million. The cost is projected to 2023, bringing HEF coverage to 5 million people increase to US$87.9 million when applying the same or 31 percent of the population (Table 3.2). While payment rates as NSSF.62 Table 3.2. Access to social health protection CSES Year None HEF NSSF Other Total 2009* Above 93.2% - - - - 2014 ** 89.5% 9.4% - 1.1% 100% 2019/20 75.2% 9.6% 14.6% 0.7% 100% 2023*** 43.0% 31% 25.0% 1.0% 100% Source: Analysis of financial protection in Cambodia’s health system using Cambodia Socioeconomic Survey data: 2009 – 2021. Phnom Penh: General Secretariat for the National Social Protection Council, Ministry of Economy and Finance Cambodia, and Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH; 2023. *2009 CSES data does not unambiguously identify all the households that have access to some form of social health protection, but from the data, at least 93.2 percent of the population does not have access to social health protection (whether from a public initiative, community health insurance or private health insurance). Access to SHP is not included in the analysis of financial protection for CSES 2009. ** NSSF’s health insurance commenced in 2016, there is no data for households with access to NSSF in the 2009 and 2014 datasets. ** 2023 access to social health protection is projected based on data from NSPC. HEF coverage is projected to reach 5 million (31 percent of population. NSFF health insurance coverage is projected to reach 4 million, an increase of 1 million. 61 At risk households are those who are near-poor having at least one of vulnerabilities: either low level of per capita consumption as proxie by IDPoor, or household with at least one of the following health vulnerabilities: elderly, children under 2 years olds, all members under 18 years olds, or female single head of households. 62 Kolesar et al. Advancing Universal Health Coverage in in Cambodia, 2021. Cambodia Public Finance Review - From Spending More to Spending Better 85 The benefit package also needs to be revisited expanding population coverage under its social to address the evolving burden of disease health protection schemes. The government has and ensure service providers have sufficient launched a process titled the “UHC Roadmap” that resources to provide appropriate care. For aims to improve social health protection and move example, the current HEF reimbursement rate for towards universal health coverage in the country, diabetes (KHR 4,000) would need to increase at least which is expected to be approved in 2023. Envisioned eightfold to cover the costs of supplying one month intervention measures include (i) Expanding of medicine for diabetes patients (estimated at KHR population coverage from existing schemes (NSSF 30,000) (Table 3.3). Due to the current reimbursement & HEF); (ii) Improving service coverage through rate and resource constraints at the health facilities, strengthening licensure, implementing accreditation HEF patients, who already face financial constraints systems, and practicing strategic purchasing; and (iii) to cover transportation costs, are often requested to Improving financial coverage through channeling out return to facilities on a weekly basis to receive their of pocket spending into coverage mechanisms. treatment. The challenge for integrating all existing social Cambodia has committed to advancing protection schemes includes (i) establishing an universal health coverage (UHC) which requires administration and governance body to oversee a Box 3.2. Health Equity Fund HEF is a key health financing scheme introduced by the government in 2000, aimed to increase health financial protection for the poor. HEF purchases health care for the poor from public health facilities using a performance-based payment mechanism. Payment for HEF user fees is made directly by MOH to public health facilities. HEF covers all health facilities across the country. In 2021, HEF costs reached US$ 18.81 million, of which 68 percent was covered by MOH’s budget.63 HEF covered two beneficiary groups: (i) ID Poor pre-identified by the Ministry of Planning and (ii) Priority Access Card for a beneficiary group that do not have ID Poor cards but are post identified as poor eligible for as a HEF beneficiary. HEF covers 100 percent user fees for in-patient and out-patient care. The table below describes the services covered by HEF at health facility level. National hospitals District hospitals Health centers &CPA1-3 hospitals IPD OPD IPD OPD OPD User fees Yes Yes Yes Yes Yes Transport Yes Yes Delivery of baby, No Delivery of baby, reimbursement attempted delivery, and attempted delivery, and post abortion care only post abortion care only Care taker food Yes No Delivery of baby, No No support attempted delivery only Funeral Yes Yes No No No support Source: WB, Impact Evaluation of Service Delivery Grants to Quality of Health Care Delivery in Cambodia (2022) Note: HEF=Health Equity Fund; IPD=Inpatient Department; OPD=Outpatient Department; Yes= services covered; No=services not covered. 63 2021 MOH annual performance report 86 Cambodia Public Finance Review - From Spending More to Spending Better Table 3.3. Comparison of benefit packages and payment rate of HEF and NSSF Description of benefit HEF NSSF packages Health benefits • For ID Poor, the benefits include outpatient • In-patient and out-patient and inpatient service fees, transport fees, services, emergency and funeral contribution. services, medical assistant • For informal workers, the health benefits services, delivery pre- cover only outpatient and inpatient service natal and post-natal care, fees. rehabilitation services, transport of corpse. • Services not covered by HEF: (i) cancer treatment by chemical stance, radioactive, and hormone (ii) organ transplantation (iv) beauty surgery (v) acupuncture, (vi) infertility treatment. Payment rates of • General consultation: KHR 4,000 • General consultation: KHR selected services (KHR) • Long-term birth control: KHR 20,000 6,000 • Inpatient treatments at CPA1: KHR 80,000 • Long-term birth control: KHR 30,000 • Birth delivery, abortion/ miscarriage at CPA 1: KHR 80,000 • Inpatient treatments at CPA1: KHR 100,000 • Birth delivery, abortion/ miscarriage at CPA 1: KHR 100,000 Source: MOH and NSSF Note: MOH=Ministry of Health; HEF= Health Equity Fund; NSSF=National Social Security Fund; KHR=Khmer Riel, CPA= Complementary package of activities unified health protection scheme, (ii) harmonizing challenges can be characterized by both demand health benefits packages of all health protection and supply constraints which include financial schemes; the benefits package and payment rates resources, limited physical access to health facilities, are different between HEF and NSSF (Table 3.3), and inadequate pool of trained healthcare workers, and (iii) ensuring government sufficient budget to finance availability of medical equipment and medicine. health care under the unified health scheme. Challenge #1: High out-of-pocket 3.2.3. Highlighted health sector expenditure challenges Cambodia’s out-of-pocket spending is high Despite notable improvements in some compared to other ASEAN countries. In 2019, health outcomes, Cambodia’s health system Cambodia’s out-of-pocket spending as a percentage continues to face a number of challenges. First, of total health spending was the second highest in the system is characterized by high out-of-pocket (OOP) payments due to low levels of public financing ASEAN countries with Myanmar the highest. Globally, and a propensity for use of private sector healthcare Cambodia ranks the 15th highest in the world. Out- providers. Second, inequities in healthcare outcomes of-pocket spending accounted for 64.4 percent of still occur between rural and urban areas despite the total health expenditures in Cambodia, in contrast narrowing in income differentials. Third, access and with 8.7 percent in Thailand, 43 percent in Vietnam, utilization rates for public health facilities, and the 41.8 percent in Lao PDR, and 40 percent across quality of those facilities, remains low. Overall, these other lower-middle income countries (Figure 3.8). Cambodia Public Finance Review - From Spending More to Spending Better 87 Figure 3.8. Cambodia’s health spending and out-of-pocket expenditures as compared to other ASEAN countries (2019) Health spending per capita (US$) 700 Government health spending (% of Total Health Spending) (RHS) 90% Out-of-pocket spending (% of Total Health Spending) (RHS) 80% % of health spending 600 70% 500 60% US$ 400 50% 300 40% 30% 200 20% 100 10% 0 0% Myanmar Lao PDR Cambodia Indonesia Philippines Vietnam Thailand Malaysia Brunei D Sources: WHO, Global Health Expenditure Database (2022) Note: RHS= Right-hand scale; US$=United States Dollar Health spending per capita in Singapore accounted for US$ 2,633; government health spending as a share of Total Health Spending and Out-of-Pocket spending as a share of Total Health Spending represented 50.2 percent and 30.2 percent, respectively. High out-of-pocket spending means the poor population pushed below, or further below the are more vulnerable to income shocks and are poverty line, because of health OOP. In Cambodia, less likely to have access to health services. close to 17 percent of the population spends 10 In 2013 the highest quintile of income (richest) percent or more of their budget on health, and this spent 16 times more on healthcare than the lowest includes over 6 percent spending 25 percent or quintile (poorest), being more able to afford to pay more of their budget on health OOP. The incidence for healthcare services owing to larger disposable of catastrophic health payments is typically higher non-food income. 64 Achieving universal health for the poorest households (24 percent for the first coverage will necessitate larger public resources quintile against 8 percent for the richest 20 percent) (government and/or development partners funded) (Figure 3.9). to expand health coverage to the poorer segments of the population. Most of OOP health expenditures Turning now to the effect of health expenditure in Cambodia are spent at the private health care on poverty, OOP contributes to pushing, on providers. average, 4 percent of the population under the US$5.5 poverty line, and another 15 percent of The incidence of financial hardship is high the population already in poverty are pushed in Cambodia, and concentrated towards further below the poverty line. While catastrophic the poorest, threatening progress towards and impoverishing health expenditure capture two achieving UHC and towards faster progress different dimensions of financial hardship, some in poverty reduction. Two types of indicators are households are exposed to both dimensions of typically used to measure financial hardship due to financial hardship, especially among the poorest 20 large health spending: i) the incidence of catastrophic percent of the population. One dimension through payments, or the share of the population spending which we see important variations in the incidence more than a given share of their budget (10 percent and in the composition of financial hardship is or 20 percent) on health OOP, and ii) the rate of the household demographic structure. Indeed, impoverishment due to OOP, or the share of the households with elderly members, especially 64 WHO, The Kingdom of Cambodia Health System Review (2015) 88 Cambodia Public Finance Review - From Spending More to Spending Better Figure 3.9. Expenditure on health and catastrophic health spending, by quintile (2019) 30 Health share (%) Catastrophic 10% Catastrophic 20% 25 20 % of household 15 10 5 0 Q1 Q2 Q3 Q4 Q5 Wealth quintile Sources: CSES (2019); Authors’ analysis Note: Health share (percent) means out of pocket expenditure on health as a share of total expenditure; Catastrophic 10 percent means catastrophic health occurs when a household’s out-of-pocket spending equal or exceed 10 percent of household’s total consumption; Catastrophic 20 percent means catastrophic health occurs when a household’s out-of-pocket spending equal or exceed 20 percent of total household consumption. Wealth quintiles are ranked from Q1 (poor) to Q5 (wealthy), based on ranges of mean household income and grouped into five categories with each quintile assigned to approximately 20 percent of the population. multigenerational households, are particularly Enrollment in existing social health protection exposed to financial hardship in Cambodia. The schemes was found to have a limited impact on share of out-of-pocket spending of total consumption financial protection (Figure 3.11). The CSES 2019 is highest amongst the poorest (7 percent). also highlighted the level of coverage by insurance Catastrophic health expenditures were found to schemes across socio-economic quintiles. While the be relatively high at 16.8 percent at the 10 percent prevalence of ID Poor cards was higher in the poorer threshold and 6.5 percent at 20 percent threshold quintiles and use of NSSF card is higher in wealthier for the total sample, and even higher amongst the quintiles (Figure 3.12), the existence of ID Poor cards poorest quintile, with 24.4 percent at the 10 percent in the third and fourth quintiles was still notable and 10 percent of respondents at 20 percent of total (ranging from 5-10 percent of respondents) perhaps spending. In other words, in Cambodia, close to 17 due to targeting issues of the ID Poor scheme. A percent of the population spends 10 percent or more comparison of reported health care needs, and out- of their budget on health (Figure 3.10). of-pocket spending across ID Poor beneficiaries, and the general population (non-ID Poor), shows Figure 3.10. Financial hardship composition, by quintile (left); by family structure (right) Financial Hardship Composition Financial Hardship Composition Financial hardship composition Financial hardship composition Catastrophic (10%) and Impoverishing ($5.5) OOP Catastrophic (10%) and Impoverishing ($5.5) OOP 100 100 80 80 Percentage 60 Percentage 60 40 40 20 20 0 Adult only Adults Adults, Adult Old only <= 19 only 0 and kids kids and and old Poorest Poorer Middle Richer Richest old No hardship Catastrophic only (10%) No hardship Catastrophic only (10%) Catastrophic and impoverishing Impoverishing only ($5.5) Catastrophic and impoverishing Impoverishing only ($5.5) Source: CSES (2019) Source: CSES (2019) Cambodia Public Finance Review - From Spending More to Spending Better 89 that while the reported use of health care was higher Figure 3.11. Impoverishment from health among ID Poor cardholders, the rate of out-of- expenditures, ID Poor card holders and pocket spending was the same for the two groups non-ID Poor holders (2019) (around 64 percent of cases). 35 30 The ID Poor cards seem not to provide much ID Poor Not ID Poor 25 % of household financial protection, with similar prevalence of 20 catastrophic spending across the ID Poor and non-ID Poor respondents. Among ID Poor card 15 holders, 18 percent of respondents spent more 10 than 10 percent of total consumption on health care 5 while 16.6 percent did among non-ID Poor. At the 0 Impovrishment Further Impovrishment Further 20 percent of spending threshold the distribution is US$3.65 impovrishment US$5.5 impovrishment US$3.65 US$5.5 similar with 8 percent of ID Poor card holders vs. 6.3 Sources: CSES (2019); Authors’ analysis percent non-ID Poor. Finally, impoverishment rates Note: the share of households who are impoverished or further remain high for ID Poor, but lower than Q1, with Q1 impoverished after out-of-pocket payments measured using a relative poverty line reflecting basic needs (food, housing, including almost 25 percent of ID Poor cardholders. utilities). Analysis was conducted at the US$3.65/day and Unfortunately, while HEF and NSSF schemes seem US$5.50/day poverty lines to have contributed to incentivizing service utilization among the covered population, the effectiveness of significantly declined in both urban and rural areas, these schemes in providing protection from financial however this decline was not equal. Rural areas hardship due to health OOP seems limited. continue to experience higher mortality rates for infants and children under five. Although there has Challenge #2: Inequity in health outcomes been a decrease in mortality for infants and children persist under five, rates are still much higher among the Despite child mortality rates significantly poor. In 2021, infant mortality in the poorest quintile declining over the last two decades the disparity was 22 deaths per live 1,000 births compared to in health outcomes across rural and urban 6 per 1,000 in the richest quintile and mortality for areas and wealth quintiles remains a challenge. children under 5 in the poorest quintile 21 per 1,000 As noted in chapter 2, between 2005 and 2021, the live births comparted to 7 per 1,000 live births in the rates of mortality for infants and children under-5 highest quintile. Figure 3.12. Existence of insurance cards by income quintile ID Poor, NSS, and subsidized health care, by expenditure quintile 25 Percentage of population 20 15 10 5 0 Poorest Poorer Middle Richer Richest ID Poor card Subsidized health care NSS card Sources: CSES (2019); Authors’ analysis 90 Cambodia Public Finance Review - From Spending More to Spending Better Figure 3.13. Percent distribution of household members who were ill or injured in the past 30 days and sought health care for the first treatment, by place of treatment, by urban and rural residents (2021) 90% 83.9% 84.7% 84.2% 80% 74.4% 70% 60% 50% Total Phnom Penh Other urban Rural 40% 30% 24.9% 20% 15.3% 14.4% 15.0% 10% 0.6% 0.0% 0.7% 0.6% 0.2% 0.7% 0.2% 0.2% 0% Public health care Private health care Non-medical health care Outside provider provider provider country/other Source: CSES (2021) Note: Public care includes national hospital, provincial hospital, district hospital, health center, health post, provincial or community based rehabilitation center, and other public service. Private care includes private hospital, private clinic, Private pharmacy, visit to home/office of trained health worker/nurses, visit to trained helath worker/nurses, other private medical service, shop selling drugs/ market. Non-medical health care includes Kru Khmer/magician, monk/religious leader, traditional birth attendant. Challenge #3: Limited utilization rate of transportation costs. On average, the distance from public health facilities and challenges in a household to public health facilities is estimated quality of care at around 3.2 km, with one-way transport costs approximately KHR 3,614 (US$0.9); a relatively high Utilization of health care services provided and potentially prohibitive cost for poorer and rural by public health facilities is still low. There is a communities.67 preference to use private sector facilities 65 in both urban and rural areas, ranging from 74.4 percent There is a low density of trained health workforce in Phnom Penh to 84.2 percent in rural parts of the in rural areas. In 2022, on average there were 7.4 country, with around 25 percent of care seekers staff allocated per health center,68 close to MOH going to the public sector in Phnom Penh, 14.4 guidelines of 8 to 11 staff per health center. However, percent in other urban areas and 15 percent in rural according to the government’s reports, 63 percent areas (Figure 3.13).66 Key factors that contribute to of health centers in rural areas and 35 percent of the limited utilization of public health services are health centers in urban areas are understaffed. The described below. geographical distribution of doctors across provinces Patients have limited physical access to health is more heavily weighted in urban rather than rural facilities. The distribution of health facilities may areas, and with most doctors working in tertiary not always reflect the distribution and needs of the settings. There is a very low number of doctors population. This can include the lack of specialist working in health centers. While the national average services in many areas, such as mental health or ratio of doctors per health center is 1 to 5, the ratio is 1 oncology, and physical access to services due to to 6 for rural centers and 1 to 1 in urban areas (Figure 65 Private sector facilities include private health care provider and pharmacies. List of private health care providers and pharmacies is shown in Table A.3.2 66 CSES 2020 67 WB, Impact evaluation of service delivery grants to improve quality of health care delivery in Cambodia (2022) 68 In 2022, total staff at health center accounted for 9,483 while the number of health center reached 1,269. Cambodia Public Finance Review - From Spending More to Spending Better 91 3.14).69 When health professionals are available, lack of routine data collected on the adequacy and they may not possess adequate training or skills for quality of consumables and commodities that are different health needs. For example, there remain essential to health service delivery, such as medical a limited number of public sector facilities offering equipment and medicines. However, a survey screening and treatment for NCDs such as diabetes of health workers found that less than half (41.9 and hypertension (approximately 20 percent), largely percent in 2019) were satisfied with the availability due to gaps in training needs for health workers. In of drugs at the facilities.73 However, 71 percent of addition, some households in rural areas seek care surveyed health workers showed strong satisfaction from untrained/non-medical providers (8.6 percent with the quality of drugs when they were available of rural households versus 0.3 in Phnom Penh and at the facilities. Issues of availability of medicines 6.5 percent in other urban areas).70 Private providers in public facilities contributes to high levels of out- have similarly low competencies with only 54 percent of-pocket spending (81.4 percent of surveyed of private providers having formal health training.71 households in 2019) continue to bear out-of-pocket There have been multiple efforts by the government expenses on health care, with the key driver of costs to improve the quality of pre-service training and being medicines. On average, households spent education which aims to improve the quality of health approximately KHR 33,436 (US$8.2) on medicines professionals in Cambodia. Patient satisfaction has for a two-week treatment.74 A similarly low rate been reported to be higher in the private sector, of satisfaction was found around the availability with private healthcare providers providing better of medical equipment, with only 47.7 percent of communication and more consistent healthcare surveyed health workers expressing satisfaction with services for patients.72 the condition and availability of health equipment at health care facilities.75 This is compounded by a Basic medical equipment and essential lack of medical equipment maintenance sites; and medicines are not always available. There is a a limited capacity of health facilities to undertake Figure 3.14. Health facilities and doctors at facilities by province (2022) 120 140 Number of health facilities 100 Hospital 120 80 HC 100 Number of doctors Medical doctors at hospital 80 60 Medical doctors at health center 60 40 40 20 20 0 0 Kep Pailin Mondul Kiri Koh Kong Stung Treng Ratanakiri Preah Vihear Oddar Meanchey Preah Sihanouk Kratie Pursat Svay Rieng Kampong Chhnang Kampot Kampong Thom Tbong Khmum Banteay Meanchey Kampong Speu Kampong Cham Takeo Battambang Siemreap Prey Veng Kandal Phnom Penh Sources: MOH (2022) and World Bank staff estimates Note: HC = Health Center 69 MOH guideline suggests an allocation of at least one physician and one physician assistant per health center. 70 Cambodia Socio-Economic Survey 2019/2020 71 Cambodia Health Financing System (September 2020) 72 World Bank, Cambodia’s Rural Health Markets Study and Quality of Care Study (2013) 73 World Bank, Impact Evaluation of Services Delivery Grants to Improve Quality of Health Care Delivery in Cambodia (2022) 74 Ibid. 75 Ibid. 92 Cambodia Public Finance Review - From Spending More to Spending Better maintenance. The lack of maintenance at provincial social health protection schemes (HEF and NSSF). hospitals means that without repair many of the This creates provider incentives for over-servicing. donated medical equipment items are left unused. Consequently, the use of private health providers is often associated with higher levels of OOP. While The corollary of supply-side constraints is an HEF enrollees are entitled to receive free care in adverse impact on the quality of health care public health facilities, currently private providers are provision. Between 2016 and 2019, health workers’ excluded from the program’s benefit package. On advice for women to undertake laboratory standard the other hand, NSSF beneficiaries are entitled to tests during pregnancy, showed no improvement in the use of public health facilities and a selection of a protocol-based vignette assessment. For instance, over 120 private health facilities. The use of private a low percentage of surveyed health workers pharmacies has grown dramatically over the past recommended standard laboratory tests, such as decade, particularly among poorer households, HCG pregnancy tests (31 percent in 2019), Hgb where use increased from 14 percent in 2009 to 40 hemoglobin tests (23 percent in 2019), urine tests for percent in 2021, surpassing wealthier households of gestational diabetes (38.9 percent in 2019), tests for which only 39 percent used private pharmacies in platelet counts (6.1 percent in 2019). 76 Only one in 2021 (Figure 3.15). three patients is estimated to have been diagnosed correctly, using five basic vignettes tools, and only Implementation of the HEF and NSSF health 82 percent of the correctly diagnosed cases were insurance schemes should have the effect correctly prescribed medical treatment. of increasing the utilization of public health facilities for their target populations. Increased Cambodia’s private health sector continues use of private health services is likely to result in an to grow rapidly and the propensity to use increase in OOP and less financial protection, even private providers, even among insurance in the face of financial protection mechanisms that enrollees, remains high. The use of private health have been introduced. Figure 3.16 below shows the care providers is of particular concern for financial proportion of the population that used different types protection. This is because private services are of health services for the years under assessment. paid through OOP on a fee-for-service basis and Based on the results, there was an initial increase in for the most part do not participate in the existing the use of public health facilities from 2009 to 2014. Figure 3.15. Use of private pharmacies for wealthier and poorer quintiles 50% 40% 39% 40% 30% 27% 26% 25% 25% 21% 20% 14% 10% 0% 2009 2014 2019/20 2021 Quintiles 1 & 2 Quintiles 4 & 5 Sources: Analysis of financial protection in Cambodia’s health system using Cambodia Socioeconomic Survey data: 2009 – 2021. Phnom Penh: General Secretariat for the National Social Protection Council, Ministry of Economy and Finance Cambodia, and Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH; 2023. 76 World Bank, Impact Evaluation of Service delivery grants to improve quality of health care delivery in Cambodia (April 2022) Cambodia Public Finance Review - From Spending More to Spending Better 93 Figure 3.16. Type of health care provider consulted first 90% 75.3% 83.9% 80% 76.5% 69.1% 70% 60% 2009 2014 2019/20 2021 50% 40% 23.5% 30% 20.1% 22.7% 20% 15.3% 7.4% 0.7% 10% 3.4% 0.5% 0.6% 0.6% 0% 0% Public Private Non-medical Overseas Sources: CSES (2009-2021) Note: Public care includes national hospital, provincial hospital, district hospital, health center, health post, provincial or community- based rehabilitation center, and other public service. Private care includes private hospital, private clinic, Private pharmacy, visit to home/office of trained health worker/nurses, visit to trained health worker/nurses, other private medical service, shop selling drugs/ market. Non-medical health care includes Kru Khmer/magician, monk/religious leader, traditional birth attendant. Thereafter, the proportion of those that first sought around 6 percent and remained stagnant. Against health care in the public sector declined, followed by the backdrop of low government expenditure and a clear trend of increasing use of private health care high OOP spending, Cambodia relied on DPs to providers as the first point of contact in the health finance its health care need. However, DP funding system for all income groups. Even those from was on the decline. Third, Cambodia’s health system households with access to HEF and NSSF, between is facing challenges to finance dual diseases burden- 2009 and 2021, there was a reduction in the likelihood communicable diseases and non-communicable of seeking care from a public healthcare provider. diseases. Addressing all these needs will require an This is contrary to expectations given that the HEF expansion of funding. Despite some improvements (and to some degree, the NSSF), should promote in SHP, only approximately a third of population is and increase the use of public health facilities. covered by SHI. However, the social health protection schemes seemed have no impact on reducing OOP 3.3. Health sector financing spending since most of OOP are spent on private health care providers. Improving spending efficiency 3.3.1. Total health sector spending is vital for improved service delivery. There is a need to improve resource allocation, both budget Cambodia is striving to achieve UHC target and human resources. Efficiency of health facility and improve its health service delivery. Both operations and social health protection schemes additional budget allocation and spending need to be strengthened. efficiency are required for a number of reasons. First, Cambodia relied heavily on OOP spending Cambodia’s spending on health care as a share to finance health care. OOP accounted for around of total government spending is relatively low. 60 percent of Total Health Expenditure (THE) while Despite being a priority sector, the government countries of similar development level have an budget allocation on MOH as share of total average OOP spending of around 30 percent. government budget over the last five years slightly Second, government spending on health per decreased from 8.6 percent in 2017 to 7 percent in capita and as a share of total government spending 2020, before it picked up to 8.9 percent in 2021. remains relatively low. Albeit slightly fluctuated, Cambodia spent relatively low levels of government government spending on health as share of total funds on health in both per capita terms and as government spending over the last decades was a share of the government budget, both lower 94 Cambodia Public Finance Review - From Spending More to Spending Better than nearly all nations in the region. This means Over the last decade, external funding for the remaining expenditure is financed by external communicable diseases such as HIV/AIDS, financing and out-of-pocket spending. tuberculosis, and malaria was 46 percent, representing the largest share of total external The health system of Cambodia relies to an funding.77 Programs for child and new born, extent on external (development partners) reproductive and mental health took up the second funding, which is likely to decrease. Traditionally, largest share of external funding (32 percent). This financing from development partners for health was followed by health system strengthening (11 has been allocated largely as earmarked funds for percent), other infectious diseases (2.4 percent), disease-specific national programs such as malaria NCDs (1 percent), and ‘other’ at 10 percent of total control or the HIV/AIDS programs. However, under external funding (Figure 3.17). Despite its importance the Health Equity Improvement Project (H-EQIP), in for health sector financing, external funding has been 2021 development partners in Cambodia contributed on the decline. Between 2008 and 2018, external 32 percent of the finances required to fund the HEF funding as a share of GDP dropped from 1 to 0.42 while the government funded the remaining 68 percent. Similarly, external funding as a share of the percent. Data from the 5-year Cambodia National overall government health expenditure fell from 96.4 Health Account 2012–16 indicates that development percent to 33 percent (Figure 3.18). partners’ funding for health is around US$200 million or 17 percent of current health expenditure The spending on health system administration (CHE). With the continuing growth of Cambodia’s stood at 15 percent, nearly as high as the economy and maturation to middle-income status, spending absorbed by national hospitals (16 it is expected that many development partners percent) (Figure 3.19). Increasing resources are spent will reduce or withdraw their health sector funding on both infectious and non-communicable diseases support. Global health initiatives, such as the Global (Figure 3.20). Spending on non-communicable Fund or GAVI, the vaccine alliance, have already diseases has nearly doubled between 2013 and 2019, requested increased government co-financing. from US$18 to US$34 per capita. During the same Figure 3.17. Share of total external funding Figure 3.18. Total external funding percent by health focus area (2008-2018) of GDP and percent of government health expenditure (2008-2018) 100% 1.25% % of government health expenditure 10.0% 90% 16.2% 80% 1.00% 10.8% 70% 60% 0.75% % of GDP 26.2% 50% 15.9% 40% 0.50% 30% Child and new born health 1.1% 11.2% HIV/AIDS 20% 0.25% 6.2% Malaria 10% 2.4% Tuberculosis 0% 0.00% Other infectious disease Non-communicable disease Reproductive and mental health Swaps/health system strengthening External funding (% of Government Health Expenditure) Other External funding (% of GDP) (RHS) Sources: IHME, Financing Global Health Database, (extracted January 9, 2023) Note: Government health spending refers to healthcare that is derived from domestic sources. Government spending includes spending on public health system infrastructure and government-provided social health insurance. 77 External funding for health forms part of total health expenditure; however, that external funding for health does not include expenditures on health infrastructure. Cambodia Public Finance Review - From Spending More to Spending Better 95 Figure 3.19. Largest share of total Figure 3.20. Current health expenditure by health expenditure was spent on private diseases and conditions, in current US$ healthcare providers per capita, 2013-2019 Health system administration and 60 financing 1% National hospital 2013 2014 2016 US$ per capita (current) 50 4% Provincial hospital 2018 2019 8% 15% 40 District hospital Health center 30 16% NGO hospital 20 31% Private hospital 5% 10 Private clinic 5% 6% Pharmacy/drugstore 0 6% Infectious Non- Reproductive Injuries Other and Nutritional 2% Other private/non-medical and communicable health unspecified Deficiencies parasitic diseases diseases Provider of preventive care/NGOs diseases (NCDs) and conditions Sources: WHO, Global Health Expenditures (2022), National Source: Global Burden of Disease Study (2019) Health Account (2019), and World Bank staff estimates Note: WHO= World Health Organization; NGO=Non-government Organization period, spending on infectious diseases increased decreases between 2017 and 2020. Prior to from US$37 to US$54 per capita. This is consistent the COVID-19 outbreak, MOH budget allocations with the “dual burden of disease”, associated with gradually decreased, from 8.6 percent in 2017 to the continued prevalence of infectious diseases and 7 percent of the total government budget in 2020. rise in non-communicable diseases resulting lifestyle In response to the pandemic, public resources and environmental changes as Cambodia climbs the allocated to MOH increased by 58 percent, from income ladder. KHR 1,729,987 million (1.9 percent of GDP) in 2017 to KHR 2,724,907 million (2.5 percent of GDP) in 3.3.2. Public health spending by the 2021. The government allocated a higher budget to MOH MOH in 2021 as part of its plan to respond to the COVID-19 pandemic.78 Data availability is a constraint to efficiency analysis. It is, for example, difficult to assess the Health infrastructure is primarily financed by extent to which externally financed projects are development partners. Spending by externally able to address the gaps in access to physical funded capital declined to the lowest level in the last infrastructure. Data on domestic spending obscures, five years at 15 percent of total MOH spending in 2018 for example, through the design of program budgets, and that consistently increased to reach 28 percent the degree to which public spending addresses in 2021. DP-financed capital spending accounts supply-side, service quality, and equity challenges in for the dominant share of total capital spending. the sector. However, preliminary analysis of available The MOH capital spending has represented a small data suggests that while spending is pro-poor, it is share of the total MOH spending, except in 2021 not progressive and that resource allocation across when capital investment reached 18 percent of provinces may not be addressing regional disparities. its spending. This spike is explained as part of the broader COVID-19 response. This increased capital Budget allocations to healthcare have spending was meant to cover health infrastructure increased in response to COVID-19, following as well as other expenses related to COVID-19 78 Significant health sector COVID spending was also conducted outside of MOH budget. 96 Cambodia Public Finance Review - From Spending More to Spending Better Figure 3.21. MOH spending, by category (KHR, million) (2017-2021) 3,000,000 Recurrent 2,500,000 Capital, domes�cally financed Capital, externally financed KHR, million 2,000,000 1,500,000 1,000,000 500,000 0 2017 2018 2019 2020 2021 Sources: Budget Settlement Law (2017-2021); World Bank staff estimates Note: MOH=Ministry of Health; KHR= Khmer Riel. Recurrent expenditure excludes Covid-19 spending in 2020 and 2021 and included estimated wage bill at provincial level in 2021. Capital expenditures budget is used to acquire infrastructure fixed assets such as hospitals or health center buildings. (Figure 3.21). Reliance on donor financing for capital in-year capital budget appropriation between 2017- spending has implications on the sustainability of 2021. the provision of health infrastructure and services, as external financing is expected to decline as Figure 3.22. MOH budget out turns Cambodia becomes richer. Against the backdrop of (percent of approved budget by budget type) low levels of government spending, this raises critical Recurrent policy questions related to domestic financing for 180% Capital, externally financed both the continued provision of health infrastructure 160% 153% % of approved budget 140% as well as the resources to operate and maintain it. 132% 120% 100% 100% 93% Externally funded capital investments are 80% less predictable relative to recurrent and 60% 40% 44% domestically financed capital spending. While 20% the aggregate level of recurrent spending broadly 0% 2017 2018 2019 2020 2021 mirrors the amounts planned in the budget (Annex to Chapter 3, Table A.3.5 reveals detailed recurrent Sources: Budget Law (2017-2021) and Budget Settlement Law (2017-2021); World Bank staff estimates budget outturn), donor-financed capital spending Note: Budget outturns over 100 percent of the approved fluctuates between over and under-spending relative budget reflect actual spending exceeding the planned budget to what was budgeted for between 2018 and 2020 (overspent); under 100 percent means actual spending was below the planned budget (underspent). (Figure 3.22). Two reasons are considered to explain The recurrent spending excluded Covid-19 expenditures. this. First, budgeting for capital spending can be overly optimistic and includes funds that are yet to Expenditure by functions be committed by donors. Second, DPs’ funding disbursement can be delayed due to procedural The increase in government spending in nominal misalignment with government processes, including terms between 2016 and 2019 was driven by different procedural timelines across the project cycle. spending on curative care and health system Domestically financed investment was not presented administration.79 Over the period, there was a in the Budget Law; however, it is reflected in MOH’s 33 percent increase in spending on administration, 79 Nominal figures are PPP adjusted Cambodia Public Finance Review - From Spending More to Spending Better 97 while spending on curative care rose by 92 percent. while spending on health service delivery This has resulted in a shift in the composition of (particularly preventive care) remains very low government health expenditure, with curative care as a total share of public expenditures.80 As accounting for a larger share. In 2019, spending noted in Chapter 2, in 2019, Cambodia’s spending on curative care represented 61.9 percent of total on health system administration as a share of government health expenditure, up from 55.2 percent government health expenditures (35 percent), was in 2016. Meanwhile, the proportion of expenditure on twice as high as in Myanmar, more than four times health system administration decreased from 44.5 to higher than in Vietnam, and nearly three times higher 34.6 percent during the same period (Figure 3.23). than in Thailand and Malaysia. On the other hand, as a share of government health expenditures, Spending on health system administration Cambodia’s public spending on curative care (61.9 remains high compared to ASEAN countries, percent) is only second lowest to Lao PDR, and lower Figure 3.23. MOH spending on health administration gradually reduced, but its share of total government expenditure remained higher than its peer countries in ASEAN Cambodia’s health spending by health care functions (left) and percent of government spending (right) (2016-2019) 2,000,000 80% % of government health expnediture 55% KHR, million 1,500,000 60% 61% 62% 60% 40% 1,000,000 20% 45% 35% 35% 35% 500,000 0% 2016 2017 2018 2019 0 Government expenditure on preventive care 2016 2017 2018 2019 Government expenditure on curative care Government expenditure on preventive care Government expenditure on governance, health system & Government expenditure on curative care financing administration Cambodia’s health spending as compared to its peer countries in ASEAN, by health care functions, ( percent of government health expenditure) (2019) 100% 4% 3% 1% 2% 1% 7% 7% 90% 10% 11% 7% 15% % of government budget Government expenditure 80% on other health care 70% 62% Government expenditure 60% 51% on preventive care 50% 66% 84% 75% 76% 40% Government expenditure 30% on curative care 20% 35% 35% Government expenditure 10% 17% on governance, health system 8% 13% 12% 0% & financing administration Cambodia Lao PDR Myanmar Vietnam Thailand Malaysia Sources: WHO-Global Health Expenditure (2022) and World Bank staff estimates Note: MOH=Ministry of Health Government health expenditures from 2016 to 2019 (graph a) are PPP-adjusted nominal values. Other healthcare includes rehabilitative care, long-term healthcare, ancillary services, medical goods, and other healthcare services. 