Report No: ACS18454 . West Bank and Gaza Public Expenditure Review of the Palestinian Authority Towards Enhanced Public Finance Management and Improved Fiscal Sustainability September 2016 . GMF05 MIDDLE EAST AND NORTH AFRICA . . Standard Disclaimer: . This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. . Copyright Statement: . The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ii CONTENTS Acknowledgments ...................................................................................................................................... ix Acronyms and Abbreviations ................................................................................................................... xi Executive Summary ................................................................................................................................. xiii Chapter 1: Recent Macroeconomic and Fiscal Trends and Key Policy Challenges ............................. 1 1.1 Recent macroeconomic trends, growth drivers, and prospects .............................................. 1 1.2 Broad fiscal developments, 2006-2014 ................................................................................. 4 1.3 Key fiscal policy issues and challenges facing the PA .......................................................... 5 1.4 Policy ingredients for successful deficit reduction ................................................................ 6 Chapter 2: Analysis of the Wage Bill ...................................................................................................... 10 2.1 The large wage bill, its impact and trends ........................................................................... 10 2.2 Wage bill growth in historical perspective .......................................................................... 12 2.3 Main issues with the size and structure of public sector employment ................................. 13 2.4 Institutional contributors to wage bill growth ..................................................................... 21 2.5 Main issues with pay practices in the Palestinian public sector .......................................... 21 2.6 Summary of findings and ingredients for controlling the PA’s wage bill ........................... 28 Chapter 3: Health Expenditures .............................................................................................................. 32 3.1 Health Sector Context.......................................................................................................... 32 3.2 Health Expenditures ............................................................................................................ 32 3.2.1 Sources of Funding ......................................................................................................... 34 3.2.2 Salaries ............................................................................................................................ 38 3.2.3 Outside Medical Referrals ............................................................................................... 38 3.2.4 Pharmaceutical Expenditures .......................................................................................... 42 3.2.5 Arrears ............................................................................................................................. 44 3.3 Government Health Insurance ............................................................................................. 44 3.4 Health System Inputs ........................................................................................................... 47 3.4.1 Infrastructure ................................................................................................................... 47 3.4.2 Health Workforce ................................................................................................................ 51 3.5 Health Services Utilization .................................................................................................. 56 3.5.1 Utilization of Primary Health Care Services ................................................................... 56 3.5.2 Utilization of Secondary and Tertiary Care Services ...................................................... 57 3.5.3 Efficiency of Service Delivery ........................................................................................ 58 3.6 Discussion & Recommendations ......................................................................................... 61 3.6.1 Financing ............................................................................................................................. 62 3.6.2 Governance.......................................................................................................................... 63 3.6.3 Service provision ................................................................................................................. 64 3.6.4 Resource generation ............................................................................................................ 65 Chapter 4: The Palestinian Public Pension System ............................................................................... 66 4.1 An overview of the current pension system and recent reform efforts ................................ 66 4.2 Implications of maintaining the status quo .......................................................................... 76 4.3 Reform options and their expected impact .......................................................................... 78 4.4 Potential cost savings from implementing the 2011 Pension Reform Action Plan ............. 82 4.5 Conclusions and recommendations ..................................................................................... 83 iii Chapter 5: Intergovernmental Fiscal Relations ..................................................................................... 84 5.1 Introduction ......................................................................................................................... 84 5.2 Functional and expenditure assignments of municipalities and VCs .................................. 87 5.3 Key revenue assignment issues ........................................................................................... 91 5.4 The intergovernmental transfer system ............................................................................... 95 5.5 The net lending issue ........................................................................................................... 98 5.6 Significant budgeting issues .............................................................................................. 103 5.7 Conclusions and recommendations for the short and medium term .................................. 110 5.7.1 Disentangle and solve the net lending issue ...................................................................... 110 5.7.2 Address the problem of fragmentation and suboptimal scale of VCs ............................... 112 5.7.3 Strengthen the pillars of fiscal decentralization ................................................................ 112 Chapter 6: Public Investment Management ......................................................................................... 116 6.1 Introduction ....................................................................................................................... 116 6.2 What is public investment?................................................................................................ 116 6.2.1 Definition of public investment ......................................................................................... 116 6.2.2 PIM and the budget process .............................................................................................. 116 6.2.3 Key requirements of a PIM system ................................................................................... 117 6.2.4 Public investment and development expenditure .............................................................. 117 6.3 Public investment in the West Bank and Gaza – recent trends and performance.............. 118 6.3.1 Comprehensiveness of data on public investment............................................................. 118 6.3.2 Aid-financed public investment ........................................................................................ 118 6.3.3 On-budget public investment – trends and composition ................................................... 119 6.4 Public investment in the West Bank and Gaza – legislative and institutional framework 123 6.4.1 Legal and regulatory framework for PIM ......................................................................... 123 6.4.2 Institutional mandates for PIM .......................................................................................... 125 6.5 PIM at local government level........................................................................................... 129 6.5.1 Public investment by LGUs .............................................................................................. 129 6.5.2 Local governments ............................................................................................................ 130 6.5.3 Oversight and consolidation of public investment in LGUs ............................................. 132 6.6 Putting in place the key features and elements of a PIM system ....................................... 133 6.6.1 Feature 1 -- Strategic priorities, project development, and preliminary screening ........... 133 6.6.2 Feature 2 -- Project preparation and appraisal................................................................... 135 6.6.3 Feature 3 -- Independent review of appraisal .................................................................... 138 6.6.4 Feature 4 -- Project selection and inclusion in the budget................................................. 138 6.6.5 Feature 5 -- Project implementation .................................................................................. 139 6.6.6 Feature 6 -- Project monitoring and adjustment ................................................................ 141 6.6.7 Feature 7 -- Facility handover and operation .................................................................... 141 6.6.8 Feature 8 -- Basic completion review and evaluation ....................................................... 142 6.7 Priority next steps .............................................................................................................. 143 References ................................................................................................................................................ 146 Annex I: Health Status ........................................................................................................................... 150 Annex II: Health Expenditure ............................................................................................................... 154 Annex III: Workforce and Infrastructure ............................................................................................ 168 Annex IV: Basic Concepts of Pension Systems .................................................................................... 170 Annex V: Notional Defined Contribution Plans as a Pension Reform Strategy ............................... 173 Annex VI: Financial Sustainability of a Pension System .................................................................... 174 iv Annex VII: Survivorship Pensions ........................................................................................................ 176 Annex VIII: Key Features of a PIM System – Existing Situation and Recommendations .............. 177 Boxes Box 1: The PA’s Salary Scale ..................................................................................................................... 22 Box 2: History of Reforms in the Palestinian territories ............................................................................. 76 Box 3: Different Approaches Have Been Successfully Tried to Address the Issue of Excessive Local Government Fragmentation ........................................................................................................................ 87 Box 4: Methodology Used for Arriving at Payments Made by Each Municipality in 2012 and Total Municipality Arrears as of December 31, 2012 ........................................................................................ 101 Box 5: Nablus FY 2010 Budget Execution performance--The Case of an Outlier................................... 110 Box 6: Organic Budget Law (OBL) Provisions Relating to Public Investment ....................................... 123 Box 7: Key Elements of a Legal Framework for Procurement ................................................................. 125 Box 8: Capabilities in the Construction Sector ......................................................................................... 127 Box 9: Oversight Responsibilities for Key Stages in Capital Project Cycle ............................................. 128 Box 10: Deir Al-Hatab Health Centre....................................................................................................... 130 Box 11: MDLF - Capital Investment Cycle Management ........................................................................ 131 Box 12: Typical Format for a Project Concept Paper (PCP)# ................................................................... 135 Box 13: Prefeasibility and Feasibility Studies .......................................................................................... 137 Box 14: Capital Investment Project Annex to the Budget ........................................................................ 139 Figures Figure 1: The PA’s wage bill as a percentage of GDP, 1995-2013 ........................................................... xvi Figure 2: Average public salary (as a multiple of GDP per capita) in the Palestinian territories compared to MENA and most other regions ................................................................................................................. xvii Figure 3: The impact of controls on PA hiring, wage growth containment and reduced over grading in the security sector plus and wage growth, plus strong economic growth ...................................................... xviii Figure 4: Total health expenditure as a percentage of GDP in 2012 ......................................................... xxi Figure 5: GHI expenditure 2000-2012 ..................................................................................................... xxiv Figure 6: Consumption, investment, net exports, and GDP, Palestinian territories, 2006-2014 (annual growth rate) ................................................................................................................................................... 1 Figure 7: Ratio of exports and imports to GDP (L) &total value of exports and imports (R), 2006-2013 ... 2 Figure 8: Real GDP Growth for West Bank and for Gaza ............................................................................ 3 Figure 9: Growth in government expenditures and West Bank consumption versus government expenditures/consumption, 2006-2014 ......................................................................................................... 3 Figure 10: Electricity Purchases, Sales, Losses, Net Lending & Debt to Suppliers (in million NIS) .......... 9 Figure 11: General government wage bill/GDP, select countries ............................................................... 10 Figure 12: Salaries as a percent of PA’s expenditures and revenues .......................................................... 11 Figure 13: The PA’s wage bill as a percentage of GDP, 1995-2013 .......................................................... 12 Figure 14: Estimated general government employment as a percentage of total country population, select countries ...................................................................................................................................................... 14 Figure 15: Wage bill of security sector as a percent of GDP, select countries ........................................... 15 Figure 16: Number of security (non-police) personnel, select countries .................................................... 16 Figure 17: Distribution of Palestinian paramilitary forces by military rank, 2010 ..................................... 17 Figure 18: Number of staff in the education sector in the West Bank and Gaza, 2007-2013 ..................... 18 Figure 19: PA’s health sector staff, 2007-2013 .......................................................................................... 19 Figure 20: Healthcare professionals (per 10,000 inhabitants) .................................................................... 19 Figure 21: Increase in staffing (number of personnel) by ministry, 2007-2013 ......................................... 20 v Figure 22: Components of the PA’s public sector salary, 2013 .................................................................. 23 Figure 23: Number of temporary public sector personnel, 2006-2013 ....................................................... 24 Figure 24: Average public salary (as a multiple of GDP per capita) in the Palestinian territories compared to MENA and most other regions ............................................................................................................... 25 Figure 25: Wage gap in the West Bank (%) ............................................................................................... 25 Figure 26: Wage gap in Gaza (%) ............................................................................................................... 25 Figure 27: Wage gap at different quintiles in the West Bank ..................................................................... 26 Figure 28: Wage gap at different quintiles in Gaza .................................................................................... 26 Figure 29: Real daily public pay in the West Bank, 2000-2013 ................................................................. 27 Figure 30: Real daily public pay in Gaza, 2000-2013 ................................................................................ 27 Figure 31: Breakdown of PA civil servant employees by seniority............................................................ 27 Figure 32: Total health expenditure as percentage of GDP (%), International Comparisons, 2012 ........... 33 Figure 33: Public spending on health as a percentage of GDP (%), Palestinian territories 2000-2012 ...... 33 Figure 34: Composition of OOP expenditures, Palestinian territories, 2012 .............................................. 34 Figure 35: Source of health financing as share of total health expenditure, Palestinian territories, 2000-2012 .................................................................................................................................................................... 35 Figure 36: Approved MoH budget vs. actual MoH spending in US$ millions, Palestinian territories, 2010, 2011, 2012 and 2013 ................................................................................................................................... 36 Figure 37: Distribution MoH expenditures, Palestinian territories, 2013 ................................................... 37 Figure 38: Public expenditure on health by function of care, Palestinian Territories, 2000-2012 ............. 37 Figure 39: MoH wage bill, Palestinian territories, 2006-2012.................................................................... 38 Figure 40: Top 10 costliest referral procedures, Palestinian territories, 2013 ............................................ 40 Figure 41: Trends in medical referrals, Palestinian territories, 2000-2013 ................................................. 40 Figure 42: Cost and volume of referrals by service provider, 2013............................................................ 41 Figure 43: Utilization of referral services per insurance scheme, Palestinian territories, 2011 .................. 41 Figure 44: Drug tender cost, overall pharmaceutical spending and total health spending, Palestinian territories, 2010-2013 .................................................................................................................................. 42 Figure 45: Withdrawals from CMS by hospital and PHCC, 2013 .............................................................. 43 Figure 46: MoH arrears, Palestinian territories, 2013 ................................................................................. 44 Figure 47: GHI enrollees by type, West Bank, 2009-2013 ......................................................................... 45 Figure 48: GHI contributions by enrollee type, West Bank, 2012.............................................................. 46 Figure 49: GHI revenue and public health expenditure gap, Palestinian territories, 2009-2012 ................ 46 Figure 50: GHI expenditure, Palestinian territories, 2000-2012 ................................................................. 47 Figure 51: Geographic Distribution of Four Levels of PHCCs Operated by MoH, 2012 .......................... 48 Figure 52: Geographic distribution of all primary health care providers, Palestinian territories, 2012 ...... 49 Figure 53: Distribution of hospitals by provider type, Palestinian territories, 2012 ................................... 50 Figure 54: Distribution of hospital beds per 10,000 population, Palestinian territories, 2012 ................... 50 Figure 55: Trends in human resources for health, Palestinian territories, 2009-2012 ................................ 51 Figure 56: Human resources for health (per 10,000 population), International Comparisons, 2012 .......... 51 Figure 57: Human resources for health (per 10,000 population), West Bank vs. Gaza, 2009-2012 .......... 52 Figure 58: Geographic distribution of MoH employees, West Bank and Gaza, 2006-2013 ...................... 53 Figure 59: MoH human resources as share of the total health workforce (%), Palestinian territories, 2009- 2012 ............................................................................................................................................................ 53 Figure 60: Health personnel, Palestinian territories, 2012 .......................................................................... 54 Figure 61: Composition of MoH personnel, Palestinian territories, 2012 .................................................. 55 Figure 62: Unemployment rate among graduates, Palestinian territories, 2010 ......................................... 55 Figure 63: Number PHCC centers and number of PHCC visits per governorate, West Bank, 2012 ......... 56 Figure 64: Number of patients per physician in MoH hospitals, Palestinian territories, 2012 ................... 57 Figure 65: Inpatient and outpatient services in public hospitals (%), Palestinian territories, 2012 ............ 58 Figure 66: Number of visits per primary health care center by governorate, West Bank, 2012 ................. 59 Figure 67: ALOS at MoH hospitals, Palestinian territories, 2012 .............................................................. 59 vi Figure 68: Bed occupancy rate (%), Palestinian territories, 2012............................................................... 60 Figure 69: Bed occupancy in MoH hospitals (%), Palestinian territories, 2012 ......................................... 60 Figure 70: Pension expenditure and % of elderly population ..................................................................... 67 Figure 71: Pension expenditure and % of beneficiaries over total population............................................ 67 Figure 72: Current balance of the four pension schemes in the Palestinian territories (% of GDP), 2013- 2050 ............................................................................................................................................................ 77 Figure 73: Projected gross expenditures for Scheme II under several reform options ............................... 79 Figure 74: Current balance of Scheme II under several reform options ..................................................... 79 Figure 75: Fiscal decentralization in the Palestinian territories and selected countries and world regions, 2012 ............................................................................................................................................................ 85 Figure 76: Total revenue and expenditure as a share of GDP (%) by level of government, 2010-2013 .... 85 Figure 77: Population distribution in VCs (L) and municipalities (R) ....................................................... 86 Figure 78: Municipalities actual expenditure allocation by budget type, 2010-2012 ................................. 89 Figure 79: VC's actual expenditure allocation by budget type, 2011-2013 ................................................ 89 Figure 80: VC's operating budget expenditure ........................................................................................... 90 Figure 81: VC's operating budget expenditure categories' share in total operating expenditure, actual 2011- 2013 ............................................................................................................................................................ 90 Figure 82: Municipal Revenues by Type .................................................................................................... 92 Figure 83: Distribution of total expenditure per capita in VCs (L) and municipalities (R) ........................ 97 Figure 84: Total revenue per capita for VCs with and without electricity revenues, actual 2011-2013 ..... 99 Figure 85: Total revenue per capita of municipalities with and without electricity revenues, actual 2010- 2012 ............................................................................................................................................................ 99 Figure 86: Total operating and development expenditure per capita for VCs with and without electricity revenues, actual 2011-2013 ...................................................................................................................... 100 Figure 87: Total operating and development expenditure per capita of municipalities with and without electricity revenues, actual 2010-2012...................................................................................................... 100 Figure 88. Net revenue position from electricity distribution for municipalities providing electricity services .................................................................................................................................................................. 102 Figure 89: VCs' operating budget per capita averages, approved 2011-2013;.......................................... 104 Figure 90: Municipalities' operating budget per capita average, approved and actual 2010-2012 ........... 105 Figure 91: VCs' enterprise fund per capita averages (approved 2011-2013; 3/4 of approved 2011-2013; and actual 2011-2013) ..................................................................................................................................... 105 Figure 92: Municipalities' enterprise fund per capita average (approved and actual 2010-2012) ............ 106 Figure 93: VCs' development budget per capita averages (approved 2011-2013; 3/4 of approved 2011-2013; and actual 2011-2013)............................................................................................................................... 106 Figure 94: Municipalities' development budget per capita average (approved and actual 2010-2012) .... 107 Tables Table 1: The Fiscal Operations of the PA, 2006-2014............................................................................... xiv Table 2: MoH budget expenditure 2010-2013 ......................................................................................... xxiii Table 3: Fiscal operations of the PA ............................................................................................................. 5 Table 4: General government revenues/GDP, selected regions .................................................................... 7 Table 5: West Bank-based personnel increases in organizations that contributed the most to employment growth between 2006 and 2013 .................................................................................................................. 20 Table 6: Potential impact on the PA’s wage bill of various policy recommendations ............................... 31 Table 7: Trends in health financing, Palestinian territories, 2000-2012 (millions US$) ............................ 32 Table 8: MoH expenditure, Palestinian territories, 2010-2013 ................................................................... 36 Table 9: Discrepancies in salary expenditure reporting, Palestinian territories, 2010-2013 ....................... 38 Table 10: Pharmaceutical Withdrawals from CMS, 2013 .......................................................................... 43 vii Table 11: Comparing drug tender costs with international benchmark prices ............................................ 43 Table 12: Summary of key characteristics of the four pension schemes operating in the West Bank and Gaza ............................................................................................................................................................ 68 Table 13: Basic pension aggregated indicators for civil servants, security personnel, and nongovernmental employees in the Palestinian territories, December 2013 ........................................................................... 69 Table 14: Contributors, beneficiaries, wages, and pensions, December 2013 ............................................ 71 Table 15: Contributors, beneficiaries, wages, and pensions, September 2014 ........................................... 72 Table 16: Annual benefit amounts by scheme (NIS), 2013 ........................................................................ 73 Table 17: Mandatory contributions received by PPA (from the MoF), June 2009 versus April 2013 ....... 74 Table 18: Collected contributions and transfers from PA/MoF to PPA, December 2013 .......................... 75 Table 19: Actual revenues, expenditures, and balance of each pension scheme, December 2013 ............. 77 Table 20: Hypothetical revenues, expenditures, and balance of each scheme assuming consistent transfers by the PA as required by law, December 2013 ........................................................................................... 77 Table 21: Actuarially fair reduction coefficients (%) for early retirement for men .................................... 80 Table 22: Actuarially fair reduction coefficients for early retirement for women ...................................... 81 Table 23: Estimated public pension expenditures (NIS million) for each scenario, 2013-2050................. 82 Table 24: Revenues as % of GDP for different government units .............................................................. 91 Table 25: Revenue items’ share in overall revenue in VCs ........................................................................ 94 Table 26: Revenue items’ share in overall revenue in municipalities......................................................... 95 Table 27: VCs’ Availability of utility services, 2013 ................................................................................. 97 Table 28: Municipalities’ availability of utility services, 2013 .................................................................. 97 Table 29: VCs’ average budget balance by budget type ........................................................................... 108 Table 30: Municipalities’ average budget balance by budget type ........................................................... 109 Table 31: Official Development Assistance receipts, 2011-12 ................................................................. 119 Table 32: PA expenditures, 2009-2013 (commitment basis) .................................................................... 120 Table 33: Development expenditure by function (cash basis), 2011-2013 ............................................... 121 Table 34: Development expenditure by major economic group (commitment basis), 2011-2013 ........... 122 Table 35: Development expenditure implementation performance (commitment basis), 2011-2013 ...... 