80 Further examination may be necessary given some challenges in budget classification in Cambodia. 98 Cambodia Public Finance Review - From Spending More to Spending Better than Myanmar, Vietnam, Thailand, and Malaysia (65.5 at Provincial Health Department- PHD or Operational percent, 84.4 percent, 74.6 percent, 75.9 percent District-OD level (for description of PHD and OD, see respectively). When expenditure on preventive Annex to Chapter 3, Box A.3.1). Finally, introducing care is measured as a share of government health this type of provider payment modality could also expenditure, Cambodia was the lowest. The share be a way to gradually crowd out OOP (at least the of government spending on preventative care is the portion directed to public facilities). lowest among comparator countries. Expenditure by economic classification There should be a transition towards higher prioritization for preventive care, including the MOH spending appear well-balanced between introduction of policies to increase targeted wage and non-wage expenditures. The nominal prevention measures where health spending wage bill increased by around a third (or 6 percent is high and institutionalization of incentives for annually) between 2017 and 2021, reaching KHR people to seek preventive services. This could 522,135 million (0.47 percent of GDP) in 2021 from be achieved by introducing some form of capitation KHR 396,566 million (0.44 percent of GDP) (Figure in the provider payment mechanisms. Transitioning 3.24). Wage-related spending accounted for around from line item and performance payment (Service 30 percent of total spending in 2021. The increase Delivery Grant-SDGs81) to a form of blended in the wage bill was driven by salary increases as capitation with performance payment could help to the total number of MOH staff remained constant improve the allocative efficiency of public expenditure on health and the equity of public spending in health. until 2020. For instance, in 2020, the annual starting This could also be a way of improving technical salary for a doctor increased from KHR 12 million in efficiency by increasing the providers incentives to 2007 to KHR 16.3 million due to changes in base focus on preventive versus curative service delivery salary and functional allowances (Figure 3.25). Figure 3.24. MOH spending, by economic Figure 3.25. Starting salary for doctors function (2017-2021) (KHR/year) (2017-2023) 20,000,000 2,500,000 Wages 18,000,000 Non-wage recurrent 16,000,000 2,000,000 Capital, domestically financed 14,000,000 KHR/year 12,000,000 KHR, million 1,500,000 10,000,000 8,000,000 1,000,000 6,000,000 4,000,000 500,000 2,000,000 0 2017 2018 2019 2020 2021 2022 2023 0 2017 2018 2019 2020 2021 Basis salary Functional allowance Sources: MEF and World Bank staff estimates Note: MOH=Ministry of Health; MEF=Ministry of Economy and Finance; KHR=Khmer Riel; RHS=Right Hand Scale. Data on MOH staff and a starting salary for doctors in 2022 and 2023 are projections by World Bank staff. It is assumed that the number of staff will increase by one percent from 2022 to 2023. Salaries and allowances are paid from the MOH budget. Basis salary is determined by a civil service rank. Each doctor is appointed at a particular civil servant rank in accordance with criteria set out by a royal decree. Doctors also receive a functional allowance which is determined by their positions and their education levels. 81 Service Delivery Grant-SDG refers to payment from the government to health centers, or hospitals as a form of fixed lumpsum grant or performance-based payment. Cambodia Public Finance Review - From Spending More to Spending Better 99 Figure 3.26. Program 4 (Strengthen Health System) dominated MOH program budget allocation Program-based budgets, by program in million KHR (left) and percent of total (right) (2017-2022) 2,000,000 100% % of total program budget 80% 1,500,000 KHR, million 60% 92% 92% 91% 91% 92% 93% 1,000,000 40% 500,000 20% 0 0% 2017 2018 2019 2020 2021 2022 2017 2018 2019 2020 2021 2022 Program 1 Program 2 Program 3 Program 4 Program 1 Program 2 Program 3 Program 4 Source: MEF Note: MEF=Ministry of Economy and Finance; KHR=Khmer Riel Program 1=Reproductive, Youth, Maternal, Newborn and Child Health and Nutrition; Program 2=Communicable disease control; Program 3=non-communicable disease control and other public health issues; Program 4=Strengthen Health System Expenditure by program Expenditure at the provincial level Program 4 (Strengthening of Health System) Provincial budgets account for a third of total dominates the program budget, raising MOH spending and are dominated by payroll. questions about the program design and MOH recurrent spending at the provincial level allocative efficiency. Since 2017, budget increased from 0.5 percent of GDP in 2017 to 0.6 allocation for Program 4 consistently accounted for percent in 2020 and declined to 0.5 percent in 2021. over 91 percent of the total program budget (Figure These consistently account for over a third (around 3.26). Other programs account for a minor share 0.6 percent of GDP) of the total subnational spending of expenditure. In 2022, 5.3 percent was allocated during the same period. Provincial spending are to Program 1 (Reproductive, Youth, Maternal, dominated by staff salaries with the wage bill (Chapter Newborn, and Child Health and Nutrition), 1.3 64). The wage bill in the provinces increased by 33 percent was allocated to Program 2 (Communicable percent between 2017 and 2021—from 0.3 percent disease control), and 0.07 percent was allocated to of GDP to 0.4 percent of GDP—and accounted Program 3 (Non-communicable disease). However, for over two-thirds (79 percent) of the total MOH this does not reflect the true program costs. The recurrent spending at the provincial level (Figure large share of spending in Program 4 is inflated by 3.27). At the program level, they predominantly focus the consolidation of all the wage expenses. This on programs to address non-communicable disease pattern is also observed in other ministries, where control (Program 3) and broader health system wage expenses are not linked to a given program strengthening (Program 4) (Figure 3.28). However, that staff are meant to support, but is rather as indicated earlier, the inability to track expenses grouped into one program – often along with other across programs may not provide the true program administrative expenses. This suggests weaknesses cost. in program design with few resources appropriated to the programs or in the budgeting and accounting Expenditure at the provincial level reflects systems that are unable to reflect the actual level of staffing and population size. Figure 3.29 illustrates resources flowing to achieve objectives under the a close association between the distribution of first three Programs. health facility staff and the corresponding spending 100 Cambodia Public Finance Review - From Spending More to Spending Better Figure 3.27. Program budgets are centralized Recurrent budget, by level of administration Program budget, by program across level of (percent of total) (2017-2022) administration ( percent of total) (2020) 100% 100% 9% 9% 80% 40% 41% 40% 41% 41% 36% 80% 39% 44% % of recurrent budget % of recurrent budget 60% 60% 91% 91% 40% 40% 60% 59% 60% 59% 59% 64% 61% 56% 20% 20% 0% 0% 2017 2018 2019 2020 2021 2022 Program 1 Program 2 Program 3 Program 4 Central (ministry) Provincial Central (ministry) Provincial Sources: MEF and World Bank staff estimates Note: Central includes MOH ministerial administration, national hospital and national center. Program 1=Reproductive, Youth, Maternal, Newborn and Child Health and Nutrition; Program 2=Communicable disease control; Program 3=non-communicable disease control and other public health issues; Program 4=Strengthen Health System Figure 3.28. MOH spending at provincial level (2017-2021), by economic classification (percent of total; left) (KHR, million; right) 700,000 100% 600,000 90% KHR, million 80% 500,000 70% 400,000 60% 78.8% 300,000 50% 200,000 40% 30% 100,000 20% 0 10.2% 2017 2018 2019 2020 2021 10% 10.7% Chapter 60: Purchases Chapter 61: Services 0% Chapter 64: Payroll expenses Chapter 62: Social benefits expenses 2021 Chapter 65: Susidy expenses Chapter 63: Taxes and tarrifs Sources: MEF and World Bank staff estimates Note: MEF=Ministry of Economy and Finance; KHR=Khmer Riel across the provinces. This finding is expected, as a of targeting public resources, however, significant portion of spending is allocated towards aggregate analyses suggest budget allocations salaries, which would be expected to be higher in across provinces may not be addressing provinces with a larger number of health workers. regional disparities. The evaluation of targeting Staffing decisions in turn reflect the population size of quality is hindered by data availability constraints, a given province with more doctors working in more which prevent the inclusion of more up-to-date populous provinces (Figure 3.30). Consequently, a health outcome data—as well as data at the DMK positive correlation is observed between population and CS levels—to evaluate the equity of access size and the share of budget that is allocated to a within provinces. However, correlations across 25 province (Figure 3.31). provinces between the size of the recurrent budget against poverty levels and health outcomes suggest Data availability limits an efficiency assessment that government resources may not be effectively Cambodia Public Finance Review - From Spending More to Spending Better 101 Figure 3.29. Recurrent spending (percent of total) 2021 vs distribution of staff at health facilities (2022) vs distribution of population (2020) by province 8% % of budget/% of staff at health facili�es % of total budget 16% 7% % of staff at health facili�es 14% 6% % total population (RHS) 12% % of population 5% 10% 4% 8% 3% 6% 2% 4% 1% 2% 0% 0% Kep Pailin Mondulkiri Stung treng Koh Kong Rattanak Kiri Preah Vihear Oddor Meanchey Pursat Kampong Chhnang Kratie Kampong Thom Svay Rieng Kampong Speu Kampot Thbong Khmum Banteay Meanchey Phnom Penh Siem Reap Takeo Kampong Cham Battambang Kandal Prey Veng Preah Sihanouk Sources: MEF, MOH (2022), CSES (2019/20) and World Bank staff estimates Note: MEF=Ministry of Economy and Finance; MOH=Ministry of Health; CSES=Cambodia Socio-Economic Survey; RHS=Right- Hand Scale The staff at health facilities do not include those at the newly established hospital such as Prek Prasab referral hospital, Sambo referral hospital, and Thbong Khmum provincial referral hospital. The budget allocation at the provincial level covers PHD, OD and health facilities in the provinces; that excludes budget allocation for regional health training centers in the provinces. Figure 3.30. Medical doctors in provinces Figure 3.31. Recurrent spending in vs population in provinces (2019) provinces (% of total) vs population in provinces (2019) 2,500,000 250 2,000,000 200 r² = 0.9237 Toal population Number of doctors 1,500,000 150 r² = 0.4422 1,000,000 100 500,000 50 0 0 2% 3% 4% 5% 6% 0 500,0001,000,0001,500,0002,000,0002,500,000 % of spending Population size Sources: MEF, General Population Census (2019) and World Bank staff estimates targeting some of the poorer provinces. Nonetheless, 3.3.3. Efficiency and equity of health without more detailed data on the functional level of spending spending and within-province targeting, the analysis cannot determine whether the level of spending When looking at the distribution of MOH targets the different needs across provinces. expenditures by level of health facility/health Furthermore, the analysis presents only a partial view administration, most of the resources flow to of spending that excludes spending by the central- level ministry as well as capital investments, which health centers, albeit with variations across are critical to addressing sector challenges. provinces.82 Figure 3.32 illustrates facility-level 82 This includes government budget expenditures allocated to provinces through PHD. It does not include insurance and user fees. Main expenditures for PHD are purchases and services. For hospital and HC, main expenditures are salaries and social benefits expenses. 102 Cambodia Public Finance Review - From Spending More to Spending Better Figure 3.32. Spending by level of health Figure 3.33. Spending in provinces, by facilities (2017-2021) economic classification, by level of health facilities (2021) 100% 100% 36% 35% 34% 33% 31% 80% 80% 43% 55% 68% 76% 73% 60% 60% 16% 17% 18% 19% 20% 4% 40% 31% 40% 17% 17% 17% 17% 17% 35% 20% 19% 8% 17% 13% 14% 14% 14% 14% 20% 21% 2% 9% 3% 10% 7% 7% 6% 18% 17% 17% 16% 17% 0% 4% 0% PHD Provincial Operational District Health 2017 2018 2019 2020 2021 hospital district hospital center/health Chapter 64: Payroll expenses post Provincial health department Provincial hospitals Chapter 63: Taxes and tariffs Chapter 62: Social benefits expenses Operational districts District hospitals Chapter 61: Services Health centers/health posts Chapter 60: Purchases Source: MOH (2022) Source: MOH (2022) Note: PHD=Provincial health department, OD=Operational district, Note: PHD=Provincial health department, OD=Operational district, HC=Health center HC=Health center In Preah Sihanouk province, OD is embedded in PHD. Only one In Preah Sihanouk province, OD is embedded in PHD. Only one hospital exists in the province. hospital exists in the province. spending across a sample of 17 provinces.83 On Income from patient fees remains a relatively average, spending on health centers and health small source of financing for all types of posts has declined from 36 percent to 31 percent facilities. Costing data from a representative as a share of general spending in the provinces, but sample of 60 facilities in 201984 provides insights. still accounts for the largest share of MOH resources. First, revenue from patients is a small fraction relative There are, however, variations across provinces. to government transfers, ranging from 11 percent For example, in 2021 Kandal and Preah Sihanouk at health centers to 25 percent at CPA3 hospitals provinces spent 41 percent and 46 percent on (Figure 3.34). A further breakdown shows that CPA1 health centers, respectively, while spending on hospitals were also the least dependent on user health centers in Battambang has been considerably fees as a share of patient revenue, with the largest lower despite the high number of health facilities in source being HEF payments. Health centers’ patient the province. For provincial health departments, the revenue was largely from user fees, followed by main expenditure includes purchases and services, NSSF at normal health centers and HEF at health represented respectively 21 percent and 31 percent, centers with beds. With regards to categories while wages accounted for only 43 percent. In of income, contributions or transfers from the hospitals, wages accounted for around two-thirds government, direct cash transfers such as fixed lump of total expenditures (Figure 3.33). The second sum grants (FLSG) and performance-based grants biggest expenditure item is on social benefits. For (PBG), were introduced to provide additional flexible health centers, wages were the major expenditures, financing for providers to improve the quality of care accounting for 73 percent of total expenditures, and overall functioning of facilities, and are a small followed by social benefit expenses (17 percent), fraction of transfers from government ranging from services (6 percent) and purchases (4 percent). 2-10 percent for FLSG and 1-4 percent for PBG of 83 Spending at health facility level in the 17 provinces, including Banteay Meanchey, Kampong Cham, Kampong Chhnang, Kampong Speu, Kampong Thom, Kandal, Koh Kong, Kratie, Phnom Penh, Prey Veng, Preah Sihanouk, Steung Treng, Svay Rieng, Takeo, Oddar Meanchey, Pailin, and Tboung Khmum. 84 Source: NIPH, 2021; GIZ estimates from the second round of health services costing study: results from 60 public health facilities (unpublished) Cambodia Public Finance Review - From Spending More to Spending Better 103 Figure 3.34. Sources of financing by facility level, 2019 Sources of income, by type of facilities (US$) Sources of income, by type of facilities (percent (2019) of total) (2019) 6,000,000 100% 90% 5,000,000 80% User fees 70% 4,000,000 60% 80% 80% 75% US$ HEF 89% 87% 3,000,000 NSSF 50% Other 40% 2,000,000 30% Government income 1,000,000 20% 10% 20% 20% 25% 11% 13% 0 0% HC HCwB CPA1 CPA2 CPA3 HC HCwB CPA1 CPA2 CPA3 Types of facilities Patient fees Government income Sources: GIZ (2021); GIZ estimates from the second round of health services costing study: results from 60 public health facilities (unpublished) Note: HC=Health center; HCwB= Health center with beds; CPA= Complementary Package of Activities Figure 3.35. Breakdown sources of income by types of facilities (2019) Patient fees income, by type of facilities Government budget income, by type of facilities (% of total) (2019) (% of total) (2019) 100% 100% 90% 90% 80% 80% 70% % of total 70% 60% % of total 60% 50% 50% 40% 40% 30% 30% 20% 10% 20% 0% 10% HC HCwB CPA1 CPA2 CPA3 0% Types of facilities HC HCwB CPA1 CPA2 CPA3 Payroll expenses Drugs and materials Types of facilities Depreciation Allowance Other government grants Fixed lump-sum grants User fees HEF NSSF Other Performance-based grants Other income (Vaccine) Sources: GIZ (2021); GIZ estimates from the second round of health services costing study: results from 60 public health facilities (unpublished) Note: HC=Health center; HCwB= Health center with beds; CPA= Complementary Package of Activities government transfers, respectively. Grants had the financial reports, but also to improve the efficiency of highest proportion in CPA1 hospitals. Drugs and their operations by reducing waste or unnecessary materials, followed by salaries and wages, had the expenditures, and by doing better planning for all highest proportion across all types of facilities (Figure available resources. 3.34 and Figure 3.35). Health facilities are accountable for a variety of There has been a growing demand for public health funds. These include user fees, HEF, Service Delivery facilities to better manage the financial resources Grants (SDG) (Annex to Chapter 3, Table A.3.3) which entrusted to them so as to enable them to have is comprised of PBG and FLSGs, funds claimed credible financial data to make informed decisions NSSF, and the Government’s funds for the Budget to improve service delivery. A proper accounting Entity-entitled Provincial Referral Hospital’s (PRH) system not only allows HEFs to get complete operations. Some other operating expenditures 104 Cambodia Public Finance Review - From Spending More to Spending Better Box 3.3. Strengthening financial management at health facilities under HEQIP During the implementation of the Health Equity and Quality Improvement Project (H-EQIP, 2017- 2023), the SDG manual accounting system was put into practice. The system incorporates necessary accounting controls and allows for production of the statement of receipts, expenditure and fund balance in accordance with the cash basis of accounting for all SDG funds. Although health facilities (HF) are commitmented to use this system, a majority of them are unable to do so due to an inadequate training and support. The application of this system will be further strengthened in HEQIP Phase 2 which began in 2022. The introduction of the SDG manual accounting system is a steppingstone to equip health facilities with some basic knowledge on financial controls and consolidated financial reports of the SDG funds. This system can be subsequently built upon to further integrate the HFs’ other fund sources to enable a production of the consolidated financial reports that present comprehensive receipts, expenditures and financial positions. Inherently, the more fund sources are added to the manual system, the more complicated it may be for HFs to implement. It is, therefore, suggested that a simple web-based accounting application adopting a cash basis of accounting be developed in the future for timely and reliable reports of financial resources. Source: Authors such as delivery fees, on-duty allowances, food for care, the costs rise as anticipated from health inpatients, medicine, vaccine, and equipment are centers with beds to the three levels of hospitals directly paid for or provided by PHD/MOH. Staff (health centers without beds do not offer inpatient salaries, and related staff’s functional allowances and care). By far the highest unit costs per service for others are paid directly by the government. The current health centers are for chronic diseases (NCDs) such systems used to manage those financial resources as diabetes and hypertension, and birth deliveries, are disaggregated and in a manual approach (Box which may primarily be due to the low volume of 3.3). Each fund source is generally managed in the deliveries and NCD care at the primary care level. form of a cash book and report on expenditures, Services such as birth delivery were found to be which are mostly paid in cash and supported by less costly per unit at the hospital level, and when approved vouchers and invoices/receipts. The looking from the quality of care angle, the quality and fundamental accounting controls such as monthly comprehensiveness of care are better due to more cash count and reconciliation, monthly bank qualified professionals, better equipment, and staff reconciliation, and general ledger are not regularly with more hands-on experience in providing care carried out. Due to the nature of a separated type of due to greater volumes of patients. For NCD care, systems and practices, the complete consolidated given screening and treatment is largely provided receipts, expenditures, and fund balance statement by the private sector and is a relatively new services for all fund sources cannot be produced in a timely for government health centers (particularly in 2019 and accurate manner. Hence, it is not possible for when the data was collected), the low volume is HF’s management and the Provincial Administration to view the entire picture of the financial operations largely driving the unit costs. As more health centers and situation. expand screening and treatment in the coming years, as the government invests in its NCD control Unit costs to deliver health services vary by level program at the primary care level, greater efficiencies of care (Figure 3.36). First, for services like inpatient will be achieved. Cambodia Public Finance Review - From Spending More to Spending Better 105 Figure 3.36. Unit costs for health centers (in US$ 2019 values) 250 200 HC HCwB CPA1 CPA2 CPA3 150 US$ 100 50 0 Inpa�ent IDP OPD Delivery Chronic Vaccina�on ANC/PNC/FP HC visit Surgery Paediatric General day inpa�ent case case/ pa�ent case case medicine maternity case Type of services Sources: GIZ (2021); GIZ estimates from the second round of health services costing study: results from 60 public health facilities (unpublished) Note: HC=Health center; HCwB= Health center with beds; CPA= Complementary Package of Activities Inpatient day means inpatient cot per day; IDP inpatient means inpatient cost per case. 3.4. Summary of key findings Several inefficiencies are observed in Cambodia’s health sector, starting with the implementation While progress can be seen in some health of program-based budgeting. The objectives outcomes, increasing health spending together behind the transition to program-based, output- with improving spending efficiency should focused budgeting is meant to focus on the delivery of be given priority to achieve better health services and targeted results, rather than just inputs, outcomes and increase equitable access to including the services to be delivered by ministries. healthcare. First, Cambodia spent less on health In practice, program-based budgets in Cambodia as a share of total government spending compared allocate funds according to specific activities, and to its neighboring countries. High OOP (around 60 these allocations are aggregated to create budget percent) as share of total health expenditure remains lines. However, the link between budget lines and a key constraint for Cambodia to achieve UHC. programs is constrained. For example, government OOP is not an efficient financing channel for health budget spending, like for salaries, that are common because it prevents risk sharing, and it is also not to more than one program, are often grouped into equitable as it imposes a relatively higher welfare a single (often the largest) program. This lack of a burden on the poorest. Although the government clear allocation methodology complicates efforts to put in place social protection schemes including HEF isolate program costs and assess their contribution and NSSF, OOP spending does not decline. Second, to the overall budget. Introducing some form of there is a need for the government to improve the capitation payment at primary care level, could add quality of spending to ensure that the spending on more clarity/transparency to public health expenditure health will translate into better health service delivery allocation and would contribute to a gradual transition and improved equitable access to health. Further, continuous increases in government staff salary towards the establishment of mixed provider payment in the coming years, and increasing the burden of mechanisms. A transition to capitation payment would communicable and non-communicable diseases however need to be accompanied by measures will exert expenditure pressure for the sector. Below aiming at strengthening the capacity to manage funds are several recommended measures to improve at the frontline level, and at reinforcing the digital allocative and technical efficiencies of spending. infrastructure to facilitate population empanelment. 106 Cambodia Public Finance Review - From Spending More to Spending Better There are notable disparities in access and in the rural areas. As primary healthcare is the first availability of health services across the country. contact with the health system, available resources The review of health outcomes and funding in the should be increased to reduce gaps in the availability report sections has shown that access to health care of affordable, quality, safe and effective health is generally pro-poor but not sufficiently progressive, services across geographical locations, particularly and differs across socioeconomic status. Distribution in rural areas. Core inputs seen as necessary for of health resources is more equitable on per capita health service delivery include financial resources, basis, but likely fails to address gaps in quality. competent healthcare staff, adequate physical Aggregate data suggests that poorer provinces facilities and equipment, and essential medicines with the worse health outcomes appear to receive and supplies. fewer resources, however, further analysis is required depending on data availability. 3.5. Recommendations The inadequate availability of the healthcare Increasing government health spending workforce, medicines, and equipment reduces through expanding beneficiary of the technical efficiency of healthcare spending. financial protection schemes These inefficiencies refer to the extent to which government resources and inputs are able to influence To achieve the UHC target, it is important to health outcomes. Problems of allocative efficiencies increase health coverage by building on existing described earlier are interrelated. Nonetheless, these social health protection (SHP) schemes. The are important to highlight as international experience social health insurance coverage is estimated to provides several potential entry points to tackle reach 57 percent of the total target population in technical efficiency challenges. 2023, with HEF covering 31 percent, NSSF covering 25 percent, and other 1 percent.85 Building on these Public capital spending on key public health programs is almost entirely financed by achievements, efforts should be made to expand external partners. The reliance on external tax-financed non-contributory coverage to the non- funding is a concern for the sustainability of poor informal sector.86 The expansion could target infrastructure programs, and potentially raises issues the currently uncovered first to third wealth quintiles in harmonization and overall efficiency – external and those 250,000 households identified in late 2022 spending fell outside of the report’s scope due to as “near poor”. A “HEF Plus” package that includes data limitations. Spending priorities do not give a slightly expanded benefit package may be more sufficient importance to public health programs with appropriate. The package could be fully subsidized high potential effectiveness and emerging challenges or with co-payment options for non-poor (some like non-communicable diseases. To more effectively revenue generation), but could possibly increase address the country’s public health objectives and challenges in enforcement. challenges described in this report, the RGC needs to better prioritize the health sector in its budget Increase government health expenditure for by increasing domestic financing to the sector and HEF service payments, including for current dedicating resources to the health centers, which enrollees, and to expand population coverage. provide primary health care for most of the population This could be through aligning HEF payment rates 85 Ibid 86 (1) Yazbeck et al 2019 The Case Against Labor-Tax-Financed Social Health Insurance For Low- And Low-Middle-Income Countries; (2) Yazbeck et al 2021 Making Health Insurance Pro-poor: Lessons from 20 Developing Countries (https://www.healthaffairs.org/ doi/10.1377/hlthaff.2019.00874) https://www.tandfonline.com/doi/full/10.1080/23288604.2021.1917092); (3) Yazbeck et al 2023 Addiction to a bad idea, especially in low- and middle-income countries: Contributory health insurance (https://pubmed.ncbi.nlm.nih. gov/36822716/) Cambodia Public Finance Review - From Spending More to Spending Better 107 with NSSF rates to optimize the incentive for accompanied by robust population coverage increased service provision and eliminate the 2-tier policies and targeting to reduce the financial system where currently payment rates are much hardship on households, especially among the lower for HEF than NSSF for the majority of services. poor and vulnerable. There should be a particular NSSF coverage should be expanded to dependents focus on reducing OOP related to medical products and household members of enrollees. and outpatient care through tangible benefit packages, bridging the gap between what is on The current UHC Roadmap agenda put forward paper and what is available at the point of service by the Government of Cambodia represents a delivery. This should be combined with strengthened unique opportunity to strategically think about regulatory measures for procurement and the pricing the most effective ways to enroll the missing of medicines. middle population (informal non-poor sector), as well as to revisit the benefit packages included Rebalance resource allocation towards primary in current SHP schemes in order to improve the health care, with focus on poorer provinces protection of these schemes regarding financial with lagging health outcomes. Address technical hardship. There is a need to establish a mechanism inefficiencies by revamping training programs to to regularly review, harmonize, and update the benefit address gaps in the workforce, further enhancing packages of social health protection schemes to coaching, mentorship and supportive supervision ensure an integrated service delivery model. In the approaches while continuing to expand and update face of the current demand confronting the sector, in-service training opportunities. the regulation and contracting modalities for private sector providers will be an important element of Increasing allocative and operational future health financing reforms in Cambodia. efficiency Reducing out-of-pocket expenditure Strengthening performance-informed budgeting through improved program budget structure. Shift from heavy reliance on OOP spending The current program budget structure for MOH is to increasing public spending on health and not responsive to performance-informed budgeting. applying it more efficiently. Overall, the volume of Making staff more responsive can be accomplished funding invested in the health sector in Cambodia by restructuring the program and allowing program is on par with what other countries are spending at managers to have authority over the resources similar income levels, but the remaining challenge needed to deliver results. in terms of health financing transition will be to shift the health financing composition from a model Developing methodology and budgeting tools to heavily dominated by health OOP spending to a isolate and allocate program costs to individual health financing scheme where prepayments in the programs. The government would benefit from re- form of increased public funding or increased social evaluating program design and seeking to prioritize health insurance contributions are the principal resources for emerging challenges. Strengthen financing channel. In the face of the current demand monitoring and evaluation systems by applying faced by the sector, the regulation and contracting modalities for private sector providers will be an more consistent and realistic performance matrices important element of future health financing reforms that include quality-related metrics. Measurement in Cambodia. is the key to accountability and improvement, but available measures typically do not capture many The shift away from OOP spending to of the processes and outcomes that matter most to tax-financed health financing should be people. 108 Cambodia Public Finance Review - From Spending More to Spending Better Improving access to quality drugs at affordable effective ‘regulatory approach’ to modify provider prices. Setting the right prices for quality medicines behavior. Several tools exist that Cambodia could should be the foundation for service delivery, whether use to tap in to private health sector capacities to through public or private avenues for procurement, improve supply side readiness of the health system. distribution and supply. As coverage of social health protection schemes expands, there will be a need to ensure that quality Harness the private sector as part of the health services, medicines and medical commodities are service delivery system. Harnessing the private available to respond to the growing demands from sector is a must for Cambodia’s UHC objectives; the population to access essential health services expanding coverage of financial protection schemes they will be entitled to through health benefit without engaging the private sector will not go far. packages under the different schemes. A few areas Contracting of private providers has been successful for exploration are presented below in Figure 3.37. elsewhere with proper regulatory measures and incentives in place. Purchasing delivery of essential Table 3.4 presents summary of recommendation for services by engaging private providers can be an health chapter. Figure 3.37. Tools of government for private sector engagement for UHC in Cambodia Large and rapidly growing Licensing procedures Untapped potential for private sector for private health facilities private provider contracting • Strengthen public-private • Minimum standards for • Unique opportunity of new PCA dialogue private sector provide new and SHI-power of single opportunity purchaser for prices and benefit • Legal, regulatory, financial package in both public & arrangements to harness • Link licensing with reporting and regulatory requirements private sector both public and private sector capacity in health • Strategic “contracting in” system models can strengthen public health facilities (e.g., diagnostics) Insuf ciencies in Lack of suf cient Insuf cient access to procurement practices legislation and regula- essential medicines in Dual practice and price setting tion on accreditation public sector for pharmaceuticals • Move forward on • Models to expand SHP • Price regulations for • Revisit regulation and pending “Law on benefit packages to procurement, incentives Administration of include supply from distribution and retail • Model performance Health Services” for private sector/ framework, task shifting national accreditation pharmacies and technology body innovation Cambodia Public Finance Review - From Spending More to Spending Better 109 Table 3.4. Options to improve health outcomes and increase health equitable access Options Challenges Short-term recommendations Medium-term recommendations (1-2 years) (3-5 years) Reducing Increase health coverage by building Revisit the benefit packages included in out of pocket on existing social health protection current SHP schemes in order to improve the (SHP) schemes protection of these schemes regarding financial expenditure • Efforts should be made to expand hardship through tax-financed non-contributory Harness the private sector as part of the health expanding coverage to the non-poor informal service delivery system sector beneficiaries • Contracting of private providers has been • Introduce co-payment options of financial successful elsewhere with proper regulatory for non-poor (some revenue protection generation) measures and incentives in place schemes • Purchasing delivery of essential services Increase government health by engaging private providers can be an expenditure for HEF service payments, effective ‘regulatory approach’ to modify including for current enrollees and to provider behavior expand population coverage • Ensure that quality services, medicines • Align HEF payment rates with and medical commodities are available to NSSF rates to optimize the respond to the growing demands of the incentive for increased service population through effective private sector provision engagement frameworks and models • Expand to dependents and household members of enrollees Improving Rebalance resource allocation The shift away from OOP spending to spending towards primary health care, with tax-financed health financing should be focus on poorer provinces with accompanied by robust population coverage efficiency policies and targeting to reduce the financial lagging health outcomes hardship on households, especially among the • Address technical inefficiencies by poor and vulnerable revamping training programs to • Focus on reducing OOP related to medical address gaps in the workforce products and outpatient care through • Enhance coaching, mentorship tangible SHP benefit packages and supportive supervision • Bridge the gap between what is on paper approaches while continuing to and what is available at the point of service expand and update in-service delivery to reduce risks of unanticipated training opportunities health expenditures for SHP beneficiaries • Strengthen regulatory measures for procurement and the pricing of medicines to improve availability, affordability, and quality Strengthening Improve program budget structure. Incorporate performance information into performance- • Restructure the current Program budget formulation and negotiation processes Budgets informed • Allow program managers to have budgeting more authority over the resources needed to deliver results Develop methodology and budgeting tools to isolate and allocate program costs to individual programs Strengthen monitoring and evaluation systems by applying more consistent and realistic performance metrics 110 Cambodia Public Finance Review - From Spending More to Spending Better 4 STRENGTHENING PERFORMANCE ORIENTATION OF PUBLIC SPENDING TO BOOST PRODUCTIVITY IN AGRICULTURE 4.1. Introduction (MOWRAM) oversees the development of irrigation systems and coordinates with the MAFF through The agricultural sector continues to play an the Technical Working Group on Agriculture and important role in Cambodia’s economic and Water (TWG-AW).87 Other ministries with important social development and provides a major source roles related to agriculture and rural development of employment and income, particularly for the rural include the Ministry of Rural Development (MRD)— poor, with 35 percent of the labor force employed in responsible for rural roads—and the Ministry of agriculture. The sector contributes approximately 22 Commerce (MOC)—responsible for business and percent to the country’s GDP. trade promotion. Multiple public sector organizations are active 4.2. Sectoral context in the agriculture sector but the government’s budget for this sector is managed by three Agriculture was the only sector that showed ministries. The Ministry of Agriculture, Forestry, and real annual positive growth—albeit at the slower Fisheries (MAFF) is the key institution responsible pace of 0.6 percent—during the COVID-19 for the regulation and promotion of agriculture pandemic. In 2020, the economy contracted by production and in 2020 accounted for almost a 3.1 percent which marked the end of annual growth quarter of the budget allocation in the sector. The over the past two decades. However, the agriculture Ministry of Water Resources and Meteorology sector was less affected by the pandemic-induced 87 The TWG-AW was established in 2004 to serve as a coordinating body and dialogue mechanism for agriculture and water resource management and is co-chaired by MAFF and MOWRAM with the support of a secretariat and coordinated by FAO and IFAD. The TWG-AW brings together government, development partners and civil society representatives to support the attainment of the National Development Goals set out in the Rectangular Strategy, the NSDP and the reform programs associated with the sector. Cambodia Public Finance Review - From Spending More to Spending Better 111 Figure 4.1. Employment in agriculture (% of Figure 4.2. Evolution of agriculture’s total employment, 2019) share of GDP and employment in selected countries (2000-2019) 2019 2000 70 40 60 50 30 40 30 20 20 10 10 0 0 0 20 40 60 80 100 Employment in agriculture (% of total employment) Sources: WDI; WB staff estimates Sources: WDI; WB staff estimates Note: Countries include Cambodia, Brazil, Chile, China, Lao PDR, Philippines, South Africa, Turkey, and Vietnam. slowdown than the rest of the economy. Other At the beginning of the pandemic, in May 2020, sectors which provide employment opportunities to nearly two-thirds of job switchers changed sectors, the population was affected significantly. The services most commonly leaving the service sector to enter sector, for example, saw the sharpest decline of the agricultural sector. Workers continued to shift 6.5 percent, driven primarily by the collapse of the into agriculture during most of 2020 and early 2021 tourism sector. (World Bank, forthcoming),89 potentially contributing to a larger share of employment in agriculture. Agriculture is one of the largest sources of employment, accounting for 35 percent of While Cambodia has been able to raise employment (Figure 4.1), with the larger share of agricultural productivity in line with its regional employment amongst women (37 percent of female neighbors, the sector faces challenges in employment in agriculture versus 32 percent for remaining competitive in export markets. The men).88 More than 5 million people (out of Cambodia’s farmgate prices of agricultural products in Cambodia 16.6 million) depend on agriculture and fishing for continue to be competitive; however, the cost of their livelihood, income, and food security. The share electricity for agro-processing enterprises, the of agriculture in total employment has decreased logistics and shipping costs for agricultural product from 73 percent in 2000, mirroring a common trend during the structural transformation of the economy. exports are not as competitive. In addition, the Overall, agriculture’s contribution to Cambodia’s GDP sector faces risks to growth and discrete challenges also declined from 36 percent to 21 percent during posed by climate change and labor shortages due the same period (Figure 4.2). Similar declines are also to both demand- and supply-side factors. Enabling observed in Vietnam (from 25 percent to 15 percent), agricultural policy and pro-growth public expenditures and Lao PDR (34 percent to 16 percent) during the in the sector will be critical in stimulating further same period. However, the pandemic may have agricultural productivity growth and promoting a paused the shift of employment out of agriculture. vibrant rural economy. 88 When expanding the definition of agriculture employment, 50 percent of the population above 18 years old have their primary occupation in the farming, livestock, fisheries or forestry. 89 EAP Covid-19 report. 112 Cambodia Public Finance Review - From Spending More to Spending Better Figure 4.3. Public spending in the agriculture Figure 4.4. ODA in Agriculture sector (US$ sector, (US$ million and % of GDP) million actuals) 2.00% 200 500 1.98% 400 US$, million US$, million % of GDP 300 1.95% 100 200 1.93% 100 0 1.90% 0 2018 2019 2020 2018 2019 2020 2021 GOC spending in Agriculture sector (US$, million) Water & irrigation Agriculture sector policy and management Extension services Fisheries ODA (US$, million) Food security, nutrition Agriculture inputs % of GDP (RHS) Other Sources: MEF; WB staff estimates Sources: Development partner reporting; WB Staff estimates Note: RHS = Right-hand scale 4.3. Agriculture public means that PFM reforms may have limited reach in strengthening budget effectiveness since the large expenditure shares of externally funded resources are not directly impacted by reformed processes and systems. Combining both MAFF and MOWRAM, Cambodia spends around 8 percent of the MAFF, MOWRAM, and MRD are three ministries total government budget on agriculture, most directly relevant to agriculture sector. while the average spending by the EAP and MOWRAM spending accounted for the biggest ASEAN countries was around 7 percent and 4 share of the sector, from 49 percent of the sector percent respectively between 2016 and 2020. spending in 2017 to a peak of 51 percent in 2019 Development partners provide between US$200 before declining to 43 percent in 2021. MRD is the million to US$300 million each year to the agriculture second biggest of the sector in terms spending. Its sector (Figure 4.3 and Figure 4.4). Reliance on spending accounted for around a third of the sector, external financing creates a considerable degree followed by MAFF spending (around one fourth of of budgetary fragmentation, creating challenges to the sector spending) (Figure 4.5). More than half of translating public expenditures into outcomes. It also ODA resources fund water and irrigation under the Figure 4.5. Agriculture sector spending (domestically financed and DP-financed), % of sector by ministry 100% 90% 30% 32% 29% 32% 32% 80% 70% 60% 50% 43% 49% 46% 51% 47% 40% 30% 20% 10% 21% 22% 21% 25% 20% 0% 2017 2018 2019 2020 2021 MAFF MOWRAM MRD Sources: Budget Settlement Law (2017-2021); World Bank staff estimates Note: MAFF = Ministry of Agriculture, Forestry, and Fisheries; MOWRAM = Ministry of Water Resources and Meteorology; MRD = Ministry of Rural Development Cambodia Public Finance Review - From Spending More to Spending Better 113 Figure 4.6. Spending by source in 2021 (% of total spending for Agriculture sector) Capital, domestically fianced 24% 49% 27% Capital, externally nanced 10% 47% 42% Recurrent 42% 27% 31% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% MAFF MOWRAM MRD Sources: Budget Settlement Law (2021); World Bank staff estimates Note: MAFF = Ministry of Agriculture, Forestry, and Fisheries; MOWRAM = Ministry of Water Resources and Meteorology; MRD = Ministry of Rural Development mandate of MOWRAM, which is by far the largest MAFF’s mandate provides it with a wide platform ODA expenditure. In 2021, less than 20 percent of to lead agriculture policy. The ministry is made up ODA was channeled through MAFF; these funds paid of four major departments: i) General Directorate of for extension services for productivity improvement, Agriculture, ii) General Directorate of Animal Health food security and nutrition, fisheries, agriculture and Production, iii) Fisheries Administration, and iv) sector policy and management, climate-resilient Forestry Administration. The first three are reviewed practices, and food safety. as part of this analysis. The mandate of MAFF is defined in a sub-decree90 on the organization and MAFF accounts for 42 percent of the recurrent functioning of the ministry.91 This covers both the spending in the sector (Figure 4.6). MOWRAM’s policy-guiding role as well as the implementation of domestically financed capital spending took up activities for broader agricultural development. The almost half (49 percent) of government investment ministry’s functions and duties are defined as the spending for the sector in 2021, followed by MRD following: 27 percent and MAFF 24 percent. DP financed capital spending for MOWRAM was higher at 47 • enhancement and implementation of agricultural percent, while MRD slightly lower that 42 percent, development policy in order to improve the and MAFF at 10 percent. The interconnected nature standard of living; of the institutional mandates poses challenges to • participation in the formulation of land reform and coordinating activities across three ministries. For land use policy; example, MAFF’s mandate in the development sector • orientation and planning for the development of is superseded by capital expenditure undertaken all agricultural sectors; by MRD for the development of the adequate road • coordination of the monitoring and evaluation of infrastructure. Coordination among these institutions the implementation of policies and activities for is vital to boosting productivity and effective service agricultural development; provision. 90 Dated 07 April 2000 91 As stated in a sub-decree on organization and functioning of the Ministry, MOWRAM has functions dealing with inland water resources, which includes defining policy and plans on water development, developing required legislation and regulations concerning the water resource sectors; setting short-, medium- and long-term targets for the exploitation and development of water resources and for conservation purposes to meet the national need and community livelihoods, rehabilitating major irrigation infrastructure to provide enough water for agriculture, daily consumption of humans and animals and tourism, and constructing flood control and polders to protect agricultural land and public property. It is logical for MOWRAM and MAFF to work closely together to carry out such activities. The two ministries would benefit from more collaborative work. 114 Cambodia Public Finance Review - From Spending More to Spending Better • monitoring the development of the natural 43 percent in 2021 (Figure 4.8). The domestic capital resources of the agricultural sector and facilitating spending has decreased from 20 percent of total the exploitation of natural resources to meet the spending in 2017 to 2 percent in 2019, before reaching needs of the country, while having respect for 21 percent in 2021. The increase of domestic capital ecosystems; spending was driven by a rise in investment subsidies • defining regulations on the management, from DPs. Most capital-intensive agriculture-related preservation of natural resources of the agricultural spending92 is implemented by other ministries, such sector, and monitoring of implementation; and as MOWRAM and MRD. Rising labor costs have the • providing training for agricultural development potential to crimp the fiscal space available for the to increase the level of technical knowledge and provision of key agricultural services and MAFF’s key understanding. objective of promotion of agriculture production. 4.4. Public spending in 4.4.1. Expenditure by program agriculture by MAFF Budget allocation that aims to boost productivity MAFF’s nominal spending has remained and crop diversification decreased between relatively constant around 0.3 percent of 2017 and 2021. However, assessing the true program GDP between 2017 and 2021. Increases in cost is constrained by difficulties in allocating and wage spending have been offset by reductions tracking wage expenses across programs. Program in domestically financed capital spending (Figure 5 “Increasing effectiveness of supporting services 4.7). Between 2017 and 2021, recurrent spending and human resource development” dominates the increased by 9 percent during the period, driven by program-based structure, accounting for more than growth in payroll expense (and a rise in wage) from two-thirds MAFF’s total (Figure 4.9 and Figure 4.10). KHR 103,141 million (0.11 percent of GDP) in 2017 This is explained by the inclusion of payroll expenses to KHR 149,678 million (0.14 percent of GDP) in (75-80 percent of program’s total, Chapter 64) that 2020 and slightly declined to KHR 125,929 million include all staffing costs for the other 4 programs. (0.11 percent of GDP) in 2021. Wages now take a Grouping all salary expenses under Program 5 larger share of domestic resources, accounting for underestimates the true expense of each of the other half of total MAFF spending in 2020 and declined to programs. Figure 4.7. MAFF spending, economic Figure 4.8. Economic composition of MAFF composition by chapter (KHR, million) spending (% of total domestic spending) 350,000 100% 2% 2017 2018 2019 2020 2021 10% 300,000 90% 20% 18% 21% 250,000 80% KHR, million 70% 48% 200,000 40% 150,000 60% 41% 41% 36% 100,000 50% 40% 50,000 30% 0 49% 50% 20% 39% 40% 43% s se en 10% p S ex P I y id bs T 0% P Su 2017 2018 2019 2020 2021 I S Wages Non-wage recurrent Capital, domestically financed Sources: Budget Settlement Law (2017-2021); World Bank staff estimates 92 Externally financed resources (from development partners) are increasingly important to MAFF but are not included in the analysis due to data constraints Cambodia Public Finance Review - From Spending More to Spending Better 115 Figure 4.9. Program-based budget, by Figure 4.10. Program-based budget, by program (KHR, million) program (% of total, excluding externally financed projects) 300,000 100% 250,000 80% KHR, million 45% 200,000 56% 68% 69% 71% 60% 150,000 6% 6% 100,000 40% 6% 6% 6% 5% 6% 5% 5% 50,000 20% 6% 6% 5% 37% 6% 5% 4% 27% 0 14% 15% 15% 2017 2018 2019 2020 2021 0% 2017 2018 2019 2020 2021 Program1 Program 2 Program 3 Program4 Program 5 Program 1 Program 2 Program 3 Program 4 Program 5 Sources: MAFF; World Bank staff estimates Notes: The five programs are: Program 1: Increasing productivity, diversifying agricultural crops, and agribusinesses; Program 2: Promoting animal production and animal health; Program 3: Fisheries management and aquaculture development; Program 4: Management and development of forest and wildlife resources; Program 5: Increasing efficiency, supporting services and human resources Annual budget outturns under programs at productivity and crop diversification. These execution the central level are considerably lower than challenges are not observed at the Provincial level. budgeted amounts, particularly for productivity- increasing objectives under Program 1 (Figure 4.4.2. Central vs provincial level 4.11). Comparing the allocation in the budget expenditures law, spending under Program 1 accounted for 15 percent, 34 percent, and 37 percent of budgets in Provincial-level budgets account for over a 2019, 2020, and 2021, respectively. Programs 2, third of MAFF’s total and are also dominated 3, and 4 perform better but are still underspent at by human resource (HR)-related expense under about 80 percent of their program budget allocations Program 5 (Figure 4.12 & Figure 4.13). The 25 in 2021. This is likely explained by the austerity provincial units (Provincial Department of Agriculture measures following the COVID-19 pandemic, which Forestry and Fisheries – PDAFFs) follow the same 5 saw a large budget reallocation to fund vaccination programs and are responsible for covering payroll and treatment of patients. Combined with declining expenses. Despite taking about a third of total allocations to Program 1, weak budget execution resources, provincial offices account for nearly half limits the ministry’s capacity to contribute to boosting of Program 5. In 2020, Program 5 accounted for 82 Figure 4.11. Program budget outturns 120% Program 1: Increasing productivit diversifying agricultural crops, an 100% agribusinesses Program 2: Promoting animal pro 80% and animal health Program 3: Fisheries managemen 60% underspend aquaculture development 40% Program 4: Management and dev of forest and wildlife resources 20% Program 5: Increasing efficiency, supporting services and human 0% resource 2017 2018 2019 2020 2021 Source: Budget law, MEF; WB staff estimates Sources: Budget law, MEF; WB staff estimates 116 Cambodia Public Finance Review - From Spending More to Spending Better Figure 4.12. MAFF recurrent spending, by level Figure 4.13. Program budget, by program across of administration (% of total) (2017-2020) level of administration, (% of total) 2020 100% 100% 90% 18% 26% 26% 80% 38% 42% 40% 39% 80% 35% 45% 70% 60% 60% 50% 40% 40% 82% 74% 74% 65% 30% 62% 58% 60% 61% 55% 20% 20% 10% 0% 0% Program 1 Program 2 Program 3 Program 4 Program 5 2017 2018 2019 2020 Central (Ministry) Provincial Central (ministry) Provincial Sources: Budget Settlement Law (2017-2021); World Bank staff Sources: MEF; World Bank staff estimates estimates percent of the provincial-level budget with the other This roughly aligns with staffing across the levels four programs taking up 7 percent, 3 percent, 4 of administration, with 44 percent employed by the percent, and 4 percent respectively. Data availability center and 43 percent working in provincial offices. limits the analysis of provincial-level spending (see MAFF’s deconcentration of its staff and wage bill is World Bank 2021, “Cambodia intergovernmental comparable to the MRD but more centralized than fiscal architecture” for the summary of decentralization other agencies such as education and health (Figure reforms and challenges).93 4.14). The wage bill split between the center and Budget allocations to the provincial level provinces reflects staffing across the levels of appear relatively equitable, however, data administration. In 2019, MAFF’s wage bill at the center availability limits the assessment of the quality accounted for 47 percent of its total with 41 percent of targeting. A positive correlation between the attributed to the provincial offices (Figure 4.15). size of the recurrent budget in a given province Figure 4.14. Civil servants in selected LMs Figure 4.15. Wage allocations in selected LMs By administrative level (2020) By administrative level (% of LM total, 2019) 3,696 100% 100% 3% 45,619 11,953 4,383 20% 9% 80% 80% 40,730 688 39% 2,867 5,467 947 23% 43% 60% 60% 47% 18% 40% 40% 2,806 853 838 20% 41% 46% 20% 132,372 866 301 107,800 13,291 364 12% 15% 57% 38% 87% 0% 0% Total MAFF MRD MOEYS MOH MOSVY MAFF MRD MOH MOSVY MOEYS District Provincial Central District Provincial Central Sources: Cambodian authorities; World Bank staff estimates (World Bank 2022) Note: MAFF = ministry of agriculture, forestry, and fisheries; MRD = ministry of rural development; MOEYS = ministry of education, youth, and sports; MOH = ministry of health; MOSVY = ministry of social affairs, veterans, and youth rehabilitation; LM = line ministry 93 https://documents1.worldbank.org/curated/en/099440003152223052/pdf/P168407082902104a0b28a0d00b4ec6625f.pdf Cambodia Public Finance Review - From Spending More to Spending Better 117 Figure 4.16. Recurrent budgets and poverty Figure 4.17. Recurrent non-wage budgets and levels in provinces poverty levels in provinces 7,000 2,500 6,000 r² = 0.6455 Non-wage recurrent r² = 0.3099 Total recurrent expenditure 2,000 (KHR, million) 5,000 expenditure (KHR, million) 4,000 1,500 3,000 1,000 2,000 500 1,000 0 0 0 100,000 200,000 0 100,000 200,000 Number of poor Number of poor Sources: MAFF; World Bank staff estimates Table 4.1. Functional composition of MAFF recurrent spending Selected functional categories (KHR, million) (% of recurrent expenditure) of MAFF 2015 2016 2018 2019 2015 2016 2018 2019 Provincial programs under Program 5 44,033 56,965 83,136 89,476 29.3% 30.3% 35.8% 33.4% Planning and management capacity 37,322 43,667 80,448 104,436 24.8% 23.2% 34.6% 39.0% Human resources development 18,952 28,615 59,780 66,661 12.6% 15.2% 25.7% 24.9% Education 8,649 11,431 7,600 10,043 5.8% 6.1% 3.3% 3.7% Production support to fisheries 1,627 2,226 16,675 17,579 1.1% 1.2% 7.2% 6.6% Production support to other products 6,840 7,113 10,909 11,334 4.6% 3.8% 4.7% 4.2% Research (by CARDI and other 6,464 7,632 9,310 10,289 4.3% 4.1% 4.0% 3.8% entities) Inspection services (law enforcement 6,464 6,454 8,960 8,720 4.3% 3.4% 3.9% 3.3% and internal audits) Extension and farmer organizations 3,437 4,888 7,014 7,430 2.3% 2.6% 3.0% 2.8% (for crops, aquaculture, and livestock) Animal health services 4,108 4,453 7,056 7,046 2.7% 2.4% 3.0% 2.6% Rubber programs 3,969 4,420 3,386 4,071 2.6% 2.4% 1.5% 1.5% SPS and laboratories 2,776 3,197 2,356 2,435 1.8% 1.7% 1.0% 0.9% Agro-industry department 2,361 2,954 1,906 1,957 1.6% 1.6% 0.8% 0.7% Rice production development 2,095 2,599 1,597 1,519 1.4% 1.4% 0.7% 0.6% Horticulture (subsidiary and industrial 1,200 1,412 1,468 1,575 0.8% 0.8% 0.6% 0.6% crops) Total recurrent expenditure 150,297 188,026 232,345 267,988 100% 100% 100% 100% Sources: Cambodian authorities; WB staff estimates Notes: CARDI = Cambodian Agricultural Research and Development Institute 118 Cambodia Public Finance Review - From Spending More to Spending Better and its number of poor (Figure 4.16) suggests that 4.5. Public financial the budget does target at least some of the poorer provinces. However, this correlation weakens when management salary is excluded (Figure 4.17). This could indicate The Royal Government of Cambodia (RGC) that the deployment of staff may be calibrated with developed and launched a public financial the needs across provinces (as proxied by poverty management reform program (PFMRP) with a levels) but isn’t necessarily translated to the level detailed and sequenced action plan in 2004. The of service provision through non-wage operational PFMRP is a sector-wide tiered (multi-stage or multi- expenditure. Nonetheless, without more detailed platform) approach that aims to instill high standards of data (functional level of spending and within-province management and accountability in the mobilization of targeting), the analysis cannot determine whether resources and ensure their effective and efficient use. the level of spending targets the different needs Piloting of program budgeting was introduced to 10 across provinces. Furthermore, the analysis presents LMs in 2015 and expanded to 39 LMs/institutions94 only a partial view of MAFF spending that excludes in 2018, with the aim to gradually strengthen spending by the central-level ministry as well as accountability and the availability of performance capital investments, which are critical to addressing and outcome information enroute to the planned sector challenges, boosting productivity, access to implementation of full performance budgeting. The markets, and strengthening climate resiliency. experience of MAFF’s PB implementation showed that the current practices are not responsive to 4.4.3. Expenditure by functional performance informed budgeting. classification Spending on activities promoting agricultural 4.5.1. BSP and program budget diversification is low and falling (Table 4.1). implementation challenges Looking at the functional breakdown of spending limit effectiveness of budget (across all programs), nearly a third of the MAFF management recurrent budget has gone toward provincial programs, consistent with the aim of increasing Strategic guidance and structure the decentralization of essential services. Spending on core public goods through education, research, MAFF’s program budget structure is organized inspection services, laboratories, and extension along its administrative architecture (Annex to declined from 18 percent to 15 percent of recurrent Chapter 4, Table A.4.1). On the one hand, designing spending between 2015 and 2019. Spending on programs on the basis of administrative factors rather increasing productivity through the support of different than program logic may be limiting in operationalizing products and extension services as well as agro- the budget towards policy objectives. On the industry development is likely inadequate to meet other hand, a program structure that allows one the sector‘s objectives. Spending on animal support spending unit or department to draw funding from (2.6 percent of recurrent in 2019) and horticultural multiple programs and subprograms can complicate products (0.6 percent of recurrent in 2019) remain accountability and becomes difficult to manage small despite the challenges in boosting domestic in the nascent stages of performance-oriented production and competitiveness. As highlighted in the reforms. A practical approach would seek a balance Bank’s previous analysis of the sector, the growing between organizing programs along with the desired demand for higher-value foods has mainly been met policy objectives and recognizing the need for stable by imports from Vietnam, Thailand, and elsewhere. administrative structures to implement the agency 94 PEFA 2021 Cambodia Public Finance Review - From Spending More to Spending Better 119 budget each year. As such, MAFF’s current program relating to MAFF subprograms are nested under budget design is pragmatic at this stage in the reform the provincial-specific subprogram which creates a in that it allows a clearer chain of accountability and clearer separation between programs implemented data flow, but it could be changed with maturity of at the central level and provincial levels. The 25 reforms. PDAFFs follow the standard template provided by the MAFF. There are, however, challenges with the The structure is impractical in identifying quality of data and consistency of information such available funding for climate adaptation. It as the definition of indicators, activity clusters, budget is impossible to identify climate-related spending breakdown, and use of supporting data which varies under the current budget program architecture. 95 significantly from one province to another. Some The current program structure could potentially evidence suggests that not all PDAFFs use the incorporate climate-related budget allocations, same coding for each of their activity clusters. This however, there currently isn’t a defined methodology fragmentation means there remains a significant to enable MAFF to tag climate spending. Climate challenge to systematically link provincial activities to adaptation funding could target government strategic plans. activities such as providing farmers with information about climate change and addressing market failures Budgeting and planning where information is not available. Introduction of the BSP has strengthened the BSP 96 is not utilized for budget appropriations. articulation of ministry policy ambitions and At the aggregate level, BSP follows the exact five expenditure planning; however, it is not yet program structures. There are, however, differences operationalized in the budget. BSP can act as in expenditure reporting that obscure the comparison a valuable multi-year planning tool that already has between the planning under the BSP and the approved a strong degree of ownership from the ministry budgets that relate to the PB. The differences are during its preparation. However, several layers of rooted in the processes for budget negotiations and fragmentation and weaknesses in its content limit appropriations that are largely divorced from the its operational value. This is at least in part a result planning efforts of the BSP. The differences between of interrelated challenges, some of which stem from BSP and annual budgets diminish the usefulness of the broader budgetary fragmentation in Cambodia the former as a budget document where comparisons that are not unique to the agriculture sector. The between the approved annual budget and the BSP end result is that the contents of the BSP currently are impractical. have weak linkages to annual PBs. For MAFF, it also means that investing greater effort and ministry MAFF PB fragmentation across the provincial resources into the BSP process will not substantively level weakens the quality of planning. The 25 impact subsequent annual PB allocations. Figure PDAFFs follow the same program structure, however, 4.18 illustrates some of these challenges: there are particularities in how programs are coded in practice for each level of administration. At MAFF’s 1. Weak link between the BSP and annual budget national level PB, the provincial spending is presented discussions/ negotiations. Annual budget as separate subprograms under each of the five negotiations tend to focus on controlling total programs, despite both levels of administration amounts and incremental changes to economic nominally working towards achieving the same expense categories rather than on programmatic program objectives. In practice, provincial activities logic (program classification, subprograms, 95 Climate-related spending could be reflected in donor-funded projects, which are beyond the scope of this analysis. 96 BSP provides a platform to align MAFF’s strategic planning with PB and support the transition toward performance orientation. BSPs have been issued annually covering a rolling three-year period. 120 Cambodia Public Finance Review - From Spending More to Spending Better activities, outputs, indicators). This means that support the final selection decision. These the efforts to strengthen program-based strategic feasibility studies are largely outside of MAFF’s planning and costing through BSP by the ministry reach (fiscal and capacity constraints), driving a are not yet integrated into the processes that further distance between capital investments and shape the annual budget and ultimately have little ministerial planning and budgeting processes. impact on annual PB allocations. The resource Furthermore, the misalignment between timing envelope proposed by the MAFF tends to exceed when decisions for capital investments are made the MEF-issued ceilings. BSP does not include and planning processes mean that key decisions an explanation for the additional resources and for approval and funding to projects and ministries whether they are requested for ongoing programs are made after the deadlines for LM planning or new policy priorities. processes. Consequently, capital spending (and A key level of fragmentation is the absence of 2. by extension the associated recurrent costs) are capital expenditure during the planning not meaningfully integrated into annual strategic stage. Externally financed capital projects are planning through the BSP. not included in the BSP and largely sit outside of MAFF’s annual budget processes. Screening of The disconnect between BSP and annual externally financed investments/capital projects budget is reflected in the large gaps between is based on macro-fiscal and debt framework MAFF budget allocations, actual spending, and considerations. This helps support more prudent the BSP amounts (Figure 4.18). Programs 1 to 4 debt management but can contribute to a get allocated less than 50 percent of the amounts less robust alignment with sector and ministry proposed in the BSP. On other hand, program 5 is objectives. Proposed projects that pass the allocated 152 percent of the amount stated in the initial screening undergo feasibility studies to BSP. This can be at least partly explained by the Figure 4.18. BSP and budget planning BSP not part of budget negotiations  Budget negotiations focus on economic classifications (line items) rather than the program logic  Comparison of budget expenditure between BSP and annual budget negotiations are impractical Link top-down BSP policy with bottom-up priorities and expenditure Policy objective estimates Annual budget (PB)  Capital budgets and PB not integrated  Screening of externally nanced ASDP objectives investments based on macro scal & debt  Promoting agricutule and rural development considerations  Sustainable management of natural and cultural resources  Limited MAFF role in feasibility  Ensuring environmental sustianable change studies ( scal and capacity Capital constraints) budget Program 1 Program 2 Program 3 Program 4 Program 5 Capital investments are not meaningfully considered as part of strategic spending Provincial level processes not well integrated into the BSP  Lack of clarity of roles and information sharing between MAFF and PDAFF  MAFF has a limited view of PDAFF planning and budgeting Source: Authors Note: ASDP = agriculture sector development program; BSP = budget strategic plan; MAFF = ministry of agriculture, forestry, and fisheries; PDAFF = provincial department of agriculture, forestry, and fisheries; PB = program budget. Cambodia Public Finance Review - From Spending More to Spending Better 121 Table 4.2. Budget strategic plan, annual budget and actual spending of MAFF (KHR, million) Budget Strategic Plan Budget Strategic Plan Program State DP Other Capital Projects Total Law Actual Law as % Actual as budget budget budget with no of BSP % of BSP financing 2020 Program 1 134,933 170 86,517 86,517 135,103 37,991 33,960 28% 25% Program 2 123,897 104,000 104,000 123,897 14,491 13,118 12% 11% Program 3 22,733 9,280 9,280 32,013 15,044 12,073 47% 38% Program 4 22,340 6,690 6,690 5,852 29,030 12,799 10,407 44% 36% Program 5 53,530 11,225 4,142 68,897 104,749 105,875 152% 154% 2021 Program 1 135,585 170 86,011 86,011 135,755 43,553 28,249 32% 21% Program 2 27,897 8,000 8,000 27,897 13,152 9,932 47% 36% Program 3 23,985 11,880 11,880 35,865 13,838 9,635 39% 27% Program 4 24,484 6,014 6,014 5,648 30,498 10,278 7,465 34% 24% Program 5 54,784 10,805 4,338 69,927 99,456 92,685 142% 133% Source: MAFF Budget Strategic Plan (2022-2024) and data provided by MAFF difference in accounting for wages between the BSP Performance and program-based budget structure (described above). Equally important, there is an even larger Despite the inclusion of performance metrics divergence between BSP and actual spending, in the BSP, performance monitoring is not confirming the execution challenges documented sufficiently structured to inform the budget in the previous section. The disconnect ultimately or assess budget effectiveness. Inclusion of contributes to implementation inefficiencies that performance information and indicators in the BSP manifest themselves through these budgetary is an important step toward the reform objectives deviations. MAFF officials point to the persistent of performance-informed budgeting and enhanced issue of the disconnect between BSP and annual accountability for results. However, weaknesses budgeting, especially for capital or investment in the design of indicators and inconsistency in expenditure. One key reason for large deviations with reporting performance information weaken the the BSP is the lack of financing for capital projects, operational value of BSP’s performance monitoring. which largely rely on development partners. In 2020 and 2021, programs 1 and 2 have proposed capital Performance matrix design is not tailored to investment projects without available resources (i.e. budget resourcing and is difficult to administer. placed in the “no financing” category in Table 4.2). Most of the indicators are a mix of inputs, outputs Excluding these projects, the disconnect between and outcomes (Table 4.3). At one extreme, several planning and actual spending would diminish, indicators track inputs like the number of civil aligning to nearly 80 percent of the BSP amounts servants trained. On the other, the indicators are for programs 1 and 2 but would still be less than 50 not reflective of a reasonable reach of MAFF’s percent for programs 3 and 4. budgetary resources in areas where outcomes are 122 Cambodia Public Finance Review - From Spending More to Spending Better Table 4.3. BSP performance reporting structure and examples Main purpose Year Indicators / Targets Final result policy and Indicators Most recent Current year Next year and Indicators Goal implemented beyond programs year Program 1 1) Gross agri- 29.9 31.5 33.1 Growth of all 5% cultural product types of crop (million tons) production (%) Subprogram Amount of local 70 78 80 {Subprogram targets do 1.2 vegetable supply not include “final result” (%) indicators} Subprogram Number of local 3,000 3,000 3,000 5.3 and foreign- trained officers … Source: adapted from BSP (2022-2024) not easily attributable to MAFF’s fiscal policy. For reported using the same metrics (e.g., million tons example, two of the program-level indicators target vs. percent), with the former reported against a the export of agricultural products (million tons and baseline, for example, a comparison to the most growth of exports). MAFF’s relatively small budget recently implemented year. Reporting for the final is unlikely to wield a direct and significant influence result indicators does not always include a baseline in expanding export volumes. Similarly, reaching or a reference period. Similarly, units of measure the target is not directly attributable to MAFF, but are not consistent across indicators. For example, likely an outcome of the broader macroeconomic both meters and hectares are used as targets for the context, changes to the regulatory environment, and coverage of an area. One notable positive change broader investments in the sector. Furthermore, the has been the time coverage of the performance sheer number of indicators may stretch the ministry’s indicators in the BSP. The performance matrix for capacity. The two most recent BSPs (2020-22 and the 2020-22 BSP included future targets for only 2022-24) include over 90 indicators across MAFF’s one year into the future, despite the BSP’s three- five programs. This creates a strain on a meaningful year perspective. This has since changed for the integration of performance metrics into the budget 2022-24 BSP, which includes yearly targets for each process, with scarce resources used to undertake year covered by the BSP (up until 2024). The final reporting functions across the many indicators result targets, however, do not indicate whether the that could instead focus on the design of better- targets are meant to be met by the end of the BSP tailored targets, improved monitoring over a smaller period or whether they are targets for average annual list of indicators, and integration of performance performance. information into the BSP and budget proposals. Performance indicators are not easily Performance information is not consistently comparable between years and BSPs. Many presented. The performance targets are broken indicators change between BSPs. Most of the down into two tables covering program-level and program-level indicators remained the same, with subprogram-level objectives. Program-level metrics five changes out of 25 indicators between the two are typically reported as targets for a given year most recent BSPs. Overall, across over 90 indicators, and as final result indicators (Table 4.3). The targets nearly 40 percent have been changed. While