122 viii Acknowledgments This document has benefited from insights and inputs from various sources. The core PER team was led by Orhan Niksic (Senior Economist) and included Nur Nasser Eddin (Economist). The chapter on the PA’s wage bill was prepared and written by Orhan Niksic, Nur Nasser Eddin, and Michael Stevens (Consultant) while Sami Miaari (Consultant) assisted in conducting the relevant research and analysis. The health chapter was authored by Ece A. Özçelik and Emre Özaltın, and was furthered by the technical input received from Samira Hillis and Enis Barış. Support and comments received by Jumana Alaraf, Doruk Y. Kıroğlu Adrienne Larson and Janee Simms were also very beneficial to the Health chapter. The chapter on the Palestinian public pension system was drafted by Montserrat Pallares-Miralles and Nur Nasser Eddin. The chapter on intergovernmental fiscal relations was drafted by Jorge Martinez-Vazquez (Consultant), Orhan Niksic and Dana Almubaied, and received valuable technical inputs from Bjorn Philipp and Noriko Oe. The final chapter on Public Investment Management was authored by Andrew Bird (Consultant) with technical input from Pierre Messali. The report benefited considerably from overall guidance and comments provided by Steen Lau Jorgensen, the World Bank Country Director for the Palestinian territories and Auguste Kouame, Practice Manager of Macroeconomics and Fiscal Management for MENA. The report also benefitted from most valuable comments received by Clelia Rontoyanni, Gallina Vincelette, Rama Venkateswaran, Shantayanan Devarajan, Zeyad Abu Hassanein, and Massimiliano Cali. Furthermore, much gratitude for time and efforts to enhance the quality of this report is owed to Nigel Roberts, Mark Ahern, Bernard Funck, Ranjana Mukherjee, Radwan Shaban and Jaime de Pinies. The team would also like to thank Maha Bali for the administrative support provided throughout the preparation process. This report was prepared in close collaboration with, and based on invaluable information and insights shared by the Palestinian Authority. The task team is grateful to counterparts at the Ministry of Finance and Planning including H. E. Shukry Bishara (Minister), Laila Sbaih (Director General of International Relations and Projects), Farid Ghannam (Director General, Central Budget Unit), Ahmad Sabbah (Accountant general), Dana Erekat (Special Advisor to the Minister, Head of Aid Management and Coordination Directorate), Abdel Jabbar Salem (Head of Payroll Unit), Mahmoud Nofal (Director General, PropertyTax) and Yazan Ajamieh (Team leader, Macrofiscal Unit). The team is also grateful for the extensive efforts exerted by counterparts at the Ministry of Health. The team extends its gratitude to H.E. Dr. Jawad Awwad (Minister of Health) and Dr. Anan Al-Masri (Former Deputy Minister) for their support, vision and direction. Gratitude also goes to Asaad Ramlawi (Deputy Minister); Nizar Masalmeh (Director General, Insurance Department), Abdel Kareem Hamadneh (Director General, Finance Department), Sulimam Ahmad (Director, Administrative Affairs), Osama Najjar (Director, Referrals Unit), Mohammad Abu Ghali (Director General of Hospitals), Rania Shahin (Head of Pharmaceuticals Unit), Jihad AlBadawi (Inspector General), Kamal AlShakhra (Director, Non-Communicable Diseases Unit), and Maria AlAqra (Director, International Relations). The team would like to acknowledge insights shared by H.E. Hani Abdeen (Former Minister of Health) and Shawqi Sabha (Head of Doctors’ Union). Valuable insights were also provided by officials at the Ministry of Local Government including Shukri Radaydeh (Director General, Local Entities Budget Department) and Mahmoud Zubaidi (Manager, Local Entities Budget Department). The team also wishes to thank counterparts at the Municipal Development Lending Fund including Abed Almughni Nofal (General Director), Hazem Qawasmi (General Director of Operations), Mohammad Ramahi (Finance and Administration Manager), Nizar Samhan (Strategic and Planning Manager), Osama Nabahin (Financial Analyst), and Lina Jildeh (Financial Analyst). The team also expresses gratitude to counterparts at the General Personnel Council including H.E.Mousa Abu Zeid (Chairman), Mahmoud Shaheen (Vice Chairman), Jamal Abu Shanab (Director General, Information Technology), Wail Rimawi (Director General, Career Planning), and Nisreen Zghaiar ix (Advisor/International Relations). Appreciation also goes to officials at the Palestine Pensions Authority including Dr. Majed Helo (Director General), Mohammad Mohaisen (Head of the Investment Unit and General Director of the Services), Mohammad Arafat (General Director of Contributions), Ayman Aldoqi (Head of the Internal Control and Studies Units), and Salama Abusalim (Actuarial Specialist). The team is also obliged for the collaboration extended by the various officials at the Ministry of Public Works, the Ministry of Education, the Palestine Water Authority, the Palestine Land Authority, the municipalities of Hebron, Ramallah and Salfit, Kobar Village Council, and other institutions of the Palestinian Authority. Undoubtedly, the report could not have been produced without data and support provided by the Palestine Central Bureau of Statistics. Last but not least, the report benefited significantly from consultations with donor partners and non- governmental organizations. The team thanks representatives from the Department for International Development (DFID), the United States Agency for International Development (USAID) and the European Union. Special thanks go to Mahmoud Daher (Head of the WHO Gaza Sub Office) and his team for data acquisition in the Gaza Strip. The team also thanks representatives of the Juzoor for Health and Social Development and Health Policy Forum for their insights. x Acronyms and Abbreviations ALOS Average length of stay AMCD Aid Management and Coordination Directorate BD Budget Directorate BOR Bed occupancy rate CMS Computerized Management System CPI Consumer price index CoLA Cost of Living Adjustment DARP Development Assistance and Reform Platform DB Defined benefit DC Defined contribution DISCO Electricity distribution company DRU Dispute Review Unit EMRO East Mediterranean region ENT Ear, nose, and throat FDC Funded defined contribution GDP Gross domestic product GHI Government Health Insurance (scheme) GPC General Personnel Council GoI Government of Israel HCPPP Higher Council for Public Procurement Policy HR Human resources IDP International development partner IEC Israeli Electric Corporation IFMIS Integrated Financial Management Information System IMF International Monetary Fund IRPD International Relations and Projects Directorate JSC Joint Service Council LGA Local Government Act LGU Local Government Unit MDAs Ministries, Departments, and Agencies MDLF Municipal Development and Lending Fund MDP1 Municipal Development Project 1 MENA Middle East and North Africa MoE Ministry of Education MoF Ministry of Finance MoH Ministry of Health MoLG Ministry of Local Government MoPAD Ministry of Planning and Administrative Development MoPWH Ministry of Public Works and Housing NCD Noncommunicable disease NDC Notional defined contribution NIS New Israeli Sheqel NPISH Nonprofit institutions serving households OBL Organic Budget Law OECD Organisation of Economic Co-operation and Development OMR Outside medical referrals OOP Out of pocket PA Palestinian Authority xi PAYG Pay-as-you-go PBB Program-based budgeting PCBS Palestinian Central Bureau of Statistics PCI Police Crime Intervention PCP Project concept paper PEFA Public Expenditure and Financial Accountability PER Public Expenditure Review PHCC Primary health care center PIM Public investment management PIMU Public Investment Management Unit PIU Project implementation unit PMMS Palestinian Military Medical Services PNDP Palestinian National Development Plan PPA Palestinian Pension Authority PPL Public Procurement Law PRDP Palestinian Reform and Development Plan PROST Pension Reform Options Simulations Toolkit PT Palestinian territories UNRWA United Nations Relief and Works Agency UPL Unified Pension Law VC Village Council WHO World Health Organization xii Executive Summary 1. The difficulty of pursuing a conventional market-oriented development strategy in the Palestinian territories led in the early part of the 2000s to a “second-best” reliance on public sector employment and wage bill expansion to boost aggregate demand. This approach was underwritten by the international community, in response to the reduction in employment for Palestinians in Israel and the movement and access restrictions that intensified with the Second Intifada coupled with the donors' wish to help create the institutions and services appropriate for a future Palestinian state. The main focus was on aggregate fiscal stimulus rather than the quality of public services delivered by the PA – frequently resulting in poorly targeted and high cost interventions. 2. The main objective of this Programmatic Public Expenditure Review (PER) is to inform policy and institution-building efforts of the Palestinian Authority (PA) and its donor partners about improving the sustainability of public expenditures and the efficacy and efficiency in the provision of essential public services. In particular, this PER aims to provide an assessment of public revenue and expenditure policies offering specific policy and institutional measures to reduce the size of the Palestinian territories’ fiscal deficit and make it more sustainable. This will create fiscal space for public investments that could catalyze economic growth. The PER analyzes sector-specific expenditure policies and the adequacy of existing intergovernmental fiscal relations, and provides reform recommendations to improve efficacy, efficiency, and sustainability in the provision of certain public services.1 3. The fiscal situation of the Palestinian Authority is not sustainable. The Palestinian Authority’s (PA) budget deficit stood at over 10 percent of Palestinian GDP in 2014 (See Table 1). Although this represents a reduction from almost 25 percent of GDP in 2007, it has been financed by budget support and a progressive accumulation of domestic debt including arrears, the stock of which is now approaching US$3 billion: the majority is owed to the Palestinian public pension system, while the rest is to the private sector. Without a significant correction, the public pension system will become insolvent within five years, while private sector suppliers may refuse to supply important goods and services (such as pharmaceuticals) any longer: the PA's coping strategy is thus reaching its limits. 4. The difficult fiscal situation facing the Palestinian Authority today results from a unique confluence of challenges. First, Palestinian economic growth and PA revenue potential remain significantly depressed by movement and access restrictions. The inability of Palestinians to exploit economic opportunities in Area C, for example, has been estimated by the Bank as reducing potential Palestinian GDP by up to US$3.5 billion per year, and potential revenue by up to US$800 million a year. Low tax effectiveness in Gaza and lower than hoped for compliance in the West Bank also affect the PA’s revenue potential. Furthermore, donor budget support intended to help offset fiscal impairments is in decline. Between 2012 and 2014, budget support averaged just over US$1 billion per annum, equivalent to only 10 percent of Palestinian GDP in 2013 (See Table 1). At its height in 2008, donor contributions to the Palestinian budget equaled 26 percent of GDP. In 2014, they had shrunk to 8 percent of a significantly larger GDP (in nominal terms, budget support fell by 28 percent, from the above-noted average of US$1.42 billion per annum between 2008-10, to an annual average of US$1.02 billion between 2012-14). 1 Due to time and resource constraints, the report is not a comprehensive assessment of the PA’s public expenditures. The themes/sectors covered in-depth in the report were chosen based on the following criteria: (i) their importance to the objectives of the PER; and (ii) a knowledge gap in the sector/policy area. For example, education was not specifically covered in the report, but the chapter on the wage bill does assess staffing and wage issues in the education sector given its significance in wage bill issues. xiii 5. The commendable efforts made by the PA to reduce the relative size of the recurrent deficit have not managed to offset the combined effects of the above-mentioned phenomena. It is important to note that the rapid growth in the public workforce described in the World Bank's previous (2007) Public Expenditure Review was curtailed by determined PA policy action, and that the size of the wage bill relative to GDP has been cut from a startling 24 percent of GDP in 2007 to 16 percent in 2014. As shown by the 2013 PEFA (Public Expenditures and Financial Accountability Assessment), PA public financial management practices have also improved significantly in recent years. The recurrent deficit was brought back from almost 25 percent of GDP in 2007 to just over 10 percent by 2014; this was largely achieved by a reduction in the size of the wage bill and of net lending2 relative to GDP (the wage bill peaked at 24 percent in 2006 and has since been reduced to 16 percent of GDP, in part due to hiring control and wage growth restraint; net lending dropped from nearly 10 percent of GDP in 2007 to just over 2 percent in 2014). Nonetheless, arrears continue to accumulate (See Table 1). Table 1: The Fiscal Operations of the PA, 2006-2014 (In % of GDP) 2006 2007 2008 2009 2010 2011 2012 2013 2014 Total revenues 23.4 21.7 23.5 22.0 21.4 19.5 19.8 18.7 21.5 Tax revenues 4.5 3.7 4.1 4.1 5.1 4.6 4.6 4.8 4.7 Nontax revenues 3.2 2.2 4.3 3.9 3.0 2.4 2.4 2.1 2.1 Clearance revenues 15.7 16.3 16.8 15.2 14.1 13.6 12.8 11.8 14.7 Total expenditure 47.0 46.2 43.2 43.9 34.5 31.8 33.7 29.9 31.8 Wage expenditures to GDP 24.2 23.3 21.8 20.2 18.1 17.0 16.9 15.5 16.0 Nonwage 15.2 13.2 14.8 18.6 13.8 13.4 14.1 12.6 13.6 Net lending 7.7 9.7 6.7 5.1 2.6 1.3 2.7 1.7 2.2 Current balance (23.6) (24.5) (19.7) (21.9) (13.1) (12.2) (13.9) (11.2) (10.4) Development expenditure 5.7 5.6 3.2 3.0 3.4 3.5 2.3 1.5 2.1 External financing 20.8 24.0 29.6 19.3 14.3 9.4 8.9 11.0 9.6 Budgetary support 15.0 18.4 26.4 18.6 12.9 7.8 7.4 10.1 8.0 Development financing 5.7 5.6 3.2 0.6 1.5 1.6 1.5 0.9 1.6 Net domestic bank borrowing (3.5) (2.4) (0.4) 2.4 0.9 0.9 0.3 (0.6) (0.3) Net accumulation of arrears (US$) 594 470 (416) 230 105 572 732 276 374 Source: PA MoF. 6. There is considerable further scope for reforms that would raise additional tax revenues, and reduce expenditures without compromising the quality of public services or negatively impacting public welfare. Doing so requires raising additional domestic revenue through improved tax enforcement (which will effectively increase the progressivity of the system), as well as by broadening the tax base. On the expenditure side, the review advises the PA to focus on a few key areas and emphasize the quality and value for money of expenditures. It first advocates reducing the size of the wage bill relative to GDP -- through wage restraint, grade restructuring and the gradual reduction of public sector staff in targeted categories (noting that astutely managed measures could help bring about improvements in performance). The PER further recommends measures to increase the cost-efficiency of basic services, restructure the 2 Net lending is a term that describes the deductions made by Israel from clearance revenues as a result of utility (mostly electricity) bills, which have not been paid by Palestinian municipalities and electricity distribution companies. These deductions from PA tax revenues amount to a de facto fiscal transfer to Palestinian municipal governments. xiv public pension system and cut back on wasteful energy subsidies. In all cases, reforms can both improve fiscal sustainability and potentially enhance the effectiveness of the service in question-- while freeing up resources for more equitable distribution elsewhere in the economy. 7. However, the PER notes that there are limits to what can be achieved by PA fiscal policy alone. Growth in the Palestinian territories has been unconventionally based, with investment playing a limited role in a process largely sustained by consumption and consumption-related import substitution3 -- enabled, in turn, by compensatory high rates of donor disbursements to the Palestinian budget. In fact, on the assumption that movement and access restrictions will continue in much the same form as currently, any credible PA reform program of the type advocated in this report must be met by a proportional donor response. 8. The PER is organized as follows: Chapter 1 provides an overview of recent macroeconomic and fiscal developments; it also contains a brief assessment of priority fiscal policy issues facing the PA, and serves as an introduction to the in-depth analysis of the issues that follow in subsequent chapters. Chapter 2 analyzes the factors driving the size of the PA’s wage bill, and shows how these can be tackled. Chapter 3 reviews expenditures in the public health sector. Chapter 4 analyzes the Palestinian public pension system, and looks into how its sustainability can be assured. Chapter 5 assesses the quality of intergovernmental fiscal transfers, including net lending transfers. Chapter 6 reviews the way in which public investment projects are planned and implemented, and identifies steps to improve investment quality. Further details on health and pensions are provided in the annexes. The Central Government Wage Bill 9. In relative terms, the Palestinian public sector wage bill is among the highest in the world. For the majority of countries, public sector wage bills do not exceed 10 percent of GDP; in the Palestinian territories they currently amount to 17 percent of Palestinian GDP4 (See Figure 1). Notably, these figures do not include the cost of civil servants and security staff employed by the Hamas government in Gaza, the amount of which is not known but is estimated at an additional 3-4 percent of GDP.5 The central government wage bill alone is equivalent to 16 percent of GDP, 55 percent of recurrent expenditures, and as much as 83 percent of public revenues. 3 The report notes that data quality issues make it difficult to be definitive about the exact nature of net export growth, which was positive for six of the eight years 2006-2013. It would appear that this growth, which contributed almost a half of GDP growth in the period, was driven by a significant reduction in reliance on imports to fuel growth. The share of imports to GDP fell 19 percentage points over the period. 4 The PA is currently revising its national accounts, and it is likely that estimates of the wage bill/GDP ratio will change. This is unlikely to result in a substantially lower ratio, however. 5 Available data indicates that in Gaza, Hamas employs 50,112 civil servants. This includes blue uniform security personnel (i.e., police); figures on the number of green uniform security employees (i.e., security services) are unavailable. The addition of known Hamas staff would raise the cost of the wage bill to at least 18 percent of GDP. xv Figure 1: The PA’s wage bill as a percentage of GDP, 1995-2013 2,500 30% 25% 2,000 20% 1,500 15% 1,000 10% 500 5% - 0% Wage bill (in millions of USD) Wage bill to GDP Sources: MoF and PCBS data. 10. The main contributor to the high wage bill/GDP ratio is the high level of wages paid to central government staff. In fact, contrary to what has often been claimed, the size of the Palestinian civil service is not excessive by international standards: even if Hamas employees are included, central government employees would still total less than five percent of the Palestinian population. This puts the Palestinian territories significantly below many nations, and close to the average for middle income countries (5.3 percent).6 On the other hand, wages are high relative to GDP per capita. Average PA wages amount to 3.5 times GDP per capita, significantly higher than in any region except Africa7 (See Figure 2). In addition to any PA assumptions about future levels of donor budget support, three factors have contributed to this outcome. First, the PA implemented two large salary increases, in 2004 (average pay increased by 8.4 percent) and in 2006 (by 12.3 percent). Second, PA staff have benefited from annual salary increases that have not always been in line with inflation, particularly in Gaza, and that have not taken account of private sector wage trends (in 2013, public sector salaries exceeded those in the private sector by some 18 percent in the West Bank and almost 50 percent in Gaza). Third, as a result of automatic promotions, the PA—and in particular its security sector—has a top-heavy grading structure, with more staff at senior levels than normally observed in other countries. Public sector unions have played a significant part in bringing about this situation. The labor intensive nature of many government services, the outcome is a higher than necessary cost of service provision. 6 LABORISTA database, ILO. 7 Africa, in turn, has a significantly smaller proportion of government employment to population: c. 2 percent of the population, as opposed to almost 5 percent in the Palestinian territories – see International Labor Organization’s Laborista database for 2008, and World Bank World Development Indicators database for 2008; calculations for the Palestinian territories are made by the World Bank. xvi Figure 2: Average public salary (as a multiple of GDP per capita) in the Palestinian territories compared to MENA and most other regions Average public pay as a multiple of per capita GDP Africa Palestinian territories MENA Asia LAC OECD ECA 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Source: World Bank staff calculations and World Bank 1999. 11. Although the overall numbers of PA staff are not exorbitant, pockets of excess exist. With close to 65,000 staff, the PA’s security sector may not be large in Middle Eastern terms, but is very large by global standards (at almost 10 personnel per 1,000 inhabitants, compared with a global average of some 4.5/1,000). Aligning staff numbers and skills to the security needs of the territories would allow these needs to be met at much lower cost. Both the health and education sectors have larger numbers of staff than other countries in the region, and other lower middle-income countries. UNRWA-run schools employ fewer teachers per pupil and have significantly larger class sizes, but outperform PA schools on standardized tests. Both the West Bank and Gaza have significantly higher doctor-patient ratios than comparator countries across the region. Again, the outcome is much higher costs for the health and education services delivered. 12. The report proposes measures that can help reduce the wage bill to sustainable levels. This can be accomplished by:  Containing staff increases for several years; empowering the Ministry of Finance to determine the overall wage bill and its allocation and giving it the authority to approve any new hiring (these controls should also be applied to temporary employment);  Integrating most allowances into base pay;  Freezing the wages of PA staff whose pay is substantially higher than Palestinian private sector employees at an equivalent level until the wage gap has been substantially reduced; and  Reducing the number of higher ranked staff in the substantially “over-ranked” security sector. 13. The report estimates that implementation of these measures would reduce the wage bill/GDP ratio by over 6 percent. These cross cutting measures should be supported by reforms that give more responsibility to service delivery units for personnel management, and enhance the focus on employing the right numbers of staff with the appropriate skills and remuneration, to improve the value of services provided to the population. xvii 14. An acceleration of economic growth will also help reduce the wage bill to GDP ratio further, provided wages are not increased. The removal of movement and access restrictions, in particular the restrictions on economic activity in Area C and the blockade of Gaza, will be a key determinants of the rate of growth in the Palestinian territories. Strong growth will contribute to revenue growth. Figure 3 also shows that a further 5 percent of GDP reduction of the wage bill/GDP ratio would be achieved with a strong economic growth in the 6 percent per annum range projected by the IMF as a plausible consequence of a breakthrough in the peace process. Figure 3: The impact of controls on PA hiring, wage growth containment and reduced over grading in the security sector plus and wage growth, plus strong economic growth Source: World Bank Staff Calculations based on the data from the Palestinian Ministry of Finance. Note: These savings have been calculated after five years, as compared with a baseline of 3 percent employment growth and wage growth equal to inflation plus 1.25 percent per annum—a baseline that would leave the wage bill/GDP ratio would remain relatively unchanged over the period. The Public Pension System 15. Public pension expenditures in the Palestinian territories are high and unsustainable. Total pension expenditure exceeded US$300 million in 2013, around 3 percent of GDP. Even though this is close to the MENA average of 3.6 percent, it is extremely high in the Palestinian context given the territories’ demographic profile, with only 3 percent of today's population above the age of 65. The transfers required to service the system confront the PA with an enormous fiscal burden, which it is already failing to meet (accumulated arrears already amount to some US$1.6 billion8 or 13 percent of GDP), and without parametric reforms these obligations will either fail to be financeable, and/or will crowd out more equitable spending on important social priorities: welfare, healthcare and education. 8 This amount represents arrears accumulated to civil servant schemes only since the amount owed to the security services schemes is currently unavailable. xviii 16. The pension system is fragmented, consisting of four schemes, two of which were inherited from Jordan and Egypt with another two created subsequently. All schemes provide extremely generous benefit formulae, with payment rates of up to 100 percent of final salary. The system also offers early retirement options from the age of 45 onwards. Eligibility criteria are broad, permitting large numbers to draw survivorship and disability benefits. 17. The combination of high benefits and low coverage raises important equity concerns. The system provides generous and unsustainable retirement benefits for those few Palestinians of retirement age who had worked in the public sector, who as mentioned, are relatively well remunerated. However, it leaves out more needy individuals who have not been employed or have worked in the private sector. According to World Bank estimates, the system will cease to pay for itself in five years if present management practice continues; even if the PA stops accumulating arrears, the onset of the deficit will only be put off another decade. 18. Restoring the Palestinian pension system’s sustainability requires a number of significant adjustments. The onset of the deficit could be delayed to 2041 by a combination of raising the retirement age to 65, reducing the accrual rate of pension benefits to 1.5 percent9, and indexing pensions to inflation. Achieving long-term financial sustainability, though, will require additional action. The PER recommends the adoption of 'actuarially fair' reduction coefficients in the case of early retirement10, as well as reducing the eligibility for survivorship and disability benefits along with the limitation (or gradual elimination) of the option to purchase years of service instead of serving them. For the critical Scheme II,11 the report also recommends that the funded Defined Contribution (DC) component be changed to a pay-as-you-go Defined Benefit (PAYG-DB) system, which is much easier to implement. Again, alongside these cross cutting measures, it would be useful to fully integrate the cost of the pension system into the employment costs for each area of government activity to reflect the full cost of staff and support better resource management by service delivery units. Local Government Financial Management Energy Subsidies -- The Net Lending Issue 19. Net lending, resulting in unplanned energy subsidies paid through local governments’ budget, has become one of the thorniest fiscal issues facing the PA. To a significant extent, local government units (LGUs) finance their operating budgets by selling electricity and other utility services provided to them by Israeli companies, and leaving the PA to repay some or all of the costs (these are deducted by Israel from clearance revenues due to the PA, along with an 11 percent late fee). As a result, the PA finds itself providing unplanned subsidies of over US$200 million per year (2 percent of GDP) to the LGUs. Inadequate investments in maintenance and upgrading of the electricity sector have led to significant technical losses, estimated at a further US$200 million in purchased, but lost, electricity. The MoF attempts 9 The 'accrual rate' is the rate at which a beneficiary builds up his/her pension benefits while a member of a defined benefit scheme. The rate is multiplied by the person's earnings to calculate how much money he or she will eventually be entitled to, and is typically expressed as a fraction: the bigger the fraction, the greater the pension benefit. 10 “Actuarially fair” coefficients are based on an individual’s age and gender, and take account of projected mortality rates and pension indexation rules, and apply a specific discount rate. Such coefficients equalize the net present value of a reduced benefit stream given at an earlier age, and a regular benefit stream awarded at the statutory retirement age. 11 Schemes I, III and IV are being phased out; Scheme II covers all workers who were 46 years or younger in 2006 (the other three cover civilian and security employees who were older than 46 in 2006). Projections show that by approximately 2020, all PA government workers will be covered under Scheme II. xix to recover those losses by withholding revenues otherwise due to LGUs (municipal property tax, professional permit fees, transportation tax for example), but these intercepts by no means offset utility non-payments and lead to disputes and chaotic budgeting. 20. Reducing this drain on the PA budget will require steps in two directions: first, reforms that will increase own-source LGU revenues, and second, the commercialization of electricity and water distribution services: utility management requires skills and governance structures not generally found in local governments. The transfer of these functions to dedicated public distribution corporations would also reduce the fragmentation of service provision, leading to some efficiency gains. Resolving the net lending problem cannot be undertaken overnight and will need to be managed over time. Arrears are in some cases massive in comparison to actual or potential own-source revenues, and the PA will need to consider writing off a portion of the debt in order to put a stop to the current dysfunctional system of intergovernmental finance. Intergovernmental Fiscal Relations 21. From an LGU perspective, the availability of utility subsidies focuses excessive attention on the provision of electricity and water as a means of acquiring revenue. In consequence of this and of MoF intercepts of own-source revenues, attention to other local public services is neglected or is of substandard performance. As mentioned earlier, ensuring that LGUs are financially better-situated will require an increase in local own-source revenues. Bank analysis, however, shows that some LGUs will remain unviable without access to intergovernmental transfers to replace their net lending income. 22. As a result, LGU revenues are insufficient to fulfill many assigned mandates (80 percent of LGUs deliver an average of only 12 of the 27 services they are mandated to provide). Local government revenues amount to only 11 percent of total revenues, while local government expenditures account for 6 percent of total public expenditures: this compares with 18.3 and 19.7 percent respectively for the Eastern Europe/Central Asia region, for example. Put another way, LGU revenues amount to less than 5 percent of GDP, roughly one third of the level observed in East Asia and one half of that observed in Europe. The revenue scarcity is in fact worse than this, because roughly half of all municipal revenue comes in the form of utility income. In many countries, electricity, water and sewerage are provided by public utilities and are not part of the municipal budget. 23. Increasing LGU own-source revenues is essential. Theory and best international practice in fiscal decentralization show there are many advantages to own-source revenue financing, including increased accountability and better fiscal responsibility and decreased corruption by local officials.12 If local revenue collections are increased, LGUs will need to address local residents’ probable lack of willingness to pay (given that they may be concerned at the excessive costs in the provision of some services and are unlikely at this stage to see any clear link between tax and fee payments to LGUs, and the services they receive). This will require increased transparency alongside improved service delivery. 24. An obvious source of additional local revenue is property tax, which is currently collected in only about 20 percent of all municipalities, and not at all by village councils. The PER estimates that LGUs could easily double or even triple property tax collections. Additional revenues will also need to be mobilized, for example through the modernization and enhancement of the professional tax as well as the 12 Martinez-Vazquez, J., et al., The Impact of Fiscal Decentralization: A survey. International Center for Public Policy Working Paper 15-02, 2015. xx gradual introduction of new taxes such as the so-called 'betterment levies'13. There is also scope to increase revenues from service fees, but it is important to prevent the emergence of 'nuisance levies' that impose heavy burdens on individuals and businesses without raising significant revenues. 