the across the two types of indicators are not always performance of indicators that remain from one BSP Cambodia Public Finance Review - From Spending More to Spending Better 123 Table 4.4. BSP performance reporting for program-level objectives Main purpose politics 2020 (target, 2020 (actual, 2020 BSP Indicators 2022 BSP Indicators and programs BSP 2020) BSP 2022) Program 1: 1) Gross agricultural 31.5 1) Gross agricultural 29.9 Increasing product (million tons) product (million tons) productivity, 2) Ratio (%) of crop 64.5 2) Ratio (%) of crop 63.4 diversifying diversification diversification agriculture and agribusiness 3) Export of agricultural 6.3 3) Export of agricultural 5.8 products (million tons) products (million tons) Program 2: 1) Gross meat product 325 1) Gross meat product 256 Promoting animal Increase (thousand tons) Increase (thousand tons) production and health 2) Increased rate of 20.7 2) Increased rate of 35 commercial breeding (%) commercial breeding (%) 3) The rate of raising 23 3) Reduce the rate of 2.8 cattle has increased. animal diseases to (%) Program 3: 1. Natural catch 600 1. Natural catch 600 Fisheries resource (thousand tons) (thousand tons) management 2. Aquaculture ratio 36 2. Aquaculture ratio 36 and aquaculture relative to total catch (%) relative to total catch (%) development Program 4: 1. Public and private 20,000 1. Quantity of trees 225 Management and plantation area (hectares) harvested (cubic meters) development of 2. Community forestry 20 2. Community forestry 12 forest and wildlife income rate is increasing income rate is increasing resources Program 5: 1. Activity ratio (%) for 95 1. Activity ratio (%) for 88 Increasing program budget program budget effectiveness of 2. Percentage of students 95 2. Percentage of students 94 supporting services who get a job or create who get a job or create and human resource a new business after a new business after development graduation (%) graduation (%) 3. Effectiveness ratio of 75 3. Effectiveness ratio of 57 auditee improvement auditee improvement recommendations (%) recommendations (%) Source: adapted from BSP (2022-2024) 124 Cambodia Public Finance Review - From Spending More to Spending Better to another can be cross-referenced across BSPs frameworks. Any introduction of performance-related (for the last year implemented only), it is impossible guidelines should seek to ensure that the capacity to to determine whether a target has been met for design, monitor, report, and evaluate performance indicators that have been dropped. Furthermore, information is enhanced within the MEF responsible the BSP does not comment on whether its own agencies. targets have been met nor provides a summary, such as the number of indicators met within a given The internal audit currently plays a very limited program. Consequently, this makes the performance role in performance. The role and responsibilities measurement process hard to manage and holds of internal audit staff are not clearly defined. little operational value in informing budget decisions. Strengthening the internal control framework could strengthen the role of audits in monitoring Table 4.4 illustrates some of the weaknesses in performance, the quality and credibility of such performance matrix design and achievement against audits are essential to safeguard the integrity of the indicators. It compares the 13 program-level budget processes. Currently, MAFF audit reports by yearly indicators (excluding the “final result” indicators) the National Audit Authority are not publicly available. between the two most recent BSPs. Two of the 13 Further analysis is required to identify constraints to program-level indicators have changed, therefore, effective internal audit. comparison could only be made for the remaining 11. Only three of these remaining indicators have Implementation been met. Relatively better performance is observed across subprogram indicators, where 28 indicators MAFF’s pilot of PB emphasizes the tension have been met out of 58 that could be directly in budgeting choices for Cambodia’s control measured between the BSPs. Importantly, the systems. Transition to PB highlights the choice table affirms the challenges of outcome attribution between traditional budgeting focus on input controls when many of the performance indicator targets are with flexibility in outputs and greater flexibility on the not within reasonable reach of MAFF’s budgetary input side with stricter expectations in output. In resources. many countries, this choice can be followed by a perceived tension between controlling expenditure Performance information is not used in on the one hand and letting managers manage budget negotiations, but BSP experience with on the other. Over the long term, the objectives of performance metrics provides a platform to performance accountability assume that managers build toward performance-informed budgeting. have reasonable discretion over the use of the Performance information is also not used during resources provided, without which it would be budget preparation and implementation. At present, impractical to hold managers to account. budget negotiations in MEF include unstructured references to the performance information. The Implementation of Cambodia’s budget current reform efforts to introduce a PIB framework commitment and virement rules contribute provide an opportunity to structure the procedures to the unpredictability of funds and contain and methods that guide the design, monitoring the transition towards greater flexibility for and reporting of performance information within budgetary inputs. The virement rules leave little the BSP, and help address the weaknesses discretion to the line ministries and line managers to pointed out above. Critically, implementation of any move funds within the ministry budget. Some evidence performance framework would require strengthened suggests that virements between sub-accounts could administrative capacity. MAFF officials have pointed take as long as two quarters. This disincentivizes line to limited administrative and technical capacity in managers from applying for virement during budget the design and implementation of performance execution of a given program. Similar cumbersome Cambodia Public Finance Review - From Spending More to Spending Better 125 and lengthy processes are observed in the release reduce the time taken up by approval processes. of funds for program activities. Consequently, these Business process review in payment processes by create budget execution rigidities that can lead to MEF can serve as an example to approach the issue. expenditure delays and discounting of unforeseen program needs, which can be particularly costly in the Misalignment between development partner agriculture sector due to the higher time sensitivity of and government rules and budget calendars some expenditure items. Deepening performance- compounds the issue. The unpredictability of informed budgeting reforms with existing virement external funds owing to different, and sometimes rules risks overwhelming MAFF and other LMs complex, sets of rules complicate the implementation with additional reporting without meaningful gains of projects. This is particularly important in agriculture in budget effectiveness and service delivery. The and for MAFF, given the overwhelming share of misalignment between the PB budgets and line-item capital projects funded by development partners and controls could have costly results, with resources the declining outturns in externally financed projects, being overconsumed by some spending units and as illustrated in the previous section. under-consumed by other units, with consequences for delivery and performance objectives. Ultimately, 4.6. Recommendations program managers have little discretion over funds at the program level, thus impacting their ability to The introduction of the Budget Strategic Plan deliver on program objectives. Attempts to hold (BSP) and the piloting of Program Budgeting program managers to account would necessarily (PB) are important strategic steps under the need to re-visit the balance of control. Government’s PFMRP to improve the quality of public spending. MAFF’s experience with the Tender processes can last for months, reforms demonstrates that these steps have helped introducing unpredictability of fund flow, and strengthen elements of the planning process, reducing time to execute programs, which are but are yet to be integrated into budgeting and time-sensitive to seasonality, thereby impacting budget management processes. Some of the key the agricultural sector. Funds remaining unspent recommendations are in line with earlier PER’s at the end of the fiscal year are surrendered. These recommendations, which have been partially delays could be explained by the misinterpretation of implemented (see Box 4.1). existing regulations and/or lengthy approvals for the procurement process and budget requests – both To support successful implementation of the cited as reasons for delayed disbursement of funds, performance-informed budgeting (PIB) under the impacting implementation and budget credibility. 97 PFMRP toward improving spending outcomes, it Specifics around the rules and functions guiding will be necessary to (i) strengthen of BSP content; the procurement process and release of funds fall (ii) align BSP documentation and processes with the outside of the report’s scope but could warrant further annual budget and externally funded projects; and (iii) examination. A detailed business process review of integrate BSP and program logic into negotiations, procurement-related approvals across MAFF and appropriation, monitoring, and evaluation, and the MEF would help identify the bottlenecks (whether legal framework (Table 4.5). misinterpretation of procedures or the processes themselves) and actions that could be taken to 97 A large number of activities across ministries means that it can take the budget controller several months to complete the process of allocating at this level (sometimes referred to as “loading the budget”). Experience suggests that during the period while this work is being done, line managers are either starved of actual funding or operate on unpredictable interim releases. This can have an adverse effect on program and subprogram service delivery (World Bank 2016) 126 Cambodia Public Finance Review - From Spending More to Spending Better Box 4.1. Agriculture in the 2019 Public Expenditure Review This box summarizes key findings and recommendations related to agriculture and irrigation in the previous Public Expenditure Review for Cambodia (World Bank, 2019). The previous PER focused on drivers of inclusive growth nurtured through the provision of core public goods and provided policy options aimed at orientating spending towards these broader goals. The PER recommended to: Continue consolidating the number of BSP subprograms and improving alignment with national priorities, while systematically collecting and monitoring performance data (partially implemented). The process of strengthening the BPS and program budgeting process is ongoing, and many improvements have been made. For MAFF, and the agricultural sector more broadly, challenges remain in the inclusion of external donor financing into PFM systems and reforms. The performance matrix has been strengthened with the regular collection and reporting on indicators. However, gaps remain, not least unrealistic and unclear outcome indicators. The design of the performance matrix and its operational value throughout the budgeting cycle could be improved. Enhance strategic engagement and coordination among sector actors by revamping the TWGAW (partially implemented). Coordination has been strengthened between public sector actors through the inter-ministerial committees, however, more could be done including through the efforts to prioritize and integrate potential DP pipeline projects into budget planning. The Technical Working Group on Agriculture and Water (TWGAW) would be the natural forum for discussion. Undertake an institutional assessment of current capabilities and future resource needs in inspection services, plant and animal health, and water usage (not implemented). An institutional assessment of capabilities has not been undertaken since the previous PER. Elaborate on a national agricultural research strategy addressing research priorities as well as institutional and financing matters (not implemented). There is low level of spending on research and education. Horticulture Development Policy, which includes research priorities, has been drafted with the assistance of the USAID Project but remains to be finalized and adopted by MAFF. Undertake regular project unit-cost analysis, combining the information with that in the CISIS (on condition of schemes), and use it to support informed decision-making (not implemented). Projects are formulated by development partners with close consultations with MAFF and MEF responsible units. Source: 2019 Public Expenditure Review for Cambodia; World Bank staff Cambodia Public Finance Review - From Spending More to Spending Better 127 Table 4.5. Options to strengthen budget management in agriculture Options Challenge Medium-term recommendations Short-term recommendations (1-2 years) (3-5 years) PB design Improve design of PB, focusing on better Further strengthen PB design challenges and costing, while aligning PB with BSP. For by linking it with BSP to deliver weak link between example: agriculture policy objectives, BSP and the PB. • Improve realism of BSP outcome administrative or support Budget negotiations indicators services which are prioritized do not follow the • Review and refine PB structure to capture Align BSP to the formally program logic all relevant expenditure associated with adopted MTBF and MTFF specific program including personnel cost Explicit costing of new proposed and capital expenditure subprograms, activities, policy • Develop guidelines to clarify the roles initiatives, capital investments, and responsibilities of PDAFFs in and required funding to continue the development of the MAFF BSP; with existing activities. This strengthen information-sharing protocols could include expenditures related between PDAFFs to MAFF at the center to risks such as climate events Weak LM Consider adjusting timelines of project Pilot an indicative ministry-level involvement in screening and selection processes to inform annual capital budget envelopes capital budget MAFF’s planning functions and development over a 3-year medium-term planning of the BSP. This may involve moving horizon for MAFF (or another capital budget planning forward and earlier prioritized LM). Align timing determination of ministry-level indicative with the preparation of the BSP capital budget allocations (issuance of circular) Strengthen agriculture-specific sector-wide cross-ministry coordination of externally funded capital projects in the three key sector ministries through the recently formed Inter- Ministerial Committees (IMCs) Complex and Reduce and simplify the number of Align with the Sectoral Climate impractical performance indicators (e.g., 5-15 key Change Strategic Plan (SCCSP) performance matrix indicators). Balance indicators between and strategic objectives of the outputs and outcomes Cambodia Climate Change Strategic Plan (CCCSP), the Encourage consistency in the indicators from performance matrix could one BSP to another include 1 (or) 2 climate change indicators MAFF to conduct a comprehensive annual review of performance against output and outcome targets as reported in the BSP. This could be complemented by the development of a mechanism to utilize the results as input in the subsequent BSP Re-design performance indicators to reflect more realistic and verifiable targets that correspond to available resources 128 Cambodia Public Finance Review - From Spending More to Spending Better Options Challenge Medium-term recommendations Short-term recommendations (1-2 years) (3-5 years) Build capacity on the design, monitoring, reporting, and use of performance metrics (not limited to MAFF) Strict virement rules MEF and MAFF to jointly conduct business Annual budget appropriations and lengthy budget process review(s) and improvement for on a program basis. At the release budget commitment and procurement spending unit levels, funds process should not, as far as possible, be allocated by activity or line- Delegation of authority to support managerial item basis; instead, spending flexibility in budget execution – i.e., discretion units should preferably operate over virements between some expenditure on a “global budget” with the items during implementation, particularly flexibility to allocate funds across those under the same subprogram. Spending activities and line items where units would report on the use of funds to the necessary subprograms. Subprogram managers would The budget execution stage report upward on program execution. should support the use of MAFF to prepare the procurement plan resources consistent with the in advance and adjust timelines for the budget presentation passed necessary documentation along with the by the legislature. Mechanism procurement calendar needs to be devised to provide direct access to approved budgetary resources in the TSA supported by FMIS Internal control, Strengthen MAFF Internal Audit and Develop and implement Internal performance audit, Inspection Teams’ capacity, roles, and control framework roadmap. and oversight responsibilities for effective internal budget Annually publish the MAFF control and focus on performance including audit report which includes deliveries of prioritized agriculture policy performance audit, targeted objectives, administrative or support services under Stage 4/Platform 4 of the PFMRP Cambodia Public Finance Review - From Spending More to Spending Better 129 5 IMPROVING CAPITAL BUDGET MANAGEMENT FOR IMPROVED IRRIGATION AND WATER RESOURCES MANAGEMENT 5.1. Introduction traditional agriculture to modern agricultural practices. As a part of such transformation, Cambodia needs Irrigation and water management are essential to go beyond rice to focus on other high value crops to promoting agricultural productivity in such as maize, cassava, mung beans, mango, Cambodia. Improved water resource management cashew, pepper, and vegetables.99 and access to water service, which are a precondition for switching from rain-fed to irrigated agriculture, More effective budget management could can raise rice yields by up to 125 percent.98 The total facilitate such a transformation by increasing arable land area in Cambodia currently is about 4.5 irrigation services’ and water resource’s million hectares (ha) of which 4 million ha are used contribution to the agriculture sector and for seasonal crops, mostly rice, and 0.5 million ha beyond. Irrigation and water resource management are under permanent crops (rubber, fruit trees, absorbed a major share of total expenditure for commercial woodland). In 2016, a total of 1.5 million the agriculture sector. However, there are major ha were irrigated, 1.1 million ha in the wet season challenges in front of effective budget management and 0.5 million ha in the dry season. in the sector: Improved irrigation and water resource • Coordination challenges between MOWRAM management can also provide opportunities and MAFF, including the lack of clear delineation for diversification to different crops other than of roles and responsibilities between the two rice. Cambodia needs to embrace a structural ministries, and duplication of effort reduce the transformation of the sector from one based on efficiency of funding allocation. 98 (ADB, 2014) (JICA, 2009). It is also recognized also that, in addition to good water management, modern farming practices are an important determinant of productivity (World Bank, 2015). Cambodia needs to introduce modern farming practices, such as controlled irrigation, machinery, improved seeds, and chemical fertilizers to achieve higher yields. 99 See for instance (Gelotti & Sin, 2016) 130 Cambodia Public Finance Review - From Spending More to Spending Better • The lack of linkages between planning and facilitated recessional cultivation at the end of and budgeting creates challenges in the the wet season. However, they were poorly designed implementation of program budgeting. and incomplete, and could only provide irrigation for • Limited planning practices and poor quality of a small dry season crop area at best. medium-term investment estimates are key challenges for expenditure execution. A majority of the current functional schemes • Limited deconcentration within MOWRAM has are either medium or small size. As of 2022, resulted in limited availability of financial and there are 2,355 irrigation schemes in Cambodia human resource at the subnational levels for categorized as small (50 to <200 ha), medium (200 frontline service delivery. to 5,000 ha) and large (>5,000 ha). Among these, 1,131 (48 percent) are small, 1,171 (50 percent) are 5.2. Sectoral context medium, and the other 53 (2 percent) are large.102 These schemes altogether cover an annual irrigated Shortcomings in irrigation infrastructure area of over 1.5 million ha, about 498,200 ha by limit agriculture productivity and crop export large schemes (32 percent), 931,900 ha by medium potential. Reliance on rainfed agriculture, limited (60 percent), and 131,290 ha (8 percent) by small- diversified crops to meet broader market demands, scale schemes.103 and higher export and electricity costs compared to neighboring countries limit the full potential of Climate change is a key factor affecting the agriculture land use and productivity.100 irrigation and water resource management sector. It intensifies storms and extends the number Variable water availability has been a key of consecutive dry days during both dry and wet challenge to the irrigation sector development seasons. Existing data indicate that Cambodia will in Cambodia. Although Cambodia has abundant experience a generally wetter climate with longer water resources, its availability is highly variable both dry periods in the future. During the dry season, temporally and spatially. In the wet season, there is the potential impact of climate change will likely ample water, though supplementary irrigation may exacerbate water scarcity for certain parts of the be required to avoid crop moisture deficits between country. During the wet season and across the rainfall events. In the dry season, many rivers and country, it is projected that more severe, frequent, streams dry up, especially in the northwest of the and longer-duration flooding will occur, especially in country, unless there are storage reservoirs in the the Mekong Delta areas.104 upper catchments.101 5.2.1. Irrigation and water resource Much of the public investment in irrigation management sectoral policy has been to rehabilitate and maintain the old schemes built during the 1970s. The majority of Policy-making for irrigation is complicated irrigation schemes that were built during the period by the many agencies involved and by of civil unrest were created by constructing 1-2 meter the inadequacy and ineffectiveness in the high earth bunds across the predominantly flat slope distribution of responsibilities across levels of areas. These structures enabled water to be stored government. MOWRAM has been the lead agency during the wet season for supplementary irrigation for implementing policies, but the division of labor 100 (RGC, 2019) 101 (MOWRAM, 2019) 102 Data provided by MOWRAM (2023), Cambodia Irrigation Scheme Information System (CISIS) database 103 (MOWRAM/AFD, 2016) 104 (MOWRAM, 2019) Cambodia Public Finance Review - From Spending More to Spending Better 131 among the different tiers within the ministry needs two ministries and participation from development more clarity. More importantly, there is no systematic partners. Various sub-committees have been set up coordination between MOWRAM at the provincial under the TWG, including one related to irrigation level with other relevant agencies. management.107 The RGC has a comprehensive range of MOWRAM has adopted several key policies policies on developing the irrigation sector to and strategies, including the National Water better support agricultural development. In the Resource Policy (2004), MOWRAM’s five-year medium- and long-term development strategies, Strategic Development Plan 2019-2023, the such as the RS4 and the NSDP (2019-2023), Strategic Framework for the Irrigation Sector (2017), priorities are given to promoting crop diversification, and the National Water Resources Management and increasing the number of operational irrigation Sustainable Irrigation Road Map and Investment schemes, constructing storage reservoirs, enhancing Program 2019-2033 (2019). These policies refer to standards for and quality of irrigation infrastructure, the RS4 and the NSDP (2019-2023), and MOWRAM’s improving the effectiveness of investment, legal mandate108 to justify their focus and priorities. strengthening government institutions involved in In line with the government’s program-based irrigation, solidifying the operations and maintenance budgeting approach, these policies, particularly (O&M) of irrigation systems, and supporting farmer MOWRAM’s 5-year Strategic Development Plan participation in O&M of irrigation schemes.105 2019-2023, have also been used to guide the 3-year Project Investment Plan (PIP) and Budget Strategic These policy objectives and strategies have Plan (BSP), which in turn determine annual budget been translated into programs of the ministry. formulation. MOWRAM’s program structures, which have remained unchanged since 2019, are presented The five-year Strategic Water Resource in the BSP and annual program budget. There are Development Plan (2019-2023) sets out five programs, each with a specific objective and five strategic areas. These are 1) improving divided into subprograms.106 In addition, since 2020, administrative management and human resource the ministry has introduced functional classification development, 2) managing and developing water according to IMF Government Financial Statistics resources, including sustainable irrigation, 3) (GFS), matching each to the relevant subprogram. managing floods and droughts, 4) managing The primary governmental stakeholders data on water resources and meteorology, and 5) engaged in the irrigation sector are MOWRAM preservation and conservation of water resources. and the Ministry of Agriculture, Forestry and For each strategic area and sub-areas, specific Fisheries (MAFF). MOWRAM has responsibility indicators, and targets (mostly in output terms) for the irrigation water supply while MAFF has were identified for each year (Table 5.1). There has responsibility for assisting farmers in making the best been strong emphasis on O&M and rehabilitation of use of water. To facilitate better co-ordination between irrigation schemes and less on overall water resource the two ministries, the government established management; this is reflected in expenditure analysis the Technical Working Group on Agriculture and in the next section. Water (TWGAW), with co-chairs from each of the 105 (RGC, 2019) (RGC, 2019) 106 P1. Management and development of water resources, P2. Management of flood and drought, P3. Preservation and conservation of water resources; P4. Management of data on water resource and meteorology; P5. General administration 107 (World Bank, 2017) 108 It is based on Sub-decree # 58 dated 30 June 1999. 132 Cambodia Public Finance Review - From Spending More to Spending Better Table 5.1. Key indicators and targets for 2019-2023 of MOWRAM No. Strategic areas and indicators Unit Targets Notes 1 Improving administrative management and human resource development • No. of staff trained No. 3,000 In & outside country • No. of workshops & training No. 150 In-country 2 Managing and developing water resources, including sustainable irrigation services • Increased irrigated areas for rice ha 150,000 Dry and rainy seasons • Increased irrigated areas for other crops ha 2,500 • Established functioning FWUC No. 35 From 548 in 2018 3 Managing floods and droughts • Reduced and prevented floods ha None • Reduced and prevented salt seawater intrusion ha None • Prevented rice farming from floods and droughts ha 450,000 Dry and rainy season • Constructed pumping stations No. 25 • Repaired pumping stations No. 45 • Repaired water pumps No. 90 4 Managing data on water resources and meteorology • Installed water resource monitoring stations No. 85 From 44 in 2018 • Repaired water resource monitoring stations No. 60 • Installed water quality monitoring stations No. 5 From 19 in 2018 • Installed hydrological and meteorological No. 25 From 65 in 2018 ( Hydromel) stations • Repaired hydrological and meteorological No. 60 (Hydromel) stations 5 Preserving and conserving water resources • Drafting and amendment of sub-decrees and Various legal other regulations relating to water resource documents management, • Enforcement of implementation of endorsed legal documents Source: MOWRAM (2019) Note: No.=Number; ha= Hectare. 5.2.2. Institutional context The PDWRAM are responsible for monitoring water resources and supporting the management of water MOWRAM’s administrative structure is organized infrastructure. A typical PDWRAM has a few technical around departments for implementing the offices and no representation at the lower DMK level. strategic plan. At the national level, there is the The offices under PDWRAM include 1) Office of General Department of Administration, responsible Administration and Personnel, 2) Office of Irrigated for finance issues, the General Department of Agriculture, 3) Office of Water Resource Management Technical Affairs, the Tonle Sap Authority, and the Cambodia National Mekong Committee (CNMC). At and Conservation, 4) Office of Meteorology, and 5) the provincial level, there are Provincial Departments Office of Water Supply and Sanitation. It is common of Water Resources and Management (PDWRAM). that each office has just one or two staff members.109 109 This is based PDWRAM BSPs for various years. Cambodia Public Finance Review - From Spending More to Spending Better 133 Sustainable Farmer Water User Communities 5.3. Public expenditure in (FWUCs) are integral parts of the institutional setup for the irrigation sector. In 2015, the irrigation and water resource Government adopted a Sub-decree and guidelines Financing for irrigation and water resources that set out principles, objectives, and activities for comes from both the national budget and achieving sustainable FWUCs. This includes activities external financing. There is a heavy reliance on for establishing new FWUCs and strengthening external financing, which accounts more than half of established FWUCs at the national, sub-national, total MOWRAM spending. Overall spending declined and FWUC levels. FWUC budget request guidelines in 2019 and 2021 due to a reduction in DP-financed are also included to assist FWUCs in sourcing funds capital spending. Government domestic investment from MOWRAM’s annual FWUC budget to support remained flat at 0.3 percent of GDP between 2017 the establishment and strengthening (e.g., training) and 2021. DP-funded capital spending declined of FWUCs. from 0.6 percent of GDP in 2017 to 0.5 percent of GDP in 2021 (Figure 5.1 and Figure 5.2). Sub-national administrations (SNAs) are supposed to play critical role in water resource Domestically financed investment expenditure management and irrigation O&M. In 2019, the deviate substantially from original budget (Figure Government took a significant step in transferring 5.3). MOWRAM experiences sizeable fluctuations various tasks to the newly integrated DMK in budget execution because of in-year budget administrations.110 SNAs at all three levels have also adjustments. Every year MEF allocates part of the contributed part of their budget to the irrigation sector unallocated budget to MOWRAM to meet ‘emergency operations. SNAs have been playing important roles needs’ to repair irrigation schemes affected by annual in supporting the FWUCs and working with private floods. This in-year budget increase applies mainly to sector to deliver irrigation O&M and water supply the investment expenditure budget (Chapter 21). A services. Despite the transfer of O&M functions, large but unpredictable amount of budget allocated budget allocation for O&M at sub-national agencies during the in-year implementation can undermine and/or community organizations remain very limited. allocative efficiency because it does not fully follow MOWRAM has made little progress in transferring the due process of budget allocation stages and the functions and resources to SNAs. decisions tend to be made under time constraint. Figure 5.1. Trends of MOWRAM’s spending Figure 5.2. Economic composition (% of GDP) (2017-2021) MOWRAM’s spending (% of total) (2017-2021) Capital, externally nanced 100% 1.2% Capital, domestically nanced 90% Recurrent 80% 1.0% 70% 64% 65% 69% 71% 68% 60% 0.8% % of GDP 0.6% 0.6% 50% 0.4% 0.5% 0.6% 0.5% 40% 30% 0.4% 28% 27% 20% 25% 25% 0.3% 0.3% 22% 0.3% 0.3% 0.3% 10% 0.2% 6% 6% 8% 7% 8% 0% 0.1% 0.1% 0.1% 0.1% 0.1% 2017 2018 2019 2020 2021 0% 2017 2018 2019 2020 2021 Wages Non-wage recurrent Capital, domestically nanced Sources: Budget Settlement Law (2017-2021); World Bank staff estimates 110 (RGC, 2019) 134 Cambodia Public Finance Review - From Spending More to Spending Better Figure 5.3. Expenditure outturns compared to original investment expenditure budget (Chapter 21) (2017-2021) 250% % of approved budget 212% 200% 174% 150% 132% 130% 100% ........................................................................................................................... 100% 50% % of approved budget 0% 2017 2018 2019 2020 2021 Sources: Budget Law (2017-2021) and Budget Settlement Law (2017-2021); World Bank staff estimates A large share of the MOWRAM expenditure was is small and managed mainly by the central ministry. for the irrigation sector and a limited amount Programs 2, 3, and 4, which are more concerned was spent on water resource management. with water management receive a small amount of Among the five programs, Program 1 has received the budget. About 15 percent to 20 percent of the the highest budget, followed by Program 5. Program budget has gone to Program 5. As Cambodia has 1 absorbed over 75 percent of the total budget in persistently suffered from the problem of too much the last five years and spent under seven sub- water during rainy seasons and too little water during programs, six of which are under the central control dry seasons, more attention on water management of MOWRAM (Figure 5.4). The largest of them are is necessary. the irrigation scheme rehabilitation and O&M sub- programs. Program 1 also covers activities to At the subprogram level (SP), rehabilitation establish and support FWUC, but the FWUC portion of large-scale irrigation schemes (SP1.2), a Figure 5.4. MOWRAM budget by programs (2017-2022) 100% 90% 16% 16% 17% 18% 20% 20% 80% 70% 60% 50% 40% 80% 80% 79% 78% 77% 77% 30% 20% 10% 0% 2017 2018 2019 2020 2021 2022 P5: General administration P4: Management of data on water resource and meteorology P3: Preservation and conservation of water resources P2: Management of ood and drought P1: Management and development of water resources Source: MEF (2023) Cambodia Public Finance Review - From Spending More to Spending Better 135 Table 5.2. Budget by key subprograms Strategic areas and indicators 2017 2018 2019 2020 2021 2022 SP1.1: Irrigation O&M 20% 19% 23% 23% 27% 27% SP1.2: Irrigation rehabilitation and 55% 54% 50% 49% 43% 43% construction SP1.3: Establishment of FWUC 0% 0% 0% 0% 0% 0% SP1.4: Managing FWUC sustainably (it 2% 3% 3% 3% 3% 3% was SP3.2 in 2017) SP1.7: Irrigation O&M by PDRAM 3% 3% 3% 3% 4% 4% SP2.1: Intervention to help pump water 1% 1% 1% 1% 1% 1% during drought and natural disaster SP2.2: Construct, O&M on mobile 0% 0% 0% 0% 0% 0% pumps and pumping stations SP2.3: Flood, and drought management 1% 0% 0% 0% 0% 0% by PDRAM SP3.1/3.2: Preservation and 0% 1% 1% 1% 1% 1% conservation of water resources Source: MOWRAM (2019) technical task under the responsibility of the budget (93 percent) is used for agriculture Department of Engineering within MOWRAM (Figure 5.5). The cooperation between MOWRAM has received the largest share in the previous and MAFF at both the national and subnational levels years (Table 5.2). The next biggest subprogram is is critical for achieving allocative efficiency. As such, O&M of irrigation schemes (SP1.1), which is also it is essential that MOWRAM and MAFF function in the responsibility of the Department of Agriculture close coordination at both the policy and institutional Irrigation at the central level. The third biggest levels. is SP1.7, which is the O&M budget for irrigation schemes (medium and small) is the responsibility of The largest expenditure item is investment PDWRAM. expenditure (Chapter 21), which is used for rehabilitating large-scale irrigation schemes, currently In terms of functions, nearly all of MOWRAM’s managed under the central Ministry (General Figure 5.5. MOWRAM budget by functions (KHR, million) 310,000 94% 96% 93% 260,000 KHR, million 210,000 160,000 110,000 60,000 1% 2% 2% 1% 2% 2% 0% 1% 1% 0% 2% 0% 3% 0% 10,000 70421: 70540: 70560: 70620: 70630: 70950: 70980: Agriculture Protection of Environment Community Water Education Education biodiversity protection development supply n.e.c 2020 2021 2022 Source: MEF (2023) 136 Cambodia Public Finance Review - From Spending More to Spending Better Figure 5.6. MOWRAM spending by economic classification (% of total) (2017-2021) 100% 90% 80% 70% 57% 60% 59% 61% 55% 60% 50% 40% 8% 9% 10% 30% 9% 9% 20% 27% 30% 25% 27% 26% 10% 0% 2017 2018 2019 2020 2021 Chapter 60: Purchases Chapter 61: Services Chapter 62: Social bene ts expenses Chapter 64: Payroll expenses Chapter 65: Subsidy expenses Chapter 63: Taxes and tariffs Chapter 21: Invetsment expenditure Sources: Budget Settlement Law (2017-2021); World Bank staff estimates Department of Engineering). Irrigation O&M captured government investment (Chapter 21) as a share of under service expenses (Chapter 61) comes second, total domestic spending slightly increased from 57 followed by personnel expenditure (Chapter 64) percent to 61 percent. Wage bill and O&M were (Figure 5.6). stable during the same period. However, in 2021, wage bill and O&M respectively increased to 10 Effective O&M is critical to ensuring the percent and 30 percent. This was offset by a decline continued functioning of irrigation schemes. in government investment (Chapter 21), reaching 55 The Government has ensured that MOWRAM takes percent in 2021 (Figure 5.6). on O&M jobs for newly completed schemes and has them budgeted under service expenses (Chapter 5.3.1. Central versus PDWRAM 61). At the same time, it has also allocated part of the expenditure allocation national budget for investment expenditure (Chapter 21) to rehabilitate irrigation schemes if the cost is Central level expenditure has taken up the largest less than US$3 million. Between 2017 and 2020, share of total MOWRAM’s budget.111 This is mainly Figure 5.7. Central vs PDWRAM spending Figure 5.8. Central vs PDWRAM spending (% (KHR, million) (2017-2021) of total) (2017-2021) 600,000 100% 7% 7% 8% 8% 8% 90% 500,000 80% 70% KHR, million 400,000 60% 300,000 50% 93% 93% 92% 92% 92% 40% 200,000 30% 20% 100,000 10% 0 0% 2017 2018 2019 2020 2021 2017 2018 2019 2020 2021 Central Provincial Central Provincial Sources: Budget Settlement Law (2017-2021); World Bank staff estimates Note: Capital spending integrated in central spending The Department of Engineering plays the most important role as a force account to implement newly designed and rehabilitation 111 irrigation projects. The provincial level has very limited capital and human resources for the time being. Cambodia Public Finance Review - From Spending More to Spending Better 137 Figure 5.9. Central vs PDWRAM personnel number (2017-2021) 100% 90% 80% 38% 37% 37% 37% 37% 70% 60% 50% 40% 30% 62% 63% 63% 63% 63% 20% 10% 0% 2017 2018 2019 2020 2021 Personnel at central MOWRWAM Personnel at PDWRAM Source: MOWRAM (2023) Figure 5.10. Central vs PDARAM spending – Wage expenditure (% of total) (2017-2021) 100% 90% 80% 40.2% 45.5% 46.2% 45.9% 43.6% 70% 60% 50% 40% 30% 59.8% 54.5% 53.8% 54.1% 56.4% 20% 10% 0% 2017 2018 2019 2020 2021 Central (Wages) PDRAM (Wages) Sources: Budget Settlement Law (2017-2021); World Bank staff estimates because of investment expenditure (Chapter 21), total of over 1,400 staff members in the ministry, with which is managed at the central level and has the the rest spread across 25 PDWRAMs (Figure 5.9). largest chunk of the total MOWRAM spending. MOWRAM’s central spending slightly declined from The shortage of staff at the PDWRAM level is 93 percent of MOWRAM spending in 2017 to 92 significant. Based on a 2017 assessment, each percent 2022, while PDWRAM spending marginally PDWRAM has around 25 staff on average. For a increased from 7 percent of total MOWRAM large irrigation scheme (e.g., a 40,000-ha scheme), spending in 2017 to 8 percent in 2022 (Figure 5.7 a minimum of 62 staff is estimated as being required, and Figure 5.8). more than double the average staff number of a PDWRAM, and that is for only one scheme.112 Scaling Within the recurrent budget, the central level up to the provincial level, Siem Reap’s PDWRAM also accounts for a significantly larger share. could potentially require over 200 staff, but has only For example, the central level accounted for over 50 40 staff, of which 33 are permanent, and the other percent of the wage bill between 2017 and 2022, seven are contracted. Only 18 of these staff have primarily because it controls over 60 percent of the irrigation scheme O&M responsibilities. This shortage 112 (World Bank, 2017) 138 Cambodia Public Finance Review - From Spending More to Spending Better Figure 5.11. Central vs PDRAM spending – Non-wage recurrent spending (% of total) (2017-2021) 100% 90% 19% 19% 16% 17% 18% 80% 70% 60% 50% 40% 81% 81% 84% 83% 82% 30% 20% 10% 0% 2017 2018 2019 2020 2021 Central (Non-wage recurrent) PDRAM (Non-wage recurrent) Sources: Budget Settlement Law (2017-2021); World Bank staff estimates of staff at PDWRAM has been recognized but has whereas the national budget focuses more on yet been resolved. 113 More than 50 percent of wage medium-scale rehabilitation (US$1-3 million). expenses are at central level (Figure 5.10). The ministry’s Department of Irrigated Agriculture is responsible for the O&M of large-scale irrigation The vast majority of non-wage spending has schemes after they are completed, while PDWRAMs also been at the central level (Figure 5.11). The oversee the O&M of smaller-scale schemes. This central level controls more than 80 percent of the division of labor, while understood among key total non-wage recurrent spending. This has resulted stakeholders, is not put into writing as a reference. largely from the fact that the central level controls Therefore, it creates uncertainty on roles and most of service expenses (Chapter 61) intended for responsibilities, especially for PDWRAM. O&M activities. The irrigation rehabilitation budget is registered 5.3.2. Expenditure in selected under investment expenditure (Chapter 21) and is placed under the Department of Engineering subprograms and priority areas (Table 5.3). In addition, there are some resources Irrigation rehabilitation and construction under purchase and service expenses (Chapter 60 and Chapter 61), but these are small amounts. The Rehabilitation of large-scale irrigation schemes department is directly involved in the whole process (over US$5 million) rely more on external-funds, of planning and implementing projects funded by Table 5.3. Budget allocation for irrigation rehabilitation (KHR, million) 2017 2018 2019 2020 2021 2022 SP1.2: Total 145,342 155,127 158,161 158,336 120,077 120,057 Irrigation rehabilitation Chapter 21: Investment 145,000 155,000 158,028 158,203 120,000 120,000 and expenditure construction Chapter 60: Purchases 24 9 15 15 12 11 – MOWRAM Chapter 61: Services 318 118 118 118 66 46 Source: MOWRAM (2023) 113 (World Bank, 2017) Cambodia Public Finance Review - From Spending More to Spending Better 139 Table 5.4. Budget allocation for irrigation O&M, by level of administration (KHR, million) 2017 2018 2019 2020 2021 2022 SP1.1: Total 53,073 54,593 73,108 74,110 74,068 74,056 Irrigation O&M – Chapter 60: Purchases 24 44 59 59 32 26 MOWRAM Chapter 61: Services 53,049 54,549 73,049 74,051 74,036 74,030 SP1.7: Total 8,532 9,794 10,015 10,453 10,352 10,795 Irrigation O&M – Chapter 60: Purchases 22 22 20 22 13 24 PDRAM Chapter 61: Services 8,510 9,772 9,995 10,431 10,339 10,771 Source: MOWRAM (2023) Table 5.5. Budget allocation for establishment and support to FWUC (KHR, million) 2017 2018 2019 2020 2021 2022 SP1.3: Total 38 47 60 60 43 38 Establishment of Chapter 60: Purchases 4 5 18 18 16 15 FWUC by central MOWRAM Chapter 61: Services 35 42 42 42 28 23 SP1.4: Managing Total 4,088 7,579 9,081 8,081 8,041 8,026 FWUC sustainably Chapter 60: Purchases 5 3 5 5 3 2 by central MOWRAM Chapter 61: Services 4,084 7,576 9,076 8,076 8,038 8,024 Total 4,126 7,626 9,141 8,141 8,084 8,064 Source: MOWRAM (2023) investment expenditure (Chapter 21). In most cases, (Chapter 61) for both the central level and PDWRAMs. PDWRAMs only provide a ‘coordination role’ in The service expenses (Chapter 61) budget under irrigation rehabilitation projects. the central level is about six to seven times that of the service expenses (Chapter 61) budget for all 25 Irrigation O&M PDWRAMs combined. The budget for O&M is split between the PDWRAMs have acted as ‘contractors’ for MOWRAM and PDRAMs (Table 5.4). Nearly all MOWRAM for some of the O&M work. The of the O&M is registered under service expenses working arrangement is that MOWRAM directly Table 5.6. Activities and budget under establishment of FWUC (SP1.3) and managing FWUC (SP1.4) (KHR, million) 2021 2022 60021: Office supplies 3 2 61103: Meeting 2 1 6112: Domestic mission 36 23 61055: Irrigation O&M 8,000 8,000 Total 8,041 8,026 Source: MOWRAM (2023) 140 Cambodia Public Finance Review - From Spending More to Spending Better Table 5.7. Budget allocation for emergency responses (KHR, million) 2017 2018 2019 2020 2021 2022 SP2.1: Intervention to help Total 3,850 3,750 3,750 3,750 2,245 1,682 pump water during drought and natural disaster by Chapter 60: Purchases 3,800 3,700 3,700 3,700 2,221 1,666 central MOWRAM Chapter 61: Services 50 50 50 50 25 16 SP2.2: Construct, O&M of Total 850 850 845 841 816 807 mobile pumps and pumping stations by central MOWRAM Chapter 61: Services 850 850 845 841 816 807 SP2.3: Flood and drought Total 2,143 1,333 1,254 1,144 1,173 1,075 management by PDRAM Chapter 60: Purchases 54 54 54 55 43 43 Chapter 61: Services 110 99 76 85 87 69 Chapter 62: Social 1,979 1,181 1,124 1,004 1,043 963 benfits expenses Source: MOWRAM (2023) implements the O&M work for large scale schemes and medium irrigation schemes (as shown in the and the cyclical O&M of medium schemes. For the Table 5.6). The establishment of FWUC (under SP1.3) regular O&M of medium and small schemes, however, takes only a small amount of budget, mostly to cover PDWRAMs are brought in and get paid as contractors. cost of office supplies, meetings, and mission. Such division of labor is not always clear-cut and is subject to negotiation on a case-by-case basis. Emergency responses Support to FWUC The emergency responses are covered by both the central level and PDWRAMs. For the Although FWUCs are community-based central level, the main corresponding budget item organizations, the budget relating to FWUC has is purchase expenses (Chapter 60), which involves been allocated to the central level (Table 5.5). procuring petroleum to be distributed to PDWRAMs There are two related subprograms: SP1.3 which to fuel water pumps for farmers (Table 5.7). The focuses only on establishment of FWUCs, and central level also uses a large amount of service SP1.4, which is about supporting FWUCs to perform expenses (Chapter 61) funds for the construction their jobs, a big part of which is to ensure regular and O&M of mobile pumps and pumping stations. O&M of irrigation schemes. Much of the budget for For PDWRAMs, the main budget item is social the FWUC is to provide O&M funding for mostly small benefits expenses (Chapter 62), which is spent Table 5.8. Reported outputs relating to rice versus other crops Reported outputs in irrigation coverage 2019 2020 2021 Increase in irrigated areas for rainy season rice (ha) 19,432 17,902 15,601 Increase in irrigated areas for dry season rice (ha) 9,057 10,060 6,987 Other crops (ha) 412 535 1,261 Other crops (as % of total) 1.4% 1.9% 5.3% Source: MOWRAM annual report (2019-2021) Note: ha=Hectare Cambodia Public Finance Review - From Spending More to Spending Better 141 Figure 5.12. Irrigation investment budget 2019 and poverty level in 2019/20 45 40 Budget (KHR, billion) 35 30 25 20 r² = 0.2434 15 10 5 0 0 50,000 100,000 150,000 200,000 250,000 300,000 Number of the poor Sources: Authors’ estimation based on figures from MOWRAM annual reports and CSES (2019/20) to support travel expenses for PDWRAM officials help guide its annual budget allocation. It has also and purchasing of various items needed to provide implemented program-based budgeting (PB), with support to farmers. five programs divided into subprograms and activity clusters. It has adopted the 5-Year Strategic Plan 5.3.3. Outcomes from public investment for Water Management and the Meteorology Sector (2019-2023) and the 3-year rolling Budget Strategic Most MOWRAM’s investment in irrigation Plan (BSP) to guide its annual budget allocation by benefited rice cultivation (Table 5.8). While programs and subprograms. One key challenge in rice has been the priority in the last decade, the budget management is that wages (Chapter 64) Government’s intention to diversify agricultural crops have not been assigned to each program but are to include other high value crops and export markets subsumed under one program (Program 5 – General suggests that the irrigation sector too needs to administration), similar to MAFF. This limitation has adjust its priorities. prevented a comprehensive capture of expenditure by programs and subprograms. Recent irrigation budget allocations to the provincial level seem relatively equitable. Using 5.4.2. High in year allocation for irrigation-related expenditure under investment expenditure (Chapter 21) and service expenses investment budget for emergency (Chapter 61) (which is by far the largest), Figure 5.12 expenses below suggests that the poorer the province, the There is a high in-year budget allocation for higher the budget allocated for it (r2 = 0.2434). This investment for emergency expenses that result is in line with what was found for the recurrent occur regularly. Investment expenditure (Chapter non-wage budgets of MAFF and poverty levels at the 21) budget comes from two sources: the budget provincial levels.114 allocated to MWRAM in the annual budget law and 5.4. Budget management the budget allocated in-year from the ‘unallocated budget line’ under the MEF. For both sources, 5.4.1. Program budget structure the budget is managed and implemented by the Department of Engineering under MOWRAM. MOWRAM, similar to MAFF and other ministries, MEF, through the Inter-Ministerial Committee on has adopted long- and medium-term policies to Investment Projects, has played active roles in the 114 See Agriculture Section for more detailed information on the MAFF budget allocation and poverty. 142 Cambodia Public Finance Review - From Spending More to Spending Better Table 5.9. Budget execution rate for the central MOWRAM Chapter 2017 2018 2019 2020 2021 Overall 146.8% 173.9% 99.5% 118.2% 116.6% Chapter 60: Purchases 94.8% 95.8% 94.8% 62.1% 72.1% Chapter 61: Services 98.9% 99.4% 99.2% 95.4% 95.8% Chapter 64: Payroll expenses 97.7% 103.3% 98.6% 97.2% 99.9% Chapter 21: Investment expenditure 174.0% 212.2% 99.9% 131.9% 130.2% Source: MEF Table 5.10. Budget execution rate for PDRAM Chapter 2017 2018 2019 2020 2021 Overall 96.6% 104.7% 98.5% 95.9% 95.6% Chapter 60: Purchases 95.3% 97.4% 97.5% 88.6% 95.7% Chapter 61: Services 97.6% 97.8% 98.5% 96.0% 97.6% Chapter 64: Payroll expenses 95.8% 116.8% 98.7% 98.0% 93.5% Chapter 62: Social benefits expenses 98.7% 97.9% 99.7% 96.1% 99.5% Source: MEF execution of investment expenditure (Chapter 21). Given its significance, formal processes The engagement by PDWRAM is said to be limited should be in place to guide the allocation – mostly in the form of ‘coordination’ during project and execution of the investment expenditure feasibility study and project implementation. The (Chapter 21) budget from the unallocated high over-execution rates have been due mainly budget. For MOWRAM, the main cause leading to to the in-year budget allocation to MOWRAM from the need for such in-year budget allocations is the the ‘Unallocated budget’ line item under MEF as an damages inflicted on irrigation schemes by floods emergency response budget. Given the frequency of or droughts. As the events occur annually, to some natural disaster as a result of climate change, it is extent, predictably, the expected damage and cost useful to improve planning for contingency. could be planned in budget. The possibility of such planning reduces the proportion of the emergency- There is a smooth execution of most budget line items, except for investment expenditure based in-year budget allocation and execution (Chapter 21) (Table 5.9). Overall, at both the central and, thus, increase overall budget allocation and ministry and PDWRAM, service and payroll expenses operational efficiency. (Chapter 61 and Chapter 64) have been executed mostly according to plan (Table 5.10). This was the 5.4.3. Management of investment case even during the COVID-19 pandemic – an budget indication that the Government gave priority to the welfare of civil servants during the crisis. Purchase The execution of the budget law based on expenses (Chapter 60) was the most affected during investment expenditure (Chapter 21) has the COVID-19 period, especially for the central level followed a more systematic process. The (with only 62 percent and 72 percent executed rates process is stipulated in the Standard Operation for 2020 and 2021, respectively). Procedure (SOP) for the Implementation of a National Cambodia Public Finance Review - From Spending More to Spending Better 143 Figure 5.13. The key steps in the PIM process Initial project identi cation Project feasibility and and proposal proposal Project proposal review and evaluation (Project pre-appraisal) (Project Appraisal) Project selection and Project management and Project adjustment budgeting implementation The use and operation of Project completion and ex- the completed project post review Source: MEF (2022) Funded Investment Project (2022)115 and MEF’s deserve more attention. The first is the quality Prakas # 207 on the procedures for implementing of the medium-term estimate of capital budget investment budgets from the national budget. 116 needs for the ministry. Both the methodology and The 2022 SOP outlines eight steps for the PIM the data used for such estimation exercises need cycle (Figure 5.13). Partly because it is only recent, to be continuously improved. The second area in it does not provide needed details on some of the need of attention is the integration between capital key steps. Management of investment budget could and associated recurrent budget, especially O&M. be improved for some of the steps, grouped into 1) Planning for the O&M budget should be included in project and budget preparation phase, 2) project and the long-term economic life of the projects. budget implementation, and 3) project monitoring, ex-post review and asset management phase. Inclusion and integration of O&M with the capital budget is needed for a functioning Project and budget preparation PIM system. Poor integration often leads to the planned benefits of investment not being realized in Better integration of the investment budget full because of a shortage of funding for operating at the planning stage is needed. International expenses or maintenance prematurely ending experiences suggest that investment budgets are an asset’s planned operating life.117 International best accounted for in medium-term budgeting and experience also suggests that such shortages are in the program-based budgeting process. For the particularly problematic where a large share of public MOWRAM case, progress has been made on both investment is planned and financed by donors, and fronts. Since 2021, the capital budget has been inadequate attention is paid to the financial operation included in the BSP and integrated into respective and maintenance through domestic financial programs and subprograms. Two specific areas resources. 118 115 (MEF, 2022) 116 (MEF, 2022) 117 (World Bank, 2020) 118 (World Bank, 2020) 144 Cambodia Public Finance Review - From Spending More to Spending Better Table 5.11. Number of investment projects by types of procurement methods used Procurement methods used 2017 2018 2019 2020 2021 Bids by international competition 0 0 0 0 0 Bids by domestic competition 2 0 2 4 0 Price consulting 0 0 0 0 0 Price surveys 0 0 0 0 0 Separate procurement (i.e., internal) 34 52 34 35 38 Total 36 52 36 39 38 Source: MOWRAM Project and budget implementation regular monitoring of project implementation. In the case of investment expenditure (Chapter There has been a noticeable improvement 21), these are provided in the MEF’s Prakas # 207 in the execution of investment expenditure on procedure of public investment and the newly (Chapter 21). MEF Prakas # 207 dated 26 February adopted SOP on investment project implementation. 2019 provides clear guidance on how the capital For service expenses (Chapter 61) on irrigation, these budget should be managed in tandem with the provisions are found in the recently adopted Prakas overall project implementation cycles. The fact that on irrigation O&M. Requirements on monitoring and the capital budget has been prepared on time to reporting are also found in the regular annual budget be reviewed and approved at the same time as the implementation circular issued by the MEF. This current budget has allowed for sufficient time for study also reviewed MOWRAM activity reports from project implementation during the fiscal year. The previous years and makes note of the rich information timeliness element has been particularly important being reported on the activities performed and for irrigation projects because it means they can be outputs produced. implemented with minimum disruption from the rainy season, which usually starts in May or June. The main gap in the current system is the lack of information on the impacts resulting from One main area for improvement is public the investment and spending. There is necessity procurement of the investment expenditure for such impact evaluation information to support (Chapter 21) projects. MOWRAM has been budget management effectiveness since a lot of implementing key reform tasks relating to public resources have already been spent on rehabilitation procurement. However, a large majority of and O&M of irrigation schemes in the past decade. investment expenditure (Chapter 21) projects (and While much information on the outputs has been other procurement works as well) have not been produced (Table 5.12), there is virtually no information done through bids by international competition, bids on the quality of impacts that have been achieved. by domestic competition, price consulting, or price surveys. Instead, most of them have been done via Figure 5.14 and Figure 5.15 show that agriculture ‘separate procurement’ which basically means giving land without irrigation facilities remained high them to a general department or a department within at 54 percent of total agricultural land, despite MOWRAM to implement as contractors (Table 5.11). declining from 62 percent in 2015. Irrigated areas of agricultural land increased from 38 percent of total Project monitoring, ex-post review and agricultural land in 2015 to 46 percent in 2021, driven asset management by irrigation during wet seasons. Irrigated areas of agricultural land during the dry season remained low Guidelines have been put in place to ensure and tended to decline over the last five years, while Cambodia Public Finance Review - From Spending More to Spending Better 145 Table 5.12. Outputs reported from investment expenditure (Chapter 21) 2019 2020 2021 No. of projects 38 34 28 No. of provinces 16 16 16 Budget spent (KHR, million) 166,842 153,829 122,844 Increase in irrigated areas for rainy season (ha) 19,432 17,902 15,601 Increase in irrigated areas for dry season (ha) 9,057 10,060 6,987 Other crops (ha) 412 535 1,261 Ensuring sustainability for rainy season (ha) 27,355 10,372 9,090 Ensuring sustainability for dry season (ha) 4,185 2,819 1,090 Source: MOWRAM annual report (2019-2021) Note: ha = Hectare Irrigated areas of agricultural land for both seasons The information on irrigation schemes in slightly increased during the same period. Cambodia has been kept in a separate database, not yet a part of the asset register. As Information is also lacking on irrigation mentioned earlier, Cambodia has over 2,500 irrigation schemes as assets to be managed. Asset schemes, categorized as small, medium, and large. management is one key area in the standard PIM Information about these schemes has been collected process and is mentioned in the PIMSRS and and updated in a system called Cambodia Irrigation other legal documents in Cambodia. A key part of Scheme Information System (CISIS). CISIS has not asset management is to have an asset register. For been made publicly available, and it is not clear what MOWRAM, and other ministries in the government, detailed information is available and whether it is asset registration has been done with the use of the same kind of information that an asset register the SARMIS (State Asset Registration Management should have. Information System). Despite progress, however, the current rules do not require irrigation schemes to be 5.5. Summary of key findings included in the asset register. A large share of the MOWRAM expenditure Figure 5.14. Agricultural land, by irrigation Figure 5.15. Agricultural land, by irrigation facilities (000 hectare) (2014-2021) facilities (% of total) (2014-2021) 4500 2014 2015 2016 2017 2019-20 2021 100% 4000 90% 3500 80% 50% 52% 54% 70% 62% 58% 000 hectare 3000 63% 2500 60% 50% 2000 7% 14% 40% 10% 11% 12% 1500 30% 9% 8% 8% 7% 1000 8% 11% 11% 20% 32% 27% 27% 500 10% 22% 19% 20% 0 0% Irrigation Irrigation Irrigation No Total 2014 2015 2016 2017 2019-20 2021 during wet during dry during both irrigation or Irrigation during wet season Irrigation during dry season season season seasons water pump Irrigation during both seasons No irrigation or water pump Sources: CSES (2014-2021) and World Bank staff estimates 146 Cambodia Public Finance Review - From Spending More to Spending Better has been allocated to the irrigation sector, in particular the limited practice of open bidding in leaving a small percentage for water resource public procurement. management. External funding, on the other hand, focuses on both irrigation and water management. Irrigation expenditure allocations to the provincial level is relatively equitable. Using There is a high in-year budget allocation irrigation-related expenditure under investment for investment expenditure (Chapter 21) for expenditure and service expenses (Chapter 21 and MOWRAM. The in-year additional budget allocation Chapter 61) (which is by far the largest), available can range from 0 percent to 100 percent of the data suggests that the poorer the province, the investment expenditure (Chapter 21) allocation to higher the budget that had been allocated to it. MOWRAM in the Budget Law. Such a large but unpredictable amount of extra budget allocated 5.6. Recommendations during in-year implementation can undermine allocative efficiency because it undermines the Allocative efficiency can be enhanced as comprehensiveness of the due process of budget budget allocation is further aligned with formulation. policy objectives. Given that the government is placing more focus on rice and other high value By function, the overwhelming majority of crops, MOWRAM, through its existing program the MOWRAM budget has been spent on and subprogram structure, should adjust its budget ‘agriculture.’ This suggests the importance of allocation to reflect these changing policy priorities. coordination between MOWRAM and MAFF at both national and provincial levels. Such coordination has It is expected that irrigation will still be a priority been limited, especially at the provincial level. of the government. However, in that priority, more attention should be given to irrigation scheme O&M One key systematic challenge is the consistently as opposed to rehabilitation and construction. limited budget allocated to PDWRAM. The To identify the optimal balance between the two problem of severe staff shortages at the PDWRAM investments, however, more data and analysis are level has been brought to the forefront by the needed. data analysis. The limited de-concentration, if not addressed adequately, can systematically constrain Allocative efficiency can be ensured if the the ability of MOWRAM to respond to local needs, different financing sources complement one especially on irrigation O&M, support to FWUC, and another. For this to happen, the first step is to have emergency responses. comprehensive data on those different financing sources and how the budget has been allocated. It The operational efficiency of investment is therefore recommended that MOWRAM, MEF and expenditure (Chapter 21) also depends on the other relevant partners support a comprehensive quality of the PIM. In the case of MOWRAM, limited sectoral financing mapping study, covering not only planning practices, data, and regulatory constraints state budget, but also development partners, SNAs have undermined the quality of the project pre- and the private sector. appraisal phase. Also limited are the quality of the medium-term estimate of capital budget needs for The limited de-concentration within MOWRAM the Ministry and the integration between capital and should be systematically addressed. The recurrent funding, especially O&M. breakdown of roles and responsibilities between the center and the local governments should The execution of the capital budget, while follow international practices where the central improving, still faces specific technical challenges level focuses more on setting policies, technical Cambodia Public Finance Review - From Spending More to Spending Better 147 standards, technical support, M&E, and data works, on results is at output level. Therefore, it is and the lower levels focus on implementation and are recommended that the Government consider accountable to the higher level for results. There are conducting an impact assessment on the spending various technical guidelines, standards, and M&E that of investment expenditure (Chapter 21) to generate need to be developed. The central MOWRAM should new lessons learned. focus on those and less on actual implementation. In addition, the question of personnel deployment and Improving PIM could help to improve operational decentralization needs to be addressed as a matter efficiency. Going forward for irrigation and water of urgency. resources, key areas that require improvement include project pre-appraisal and selection, and The function transfer to the DMK administration open bidding procurement. Experiences from should be accelerated in accordance with the other countries suggested that strategic sourcing Sub-decrees # 182, 183, and 184 (2019). Certain methodology for procurement has yielded substantial key functions to be transferred, such as small-scale saving and improved efficiency on goods and service irrigation scheme O&M and support to FWUCs, are spending. Another important aspect of investment appropriate for SNAs, given that some SNAs have budget management is to improve the forecast of already performed them as their de facto functions. ‘flood damage repair’ needs in order to reduce the ad- hoc /in-year budget from the ‘unallocated budget.’ Significant public resources (domestic and Short-term and medium-term recommendations are external) have been invested in irrigation presented below in Table 5.13. projects, but much of the available data 148 Cambodia Public Finance Review - From Spending More to Spending Better Table 5.13. Recommendations for irrigation sector Options Challenges Medium-term Short-term recommendations (1-2 years) recommendations (3-5 years) Capital budget Improving coordination between MOWRAM Conduct an impact planning and and MAFF on strategic water and agriculture assessment on the management development to ensure effective expenditure and investment expenditure coordination (Chapter 21) Improve the forecasting, planning and management of emergency response interventions such as flood and drought events which could reduce the ad-hoc / in-year budget from the ‘unallocated budget’ Integrate irrigation scheme inventory in state asset registry Weak Key areas that require improvement include the implementation of project appraisal, and implementation of procedure Public Investment for open bidding procurement. Management Optimizing uses of built infrastructure to provide active service delivery to respond to the needs and for economic outcomes Limited Clarifying the roles and responsibilities of MOWRAM Develop a de-concentration and PDWRAM. comprehensive roadmap within MOWRAM • Central level MOWRAM focuses more on setting for sectoral financing and decentralization policies, technical standards, technical support, covering not only M&E, and data works state budget, but also • The sub-national levels focus on implementation, development partners, service operations, operation and maintenance SNAs and the private and are accountable for results sector • Adequate personnel and technical staff deployment at PDWRAM Accelerating the function transfer to the DMK administration • Transfer the roles on small-scale irrigation scheme O&M and provide technical and financial supports to FWUCs to strengthen their operational functions to ensure sustainable operation management of the schemes and better service delivery Cambodia Public Finance Review - From Spending More to Spending Better 149 6 KEY LESSONS ON SUBNATIONAL PFM BASED ON CASE STUDY OF PREAH SIHANOUK PROVINCE 6.1. Introduction 6.1.1. Development financing and budget allocation to Preah Sihanouk Subnational administration plays an important province role in public service delivery. Preah Sihanouk province is chosen as a case study province to The key public sector agencies operating at provide insight into how public finance (both revenue the sub-national level include the SNAs and and expenditure) is managed at the subnational line ministries’ departments/agencies (hereafter level. With a population of 223,286 in 2021, Preah referred to as line departments). The number of SNAs Sihanouk province is a coastal province with a in the province includes: a provincial administration, high level of economic potential in the service and three districts, two municipalities, 23 communes, tourism sectors. The international port allows Preah and six sangkats. In addition, there are 22 line Sihanouk province to welcome international trade departments (LD) and three public administration connections and tourism. Given its economic establishments (PAE) present, namely, electricity and potential, the government plans to transform Preah water supply utilities, and a public sector training Sihanouk province into a multi-purpose economic center. These public sector agencies have their own zone. Economically, Preah Sihanouk province is budgets and participate in the joint planning exercise one of the most vibrant provinces in the country. to develop the 3-year rolling investment program (3y- The province has experienced impressive economic IP) and the Budget Strategic Plan (BSP). growth, leading to a continuous increase in per capita income and a steady decrease in the poverty rate, Preah Sihanouk province is among the top from 12.7 percent in 2015 to 3.3 percent in 2019. provinces in terms of total budget. In 2019, it The main economic engines are industry, service stood at 4th place in term of SNA total budget, coming sector, and agriculture, which respectively accounts after Phnom Penh, Kandal, and Siem Reap (Figure 6.1). for 51 percent, 30 percent, and 19 percent of the Its ranking comes third (after Phnom Penh and Kep) if province GDP. SNA budget per capita is considered (Figure 6.2). 150 Cambodia Public Finance Review - From Spending More to Spending Better Figure 6.1. SNA total budget in 2019 (KHR, Figure 6.2. SNA budget per capita in 2019 million) (KHR) Tbong Khmum Tbong Khmum Takeo Takeo Svay Rieng Svay Rieng Stung Treng Stung Treng Preah Sihanouk Preah Sihanouk Siem Reap Siem Reap Ratanak Kiri Ratanak Kiri Pursat Pursat Prey Veng Prey Veng Preah Vihear Preah Vihear Phnom Penh Phnom Penh Oddar Meanchey Oddar Meanchey Mondul Kiri Mondul Kiri Kratie Kratie Koh Kong Koh Kong Kep Kep Kandal Kandal Kampot Kampot Kampong Thom Kampong Thom Kampong Speu Kampong Speu Kampong Chhnang Kampong Chhnang Kampong Cham Kampong Cham Battambang Battambang Banteay Meanchey Banteay Meanchey 0 500,000 1,000,000 0 200,000 400,000 Source: MEF, GDSNAF (2022) The provincial administration has a substantial Figure 6.3. Proportion of executed budget of amount of budget, accounting for 45 percent of total SNAs and LDs in Preah Sihanouk province government budget allocated to both SNAs and LDs 200,000 in Preah Sihanouk province in 2020 (Figure 6.3). It is 180,000 important to note that the provincial administration 160,000 5% budget does not include that of the line departments 140,000 4% 4% 3% 5% 120,000 and agencies. 4% 42% 45% 100,000 38% 80,000 Information on the public investment program 60,000 40,000 of the LMs in Preah Sihanouk province is not 56% 50% 47% 20,000 available—an important weakness in the 0 planning process. Two types of new investment 2018 2019 2020 LDs (22 units) Provincial administration (1 unit) projects are initiated every year: 1) those funded by DMK administration (5 units) CS administration (29 units) development partners, most of which are included in the Public Investment Program (PIP), and 2) Source: MEF, GDNT (2022) those funded by state capital budget which are monitored by the Department of Investment of the investment projects is not publicly available. At the General Department of Budget under the MEF. provincial level, however, the information on both At the national level, information about the donor- types of investment projects is neither available nor funded projects is available on the website of the recorded by the provincial administration despite the MOP, but the information about the state-funded significant size of these projects. Cambodia Public Finance Review - From Spending More to Spending Better 151 Figure 6.4. Preah Sihanouk province revenue (including transfer), planned and actual (2018-2020) 200,000 150,000 KHR, million 100,000 83% 81% 89% 77% 50,000 87% 89% 0 Plan Actual Plan Actual Plan Actual 2018 2019 2020 Tax revenue Non-tax revenue Transfer Source: MEF, GDNT (2022) 6.2. The provincial plan, and in 2019 it was 401 percent higher before going down to 111 percent in 2020, mostly because administration of the COVID-19 pandemic. 6.2.1. Budget allocation and There is a high level of repair and maintenance expenditure outlook expenditures. More than 70 percent of the Preah Sihanouk provincial administration budget has been Provincial administration revenues come mostly spent on the repair and maintenance of infrastructure, from taxes (Figure 6.4). Tax revenues account for mostly roads in municipal areas (Figure 6.5). Road more than 80 percent of the total revenue for all three maintenance was considered money worth spending, years. In 2018 and 2019, the administration planned given the poor road condition after the floods of only a small amount of non-tax revenue, however, 2019 and the need for road rehabilitation to support this was substantially increased in 2020 by about tourism and facilitate travel in the province. However, 20 times compared to 2019. Furthermore, in 2018, the high percentage of spending by the provincial the actual revenue was 384 percent higher than the administration on road repairs in municipal areas Figure 6.5. Preah Sihanouk province expenditure, planned and actual (2018-2020) 120,000 100,000 KHR, million 80,000 60,000 75% 40,000 81% 70% 81% 20,000 70% 68% 0 Planned Actual Planned Actual Planned Actual 2018 2019 2020 Chapter 60: Purchases Chapter 6105: Repair and maintenance Chapter 61: Services Chapter 64: Payroll expenses Chapter 62: Social bene ts expenses Chapter 65012: Transfer to DMK Chapter 65013: Transfer to CS Chapter 69: Reserve Chapter 21: Investment expenditure Source: MEF, GDNT (2022) 152 Cambodia Public Finance Review - From Spending More to Spending Better Figure 6.6. Budget allocation by programs for 2020 (KHR, million) 90,000 80,000 KHR, million 70,000 60,000 50,000 40,000 85,470 30,000 20,000 10,000 3,968 0 230 2,613 P1: Economic P2: Social P3: Public order P4: General administration development development and safety and services Chapter 60: Purchases Chapter 64: Payroll expenses Chapter 21: Investment expenditure Chapter 61: Services Chapter 65: Subsidy expenses Chapter 62: Social bene ts expenses Chapter 69: Reserve Source: MEF, GDNT (2022) raises questions related to the potential duplication all of which is for road repair and maintenance in the of roles between provincial and DMK levels. There is municipal areas. Additionally, personnel expenditures also a high variation between the planned and actual are not captured in every program. For instance, all expenditures of around 170 percent in 2018 and personnel expenditure (Chapter 64) were captured 2019, but this dropped to only 75 percent due to the under General Administration Program (Program 4) impact of COVID-19 related restrictions. (Figure 6.6). The percentage of actual personnel expenditure 6.2.2. Development planning, spatial (Chapter 64) in the budget increased noticeably planning, budget strategic planning, from 4 percent in 2019 to 12 percent in 2020 and allocative efficiency because of incentive sharing payment (under the account 6403), which provincial administration Despite the current legal and regulatory made to its own staff and those seconded from line framework and the progress over the years, departments to work in the One Window Service the current planning process at Sub-national Unit (OWSU). 119 Despite the increase, the overall administration (SNA) is not adequately efficient percentage of personnel expenditure (Chapter 64) is for budget allocation purposes. SNAs, including only a small proportion of the total budget (Figure provincial administrations, are required to prepare 6.6). 120 The shares of these other budget chapters various local planning documents, following specific (besides the repair and maintenance) in total were rules and calendar. However, the plans of different similar in 2018 and 2019. entities at SNAs are not yet fully integrated. SNA plans and line department plans are also not yet The program budget data, while useful in aligned, and spatial planning has not been given showing where the money was allocated, is not sufficient attention. Efforts are being made to improve yet helpful to assess allocative efficiency. 121 Out the current planning process, but more attention is of the KHR 117,867 million, about 73 percent was needed to improve data availability, management, spent on economic development (Program 1), virtually sharing, and use for SNA planning purposes. 119 OWSO is operated by two groups of staff: front office mainly SNA officials, who directly interact with citizens and back office staff, who processes different requests. These second group includes secondment staff from line departments that are not part of SNA. The OWSO generate revenues from user fee. Part of this user fees is used as additional incentive for staff in OWSO. 120 Actual expenditure for personnel cost at provincial administration would appear to be even smaller if incentive sharing with staff of line department at OWSU was not captured under provincial budget. 