25. Even if own-source revenue potential is maximized, it will be difficult for LGUs to cover their expenditure needs. Thus, the PA needs to reform the current system of irregular and small-scale transfers to LGUs. The centerpiece of a new transfer system should be a conventional equalization grant across municipalities and VCs -- one that uses an objective, stable, and explicit formula based on the expenditure needs and revenue capacity of LGUs. Since it is supposed to equalize, only those LGUs with predictable deficits should receive funds. Although arriving at an acceptable formula will not be easy, the larger difficulty will be generating a sizable pool of stable and predictable resources. As already discussed in the Public Expenditure and Financial Accountability (PEFA) report, one possibility would be to dedicate 100 percent of transportation taxes to the equalization pool14, though this alone would likely be insufficient. The practice followed in many countries is to set aside a certain percentage of central tax revenues, often lagged by one year so that planning can be carried out within a defined fiscal framework. This can help create the firm budget constraint necessary to encourage the service delivery units within LGUs to focus on better resource management to enhance the quality of service and minimize their costs. Any new system of transfers would also benefit from conditional grants to incentivize LGUs as well as to address inter- jurisdictional issues such as environmental protection. The Public Health System 26. The Palestinian health sector is at a crossroads. The financial sustainability of the healthcare system is in doubt. The recent conflict in Gaza exposed major weaknesses in institutional and regulatory systems, and highlighted the precarious fiscal position in which the sector finds itself. Moreover, uncertain foreign aid flows, the increasing costs of referrals, inefficiencies and duplications of service, and an excessive focus on tertiary care are together straining the fiscal position of the health sector. Health expenditures are on the rise, while health outcomes are below potential for current levels of spending. Overall health expenditures (public and private) more than tripled in the last decade, reaching US$1.3 billion in 2012, or 12 percent of GDP—one of the highest shares of GDP in the world (See Figure 4). Public spending on health is close to 5 percent of GDP, and exceeds the MENA average of 2.6 percent and the Low and Middle Income Country (LMIC) average of 1.7 percent of GDP. Figure 4: Total health expenditure as a percentage of GDP in 2012 20 17.9 percentage 15 12.4 12.3 11.7 11.5 10 4.6 4.6 5 0 United States Netherlands Palestinian France Austria MENA LMICs Territories Sources: World Bank, 2014; PCBS & MoH, 2014. 13 Betterment levies, used in many countries at the local level, are one-time charges on the increased value of properties associated with urban improvements, such as the introduction of street lighting, sidewalks, or the construction of new roads, drainage, etc. in newly developed areas. 14 The West Bank and Gaza PEFA Public Financial Management Performance Report, 2014 update. xxi 27. The rise in public spending on health is being driven mainly by the salary bill, the cost of medical referrals outside the public health system, and high spending on pharmaceuticals (See Table 2). At the same time, public expenditures are allocated mostly to curative care, with hospital inpatient treatment representing a significant proportion of spending. In addition to introducing economies in these three areas (see below), global evidence shows that non-communicable diseases are best addressed at the primary health care level through cost-effective preventative interventions15, and the PA can improve the allocative efficiency of the health system significantly by shifting resources into preventative care and disease prevention.  The public sector wage bill has been expanding as a result of PA wage policies and by the expansion of the Palestinian health workforce (see earlier). In 2013, MoH employees totaled around 14,000 staff, 11 percent more than in 2006; this corresponded to 15 percent of all public sector employees. Staff numbers exceed other regional comparators, in part because of duplicate MoH personnel in Gaza, a large number of whom is not working.  The cost of outside medical referrals is heavy and has been increasing. The lack of availability of certain treatments, medications, medical staff, equipment and infrastructure within the public system has led to the creation of a referral system whereby large numbers of patients requiring tertiary care are referred to not-for-profit or commercial providers. Between 2000 and 2013, expenditure on referrals increased from US$8 million to US$52 million, and now corresponds to 48 percent of non-salary public health spending. MoH lacks a clear decision making mechanism to determine which services it would like to develop, which it would like to purchase from other service providers within the Palestinian territories, and which it should refer abroad. A 2013 World Bank study reviewed the appropriateness of outside medical referrals and concluded that a significant majority of referral documents were of doubtful quality, and that a large proportion of patients had circumvented the existing referral system. In 2015, the PA started to implement measures to control outside health referrals, and has already achieved some success in lowering the cost of referrals to Israeli hospitals.  Pharmaceutical expenditures accounted for 44 percent of non-salary public expenditures in 2013. While expenditures on drugs have shown large fluctuations, they have been a significant factor in driving up public health expenditure. There is an urgent need to improve the efficiency of pharmaceutical purchasing through more competitive procurement and attentive international price benchmarking. 15 World Economic Forum/World Health Organization, From Burden to "Best Buys": Reducing the Economic Impact of NCDs in Low- and Middle-Income Countries, 2011. xxii Table 2: MoH budget expenditure 2010-2013 (in US$ millions) 2010 2011 2012 2013 Recurrent Salary Expenditure 91 159 164 176 Recurrent Non-Salary Expenditure 104 97 129 108 Pharmaceutical Expenditure 50 48 70 47 Medical Referrals 46 40 49 52 Other 8 9 10 9 Capital Investment Expenditure 2 1 1 1 Total MoH Expenditure 197 257 295 286 Source: MoH, 2014. 28. Health system arrears are also of concern. MoH reports that in 2013 arrears associated with spending on pharmaceuticals and consumables, referrals, capital and other running expenditures reached US$193 million, corresponding to 67 percent of actual spending. Arrears due to referrals alone represented 44 percent of MoH arrears (excluding salaries) and reached US$85 million. 29. The financial sustainability of the Government Health Insurance (GHI) scheme is also in doubt, placing further pressure on the finances of the sector. The GHI scheme was established in 1994 to provide health insurance, including on a non-contributory basis to those facing hardship. Over the past decade, the number of non-contributing GHI enrollees increased compared to contributing members, damaging the scheme’s financial structure. In 2000, GHI revenues reached US$35 million; by 2009 they had declined to US$23 million, and continue to do so. In the meantime, GHI expenditures have been increasing rapidly (see Figure 5): as a percentage of total health expenditures, GHI costs amounted to 5.2 percent in 2000 but have increased steadily since 2005 and reached 15 percent of total health expenditures by 2010. Thus, the health sector faces a growing internal financial crisis,16 as well as placing considerable pressure on the PA budget. Addressing this situation will require significant adjustments to budget planning, budget discipline, staff deployment and management, and a set of policy priorities that currently prioritize unaffordable levels of tertiary care. It will likely require institutional changes that promote an increased focus on resource management at the front line to deliver the appropriate levels of service at the least cost. 16 The report also points to the severe welfare implications for the poor caused by rising outofpocket expenditures. Outofpocket spending accounts for 40 percent of total health expenditure and exceeds public spending. The poorest are at the greatest risk of impoverishment since they bear a higher share of outofpocket expenditures as compared to their share in national income. Expenditures on medications needed to treat chronic conditions drive these expenditures. xxiii Figure 5: GHI expenditure 2000-2012 250 in US$ millions 200 150 100 50 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS & MoH, 2011, 2012, 2014. Public Investment Management 30. One of the key issues facing the PA’s Public Investment Management (PIM) is the lack of a coherent system applying to both donor-funded and domestically-funded projects. Public investment in the Palestinian territories has until recently been funded almost entirely by international development partners (IDPs). While the implementation of capital investment projects has either been directly by the IDPs or through PA ministries, direct IDP funding for public investment has been channeled outside of the Treasury system and for the most part has not been reflected in the PA’s budget. Since 2009, the PA has funded a significant capital investment program from domestic resources. This situation has resulted in an incoherent approach to PIM in addition to a lack of consolidated data on commitments and spending. 31. Responsibilities for PIM have similarly been fragmented. In practice, prior to the recent merger between the Ministry of Finance and Ministry of Planning, responsibility for PIM has been divided between the MoF for domestically financed development expenditure and the Ministry of Planning and Administrative Development (MoPAD) for IDP financed projects. The situation is further complicated since on IDP funded projects many PIM functions, including project identification, appraisal and implementation, are carried out directly by IDPs or with very limited PA involvement. Also, a significant capital spending program is undertaken by the local government units (LGUs). Even though the Ministry of Local Government (MoLG) is required to approve LGU budgets and development projects, it currently undertakes no consolidated analysis of LGU budgetary operations. 32. Putting in place a comprehensive PIM system requires a combination of measures. Initially, a clear definition of public investment that is based on expenditure on fixed capital assets and related equipment costs needs to be agreed. Also, formal procedures should be established for screening projects at the identification/concept stage so that only those which are consistent with PA priorities and appear technically and economically sound proceed further to detailed preparation and appraisal. Procedures should also be developed for project selection and approval for financing. The PA should also put in place procedures for budgeting, managing, and monitoring projects against their total approved cost and implementation plans. A plan should be made to work with donors and bring all IDP funded projects into the budget. The responsibilities and functions of the MoLG with respect to the oversight and reporting of public investment by local governments should be fully elaborated and the required analytical capacities put in place. xxiv Chapter 1: Recent Macroeconomic and Fiscal Trends and Key Policy Challenges 1.1 Recent macroeconomic trends, growth drivers, and prospects 1.1. Between 2007 and 2012, the West Bank and Gaza enjoyed strong economic growth driven by consumption and net exports. The Palestinian economy rebounded in 2007, following the recession caused by the outbreak of conflict between Fatah and Hamas. Large amounts of donor funding, equivalent to 26 percent and 30 percent of Palestinian GDP in 2007 and 2008, respectively, boosted consumption and investment. GDP grew by 8 percent on average annually over the 2007-2012 period. Growth in consumption was the dominant contributor (adding 53 percent) to GDP growth over the period. Net exports contributed 45 percent of total growth, while the contribution of investment was very modest (only 2 percent of the total). Figure 6: Consumption, investment, net exports, and GDP, Palestinian territories, 2006-2014 (annual growth rate) 14.0% 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% 2006 2007 2008 2009 2010 2011 2012 2013 2014 -2.0% -4.0% -6.0% Consumption Investment Net Exports Gross Domestic Product 1.2. Growth decelerated substantially in 2013 and turned negative in 2014 due to the war in Gaza. A significant fiscal retrenchment, with a drop in government spending of almost 4 percentage points, caused a sharp fall in consumption in 2013, with the overall GDP growth sharply falling to 2.2 percent compared with 6.3 percent the previous year. This drop was a clear sign that the PA’s growth drivers of the 2007-12 period -- consumption supported by fiscal expansion -- had run out. In 2014, the overall economy fell into recession due to Gaza’s deep recession, caused by the outbreak of war and the bombing campaign in July and August. The depth of the recession was offset by surprisingly strong consumption growth in the West Bank driven by government spending, in particular a rapidly expanding wage bill, as well as growth in private consumption supported by credit growth, as local banks, flush with cash, launched a campaign to expand their lending portfolios. 1 1.3. The significant contribution of net exports to growth over this period—apparently due to import substitution-- needs to be taken with caution due to data quality issues. In six out of eight years, net exports made a positive contribution to growth. The main reason was a significant reduction in reliance on imports since growth in exports contributed only marginally. In real terms, the share of imports in GDP dropped 19 percentage points between 2006 and 2013. Nevertheless, these results have to be interpreted with caution as they may be due to statistical errors.17 Figure 7: Ratio of exports and imports to GDP (Left) and total value of exports and imports (Rght), 2006-2013 Source: PCBS and World Bank staff calculations. Note: Figure on the left is based on real values; figure on the right is based on nominal values. 1.4. Investment levels have been low. The average level of investment dropped significantly from 28 percent of GDP during the 2000-2005 period to 21 percent during the 2007-2013 period. The investment to GDP ratio in the West Bank and Gaza is more than 8 percentage points lower than the average for other low middle income countries. Moreover, most of the investment is in low-productivity activities. While data on the structure of investments are relatively scarce, the majority is in buildings, a good indication that it is not going to high-productivity activities; investment in non-building activities averaged 8 percent of GDP in the West Bank and 1 percent in Gaza during the 2007-2013 period. Wide consensus exists among economists that movement and access restrictions have inhibited private sector investments in both the West Bank and Gaza. 1.5. Growth patterns clearly differ between the West Bank and Gaza. While growth has been positive and relatively strong in the West Bank with the exception of 2013, it has been very volatile in Gaza. Deep recessions have given way to strong rebounds funded primarily through inflows of donor aid (in recent years, close to 80 percent of Gaza’s GDP).18 However, investment levels in Gaza are expected to stay relatively low in size and productivity until the political and security climate improves. 17 The breakdown of data between the West Bank and Gaza shows a smaller drop of imports/GDP in the West Bank (5 percentage points), while the drop in Gaza is much larger. However, national accounts data do not adequately capture the tunnel trade between Egypt and Gaza and the errors are likely to be significant. It is possible that the contribution of consumption to growth is much more significant and that of net exports smaller. 18 Exact levels of aid flowing into Gaza are not known. No estimates of the level of remittances flowing into Gaza exist, although they are thought to be significant. 2 Figure 8: Real GDP Growth for West Bank and for Gaza 20.0% 15.0% 10.0% 5.0% 0.0% 2006 2007 2008 2009 2010 2011 2012 2013 2014 -5.0% -10.0% -15.0% -20.0% West Bank Gaza Sources: PCBS and World Bank staff calculations. 1.6. Government spending has been a significant factor in economic growth, but has been declining. Tracking the relationship between government spending and growth in Gaza is difficult due to the lack of data on expenditures of the de facto authority, but for the West Bank it is clear that up to 2008, increases in government expenditures played an important role in growth by raising public and private consumption levels. Since 2008, however, final consumption expenditure and government expenditure have not always moved in the same direction (e.g., in 2009 and 2012). Furthermore, government expenditure significantly declined by 15 percentage points between 2006 and 2014 relative to consumption, which subdued its impact on final consumption. Nevertheless, growth in final consumption dropped substantially during the past three years and recent reductions in the growth of government spending were likely a factor. Figure 9: Growth in government expenditures and West Bank consumption versus government expenditures/consumption, 2006-2014 30 60% 25 50% 20 40% 15 30% 10 20% 5 0 10% 2006 2007 2008 2009 2010 2011 2012 2013 2014 -5 0% Government expenditures growth Final Consumption growth WB Government expenditure/Consumption 3 1.7. With the government’s reduced capacity to affect consumption growth, economic growth is expected to be sluggish in the coming years unless the political and security climate improves. As discussed below, with limited aid, the PA will have to further reduce its recurrent expenditures to ensure the sustainability of its finances. On the other hand, private sector growth is constrained by movement and access restrictions19. 1.8. Significant reduction in the rate of unemployment and poverty in the West Bank and Gaza can only be achieved with strong and broad-based economic growth that exceeds population growth. Evidence from around the world, and specific evidence from the West Bank and Gaza, shows that inclusive economic growth that reduces unemployment is the most effective contributor to poverty reduction. In both the West Bank and in Gaza, poverty is closely correlated with unemployment, with the poverty rate twice as high for the unemployed compared to employed individuals in both the West Bank and in Gaza. Although some employed individuals in the West Bank and Gaza are poor (almost exclusively those employed in the private sector), the most effective instrument to reduce poverty, even among the working poor, is to adopt policies that support growth and jobs. 1.9. Fiscal policy can contribute to improved growth prospects. Two main channels through which fiscal policy could support inclusive growth and development are: (i) improving fiscal sustainability to stop the accumulation of government debt and arrears; and (ii) creating fiscal space to increase public spending on infrastructure, which can both support private sector growth and directly contribute to improving the quality of life. As shown below, neither of these objectives has been met. 1.2 Broad fiscal developments, 2006-2014 1.10. The most noteworthy fiscal development over the period between 2006 and 2014 is that the PA has managed to significantly reduce the relative size of its recurrent fiscal deficit. The reduction in the recurrent deficit from 25 percent of GDP in 2007 to 10 percent of GDP in 2014 (Table 3) is significant. It was achieved mostly through reduction in the wage bill and net lending to GDP. The wage bill peaked at 24 percent of GDP in 2006 and has since been reduced to 16 percent, largely thanks to strong GDP growth, but also due to hiring control and wage growth restraint. Net lending has been a significant source of fiscal burden and the PA has recently taken a number of actions to reduce it, dropping from nearly 10 percent of GDP in 2007 to 2 percent in 2014. Analysis shows that it is possible to further reduce substantially net lending. Other categories of recurrent spending have fluctuated in the 13-14 percent range in recent years. Government revenues did not contribute substantially to deficit reduction. 1.11. Although the PA’s revenues increased by nearly 3 percentage points in 2014, they are still historically low. Revenues increased from 18.7 percent to 21.5 percent of GDP between 2013 and 2014, but were still about 2 percentage points below the levels observed in 2006 and 2008. The main reason revenues remain below recent levels is a reduction in the imports to GDP ratio and the resulting low level of customs revenues. At below 5 percent of GDP, tax revenues collected internally in the West Bank and Gaza remain rather small. 1.12. The wage bill remains very large by international standards. At 16 percent of GDP for the PA and around 17 percent of GDP for the general government sector in the West Bank and Gaza, the wage bill 19 Lack of access to conduct economic activity in Area C, which comprises 60 percent and is the only contiguous, resource-rich territory in the West Bank, costs the Palestinian economy an estimated US$3.4 billion per year (World Bank 2013). Recurrent violence and the economic blockade of Gaza have reduced Gaza’s economy to 40 percent of its potential by a conservative estimate. 4 is among the largest in the world. Its size is mainly the result of high average wages in the public sector, which in turn is the consequence of substantial overgrading, particularly in the security sector, and high wages at the lower end of the pay scale (see Chapter 2 for details). In 2015, the PA froze staffing growth, but due to automatic increases and additional increases in the health and education sectors, the wage bill grew by 1.4 percent in nominal NIS terms. 1.13. There exists a very low level of public investment. In 2006 and 2007, public investment spending amounted to 5.6 percent and 5.7 percent of GDP, respectively, but the level was reduced gradually to an average of 2 percent between 2012 and 2014. While total public investment (i.e., including local governments and out-of-budget public investment) is slightly higher in the range of 3 percent, public investment is still relatively low compared to well-performing developing countries and significantly below Palestinian needs, especially if the Palestinian population continues to grow at an annual rate of 3 percent. 1.14. External financing both for recurrent and development expenditures has drastically reduced since the peak of 30 percent of GDP in 2008, creating a significant fiscal adjustment challenge for the PA. By 2014, external financing—almost entirely grants—dropped to 10 percent of GDP; if one deducts the funding allocated for Gaza reconstruction, it dropped further in 2015 to 7 percent of GDP. Measured in nominal terms, external donor funding for the PA’s budget now stands at around US$800 million, about US$1 billion lower than it was in 2008. While both budget support and donor aid directed toward development projects were reduced, the most significant drop occurred in budget support grants. 1.15. Since the PA has not been able to make sufficient fiscal adjustments following the decline in donor budget support, it has relied upon the accumulation of payment arrears to close this gap, thus increasing the stock of debt. Total public debt held by the PA is close to US$5 billion, equivalent to 39 percent of GDP and close to the legal ceiling of 40 percent. Arrears to the pension fund (US$1.8 billion at end 2014) and private sector suppliers (US$700 million) comprised about half of total public debt. Foreign debt amounted to about 9 percent of GDP (US$1.26 billion) and domestic debt to the banking sector was only slightly below (US$1.23 billion). With limited access to external debt and a limit imposed by the Palestinian Monetary Authority on bank debt, the stock of debt increased entirely through the accumulation of payment arrears, which during the past three years increased by 70 percent. The rapid accumulation of arrears squeezes liquidity out of the private sector and leads to unsustainable debt levels. Table 3: Fiscal operations of the PA (In % of GDP) 2006 2007 2008 2009 2010 2011 2012 2013 2014 Total revenues 23.4 21.7 23.5 22.0 21.4 19.5 19.8 18.7 21.5 Tax revenues 4.5 3.7 4.1 4.1 5.1 4.6 4.6 4.8 4.7 Nontax revenues 3.2 2.2 4.3 3.9 3.0 2.4 2.4 2.1 2.1 Clearance revenues 15.7 16.3 16.8 15.2 14.1 13.6 12.8 11.8 14.7 Total expenditure 47.0 46.2 43.2 43.9 34.5 31.8 33.7 29.9 31.8 Wage expenditures to GDP 24.2 23.3 21.8 20.2 18.1 17.0 16.9 15.5 16.0 Nonwage 15.2 13.2 14.8 18.6 13.8 13.4 14.1 12.6 13.6 Net lending 7.7 9.7 6.7 5.1 2.6 1.3 2.7 1.7 2.2 Current balance (23.6) (24.5) (19.7) (21.9) (13.1) (12.2) (13.9) (11.2) (10.4) Development expenditure 5.7 5.6 3.2 3.0 3.4 3.5 2.3 1.5 2.1 External financing 20.8 24.0 29.6 19.3 14.3 9.4 8.9 11.0 9.6 Budgetary support 15.0 18.4 26.4 18.6 12.9 7.8 7.4 10.1 8.0 Development financing 5.7 5.6 3.2 0.6 1.5 1.6 1.5 0.9 1.6 Net domestic bank borrowing (3.5) (2.4) (0.4) 2.4 0.9 0.9 0.3 (0.6) (0.3) Net accumulation of arrears (in US$) 594 470 (416) 230 105 572 732 276 374 Source: MoF. 1.3 Key fiscal policy issues and challenges facing the PA 1.16. A top fiscal policy priority for the PA is to stop increasing the stock of arrears. Although largely anecdotal, evidence suggests that arrears to the private sector are creating financial problems for 5 some enterprises and are certainly a drain on private sector liquidity. Without adjustment, the PA’s debt will become unsustainable and further fiscal adjustment will be needed to reduce the primary fiscal deficit to the debt-stabilizing level of 1.4 percent. 20 1.17. Further reduction in the recurrent fiscal deficit is required to create fiscal space to increase public investment and repay all existing arrears. While the reduction in the overall deficit level would constrain growth in the short term, the reduction in the recurrent deficit would help create fiscal space for investment, address infrastructure gaps and potentially help boost growth in the long term. The objective should be to increase public investment to about 5-6 percent over the medium term to start addressing a significant infrastructure deficit to improve the quality of life for Palestinian citizens and facilitate private sector growth. Ideally, this target should be achieved through a combination of deficit reduction and increased donor support for development spending. The PA also needs to create adequate fiscal space to repay its arrears to the private sector, which amount to about a quarter of its revenues. Arrears to the Pension Fund also need to be cleared along with reform of the pension system to reduce future liabilities, as the pension system is otherwise unsustainable even if the PA clears all past arrears (see Chapter 4). 1.18. Greater public investment ought to be accompanied by substantial improvement in the capacity for investment management. As mentioned earlier, public investment in the Palestinian territories is low and inadequate to improve public infrastructure, which is needed to improve the quality of life of Palestinians and facilitate private sector investment by reducing the cost of doing business. However, even if significant fiscal space were created now to increase public investment, it is unlikely that investment would be managed effectively and efficiently to optimally serve Palestinian development priorities. The PA’s public investment management capacity is weak and needs strengthening. Chapter 6 provides a detailed assessment of the Palestinian PIM capacity and gives specific reform recommendations to strengthen it. 1.4 Policy ingredients for successful deficit reduction 1.19. The relatively low level of tax revenues is a key issue facing Palestinian policymakers and should be given a high priority. The general government revenues to GDP ratio in the West Bank and Gaza is significantly lower than in well-performing countries that do not finance their recurrent deficits with arrears. For example, the ratio in the West Bank and Gaza is significantly below the world average, the Middle East and North Africa (MENA) average, and the average of emerging and developing countries in Europe (Table 4). An important reason behind the low revenue ratio is related to the amount of tax revenues collected in Gaza’s tax base which is significantly lower than that in the West Bank. This is a result of the internal division between Fatah and Hamas, which takes off almost 4 percentage points of general government revenue. Nevertheless, even without these losses, general government revenue in the West Bank and Gaza would still be lower than in the vast majority of relevant comparators. Another important reason is the very small amount of revenues collected by local governments, amounting to a mere 2 percent of GDP. Inadequate municipal revenues contribute to the PA’s deficit by increasing the need to finance local government operations through transfers.21 1.20. Significant potential exists for increasing central government revenues by increasing the tax base, which is still narrow despite recent expansions. In particular, the tax base for internally collected 20 Based on the latest debt sustainability analysis conducted by the IMF in early 2015. 21 Currently, there are no significant mandatory transfers to sub-national levels of government, but the PA indirectly finances municipal budgets through the deductions Israel makes from revenues it collects on behalf of the PA to clear unpaid utility bills that local government units owe to Israeli suppliers (a process called net lending). 6 direct and indirect taxes is small as a result of weaknesses in tax legislation that provide generous tax holidays (e.g., investment promotion law), as well as weaknesses in tax enforcement, as a consequence of which many individuals and companies are not registered to pay taxes. By reducing tax holidays and improving tax collections, the PA could generate sufficient additional revenues (2-3 percent of GDP) to end its reliance on arrears for deficit financing. Recent efforts by the PA to improve tax collections are welcome and need to be intensified. Table 4: General government revenues/GDP, selected regions Comparator General government revenue/GDP European Union 45.2 Advanced economies 37.0 Emerging and developing Europe 36.9 Middle East 35.9 Middle East, North Africa, Afghanistan, and Pakistan 34.4 World average 31.6 Emerging and developing economies 28.6 West Bank and Gaza actual 23.9 Source: Economy Watch database, www.economywatch.com and World Bank staff calculations. 1.21. The PA needs to reduce the size of its recurrent expenditures, primarily by reducing the size of its wage bill, which is currently far above an affordable and reasonable level. A top priority for the PA ought to be reducing the size of its wage bill since it is at least 6 percent of GDP (or some US$740 million) higher than usually observed in other countries. If the wage bill of the civil and security services in Gaza (paid by Hamas) were added to the total, another US$450-500 million a year would be added to the wage bill. A chapter of this PER is devoted to the analysis of main drivers behind the bloated wage bill. The analysis shows that a significant reduction of the wage bill over the medium term is possible (about 4- 5 percentage points), but will require a systematic pay and grading reform and relatively strong economic growth averaging at least 5 percent per year. 1.22. The eventual reintegration of the public sector in Gaza will pose a fiscal challenge in the short run, but the benefits should outweigh the costs. Currently, only information on civil servants hired by the de facto authority is available, whereas, the number of security sector staff is unknown. Civil servants hired by Hamas in Gaza would raise the PA’s payroll by approximately USD 389 million.22 In order to finance these additional costs, a mix would be needed of significant downsizing of the combined public sector, an increase in tax revenues related to economic activity in Gaza and/or incremental sources of donor financing. Preliminary estimates indicate that the incremental revenues that could be collected from Gaza would eventually be sufficient to cover the costs of adding those staff to the PA’s payroll, but at this stage it is difficult to forecast the pace of growth of those revenues and the time when they will be adequate to finance the increased staff on the PA’s payroll in Gaza. Irrespective of the amount of revenues which could be collected from Gaza, the unification will lead to some staff redundancies that will have to be addressed through civil service reform. 1.23. Pension reform is another important policy agenda facing the PA. Public expenditure on pensions amounts to about 3 percent of GDP. This is extremely high given that only 1 percent of the total Palestinian population receives pension benefits, which are roughly 80 percent of preretirement income for current beneficiaries. In addition to being overly generous, the pension system is unsustainable. As noted, the PA currently accumulates substantial arrears from the pension system, which by the end of 2014 were 22 World Bank staff estimates based on data obtained indirectly on the number of civil servants and salaries for Hamas staff in Gaza. 7 about US$1.8 billion. Excluding the arrears, the pension system would still be insolvent in about 15 years in the absence of reform. The earlier reforms are conducted, the easier it will be to ensure the pension system’s sustainability. Chapter 4 provides an in-depth assessment of the key issues facing the Palestinian pension system and specifies policy recommendations to render it sustainable. 1.24. Energy subsidies are another significant source of inefficiency in the PA’s expenditures. Although the PA has significantly reduced the amount of fuel subsidies over the past year, subsidies still cost US$150 million per year. While incidence analysis on subsidies is not available to determine the extent to which different segments of the West Bank and Gaza population benefit from fuel subsidies, ample evidence from other countries shows that fuel subsidies disproportionately benefit the wealthier segments of a population. Thus, they represent a highly inefficient instrument to offset the impact of fuel prices on the poor. The PA is strongly advised to fully eliminate the fuel subsidy while at the same time allocate additional funds to its well-targeted cash transfer program to expand the depth and scope of coverage. Even if it were to offset the impact of higher fuel prices on the poor by increasing expenditures on social benefits23 by 15 percent, the PA would still save about US$100 million per year. Furthermore, the PA and municipalities provide substantial implicit electricity subsidies to individuals and private companies that do not pay their electricity bills in full. Unpaid electricity bills cost the PA about US$150 million per year (based on 2013 data); roughly two-thirds stem from unpaid bills in the West Bank and one-third from Gaza (World Bank 2014).24 Significant savings could also be made by reducing electricity losses, the result of electricity theft and underinvestment in the electricity grid, which together amount to losses of about US$200 million.25 23 Social benefits were US$358 million in 2014. 