121 The program budgeting exercise began in Preah Sihanouk province fairly recently and only one year of data is available (2020). Cambodia Public Finance Review - From Spending More to Spending Better 153 SNA development and spatial planning planning in accordance with the relevant policy and regulations.122 As of 2021, the Preah Sihanouk The provincial three-year implementation plan provincial administration, the Sihanoukville municipal (3y-IP) includes the development projects of administration, and seven communes and sangkats line departments, but very little information are known to have developed spatial planning, while is available on national-level projects. The the others are in the process of preparing one.123 review of the 2020 and 2021 3y-IP provides a list The challenge is that SNAs need to have a better of various projects to be funded by deconcentrated understanding of how to link their socio-economic line departments. However, not much information is planning (i.e., the 5y-DP and 3y-IP) more explicitly available on major projects managed by national line and systematically with their land-use planning. ministries (including those funded by development partners). This implies that a large part of development Limited effectiveness in data storage, sharing, interventions in the province are not included in the and use has increasingly become a key 3y-IP. Given that most line ministry projects tend challenge in the current SNA planning exercise. to be large in scope, the lack of information about In preparing the 3y-IP, the current guidelines124 them is a major weakness in subnational planning. project the production and use of a wide range of In contrast, line department projects focus more on data. There is a Planning Working Group established minor scale repair, maintenance, and operations. to coordinate the planning process. The group’s main task is to collect and compile data from various The information in the 3y-IP is also limited SNAs and provincial line departments and coordinate when it comes to investment and financing by the planning exercises among these agencies. NGOs and the private sector. The Preah Sihanouk In practice,125 the workload falls on the Provincial provincial administration has not been able to keep Department of Planning (PDP) and the Planning and track of all the critical data on these investment and Investment Unit of the provincial administration. The financing sources and this limitation is reflected in the PDP, as the permanent member of the Planning 3y-IP. There is information about only the projects with Working Group, possesses and stores much of available financing from the provincial administration the data, but it also needs to get data from line and line departments, and none from NGOs or the departments and SNAs. The practice of data sharing private sector. This lack of data reflects the current and data storage has so far not been systematic. situation where central ministries still hold much Data are kept in MS Word or Excel format on various authority over investment approval and licensing, as computers, vulnerable to data loss, unauthorized well as the operation of NGOs. When it comes to changes, or clerical errors. interacting with these non-state actors, the provincial administration is mainly limited to a coordinating role, Budget Strategic Planning (BSP) which is yet to be clarified. The Preah Sihanouk provincial administration Spatial planning has been more advanced has followed the BSP preparation process, but in Preah Sihanouk province compared to many challenges remain at the implementation other provinces, although some challenges level. The administration started preparing the remain. SNAs in Preah Sihanouk province have BSP in 2020 by following all the guidelines and made good progress with sub-national spatial program structure discussed above. However, in 122 RGC (2011) 123 Interview with the Preah Sihanouk Provincial Department of Land Management, Urban Planning, and Construction (March 2022) 124 Please see MOI and MOP (2020), MOI (2017) 125 For instance, the Planning Working Group for the preparation of the 3y-IP for 2021-2023 has 53 members. 154 Cambodia Public Finance Review - From Spending More to Spending Better actual implementation, it faced several constraints. Tax revenue forecast and management The first challenge is the level of awareness of the new guidelines among technical staff. Second is The provincial tax branch in Preah Sihanouk the challenge of collecting data from DMK and CS province collects and manages the tax revenue administrations, line departments, NGOs, and other in its role as a subordinate of the national stakeholders to fill in the template. Consequently, as General Department of Taxation (GDT) and of 2021, the provincial BSP has only data relating to in accordance with the relevant rules and the provincial administration. regulations. The tax branch office is not under the supervision of the provincial administration, although Besides the SNA capacity and implementation two agencies are expected to work together to challenges, a number of gaps in the current ensure effective tax collection. According to the process are identified. To integrate all the SNA MEF annual circular, the provincial administration 3y-IP into the CP BSP, three structural consistencies needs to also appoint a statistical official in charge of need to be addressed: collecting and updating data relating to tax collection and collaborating with tax officials when required.127 1) The timeframe by which the two planning actions are carried out. The timeline for preparing Preah Sihanouk provincial administration the 3y-IP was already adjusted per the ministerial prepares its revenue forecast following the Prakas PK #0149 (2020) so that necessary legal provisions in the 2011 SNA finance law and information from the 3y-IP of all SNAs is ready other regulations, including the MEF’s annual by the end of March to be included in the annual circular. For the planned tax revenue (N+1), the plan BSP preparation process. needs to be based on the following considerations: 1) actual performance in the previous year (N-1), 2) 2) The inconsistent sectoral classification under actual and expected performance in the current year the 3y-IP and the BSP, especially for P3 and P4.126 (N), 3) estimated economic performance in the next year, 4) any expected changes resulting from the 3) The inconsistency in the budgetary scope of changes in tax rate and tax bases, and 5) expected the two planning tools. The 3y-IP covers mostly improvement in tax administration. The tax revenue development or investment, whereas the BSP planning needs to also refer to the broader macro- covers all types of expenditure categories, both economic forecast and the Revenue Mobilization recurrent and capital. Strategy (RMS). 6.2.3. Revenue forecasting and management In practice, tax revenue forecasting has been done with unclear methodology and limited Inaccurate revenue forecasts from both tax and inputs from the provincial administration. non-tax revenue (NTR) sources is impacting both Two limitations were identified. First, there was allocative efficiency and operational efficiency. no clear data or strategy to assist the forecasting Inaccurate forecasting limits the ability to plan and exercise, except for the last year’s actual collection utilize available resources efficiently. performance (consideration #1 above).128 Second, 126 In the BSP, P3 is “Public Safety and Order,” but in the 3y-IP, it is “Land Use, Environment, Disaster Management, and Climate Change’; in the BSP, P4 is ‘General administration’, but in 3y-IP, it is “Public order and general administration.” 127 The interviews also suggest that the tax branch needs to submit a regular (either weekly or every two weeks) summary table report on the tax collection to the provincial administration. The interviewed tax officials also claimed that the provincial governor could also ask for clarification or justifications for, say, lower than expected tax collection, and the tax branch would provide the answers. In reality, it is not clear how vigorous such accountability arrangements have been, given the still centralized arrangement of tax administration in Cambodia. 128 Due to data limitations, the other considerations (from #2 to #5) were largely subjective. Cambodia Public Finance Review - From Spending More to Spending Better 155 the roles and accountability line between the on property rental, which accounted for 6 percent provincial administration and the GDT branch have of total revenues in 2018, 14 percent in 2019, and not been clearly provided or understood. As a result, 9 percent in 2020. It should be noted also that the approved tax forecast for the province was there is a huge variation between the planned and decided mainly by the suggestion of the GDT and actual revenues for these two revenue handles. The the GDP at the national level. variation was over 400 percent in 2018 and 2019 for the stamp tax on property, and above 550 percent A large share of tax revenue of Preah Sihanouk for tax on property rental in 2019. Other tax items provincial administration comes from real such as stamp tax on legal documents and tax on estate sales and rentals. The largest revenue item unused land already show higher variation between is the stamp tax on property transfer and ownership, plan and actual, but their shares in the overall tax are which accounted for 74 percent of total revenues small. Such a high variation between planned and in 2018, 71 percent in 2019, and 66 percent in actual revenue suggested limited revenue forecasting 2020 (Figure 6.7). The next largest item is the tax capability. Figure 6.7. Tax revenue by composition – plan and actual (2018-2020) 160,000 KHR, million 110,000 71% 60,000 74% 66% 74% 10,000 56% 62% 14% 0% 6% 9% 10% 9% Plan Actual Plan Actual Plan Actual 0 2018 2019 2020 Taxes on property rental Taxes on accommodation Patent taxes Taxes on public transportation Stamp on property transfer and ownership Stamp on vehical transfer Taxes on property (individual) Taxes on property (Company) Taxes on unused land Other taxes Source: MOI (2021) Figure 6.8. NTR of Preah Sihanouk province by composition, plan vs actual (KHR, million) (2018-2020) 15,000 KHR, million 10,000 94% 5,000 93% 0 50% 52% 47% 51% Plan Actual Plan Actual Plan Actual 2018 2019 2020 Chapter 74: Revenue from ne Chapter 73: Revenue from services and asset rental/sales Chapter 72: Revenue from state asset Source: MOI (2021) 156 Cambodia Public Finance Review - From Spending More to Spending Better Figure 6.9. Revenue of Preah Sihanouk province OWSU as % of national total (KHR, million) (2018-2020) 16,000 14,712 20% 14,000 17.6% 12,000 15% KHR, million 10,000 8,000 10% 6,000 7.3% 4,000 5.2% 5% 2,622 2,000 429 0 0% 2018 (5 months) 2019 2020 Revenue of OWSU Revenue of OWSU (as % of total) Sources: MOI (2021) data on OWSU revenue Figure 6.10. Other NTR items besides OWSU fee (KHR, million) plan and actual (2018-220) 900 56 KHR, million 700 400 500 - - - 300 - 543 100 100 100 141 112 69 0 Plan Actual Plan Actual Plan Actual 2018 2019 2020 72004: Mining concession 72006: Transportation parking/station 72007: Water vehicle parking/station 73055: Revenue from market pheasi 74012: Revenue from ne Source: MEF, GDNT (2022) Non-tax revenue planning and in Figure 6.9, the share of the revenue from Preah management Sihanouk province OWSU compared to the total national OWSU income increased from just about The non-tax revenue (NTR) 129 of the Preah 5.23 percent in 2018 to almost 17.63 percent in Sihanouk provincial administration significantly 2020. increased in 2020 due to the sharp increase in fees from administrative services (Chapter 73), While revenue from OWSU is significant, those especially those provided via OWSU (Figure 6.8 from state assets and fines should also be given and Figure 6.9). The OWSU was introduced in late more attention. The main revenue items include 1) 2018, but it became fully operational only in 2020. transportation parking/station, 2) market pheasi130, That explained the sharp increase in revenue (as 3) mining concession royalty, and 4) revenue from illustrated in the diagram) starting from that year. fines. The first two items are planned and collected But it was also due to better administration and based on contracts and thus have been stable, with more demand for certain services, especially land slight variation between plan and actual. registration and construction permits. As illustrated 129 Preah Sihanouk provincial administration plans and collects NTR from public service fees, state assets, and fines. 130 The revenue refers to the fee that the provincial administration receives from a private company for its business rights to manage a market (mostly on matters relating to hygiene and security) and receive regular fees from market venders. Cambodia Public Finance Review - From Spending More to Spending Better 157 The royalty from mining concessions shows a building and technical assistance support (in the form significant variation between plan and actual of a help desk) is needed. In addition, management (Figure 6.10). It was under-executed by about 30 of NTR would benefit from (1) clarification on how the percent in 2018, and over-executed by about 40 NRMIS can be linked to the management information percent in 2019, but the execution rate went up to system (MIS) that the MOI has set up for OWSU to more than 500 percent in 2020. The revenue from record the number of services and fees at OWSU, fines was not planned until 2020, and the execution (2) reviewing the current regulation and estimate of was well below 20 percent due to COVID-19 and collected revenue that can be kept at OWSU, and challenges in enforcing the fines. (3) clarification/reviews of recording of NTR made via e-payment (For instance, a part of service fees There are three related areas for improving NTR for vehicle registration is made through the banking management. The first is the need to improve the system to the MPWT accounts, although the OWSU system, process, and capacity of those in charge of is the one that provides the services). managing the revenue. The second is to continue transferring more NTR functions from line ministries More NTR functions should be considered to provincial administration and do so in a more for transfer from line ministries to the SNAs. coherent manner. As relevant capacity at the SNA Significant NTR sources have already been improves, the third aspect, the government could transferred from line ministries and line departments consider transferring some of the NTR to the DMK to the provincial administration (a large part is to the level. OWSU). However, the overall NTR revenue sources at the provincial level are still limited compared to NTR On the first point, different areas need to be at the national level. While a careful review is needed improved regarding the NTR management before more NTR functions can be transferred, there at the provincial administration. The first is the are some obvious candidates to consider for transfer regulatory framework and establishing a system to in the short term. One example is the revenue from update procedures governing the management of the penalty on fishery offenses (sub-account 74012) NTR handles. The administration needs to follow which is currently collected by the provincial line through with the provisions of various laws and department; the collected cash is then physically regulations on NTR management, including Sub- transported to the Ministry in Phnom Penh. decree #25 (2020) and various Prakas concerning the operation of the OWSU and incentive sharing. The The transfer process should also be systematic NTR management at the SNA level relates closely to to minimize disruptions at the implementation the functional delegation process from line ministries, level. One example identified by the study is how officials at the provincial administration (especially NTR is collected and managed from a local soil/dirt those in the OWSU) need to continuously update extraction business. For this NTR, two functions the procedures governing the management of NTR are involved: 1) issuance of a business license to handles as new Prakas from line ministries come out, businesses (NTR from public service fees), and making changes to the fees and/or procedures on 2) collecting royalty from businesses (NTR from how a service can be applied and processed. state assets). Currently, the first function is already transferred from the Ministry of Mine and Energy The NTR collection, recording process and (MME) to the DMK level, but the second function is implementation have been improved, but kept under the line department. To ensure proper more needs to be done. One noticeable progress collection of the royalty, the line department relies in this regard is using the NRMIS at the provincial on the licensing information from the DMK level. In administration level (not just at the OWSU). This new practice, the SNAs do not update line departments system is broadly appreciated but more capacity about new licenses, thus leading to lower royalty 158 Cambodia Public Finance Review - From Spending More to Spending Better revenue collection. While it is reasonable to suggest and NTR items from their own NTR list of services ‘better coordination and reporting between the two and fees, which implies that many relevant inter- agencies’ as a solution, it might be more effective to ministerial Prakas on NTR need to be revised. Third, transfer the two related functions to the SNAs and the Prakas should cover not only NTR from public make them responsible for the collection of revenue. services, but from state assets and fines/penalty as well. The vertical imbalance between the provincial and DMK level also needs to be addressed. While In the NP2, both short- and long-term strategies the increase in NTR for the provincial administration are set out for improving the NTR of SNA. For is a positive development, there are opportunities to the short term, the priority is to build on the success further transfer additional revenues to the DMK level, of the OWSU, including the legal framework, service where more service delivery functions are expected delivery systems (including e-services), management to be transferred and where own-resource revenue system (including e-administration in the form of MIS), collection has been minimal. Before 2020, the NTR and feedback mechanisms (including e-feedback). of the provincial level was comparable to the DMK For the long-term, the reform of SNA NTR needs level, but since 2020, with more revenues assigned to be coordinated with the development and to the OWSU, the gap has become much wider implementation of the 10-year Blueprint for Non-Tax (Figure 6.11). The planned NTR of the province was Revenue (2021-2030) that MEF has been working about ten times that of the DMK level, and the actual on. To ensure smooth coordination, decentralization is about 14 times. aspects should be integrated throughout the critical aspects of NTR management, including the legal The MEF has issued a Prakas on the transferring framework, budget management and reporting, of NTR from the national to the sub-national e-governance, and human resource development. level. While the details are to be finalized, a few key As such, the National Committee for Subnational points are noted. First, the Prakas seeks to make Democratic Development Secretariat (NCDD-S) an explicit linkage between functional transfer and needs to engage in the development of the Blueprint NTR transfer by making clear that when the function as much as possible. is transferred, any associated revenue would also be transferred. Second, the Prakas should provide State asset management instructions for the transferring line ministries. For instance, it should require that all the transferring State asset management for SNAs, including at line ministries drop the already transferred functions the provincial level, is subject to a number of Figure 6.11. Provincial NTR versus DMK NTR (2018-2020) 1600% 12,000 1,378% 1400% 10,000 1200% KHR, million 937% 8,000 1000% 800% 6,000 600% 4,000 400% 2,000 203% 200% 129% 140% 78% 0 0% Plan Actual Plan Actual Plan Actual 2018 2019 2020 Provincial NTR DMK NTR Provincial NTR (% of DMK NTR) Source: MOI (2021) data Cambodia Public Finance Review - From Spending More to Spending Better 159 Table 6.1. State asset information compiled by PDEF (2021) Agencies Unit Asset values 2021 Line departments/agencies 32 KHR 7,870 million Out of the 244 asset units reported, only 116 (48%) have ownership titles. Provincial administration 1 KHR 2,891 million Out of the 29 asset units reported, only 16 (55%) have ownership titles DMK administrations 5 KHR 105 million Out of the 13 asset units reported, only 5 (38%) have ownership titles CS administrations 29 KHR 10,874 million Out of the 323 asset units reported, only 152 (47%) have ownership titles Source: Preah Sihanouk province, PDEF (2022) key laws and regulations. These include the Public The Circular #007 (2014) requires SNAs to prepare Finance Law (2008), the Organic Law (2008), the SNA ownership certificates for all state immovable Finance Law (2011), the Sub-decree #66 (2017), and, properties in their inventories and list them in their more recently, the State Asset Law (2020). MEF’s inventory book. For the concession agreements on Circular #007 (2014) is another important regulatory every type of state property, SNAs must update them reference on state asset management. According to annually before the end of the first quarter of each these laws and regulations, there are three layers of fiscal year and send them to the PDEF, which is the authority in state asset management: (i) a controlling deconcentrated arm acting on behalf of the MEF as authority (CA); (ii) managing authorities (MAs); and (iii) the CA. This updating process involves two types of utilizing authorities (UAs). The MEF is a CA, which is documents, namely a five-year inventory book and an responsible for regulating, monitoring, and inspecting annually updated inventory list. The Preah Sihanouk all levels of authorities. MAs refer to line ministries, provincial administration has maintained a register of institutions, SNAs, other qualified public entities, etc. its assets, which are recognized at their acquisition MAs can have many UAs; UAs are reporting to MAs, cost or fair value. However, officials recognized that and MAs are reporting to CA. the register only has partial information on the assets’ usage, value, location, and age. Assets of an SNA include the state assets that are transferred to the administrations and the The PDEF, acting as the CA on behalf of the assets that it acquires on its own. The assets that MEF, compiles the annual asset registry from all are transferred to SNAs are the assets that the SNAs MA, SNAs, line departments, and agencies. In can manage, utilize and generate revenues from. the 2020 annual inventory list made available to the SNAs maintain and protect these assets that the World Bank team by the Preah Sihanouk province state transferred to them, and they have no right to PDEF, all necessary information was included. The sell or transfer ownership and transfer the utilization Table 6.1 summarizes the key data from the inventory without the agreement from MOI and the approval list. Asset management at SNA is a relatively new from MEF.131 There are four types of state assets: practice, and therefore requires more long-term i) lands and buildings, ii) technical and industrial support. equipment, iii) vehicles and machinery, and iv) office equipment and furniture. The completeness and reliability of the current asset registration process is limited. As Preah Sihanouk provincial administration is the presented earlier, relevant rules and regulations are managing authority (MA) over its state assets. followed when asset inventory is prepared. However, 131 Article 254-255 of the Organic Law (2008) 160 Cambodia Public Finance Review - From Spending More to Spending Better Table 6.2. Percentage and timing of budget execution (2018-2020) 2018 2019 2020 Procurement methods used executed executed executed executed executed executed (% of total) (% of Q4) (% of total) (% of Q4) (% of total) (% of Q4) Total 373% 90% 171% 40% 75% 78% Chapter 60: Purchases 104% 54% 124% 45% 89% 41% Chapter 61: Services 191% 80% 196% 41% 69% 81% Chapter 6105: Repair and maintenance 195% 81% 205% 41% 70% 82% Chapter 64: Payroll expenses 103% 27% 105% 26% 102% 75% Chapter 62: Social benefits expenses 316% 88% 151% 76% 253% 82% Chapter 65012: Transfer to DMK 100% 79% 128% 9% 100% 60% Chapter 65013: Transfer to CS 100% 100% 100% 11% 100% 100% Chapter 21: Capital investment 434% 81% 178% 100% 136% 74% Source: MEF, FMIS (2022) there are questions about the completeness and 6.2.4. Overall expenditure execution reliability of information (e.g., if the current asset classification fully covers all types of assets that Despite the introduction of the program-based should be covered, if the valuation is appropriate). budget, the expenditure execution rules still There is a need for more information on the use of focus on the budget line items. The budget the assets and the revenue generated from them. implementation circular emphasizes the need to This is even more relevant given that SNAs, including spend, for instance, personnel, petty advance, and Preah Sihanouk province, are being asked to identify procurement budgets according to specific rules new potential NTR. Preah Sihanouk provincial and regulations. It also emphasizes the rules around administration has not had a centralized database budget virement and the spending of budget reserve system for collating key details on state property (Chapter 69). management information. However, the MEF’s plan to roll out the State Property Inventory Management During 2018 - 2020, a few positive points are Information System (SARMIS) to C/P administration observed in the expenditure execution at the is expected to help address this challenge. provincial level. First, there has been smooth execution of key budget Chapters, including For Preah Sihanouk province, limited legal personnel (Chapter 64), purchase of goods (Chapter ownership over land coupled with land conflicts 60), services (Chapter 61) (excluding infrastructure is an immediate and sensitive issue. This is partly repair and maintenance (Chapter 6105), and others. a reflection of the overall limitation regarding titling of Second, the on-time approval of the procurement land in the country and partly because of the high plan (i.e., before the budget starts) has also been land price speculation in Preah Sihanouk province. There are still many cases where a public institution helpful in minimizing procurement and spending (including SNAs) could not move ahead with having delay. Third, the transfer of the shared tax revenue their land titles approved because there are still many from the provincial administration to the DMK (Sub- land conflicts involved. Addressing these conflicts account 65012) and CS (Sub-account 65013) requires intervention from both the provincial and level have not met any major challenges. These national levels. are reflected in the percentage of annual budget Cambodia Public Finance Review - From Spending More to Spending Better 161 execution for each budget chapter and the percent except a general understanding that new investment of budget executed in the last quarter, as presented is when completely new infrastructure is built, in Table 6.2. However, there is persistent over- whereas repair and maintenance is an investment execution or under-execution for the “investment” done on an existing project. With this definition, for budget, which was understood at SNA as Preah Sihanouk provincial administration, any work ‘infrastructure repair and maintenance’ (Account on road, irrigation, water system, etc., is considered 6105), capital budget (mainly Chapter 21), and as ‘repair and maintenance’ because most of the projects (either infrastructure or non-infrastructure) work is done on locations that used to have a similar funded by development partners. project before.132 New capital investment, on the other hand, tends to be associated with new building 6.2.5. Management of investment construction (e.g., offices, schools, hospitals). projects Therefore, a repair and maintenance (Chapter 6105) or new capital investment (Chapter 21) is referred to In the current process, repair and maintenance as “investment projects”. and new capital investment are treated the same from the budget execution perspective. Preah Sihanouk provincial administration There are no clear criteria to distinguish the two, established the required rules and regulations Box 6.1. Key rules and regulations for investment projects at provincial levels 1- The preparation and approval of a Capital/provincial project is guided by the 2014 Guideline on CP Project Preparation Manual. This over 200-page document provides details on how a CP project is defined and prepared, including the setup of a project preparation committee, identification of technical people, land and environment impact assessment, project cost estimation, local contributions, project operation and maintenance, and various templates that a CP administration needs to use. It is important to note that, according to the manual, all the CP project plans need to go through two main technical and legal reviews: 1) by relevant provincial line departments, and 2) by the Ministry of Interior (MOI), General Department of Administration. 2- Public procurement covers procurement planning, implementation, and payment. According to the SNA-BSRS and the MEF annual circular, the provincial administration needs to prepare its annual procurement plan and have it approved by the start of the budget year. As of now, a provincial administration is given authority to manage a project procurement if it is under KHR 500 million. For those of higher value, additional approval from the General Department of Public Procurement (GDPP) in Phnom Penh is needed. However, to support further fiscal decentralization, the project value threshold will be further increased, and the administration will be given more authority over the procurement process. A budget spent through the procurement process is also subject to slightly different rules regarding payment, especially relating to the submission of supporting documents. 3- How an investment budget execution fits within the broader annual planning and budgeting process and calendar is guided by two particular rules: the procurement plan, including investment projects, approved before the start of a budget year; and all budgets, including the investment, which need to be executed by the year-end. Sources: World Bank staff compilations based on various law and regulations 132 For the same types of projects conducted by CS, they are classified under investment expenditure (Chapter 21), not repair and maintenance (Chapter 6105). 162 Cambodia Public Finance Review - From Spending More to Spending Better Figure 6.12. Monthly execution rate of repair and maintenance (Chapter 6105) and investment expenditure (Chapter 21) in 2020 82% 50,000 40,000 KHR, million 30,000 20,000 10,000 10% 6% 74% 26% 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Chapter 6105: Repair and maintenance Chapter 21: Capital investment Source: MEF, FMIS (2022) (Box 6.1), but key challenges remain. The for investment projects -both repair and maintenance administration reportedly followed the 2014 CP (Chapter 6105) and investment expenditure (Chapter Project Preparation Manual and had their project 21). Supplemental requests were usually made in plans approved by both technical line departments September or even October of each year, a time and the MOI. They have also managed to prepare when the provincial administration could make a and have its procurement plan approved by the reasonable estimate on how much extra tax revenue time each budget year starts. However, for repair they can collect compared to the tax collection plan. and maintenance (Chapter 6105) and investment expenditure (Chapter 21), there have been significant Second, the severe under-forecast of tax revenue variations in execution as compared to the planned undermines the tax revenue plan’s credibility budget, which implies high unpredictability of the and results in high additional budget requests budget itself. In addition, there was a significantly high late in the year. According to the 2017 Guideline on percentage of budget execution in the last quarter PB Implementation, a provincial administration can and the last month of the fiscal year, which suggests request a supplemental budget when it becomes inefficiencies in the execution process. For instance, clear that its revenue performance will exceed the 82 percent of repair and maintenance (Chapter 6105) planned amount. However, the guideline does not and 74 percent of investment expenditure (Chapter provide detailed guidance, except to say that the 21) were executed in December 2020 (Figure 6.12). MEF will examine such requests accordingly. In the case of Preah Sihanouk province, as indicated earlier, A combination of three factors has contributed the administration started to get a rough estimate of to the higher execution rate and high payment tax revenue over collection around September or percentage in the last quarter. First, overall October. Better information about revenues is the planning and common practices push more basis for the additional spending request. The late spending toward year end. It is common for the request means late approval and, in turn, a rush to Preah Sihanouk provincial administration to request finish the projects by the year-end. The resulting time additional budget credit late in the year. According to constraint has hypothetically undermined proper the MEF’s Guideline of PB implementation for SNA public procurement and project execution. (2017), an SNA can ask for in-year budget virement and supplemental budget starting from the second Third, delay in procurement process impacts semester/ third quarter. For the period of 2018 -2020, budget execution of investment projects. The a large part of the supplemental requests had been over-threshold procurement projects (Table 6.3) Cambodia Public Finance Review - From Spending More to Spending Better 163 Table 6.3. Procurement thresholds for SNA SNAs Maximum procurement values Provincial administration KHR 500 million Capital administration KHR 800 million DMK administration KHR 100 million Source: MEF (2019) mean longer time is needed not only for procurement investment projects). Additionally, there are other but also for payment, resulting in delays in budget ongoing legal and policy changes with relation to execution toward year end. The procurement SNA planning and budgeting, and it is critical that process in such cases needs approval from the the new PIM guideline be consistent with those new provincial administration, the PDEF, and the GDPP changes. of MEF in Phnom Penh. Similar steps also apply when a contractor seeks verification of the project 6.2.6. Expenditure arrears deliverables for payment. Because of the time In Cambodia, the Government needs to pay a involved, in many cases contractors chose not to financial obligation within 60 days of receipts claim the first payment but waited to receive the final of payment request at the General Department and 100 percent payment at year-end. This practice of National Treasury or it would be considered has systematically created a high budget execution as an arrear.133 For Preah Sihanouk provincial rate in the last month of each fiscal year. In 2020, for administration, no payment arrears were reported for instance, nine out of the eleven investment project recurrent expenditures following the above definition. procurements that the Preah Sihanouk provincial Salary, utility, and goods and services purchased administration had were considered over-threshold, through direct payment (not procurement) were according to the criteria in the table below. reportedly done on time. For all non-procurement spending, reported times for payment processing A new Guideline for SNA Public Investment are in line with the data provided by the FMIS which Management (PIM) has been drafted. The main shows that, on average, the number of days taken idea is to align public investment at SNAs with the from the date a payment order is signed by provincial national strategy on PIM reform and the new sub- administration to payment made by the treasury was decree #41 on PIM. It may take more time for the about 7.5 days for 2019 budget year and about 10.5 new guideline to be approved, partly because of the days for 2020 (Table 6.4). need to agree on the definition, scope, and value thresholds of what constitutes ‘investment’ and The extent of the expenditure arrears for the ‘SNA investment projects’ (as opposed to national spending made through procurement could Table 6.4. Number of days between payment orders and payments (non-procurement) # of days from payment orders signed by provincial Budget years # of transactions administration to payment made by the treasury Average # Min # Max # 2019 151 7.5 days 0 70 2020 132 10.5 days 0 59 Source: MEF, FMIS (2022) 133 Please see, for instance, RGC (2020) 164 Cambodia Public Finance Review - From Spending More to Spending Better not be identified. This is because of a number of Portal (Generation 1) is also rolled out in 2022 but specific reasons. First, for procurement spending, focuses only on selected districts and communes. the rules as indicated in the MEF Circular #006 (2016) explicitly indicate that an ‘invoice’ is not a required All seven budget classifications were included supporting document to be submitted for payment in the FMIS system design, but not all of processing. Instead, for the Provincial Treasury, them have been used. Since the adoption of the when calculating the number of days taken for a program-based budget, program classification has payment to be processed, the date of the “payment been used along with the economic classification. order” (not of an invoice) submitted from a spending However, the functional classification has not been agency is used to compare with the date of actual used. Other specific FMIS system-related issues that were raised include a request to have a better payment. The payment order date is not the same as synchronization between FMIS and NRMIS and a the invoice date. Therefore, the difference between request for people who work on account receivable payment order and invoice date cannot be verified and account payable in FMIS to have access to the as contractors are not required to submit invoices. system. Second, the ‘goods/service receipt’ date The Preah Sihanouk provincial administration is not helpful either. The Circular #006 (2016) has since 2021 started using the NRMIS to requires that ‘goods/service receipt’ be included in record non-tax revenue. Preah Sihanouk province the payment request documentation. However, at officials appreciate having this new system, but least for the investment budget - including repair they need more capacity building, technical support and maintenance (Chapter 6105) and investment and guidance on how the NRMIS is linked and expenditure (Chapter 21) (which accounts for well complements the MIS of the MOI. They also seek to over 70 percent of total provincial administration), as better understand how the NRMIS will be connected discussed earlier, the procedures for ‘goods/services with the FMIS. Other specific NRMIS system-related receipt’ verification have not been strictly followed. problems that were raised include how to make the printing from the system easier and reviewing or Third, invoice date and payment order date are removing specific data entries in the NRMIS which used interchangeably in the FMIS entries. One does not exist in the paper-based process. of the inconsistencies is that while there is an entry column titled ‘invoice date’ in the system, in practice, There are four specific issues relating to the the entered data is sometimes a ‘payment order’ financial recording that should be addressed date, if there is no invoice required. in the short term. First, there is a recorded difference between revenue and spending, for 6.2.7. Transparency and oversight both plan and actual. According to the 2011 SNA budget law (Article 18), the budget of SNAs needs Reporting, recording, and FMIS to be prepared by showing the balance between revenue and expenditure. For Preah Sihanouk The roll-out of FMIS is still new to Preah province, the planned budget shows recorded Sihanouk provincial administration. Up until balance for 2018 (i.e., the proposed revenue and 2022, the Provincial Department of Economy and expenditure are equal), but not in 2019 and 2020 Finance (PDEF) helped enter financial information to (Figure 6.13). FMIS for the provincial administration. Since 2022, per MEF’s instruction No. 002, the FMIS is rolled Second, more clarification is needed on the out to all CP administrations, covering four main recording of the surplus. According to Article 41 functions, namely: budget allocation (BA), purchases of the 2020 Budget Law, the budget surplus needs (PO), account payable (AP), and account receivable to be transferred back to the national level into the (AR). For the DMK and CS level, in 2022, an FMIS ‘SNA Fund.’ However, the details on how exactly the Cambodia Public Finance Review - From Spending More to Spending Better 165 Figure 6.13. Recorded revenue and expenditure difference, plan and actual (2018-2020) 250,000 140,000 120,000 200,000 KHR, million KHR, million 100,000 150,000 80,000 100,000 60,000 40,000 50,000 20,000 0 0 Plan Actual Plan Actual Plan Actual 2018 2019 2020 Revenue Expenditure Difference (RHS) Source: MEF, FMIS (2022) Note: RHS = Right-hand scale transfer shall be made, what recording is done, and 61) (not payroll expenses-Chapter 64). Although this how the SNA Fund is allocated back for the following is just one example, the study recommends that the year’s budget are not provided. budget recording be thoroughly reviewed to ensure that the current chart of accounts can accommodate The third is the recording of transfers from the the more complicated budget transactions arising national level. Currently, the transfer is recorded from more fiscal transfers to SNAs. under Chapter 75, with three main accounts mainly in use: 1) Account #75011 for transfer from national level, 6.3. The district and 2) Account #7502 for transfer from other SNAs, and 3) Account #7508 for transfer from other sources. As municipality administrations fiscal decentralization progressed, different types of transfers were used, including unconditional transfer, 6.3.1. Overall revenue and conditional transfer, and budgeting for performing expenditure at the DMK level delegated functions. The current chart of accounts is not sufficient to differentiate these different transfer Like other DMKs in the country, the DMK mechanisms and thus cannot generate needed administrations in Preah Sihanouk province disaggregate data for analyzing budget allocation rely heavily on the budget transfer from the and execution in an intergovernmental setting. national level and provincial level. From 2018 to 2020, the national transfer, which accounts for The fourth issue is the use of specific more than 70 percent of total actual revenue, has accounts and sub-accounts to record certain steadily increased, although at a lower rate in 2020 transactions. The study did not exhaustively focus (Figure 6.14). The transfer from the provinces, which on this issue, but did identify one example wherein was set at 4 percent of the collected tax revenue the use of certain accounts might not reflect the of the province, also increased between 2018 and nature of the recorded transactions. The example 2019 but dropped in 2020. The provincial shared tax is the use of Sub-account #6403 (reward) to record was not included in the planned budget of the DMK, incentive sharing that Preah Sihanouk provincial but was treated as additional budget credit during administration made to the staff of line departments the year, and therefore, is shown only in the actual who help provide the services through OWSU. The budget execution column. The revenue from service use of this account can be misleading because they fees generated by the DMK OWSO also increased are not for the administration’s own personnel, but steadily, although at a lower rate in 2020. for the purchase of external services, in which case, it should be recorded under service expenses (Chapter On the expenditure side, a large proportion of 166 Cambodia Public Finance Review - From Spending More to Spending Better Figure 6.14. Revenue of the Preah Sihanouk province DMK (including transfer), plan and actual (2018-2020) 9,000 8,000 20.0% 9.8% 7,000 0.0% 6,000 0.0% KHR, million 20.4% 5,000 0.0% 76981: Others 4,000 85.7% 77.3% 70.1% 75061: Revenue from performing 3,000 94.7% delegated functions from line ministries 95.3% 74.2% 75021: Provincial shared tax 2,000 75011: National transfer 1,000 9.0% 14.3% 12.1% 73048: Revenue from other services 0 4.7% 5.3% 5.3% Plan Actual Plan Actual Plan Actua 73028: Own-source revenue (OWSO) 2018 2019 2020 Source: MOI (2021) data Figure 6.15. Expenditure of the Preah Sihanouk province DMK by economic classification (2018-2020) 8,000 7,000 KHR, million 6,000 5,000 4,000 63% 3,000 65% 2,000 70% 1,000 22% 16% 23% 0 2018 2019 2020 Chapter 60: Purchases Chapter 6105: Repair and maintenance Chapter 61: Services Chapter 64: Payroll expenses Chapter 62: Social bene ts expenses Chapter 69: Reserve Source: MEF, GDNT the Preah Sihanouk province DMK budget has of service delivery. Since 2019, the DMK level been spent on salary (Chapter 64), leaving about has gone through significant reforms, with the 20 percent on infrastructure (mostly road) repair organizational restructuring and integration and and maintenance (Figure 6.15). This is a common the transfer of 55 functions and about 20,000 staff pattern, not just in Preah Sihanouk province. 134 With (country-wide). In Preah Sihanouk province, the the small overall budget and large wage share, there DMK administrations have not yet been able to fully has not been much room for development activities. operationalize the new roles and responsibilities. Sihanoukville municipality officials suggested that during the COVID-19 pandemic, they had also spent 6.3.2. Specific gaps and limitations some of the budgets for COVID-19 prevention and on service delivery at the DMK level intervention activities, but had only a small amount for social benefits expenses (Chapter 62). Key policy gaps and capacity issues are impacting the implementation of the 2019 Despite some increases, DMK budget remains functional transfer and, therefore, a vision small and is impacting the quantity and quality for service delivery improvement has not yet 134 World Bank (2022) Cambodia Public Finance Review - From Spending More to Spending Better 167 materialized. First, there had not been the necessary illustrated through the cases of clean water and solid decisions to classify the 55 functions (and their sub- waste management below, however, the DMK do not functions) into obligatory versus permissive functions, necessarily need to provide the services themselves, and determine the relevant funding mechanisms for but can instead play coordinating, regulating, and each. Related to this point, it was also found that supporting roles in partnership with the private no costing had been done to identify the amount of sectors to ensure the services are delivered (Box 6.2). budget needed to perform the functions. Without such information, the discussion on the amount of Third, DMK officials are aware of the changes; budget to transfer had referred mainly to the budget however, their detailed understanding of how that line ministries and their provincial departments they should perform those functions varies. had allocated and spent to perform the functions For those in education, health, and solid waste before they were transferred. management where decentralization had been more advanced, the officials in charge tend to have better Second, the roles and partnerships with the awareness of the reform. However, in other sectors private sector in delivering essential services such as agriculture and water resource management, have not been adequately explored. There seems the level of understanding is lower. Regardless of to be an expectation among local level officials and sectors, DMK officials normally requested more clarity certain advocacy groups that the DMK needs to have on how they should perform their daily operations much more funding from the national level to actually under the new working arrangements. Those ‘provide’ the services assigned to them. This operational aspects include 1) regular administration, expectation partly reflects an assumption that the such as the process for approving everyday DMK will actually provide those services themselves. As administrative paperwork, permission and the use of Box 6.2. Roles and engagements of private sector in DM service delivery Existing policy and legal frameworks expect DMK administrations to collaborate with the private sector and civil society organizations (CSO) in service delivery. According to the NP-SNDD Phase 2 (Component 5), the collaboration with the private sector can take the following forms: • A company/CSO with a contract from the SNA to deliver services, funded from the budget of SNA, • A company or CSO providing fee-based services under an agreement with the SNA (the benefits are financed from the fees revenue, not from the SNA budget), or • A public-private partnership in which the SNA is the owner of assets used by a firm to deliver services; for example, a waste disposal site is owned by the SNA but operated by a private firm. The NP-SNDD Phase 2 expects the partnership with the private sector and CSOs to be delivered through a number of actions. These include 1) registry of local businesses or enterprises, 2) developing and implementing appropriate models of partnership with the private sector (including PPP), and 3) developing and implementing a framework for contracting-out of public services. Specific steps are proposed for each of the main actions, but more needs to be learned, especially when developing the mentioned models and frameworks. The engagement from the private sector with DMK is supposed to be systematic and regular. At the DMK level, according to sub-decree #182 and #184, the engagement is expected to happen through the existing participatory planning process and/or specific mechanisms and instructions created by each DMK. The sub-decrees also mention that DMK administrations need to hold regular meetings with the private sector and CSOs to identify and solve the problems of local people. In reality, the inclusion of non-state actor financing and investment in the local plans has been minimal and uneven. Sources: World Bank staff compilations based on laws and regulations 168 Cambodia Public Finance Review - From Spending More to Spending Better budget for travel and mission, the process of getting with the number of DMK and less with economic permission to work with NGO partners, reporting on activities. For instance, Preah Sihanouk province, staffing matters, and the use of budget funds for hiring despite its economic significance, gets only 137 civil contract staff; 2) the estimated additional cost that servants transferred from line ministries, a number should be covered to deliver the newly transferred similar to or lower than that of Steung Treng, Kratie, functions and the public financial management Udor Meanchey, and Koh Kong (Figure 6.16). processes that accompany these transfers; and 3) the human resource management functions (e.g., Fifth, the working relationship between the reporting lines, performance management) within the DMK and line departments requires improved revised DMK administrative structure (e.g., among coordination to support local service delivery. district technical officials, the DMK BoG/councils and There are four systematic challenges that need to line departments at the CP level). These points were be addressed. The first relates to the adequacy and also raised in the recent study by the World Bank. 135 consistency of decentralization rules and regulations on the one hand and sectoral rules and regulations Fourth, personnel shortage is still a concern on the other. As illustrated in the case of urban despite the recent transfer. According to the water supply below, while the function of managing progress report by NCDD-S, almost 20,000 staff private water operators was already transferred members were transferred to the DMK in January to the DMK, the concerned line ministry still relies 2020, the bulk of which were from MOI and MOEYS. on its 2014 Prakas to regulate the services. The For other ministries such as tourism, mining, second is the lack of personnel (and capacity) to be commerce, and labor, the number of staff members in charge of certain services. The third is the fact that transferred was less than 50 in the whole country the 2019 functional transfer has not been reflected (with more than 200 DMK). Province-wide, the into the regular planning and budgeting exercises of number of personnel transferred seems to correlate concerned line departments. The last challenge is Figure 6.16. Personnel transfer to DMK by ministry and province (2020) Civil servants transferred to DMK by Civil servants transferred to line ministry DMK by province Water resources Kandal Environment Takeo Mining Kampong Cham Kampong Speu Industry Kampot Commerce Kampong Chhnang Labour Banthey Meanchey Information Phnom Penh Pursat Planning Tbong Khmum Agriculture Udor Meanchey Tourism Kratie Siem Reap Religion Steung Treng Public work Mondul Kiri Justice Ratanak Kiri Post and telecom Preah Vihear Koh Kong Social affairs Kampong Thom Women affair Battambang Culture Preah Sihanouk Rural development Svay Rieng Prey Veng Land management Kep Education Pailin 0 500 1000 1500 2000 2500 0 200 400 600 800 Source: NCDD-S (2021) 135 World Bank (2022) Cambodia Public Finance Review - From Spending More to Spending Better 169 the inconsistency in the use of data for planning and and 72 percent for investment expenditure (Chapter M&E purposes by the DMK administrations versus 21)) presented earlier for the same year. the line departments. No serious concerns were raised on the 6.4. The commune/sangkat procurement process although a delay in CS project document preparation was administrations identified. As of 2021, the CS procurement was done following the 2005 PK#231 and the 2017 The CS is the lowest level of SNA and has Project Implementation Manual (PIM). For 2020, the longest established budget management the start of CS project document preparation, system. As of 2020, the average total budget for which is needed for procurement, experienced each CS in Preah Sihanouk province was about delays resulting in the postponement of project KHR 320 million (or about US$80,000), virtually all of which (97 percent in 2020) came from national implementation. For example, the provincial level transfer. In 2020, the CS allocated most of its budget administration sent technical officials to conduct to capital investment classified under the investment the project study in May in 2020 and completed the expenditure budget (Chapter 21) (Figure 6.17). There required project documents in August. However, has been a small variation between the planned and by the time procurement completes, it is close to actual budget for CS from 2018 to 2020. The PFM November when project implementation can begin. rules for CS are stated in the CS Administration Law, Consequently, the MEF issued Circular #002 in early sub-decree on CS fund, and various Prakas and 2022 to adjust the timeline of CS project preparation Guidelines. Specific rules on CS budget management and implementation by following the 2020 PK #0149 are also stated in the annual budget circular. (discussed earlier). A few key points are noted as follows: In this study, we focus on the execution of investment expenditure budget (Chapter 21), CS administrations need to finish preparing their 1. including procurement, at the CS level. The data 3y-IP as early as February of year N-1 and send it show that the execution of the CS budget has spread upward to the DMK level. throughout the year. Although a moderately high Provincial administrations need to send technical 2. percentage of budget (38 percent) was executed officials to conduct the project technical study in December for 2020 (Figure 6.18), this proportion starting in April and finished by August of year is much smaller than the case of the province (82 N-1. percent for repair and maintenance (Chapter 6105) CS administrations need to conduct their 3. Figure 6.17. Budget allocation for CS budget (2018-2020) 12,000 10,000 KHR, million 8,000 56% 6,000 38% 43% 4,000 45% 31% 2,000 39% 0 2018 2019 2020 Chapter 60: Purchases Chapter 6105: Repair and maintenance Chapter 61: Services Chapter 64: Payroll expenses Chapter 62: Social bene ts expenses Chapter 69: Reserve Chapter 21: Investment expenditure Source: MEF, FMIS (2022) 170 Cambodia Public Finance Review - From Spending More to Spending Better Figure 6.18. CS budget execution by months for road construction in 2020 1,800 38% 1,600 1,400 KHR, million 1,200 1,000 17% 800 13% 600 9% 400 6% 6% 2% 3% 200 2% 2% 1% 1% 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Source: MEF, FMIS (2022) procurement from January to March of year N. following the SNA-BSRS and the SNA PB CS 4. administrations cannot make advance guideline. Specific areas to consider include payments for contractors, but must follow a functional review of the current program structure payment schedule of 45 percent (for the first 50 and how to prepare a plan and report using the percent of work finished), 45 percent (for the other program and subprogram structure. 50 percent of work finished), and 10 percent after 3. Improving data management for SNA planning six months to assure project quality. by 1) introducing a more computer-based and online data sharing, 2) creating more structured 6.5. Recommendations planning data worksheets that are linked across different components, and 3) strengthening the 6.5.1. Provincial administration capacity of the PDP and the PDEF. Strengthening effectiveness of public finance 4. Strengthen spatial planning by allocating more at subnational level will require improvement budget to support the Provincial Departments of in both budget planning and execution. Key Land Management to work closely with SNA to challenges include weak spatial planning (urban/ finish the spatial plans for the province. sectoral planning); disparity between planning and 5. Clarify and concretize functional assignments for actual spending; inaccurate revenue forecasts from each tier of SNAs. The clarification of such roles both tax and non-tax revenue (NTR); and expenditure will minimize duplication of investment among bunching toward last quarter for capital investment. different SNAs. This will require a clearer legal and policy framework. To improve allocative efficiency, there is a need 6. Clarify and institutionalize SNA roles in social to improve provincial administrative planning protection, especially social assistance such as in terms of integration among different tiers of cash transfer schemes. The COVID-19 crisis has SNAs and alignment with line departments. This shed more light on the important roles of SNAs in could be achieved through the followings: pandemic prevention and intervention. However, 1. Improve the provincial BSP preparation using such roles have not been explicitly mentioned in the to-be-adopted assessment criteria, coupled the legal framework. with additional capacity building and on-the-job Improving allocative efficiency also requires training for planning and finance officials of the improvement in overall revenue (both tax and non- province. tax) forecast through the following: Strengthen program budget implementation 2. Cambodia Public Finance Review - From Spending More to Spending Better 171 7. Improve tax revenue forecast, especially for versus permissive functions for the 55 functions and the main tax revenues such as the stamp the costing of specific functions. taxes on property revenue. In addition, the accountability line between the GDT and the Encourage/require DMK administrations to set up provincial administration should be clarified and clear mechanisms and updated data on private institutionalized. sector engagement in local service delivery 8. Improve NTR forecasting, especially on NTR from and development. This can be done following the state assets that have not been well managed requirement set out in the SD#182 and #184 and the and experienced high plan-to-actual variation. strategic directions indicated in the NP2. 9. Introduce a formula for budget allocation across the three tiers of SNAs (to achieve vertical Personnel-related challenges require urgent balance) and across SNAs of different regions (to attention. There is a need to address personnel achieve horizontal balance), especially in relation shortage for key services such as water supply, solid to the allocation of the SNA Fund intended for waste, and mining. Additionally, there is a need to budget surplus and the revenue sharing from CP clarify how technical staff at the DMK administrations to DMK and CS levels. should perform their daily operations under the new 10. Improve the NTR system, processes, and working arrangements: the capacity of those in charge of managing the revenue and continue transferring more The working relationship between the DMK and NTR functions from line ministries to provincial line department is another immediate challenge administrations and the DMK level. to be addressed. This includes the need to: The improvement in revenue forecasting (both 1. Identify relevant sectoral regulations that need to tax and non-tax revenue) is also important to be revised to be in line with the D&D reform, enhance operational efficiency. This will reduce 2. Ensure that line departments factor in the recent the variance in plan and budget execution. functional transfer in their regular planning and budgeting exercises. State asset management is an area deserving 3. Address the inconsistency in the use of data attention, especially for Preah Sihanouk for planning and M&E purposes by the DMK province. This can be done by assessing the extent administrations versus the line departments. to which the current asset inventory is comprehensive and reflects reality, especially in terms of asset use 6.5.3. CS administrations and asset value. For the longer-term, however, an improved governance arrangement for SNA state After the adoption of the new Circular on CS management needs to be developed, starting with procurement, more monitoring and capacity ensuring a more coherent legal and regulatory building is needed. The circular represents the framework. policy and regulatory responses to the problem of delay associated with the CS development 6.5.2. DMK administrations project implementation. However, like many other regulations, to be effective, local officials need to be The persistently small budget at the DMK aware of its content and intended changes, and their level, despite some increase, raised concerns compliance needs to be monitored and feedback with regard to service delivery. Improvement of collected for future adjustments and strengthening. budget allocation to DMK level will be necessary for improving public service delivery. This should be Table 6.5 shows recommendations to improvement done along with the process of clarifying obligatory PFM at provincial, DMK, and C/S level. 172 Cambodia Public Finance Review - From Spending More to Spending Better Table 6.5. Recommendations by timeframe Medium-term SNA Areas Short-term recommendations (1-2 years) recommendations (3-5 years) Province Strengthen • Continue to improve the preparation of the • Invest more in spatial planning provincial BSP, the integration of DMK and planning to support the capacity CS plans into the BSP, and the program- Provincial Department budget planning of Land Management • Strengthen data management to assist to work closely with in planning and budgeting exercises with SNA to finish the spatial enhanced use of ICT for data collection, plans in the province storage, and sharing; complemented by strengthening the capacity of the Provincial Department of Planning (PDP) and the Provincial Department of Economy and Finance (PDEF) • Better collection and use of information on national-level investment, private investment, and financing by CSOs to inform more effective planning and resource allocation Improve overall • Improve tax revenue forecast, especially • Consider increasing tax revenue for the main tax revenues such as the the percentage of tax forecasting stamp taxes on property. In addition, the revenue sharing from capability accountability line between the GDT and CP to DMK and CS the provincial administration should be levels in order to reduce clarified and institutionalized the current vertical • Improve tax revenue management and imbalance among the accountability relationship between the three tiers of SNAs GDT provincial branch and the provincial administration Improve NTR • Improve forecast of revenue from state assets • As the capacity at the forecasting • Improve the system, process, the capacity SNA improves, consider of those in charge of managing the transferring some of the revenue (including on the use of NRMIS) NTR to the DMK level • Continue transferring more NTR functions from line ministries to provincial administration and do so in a more coherent manner Clarity of • Clarify and concretize functional • SNA’s roles in social functional assignment for each tier of SNAs, protection (especially social assignment especially between the provincial assistance such as cash administrations and municipalities on transfer schemes) should infrastructure development be explicitly mentioned and institutionalized Strengthen • Assess the extent to which the current • Develop a stronger state asset asset inventory is comprehensive and governance arrangement management reflects reality, especially in terms of asset for SNA state asset use and asset value management, starting (urgent with ensuring a more attention coherent legal and needed) regulatory framework Cambodia Public Finance Review - From Spending More to Spending Better 173 Medium-term SNA Areas Short-term recommendations (1-2 years) recommendations (3-5 years) Improve • Improve the timing of project preparation • Clarify and, where budget and procurement to allow sufficient time necessary, adjust the execution for quality project implementation and existing COA and processes payment recording to better reflect • Ensure timely and explicit instruction on the real nature of specific new procedures/changes (e.g., per SNA- transactions (such as BSRS) certain non-tax revenues, capital spending, and • Continue to implement the Public incentive sharing) Procurement Strategy to decentralize more authority to the SNA level • Build capacity for SNAs with better coordination (including with National School of Local Administration (NASLA)) and more use of digital technology DMK Improve • Conduct detailed costing of specific functions • Improve vertical budget • Address the budget gaps at the DMK level balance by increasing availability as a matter of urgency the percentage of tax and clarity of revenue sharing from • Identify relevant sectoral regulations that functions CP to DMK and CS need to be revised to be in line with the levels D&D reform, ensure that line departments factor in the recent functional transfer • Move toward in their regular planning and budgeting decentralized finance exercises, and address the inconsistency and HR management in the use of data for planning and M&E with centralized purposes by the DMK administrations information supported versus the line departments by modern shared platform GovTech • Encourage/ require DMK administrations system to set up clear mechanisms and updated data on private sector engagement in local service delivery and development. This can be done following the requirement set out in the SD#182 and #184 and the strategic directions indicated in the NP2 Improve HR • Address personnel shortages for key management services such as water supply, solid waste management, and mining • Clarify how technical staff at the DMK administrations should perform their daily operations under the new working arrangements CS Address delays Support CS officials to implement the circular Monitor the implementation to project associated with the CS development project and collect feedback for development implementation to prevent delays to project future adjustments and and development and preparation strengthening preparation 174 Cambodia Public Finance Review - From Spending More to Spending Better REFERENCES RGC. (2018) Kingdom of Cambodia: Rectangular strategy Phase 4. Phnom Penh. RGC. (2019) Kingdom of Cambodia: National Strategic Development Plan (2019-2023). Phnom Penh RGC. (214) Kingdom of Cambodia: National Strategic Development Plan (2014-2018). Phnom Penh ADB. (2019). Kingdom of Cambodia: Preparing the irrigated agriculture improvment project. Manila. CAVAC. (2019). Sustainable Irrigation Services Impact Evaluation. Phnom Penh. CAVAC. (2020). Structural transformation and the role of agriculture in the Cambodian economy: Past, present, and future. Phnom Penh. Gelotti, F., & Sin, S. (2016). Development of Master Plan for Crop Production in Cambodia by 2030. Phnom Penh: TSSD Project. IMF. (2001). Annex to Chapter 6: Classification of the functions of Government. Washington DC. MEF . (2022). Circular on BSP preparation for 2023-2025. Phnom Penh. MEF. (2020). Procedures for implementing investment budget from national budget (Chapter 21). Phnom Penh. MEF. (2022). Cambodia Climate Public Expenditure Review . Phnom Penh. MEF. (2022). Macro economic framework and MTFF for the national budget preparation 2023. Phnom Penh. MEF. (2022). Prakas no. 1028 on the Common principles for the management of irrigation O&M. Phnom Penh. MOWRAM. (2019). Natioanl Water Resources Management and Sustainable Irrigation Road Map and Investment Program 2019-2023. Phnom Penh. MOWRAM. (2019). Strategic framework for the irrigation sector. Phnom Penh. MOWRAM. (2022). MOWRAM BSP for 2022-2024. Phnom Penh. RGC. (2004). National Water Resource Policy for the Kingdom of Cambodia. Phnom Penh. RGC. (2017). Sub-decree No. 18 on Adminsitrative services by SNAs. Phnom Penh. RGC. (2019). Sub-decree No. 182, 183, 184 on the functional transfers to DMK adminsitrations. Phnom Penh. RGC. (2021). Royal decree No. 449 on the establishment of the commitee on public administrative reform. Phnom Penh. RGC. (2021). Strategic Framework and Program for Cambodia’s Economic Recovery (2021-2023). RGC. (2022). Medium Term Financial Framework (MTFF) for 2023-2025. Phnom Penh. RGC. (2022). SOP for PPP - Volume 1: Policies and Procedures. Phnom Penh. RGC. (2022). Vision for the coordinaiton of reform 3+1. Phnom Penh. World Bank. (2017). Technical assistance note: Camobdia Strategic Framework for the Irrigation Sector. Phnom Penh. World Bank. (2020). Public Investment Management Reference Guide. Washington DC. IMF. (2014). Prevention and management of government expenditure arrears. Washington, DC. MEF. (2017). The implement of program based budget for SNA. Phnom Penh. Phnom Penh. MEF. (2019). PK #10 on procurement authority for spending agencies including SNAs. Phnom Penh. MEF. (2020a). Guideline on the preparation of C/P BSP. Phnom Penh. MEF. (2020b). Circular on SNA budget preparation. Phnom Penh. MEF. (2021a). Progress on the five important PFM reform strategies. MEF. (2021b). Concept note on the preparation of a Master Plan to develop and transfor Preah Sihanouk province into a model multi-purpose special economic zone. Phnom Penh. MEF. (2022a). Circular #002 on CS project preparation, procurement and payment. Phnom Penh. MEF. (2022b). Instruction #2 on the use of FMIS at the Capital and provincial administration. Phnom Penh. Phnom Penh. Cambodia Public Finance Review - From Spending More to Spending Better 175 MEF. (2022c). Instruction #4 on the use of FMIS portal at DM and CS level. Phnom Penh. MISTI. (2014). PK #461 on the procedures for issuing, changing, extending, suspending, and withdrawing licenses of private water supply operators . Phnom Penh. MOI. (2017). PK #084 on the preparation of five year plan for provincial administration. MOI and MOP. (2020). PK #0149 on the additional guideline on the develpment of the 3 year investment program for CP, DMK and CS. Phnom Penh. MRD. (2017). PK #161 on the piloting of rural water supply functions in selected districts . Phnom Penh. NCDD. (2014). CP project preparational manual. Phnom Penh. RGC. (1995). SD #82 on the budget implementation. RGC. (2008). Law on Administrative Management of the Capital, Provinces, Municipalities,. Phnom Penh. RGC. (2009). SD #219 on the preparation of planning for SNAs. RGC. (2011). National Policy on Spatial Planning. Phnom Penh. RGC. (2020). PEFA Assessment 2020. Phnom Penh. RGC. (2022). Budget in brief for 2022. Phnom Penh. RGC. (2023). Budget in brief for 2023. Phnom Penh. NCDD. (2021). National Program on Sub-National Democratic Development (NP-SNDD) for 2021-2030. RGC. (2020). SD #168 on the Function and Duties of Internal Audit and Inspection. Phnom Penh. Preah Sihanouk province Provincial Administration. (2021). SHV Provincial Administration 3 year Investment Program (2021-2023). Sihanoukville. SHV provincial aministration. (2019). SHV 5-year development plan (5y-DP) for 2020-2024. SHV. World Bank. (2022). Cambodia’s intergovernmental architecture. Phnom Penh. World Bank; The Asia Foundation. (2013). Voice, Choice, and Decision 2 : A Study of Local Basic Service Delivery in Cambodia. . Washington, DC. RGC. (2022). Cambodia Demographic and Health Survey (CDHS) 2021-22. Phnom Penh. RGC. (2014). Cambodia Demographic and Health Survey (CDHS) 2021-22. Phnom Penh. RGC. (2010). Cambodia Demographic and Health Survey (CDHS) 2021-22. Phnom Penh. RGC. (2005). Cambodia Demographic and Health Survey (CDHS) 2021-22. Phnom Penh. RGC. (2021). Public Expenditure and Financial Accountability Assessment. RCG. (2019). General Population Census of the Kingdom of Cambodia. Phnom Penh. RGC. (2021). Cambodia Socio-economic Survey Survey (CSES 2021-22). Phnom Penh. RGC. (2015). Cambodia Socio-economic Survey Survey (CSES 2021-22). Phnom Penh. RGC. (2010). Cambodia Socio-economic Survey Survey (CSES 2021-22). Phnom Penh. RGC. (2005). Cambodia Socio-economic Survey Survey (CSES 2021-22). Phnom Penh. RGC. (2019). Sub-decree # 193 on “The transfer of management functions and health services delivery to municipal and provincial administration”. MOH. (2020). Annual Performance Report. Phnom Penh. MOH. (2021). Annual Performance Report. Phnom Penh. MOH. (2022). Health Management Information System. Phnom Penh. World Bank. (2020). Cambodia Health Financing System.Phnom Penh. World Bank. (2013). Cambodia’s Rural Health Markets Study and Quality of Care Study. Phnom Penh. World Bank. (2022). Impact Evaluation of Service Delivery Grants to Quality of Health Care Delivery in Cambodia. World Bank. (2019). Impact Evaluation of Service Delivery Grants to Quality of Health Care Delivery in Cambodia. World Health Organization. (‎2018)‎. Noncommunicable diseases country profiles 2018. World Health Organization. (2015). The Kingdom of Cambodia Health System Review. Kolesar et al. (2021). Advancing Universal Health Coverage in in Cambodia. 176 Cambodia Public Finance Review - From Spending More to Spending Better ANNEX Annex to Chapter 1 Annex 1.1. Fiscal position Table A.1.1. Fiscal balance (% of GDP) 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022e Overall balance (including grants) -4.9 -4 -3.2 -1.8 -0.5 -0.2 -0.8 0.4 1.5 -4.6 -7 -4.7 Prim ary balance -4.6 -3.5 -2.5 -1.1 -0.2 0.2 -0.4 0.8 1.9 -4.1 -6.5 -4.2 Total revenue 12.8 15 14.5 17.6 17.8 18.5 20.1 21.8 25.1 22.2 20.5 22.5 Tax revenue 10.1 11.4 11.8 13.8 14.8 15 16.2 17.4 20 18.3 16.7 18.8 Direct taxes 1.8 2.3 2.5 2.9 3.4 3.6 4.2 4.1 4.6 5.5 5.4 6.1 Taxes on goods and services 6 6.8 6.8 8.1 8.7 8.9 9.7 10.9 12.8 10.7 9.4 10.6 Taxes on international trade 2.3 2.4 2.4 2.8 2.7 2.5 2.2 2.4 2.6 2.1 1.9 2.2 Non-tax revenues 2 2.2 2 3.1 1.9 2.6 2.7 2.9 3 2 1.8 1.9 Grants 4.8 2.7 3.9 2.4 1.9 2.3 1.8 2 1.9 1.8 1.6 0.7 All other revenues 0.6 1.4 0.7 0.8 1.1 0.9 1.2 1.5 2 2 2 1.8 Total revenue (and grants) 17.6 17.7 18.4 20 19.7 20.9 21.9 23.8 27 24 22 23.2 Total expenditure 22.5 21.7 21.6 21.8 20.2 21.1 22.7 23.4 25.5 28.6 29 27.9 Current expenditure 10.9 11.6 11.8 12.3 11.5 12.7 14 14.3 14.6 15.5 15.6 15.5 Wages and compensation 4.3 4.6 5 5.7 6.5 6.8 7.6 8 7.4 8.1 7.2 6.6 Expenditure on goods and services 6.3 6.5 6.1 5.9 4.7 5.5 6 5.8 6.9 6.9 8 8.3 Interest payment 0.3 0.5 0.7 0.7 0.3 0.4 0.4 0.4 0.4 0.5 0.5 0.5 Capital expenditure 10.7 9.1 9.1 8.2 6.9 6.7 6.8 6.7 7.7 10.5 9.7 8.8 Other expenditure 0.9 1 0.8 1.2 1.8 1.7 1.9 2.5 3.2 2.7 3.7 3.6 Budget balance -9.7 -6.7 -7.1 -4.1 -2.4 -2.6 -2.6 -1.6 -0.4 -6.4 -8.5 -5.5 General government debt 29.7 31.6 31.7 31.8 31.2 29.1 30.3 28.7 28.1 33.9 35.2 33.7 Total financing 4.9 4 3.2 1.8 0.5 0.2 0.8 -0.4 -1.5 4.6 7 4.7 External financing 3.7 4.5 3.6 3.9 3.4 2.5 3.3 2.6 3.1 4.7 4.2 5.6 Cambodia Public Finance Review - From Spending More to Spending Better Amortization -0.3 -0.3 -0.4 -0.5 -0.4 -0.6 -0.6 -0.7 -0.8 -1 -1.1 -1.2 Domestic financing 1.4 -0.2 0.1 -1.7 -2.5 -1.7 -1.8 -2.3 -3.8 0.9 3.8 0.4 177 Memo item: Nominal GDP (billion of riels) 52,068.70 56,616.80 61,390.50 67,740.40 73,422.70 81,241.90 89,830.50 99,544.30 110,014.00 105,891.80 110,505.90 122,883.80 Table A.1.2. Selected indicators 178 Cambodia Public Finance Review - From Spending More to Spending Better Source: MFMOD, Budget Settlement Laws and World Bank staff estimates. Note: e = estimates. Table A.1.3. Loan disbursements Loan disbursement quickly increased, financing widening fiscal deficits as the pandemic hit 2019 2020 2021 2022 Amount % of Amount % of Amount % of Amount % of Top 3 bilateral creditors (US$ m) GDP (US$ m) GDP (US$ m) GDP (US$ mln) GDP China 322.1 1.2 365.5 1.4 285.5 1.1 301.1 1.0 Japan 83.8 0.3 149.8 0.6 382.0 1.4 386.2 1.3 Republic of Korea 44.8 0.2 67.0 0.3 72.1 0.3 60.4 0.2 Top 3 multilateral creditors ADB 159.1 0.6 421.9 1.6 189.0 0.7 238.3 0.8 World Bank 61.7 0.2 66.1 0.3 84.5 0.3 268.9 0.9 IFAD 10.4 0.0 19.4 0.1 11.2 0.0 29.8 0.1 Other bilateral/multilateral creditors 147.8 0.5 123.6 0.5 134.4 0.5 60.3 0.2 Total 829.7 3.1 1,213.4 4.7 1,158.6 4.3 1,345.0 4.5 Source: Public debt statistical bulletins. Note: US$ m= US$ million Table A.1.4. Fiscal projections (% of GDP) 2023p 2024p 2025p Overall balance (including grants) -6.4 -4.9 -4.2 Primary balance -5.9 -4.4 -3.6 Total revenue 22.8 23.5 24.6 Tax revenue 19.3 19.8 20.7 Direct taxes 6.3 6.4 6.7 Taxes on goods and services 10.8 11.3 11.8 Taxes on international trade 2.1 2.1 2.2 Non-tax revenues 1.7 1.6 1.7 Grants 0.6 0.5 0.4 All other revenues 1.8 2.0 2.2 Total revenue (and grants) 23.5 24.0 25.0 Total expenditure 29.9 29.0 29.2 Current expenditure 15.6 15.2 15.6 Wages and compensation 6.7 6.8 7.0 Expenditure on goods and services 8.3 7.8 8.1 Interest payment 0.5 0.6 0.6 Capital expenditure 9.5 8.7 8.4 Other expenditure 4.8 5.0 5.2 Budget balance -7.1 -5.5 -4.6 General government debt 35.1 35.6 35.9 Total financing 6.4 4.9 4.2 External financing 4.5 4.3 3.9 Amortization -1.3 -1.4 -1.4 Domestic financing 3.3 2.0 1.7 Memo item: Nominal GDP (KHR, billion) 133,268.8 145,073.8 157,731.7 Source: World Bank staff projections. Cambodia Public Finance Review - From Spending More to Spending Better 179 Annex 1.2. Tax expenditure analysis Tax expenditures are tax relief granted on a particular revenue stream to targeted beneficiaries. In many cases, tax expenditures are intended to stimulate specific sectors of the economy (e.g., automobile manufacturing in the case of Cambodia); preferential tax rates give them a competitive advantage vis-à-vis other sectors. On consumption taxes, commodities can be taxed at a lower rate or exempted; in some cases, commodities like food and pharmaceuticals can be zero-rated or exempt in a VAT system. The preferential treatment reduces the cost of a sector or commodity making goods and services more affordable, and skewing consumers’ decisions towards those sectors that produce the identified goods and services. This preferential treatment is called “non-neutrality.” Whereas the cost of direct expenditures is evident, without tax expenditure analysis relief from taxes is not measured, making a cost-benefit analysis of these decisions difficult. Cambodia tax expenditure on imports Cambodia’s tax expenditures on imports are substantial. Figure A.1.1 shows that the tax expenditures on the country’s imports were about half its revenue potential over this period. The columns show that between 2016 and 2019 the country’s share of collections lowered significantly but were reversed in 2020, likely due to COVID. In addition, the results show that a substantial amount of the relief was granted on VAT, which may not necessarily have been a tax expenditure because firms are eligible for input tax credits. However, foregone customs revenue went from 22 percent of total potential revenue to a low of 17 percent in 2019 rising back to 20 percent of potential revenue in 2020. Relief on special taxes was about 5 percent of potential revenue and petroleum taxes were negligible throughout the period. The data show that Cambodia made significant progress in rationalizing foregone revenue on imports prior to COVID. This suggests that the country can increase its collections on imported goods in the years to come. Figure A.1.2 shows the amount of foregone revenue from customs, special taxes and additional petroleum tax compared to amounts paid for the same three taxes from 2016 through 2020. Note that, VAT is excluded from this chart, because the bulk of Cambodia’s value-added tax is paid by firms that can subsequently claim input tax credits. Figure A.1.1. Forgone revenue as a share of Figure A.1.2. Foregone revenue vs. amounts collections paid (exc VAT) (KHR, billion) 5% 5% 6% 7,287 5% 5% 18 22% 20% 20% 10 19% 4,744 1,044 1,280 17% 9 875 4,078 10 8 734 611 3,084 2,675 3,968 4,083 3,443 27% 27% 25% 26% 2,851 3,032 23% 2016 2017 2018 2019 2020 2016 2017 2018 2019 2020 Customs Special Tax VAT Customs Special Additional Petrol Additional Petrol Receipts Table A.1.5 shows that Cambodia relieves significantly more VAT on imports than it collects. There is a trend toward collecting a greater share of potential VAT from 2016 to 2019; this is reversed in 2020, likely due to COVID. However, most of this relief is not a tax expenditure because the firms making imports are usually eligible for input tax credits. Granting VAT relief on imports makes the tax system unnecessarily complicated. Cambodia exempts very few commodities from VAT; charging VAT on importations would streamline the system reducing the costs of compliance for taxpayers and simplify administration for the government. 