24 World Bank, Assessment and action plan to improve payment for electricity services in the Palestinian territories: study on electricity sector contribution to net lending, 2014. 25 Network losses are equivalent to 23-30 percent of purchased electricity. See: World Bank, Assessment and action plan to improve payment for electricity services in the Palestinian territories: study on electricity sector contribution to net lending, 2014. 8 Figure 10: Electricity Purchases, Sales, Losses, Net Lending & Debt to Suppliers (in million NIS) 3,000 Purchases West Bank from IEC Gaza 2,500 900 and other Purchases suppliers Losses from IEC 800 Debt823 479 1,935 Losses 2,000 700 180 Not 247 collected 600 Debt Not 374 500 collected Net lending 1,500 638 400 166 374 Sold to Net Lending 300 Collected Sold to Customers 1,000 Collected 317 from customers 200 575 Payment to 1,972 from customers 100 suppliers customers Payment to 409 269 500 1,598 IEC - 1,002 - Sources: Israeli Electric Corporation, PA, and World Bank staff calculations. 1.25. The PA and local governments need to start intergovernmental fiscal system reform to eliminate the substantial burden of net lending on the PA’s budget (2 percent of GDP in 2014). Even if the collection rate reached 100 percent and losses were fully eliminated (unrealistic assumptions), the issue of net lending would probably remain unresolved, because local governments (i.e., municipalities and village councils) use a significant share of electricity revenues to finance their recurrent expenditures. Out of the NIS 2 billion collected in electricity bills, utilities in the West Bank and Gaza remit only NIS 1.27 billion to suppliers, primarily the Israeli Electric Corporation (IEC). This situation is a consequence of very limited municipal revenues and an outdated system of intergovernmental fiscal relations. 1.26. Aid should be linked increasingly to sound fiscal and development policies. The reduction in donor aid in recent years has certainly had a negative impact on growth and rendered any fiscal reforms more difficult to achieve. Nonetheless, donors should coordinate their support to the PA more closely and seek to use their assistance as a leverage and catalyst for essential fiscal and structural reforms. 1.27. The PA can do a great deal to enhance fiscal sustainability and improve the efficiency and effectiveness of public expenditures. While economic growth can substantially benefit from an improvement in the political climate, and in turn help to improve the PA’s fiscal outlook, growth is not a substitute for reform. Much can be achieved through fiscal reforms that help improve revenue collection, reduce ineffective expenditures and improve the efficiency of remaining expenditures. 9 Chapter 2: Analysis of the Wage Bill 2.1 The large wage bill, its impact and trends 2.1. By any standard, the PA’s wage bill is very large. As Figure 11 shows, the general government wage bill in the West Bank and Gaza amounts to 17 percent of GDP and is among the highest in the world whether compared to developed, developing, or fragile and post-conflict states. Although a comprehensive dataset is not available, the central government contributes the majority of the oversized wage bill of the general government (at least 90 percent based on the World Bank’s estimate). Figure 11: General government wage bill/GDP, select countries Jordan Denmark West Bank & Gaza Lesotho Sweden Barbados Iceland San Marino Cyprus Morocco Portugal Finland France Israel Norway St. Kitts and Nevis South Africa Seychelles Belgium Hungary Bolivia Canada Bosnia & Herzegovina Bahrain, Kingdom of Greece Honduras Cape Verde Italy United Kingdom Switzerland Belarus Costa Rica Estonia Croatia Spain Poland United States Ukraine Colombia Austria Netherlands Lithuania Paraguay Kuwait Ireland Australia Angola Latvia Iran, Islamic Republic of New Zealand Moldova Argentina Burundi Mauritius Russian Federation Romania Maldives Macedonia, FYR Luxembourg Egypt Mongolia El Salvador Germany Malta Afghanistan Bhutan Albania Jamaica Thailand Bulgaria Japan Congo, Republic of Peru India Chile China, P.R.: Hong Kong Singapore Nigeria 0 2 4 6 8 10 12 14 16 18 20 Source: World Bank staff calculations based on IMF Government Finance Statistics data for 2000-2008. Note: The West Bank and Gaza estimate is for 2012. Estimates for other countries are averages of yearly estimates during the period 2000-2008. 10 2.2. Moreover, with the formation of the “Consensus Government”, the wage bill is expected to increase by another 3-4 percentage points of GDP. This will occur if, following the reconciliation agreement between Fatah and the de facto authority in Gaza and the consequent formation of the “Consensus Government”, employees of the de facto authority are added to the wage bill.26 This has yet to happen, but the PA with support from several donors has looked into the best modalities to merge the two administrations. Data currently only exist on the number of civil servants hired by the de facto authority in Gaza (50,112). This figure includes blue uniform security personnel, while the number of green uniform security employees is still unavailable. The de facto authority has about 21,000 civil employees at the Ministry of Interior. Together with employees of the Ministry of Health (MoH) and the Ministry of Education (MoE), they represent more than 70 percent of all civil employees on the de facto authority’s payroll. If all de facto authority civil employees were added to the PA’s payroll, the annual wage bill would increase by about US$389 million, or more than 3 percentage points of GDP. Some of the de facto authority’s civil servants are temporary employees recruited into employment programs in an effort to create jobs in Gaza. These include 2,186 employees offered short-term employment opportunities at various ministries for two to three months. In addition, 7,148 employees were hired through a temporary employment program by the Ministry of Labor for less than one year. If the administrations of the PA and the de facto authority in Gaza merged, many temporary employees would not be replaced when their contracts expire. If all temporary workers were removed, the merging of the two administrations would increase the PA’s wage bill by US$362 million. 2.3. The high wage bill has been and remains a large fiscal challenge despite significant recent reduction. In 2013, the PA’s wage bill amounted to US$1.9 billion, equivalent to 49.5 percent of total expenditures, 52 percent of recurrent expenditures, and a staggering 83 percent of its revenues (see Figure 12). Whereas, the average wage bill/revenue ratio for 154 countries is 25 percent. Consequently, wage bill control has been the key focus to reduce the PA’s chronic fiscal deficits, particularly since donor aid has been insufficient to cover them. Figure 12: Salaries as a percent of PA’s expenditures and revenues Item (US$ million) Total net revenues 2312 Development Net lending, Gross domestic revenues 852 5.4% expenditures, Interest, 2.3% 4.8% Tax revenues 597 Minor capital, Non tax revenues 255 0.2% Social Clearance revenues 1690 transfers, Recurrent expenditures and net lending 3694 20.3% wages and salaries, Wage expenditure 1919 Use of goods 49.5% Non-wage expenditure 1564 & services, 13.0% Net lending 210 Recurrent balance -1381 Development expenditures 186 Social Overall balance (before external support) 1568 contributions, Source: MoF (2013). 4.6% 26 These calculations take into account the de facto authority’s civil servants, which include blue uniform security personnel. They do not, however, include green uniform security employees, whose data are unavailable. 11 2.2 Wage bill growth in historical perspective 2.4. Since the PA’s inception in 1994, the wage bill has been shaped by efforts to build functioning institutions of a future state. During the past 20 years, a main objective of the PA and the international community has been building the institutions of a future state and enabling them to deliver a wide range of public goods and services, which necessarily led to an expansion of public sector employment for many years. Public sector employment was also driven by the need to offset low labor demand in the private sector, particularly following events like the second Intifada and the Gulf Wars, which reduced employment opportunities for Palestinians in the private sector of the West Bank and Gaza, Israel, and third countries. Figure 13: The PA’s wage bill as a percentage of GDP, 1995-2013 2,500 30% 2,000 20% 1,500 10% 1,000 0% 500 -10% - -20% 2000 1995 1996 1997 1998 1999 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Wage bill (in millions of USD) Wage bill to GDP Real GDP growth Sources: MoF and PCBS. 2.5. Institution-building was one of the main drivers of the PA’s wage bill growth. Public services provided by the PA in education, health, and security expanded over the past two decades, all of which added considerably to the wage bill. For instance, the number of government hospitals increased from 14 to 25 between 1997 and 2010 to keep pace with population growth with the same number of hospital beds per inhabitant. The number of healthcare workers increased in parallel; consequently, the PA’s health sector wage bill increased from US$48 million in 2000 (no earlier data is available) to US$172 million in 2013. During the same period, the number of schools nearly doubled, from roughly 1,000 to 2,000, and the education sector wage bill increased from US$168 million to US$531 million. With these institution- building efforts, the Palestinian territories managed to achieve public health indicators on par with middle- income countries and near universal literacy (literacy increased from 85 percent in 1995 to 96 percent in 2012). 2.6. Other factors contributed to wage bill growth, particularly before 2006. With the advent of the Second Intifada in 2000, restrictions on movement and access were substantially increased. These restrictions included a policy of reducing the amount of Palestinian labor in Israel and its settlements by limiting the number of permits issued. During this period, the PA’s wage bill grew by 13 percent as it started absorbing many of the Palestinian workers previously employed in Israel and Israeli settlements. Growth of the public service and wages was also facilitated by external resources. From the early days of the PA and particularly following the Second Intifada, Western and Gulf donors provided substantial aid to the PA to expand its capacity and to deal with the socioeconomic aftermath of the Intifada, when there were few formal employment opportunities outside the public sector. Public employment was also run up ahead of the hotly contested 2006 elections, when the total size of the public service peaked at 180,000 before being reduced by the Caretaker Government in 2007. 12 2.7. The wage bill/GDP ratio dropped sharply from 26 percent in 2006 to 15 percent in 2013. In 2007, the drop was the result of shedding staff hired prior to the 2006 elections. The World Bank does not have data on staff shed in the security sector, but 1,581 civilian employees lost their jobs. Strong economic growth between 2007 and 2012, averaging 8 percent per year, significantly contributed to the continued reduction of the PA’s wage bill relative to GDP. After growth started slowing in 2012 and the PA’s fiscal challenges intensified, the PA adopted a “zero net hiring” policy that continues to be in force. According to this policy, departments are permitted to replace staff who leave but not to add new positions or fill existing vacancies. Implementation of this policy reduced staffing on the PA’s payroll in 2012 by 3,176 but has been implemented unevenly. While PA-funded staff in Gaza were reduced by 5,502 employees between 2012 and 2013 due mainly to a deliberate policy of non-replacement (and limited donor investment), this was offset by a net increase of 2,326 in the West Bank staff. 2.3 Main issues with the size and structure of public sector employment 2.8. Despite the large wage bill, the overall size of Palestinian public employment is not large relative to that of other countries.27 Public sector employment in the West Bank and Gaza is equivalent to 4.6 percent of the Palestinian population. Even if one adds the employees hired by the de facto authority in Gaza, the ratio still remains close to 5 percent—17 percent below the average of 75 developed and developing countries from a World Bank database. Thus, public sector employment—most of which is contributed by the PA—is not the main reason for the extremely large size of the PA’s wage bill relative to GDP. 27 For the purpose of this PER, the public service in West Bank and Gaza comprise the civil service and the security services, mostly uniformed. It excludes employees of state enterprises, state autonomous bodies outside the budget, and municipal government. Local government staff number about 14,000, and are employed by municipal councils on terms that may vary between councils. Their salaries are mostly paid by local revenues, though there are some financial transfers from central government. 13 Figure 14: Estimated general government employment as a percentage of total country population, select countries 0 2 4 6 8 10 12 14 16 18 Denmark Sweden United Arab Emirates Belarus Latvia Lithuania Estonia Trinidad and Tobago Hungary Luxembourg Ireland Belgium Syrian Arab Rep. Israel Czech Republic Cyprus Croatia Moldova, Rep. of Austria Mauritius Switzerland Bulgaria New Zealand Fiji West Bank and Gaza Brazil Japan Jordan Panama Sri Lanka Morocco Bolivia Cuba Indonesia India Afghanistan Armenia Source: World Bank calculations produced using public employment numbers from the International Labor Organization’s Laborista database for 2008 and World Bank World Development Indicators database for 2008. The estimate for the West Bank and Gaza is for 2013 and was produced by World Bank staff based on data from the PA. Note: The estimates for some countries may have changed significantly since 2008, but it can still be safely concluded that public employment in the West Bank and Gaza is not excessive compared to other countries around the world. The planned addition of roughly 50,000 new staff from Gaza would increase public employment in the West Bank and Gaza by about 26 percent, but this would still not make public employment an outlier in the broader international context. 14 2.9. While not high by international standards, public sector employment by the PA has increased over the past six years. Staffing increased from 149,305 employees in 2007 to 169,095 in 2013, an increase of 13 percent over the period, following the Caretaker Government’s shedding of temporary hires. In the West Bank, however, the increase was greater, with total public service employees growing from 78,419 in 2007 to 104,095 in 2013, an increase of 33 percent over the period significantly faster than the rate of population growth. By contrast, the PA’s public service in Gaza shrank from 70,886 in 2007 to 64,998 in 2013, a reduction of 8 percent over the period resulting from not filling gaps caused by retirement, resignation, sickness, disciplinary dismissal, or other forms of attrition. 2.10. Even though the aggregate level of public employment is not excessive, specific organizations within the PA appear overstaffed relative to pertinent international benchmarks. This is the case with the security sector (Figure 15), but efficiency improvements, in terms of personnel numbers, could also be achieved in the education and health sectors, as well as some other units of government. 2.11. The PA’s security sector’s wage bill is large by international standards. The total wage bill of security services amounts to 8 percent of GDP. After deducting police and prison services, the remaining paramilitary wage bill ratio is 5 percent of GDP. International comparisons show most countries in a range of 2-2.5 percent of GDP for total military spending (i.e., salaries, operating costs, and capital spending of military and paramilitary units), and a Middle East average of about 4.5 percent. Likewise, very few countries have such a high proportion (44 percent in 2010) of the total centrally funded public service in police and armed forces. Figure 15: Wage bill of security sector as a percent of GDP, select countries Sources: International Institute of Strategic Studies in London (IISS), the Stockholm International Peace Research Institute (SIPRI) and the MoF. 15 2.12. Figure 15 reveals that the PA’s paramilitary forces are large relative to GDP and to population by international standards. According to MoF payroll data, 64,934 persons were in the security forces in 2013, which is large compared to the global average and even to countries with substantial military expenditures. On the other hand, the size of the Palestinian paramilitary forces is less exceptional when compared to other countries in the region. The wage bill of the security forces is very large and undermines fiscal sustainability.28 The greatest excess appears to be in Gaza, where the PA’s force of 30,000 may have to be integrated with the de facto authority’s recruited force (number unknown). Figure 16: Number of security (non-police) personnel, select countries (military personnel per 1,000 inhabitants) Sources: International Institute of Strategic Studies in London (IISS), the Stockholm International Peace Research Institute (SIPRI), and the MoF. 2.13. Not only is the Palestinian security sector large in terms of personnel numbers, it is also substantially grade inflated. The total security force comprises one-third officers and two-thirds Non Commissioned Officers (NCOs). By any normal yardstick, this represents a disproportionately large officer corps. In other countries, structures are typically flatter with broader spans of control in the hierarchy.29 Military ranks, and their distribution in 2010 (when the total size of the security force was 63,586, including Police Crime Intervention) is shown in Figure 17. 28 Comparing the Palestinian security forces with the armed forces of other countries requires some adaptations. First, the paramilitary side of the security forces amounts to about two-thirds of the total of 65,000, after netting out the Police Crime Intervention (PCI) and the more civilian components of the uniformed branches, such as prison staff. Second, data from other countries have to be adjusted by netting out reservists (many countries have large numbers of reservists not on active duty) and counting only active military and paramilitary. 29 Armed forces of most countries conform to long established control norms: a lieutenant, the lowest ranking field officer, will normally command a platoon of 40-50 men, a captain a company of about 4 platoons, a lieutenant colonel a battalion of 4 companies, a colonel a regiment of several battalions, and a brigadier-general (the lowest ranking general officer) a brigade of several regiments. Divisions (10,000 or more men) will be commanded by a major- general. In practice, officer structures are supplemented by staff and logistical functions, and support and technical units with smaller spans of control, but by no means to the extent found in the Palestinian security forces. 16 Figure 17: Distribution of Palestinian paramilitary forces by military rank, 2010 0 2000 4000 6000 8000 10000 Major General Brigadier General Colonel Lieutenant Colonel Major Captain First Lieutenant Lieutenant Staff Warrant Officer Warrant Officer Sergeant Major Sergeant Corporal Lance Corporal Source: General Personnel Council of the PA. 2.14. Based on international benchmarks, the size of the Palestinian police appears adequate in the West Bank but oversized in Gaza. There are about 8,800 policemen in the West Bank, and a further 8,000 in Gaza. Formerly, 12,000 police in Gaza were on the PA payroll, but following 2007, officers were told to stay at home. The PA has continued to pay their wages and all allowances (except, most recently, transport), but the number has since declined to around 8,000.30 The median number of police in countries across the world, as calculated by the United Nations, is about 300 per 100,000 population.31 Using this yardstick, a reasonable size for the PCI in the West Bank would be 8,100, and 5,100 in Gaza. Accordingly, the size of the police force in the West Bank is only slightly above global norms, but disproportionately large in Gaza. Finally, if the police force recruited by the de facto authority was merged with the stay-at-home police in Gaza (on PA’s payroll), the combined force would very probably suffer from extensive redundancies. 2.15. The public education sector is also relatively overstaffed. In 2013, the education sector had 48,502 personnel on the PA’s payroll, an increase of 20 percent over the past five years, while at the same time the number of students in government-run schools dropped by 2 percent.32 The aggregate masks the fact that the number of education personnel on the PA’s payroll in Gaza stayed roughly the same over the period, while the number of personnel in the West Bank grew at an average rate of 30 percent during the past five years. Thanks to an increased number of teachers, the number of students per classroom dropped from 33 in school year 2007/2008 to 29 in 2012/2013, while the number of teachers per class (an indicator of teacher workload) stayed roughly the same at 1.4. On average, class size is significantly smaller in government-run schools than in United Nations Relief and Works Agency (UNRWA)-run schools (29 students versus 36). Smaller class size in government schools has not translated into better performance, 30 Some PCI members in Gaza switched to the de facto authority’s own police formations on their own volition, while others retired. 31 The Report of the Twelfth UN Congress on Crime Prevention and Criminal Justice, 2010, (page 19) refers to UN survey data, suggesting a median of 300 policemen per 100,000 of population globally. 32 The source for staffing numbers is the MoF and for the number of students is PCBS. 17 however, as students from UNWRA-run schools out-performed those from government schools in both math and science on standardized Trends in International Mathematics and Science Study (TIMSS) tests conducted in 2003 and 2007. In fact, research shows that within a range of class size between 20-40 students, no significant correlation exists between the number of students per class and student performance.33 Figure 18: Number of staff in the education sector in the West Bank and Gaza, 2007-2013 60,000 50,000 40,000 Staff 30,000 20,000 10,000 0 2007 2008 2009 2010 2011 2012 2013 West Bank & Gaza West Bank Gaza Sources: PA MoF and General Personnel Council. 2.16. In the interest of efficiency, the PA should consider reducing the number of teachers relative to the student population. While reducing class size may not always be possible without consolidating smaller schools into larger ones, a recent study conducted by REPIM (Research on Economic Policy Implementation and Management) concluded that 242 schools in 2010 already had an above-optimal number of teaching positions, indicating room for efficiency gains without having to consolidate schools, which would ultimately also be desirable.34 2.17. Wages in the Palestinian public health sector account for roughly half the total costs of this sector and overstaffing appears to be an issue as well. The number of health sector employees on the PA’s payroll grew until 2010, and has been stagnant thereafter. As in other sectors, the aggregate masks differences between the West Bank and Gaza, as the number of personnel in Gaza dropped by 13 percent, while the number in the West Bank increased by 59 percent. Figure 20 shows that the Palestinian territories, and in particular the West Bank, have a significantly higher number of medical professionals than comparator countries in the region and significantly more than other lower-middle-income countries. While higher numbers of staff should translate into better-quality healthcare services, the PA simply cannot sustain this level of expenditure. In addition, administrative workers account for 35 percent of all public health staff in the West Bank and Gaza, which is very high by international standards, and suggests significant overstaffing in the public health administration. 33 See Chingos 2011 and 2010. 34 Ministry of Education and Higher Education (MoEHE) officials argue that hiring is based on manpower planning that takes account of the increase in the school-age population (believed to be 3 percent per annum), the opening of new schools, and the need to supply more teachers to densely populated urban areas, where the pupil/teacher ratio can exceed 50:1, while at the same time keeping open schools in remote rural areas where the ratio may be a fraction of this. 18 Figure 19: PA’s health sector staff, 2007-2013 Figure 20: Healthcare professionals (per 10,000 inhabitants) Sources: The MoF, the MoH, WHO, and World Health Statistics 2012 . 2.18. The most significant employment growth in the past five years has not been in the security sector, health, or education, but elsewhere in the civil service. Employment in those PA organizations (78 in total) increased from 33,570 to 42,152 (25 percent growth) between 2007 and 2013.35 In the West Bank the number increased from 17,416 to 27,025 personnel (55 percent growth), while in Gaza the number dropped from 16,154 to 15,127. Overall, 54 agencies saw their staffing increase and 24 saw theirs reduced. Looking at the West Bank alone, 65 agencies saw an increase and 13 saw a reduction. The most significant personnel increases took place between 2007 and 2011 in the organizations shown in Table 5. Further assessments (i.e., functional reviews) would be needed to systematically assess the merits of increases for each organization, but the link between personnel increases for most of those organizations and the PA’s strategic priorities is not obvious and it is likely that weaknesses in hiring controls coupled with pressure to hire at the agency level are plausible explanations for the rapid increase in the number of the West Bank- based personnel in these organizations. 35 This category comprises 78 ministries, departments, and agencies (MDAs) ranging from the Ministry of Religious Affairs (3,281 staff in both the West Bank and Gaza) down to small bureaus and councils with a few dozen staff. The group also includes 14 governorates, but these are small, typically with 30-70 staff each. 19 Figure 21: Increase in staffing (number of personnel) by ministry, 2007-2013 Source: GPC of the PA. Table 5: West Bank-based personnel increases in organizations that contributed the most to employment growth between 2006 and 2013 (Excluding health, education, and security sectors) 2006-2013 2006 2007 2008 2009 2010 2011 2012 2013 Increase Number of personnel Palestine Public Broadcasting Corporation 379 372 552 667 822 876 878 888 509 Ministry of Awqaf and Religious Affairs 2174 2131 2241 2341 2519 2685 2646 2564 390 Ministry of Social Affairs 466 460 561 669 694 763 762 776 310 Ministry of Finance 1077 1057 1132 1175 1237 1276 1305 1367 290 Ministry of Higher Education 423 449 472 504 553 591 585 614 191 Palestinian Academy for Security Sciences 10 10 10 10 26 151 189 186 176 West Bank Water Department 12 12 12 113 115 170 174 174 162 Judiciary 312 301 377 382 454 479 471 465 153 General Prosecutor 96 108 162 196 216 225 223 233 137 State Audit and Admin. Control Bureau 23 26 100 145 146 146 146 150 127 Office of the President 229 228 256 283 323 319 338 333 104 Ministry of Detainees and Ex-Detainees Affairs 193 193 197 241 265 268 267 267 74 Ministry of Public Works and Housing 372 364 382 397 444 459 449 443 71 Ministry of Justice 64 65 93 94 96 107 106 128 64 Ministry of Foreign Affairs 157 157 172 178 198 203 211 214 57 Ministry of Transportation 403 400 407 437 452 465 441 453 50 TOTAL 6534 6476 7316 8094 8895 9550 9558 9615 3081 Sources: General Personnel Council of the PA and World Bank staff calculations. 20 2.4 Institutional contributors to wage bill growth 2.19. Excessive personnel growth in some sectors can be attributed to inadequate strategic planning. A needs-based approach to planning and budgeting has characterized resource allocations. The PA has a relatively decentralized strategic planning process in which departments set their goals and determine paths to get there, often unconstrained by manpower or financial resource considerations. For instance, hiring in the health sector is driven by the need to staff new hospitals and the expansion of existing facilities, consistent with the MoH’s Strategic Plan, and in large part financed by external donors. The central management agencies in the PA have yet to develop an effective organizational and staffing scrutiny capacity to critically review demands from line ministries and departments. Until now, the focus of the General Personnel Council (GPC) has been to ensure ministries and departments follow processes established by law in the exercise of their delegated human resources (HR) functions for recruitment, grading, promotion, and discipline. This is important, but the GPC has the authority to play a larger role. The MoF, on its part, has been preoccupied with finding the money to meet the wage bill or, if that was not possible, delaying the implementation of cost-of-living increases previously agreed with public employees’ unions. Managing the wage bill more effectively in the future requires a change in the budgeting rules and process from needs to availabilities and what can be achieved with existing resources. Ideally, the MoF should have the power to set staff ceilings based on the medium-term fiscal framework. 2.20. Promotion and grading practices in the public sector also contribute to wage bill growth. Employees are typically recruited at the bottom of the pay scale corresponding to the grade to which they are recruited. After a minimum of five years in a grade, they are eligible for promotion to the next grade, based on good performance. To move from, say, a Fourth Category designation to a Third Category would require an officer to obtain a higher technical or professional qualification. While in many civil services elsewhere in the world promotion requires a job vacancy at a higher level, this is only loosely applied at higher levels and in practice not at all in the middle and lower grades, where the bulk of civil servants are located. Thus the PA remains essentially a “rank-in-person” system, not unlike other civil services in the region. The grading structure is particularly distorted in the security sector, as discussed earlier. Consequently, a security sector review is urgently needed to look at functional requirements, organizational structures, and staff levels of police and paramilitary formations and to come up with a new slimmed down, fiscally affordable structure. 2.5 Main issues with pay practices in the Palestinian public sector 2.21. The PA’s pay and grading system is typical of others in the region. It is characterized by education- (not job content-) based grades, seniority advancement, compressed scales, and multiple allowances. It is based on the 2005 Palestinian Civil Service Law, which stipulates that ministers enjoy a fixed pay rate and special allowances, while permanent officials fall within 17 grades divided into six partly overlapping categories (see Box 1).36 The distribution of personnel across grades shows the bulk of personnel in the middle, between the third and seventh grades. This is typical of civil services around the world, though second and fourth grades appear underrepresented in the PA. The bottom pay in the lowest scale is NIS 1,250 per month; the bottom pay of the highest scale is NIS 4,020 per month. Based on this, the compression ratio for the base pay is calculated at 3.2 to 1, which means that a top civil servant receives in base pay about three times the amount paid to the lowest permanent employee in the PA civil service. Calculation of salary compression ratios using base pay alone puts the PA in the company of some of the more advanced countries in Europe, as measured by the OECD. But this is misleading because in practice the real range is much greater due to a system of special job allowances. 36 Different arrangements apply for the security sector, where all staff including police have military ranks. 21 Box 1: The PA’s Salary Scale The scale is divided into Categories of functionaries and scale of salaries six partly overlapping categories. In the Higher Minimum Category, there are 4 Basic scales, A1 to A4, for Financial limit to salary Categories directors general and grade stay on (NIS) general directors. Next the grade comes the First Category, Heads of Departments Fixed Special for directors and Ranking as Minister rate Category managers, comprising A1 2 Years 4020 scales A to C. Thereafter, in descending order, come A2 2 3720 Higher 10 numerical scales, with A3 2 3470 Category the Second Category (section and unit A4 2 3220 managers, and other specialized functions) A 6 2970 comprising scales 1 to 5, First the Third Category B 6 2720 Category (Clerical and Technical C 6 2470 Functions) covering scale 1 5 2220 2 down to scale 7, the Fourth Category (Skilled 2 5 2090 Second Functions) covering scale Category Third 5 down to scale 9, and, 3 5 1960 Specialized Category finally, the Fifth Category 4 5 1830 Functions Clerical (service functions such as and drivers and cleaners) 5 5 1700 ranging from scale 6 down Technical 6 5 1570 Functions Fourth to scale 10. The scales are Category for basic salary and each 7 5 1490 Fifth scale has 40 pay points Profession Category progressing from the 8 5 1410 Functions Service lowest point in the scale by 9 5 1330 Functions 1.25 percent, yielding a 10 5 1250 total range of 64 percent within each scale. The bottom of scale 10 is currently NIS 1,250 per month and the bottom of scale A1 is NIS 4,020 per month while the top is NIS 6,607. Source: GPC of the PA. 2.22. Allowances make up a significant portion of public sector remuneration and their size depends on the grade and position of a staff member. The Civil Service Law defines a system of allowances that allows the PA to pay personnel significantly more than what base salaries alone would provide. These allowances include supervisory/managerial allowances, transportation allowances, bonus allowances for special assignments, and social allowances based on family size. In addition, employees get job allowances which are considered the most significant in size and are the means by which the PA is able to bridge between the highly compressed basic pay scales and more market-relevant ones. Taking the civil service as a whole, a typical employee earns about 61 percent of his/her monthly remuneration in base pay and 39 percent in allowances. In practice, huge variation arises between different grades and positions. For example, while the job allowance for nurses and lab technicians is around 20 percent of the base pay, doctors receive 100 percent, and specialist doctors 150 percent. Managers in Grades A, B, and C all receive 50 percent and job allowances for senior civil servants in Grades A1 to A4 vary from 60 percent to 90 22 percent of base pay.37 Workers in the Fourth Category get 15 percent. Similarly, other Third and Second Category personnel get 20 percent and 25 percent, respectively, of their basic pay in job allowances. The huge variation in job allowances means that the effective ratio between bottom and top earners in the PA civil service is about three times greater than the apparent compression ratio calculated on base pay alone. Figure 22: Components of the PA’s public sector salary, 2013 0% 4% 11% 3% 0% 17% 61% 3% 1% Base Salary Wage Workers Social Allowances Job Allowances Bonus Allowances Transportation Allowances Managerial Allowances Other Allowances Reward Source: MoF. 2.23. A significant fraction of employees are engaged on contract terms with remuneration packages higher than what is offered by the PA salary scale. Many of the short-term contracts are for specialists who are difficult to attract with the typical pay offered by the PA. Their salaries are part of the PA wage bill and special contracts are approved by the GPC and the Council of Ministers.38 Figure 23 shows how the number of contract employees has evolved over time. The existence of contract personnel on special terms signals the need to address the competitiveness of government pay scales. Temporary contracts, as in other civil services around the world, will always have a place, but civil service contracts should be sufficient to attract, recruit, and retain the bulk of the skills needed by the PA to run public services. This awaits the development of job descriptions and reform of the pay scale system. Managing the remuneration of the most skilled personnel through short-term contracts and variable job allowances can only be a temporary fix. 37 Other categories with high job allowances are: professors (250 percent), legal advisors (150 percent), dentists, veterinarians, pharmacists, and engineers (60 percent), and computer programmers (45 percent). 38 At the other end of the scale, a small number of daily paid staff, undertaking unskilled tasks, are not part of the public service and have no pension rights. 23 Figure 23: Number of temporary public sector personnel, 2006-2013 20 18 16 14 12 Thousands 10 8 6 4 2 0 2006 2007 2008 2009 2010 2011 2012 2013 Sources: GPC and MoF. 2.24. Average public sector wage relative to GDP in the Palestinian territories is among the highest in the world, mainly due to government policies and wage bill management practices. Average yearly public wage in the Palestinian territories was US$11,059 in 2013 according to payroll data provided by the MoF. When measured as a multiple of per capita GDP, it amounts to 3.5. As Figure 24 shows, this ratio is higher than the MENA average of 3.4 and exceeds the ratios in other regions including Asia, Europe and Central Asia, Latin American and the Caribbean, in addition to OECD countries. The ratio in the West Bank and Gaza is only lower than that of the Africa region.39 Relatively higher average public wages in the West Bank and Gaza can be explained by the PA’s pay practices. In particular, the PA implemented two large salary increases over the last decade. The first one took place in 2004, when average pay was increased by 8.4 percent following enactment of the 2003 Civil Service Law. In early 2006, average wages were increased by as much as 12.3 percent. 39 Wage bill to GDP of African countries is not very high on average because public employment in the African continent (estimated at 2 percent of the population) is lower than in any other region. 24 Figure 24: Average public salary (as a multiple of GDP per capita) in the Palestinian territories compared to MENA and most other regions Average public pay as a multiple of per capita GDP Africa Palestinian territories MENA Asia LAC OECD ECA 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Source: World Bank staff calculations and World Bank (1999). 2.25. Public sector pay is high relative to private sector comparators. Data from labor force surveys carried out by the Palestinian Central Bureau of Statistics (PCBS) since 2000 indicate that a significant wage gap has emerged between public and private workers, to the detriment of the latter. In 2000, almost no gap existed between public and private pay in the West Bank. In 2003, however, the wage gap favored workers in the public sector. The wage gap continued to grow until 2006 and then decreased for a couple of years. However, it resumed its growth in 2009. In 2013, average wage in the West Bank was 18 percent higher in the public sector than in the private sector. In Gaza, public sector employees earned a higher premium relative to their private sector counterparts in the period between 2000 and 2013. The premium grew from an average of 28 percent in 2000 to almost 50 percent in 2013. Figures 21-24 show how the gap has evolved over time. Gaza’s bigger wage gap probably reflects an economy with limited formal employment and a large informal sector paying much lower wages. Figure 25: Wage gap in the West Bank (%) Figure 26: Wage gap in Gaza (%) 25 70 20 60 15 50 10 40 5 30 0 20 -5 -10 10 -15 0 Source: World Bank staff calculations using PCBS Labor Force Surveys. 25 Figure 27: Wage gap at different quintiles in Figure 28: Wage gap at different quintiles in the West Bank Gaza Source: World Bank staff calculations using PCBS Labor Force Surveys. 2.26. The public sector wage premium is highest for those at the lower end of the pay scale and is not related to workers’ attributes. Unsurprisingly, as Figure 27 and Figure 28 show, the public sector wage premium is highest at the lowest end of the pay scale, drops for higher grades, and even reverses for those at the 95th percentile of the pay scale in the West Bank. Notably, public and private sector workers’ attributes such as education, age, gender, and marital status are converging. This suggests that the main explanation for the wage gap is not that public workers are better educated or more experienced than those in the private sector, but simply that the public sector pays more than the private sector would to workers with similar skills for similar levels of responsibility. 2.27. Inflation has caused real public remuneration in the West Bank to decline slightly over the last decade while real public wages in Gaza have increased. The mechanism used by the PA for adjusting public service pay to inflation is an annual Cost of Living Adjustment (CoLA), agreed with the unions. The CoLA raises basic pay scales, which each year move upwards to maintain the same relativities within and between scales. World Bank staff calculations using public pay data from labor force surveys indicate that between 2000 and 2013, average growth of nominal public remuneration in the West Bank was lower than inflation, resulting in an average real wage decrease of about 0.2 percent during this period. The situation in Gaza, however, was different: real wages increased on average by 1.4 percent between 2000 and 2013. The different trends in real public remuneration in the West Bank and Gaza are understandable given that inflation in Gaza was lower on average than in the West Bank over the last decade. 26 Figure 29: Real daily public pay in the West Figure 30: Real daily public pay in Gaza, 2000- Bank, 2000-2013 2013 16 80 16 80 NIS NIS 14 70 14 70 12 12 60 60 10 10 8 50 8 50 6 40 6 40 4 30 4 30 2 2 20 20 0 0 -2 10 -2 10 -4 0 -4 0 Real daily wage Real daily wage Inflation Inflation Growth in nominal daily wage Growth in nominal daily wage Sources: MoF, PCBS, and World Bank staff calculations. 2.28. The system of automatic pay increments and promotions has driven public remuneration growth. Although an annual staff performance appraisal system is in place, it is only loosely applied. Almost all personnel were awarded the “good” or “very good” assessments needed to receive an annual increment and become in due course eligible for promotion, even if this does not reflect actual performance. The PA effectively has an automatic increment system in which all personnel’s basic salaries move up by 1.25 percent each year, as do allowances that are base pay-related. Since the civil service has been expanding strongly, retirees are only a small fraction relative to the number of new entrants each year (Figure 31). This suggests that the PA’s present system of de facto automatic increments alone causes the total wage bill to grow by at least 1 percent per annum. This would happen even if the total number of civil servants was held constant. Figure 31: Breakdown of PA civil servant employees by seniority 40,000 35,000 Total Number of Civilian Staff 30,000 Ministry of Education 25,000 Ministry of Health 20,000 Rest of the Civilian Staff. 15,000 10,000 5,000 0 Source: General Personnel Council. 2.29. The current performance review system needs to be reformed. A better system would base the annual performance review processes on individual assessments linked to unit work plans that are derived from departmental plans. In this type of modern performance review system, rewards usually take the form of one-off bonuses rather than permanent increments up a scale to avoid a sense of entitlement. At present, introducing such a system in the Palestinian territories would be difficult. This would first require 27 fundamental reforms of salary scales in addition to many others outside the range of practical measures the government can take now. In the meantime, one way to improve the current system and curb the growth of the wage bill from increments would be to institute a more rigorous performance review process (even if still trait-based) and to signal this to personnel. 2.30. Unions’ bargaining power has also led public remuneration to grow beyond affordable levels. Public sector unions have been requesting higher CoLAs in addition to increases in fixed allowances. As a result of mounting pressure and ongoing strikes, the PA agreed in 2014 to increase retroactively the 2013 CoLA from 0.75 percent to 2.78 percent as of October 2013. In addition, pressure has been building for larger adjustments in areas like health, where the MoH competes with regional markets for skilled Palestinian health care professionals, and has sought a 35 percent across-the-board increase in health workers’ pay. Teachers have been pressing for more pay and a 20 percent increase has been agreed to be gradually implemented over the coming two years. The MoF has not yet fully implemented these agreements because of lack of funds, so strikes continue. The MoF, for its part, argues that with a high level of government arrears to the private sector and a wage bill that absorbs the bulk of revenues, the PA cannot afford to increase public servants’ wages and allowances. 2.6 Summary of findings and ingredients for controlling the PA’s wage bill 2.31. At the current level, the PA’s wage bill is clearly unsustainable. In relative terms (as a share of GDP, recurrent expenditures, and revenues), the PA’s wage bill is one of the largest of any central government. Although the aggregate wage bill of local governments is much smaller, taken together, the wage bill of the general government in the West Bank and Gaza is amongst the largest in the world. Notably, the wage bill analyzed in this PER does not include personnel on the payroll of the de facto authority in Gaza—the inclusion of which would probably increase the current estimate from 16 percent of GDP to 19 percent. Given that roughly a third of the PA’s expenditures are financed by donor aid and that it still has a financing gap in its budget, which it finances by accumulating arrears to the pension fund and the private sector, the wage bill is clearly unsustainable and efforts must be made to reduce it. 2.32. The very high average salary paid to employees on the PA’s payroll is the main reason behind the exorbitant wage bill. With the exception of Africa, where government employment is on average only a third of that in the West Bank and Gaza, the PA pays higher wages than any other region in the world relative to the size of its economy (measured as average wage relative to GDP per capita). While the cost of living in the Palestinian territories seems to be slightly higher than in other lower-middle-income countries, which may partially explain its high average wage in the public sector relative to other countries, government policies and wage bill management practices are the most important factors. 2.33. Across-the-board pay increases in 2004 and 2006, coupled with generous annual CoLAs and step increases, are the main reasons behind a relatively high average wage for PA personnel. Following the enactment of the 2003 Civil Service Law and prior to the 2006 elections, the PA raised the average wage by 8.4 percent in 2004 and 12.3 percent in 2006. Annual increases in the form of the CoLAand the mandatory 1.25 percent step increase granted annually to all PA personnel also contributed to wage growth. 2.34. Unlike average salary, the overall size of public sector employment in the West Bank and Gaza is not large, but some sectors are nevertheless significantly overstaffed relative to comparator countries. The widely shared perception that public employment in the Palestinian territories is very high does not stand up to scrutiny. In fact, overall government employment in the Palestinian territories—even if one takes into account public sector employees hired by the de facto authority in Gaza —is around the international average relative to the population. In fact, relative to population, some OECD countries have more than twice as many public sector employees. Certain sectors, primarily security, education, and health, 28 do have excess staff relative to some acceptable international norms. For instance, the PA’s security sector, while not very large compared to other countries in the region, is very large in the broader international context. The number of professional and administrative staff in both the health and education sectors is high relative to other countries in the same income group. The number of teachers in government-run schools is high even relative to UNWRA-run schools in the Palestinian territories, while student performance as measured by standardized tests is worse than in UNWRA-run schools. Hiring freezes in these sectors would reduce the number of employees relative to population, while it is unlikely that those measures would have a negative impact on the quality of services provided. Furthermore, central management agency officials routinely cite mandate overlaps and organizational redundancies in government units. Based on this information, significant staff redundancies conceivably prevail in many government organizations. 2.35. The political economy of hiring pressures coupled with the implementation of strategic plans misaligned with available resources are the main cause of excessive hiring in certain sectors and organizations. One factor that has enabled the growth in public service numbers in the past decade is the inability of the central management agencies, principally the GPC and the MoF, to critically scrutinize the staffing demands of line ministries and agencies at budget time. The GPC sees its role primarily as a regulator, ensuring that ministries and departments comply with the Civil Service Law and regulations in the exercise of their delegated HR management responsibilities. The MoF deals with the consequences of wage bill expansion as best it can, but is often unable to stop political decisions on hiring made at the Cabinet level. Its focus therefore remains short term, one of finding the money for decisions on staffing already taken and endorsed by the Cabinet of Ministers. Thus, it tries to roll back implementation of CoLAs, postpone the filling of vacancies, or delay the payment of increased salaries that accompany seniority-based promotions. It does not have the tools to manage the wage bill in a more strategic way and lacks the institutional clout to stop wage bill-expanding decisions at source. 2.36. The reduction of the wage bill to a more sustainable level over the medium term will require three essential ingredients. First, staff increases, which have been the norm up until 2012, will have to be contained for several years. Empowering the MoF to determine the overall wage bill ceiling and its allocation among different units of government based on pre-agreed upon criteria, as well as authority to approve any new hiring, would strengthen the PA’s ability to control wage bill growth. Second, wage growth will have to be contained until the wage gap between the public and private sectors is substantially reduced and wage levels do not affect the PA’s ability to attract staff at any grade. Folding most allowances now paid on top of base salary into base salary would also enhance control over wage inflation. Finally, even if the PA implemented the above measures to contain the wage bill growth to a degree that seems socially and politically feasible, a significant reduction of the wage bill relative to GDP will take longer without strong GDP growth, as shown in Table 6. While GDP is constrained by the restriction regime, the PA should seek to accelerate structural reforms hoping to achieve a higher level of economic growth while awaiting a removal of the restrictions. 2.37. The success of the PA’s efforts to reduce the wage bill will, first of all, depend on strong political commitment. Thus, as the PA moves to develop a Medium Term Fiscal Strategy (MTFS), it should incorporate within that MTFS an explicit goal for the size of the wage bill, relative to both domestic revenues and GDP. In the long run, it is suggested the former should not exceed 35 percent, and the latter should not exceed 10 percent. In the medium term, goals of 60 percent and 12 percent should be the aim. These are not hard and fast targets, but useful beacons to chart a course to sustainability. 2.38. More specifically, the PA should institute improved hiring controls in all sectors and keep a net hiring freeze in the education, health, and security sectors. In the education sector, the PA should freeze hiring until the ratio of students to teaching and non-teaching staff reaches the levels of UNWRA- run schools, while at the same time conducting additional reforms to improve education outcomes. Similar measures should be made in the health sector to contain hiring and to bring the number of professional staff 29 more in line with other countries in the region, if not with peer lower-middle-income countries, at least in line with middle-income countries. The security sector is very large by international comparison and it would be worth the effort to assess whether staff reductions are feasible without compromising the security sector’s important mandate. Finally, the PA needs to assess critically other sectors’ staffing needs taking into account the budget constraint. To achieve this, a stronger partnership should be built between the GPC and the MoF, supported by the Council of Ministers. GPC’s capacity to scrutinize mandates, organizational structures, and staffing more affectively ought to be enhanced. A first step might be for the Council of Ministers to cancel all vacancies that have not been filled for over one year. 2.39. The PA should introduce differentiated wage growth controls to reduce the pay gap between public and private sector employees with similar qualifications and job responsibilities. A significant gap exists at the lower end of the pay scale between wages paid by the PA and those paid in the private sector for jobs that require similar qualifications. Given that job security is usually higher in the public than in the private sector, the wage premium paid by the PA on top of what the private sector pays is not warranted and distorts the labor market. It is recommended that the PA contain the growth of wages in the public sector—in particular those at the lower end of the pay scale—until they equalize with private sector equivalents. 2.40. The CoLA should be based on actual consumer price growth and not be subject to negotiation with trade unions. The aim of the CoLA is to prevent real wages from falling, which should be done by granting wage increases commensurate with consumer price growth. To ensure that the CoLA is not used as a bargaining chip by trade unions to increase real wages, the PA should institutionalize the use of PCBS data on the consumer price index (CPI) as the exclusive determinant of CoLA increases. The PA should be able to suspend the CoLA for a certain period of time, however, to bring its staff’s wages more in line with those in the private sector, as well as when funding is not available to implement the increase. 2.41. Finally, grading reforms are needed, particularly in the security sector. The ranks in the security sector are clearly inflated by international standards and it appears that promotions are used as an instrument for wage increases, which has resulted in a significantly larger officer corps than seen in other countries with efficiently managed security forces. A functional review should be carried out, starting in the security sector but also in other sectors where grade inflation is suspected. Based on the findings, the PA should implement reforms to align the hierarchical pyramids in the security and civil sectors with good international practices. 30 Table 6: Potential impact on the PA’s wage bill of various policy recommendations under baseline and optimistic growth projections, 2014-2018 2014 2015 2016 2017 2018 Policy recommendat ions Wage bill/GD Wage bill P # of Wage bill Wage Wage Wage Wage bill Wage Wage Wage US Wage (optimi employ US Wage bill/GDP # of bill US Wage bill/GDP # of US Wage bill/GDP # of bill US Wage bill/GDP # of million bill/GDP stic) ees million bill/GDP (optimistic) employees million bill/GDP (optimistic) employees million bill/GDP (optimistic) employees million bill/GDP (optimistic) employees Zero staff 172,71 growth in 1966.1 17.7% 14.3% 2035.7 16.9% 13.4% 176,437 2135.1 16.5% 12.1% 180,275 2241.9 16.3% 11.4% 184,228 2354.5 16.0% 10.8% 188,300 1 education sector Zero staff 173,76 1978.2 17.8% 14.4% 2060.6 17.1% 13.5% 178,568 2174.0 16.8% 12.3% 183,520 2296.1 16.6% 11.7% 188,621 2425.2 16.5% 11.1% 193,874 growth in 1 health sector Zero staff growth in civil 172,90 service outside 1971.1 17.8% 14.3% 2045.9 17.0% 13.4% 176,824 2151.1 16.6% 12.2% 180,864 2264.2 16.4% 11.6% 185,025 2383.6 16.2% 10.9% 189,311 1 health and education Wage freeze 174,16 for bottom 1929.1 17.4% 14.0% 1988.6 16.5% 12.9% 179,391 2051.3 15.9% 11.5% 184,772 2116.3 15.3% 10.6% 190,316 2183.4 14.9% 9.7% 196,025 6 50% of PA staff Reduction in 162,73 number of 1801.5 16.2% 13.1% 1783.2 14.8% 11.7% 161,449 1833.8 14.2% 10.4% 162,960 1913.6 13.9% 9.8% 166,050 2011.2 13.7% 9.2% 170,059 8 high ranks in security sector 31 Chapter 3: Health Expenditures 3.1 Health Sector Context 3.1. The Palestinian territories are experiencing an epidemiological transition, compounded by the health impact of the political, security, and economic situation. The burden of non-communicable diseases (NCDs) is increasing due to cardiovascular diseases, diabetes, and cancers. This disease burden contributes to cost escalation in the health sector and will necessitate a greater focus on health prevention, primary care, and integrated disease management. The political, security, and economic situation pose severe challenges to the Palestinian health sector. These include challenges related to health outcomes, service delivery, access to health services, quality of care, human resources for health development and planning, fiscal sustainability, governance and accountability, investment planning and management, and creating a long-term vision and strategy for the development of the sector. The recent conflict in Gaza has had severe negative consequences for the health of the population and severely exacerbated the fiscal problems in the health sector. The conflict has greatly worsened the situation and the PA is limited in its ability to pay for health services, pharmaceuticals, and recurrent expenditures and to meet immediate challenges in the health sector. 3.2 Health Expenditures 3.2. Health expenditure is increasing (Table 7). Between 2000 and 2012, total health expenditure more than tripled from US$126 million in 2000 to US$1.3 billion, corresponding to an increase in per capita health expenditure from US$126 to US$294. Public spending on health quadrupled from US$126 million to US$488 million and out of pocket (OOP) spending increased substantially from US$152 million to US$502 million. Expenditures of nonprofit institutions serving households (NPISH) also increased in the last decade from US$90 million to US$231 million. In addition, private insurance enterprises spent US$29 million in 2012, increasing from US$10 million in 2000. The contribution of Palestinians living abroad to health financing also increased from US$7 million in 2000 to US$11 million in 2012. Table 7: Trends in health financing, Palestinian territories, 2000-2012 (millions US$) Health 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Spending Total* 384 372 386 417 528 557 624 684 894 959 1,075 1,201 1,262 Public* 126 114 125 153 212 214 222 243 328 346 390 424 488 OOP* 152 154 134 151 193 190 215 293 328 403 440 518 502 Private insurance 10 8 8 8 10 12 15 14 29 13 27 18 29 enterprises* Non-profit institutions serving 90 89 98 98 109 121 134 114 189 168 196 228 231 households (NPISH)* Palestinians 7 6 21 7 5 20 38 21 20 29 23 13 11 living abroad* Health expenditure per capita (current US$) 125.9 118.4 119.7 125.8 155.0 158.9 172.8 183.9 233.6 243.7 265.5 288.1 293.9 Sources: Authors’ calculations based on PCBS & MoH (2014), (2012) and (2014). 32 3.3. Total health spending in Palestine (public and private) as a share of GDP is among the highest in the world. (Figure 32). Despite the slight decline in total health expenditure as share of GDP in recent years, overall health spending exceeds health expenditures in many advanced economies, such as France and Austria. This comparison should be interpreted with caution, however, because the PT, not being a sovereign state, does not have several line expenditures (e.g., military), which implies that total health expenditure as a share of GDP would be higher. Furthermore, the PT pays significantly higher prices for outside medical referrals and pharmaceuticals, which inflate overall health spending; the occupation and its implications on planning, management, and population mobility also significantly hinder the efficiency of the system, increasing the per unit cost of health care. Note finally that spending is largely driven by private expenditures, while the share of public expenditures is low. Figure 32: Total health expenditure as percentage of GDP (%), International Comparisons, 2012 20 percentage 15 10 5 0 United States Netherlands Palestinian France Austria MENA LMICs Territories Sources: World Bank (2014); and PCBS & MoH (2014). 3.4. Public spending on health as a share of GDP has increased over the last decade, but is still not high relative to other countries. Public spending on health increased from 3 percent of GDP in 2000 to 4.8 percent in 2012 (Figure 33). And while it is higher than the MENA average of 2.6 percent and LMIC average of 1.7 percent of GDP, it remains low and represents less than half of total spending. The largest component of health expenditures is out of pocket from households posing a significant barrier to access to care, particularly for the poorest. Figure 33: Public spending on health as a percentage of GDP (%), Palestinian territories 2000-2012 6 5.0 5.3 5.2 4.8 4.8 percentage of GDP 4.6 4.7 4.7 4.3 3.6 4.0 4 3.0 2.9 2 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Sources: Authors calculations based on PCBS & MoH (2014), (2012), and (2010). 3.5. Public spending on health relies heavily on increasingly unpredictable donor funding. While the PA has benefitted significantly from external aid provided by the international community, aid flows have shown large fluctuations. After peaking at 30 percent of the GDP in 2006, external support declined to 15 percent of GDP in 2010 and 9 percent in 2012. In the short term, donor aid is projected to decline further to 8.6 percent of GDP between 2014 and 2018.40 The heavy reliance on donor funding is a significant 40 See IMF (2014). 33 challenge for the Palestinian health sector and the unpredictable nature of the aid flow hinders the MoH’s ability to prepare and implement medium and long term health reforms and plan allocation of resources. Furthermore, the MoH faces the risk of running high arrears given that donor funding is an important means of covering recurrent expenditures. 3.6. Private expenditures comprise the bulk of health spending, most of which is OOP. The general impoverishment of the population makes it vulnerable to healthcare expenditures, and the scarcity of good quality government health services increases utilization of private, more expensive health providers; 61 percent of health expenditures are private, 65 percent of which (almost 40 percent of total) come directly OOP from households. Furthermore, while catastrophic expenditures, at all income levels, are an important driver of impoverishment, the poorest are at the greatest risk as they bear a higher share of OOP expenditures compared to their total share of income. The incidence of catastrophic expenditures41 nearly doubled between 1998 and 2007 and, while 11.8 percent of surveyed households fell into deep poverty in 1998 due to healthcare payments, 12.5 percent of households entered deep poverty for the same reason in 2006.42 3.7. Pharmaceutical expenditure is the main driver of OOP spending. In 2012, the expenditures on medical goods dispensed to outpatients (including pharmaceuticals and other non-medical durables, therapeutic appliances and other medical durables) accounted for 39 percent of all OOP spending. The other large expenditure category driving OOP expenditures is curative care, which was 39 percent of OOP expenditures in 2012, declining from 53 percent in 2000. Twenty two percent of OOP curative care expenditures go to outpatient care and 16 percent to non-class care, whereas, inpatient care represented only 0.4 percent of all OOP expenditures, reflecting the dearth of private tertiary facilities in PT. Figure 34: Composition of OOP expenditures, Palestinian territories, 2012 0.5% 13.9% Health Administration and Health Insurance 22.3% 1.3% Prevention and Public Health Services Medical Goods Dispensed to Out-Patient Ancillary Services to Health Care 16.1% Inpatient Long-Term Nursing Care Non-Class Curative Care 0.1% 38.5% Outpatient Curative Care 7.2% Inpatient Curative Care Source: PCBS (2014). 3.2.1 Sources of Funding 3.8. Palestine mobilizes resources from public, private, non-profit institutions and Palestinians living abroad (Figure 35). Households’ out of pocket spending of 39.8 percent of total health expenditure in 2012 is higher than many countries in MENA (e.g., Jordan: 28.5 percent, Tunisia: 35.5 percent) and the overall MENA average of 35.5 percent. High OOP spending indicates that Palestinians are not protected from financial shocks due to health events. The public sector, comprised of central, state and local government units, as well as social security funds administered by these units, is the second main source of health financing with 38.7 percent of total health expenditures in 2012. Public funds for health are generated through general taxes raised by the MoF and the Government Health Insurance scheme (GHI). Non- governmental organizations financed 18.3 percent of total health spending in 2012. NGOs can be 41 Greater than or equal to 40% of total household expenditures (net of food expenses) on healthcare. 42 See Mataria (2010). 34 categorized in three broad categories depending on how they provide services: (1) directly to businesses (e.g., chamber of commerce); (2) in partnership with the public sector (e.g., government owned hospital); and (3) directly to the households such as charities, trade unions, professional unions, churches, and privately financed aid organizations. Private insurance enterprises account for 2.3 percent of total health spending and funds from Palestinians living abroad account for less than 1 percent. Over the last decade, as NGO expenditures on health have declined, public expenditures have increased to fill the gap. Figure 35: Source of health financing as share of total health expenditure, Palestinian territories, 2000-2012 Public Expenditure on Health OOP Private Insurance Enterprises NPISH Palestinians living abroad 50 percent of total health 40 expenditure 30 20 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Sources: PCBS & MoH (2014), (2012), and (2010). 3.9. Public Health financing in PT relies on a combination of global and line item budgeting based principally on historical costing. Typically, systems with this financing mechanism are incentivized to (1) increase inputs (beds, equipment, staff, etc.); (2) underprovide services; and (3) spend all funds by the end of the budget cycle. International evidence and experience indicates that these systems offer little incentive to improve the efficiency of the input mix, quality or access. 