180 Cambodia Public Finance Review - From Spending More to Spending Better Table A.1.5. VAT foregone as a share of value added tax collected 2016 2017 2018 2019 2020 Forgone VAT/VAT collected 155% 168% 143% 120% 150% The bulk of the relief Cambodia grants on imports is foregone customs revenue, shown in blue, which rose from about KHR 2.9 trillion in 2016 to nearly KHR 4.1 trillion in 2020. Over the same period relief on special taxes, shown in orange, doubled from around KHR 600 billion to about KHR 1.3 trillion. At the same time, revenue receipts rose from about KHR 2.6 trillion in 2016 and peaked at about KHR 7.3 trillion in 2019, dropping back to about KHR 4.1 trillion in 2020. In revenue terms, the foregone tax on imported petroleum products is negligible. Table A.1.6 underscores that Cambodia’s tax relief on imports is generous. Between 2016 and 2020 Cambodia relieved about two-thirds of the potential customs revenue on imports whereas the share of relief on special taxes was about 22 percent in 2016, falling to 16 percent in 2019 and rebounded to 26 percent in 2022. Relief on both taxes fell steadily from 2016 to 2019 and rose sharply with the onset of COVID. Table A.1.6. Customs and special taxes relief as a share of potential collections 2016 2017 2018 2019 2020 Customs 66% 67% 66% 62% 70% Special tax 22% 20% 18% 16% 26% Between 2016 and 2020 the amount of revenue foregone on imports into Cambodia rose about 45 percent from around KHR 3.3 to nearly KHR 4.7 trillion. Moreover, figure 5.3 shows that while QIP relief dominates, foregone revenue through Special Economic Zones increased as a share of total relief from 2016 through 2020. Furthermore, the share of relief granted through mechanisms other than Qualified Investment Projects and Special Economic Zones also increased over the period analysed. The increase in average transaction value among firms benefiting from relief from methods other than Qualified Investment Projects and Special Economic Zones rose constantly from 2016 through 2019 and then quite dramatically in the last year (Figure A.1.3). Conversely, the average value of each transaction on a QIP declined fairly consistently over the horizon whereas the average relief on SEZ remains relatively consistent (Figure A.1.4). This raises the question about the homogeneity of the imported commodities. Figure A.1.3. Foregone revenue by relief Figure A.1.4. Average relief claimed per mechanism (KHR, billion) transaction 5,000 14 4,500 499 716 12 4,000 376 10 3,500 284 1,346 3,000 270 1,596 1,256 8 988 2,500 937 6 2,000 1,500 4 2,666 2,360 1,000 2,019 2,195 2,333 2 500 0 0 2016 2017 2018 2019 2020 2016 2017 2018 2019 2020 QIP SEZ Other Other relief SEZ QiP Cambodia Public Finance Review - From Spending More to Spending Better 181 Relief on importations by QIP and SEZ have very different profiles. The top five commodities are consistent among the five years and account for about 55 percent of imports for both QIPs and SEZ. However, the commodities imported by each are very different (Figure A.1.5). Typically, QIPs claim the greatest relief on fabrics used in garment production. The largest sources of relief for the SEZs are typically inputs into manufacturing processes. These results suggest that firms using SEZ may be making investments in higher value-added industries than those claiming relief through QIPs. Figure A.1.5. Top five imported commodities Figure A.1.6. Importations of construction equipment by type of investor 35% 50% 30% 45% 25% 40% 20% 35% 15% 10% 30% 5% 25% 0% 20% Vehicle parts SEZ Other knited fabric QIP Insulated wired SEZ knited fabric long QIP Woven fabric QIP Raw furskins QIP Articles of plastics SEZ Electrical apparatus SEZ Articles for packing SEZ 15% 10% 5% 0% 2016 2017 2018 2019 2020 2016 2017 2018 2019 2020 No TE Other TE QIP SEZ Interestingly, investments in construction equipment are negatively correlated with tax relief on importations. Construction equipment by QIPs declined significantly from nearly half of all importations to less than one quarter in 2020 (Figure A.1.6). Conversely, firms paying all their customs and excise duties on importations and construction equipment accounted for about 20 percent of importations in 2016 and, by 2018 they were the biggest importers of construction machinery and equipment. While the importations as a share of the total for these commodities dropped off between 2019 and 2020, the upward trend in the growth in expenditures is in stark contrast to cases where Cambodia forgoes tax on imports. Throughout the period we see that relief on construction equipment for firms in SEZs rose marginally and interestingly were below those claiming other forms of tax exemptions on there importations. This result suggests that Cambodia has additional capacity to mobilize revenue by targeting tax relief more effectively. Overwhelmingly, the relief on customs, special taxes and the additional tax and petroleum is granted through either QIPs or in SEZs. Table A.1.7 shows that in 2016 about 63 percent of the relief was granted through QIPS; this dropped to 51 percent in 2020. Conversely the use of Special Economic Zones rose from 29 percent in 2016 to 34 percent over the same period. Over the period total relief on imports rose, but not as rapidly as the value for duty. The relief granted through mechanisms other than QIPs and SEZs grew from 8 percent in 2016 and 2017 to 15 percent in 2020 (Table A.1.7). Throughout the observation period the use of Qualified Investment Projects and Special Economic Zones contributed significantly to the foregone revenue on imports. These programs are expensive and should be reviewed thoroughly to ensure that they are achieving their policy objectives. 182 Cambodia Public Finance Review - From Spending More to Spending Better Table A.1.7. Share of relief granted through QIP or SEZ 2016 2017 2018 2019 2020 QIP 63% 63% 59% 59% 51% SEZ 29% 29% 32% 30% 34% Non-QIPSEZ 8% 8% 9% 11% 15% Total 100% 100% 100% 100% 100% The top 5 exporting nations account for more than 80 percent of the customs relief granted to QIPs and SEZs. Exports from China accounted for more relief than all others combined. From 2016 through 2020 the Chinese share rose steadily from about 53 percent to more than 57 percent. Thai and Vietnamese exports traded places for second place over the period analyzed, typically around 10 percent of relief granted of imports into Cambodia among QIPs and SEZs. Figure A.1.7. Share of total QIP or SEZ relief 60% 50% 40% 30% 20% 10% 0% 2016 2017 2018 2019 2020 China Thailand Vietnam Hong Kong Japan Other Cambodia Public Finance Review - From Spending More to Spending Better 183 Annex to Chapter 3 Box A.3.1. Roles and responsibilities of public health administrations and facilities Cambodia’s health system consists of three levels of administration: (i) national-level administration, (ii) provincial-level administration, and (iii) district-level administration. • At the national level, MOH - consisting of three directorate generals - is responsible for managing the healthcare system in Cambodia. Specifically, (i) preparing, mobilizing funding, and implementing a health policy framework, (ii) developing laws and regulations governing the health services quality, infrastructure development of referral hospitals and HC/HP, (iii) preparing and implementing national preventive programs, (iv) preparing human resources development plan, core competencies of professional health care staff, and providing capacity development for health care workforce of a referral hospital, HC/HP, (v) developing standards of and procuring and distributing drugs and medical supplies/equipment to all health facilities, (vi) providing support to provincial/capital administrations in developing a budget and operational plan, and (vii) monitoring and evaluating the management and health services of provincial/capital administrations. • At the provincial level, provincial/capital administrations are responsible for (i) developing an annual plan and budget for delivering health care services at the provincial level, (ii) coordinating with DPs and NGOs, mobilizing funding, managing financial and human resources to implement the annual plan, (iii) requesting MOH for health professional staff and capacity building for health care staff, and (iv) reporting implementation progress to MOH and Ministry of Interior (MOI). • Provincial health departments (PHD) (25 provinces) are responsible for (i) coordinating with provincial administration to implement health policies and plans set out by MOH, (ii) distributing financial resources, supervising and monitoring the performance of Operational District (OD) and provincial referral hospital in delivering health care services, and (iii) providing continuous training for health care staff. From 2021, PHD received a budget allocation for provincial health from the respective provincial administration. • At the district level, 103 operational district (OD) offices - covering 208 administrative districts/khans/ krongs - were established in 2005. OD’s roles and responsibilities include (i) coordinating with PHD to implement health policies and plans within its own jurisdiction areas, (ii) allocating budget, supervising and monitoring the performance of district referral hospitals and health centers (HC)/health posts (HP), and (iii) providing in-service training to health care staff. Referral hospitals- at national, provincial, and district levels provide health care services including specialized services, diagnosis, follow-up, and treatment for the management of complex health problems. Health Center (HC) and Health Post (HP) province health care services at the district level and remote areas, respectively. Referral hospitals are located in populated areas within a two-hour drive, covering 80,000 to 200,000 population. HC is located within 10 kilometers, covering 8,000-12,000 population. HP is located in distant communes or villages, covering 2,000-3,000 population. Sources: MOH and World Bank staff 184 Cambodia Public Finance Review - From Spending More to Spending Better Box A.3.2. Summary of CDHS 2021 and government achievements Family planning • 62 percent of currently married women use a method of contraception. • The total demand for family planning among currently married women increased from 69 percent in 2014 to 74 percent in 2021–22; 61 percent of the total demand is satisfied by modern methods. • 42 percent of currently married women who are not using contraception intend to use family planning in the future. • Forty-six percent of women who currently use a modern method of contraception last obtained that method from a public sector source, primarily a government health center (37 percent), while 52 percent obtained their method from the private sector. Infant and child mortality • The infant mortality rate declined from 28 deaths per 1,000 live births in the 5 years before the 2014 survey to 12 deaths per 1,000 live births in the 5 years before the 2021–22 survey. Over the same period, the under-5 mortality rate declined from 35 deaths per 1,000 live births to 16 deaths per 1,000 live births. Maternal and newborn healthcare • Almost all women (99 percent) age 15–49 who had a live birth in the 2 years preceding the survey received antenatal care (ANC) from a skilled provider during their most recent birth. • Almost all women who received antenatal care for their most recent pregnancy had their blood pressure measured (99 percent), 88 percent had a blood sample taken, and 76 percent had a urine sample taken. • Nearly all births in the past 2 years were delivered in a health facility (98 percent). • 85 percent of mothers, 77 percent of newborns, and 74 percent of both mothers and newborns had a postnatal check during the first 2 days after delivery. Child health • 76 percent of children age 12–23 months had received all basic vaccinations by the time of the survey, and 65 percent were fully vaccinated according to national schedule. • 1 percent of children under age 5 had symptoms of acute respiratory infection (ARI) in the 2 weeks before the survey. Advice or treatment was sought for 92 percent of children with symptoms of ARI. • 6 percent of children under age 5 had diarrhea in the 2 weeks before the survey, and advice or treatment was sought for 69 percent of these children. Fifty-three percent of children with diarrhea received oral rehydration therapy (ORT), while 17 percent received no treatment. Nutrition of children and women • 22 percent of children under age 5 are stunted, 10 percent are wasted, 16 percent are underweight, and 4 percent are overweight. • Among children age 0–23 months, 54 percent were breastfeed within 1 hour of birth and 50 percent were exclusively breastfed for the first 2 days after birth; 50 percent of children under age 6 months are exclusively breastfed. • 49 percent of children age 6–23 months received meals with the minimum recommended diversity the previous day, 82 percent received meals at the minimum frequency, 42 percent were fed a minimum acceptable diet, 28 percent were given sweet beverages, and 21 percent were given unhealthy foods. Cambodia Public Finance Review - From Spending More to Spending Better 185 Box A.3.2. Summary of CDHS 2021 and government achievements (Cont.) • During the 12 months before the survey, 15 percent of children age 6–59 months were given iron- containing supplements. During the 6 months before the survey, 40 percent of children age 6–59 months were given vitamin A supplements and 47 percent of children age 12–59 months were given deworming medication. • 4 percent of women age 20– 49 and 14 percent of young women age 15–19 are of short stature. In addition, 7 percent of women age 20–49 and 29 percent of women age 15–19 are thin. The prevalence of overweight or obesity is 33 percent among women age 20– 49 and 6 percent among women age 15–19. • 57 percent of women consumed food from at least five of 10 specified food groups (minimum dietary diversity) the previous day, 63 percent consumed sweet beverages, and 33 percent consumed unhealthy foods. • 49 percent of households have iodized salt. Knowledge, attitudes, and behavior related to HIV AND AIDS • About one quarter of women and men (26 percent each) age 15–49 express discriminatory attitudes towards people living with HIV. • 88 percent of women and 64 percent of men age 15–49 have heard of PrEP approve of using it to prevent HIV infection. • Less than 1 percent of women and 4 percent of men reported having sex with a person who neither was their partner nor lived with them in the 12 months preceding the survey. • 47 percent of women and 76 percent of men reported using a condom during their last sexual intercourse with a nonmarital or noncohabiting partner. • 47 percent of women and 28 percent of men age 15–49 have ever been tested for HIV and received the test results. Disability • 21 percent of household members age 5 or above have some level of difficulty in at least one functional domain, and 4 percent have a lot of difficulty or cannot function at all in at least one domain. • The proportion of household members who have some difficulty in at least one domain generally rises with increasing age. For instance, less than 2 percent of household members in the age groups below 40 years have a lot of difficulty or cannot function at all in at least one domain, as compared with 20 percent of those age 60 and above. • 20 percent each of women and men who are widowed have a lot of difficulty or cannot function at all in at least one domain compared to 6 percent or less among women and 10 percent or less among men in other marital status categories. • 7 percent of rural women and 6 percent of rural men have a lot of difficulty or cannot function at all in at least one domain compared to 4 percent of urban women and 3 percent of urban men. • The percentages of men and women who have some difficulty or have a lot of difficulty or cannot function at all in at least one domain generally decline with increasing education and household wealth. Adult and maternal mortality • The adult mortality rate is 1.30 deaths per 1,000 population among women and 2.81 deaths per 1,000 population among men. • The lifetime risk of maternal death indicates that 1 in 250 women in Cambodia will die from maternal causes. • The maternal mortality ratio for the 7-year period before the 2021–22 CDHS is estimated at 154 maternal deaths per 100,000 live births. • The estimated pregnancy-related mortality ratio for the 7-year period preceding the 2021–22 CDHS is 163 deaths per 100,000 live births. Source: CDHS (2021/22) 186 Cambodia Public Finance Review - From Spending More to Spending Better Box A.3.3. Level of public health care services as regulated by MOH In 2006, the National Guidelines on Complementary Package of Activities for Referral Hospital Development from 2006 to 2010 stipulated the level of health care services provided by public health facilities. The first level of health care services is Minimum Packages of Activities (MPA) provided by health centers. The second level of health care is the Complimentary Package of Activities (CPA) provided by referral hospitals. The second level of health care consists of three CPAs (CPA-1, CPA-2, and CPA- 3). The guidelines set out detailed quality assurance criteria for referral hospitals to comply with to be entitled as CPA referral hospitals. MOH consistently conducts assessments on referral hospitals against the quality assurance requirements. MPA CPA-1 CPA-2 CPA-3 Description of MPA includes A referral hospital A CPA-2 referral A CPA-3 referral services preventive and is required to hospital provides hospital has the basic curative provide at least CPA-1 services highest level of services provided one of the three with 60- 100 beds health services, by HC/HP, CPAs-CPA-1, and it is required with 100-250 including neonatal, CPA-2, CPA-3. to provide beds. In addition child health CPA-1 referral emergency care to CPA-2 health services, maternal hospital is services and services, a health services, required to have grand surgery CPA-3 referral prevention and 40-60 beds and (with general hospital provides treatment of provide basic anesthesia). specialized health communicable health care services and a diseases, services without blood bank. treatment of non- major surgery or communicable blood bank. diseases, etc. National referral 0 0 0 9 hospital Provincial referral 0 0 4 21 hospital District referral 0 60 35 0 hospital Health center/ 1,269 0 0 0 health post Sources: MOH, WHO, and World Bank staff Note: MOH=Ministry of Health; MPA = Minimum package of activities; CPA = Complementary package of activities, HC=Health center; HP=Health post Cambodia Public Finance Review - From Spending More to Spending Better 187 Table A.3.1. Distribution and characteristics of public health care providers Public health Total (2021) Characteristics and provisions of health care services facilities National hospital 12 • National hospitals, the majority located in Phnom Penh, provide high level tertiary services. All 12 national hospitals are delegated as Special Operating Agency (SOA). Largely autonomous (as delegated by the MOH), national hospitals employ the workforce, monitor and evaluate the performance of the hospital and services, and make decisions how to make use of user fees to provide incentives for the health care workforce. • National hospitals provide the highest level of health care services (specialized services) which include general hospital services, and specialized services such as pediatrics and tuberculosis. • Sources of funding: (i) budget allocated across expenditure categories at the national level, (ii) Service Delivery Grant (SDG), including fixed lumpsum grant and performance-based grant, (iii) user fees, (iv) NSSF (National Social Security Fund) payment, and (v) HEF reimbursement. • Accountability framework: national hospitals are accountable for delivering health care services under the supervision of the Directorate General of Health, MOH. Provincial 25 • Provide CPA at three levels of health services (CPA1, CPA-2, CPA- referral hospital 3) at the provincial level depending on the number of factors such as staff and number of beds. Twenty out of 25 provincial referral hospitals provided CPA-3 level services. • Only Kandal provincial referral hospital has a psychiatric unit. • Only 10 out of 25 provincial hospitals are SOAs. • Conduct assessments to comply with the CPA requirements. A referral hospital is required to comply with at least CPA-1 requirements. • Sources of funding: (i) budget allocated from provincial health department (PHD), (ii) Service Delivery Grant (SDG)- including fixed lump-sum grant and performance-based grant, (iii) user fees, (iv) NSSF payments and (v) Health Equity Fund (HEF) reimbursement. • Accountability framework: provincial referral hospitals are under the direct supervision of the Provincial Health Department (PHD). Performance metrics at the facility are prepared in consultation with PHD, which is responsible for conducting performance monitoring of the provincial referral hospital. 188 Cambodia Public Finance Review - From Spending More to Spending Better Public health Total (2021) Characteristics and provisions of health care services facilities District referral 95 • Provide CPA at two level services (CPA1, CPA-2) depending on the hospital number of criteria including the number of staff and beds at the district level. • Communicable diseases services are mainly provided by referral hospitals and health centers. Conduct assessments to comply with CPA requirements. • • Sources of funding: (i) budget allocated from the operational district (OD)), (ii) SDG, (iii) user fees, (iv) NSSF payment, and (v)HEF reimbursement. ccountability framework: District referral hospitals are under the • A supervision of OD. OD is responsible for monitoring the performance of district referral hospitals. OD is accountable vis-à-vis PHD. for the performance results of district referral hospitals and health centers under its supervision. Health center 1,288 • Working much closer at a community level, provides MPA including neonatal, child health services, maternal health services, prevention and treatment of communicable diseases, and treatment of non- communicable diseases. • Covering 10,000–20,000 people. • User fees covering drugs. • Conduct assessments to comply with the MPA requirements. • Sources of funding: (i)budget allocated from PHD), (ii) SDG, including fixed lump-sum grant and performance-based grant (iii) user fees, (iv) NSSF payments, and (v) HEF reimbursement. • Accountability framework: Health centers are under the supervision of OD. OD is responsible for monitoring the performance of health centers. Health Posts, which are located in remote areas, are under the health center. Source: MOH and World Bank staff Note: MOH=Ministry of Health; MPA = Minimum package of activities; CPA = Complementary package of activities; SOA=Special Operating Agencies; DG=Directorate General, HEF=Health Equity Fund MPA includes preventive and basic curative services. CPA has three levels (CPA-1, CPA-2, CPA-3). CPA-1 hospital is the lowest level of a hospital with 40-60 beds which provides basic health without major surgery or a blood bank. CPA-2 hospital has 60-100 beds providing CPA-1 services and additional services such as emergency care, major surgery, blood transfusion, etc. CPA-3 hospital has 100-250 beds, providing CPA-2 services plus major surgery, and other specialized services. Cambodia Public Finance Review - From Spending More to Spending Better 189 Table A.3.2. List of private health facilities and pharmaceutical and medical supply institutions in Cambodia (September 2023) No Type of Private Facilities and Institutions Total I. Private Health Facilities without In-patient Bed (Small-size facilities) 1 Antenatal care room 2,327 2 Nursing care room 6,728 3 Dental care room 373 4 Physiotherapy room 34 5 Medical consultation cabinet 4,462 6 Dental consultation cabinet 965 7 Ophthalmology consultation cabinet 53 8 ENT consultation cabinet 34 9 Dermatology consultation cabinet 56 10 Mental health consultation cabinet 22 Sub-total Small-size Facilities 15,054 II. Private Health Facilities with In-patient Beds (Large-size facilities) 1 Private hospital 23 2 Polyclinic 104 3 Medical clinic 749 4 Pediatric clinic 2 5 Maternity clinic 18 6 Dental clinic 71 7 Ophthalmology clinic 4 8 ENT clinic 0 9 Dermatology clinic 0 10 Mental health clinic 3 11 Laboratory 106 12 Aesthetic center 46 13 International Hospital Representative Office in Cambodia 5 Sub-total Large-size private facilities 1,131 TOTAL PRIVATE HEALTH FACILITIES (small & large size facilities) 16,185 III. Private Pharmacies 1 Pharmacy 3,147 2 Sub-pharmacy level "A" 357 3 Sub-pharmacy "B" 248 4 Khmer and Chinese Traditional Medicine Store 155 Sub-total Pharmacies (excluding traditional medicine store) 3,752 IV. Pharmaceutical and Medical Equipment and Supply Institutions 1 Pharmaceutical Import and Export company 605 2 Branch of Pharmaceutical Import and Export company 33 3 Pharmaceutical Manufacturer 15 4 Medical Equipment and Supply Manufacturer 17 5 Branch of Medical Equipment and Supply Manufacturer 1 6 Nutritious Product Manufacturer 1 7 Aesthetic Product Manufacturer 8 8 Aesthetic Product Import and Export company 1,356 9 Branch of Aesthetic Product Import and Export company 5 Pharmaceutical and Medical Equipment and Supply Institutions 2,041 TOTAL PHARMACIES AND PHARMACEUTICAL & MEDICAL EQUIPMENT SUPPLY INSTITUTIONS 5793 Source: MOH, Department of Hospital Service, Department of Drugs, Food and Cosmetics, as of August 2023 190 Cambodia Public Finance Review - From Spending More to Spending Better Table A.3.3. Policy and key health care reform priorities, implementation progress, and challenges ahead Key priorities reforms and Key implementation progress Challenges ahead strategies Health Equity Established in 2000, HEF mobilized funds from the government and PFM challenges: Fund development partners to provide health care services and financial • Increase utilization protection for the poor (around 3 million population). HEF covered 100 rate and coverage. percent user fees related to hospitalization, and health care services, transportation and food costs at referral hospitals and health centers. The beneficiaries covered under HEF included poor as identified by ID Poor136 system and informal sector workers.137 HEF covered all public health facilities in 2015. Between 2016 to 2020, total HEF cost was estimated at US$70 million. Of the total HEF cost, 57.14 percent (US$402 million) was financed by the government and the rest 42.85 percent (US$30 million) financed by pooled donor fund.138 In 2021, HEF cost reached US$18.81 million, of which 68 percent was covered by MOH budget. 139 Health system Since 2008, MOH converted a number of health Operational District PFM challenges reform to (OD) and Provincial Hospitals into Special Operating Agencies (SOAs). • Monitoring strengthen 12 national hospitals, 11 provincial hospitals, 38 ODs, and 78 referral performance of SOA capacity of hospitals, were converted into SOAs. health facilities was public health SOA was granted with a level of management autonomy to make use of challenging due to facilities to resources-human and financial- to deliver results. SOA has authority to limited resources deliver quality hire additional workforce, provide incentives for them, and monitor their for conducting health services performance. In addition to MOH budget allocation and user fees, SOA performance received Service Delivery Grant (SDG)-a government budget for health monitoring on regular sector support. SOA budget was channeled through PHD and OD. basis. Service Delivery Grant (SDG) consisted of fixed lumpsum grant and • Although there were performance-based grant. In 2021, performance-based grant was incentives for health US$11.36 million (50 percent was covered by MOH), while fixed lumpsum workforce at SOA grant amounted to US$0.3 million. Fixed lumpsum grant can be used only health facilities, the for covering operating costs at health facilities (MOH and MEF). incentives were SOA was an internal contracting in which SOA facilities singed service not enough to delivery management contract and was accountable for delivering encourage facility results agreed upon in the performance metrics. At national level, staff to conduct dual national hospitals signed service delivery contract with MOH; and health care practices- MOH monitored their performances accordingly. At provincial level, both at public MOH singed service delivery contract with PHD; PHD, in turn, singed health facility and at service delivery contract with SOA facilities-OD and provincial referral private health care hospital- within its own jurisdiction. PHD were responsible for monitoring providers.141 performance of OD and provincial referral hospitals. OD monitored district referral hospitals and health facilities on a regular basis. Anecdote evidence suggested SOA facilities had improved on health care staff behavior such as improved punctuality of staff, 24 hours of health services at health facilities.140 136 ID-Poor is a program implemented by the Ministry of Planning that pre-identifies poor household nation-wide. 137 MOH, guidelines for the HEF Promoter, 2016. 138 World Bank, Project Appraisal Document for Health Equity and Quality Improvement Project (H-EQIP), p. 33 139 ID-Poor is a program implemented by the Ministry of Planning that pre-identifies poor household nation-wide. 140 Internal contracting of health services in Cambodia: drivers for change and lessons learned after a decade of external contracting, Vong et al. BMC Health Services Research (2018) ReBuild Consortium, understanding contracting in Cambodia: findings from interviews with key informants and health service 141 managers and providers (2015) Cambodia Public Finance Review - From Spending More to Spending Better 191 Key priorities reforms and Key implementation progress Challenges ahead strategies Moving towards The Strategic Framework for Health Financing 2008-2015 aimed • Increase health universal coverage at achieving universal health coverage by improving health system coverage through financing. The framework introduced National Social Health Protection integration of Schemes to cover informal sector social protection, including HEF, all social health CBHI, and maternal health voucher schemes. protection. CBHI’s coverage was low in both the rural and urban areas. Urban households had as low chance as rural households in receiving free and subsidized care from CBHI (1.3 percent vs 1.5 percent). Households in lowest quintile were more likely to receive free and subsidized care from CBHI (2 percent for quintile 1 versus 0.6 percent for quintile 5). Established in 2008, The National Social Security Fund (NSSF) is designed social protection schemes providing health insurance coverage for formal private sector employee. The scheme covered 1.3 million workers in the formal private sector, by 2019. The medical benefits under NSSF included all services available in the public hospitals, expect expensive procedures, malaria, HIV/AIDS, dental acre, plastic surgery and laser surgery etc. Sources: MOH, WHO, and World Bank staff Note: MOH=Ministry of Health; WHO=World Health Organization; HEF=Health Equity Fund; SOA=Special Operating Agencies; OD=Operational District; CBHI= Community-based Health Insurance; H-EQIP=Health Equity and Quality Improvement Project, SDG=Service Delivery Grant; M&E= Monitoring and Evaluation 192 Cambodia Public Finance Review - From Spending More to Spending Better Table A.3.4. Distribution of public health facilities by province and OPD cases (2022) Total National Provincial District Total OPD at Health OPD at health Provinces health Total OPD hospitals hospitals hospitals hospitals hospital center center facilities Banteay Meanchey 0 1 8 9 81,042 68 730,005 77 811,047 Battambang 0 1 6 7 178,445 82 682,328 89 860,773 Kampong Cham 0 1 8 9 63,668 91 551,101 100 614,769 Kampong Chhnang 0 1 2 3 35,251 45 238,075 48 273,326 Kampong Speu 0 1 3 4 69,031 58 331,217 62 400,248 Kampong Thom 0 1 2 3 17,639 56 357,432 59 375,071 Kampot 0 1 4 5 34,605 64 239,812 69 274,417 Kandal 0 1 10 11 124,742 106 675,563 117 800,305 Kep 0 1 1 8,434 5 14,049 6 22,483 Koh Kong 0 1 1 2 28,855 13 42,930 15 71,785 Kratie 0 1 4 5 26,175 40 85,072 45 111,247 Mondul Kiri 0 1 1 2 3,195 16 51,665 18 54,860 Oddar Meanchey 0 1 1 2 24,250 38 126,147 40 150,397 Pailin 0 1 1 12,052 7 13,890 8 25,942 Phnom Penh 0 1 8 9 88,357 43 385,367 52 473,724 Preah Sihanouk 0 1 1 19,934 17 43,578 18 63,512 Preah Vihear 0 1 1 2 37,713 30 142,983 32 180,696 Prey Veng 0 1 11 12 74,172 113 899,019 125 973,191 Pursat 0 1 3 4 39,998 47 366,990 51 406,988 Ratanakiri 0 1 1 2 16,907 30 126,807 32 143,714 Siemreap 0 1 4 5 167,602 93 743,018 98 910,620 Stung Treng 0 1 1 6,889 19 78,162 20 85,051 Svay Rieng 0 1 5 6 68,222 48 238,283 54 306,505 Takeo 0 1 6 7 150,451 85 548,459 92 698,910 Tbong Khmum 0 1 6 7 51,790 74 520,439 81 572,229 National Hospital 12 12 1,642,488 1,642,488 Total in all provinces (excluding national 12 25 95 120 1,429,419 1,288 8,232,391 1,408 9,661,810 hospitals) Sources: MOH and World Bank staff Table A.3.5. MOH budget outturn 2017 2018 2019 2020 2021 Execution Execution Execution Execution Execution BL Actual BL Actual BL Actual BL Actual BL Actual rate rate rate rate rate Chapter 60: 201,445 196,620 98% 223,416 218,093 98% 209,935 205,469 98% 231,833 223,487 96% 221,623 213,673 96% Purchases Chapter 61: 78,531 69,132 88% 82,070 73,346 89% 89,625 82,015 92% 91,963 76,737 83% 79,973 63,485 79% Services Chapter 64: 401,277 396,566 99% 466,148 444,777 95% 514,316 486,842 95% 565,189 528,035 93% 562,955 522,135 93% Payroll expenses Chapter 62: Social 469,459 467,326 100% 574,444 595,819 104% 675,727 669,666 99% 691,810 620,223 90% 801,031 637,946 80% benefits expenses Chapter 65: 50,609 47,555 94% 47,347 47,185 100% 55,402 54,599 99% 55,180 175,828 319% 54,589 53,331 98% Subsidy expenses Chapter 63: Taxes 534 219 41% 549 254 46% 520 257 49% 523 246 47% 513 277 54% and tariffs Sources: MEF and World Bank staff estimates Cambodia Public Finance Review - From Spending More to Spending Better 193 Annex to Chapter 4 Annex 4.1. Budget Strategic Plan The introduction of the Budget Strategic Plan (BSP) and the piloting of Program Budgeting (PB) are important steps forward under the Public Financial Management Reform Program (PFMRP). MAFF was one of the first ministries to pilot program-based budgeting as part of RGC’s reform roadmap. It thus provides valuable experience to Cambodia’s reformers as they seek to build a foundation to support the next stage of reforms aimed at strengthening accountability for results. This section reviews the MAFF experience with the BSP and PB in the context of the reform aims. The analysis focuses on strategic guidance, planning and budgeting, performance framework (assessment of the quality of output and outcome indicators), and implementation. The Royal Government of Cambodia (RGC) developed and launched a Public Financial Management Reform Program (PFMRP) with a detailed and sequenced action plan in 2004. The PFMRP is a sector- wide tiered (multi-stage or multi-platform) approach that aims to instill high standards of management and accountability in the mobilization of resources and ensure their effective and efficient use. Specifically, the sequenced multi-step reform process aims to strengthen budget credibility, improve financial accountability, establish budget-policy linkages, and improve performance accountability. One of the objectives at the current stage (platform 3) of the reform is to strengthen the foundation towards adopting performance-informed budgeting (PIB),142 with program budgets as a key component. The Budget System Reform Strategy (BSRS 2018‐2025) frames the implementation of program budgeting as a tool for linking budget to policy objectives and, by extension, the measuring of performance. Piloting of program budgeting was introduced to 10 LMs in 2015 and expanded to 36 LMs/institutions in 2017, with the aim to gradually strengthen accountability and the availability of performance and outcome information en route to the planned implementation of full performance budgeting from 2025 onwards. Reflecting the Government’s modular and sequential reform approach, each reform stage is supported by ministry-level pilots to gain valuable operational experience and strengthen the practical understanding of the challenges in the implementation of reforms that could take place across the whole of government. The introduction of the Budget Strategic Plan (BSP) and the piloting of Program Budgeting within MAFF is one such example to help guide the implementation of Stage 3 of the PFMRP. BSP and PB attempt to connect Line Ministry expenditure allocations to policy priorities and outcomes across a three-year horizon. The preparation of the BSP precedes the annual budget preparation and provides the background information and parameters to shape the basis for budget allocations. A joint review of the reform efforts conducted by the PFMRP Steering Committee and the WB noted that the introduction of the BSP was a key step forward and contributor to positive progress along Cambodia’s reform journey and specifically in promoting the focus on performance. The MAFF case study demonstrates that program structures and BSP have helped strengthen elements of the planning process, but are yet to impact budgeting and budget management systems. Deepening of PB reforms will require the integration of program logic to budget management processes that go beyond planning and are guided by strategic sequencing choices and targeted functional reforms in prioritized bottlenecks. 142 Performance-informed budgeting could be seen through the lens articulatesd in the Budget System Reform Strategy (BSRS), with the obejctive for to have the discussion on budget proposal, both current and capital budget, to be principally be based on program performance. 194 Cambodia Public Finance Review - From Spending More to Spending Better International experiences suggested that PB implementation has faltered mainly due to the difficulty of combining the objectives of planning, management, and control in one budget system. In some countries, instead of providing more information to the government to manage its resources, program budgeting, as implemented, may have resulted in the implementation of more, rather than less, controls and a fragmentation of the budget, which is exactly counter to the aspirations of program budgeting.143 Annex 4.2. Program budget structure Table A.4.1. MAFF program structure across administrative units MAFF entity/department Program 1) General Directorate of Agriculture Program 1: Increasing productivity, 2) General Department of Rubber diversifying agricultural crops, and agribusinesses 3) Department of Agro-Industry 4) Cambodian Agricultural Research and Development Institute 5) Cambodian Rubber Research Institute 6) Provincial Departments of Agriculture, Forestry, and Fisheries 1) General Department of Animal Health and Production Program 2: Promoting animal 2) Provincial Departments of Agriculture, Forestry, and Fisheries production and animal health 1) Fishery Administration Program 3: Fisheries management and 2) Provincial Departments of Agriculture, Forestry, and Fisheries aquaculture development 1) Forestry Administration Program 4: Management and 2) Provincial Departments of Agriculture, Forestry, and Fisheries development of forest and wildlife resources 1) Department of Planning and Statistics Program 5: Increasing effectiveness 2) Department of Finance and Accounting of supporting services and human resource development 3) Department of Personnel and Human Resource Development 4) Department of Internal Audit 5) Department of Internal Cooperation 6) Department of Agricultural Legislation 7) Department of Administrative Affairs 8) Agricultural Documentation and Information Center 9) General Inspectorate 10) Royal University of Agriculture 11) Prek Leap National Institute of Agriculture 12) Kampong Cham National Institute of Agriculture 143 Brumby,James A.; Hashim,Ali; Piattifuenfkirchen,Moritz Otto Maria Alfons. 2022. Introducing the New PPB: Pragmatic Program Budgeting: Overcoming Design Obstacles to Planning, Management, and Control (English). Equitable Growth, Finance and Institutions Insight Washington, D.C.: World Bank Group. Cambodia Public Finance Review - From Spending More to Spending Better 195 Annex to Chapter 5 196 Table A.5.1. Numbers of irrigation schemes and irrigated areas Cambodia Public Finance Review - From Spending More to Spending Better The World Bank Cambodia Country Office Exchange Square Building, No. 19-20, Street 106, Sangkat Wat Phnom, Khan Daun Penh, Phnom Penh Tel: +855 23 261300 Fax: +855 23 261301-2 cambodia@worldbank.org facebook.com/WorldBankCambodia 198 Cambodia Public Finance Review - From Spending More to Spending Better