3.10. The implementation of the MoH budget is undermined by discrepancies between approved and actual expenditures (Figure 36). Since 2008, the Palestinian territories have employed an integrated planning and budget framework.43 During key informant interviews, the MoH officials indicated that the Ministry of Health did not receive the entirety of its approved budget (typically receiving 1/3 to 1/2). Recent data shows that there are sizable discrepancies between the approved and actual MoH expenditures; in 2011, the MoH was approved to receive US$346 million while actual spending was around US$257 million. In 2012, the gap between the approved MoH budget and actual MoH expenditures narrowed to US$36 million but then increased to US$40 million in 2013. The discrepancies between approved and actual expenditures hamper recurrent expenditure planning and medium-term investment prospects. Furthermore, unpredictable revenues undermine the MoH’s credibility as a purchaser of services (from other service providers) and medical supplies (e.g., drugs and disposables). 43 At the budget preparation stage, the MoF shares a budget circular among line ministries, including the MoH, and requests a proposed budget for the upcoming fiscal year. The Central Budget Department in the MoF compares proposals received from each line ministry with historical figures and proposes an overall budget for Cabinet approval. Following the approval of the Cabinet, line ministries are responsible for allocating resources from the approved budget. 35 Figure 36: Approved MoH budget vs. actual MoH spending in US$ millions, Palestinian territories, 2010, 2011, 2012 and 2013 Approved MOH budget Actual MOH Expenditure Approved Budget vs. Actual Expenditure Gap 500,000,000 400,000,000 300,000,000 200,000,000 100,000,000 0 2010 2011 2012 2013 -100,000,000 -200,000,000 Source: MoH (2014). Note: The task team noted discrepancies between expenditure figures reported by the Ministry of Health and the Ministry of Finance. Moreover, there were inconsistencies in some spending figures reported by the MoH in different reports. 3.11. MoH spending has more than doubled in the last decade. Following the 2nd Intifada, MoH spending showed a two-year contraction from US$95 million in 2000 to US$88 million in 2002. But between 2003 and 2013, it increased from US$122 million to US$286 million, despite a small decrease between 2012 and 2013 (Table 8). Table 8: MoH expenditure, Palestinian territories, 2010-2013 (in US$ millions) 2010 2011 2012 2013 Recurrent Salary Expenditure 91 159 164 176 Recurrent Non-Salary Expenditure 104 97 129 108 Pharmaceutical Expenditure 50 48 70 47 Medical Referrals 46 40 49 52 Other 8 9 10 9 Capital Investment Expenditure 2 1 1 1 MoH expenditure as % of GDP 2% 3% 3% 3% Total MoH Expenditure 197 257 295 286 Source: MoH (2014). Notes: (i) Pharmaceutical expenditures include spending on medical equipment and spare parts. (ii) The World Bank team noted significant inconsistencies between MoH spending reported by the Ministry of Health directly to the task team and expenditures reported in the Ministry’s Annual Health Reports. The table above reports figures received directly from the MoH as these figures included more detailed breakdown of MoH expenditures for different spending categories. 3.12. MoH expenditures have been responsive to the changes in GDP. While, both GDP and MoH spending declined following the 2nd Intifada, MoH spending contracted at a much slower rate (59 percent less rapidly per year) compared to GDP (elasticity of 0.41). MoH expenditures surged quickly in the following period and grew 39 percentage points more rapidly than GDP per year (elasticity of 1.39) between 2003 and 2005. However, the increase in the MoH expenditures compared to GDP has slowed down since 2005, growing 28 percent less rapidly than GDP (elasticity of 0.72). 36 3.13. The main drivers of the MoH spending are salaries, medical referrals to providers outside of the public health system, and pharmaceutical expenditures44 (Figure 37). Salaries represented 62 percent of all MoH spending in 2013, an increase from 46 percent in 2010. Recurrent salary expenditures nearly doubled from US$91 million in 2010 to US$176 million in 2013. Mirroring the increase in the volume of referrals, MoH expenditures on OMR reached US$52 million in 2013 and now account for 18 percent of overall MoH expenditure and 48 percent of all non-salary recurrent spending. Pharmaceutical expenditures were US$50 million in 2010 and US$47 million in 2013. Figure 37: Distribution MoH expenditures, Palestinian territories, 2013 3.3% 0.3% Recurrent Salary Expenditure Pharmaceuticals 18.1% Referrals 16.5% 61.7% Other Recurrent Non-salary Expenditure Capital Investment Expenditure Source: MoH (2014). 3.14. Public expenditures are mainly allocated to curative care (Figure 38). Health expenditure data disaggregated by function indicate that the Palestinian health system is geared towards inpatient care at hospitals. In 2012, 82 percent of all public expenditures on health were allocated to curative care, while only 14 percent went towards preventative care and public health services; 66 percent of curative care expenditures were for inpatient care and 34 percent for outpatient care. While these figures should be interpreted with care due to issues rising from definitions and measuring of these expenditure categories, given the rising burden of non-communicable diseases in the Palestinian territories, there is significant room for improving the allocative efficiency of the Palestinian health system by shifting the allocation of resources toward primary care, preventative care and disease prevention. Many NCDs can both be prevented, through cost-effective interventions, and effectively managed at the primary health care level. Figure 38: Public expenditure on health by function of care, Palestinian Territories, 2000-2012 600,000 Inpatient Curative Care 500,000 Outpatient Curative Care Non-Class Curative Care in US$ thousands 400,000 Rehabilitative Care Inpatient Long-Term Nursing Care 300,000 Ancillary Services to Health Care Medical Goods Dispensed to Out-Patient 200,000 Prevention and Public Health Services 100,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Sources: PCBS & MoH (2014), (2012), and (2011). 44 The task team noted inconsistencies between expenditure figures reported by the MoH directly and figures published in the Annual Health Report of the MoH. The team reports figures from the former as they present a more detailed breakdown of expenditures. 37 3.2.2 Salaries 3.15. The health sector wage bill has been increasing (Figure 39). The Ministry of Finance reports the health sector wage bill has increased from US$97.8 million in 2006 to US$176.8 million in 2013. The overall wage bill increased in this period, despite significant fluctuations in the MoH expansion rate of the workforce. For example, the largest increase in the wage bill took place between 2007 and 2008, increasing from US$116.7 million to US$147.8 million when the expansion rate of the MoH workforce was slowing compared to the previous years. The health sector wage bill increased from US$156.7 million in 2010 to US$170.5 million in 2011 (9 percent) while the MoH workforce contracted by 0.6 percent. Thus, in addition to staff increases, salary raises have also contributed significantly to the growth of the PA financed health sector wage bill. Figure 39: MoH wage bill, Palestinian territories, 2006-2012 Health Sector Wage Bill (right axis) Rate of expansion in MOH employment (left axis) 10% 180 160 5% 140 120 0% 100 -5% 80 2006 2007 2008 2009 2010 2011 2012 Source: MoF (2014). 3.16. There are discrepancies in salary expenditure reporting (Table 9). The MoH and MoF report different figures for recurrent salary expenditure. The MoH reports that salary expenditure was US$91 million in 2010 while the MoF records indicate US$157 million. Similarly, the former reported that in 2011 salary expenditure was US$159 million, while the latter showed that it was around US$170 million. The discrepancies persisted in 2012, though narrowing considerably. These discrepancies undermine the Ministries’ ability to estimate accurately the recent dynamics in salary expenditure and the share of salary spending with respect to the overall MoH expenditure and hinders the MoH’s ability to formulate information based policy. Table 9: Discrepancies in salary expenditure reporting, Palestinian territories, 2010-2013 In US dollars (millions) 2010 2011 2012 Ministry of Health 91 159 164 Ministry of Finance 156.7 170.5 168.8 Sources: MoH (2014); and MoF (2014). 3.2.3 Outside Medical Referrals 3.17. The lack of availability of certain treatments, medications, medical staff, equipment, and infrastructure led to the creation of an outside referral system. In the PT, the lack of availability of certain treatments, shortage of medications, and inadequate staff, equipment, and infrastructure, compromise the MoH’s ability to provide services. This has resulted in a referral system where a large volume of patients requiring tertiary-level care are referred to a mix of private (for profit) and non-profit service providers outside of the public health system. Currently, the MoH lacks a clear strategic decision- making mechanism to define which services it would like to invest internally and which services it would 38 like to purchase from other service providers within the Palestinian territories, as well as neighboring countries. 3.18. The number and cost of referrals has increased dramatically over the last decade. The number of patients referred to outside providers increased from 8 thousand in 2000 to 62 thousand in 2013 (44 thousand from the West Bank and 18 thousand from Gaza). Eighty-two percent of patients were referred to health facilities operating in Jerusalem, the West Bank, and Gaza while the remainder were sent to hospitals in Israel, Egypt and Jordan. The pattern of patient flows varied significantly between the West Bank and Gaza. In 2013, 47 percent of patients in the West Bank were referred to the six NGO hospitals in Jerusalem, 43 percent to private and NGO hospitals within the West Bank, 10 percent to hospitals in Israel, and less than 1 percent to providers in Jordan and Egypt. In Gaza, 34 percent of patients were referred to Jerusalem hospitals, 22 percent to Israel, 16 percent to Egypt, 14 percent to NGO hospitals in Gaza, and 13 percent to the West Bank. Between 2000 and 2013, the cost of referrals increased from US$8.5 million to US$144.4 million. 3.19. The pattern of referrals has been changing. Since 2010, the number of patients referred to Egypt and Jordan declined, while referrals to Israel have increased. In 2010, Egypt received 6 percent of all referral patients in the PT, the majority from Gaza. Similarly, Jordan was among the top destinations for referrals, especially from the West Bank, receiving 6 percent of referral patients in the PT in 2010. Due to significant delays in payments in recent years, however, Jordan has almost halted receiving Palestinian patients and, in 2013, only 0.4 percent of all referral cases were to Jordan. Conversely, in 2013, 13 percent of all referral patients were sent to hospitals in Israel, doubling from 7 percent in 2010, driven by an increased number of patients referred from the West Bank. Between 2010 and 2013, the number of referrals from the West Bank to Israel nearly quadrupled while, in the same period, the number of referrals from Gaza to Israel increased by 37 percent. In the last decade, Jerusalem hospitals have played an increasingly important role in providing services to referral patients. In 2013, they received 44 percent of all referrals in the PT, compared with 19 percent in 2003. A recent World Bank study indicated that this changing pattern in referrals in recent years is driven by a number of factors, including (i) the impact of the Israeli occupation; (ii) mobility restrictions imposed by Israel (especially for patients residing in Gaza); (iii) perceived quality of care and cost concerns; and (iv) the social value attributed to the East Jerusalem hospitals as resilient Palestinian institutions.45 3.20. Noncommunicable diseases account for the majority of referral expenditures (Figure 40). The cost of referrals associated with tumors has been increasing. It rose from US$16 million in 2009 to US$28 million in 2013 (US$20 million in the West Bank and US$8 million in Gaza), corresponding to 19 percent of overall referral expenditures. The referral expenditure due to blood diseases is also high. In fact, it more than tripled between 2009 and 2013 from US$5 million to US$16 million, representing 11 percent of the overall referral bill. Neurological surgery accounts for 7 percent of the total referral expenditures. Human resource planning is a crucial element in addressing outside medical referrals (OMR) and the PT has shortages in certain sub-specialties, including hematology, oncology and cardiology. In the West Bank, there were only 5 hematologists, 6 cardiologists, and 6 oncologists in 2013. 45 See World Bank (2013). 39 Figure 40: Top 10 costliest referral procedures, Palestinian territories, 2013 30 West Bank Gaza in US $ millions 20 10 0 Tumors Blood diseases Neurological Child diseases Cardiac Cardiovascular Cardiology cases Internal Orthopedic General surgery surgery catheterization surgery medicine surgery Source: MoH (2014). 3.21. The West Bank spends 2.5 times more on referrals than Gaza (Figure 41). The geographical distribution of expenditure on referrals has seen a shift in recent years. In 2009, the cost of referrals was US$75.7 million, US$49.4 million of which was incurred by the West Bank and US$26.3 was by Gaza. While spending associated with outside referrals increased both in the West Bank and in Gaza, it has grown 74 percent more rapidly in the West Bank than Gaza between 2009 and 2013. In 2013, the cost of referrals from the West Bank was 2.5 times higher than the cost of referrals from Gaza (US$102 million in the West Bank and US$42 million in Gaza). The geographic disparity in expenditures per capita reflects the inequitable process and approval system associated with referrals. Figure 41: Trends in medical referrals, Palestinian territories, 2000-2013 West Bank Gaza Cost of Referrals (right axis) number of referred 80,000 200,000,000 patients in US dollars 60,000 150,000,000 40,000 100,000,000 20,000 50,000,000 0 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Source: MoH (2014). Note: No data available for referrals from Gaza for 2000-2002 & 2004. 3.22. Israeli hospitals represent a high cost burden relative to the number of treated patients (Figure 42). Jerusalem hospitals, despite being the top destination for outside medical referrals (44 percent of referral cases in 2013), accounted for 34 percent of overall referral expenditures while Israeli hospitals saw 13 percent of referral cases but represented 33 percent of the total referral expenditures. The hospitals operating in the West Bank are the second destination for referrals (34 percent of referral cases in 2013) and represent 28 percent of expenditures. While Israeli hospitals do see more complex cases, these do not account for the observed price differential; a study by USAID revealed that over half of the referral bills from Israeli hospitals are either unauthorized or have expired. Israeli hospital fees are deducted directly from Palestinian tax revenue often without Palestinian authorization. The MoH has launched an audit of Israeli medical referrals to recover charges that have been improperly levied by these hospitals. It has also signed MoUs with several Israeli hospitals in an effort to reduce referral costs. 40 Figure 42: Cost and volume of referrals by service provider, 2013 Cost of referrals % of total referral cases (right axis) 60 50% 50 in US$ millions 40% 40 30% 30 20% 20 10 10% - 0% Jordan Jerusalem Israel Egypt West Bank Gaza Source: MoH (2014). 3.23. A large proportion of referral services are used by the enrollees of the non-contributing health insurance scheme with important consequences for fiscal sustainability.46 In 2011, 35 percent of all outside medical referrals were used by the enrollees of the non-contributing health insurance scheme47 and 8 percent was used by the members of the Al-Aqsa scheme. The enrollees of the compulsory health insurance scheme, public sector employees and their dependents, represent 15 percent of all referral cases. In addition, 14 percent of all referrals were used by members of trade unions (Figure 49). Figure 43: Utilization of referral services per insurance scheme, Palestinian territories, 2011 Compulsory 1.8% 4.4% Social affairs 7.9% 15.4% Voluntary 12.0% Workers in israel 2.1% Group contracts 34.8% Trade Unions 14.1% 5.2% Free of charge 2.4% Intifada Min of Prisoners Others Source: WB (2014). 3.24. Persistent weaknesses in planning, organization, management, and financing of outside medical referrals undermine the fiscal sustainability of the health sector. A recent World Bank study, which audited the appropriateness of OMR, concluded that a significant majority of the referral documents 46 See section 5.4 Government Health Insurance. 47 MoH acts as the last resort for the uninsured through the Al-Aqsa scheme. 41 lacked accurate and readable information. Approximately 90 percent of all audited referral files lacked a reason for referral; 73 percent missed a clinical summary and 35 percent lacked an explanation of the initial diagnosis. In documents where the diagnosis was available, it consisted of general symptoms, such as chest or abdominal pain, rather than specifics. Furthermore, 31 percent of the referral files lacked sufficient information to determine whether the case was referred for inpatient or outpatient services. While the findings pertaining to medical information/justification for referral might be partially justified by weak diagnostic capacity of the referring hospitals, the administrative data were attributed to organizational and managerial deficiencies. 3.2.4 Pharmaceutical Expenditures 3.25. Pharmaceutical expenditures accounted for 47 percent of non-salary public expenditures in 2013. Expenditure on drugs has shown large fluctuations, increasing by 29 percent between 2010 and 2011 mirroring, and partially driving, the increasing overall public expenditures in health (Figure 44). Additionally, there is a discrepancy between the cost of drug tenders and overall pharmaceutical spending. For example, tender costs were US$37 million and overall pharmaceutical expenditures US$50 million in 2010, whereas in 2011 the values were US$59 million and US$48 million respectively. The differences arise due to a number of factors including: (1) pharmaceutical expenses include expenditures additional to drugs, including medical equipment; (2) the full amount of the tender is not often delivered; and (3) poor records and expenditure reporting lead to data discrepancies. Figure 44: Drug tender cost, overall pharmaceutical spending and total health spending, Palestinian territories, 2010-2013 Value for the Annual Drug Tenders Total Pharmaceutical Spending Total Public Health Spending 400 in $US millions 300 200 100 0 2010 2011 2012 2013 Source: MoH (2014). 3.26. Total withdrawals from the CMS totaled US$62 million in 2013 (Table 10). Expired drugs cost the PT US$2 million. Hospitals accounted for US$32.5 million of withdrawals and PHCCs (including warehouses in Nabulus and Gaza) accounted for US$29.1 million, the remainder being withdrawn by other Medical Center Units. There was a wide variation in withdrawals between hospitals and also within PHCCs (Figure 41). The top three hospitals in pharmaceutical consumption are National hospital, Beit Jala, and Ramallah and the top three PHCCs are Ramallah, Hebron and Nabulus48. 48 The distribution of withdrawals is only known for Gaza Warehouse not within Gaza PHCCs. 42 Table 10: Pharmaceutical Withdrawals from CMS, 2013 Classification Expenditure Cost of expired units Gas 61,755 205 Lab Materials 4,797,274 168,365 Drugs 53,250,971 1,837,842 Vaccines 4,011,749 243 Total 62,121,748 2,006,655 Source: MoH (2014). Figure 45: Withdrawals from CMS by hospital and PHCC, 2013 10 9 in $USD, millions 8 7 6 5 4 3 2 1 0 Emergency and Safe Child… National center for chronic… Darwish Nazzal (Qalqilia) National Blood Bank - Ramallah Gaza warehouses Northern Hebron warehouse Jenin Southern Hebron Bethlehem Ibn Sina Institute Qalqilia Ramallah Hebron Jericho Toubas Nablus Salfeet Tulkarem Jerusalem General Health Lab Bethlehem National Rafidia Jenin Yatta Ramallah Jericho Salfeet Emergency Beit Jala Tulkarem Hebron Hospital PHCC Source: MoH (2014). 3.27. The MoH pays relatively high prices for pharmaceuticals, but import restrictions and large arrears owed to suppliers hamper its ability to negotiate better prices. Pharmaceutical companies factor in the cost of delayed payments in their pricing. In addition, MoH officials and NGO representatives claim that restrictions on pharmaceuticals that can be imported limit the choices available.49 The review compared the top 20 tenders by cost from the 2013 CMS list with international benchmark prices based on the WHO’s International Drug Price Indicator Guide [33]. Of the top 20 expenditure items, 6 were in the WHO index50 and reflected a variety of price differentials with the PT typically overpaying for drugs (Table 11). Table 11: Comparing drug tender costs with international benchmark prices Unit Price Benchmark Price Percent Over-paid Human albumin serum 20% 50ml 36.50 12.00 204.1 Imatinib 400mg tab 123.13 79.32 55.2 Imatinib 400mg tab 35.32 79.32 -55.5 Rituximab 500mg /50ml vial 909.88 206.75 340.1 Tacrolimus 1mg cap 2.60 2.60 0.2 Tacrolimus 1mg cap 2.86 2.60 10.1 Sources: MoH (2014); and WHO & MSH (2010). 49 It was not possible to verify independently these claims 50 The review notes that Imatinib and Tacrolimus appear twice due to multiple tender costs. 43 3.2.5 Arrears 3.28. In 2013, arrears corresponded to more than half of MoH’s spending. The MoH reported that in 2013, arrears due to spending on pharmaceuticals and consumables, referrals and capital and other running expenditures reached US$193 million, corresponding to 54 percent of MoH’s actual spending.5152 More specifically, almost half of MoH arrears, excluding salaries, were due to spending on pharmaceuticals and consumables. Arrears due to referrals represented 44 percent of MoH arrears, excluding salaries, and reached US$85 million. Lastly, arrears from capital and other current expenditures were almost US$13 million.53 Figure 46: MoH arrears, Palestinian territories, 2013 7% Purchasing Service 44% Pharmaceuticals, consumables 49% Capital and other running expenditures Source: MoH (2014). 3.3 Government Health Insurance 3.29. The Government Health Insurance (GHI) scheme was established in 1994 in order to provide health insurance coverage to the Palestinian population. Administered by the Ministry of Health, GHI provides coverage to several population groups: (i) public sector employees on a compulsory basis; (ii) private individuals and households on a voluntary basis; (iii) Palestinian workers in Israel on a compulsory basis; (iv) businesses and employers (contract groups); (v) eligible Palestinians through the Ministry of Prisoners Affairs; and (vi) social hardship cases. In addition to these population groups, the MoH acts as the insurer of last resort through the Al-Aqsa scheme and all Palestinians residing in Gaza are covered by the GHI scheme free of charge.54 All Al-Aqsa enrollees must be identified as “unemployed” by the Ministry of Labor. The benefits package is comprehensive and includes maternal and child health services, primary health care, prescription drugs, as well as services purchased from non-MoH providers such as East Jerusalem Hospitals or hospitals operating in neighboring countries (e.g., Egypt and Israel). 51 This calculation is based on expenditure figures reported in the MoH Annual Health Report 2013. The MoH’s total expenditure in 2013 used for this calculation differs from the figure listed in Table 5 due to the noted inconsistencies between different reports. The team decided to use this figure in order to maintain the internal consistency of the calculation. 52 See MoH, Health Annual Report Palestina 2013, June, 2014. 53 See MoH, Health Annual Report Palestina 2013, June, 2014. 54 See ibid. 44 3.30. The GHI functions primarily as a list of beneficiaries and revenue collection mechanism with no purchasing function. There is no connection between health insurance and provider payment, and the health insurance department at the MoH does not function as a traditional insurer. Rather, it functions simply to catalogue beneficiaries under various schemes and to collect copayments from them, where applicable. 3.31. The number of non-contributing GHI enrollees increased in the last decade compared to the number of contributing members, damaging the scheme’s finances. In 2000, the majority of GHI enrollees were public employees covered on a compulsory basis, followed by contract groups, and workers in Israel. Between 2000 and 2007, the number of Al-Aqsa enrollees (non-contributing members) increased while the number of insured Palestinian workers in Israel declined significantly. The share of Palestinian workers in Israel as percentage of overall GHI enrollees declined from 11.3 percent to 4.8 percent while, concurrently, the number of Al Aqsa enrollees doubled between 2009 and 2012 from six thousand families to eleven thousand families. In 2013, Al Aqsa enrollees represented 7.4 percent of all GHI enrollees, compared to 5.2 percent in 2009. On the other hand, the number of GHI enrollees covered by the Social Welfare declined considerably from thirty seven thousand families to twenty eight thousand between 2012 and 2013. In 2013, public sector employees accounted for 40.3 percent of all GHI enrollees, followed by group contracts (19.2 percent), hardship cases receiving social assistance (18.6), and workers in Israel (4.8 percent).55 Figure 47: GHI enrollees by type, West Bank, 2009-2013 Voluntary Compulsory Workers in Israel Group Contracts Ministry of Prisoners Affairs Social Welfare Free of Charge (Al Aqsa) number of households 180,000 130,000 80,000 30,000 -20,000 2009 2010 2011 2012 2013 Source: MoH, (2009-2014) Notes: (1) The MoH Annual Report does not contain data from the Gaza Strip. (2) The population in Gaza Strip is covered by the GHI scheme for free.56 3.32. The GHI revenue collection mechanism is highly fragmented. A large proportion of the GHI revenues are pooled at first by the Ministry of Health. The MoF deducts the contributions of the public sector employees automatically from their salaries and makes annual transfers from the Ministry of Social Affairs budget to the GHI pool in order to cover the contributions of the social hardship cases. The MoF also receives the contributions of the Palestinian workers living in Israel through the Israeli Authority on a monthly basis. In addition, those who are enrolled in the GHI scheme on a voluntary basis pay their contributions to the health directorates every six months or once a year. The contributions of contract groups are received collectively through employers either on a monthly or on a yearly basis. 55 See MoH, Health Annual Report Palestina 2013, June, 2014. 56 Ibid. 45 Figure 48: GHI contributions by enrollee type, West Bank, 2012 Voluntary Compulsory Group Contracts Ministry of Prisoners Affairs Social Welfare Copayments 1.5% 17.5% 11.6% 47.3% 5.5% 16.5% Source: MoH (2014). Note: Contributions of Palestinian workers in Israel not included. 3.33. The imbalance between the GHI revenues and public expenditures on health is growing. In 2000, public expenditure on health was US$125 million while GHI revenues reached US$35 million.57 By 2009, public expenditure on health reached US$346 million, while GHI revenues declined to US$23 million. Since 2009, the imbalance between GHI revenue and public expenditure on health has been widening as GHI revenues have not been able to match the increase in public expenditure on health. In 2012, public expenditure on health reached US$488 million and GHI revenues were US$31 million, undermining the sustainability of health financing. Figure 49: GHI revenue and public health expenditure gap, Palestinian territories, 2009-2012 Public expenditure on health GHI Revenues Revenue-Expenditure Gap 500 346 390 424 488 in US$ millions 300 100 23 27 30 31 -100 2009 2010 2011 2012 -300 -500 (323) (363) (457) (394) Sources: MoH (2014), (2013), (2012), (2011) and (2010); and PCBS and MoH (2014) and (2012), MoH. 3.34. GHI expenditures have been increasing rapidly exacerbating the already constrained fiscal position. GHI expenditures, as percentage of total health expenditures, were 5.2 percent in 2000 and fluctuated between 6.2 percent and 4.7 percent until 2005. Starting from 2005, they have increased steadily and reached 15 percent of total health expenditures in 2010. This sizable increase in the GHI expenditures adds additional pressures to the already constrained fiscal position.58 57 See World Bank (2008). 58 See PCBS and MoH (2014). 46 Figure 50: GHI expenditure, Palestinian territories, 2000-2012 250 200 in US$ millions 150 100 50 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Sources: PCBS & MoH (2014), (2012), and (2011). 3.4 Health System Inputs 3.35. Health service delivery and governance in PT is fragmented. While interoperable health information systems, standards of care, and regulatory capacity of the MoH are of high importance, a committee tasked with overseeing this coordination was disbanded. Adding to the challenge is a lack of separation of functions in the federal health system: the MoH is payer, provider, and regulator of the health sector, a situation difficult to reform in the context of the political uncertainty. Furthermore, the division between Fatah and Hamas has resulted in the creation of two separate governance structures with separate visions and strategies for the development of the sector. The MoH in Gaza regulates the sector directly, involving little communication with the MoH located in the West Bank; the MoH departments are duplicated and administered by different personnel. Finally, there is little coordination and cooperation between the public, private for profit and NGO sectors. This disconnect results in a system where the MoH is steward of the public system only rather than the entire sector. It sees NGOs as competitors rather than collaborators, with planning, including allocation of infrastructure and health staff, being done without consideration of overall sectoral distribution and needs. 3.4.1 Infrastructure 3.36. In the Palestinian territories, health services are delivered by a complex network of service providers. The public sector, comprised of the MoH and the Palestinian Military Medical Services (PMMS), is the main service provider. The MoH has a network of 460 primary health care centers (PHCCs), (404 in the West Bank and 54 in Gaza in 2012), and 25 hospitals (12 in the West Bank and 13 in Gaza). The PMMS focuses largely on primary health care services through 23 PHCCs (16 in the West Bank and 7 in Gaza) and provides secondary and tertiary care services in 3 hospitals located in Gaza. NGOs have historically played an important role in service delivery: in 2012, NGOs operated 206 PHCCs (140 in the West Bank and 66 in Gaza) and 33 hospitals (19 in the West Bank and 14 in Gaza). UNRWA plays a critical role in providing services to registered refugee populations, mainly through primary health care clinics: UNRWA operates 61 PHCCs (41 in the West Bank and 20 in Gaza) and one hospital in the West Bank. Finally, there are 17 private hospitals, all located in the West Bank. Overall, the private sector in the PT is limited and mainly provides maternal and child health services. 3.37. Primary health care services are provided through four levels of health care centers ( Figure 51). Level I PHCCs serve up to 1,000 population and provide mainly mean corpuscular hemoglobin (MCH) and immunization services. The Level II PHCCs serve 2,000 to 4,000 population and provide curative care and laboratory tests, in addition to services provided at level I. Primary health care centers in Level III serve 6,000 to 12,000 population and provide specialized MCH consultations. Level IV PHHCs serve more than 47 12,000 population and provide psychological consultations, dental care, as well as radiology services including X-rays and ultrasounds. In addition to these primary health centers, 5 mobile clinics operate in the West Bank.59 Between 2010 and 2012, the MoH has upgraded 4 primary health care centers from Level I to Level II; and 5 from Level II to Level III. Moreover, the MoH has introduced quality standards for primary health care centers. However, the implementation of these standards across all PHCCs remains a challenge. Furthermore, despite the size and complexity of the primary care system, they remain underfunded (see section 3.2.1) and underutilized (see section 3.5.1) relative to tertiary care. 3.38. The geographic distribution of primary health care centers does not match the population size in the governorates (Figure 43 and Figure 44). There exists a disconnect between where public primary health care centers have been constructed and where Palestinians reside. Furthermore, while epidemiological data by governorate are not available, key informant interviews suggest that the distribution on PHCCs also does not match population health needs. In 2012, PHCCs operated by MoH provided services to 6 thousand patients on average. In the West Bank PHCCs provided services to 4 thousand patients on average, while in Gaza this figure was almost threefold higher.60 Within Gaza, the highest population to PHCC ratio was observed in the Gaza governorate (14,472 people per PHCC) and the lowest was observed in the Salfit governorate (2,389 people per PHCC). Figure 51: Geographic Distribution of Four Levels of PHCCs Operated by MoH, 2012 Level I Level II Level III Level IV Mobile Clinics Population per center (right axis) 100 50,000 80 40,000 60 30,000 40 20,000 20 10,000 0 0 Source: MoH (2013). Note: *denotes the J2 area. 59 See MoH (2013). 60 Ibid. 48 Figure 52: Geographic distribution of all primary health care providers, Palestinian territories, 2012 MOH NGOs UNRWA PMMS Population per center (right axis) 100 16,000 80 12,000 60 8,000 40 20 4,000 0 0 Source: MoH (2013). 3.39. The MoH is the main provider of secondary and tertiary care (Figure 53). In the PT there are a total of 79 hospitals. In the West Bank, the MoH operates 12 hospitals, which account for 44 percent of the hospital bed capacity. The Palestinian Medical Complex located in Ramallah and Rafidia hospitals in Nablus are the two biggest MoH hospitals. In addition, there are 19 NGO hospitals representing 39 percent of hospital bed capacity. The NGO sector plays an important role in service delivery, especially maternal health and rehabilitation services. All three rehabilitation hospitals in the West Bank are, in fact, operated by NGOs. The only UNRWA operated hospital in the Palestinian territories is located in the Qualqilia governorate. All 17 of the private hospitals in the PT are located in the West Bank, accounting for 15 percent of bed capacity and focusing mainly on maternal health. In Gaza, the Ministry of Health’s role in secondary and tertiary care is more pronounced and the MoH operates 13 hospitals and accounts for 68 percent of the total hospital bed capacity. The Shifa hospital61 located in Gaza is the biggest hospital in the territories and serves the population with 509 beds. There are 14 NGO hospitals in Gaza, which account for 25 percent of the hospital bed capacity. The majority of these hospitals are general hospitals, with one that focuses on rehabilitative care and two on maternal care. Lastly, the Palestinian Military Medical Services operates 3 hospitals in Gaza, which represents almost 8 percent of the bed capacity. 61 The World Bank team’s site visit found that Shifa hospital is sever ely affected by frequent electricity cuts, inadequate equipment and a dire shortage of drugs. 49 Figure 53: Distribution of hospitals by provider type, Palestinian territories, 2012 MOH UNRWA NGOs Private PMS Beds per 10,000 (right axis) 16 30 beds per 10,000 population 14 number of hospitals 25 12 10 20 8 15 6 10 4 2 5 0 0 Source: MoH (2013). 3.40. There are significant disparities in hospital beds available across governorates. Overall, there were 12.6 hospital beds per 10,000 population in 2012 (13.9 in Gaza and 11.8 in the West Bank) , higher than the MENA average of 10 beds per 10,000 population (Figure 20).62 Yet, there is a huge variation in this ratio: in Bethlehem, it is as high as 27.5 hospital beds (the highest in the PT), and as low as 5.4 beds in Deir Al Balah.63 Hospitals operated by different providers tend to be clustered in the same areas, possibly due to poor coordination between different service providers in facility planning for secondary and tertiary care. Figure 54: Distribution of hospital beds per 10,000 population, Palestinian territories, 2012 Jordan Gaza Strip West Bank Khan Younis Jerusalem Qalqiliya Ramallah & Al Bireh Tulkarem Salfit Rafah Tubas 0 5 10 15 20 25 30 35 40 Sources: MoH (2013); WB (2014); and WHO (2014). 62 See MoH (2013) and World Bank (2014). 63 See MoH (2013). 50 3.4.2 Health Workforce 3.41. The Palestinian health workforce has been expanding64 (Figure 55). The number of physicians increased from 17.4 in 2009 to 20.2 per 10,000 population in 2012. In the same period, the number of nurses and midwives increased, at a higher rate compared to physicians, from 16.4 to 19.7. The number of dentists rose from 4.6 to 6.1, and the number of pharmacists from 7.6 to 11.5 per 10,000 population. In aggregate, the availability of health workers in the Palestinian territories surpasses many countries with similar income levels and countries in the East Mediterranean region (EMRO)65 (Figure 56). Figure 55: Trends in human resources for health, Palestinian territories, 2009-2012 25 Physicians Nurses and midwives Dentists Pharmacists 23 21 per 10,000 population 19 17 15 13 11 9 7 5 3 2009 2010 2011 2012 Source: MoH (2013), (2012), (2011), (2010). Figure 56: Human resources for health (per 10,000 population), International Comparisons, 2012 EMRO LMICs West Bank Gaza 25 20 15 10 5 0 Physicians Nurses and midwives Dentists Pharmacists Sources: PT data from MoH, 2013; data on other countries are from WHO, 2014. 3.42. Regional disparities in health personnel are increasing66 (Figure 57). While in 2009, the density of physicians per 10,000 population in Gaza and West Bank were similar (17.5 and 17.4 per 10,000 respectively), in 2012 physicians in Gaza declined to 15.9, while increasing to 22.9 per 10,000 population 64 See Annex II for more detailed information. 65 WHO classification. 66 It was not possible to report the exact number of health workers that were asked not to report to work in Gaza or the total number of health workers hired by Hamas. After the split in 2008, the PA ordered the ~6,000 health workers in Gaza to not show up to work. A 2013 study estimated that 1,800 health workers continue to not work but draw salaries, which corresponds to 28 percent of the total health personnel in Gaza. Furthermore, the study showed that a significant proportion of those who do not report to work are administrative personnel (1,000), including staff working in hospitals and the MoH, followed by nurses (318), and 87 physicians. See MoH (2014). 51 in the West Bank. The number of nurses and midwives increased in PT overall, and this increase was more pronounced in the West Bank (from 18.2 to 22.4 per 10,000 population), compared to Gaza (from 13.5 to 15.4 per 10,000 population). The number of nurses and midwives in Gaza has seen a slight decline since 2010. Unlike other categories of health workers, today, more pharmacists provide services in Gaza than in the West Bank. Between 2009 and 2012, the number of pharmacists in Gaza increased more than six-fold from 2.0 to 12.8 per 10,000 population, the reason for which is not clear. In the same period, the number of pharmacists in the West Bank slightly declined from 11.0 to 10.7 per population. Figure 57: Human resources for health (per 10,000 population), West Bank vs. Gaza, 2009-2012 Source: MoH (2010), (2011), (2012), and (2013). 3.43. The MoH continues to be an important source of employment in the public sector. In 2013, there were some 14 thousand MoH employees, an 11 percent increase from 2006 corresponding to 15 percent of all public sector employees67. With this large workforce, the MoH is the second biggest employer in the public sector, after the Ministry of Education. Public employment has been capped at 3,000 jobs per year and this has impacted mainly the health and education sectors68. The MoH workforce saw the largest expansion between 2006 and 2007, growing by 5.2 percent, followed by a contraction of 0.4 percent between 2008 and 2009, 0.6 percent in 2010, and the growing again by an average of 1.2 percent annually between 2011 and 2013. 3.44. The geographic distribution of the MoH employees has seen a significant shift since 2006. In 2006, Gaza employed 63 percent of all MoH employees, whereas in 2013, this number dropped to 51 percent. In this period, the MoH workforce in the West Bank expanded on average by 5.6 percent annually and contracted by 1.5 percent per year in Gaza. More specifically, in the West Bank, the expansion of the public sector employment in health mainly took place between 2006 and 2007 when the public sector expanded on average by 13 percent. This expansion slowed down between 2008 and 2010, but again picked up since 2010. In Gaza, by contrast, the public sector workforce contracted by 3 percent between 2007 and 2008 (the most significant decline in recent years), following the call of President Mahmoud Abbas for government employees to stop reporting to work following the split between Fatah and Hamas. Public 67 The task team noted discrepancies between the number of civil servants reported by the MoH and the MoF. The report uses figures reported by the MoF in the subsequent section to allow comparison of the MoH workforce with the human resource capacity of other Ministries. 68 Palestinian National Development Plan 2011–13. 52 sector employment in Gaza continues to decline on average by 1.8 percent annually69. Although some of the health professionals who had stayed at home have recently returned to their posts, the number of MoH employees who continue to opt out from work is not known. In order to offset the adverse impact of shrinking the MoH workforce, Hamas hired health professionals to provide services in Gaza; the exact number of medical professionals hired is also not known. Figure 58: Geographic distribution of MoH employees, West Bank and Gaza, 2006-2013 West Bank Gaza number of MoH health 10,000 8,000 workers 6,000 4,000 2,000 0 2006 2007 2008 2009 2010 2011 2012 2013 Source: MoF (2014). 3.45. While both the overall health workforce and MoH employment have seen an increase in recent years, MoH human resources as a percentage of total health workforce have been decreasing (Figure 59). In 2009, the MoH employed 80 percent of all health workers in the PT and this proportion declined to 59 percent by 2012. More specifically, 34 percent of general practitioners were employed by the MoH in 2012, compared to 47 percent in 2009. Similarly, the MoH’s human resources in terms of specialist physicians declined and in 2012 only 37 percent of specialist physicians were employed by the MoH, compared to 45 percent in 2009.70 The share of MoH nurses, dentists, pharmacists, and midwives with respect to overall health workforce has also been declining. The dearth of information precludes one from ascertaining which type of health service provider is driving the expansion of the non-MoH employment. Figure 59: MoH human resources as share of the total health workforce (%), Palestinian territories, 2009-2012 non-MOH MOH 100% 80% 60% 40% 20% 0% 2009 2010 2011 2012 Source: MoH (2013), (2012), (2011) and (2010). 69 The Annual Health Reports does not provide information regarding whether these employees are hired by the Palestinian Authority or Hamas. 70 See MoH (2013), (2012), (2011), and (2010). 53 3.46. There is a perception of inadequate remuneration of health workers and disruptions in salary payments in the public sector. On several occasions during the past two years, public sector health workers, much like other public sector employees, were not paid their salaries on time or they only received partial payment.71 Delays in salary payment have resulted in large-scale strikes, with serious implications on service delivery. In addition, there is a perceived lack of stable career paths in the public sector and the fragmented institutional framework of the Ministry of Health and the absence of a well-defined and managed human resource development strategy for the PT hinders the overall development of MoH human resources.72 3.47. MoH health personnel predominantly work in hospitals (Figure 60)73. Despite recent MoH efforts to prioritize primary health care, the majority of the MoH health personnel work in hospitals; in 2012, 58 percent of all physicians, 68 percent of all nurses, 60 percent of all midwives and 45 percent of all administrative staff in the West Bank worked in hospitals. This composition is more pronounced in Gaza, where 71 percent of all physicians, 75 percent of all nurses, 62 percent of all midwives, 57 percent of all paramedics and 47 percent of all administrative staff works in hospitals operated by the MoH.74 This distribution of health personnel reflects a health system geared towards tertiary and curative care, undermining the PT’s ability to contain costs, particularly with NCDs. Figure 60: Health personnel, Palestinian territories, 2012 West Bank Gaza Hospital PHCC Hospital PHCC number of health workers number of health workers 1500 1600 1400 1200 1000 1000 800 500 600 400 200 0 0 Source: MoH (2013). 3.48. The composition of the MoH health workforce is characterized by imbalances in skill mix, with the administrative staff accounting for the majority (Figure 61). Administrative and service staff represented the largest category of MoH healthcare workers, accounting for 35 percent of all MoH employees in 2012 (30 percent in the West Bank and 39 percent in Gaza), with efficiency implications for MoH services.75 A significant majority of the administrative and service staff (63 percent) were employed in Gaza. 71 See Ozaltin and Alaref (2014). 72 See Schoenbaum, Afifi, and Deckelbaum. 73 The data on the distribution of all health workers across the PT was not available at the time of drafting this report. 74 See MoH (2013). 75 See MoH (2013). 54 Figure 61: Composition of MoH personnel, Palestinian territories, 2012 21% Physicians 35% 2% Dentists 3% Pharmacists Nurses Midwives 25% Paramedicals 12% Admin Staff 2% Source: MoH (2013). 3.49. Unemployment among graduates in health varies considerably depending on geographic location and gender. Unemployment among new health sector graduates was around 14 percent in 2010 and, while lower than the unemployment rate among all graduates (24.6 percent in 2010), it represents a challenge for the Palestinian health sector76. Unemployment among this group was almost double in Gaza (20.1 percent) compared to the West Bank (11 percent) and also varied significantly depending on the gender of the graduate with 20.2 percent of female graduates unemployed in 2010 compared to 10.3 percent of male graduates. Figure 62: Unemployment rate among graduates, Palestinian territories, 2010 40 percent of unemployed 30 graduates 20 10 0 Law Personal Health Total Natural Social and Computer Education Services Sciences Behavioral Science and Sciences Teacher Rehabilitation Source: PCBS (2011). 3.50. Dual practice, characterized by health workers engaging in clinical practice and other health- related activities concurrently in both the public and private sectors, is common across health workers in the PT. A recent World Bank study found the prevalence of dual practice in the health sector is close to 100 percent with little differentiation in the level of activity between the different cadres of health workforce, or by specialty, seniority, etc.77 The study indicated that dual practice was perceived as a coping mechanism to compensate for low salaries in the public sector and, that in the face of economic hardship, many physicians resort to private practice as a way of diversifying and in some cases supplementing their incomes. The presence of dual practice has both positive and negative consequences for the health sector. On one hand, dual practice might have helped easing the workload in the public sector, while on the other hand, the quality of services was likely to be impacted. This is largely because (i) physicians may provide better quality services at private facilities rather than in the public sector for the same condition, and (ii) long working hours may result in exhaustion and poor performance in the public sector. Furthermore, dual 76 The term graduates refer to those who hold an associate diploma certificate or higher degrees. 77 See Ozaltin and Alaref (2014). 55 practice is likely to have financial and governance implications for the Palestinian health system. For example, dual practice may lead to self-referrals (i.e., physicians referring patients to private hospitals where they work), producing a conflict of interest and inflating the referral bill for the MoH and contributing to high OOP spending.78 The reform of the dual practice was launched, but was then stopped as unworkable.79 3.5 Health Services Utilization 3.5.1 Utilization of Primary Health Care Services 3.51. On average, patients residing in the West Bank visit primary health care centers twice a year80. Between 2009 and 2012, the number of PHCC visits per population increased from 1.4 to 2.0 visits in the West Bank, corresponding to 4.8 million PHCC visits in 2012. The utilization of primary health care centers varies across governorates in the West Bank, from 3.4 visits in 2012 in Salfit governorate to 0.5 in Bethlehem governorate. In 2012, 84 percent of all primary health care patients in the public sector were received by general clinics and 16 percent were by specialized health clinics. Figure 63: Number PHCC centers and number of PHCC visits per governorate, West Bank, 2012 100 number of PHCCs (left axis) Number of visits per patient in governorate 4.0 80 3.0 60 2.0 40 1.0 20 0 0.0 Salfit Tulkarem Tubas Qalqiliya Jenin South Ramallah Nablus Jericho & Hebron Jerusalem Bethlehem Hebron & Al Bireh Al Aghwar Source: MoH, 2013. 3.52. The geographic distribution of PHCCs may have implications for utilization of primary health care services. As discussed earlier, the geographic distribution of primary health care centers in the West Bank do not reflect the population size of governorates. For example, in governorates with high volume of patients and fewer primary health care centers (e.g., Salfit, Tulkarem, Tubas, Jericho & Al Aghwar and Qalqiliya governorates), PHCCs may be crowded and the workload of medical staff may be heavier and, although services are provided at the primary health care level in public facilities free of charge, patients may perceive the quality of care in public services to be low compared to other service providers. Conversely, in governorates with a higher number of PHCCs and relatively lower patient volume, PHCCs may be underutilized. 78 Ibid. 79 The government has since focused on designing a policy that will be 1) based on evidence from well-designed studies; 2) sequenced gradually over time; 3) executed in the context of broader sectoral reforms; 4) taking into account the intrinsic as well as the extrinsic motivation of health workers; and 5) costed and sustainable within the given fiscal space. 80 There is very limited data on utilization services in the Gaza Strip; service utilization in the Gaza Strip is discussed where data is available. 56 3.53. Patient preferences towards public services differ across governorates. Despite having fewer primary health care centers, patients residing in Salfit, Tulkarem, Tubas, Jericho & Al Aghwar and Qalqiliya governorates visit public PHCCs more frequently than the West Bank average of 2 annual visits per patient. On the other hand, patients living in Hebron governorate visit PHCCs less frequently than the West Bank average. One possible explanation is that patients in this governorate prefer to go directly to one of the 5 hospitals operating in Hebron. In Bethlehem governorate, where the fewest number of patient visits to PHCC is observed, there are also fewer primary health care clinics, patients may be discouraged from utilizing primary health care due to the limited availability of services or due to a preference for seeing doctors (Bethlehem has the fewest number of physicians serving at the PHCC level in the West Bank). 3.5.2 Utilization of Secondary and Tertiary Care Services 3.54. The number of patients per physician has been increasing in public hospitals. 81 The number of hospital patients per physician in MoH hospitals increased from 1,241 patients in 2009 to 1,586 in 2012. This ratio increased faster in the West Bank than in Gaza: in the West Bank, it increased from 1,765 patients per physician in 2009 to 2,465 patients in 2012 (40 percent) and in Gaza from 1,070 patients per physician to 1,318 patients in the same period (23 percent). Hospitals in the West Bank continue to be an important destination for medical referrals from Gaza. Between 2009 and 2012, the number of patients referred from Gaza to the West Bank hospitals increased by 6 percent. Figure 64: Number of patients per physician in MoH hospitals, Palestinian territories, 2012 7,000 number of patients per physician 6,000 5,000 4,000 3,000 2,000 1,000 0 Source: MoH, 2013. Note: Red denotes admissions per physicians working in hospitals in the West Bank and blue in Gaza. 3.55. Outpatient services comprise the majority of public hospital service provision. In 2012, outpatient services accounted for 88 percent of all services provided by the public hospitals (Figure 65). Despite the high volume of outpatient services in public hospitals, the overall expenditure for inpatient services is nearly two times that of outpatient services.82 Anecdotal evidence indicates that in many cases, 81 The calculations in this paragraph are based on data available from public hospitals operating in the West Bank and in Gaza. In the West Bank, data was available for Al Watani Hospital, Yatta Hospital (Abu Al Hassan Al Kassem), Hebron Hospital (Alia), Tulkarim Hospital (Thabit), Jenin Hospital (Khaleel Sulaiman), Rafidia Hospital, Qalqiliya (Darweesh Nazal), Jericho Hospital, Beit Jala Hospital (Al Housein) and Salfit Hospital. In the Gaza Strip, data was available for the Othphalmology Hospital, Al Aqsa Hospital, Al Najar Hospital, Kamal Edwan Hospital, Tal el Sultan Hospital, Al Dorra Pediatric Hospital, Shifa Hospital, Beit Hanon Hospital and the European Hospital (Abu Jihad). 82 See section 3.2.1 Public Expenditures, for a detailed discussion on health spending by type of service. 57 patients prefer to visit outpatient wards and emergency rooms to receive treatments that could be provided by PHCCs.83 While, in theory, a referral is required, many patients bypass this system by going directly to the emergency ward or by visiting the hospital after primary clinics have closed. While more research is required to better understand utilization behavior and the composition of care in outpatient services, it is reasonable to assume that a majority of outpatient care could more cost-effectively be delivered through lower level facilities. Figure 65: Inpatient and outpatient services in public hospitals (%), Palestinian territories, 2012 Inpatient Outpatient 100 80 percent 60 40 20 0 Source: MoH (2013) 3.5.3 Efficiency of Service Delivery 3.56. The interaction time between patients and health workers at the primary health care level is limited. Figure 66 shows that PHCCs in the West Bank received on average 11.8 thousand visits in 2012. PHCCs in the Tulkarem governorate received 18 thousand visits (the highest in the West Bank); PHCCs in the Bethlehem governorate received 5 thousand visits. On average, a patient visiting a PHCC spent 9 minutes with a health professional.84 Key informant interviews indicated that, in practice, patient interactions with health professionals is much shorter, with significant time devoted to administrative tasks. The shortage of the interaction time between patients and health workers has several implications for the quality of care: health workers have limited time to inquire about the patients’ medical history and symptoms, offer a diagnosis and if necessary refer patients to secondary or tertiary care providers, and properly complete the required documentation and may also result in low patient satisfaction. 83 While referral from primary care is in theory required, many patients bypass this system, in particular after 3pm when PHCCs close. 84 This calculation assumes that a regular workday is 7 hours and the total number of business days in the public sector is 248 days. 58 Figure 66: Number of visits per primary health care center by governorate, West Bank, 2012 20,000 15,000 10,000 5,000 0 Source: MoH (2013) 3.57. The average length of stay in MoH hospitals has been relatively stable. The average length of stay (ALOS) in MoH hospitals declined slightly from 2.5 days in 2009 to 2.4 days in 2012 [9, 36]; 2.2 days in the West Bank, and 2.6 days in Gaza. Figure 67 shows that ALOS in MoH hospitals is lower than the international benchmark of around 4 days [43] and that there is significant variation across hospitals in ALOS.85 Overall, the low ALOS indicates that Palestinians may not be receiving adequate care for illness episodes. This may be due to the high volume of patients and insufficient capacity in the public sector (beds, materials, etc.) to meet demand. Figure 67: ALOS at MoH hospitals, Palestinian territories, 2012 6 Average length of stay (days) 5 4 3 2 1 0 Source: MoH (2013). Note: The purple line marks the international ALOS benchmark of 4 days. 3.58. While the average bed occupancy rate is within international standards, 74 percent of hospitals fall outside the benchmark range. Figure 68 shows that between 2009 and 2012, bed occupancy rate in MoH hospitals increased from 76.5 percent to 79.8 percent (79.1 percent in the West Bank and 78.9 percent in Gaza in 2012). Despite this slight increase in the recent years, the average bed occupancy rate (BOR) in 2012 was within the international benchmark range of 75-85 percent. However, a majority of 85 ALOS is higher in certain long-term care specialty hospitals/departments, e.g., geriatrics. 59 hospitals remain outside the benchmark range (Figure 69). Beit Hanon Hospital in Gaza and the Hebron Hospital in the West Bank had the highest BOR in the Palestinian territories, surpassing full capacity in 2012.86 The BOR of the Palestinian Medical Complex in 2012 was 92 percent: 94.9 percent in the Ramallah’s Sons Wing; 91 percent in the Heart and Specialized Surgery Wing; and 85.4 percent in the Pediatric Wing. BORs over the benchmark range have important implications for infection rates and quality of services. It is worth noting that BORs in NGO hospitals are much lower, pointing to continuing inefficiencies of how the MoH takes advantage of resources across the sector. Figure 68: Bed occupancy rate (%), Palestinian territories, 2012 Gaza West Bank Palestinian territories 90 80 percentage 70 60 50 40 30 20 10 0 2009 2010 2011 2012 Source: MoH (2009-2012). Figure 69: Bed occupancy in MoH hospitals (%), Palestinian territories, 2012 120 100 percentage 80 60 40 20 0 Source: MoH (2013). 86 A higher than 100 percent BOR indicates that more patients receive treatment in these facilities than there are available beds for them. This may be due to bed turnover with more than one patient per bed per day or patients sharing beds, or patients who are housed in rooms, corridors etc. without beds. 60 3.6 Discussion & Recommendations 3.59. The Palestinian health sector is at a critical crossroad. The recent conflict in Gaza exposed major weaknesses in the Palestinian institutional and regulatory system and highlighted the untenable fiscal position of the sector. 3.60. The financial sustainability of the healthcare system is compromised. Uncertain foreign aid flows, the increasing costs of referrals outside of the public healthcare system, inefficiencies and duplication of services, a focus on tertiary care, and the increasing prevalence of non-communicable diseases requiring long-term and expensive treatments, is putting a strain on the health sector and necessitating a government reaction in the short and medium term. 3.61. The fragmented institutional framework hampers the governance and long-term development of the sector. Coordination of the health sector, particularly among the Ministry of Health (MoH), private sector providers, non-governmental organizations (NGOs), and the United Nations Relief and Works Agency (UNRWA) is a top priority. The health sector division between West Bank and Gaza impacts the overall effectiveness of sector governance, undermines efforts to implement reforms, and creates uncertainty vis-à-vis the long-term development of the sector. 3.62. There is also high fragmentation in the donor community. There has been little coordination among the donor community in the health sector and donor funds are often politically tied and uncertain. Donor interventions and policy are often not aligned with national priorities, leading to inefficiencies in service delivery and constraints on health sector governance. 3.63. Due to the lack of availability of certain treatments and medical equipment, shortages of drugs and insufficient human resource capacity in public facilities, outside medical referrals (OMR) have become a salient feature of the health system in the PT. The situation is greatly exacerbated by restrictions on mobility which impede access to health services. Financing referral treatments outside the MoH is presently unsustainable due to: (1) rapid and uncontrolled increase in the demand for services; (2) overly generous service package; (3) limited financial basis, which is exacerbated by the fact that a large proportion of the covered population is exempted from any financial contribution; and (4) inefficiencies in and weak control over referral process and expenditures. 3.64. Out of pocket expenditures (OOP) and the resulting impoverishment remain a key challenge and priority for healthcare reform. The general impoverishment of the population makes it vulnerable to healthcare expenditures, and the scarcity of good quality government health services increases utilization of private, more expensive health providers. The poorest groups are at the greatest risk of impoverishment as they bear a higher share of out of pocket expenditures compared to their total share of income. Expenditures on medications due to chronic conditions, exacerbated by the lack of prevention and poor primary care represent a major financial burden for lower income families. 3.65. A focus on tertiary care has meant that primary care is underfunded and underutilized. Many of the outpatient cases seen in hospitals could reasonably and more cost-effectively, be cared for in primary care. While both West Bank and Gaza have family medicine programs, this only functions well in Gaza. 3.66. The MoH has the long-term objective of providing quality health services to all Palestinians while protecting them from financial hardship due to health costs in a manner that is financially sustainable. Achieving the long-term objective will require measures in the short and medium terms to improve the efficiency of health spending and the budget processes. The aim of this chapter is to provide an overview and analytical underpinnings to inform such reform and to provide policy options to assist the Palestinian Authority (PA) to achieve improvements in the technical and allocative efficiency of health 61 expenditures. The recommendations that follow are organized along the four functions of health systems 87: Financing; Governance (stewardship); Service delivery; and Resource generation. 3.6.1 Financing 3.67. Increasing the efficiency of purchasing and management of OMR, which represent an important cost driver, addressing health insurance, particularly coverage and revenues, and drug spending are top priorities. Shifting of financing towards primary and preventive care is also a priority and discussed under service provision (see section 3.6.3). In the short term:  To address the immediate financial burden imposed by OMR, the MoH can develop stringent criteria for prioritizing services that should be outsourced, including measures of cost-effectiveness, efficiency, and quality of care: o Strengthen capacity for accurate diagnosing and appropriate referring;88 o Standardize the process and rules of referrals for each procedure; o Consolidate information flows and databases, with a goal of establishing an integrated eReferral system; o Establish a clear incentive structure for organizing and rationalizing the demand for referral services;89 o Strengthen regulatory and supervision capacity to manage and verify referrals; o Develop a referrals masterplan, in close consultation with stakeholders, to outline areas and strategies for reform, roles of development partners and stakeholders as well as targets and timeline.  Improve the efficiency of pharmaceutical, equipment, and medical supply expenditure through more competitive procurement and international price benchmarking. Use savings to further increase drug availability and to finance other priority programs, including steps toward universal health coverage.  Rationalize the budgeting process. The discrepancies between the approved MoH budget and actual MoH expenditures hamper MoH recurrent expenditure planning as well as medium-term investment prospects. Furthermore, unpredictable revenues undermine the MoH’s credibility as purchaser of services (from other service providers) and medical supplies (e.g., drugs and disposables). 87 World Health Organization. World Health Report 2000. Health systems: improving performance. Geneva, WHO http://www.who.int/whr/2000/en/index.html. 88 Once there is a proper referral system in place, with clear policies and monitoring procedures, the MoH may decide to abandon the system to work with a central referral system and empower hospital and medical departments who are legitimized and able to refer patients. The central committee would then focus on monitoring and quality improvement. 89 For example, higher co-payments could be introduced for certain services and providers where patients could be encouraged to accept referrals to a list of preferred providers (according to clearly defined criteria) through differential co-payments that are higher for non-preferred providers. 62 In the medium term:  Define a roadmap to Universal Health Coverage for the Palestinian territories with a detailed calendar and planned actions in order to enhance supply capacity to deliver services to those in need with better efficiency. o Develop strategies for covering the informal sector.  Focus on health insurance reform. Revise and strengthen the current health insurance system to improve its efficiency and sustainability. Improve the current system to create genuine risk pooling and purchasing functions for GHI. o Revise and rationalize the benefit package to make it more sustainable, including the costing of services as well as a criteria to include and exclude services in the package; o Define the enrollment criteria and options; o Establish provider payment options for primary and hospital care; o Aim to reduce Out of pocket expenditures by focusing on coverage of medications.  Develop a strong framework and institutional capacity for strategic and transparent purchasing of OMR: o Establish a Purchasing Agency for OMR (possibly as a first step towards implementing a national purchasing agency under GHI). This pooling and purchasing agency should be independent or autonomous. Once it is functional and developed, this institution could eventually expand its mandate to cover purchasing of all health services. o Establish an Accreditation and Technical Audit unit with a mandate to set the standards and protocols to be followed by the Purchasing agency. o Revise all contracts with outside providers to include quality standards and appropriate medical audits to ensure adherence to those standards as well as prospective case-based payments, rather than fee-for-service or simple price and volume contracts, in order to improve the incentives for performance.  Pilot output based financing mechanisms. With a separation of functions at the MoH, a move towards more advanced forms of health system financing should take place. It is advised that this be incremental. This will require a good health information system as well as monitoring and supervision capacity at the MoH. o Choose a pilot hospital and increase its financial management autonomy. o Introduce and track a small number of efficiency KPIs, linked to financial incentives, to introduce efficiency at hospitals. 3.6.2 Governance 3.68. Revise the design and organization of the health system to clarify the role of different types of facilities and providers. Top priorities are sectoral reintegration between West Bank and Gaza; integration across the different providers in the health sector; donor coordination; and the separation of functions within the MoH. 63 In the short term:  End the sectoral division between Gaza and the West Bank; merge governance structures, visions and strategies for the development of the sector.  Operationalize the NHS90 goal of enhancing coordination among service providers to ensure integrated health service delivery by defining a set of steps and calendar to achieve this goal.  Reinstitute the committee tasked with overseeing coordination of the health sector, particularly of the Ministry of Health (MoH), private sector providers, non-governmental organizations (NGOs), and the United Nations Relief and Works Agency (UNRWA) with a focus on interoperable health information systems, standards of care, and regulatory capacity of the MoH.  Enhance donor coordination by ensuring a monthly meeting of health development partners, including key stakeholders, MoH departments and chaired by the Minister of Health. Establish key sub-sectoral working groups including for referrals and health insurance. In the medium term:  Separate the purchasing, provision, and regulation of health services functions of the MoH.  With a number of donor partners (including the World Bank) set to reengage in the health sector, donor partners should explore a Sector Wide Approach (SWAp) or Sector Wide Management (SWiM), in line with the Paris declaration on aid effectiveness, to reduce duplication, improve coordination and ensure alignment with national priorities. 3.6.3 Service provision 3.69. Quality of care, prevention – particularly linked to NCDs – primary care, special risk groups and mental health are key priorities. In the Short & Medium Terms:  Focus on primary care. Shifting appropriate care to lower cost settings and strengthening the quality of primary care, particularly for the primary prevention of chronic conditions will be of top importance in meeting the health needs of the population in a sustainable manner. This may entail investing in family medicine reform, which is already functioning well in Gaza – a model which should be emulated in the West Bank, strengthening the gatekeeping function of PHCs through legislation and rules and the use of technology, such as telemedicine, to increase home based care or care and/or facilitate care by health workers without requiring the physical presence of specialists.  Focus on nutrition and primary prevention of NCDs. The increasing burden on the health system due to NCDs and the multi-generational impacts of chronic malnutrition threaten the sustainability of the health system and PTs future economic development. Exclusive breastfeeding, appropriate complementary feeding ages 0-2 and risk factors associated with youth, including sexual behaviors, drug abuse, lack of exercise and smoking, should be priorities.  Focus on quality of care. It affects the health status of Palestinians, including maternal mortality ratio (MMR) and Infant mortality ratio (IMR), and impacts the demand for referrals as well as high OOP. In order to prevent deterioration in the quality of care, the MoH should continue to invest in quality improvement programs that focus on both people and institutions, including the development of a 90 National Health Strategy 2014 – 2016. 64 system of accreditation for service providers and continuing medical education programs for healthcare professionals. It is recommended that policymakers focus on: o Infection prevention; o Understanding quality of care in public and private sectors; o Regulation of, and data on, the private sector; o Defining standard treatment protocols for key conditions; o Accreditation; and o Strengthening audit functions. 3.6.4 Resource generation 3.70. Improving human resources, integration of the sector, health management information systems (HMIS), and appropriate planning of infrastructure are top priorities. In the Short & Medium Terms:  Resolve the issue of duplicate personnel in Gaza including the large number of health staff on MoH payroll but not working.  Improve measurement, evaluation and management capacity at the MoH including a continued investment in HMIS and focus on improved analytical input into policy decisions.  Ensure that facility planning takes the entire health sector into account. The MoH is encouraged to consider the entire Palestinian health sector (this includes NGO, private, UNRWA, and Jerusalem hospitals) in its sectoral planning, including purchasing of needed services from these hospitals and playing an appropriate stewardship role.  Reorganize the supply of tertiary care to rationalize tertiary-level referrals and improve equity in the access to these services. 65 Chapter 4: The Palestinian Public Pension System 4.1 An overview of the current pension system and recent reform efforts 4.1. Public pension expenditure in the Palestinian territories is high and unsustainable. Total pension expenditure was NIS 1,089 million91 in 2013, or around 3 percent of GDP.92 Even though this is close to the MENA average of 3.6 percent, it is considered extremely high in the Palestinian context given the demographic profile, with only 3 percent of the population currently above the age of 65. Transfers currently required by the pension system pose a huge fiscal burden on the PA’s budget and future costs are projected to continue increasing and, at some point, start crowding out spending on other key areas including social assistance, healthcare, and education. Political instability, low economic growth, high unemployment, and chronic difficulties with financing the PA’s budget are all factors that dampen the government’s appetite for pension reform. Nonetheless, the system is not fiscally sustainable as currently designed and implemented; hence the urgent need for reform. 4.2. Assessing the Palestinian public pension system against best international practice93 reveals a low pension coverage in the Palestinian territories. As of September 2014, the system had 154,986 active contributors, including 86,359 civil servants, 65,277 security personnel, and a limited number of 3,350 private sector employees.94 Together, these represent around 16 percent of the total labor force, and 7 percent of the working-age population (ages 15 to 64). On the other hand, the system provides benefits - whether for old age, disability, or survivorship - to 46,596 beneficiaries who represent only 1 percent of the total population. This coverage rate is very low given the share of pension expenditure in the Palestinian economy. Countries with similar low coverage ratios, such as Paraguay and Ghana, have much lower pension spending as a percentage of GDP, ranging between 0.1 percent to a maximum of 2 percent. 91 Based on December 2013 data (the amount includes regular and lump-sum payments). 92 Estimated GDP in 2013 was NIS 40.72 billion. 93 See Annex IV for an overview of generally accepted principles of a well-designed (or reformed) pension system, based on international best practices. 94 The first NGO joined the pension system in October 2011. By the end of 2013, 38 NGOs had joined the pension scheme. Low coverage in the private sector is mostly due to high informality, unemployment, and under-employment. 66 Figure 70: Pension expenditure and % of Figure 71: Pension expenditure and % of elderly population beneficiaries over total population 16.0% 14.0% Italy Portugal Pension Spending as percentage of GDP 12.0% Slovenia Bulgaria Germany 10.0% Austria 8.0% Lithuania United Kingdom Finland 6.0% Tunisia Norway West Bank and Gaza 4.0% Iraq Morocco Lebanon y = 0.391x 2.0% R² = 0.7017 Paraguay Ghana 0.0% 0% 5% 10% 15% 20% 25% 30% 35% 40% Percentage of Beneficiaries over Total Population Source: HDNSP Pensions Database. 4.3. The pension system is also highly fragmented, consisting of four schemes. Two of these schemes cover civil servants. The first one is for the West Bank and was inherited from Jordan (hereafter referred to as Scheme IV), and the second one is for Gaza, inherited from Egypt (hereafter referred to as Scheme I). The PA has been conducting efforts to integrate the two civil service schemes since 1998, which is why in 2001 the West Bank civil servants scheme (Scheme IV) was closed off and new entrants were enrolled into the Gaza scheme (Scheme I). Security services personnel were excluded from the public sector pension system until January 2005, when a new Security Services Pension Law (SSPL) was enacted to provide pensions for all security personnel aged 46 and above through a separate scheme (hereafter referred to as Scheme III). Since September 2006, and as a result of the unified pension law (UPL) that was enacted in 2005,95 all government employees including security staff under the age of 46 are members of a new scheme (hereafter referred to as Scheme II).96 Older employees were “grandfathered” and remain beneficiaries of their original pension schemes (Schemes I, III, and IV). Hence, beneficiary coverage in different schemes is age dependent. Scheme II only covers workers who were 46 years or younger in 2006. The other three schemes cover workers who were older than 46 in 2006. Projections show that by approximately 2020, all PA government workers will be covered under Scheme II. Schemes I, III, and IV were designed traditionally on the principle of defined-benefit (DB). Scheme II, on the other hand, was designed with two components: one as DB, and the other as defined-contribution (DC). In practice, however, all schemes operate on a pay-as-you-go, defined-benefit basis (PAYG-DB). 95 The preparation of this draft law was done through an Institutional Development Grant (IDG) funded by the World Bank. 96 This scheme was designed to have two pillars: a DB and a funded DC, but the latter pillar was never put in place. In 2011, the Cabinet adopted an action plan to reform the pension system which required the DC component of Scheme II to be modified to introduce Notional Accounts instead. However, there has been no progress in this regard, and given the complexity of this type of scheme, it is recommended that the switch is made to a PAYG-DB system. 67 Table 12: Summary of key characteristics of the four pension schemes operating in the West Bank and Gaza Scheme Population covered (members) Scheme I (gradually Gaza civil service employees plus those from the West Bank hired May 1, 2000 or later, phasing out) with the exception of those who are members of Scheme II Scheme II (“new” All PA public workers (civil servants and security services, and a few private sector scheme) employees) who were under the age of 45 by September 1, 2006 Scheme III (gradually Old security services personnel (ages 46 or older in 2006) phasing out) Scheme IV (gradually West Bank civil service employees hired before May 1, 2000, with the exception of phasing out) those who are members of Scheme II 4.4. All schemes offer an overly generous benefit formula by international standards. In Schemes I and IV, replacement rates are in excess of 90 percent of the final salary. Pension benefits are also tied to wage levels, as opposed to a cost of living index. Scheme III provides a benefit formula that is even more generous whereby retirees receive 100 percent of their final salary. The Palestinian public pension system also offers early retirement benefits. The mandatory retirement age for Schemes I, II, and IV is 60. In Scheme III, however, the retirement age for a full pension is 50. A civil servant may be eligible for early retirement after completing 15 years of contributory service and reaching the age of 55, while a security service employee may receive a complete pension after spending 15 years of contributory service and reaching 50 years of age. Early retirement at 45 years of age is allowed under certain conditions. Members are also allowed to purchase years of service. Further details on the benefit formula for each scheme are included in Table 13. 68 Table 13: Basic pension aggregated indicators for civil servants, security personnel, and nongovernmental employees in the Palestinian territories, December 2013 Indicators Scheme I Scheme II Scheme III Scheme IV Also known as 10% scheme, it covers Also known as 7% scheme, it covers Also known as 2% Gaza employees public sector and security services scheme. It covers plus West Bank workers under age 45 in 2005. Other Also known as civil service West employees hired employees of local entities and public Security Services Bank employees since 2001 institutions participate in this scheme Scheme hired before 2001 Current Situation Phasing-out On-going Not yet effective Phasing-out Phasing-out SCHEMES DESIGN Defined-Benefit Defined BenefitDefined-Contribution Defined Benefit Defined Benefit Contribution Rates from E & E: 22.5% 16.0% 6.0% 22.5% 2.0% (as % of covered wage) From Employees 10.0% 7.0% 3.0% 10.0% 2.0% From Employer (Gov.) 12.5% 9.0% 3.0% 12.5% - Retirement Age and/or qualifying conditions 60 60 60 60 60 Required Length of Service for Basic Rep. Rates 15 15 n.a. 15 15 Basic Replacement Rate 37.5% 30.0% n.a. 37.5% 37.5% Incremental Replacement Rate 2.5% 2.0% n.a. 2.5% 2.5% Maximum Replacement Rate 70% 70% n.a. 70% 70% Number of Last Years for Wage Base Calculation 3 3 n.a. 3 3 Lump-sum payments Returns of contributions Yes Yes - Yes Yes For those with more than 28 years of contribution Yes - - Yes Yes Insurance payments Yes - - Yes Yes Post-retirement Indexation Adhoc Adhoc Adhoc Adhoc Adhoc (as wage growth) (as wage growth) (as wage growth) (as wage growth) (as wage growth) Survivors Pensions (as % of Old Age Pensions) 100% 100% 100% 100% 100% SCHEMES DEMOGRAPHICS Number of Contributors (total) 6,465 141,216 Same 1,202 4,670 Civil Servants 6,465 74,438 - 4,670 Security Services - 63,724 1,202 - Private sector employees 3,054 Number of Beneficiaries (total) : 12,252 5,697 Same 8,825 15,004 Civil Servants 12,252 1,942 15,004 Security Services - 3,755 Private sector System Dependency Ratio: 190% 4% Same 734% 321% SCHEMES PERFORMANCE (in NIS) Average Monthly Wage Civil Servants 3,706 2,967 Same - 3,891 Security Services - 2,901 Same 4,988 Average Monthy Pension Civil Servants 1,385 350 - - 2,056 Security Services - 1,185 - 3,655 - Old Age Replacement Rate 37.4% 26.2% - 73% 52.8% 4.5. The eligibility criteria are broad, hence more than 50 percent of all beneficiaries in the public pension system are either disabled or receive survivorship benefits, but are not retirees. A wide range of cases are entitled to receive benefits from the survivorship program including: (i) the widow or widows of a participant; (ii) children and brothers younger than 21 who were dependent on the participant at the time of death; (iii) any children and brothers between the ages of 22 and 25 years old who were dependent on the participant at the time of death and who are continuing their university or higher education; (iv) any children and brothers who were dependent on the participant at the time of death and are incapable of earning a living due to health issues; (v) any daughters and sisters who are unmarried, widowed, or 69 divorced; (vi) parents of the participant; and (vii) husband of the participant if, at the time of her death, he was medically unfit to support himself. The high number of survivorship beneficiaries compared to old-age beneficiaries in the Palestinian public pension system is also due to the fact that Schemes I, III, and IV are phasing out,97 while Scheme II is still very young.98 On the other hand, the number of disability beneficiaries has been decreasing over the last few years, as the qualifying conditions were made more stringent. According to the law, one can receive a disability benefit when he/she is: (i) under the age of 60; (ii) not eligible for early retirement or old-age retirement; and (iii) approved by the medical committee.99 Details on all four schemes are provided in Table 14 and Table 15, where indicators of two representative months (December 2013 and September 2014) are presented. 97 Eventually older agedbeneficiaries will be dying faster than survivorship beneficiaries. 98 Most contributors have not yet reached retirement age, however there are already survivors from active contributors. 99 The Palestinian Pension Authority (PPA) has been reevaluating these conditions, particularly the third one, expecting to further decrease the number of disability beneficiaries. 70 Table 14: Contributors, beneficiaries, wages, and pensions, December 2013 Total wages / Average Scheme I Number pensions (NIS) wages/ pensions (NIS) Active contributors 6,455 23,924,281.51 3,706 Old age and early retirement 4,954 12,476,757.29 2,519 Disability 347 582,532.85 1,679 Survivors 6,951 3,912,950.04 563 Total beneficiaries 12,252 16,972,240 1,385 Total wages / Average Scheme II Civil Number pensions (NIS) wages/ pensions (NIS) Active contributors 74,438 220,877,158.04 2,967 Old age and early retirement 6 9,436.72 1,573 Disability 159 265,060.39 1,667 Survivors 1,777 404,737.09 228 Total beneficiaries 1,942 679,234 350 Total wages / Average Scheme II Security Number pensions (NIS) wages/ pensions (NIS) Active contributors 63,724 184,864,664.78 2,901 Old age and early retirement 771 2,885,849.24 3,743 Disability 8 14,883.78 1,860 Survivors 2,976 1,548,555.11 520 Total beneficiaries 3,755 4,449,288 1,185 Total wages / Average Scheme III Number pensions (NIS) wages/ pensions (NIS) Active contributors 1,202 5,995,286.80 4,988 Old age and early retirement 5,561 28,515,134.75 5,128 Disability - - - Survivors 3,264 3,737,759.85 1,145 Total beneficiaries 8,825 32,252,895 3,655 Total wages / Average Scheme IV Number pensions (NIS) wages/ pensions (NIS) Active contributors 4,670 18,173,129.51 3,891 Old age and early retirement 10,141 26,210,635.70 2,585 Disability Survivors 4,863 4,635,093.40 953 Total beneficiaries 15,004 30,845,729 2,056 Source: PPA. 71 Table 15: Contributors, beneficiaries, wages, and pensions, September 2014100 Average Scheme I No. of Records Total Wages wages/ pensions Active 5,832 23,099,422.00 3,961 Passive - Old Age & Early retirement 5351 13,483,843.00 2,520 Passive - Disability 337 574,340.00 1,704 Passive - Survivors 9,785 3,775,900.00 386 Total Passive 15,473 17,834,083 1,153 Average Scheme II Civil No. of Records Total Wages wages/ pensions Active 75,857 238,461,849.70 3,144 Passive - Old Age & Early retirement 34 55,688.00 1,638 Passive - Disability 113 197,008.00 1,743 Passive - Survivors 2,055 413,383.00 201 Total Passive 2,202 666,079 302 Average Scheme II Security No. of Records Total Wages wages/ pensions Active 64,218 205,017,376.30 3,193 Passive - Old Age & Early retirement 782 1,408,194.00 1,801 Passive - Disability 9 18,269.00 2,030 Passive - Survivors 3,156 1,203,948.00 381 Total Passive 3,947 2,630,411 4,212 Average Scheme III No. of Records Total Wages wages/ pensions Active 1,059 5,989,491.90 5,656 Passive - Old Age & Early retirement 5517 21,543,032.00 3,905 Passive - Disability 0 0 0 Passive - Survivors 3,392 3,351,882.00 988 Total Passive 8,909 24,894,914 2,794 Source: PPA. 4.6. Beneficiaries are entitled to various types of lump suml payments that pose a fiscal burden on the system. Around 45 percent of these lump sum payments are returns of previous contributions, while the remainder 55 percent are insurance payments. Participants are entitled, in the event of death or incapacity due to a permanent physical disability resulting from an occupational accident, to receive an insurance amount equivalent to a certain percentage of the annual salary according to their age. These insurance lump sum payments are paid in addition to regular benefits. Over the past few years, lump sumpayments represented over 6 percent of total annual pension payments.101 Table 16 presents a breakdown of total benefits paid by all schemes in 2013, including pensions and lump sum payments. 100 Data for Scheme IV were not available for the reporting period. 101 In 2013, the annual payments for regular pensions for Schemes I, II, and III amounted to NIS 650 million, while the lump-sum payments were NIS 44 million. 72 Table 16: Annual benefit amounts by scheme (NIS), 2013 Lump sum Scheme Pensions salaries Total benefits payments Scheme I 198,891,632.70 19,606,899.00 218,498,531.70 Scheme II - Civil 8,902,436.42 9,600,264.00 18,502,700.42 Scheme II - Security 53,290,871.00 1,459,898.00 54,750,769.00 Scheme III 389,568,303.00 13,481,977.00 403,050,280.00 Scheme IV n.a n.a 394,302,000.00 Source: PPA. 4.7. The PA has been making sporadic contributions to the pension system and accruing considerable arrears. Contributions transferred by the PA to the Palestinian Pension Authority (PPA)102 are well below the amounts stipulated by the UPL. Since May 2011, the PA has only transferred to the PPA amounts sufficient to make pension payments, but has rarely made any contributions to the system. The MoF transfers about NIS 70 million to the PPA every month, while its monthly commitment is around NIS 96 million, according to the pension law. As a result, the MoF accrues more than NIS 300 million in arrears to the pension system every year. In total, arrears accumulated to civil servant schemes over the years are estimated at US$1.595 billion. No information is yet available on arrears accumulated to the security services schemes.103 4.8. Pension benefits have nevertheless been paid consistently to date from funds earmarked for Scheme II and from the accumulated reserve. Since Scheme II is still considered young, it is currently accumulating reserves. In the absence of consistent contributions for the older schemes (I, III, and IV), transition costs have been paid through an opaque mechanism of transfers from funds earmarked for Scheme II. This is a serious concern because the PPA has not been able to credit contributions in the individual pension accounts of the beneficiaries of this scheme, as required by the UPL. Also, funds needed to meet the future obligations of these beneficiaries are being depleted. The PPA has relied on the reserve accumulated in years prior to 2011 to pay the monthly pension obligations. This reserve, however, has been declining gradually and latest available data show that it amounted to about US$199 million on March 31, 2015. 102 According to the UPL enacted in 2005, all pension schemes should be centralized in the PPA. Centralization has been improved, but there are still a few constraints. The main ongoing challenge with the accounting systems of all pension schemes is that in practice the PA transfers to the PPA are less than what is required by the UPL. The differences between the legal and actual contributions are summarized in and Table 18. 103 A joint committee that includes members from the MoF and the PPA was set up a few years ago to calculate the total amount of arrears owed to the pension system. The committee has so far focused its work on civil servant schemes and it plans to expand its work to include the security services schemes soon. 73 Table 17: Mandatory contributions received by PPA (from the MoF), June 2009 versus April 2013 Contributions due by the current Law PPA PPA (2009) (2013) Scheme I 22.5 percent of Only 10 percent No contributions received wage bill received since May 2011 Civil service and Scheme II Security services For civil servants only 7 - No contributions received since May 2011 16 percent of wage percent received bill for DB and 6 for DB, and 3 percent for DC - Only pension payments are received for percent for DC For security service only pension security forces payments are received Scheme III 22.5 percent of Wage bill Only pension payments Only pension payments are received are received Scheme IV 2 percent of Wage bill Nothing received Only pension payments are received Source: PPA. 74 Table 18: Collected contributions and transfers from PA/MoF to PPA, December 2013 Scheme Monthly Collected Transferred by Notes Contributions Contributions the MoF (Dec Due(Dec 2013)* 2013)** Schemel 5,382,963.34 0.00 0.00 The MoF has not transferred the monthly contributions of Schemel since May 2011 till now (April 2014). The PPA pays the monthly pensions and lump sum amounts to the beneficiaries of Schemel I. Schemel II - 48,592,074.77 0.00 0.00 The MoF has not transferred the monthly contributions of Schemel II – civil since Civil May 2011 till now (April 2014). The PPA pays the monthly pensions and lump sum amounts to the beneficiaries of Schemel II-civil. Schemel II- 40,670,226.25 0.00 The MoF has never transferred the contributions of Schemel II – security (except for Security a few scattered months in 2010 and 2011). The MoF pays the monthly oensions and lump sum payments of Scheme II – security. 39,846,226.40 Schemel III 1,348,939.53 0.00 Te MoF has never transferred the contributions of Scheme III (except for a few scattered months in 2010 and 2011. The MoF pays the monthly oensions and lump sum payments of Scheme III Schmel V 363,462.59 0.00 Approximtely 31 million NIS * The reconciliation amounts of the monthly contributions are not included. ** The transferred amount for Schemel-Security and Schemel III represent pensions + lump sums. Source: PPA. 4.9. Arrears combined with an overly generous benefit formula and early retirement provisions place severe financial constraints on the pension system, and make it unsustainable. The PPA currently faces a critical financial crisis because it has continued to consistently pay pension benefits even though contributions to the system have been highly sporadic. The PPA has tried to widen the system’s coverage to include the non-governmental and private sectors, aiming to increase the level of contributions. This activity is not recommended at this stage; the focus should instead be on parametric reforms. 75 Box 2: History of Reforms in the Palestinian territories The UPL, also known as Scheme II, was enacted in 2005 and all government employees, including civil and security services who were less than 46 years of age at that time, were enrolled into this scheme. It was designed to have two pillars: a defined-benefit (DB) and a funded defined contribution (DC), but the latter was never put in place. The old schemes and their beneficiaries were grandfathered and reserves from Scheme II have been financing pensions paid under the old schemes. In 2008, the PA adopted the Palestinian Reform and Development Plan (PRDP) 2008-2011 in which the PA committed, among other things, to the preparation of a pension reform action plan covering administrative and parametric changes to drive the pension system on the path of fiscal sustainability. In 2009, the MoF issued a decree requiring all civil servants to remain in each grade for a minimum of five years before being eligible for promotion and in the case of security services personnel, a minimum of four years. All pensions as of January 1, 2009, were calculated using the average of three years of wages to determine the pensionable salary. As such, artificial end-of-career salary increases, which have a negative effect on pension liabilities, were eliminated. In July 2010, the Council of Ministers approved an action plan aimed at reducing the heavy burden that the pension system imposes on the budget. The plan, however, only consisted of small measures that were likely to produce desirable effects in the short-term, but were not enough to ensure long-term sustainability. The plan included a set of administrative reforms to strengthen the institutional capacity of the PPA in addition to a number of parametric changes aimed at improving the financial performance and fairness of the existing pension schemes. Parametric reforms covered by the plan included: i) Increasing the mandatory retirement age from 60 to 62; ii) early retirement would only be allowed if pensions are reduced according to an actuarially fair coefficient to be produced by the World Bank; iii) lump sums for those who contributed to Scheme I for more than 28 years would be calculated using the same average salary as the one used to calculate regular pensions; iv) buying additional years of contributions would only apply with a limit of up to 3 years at the time of retirement for Schemes I, III, IV; v) new entrants to Scheme II would be allowed to purchase voluntary contributions of up to 5 years in advance within the first year of joining the system; vi) pensions would be indexed to the CPI published by the Palestinian Central Bureau of Statistics; and vii) the DC component of Scheme II would be modified to introduce notional accounts instead of funded accounts as currently stipulated by the law. Please see Annex V for a detailed explanation of notional defined contribution (NDC) plans. Progress made in the implementation of the reform action plan was uneven. Most of the administrative reforms were implemented, leading to an overall improvement in the organizational aspects of the PPA. However, none of the parametric reforms were carried out since those required commitment and actions from the PPA in addition to the Cabinet, the MoF, and other actors. The implementation of such reforms is still under discussion. 4.2 Implications of maintaining the status quo 4.10. The financial situation of the four pension schemes clearly shows that all register a monthly deficit except Scheme II. Table 19 summarizes actual revenues received and expenditures paid out of each of the four schemes in December 2013. Pension payments for the security service personnel in Scheme II in addition to those in Scheme III have been transferred by the PA to the PPA. On the other hand, Scheme I and the civil servant and non-governmental component of Scheme II have not received any contributions from the PA, and therefore, monthly pension benefits are covered by the PPA from revenues generated by Scheme II and from reserves. Contributions and payments for Scheme IV are directly taken care of by the PA, even though by law this scheme, like others, should be centralized at the PPA. Table 20 constructs a hypothetical financial situation for December 2013 assuming that all due contributions were transferred to the PPA as required by the UPL. Even under this assumption, all schemes except Scheme II register deficits. 76 Table 19: Actual revenues, expenditures, and balance of each pension scheme, December 2013 Indicators Scheme I Scheme II ( CS ) Scheme II ( SS ) Scheme III Scheme IV TRANSFERS FROM PA TO PPA OR DIRECT PAYMENT 0 0 39,846,226 31,968,151 TOTAL PENSION EXPENDITURES 18,606,148 1,479,256 39,846,226 31,968,151 CURRENT BALANCE (18,606,148) (1,479,256) - - Table 20: Hypothetical revenues, expenditures, and balance of each scheme assuming consistent transfers by the PA as required by law, December 2013 Indicators Scheme I Scheme II ( CS ) Scheme II ( SS ) Scheme III Scheme IV TOTAL REVENUES FROM CONTRIBUTIONS 5,382,963 48,592,975 40,670,226 1,348,940 363,463 TOTAL PENSION EXPENDITURES 18,606,148 1,479,256 4,570,946 35,275,281 31,968,151 CURRENT BALANCE (13,223,185) 47,113,719 36,099,280 (33,926,341) (31,604,688) 4.11. The deficit generated by the old schemes is currently covered by revenues from Scheme II, but this situation is unsustainable. To understand how the pension system’s financial situation will evolve over the next 35 years (until 2050), simulations were conducted using the World Bank’s Pension Reform Options Simulations Toolkit (PROST). The simulations were made under the assumption that no reforms will be carried out over this period.104 The results of this exercise are presented in Figure 72. Figure 72: Current balance of the four pension schemes in the Palestinian territories (% of GDP), 2013-2050 2.0% (Scheme I) 1.5% (Scheme II -CS) (Scheme II -SS) 1.0% (Scheme III ) Current Balances as % of GDP (Scheme IV) .5% Notional Cash Surplusus (.5%) (1.0%) (1.5%) (2.0%) (2.5%) (3.0%) Source: Authors’ estimates using PROST. 4.12. As shown in Figure 72, Schemes I, III, and IV will continue to accumulate deficits until they are phased out. The deficit from these schemes will continue to grow in the short to medium term, but will start declining in the longer term as they are closed off to new entrants. Notably, the deficit accumulated by Scheme I in particular will further widen until 2020. These schemes will continue to pay pensions, including survivorship benefits, beyond 2050. 104 Projections assume annual public employment growth to be 2 percent since this has been the trend during the last few years. It also assumes a slight increase in private sector coverage. 77 4.13. If the PA starts making contributions to the PPA as required by the UPL, the pension system will fall into a deficit in 2031. In the short term, Scheme II will continue to generate a surplus that can be used to pay pension liabilities for all schemes. In about 10 years, however, and as Scheme II starts maturing, this surplus will start declining and will not be sufficient to finance the high deficits generated by the older schemes. In the long term, specifically in 2031, the parameters of Scheme II will lead to structural deficits. The funded DC component of Scheme II will not be able to accumulate reserves either.105 4.14. Notably, the pension system could run out of funding much earlier if the PA carries on with the practice of accumulating arrears. PROST simulations show that if the PA continues underpaying the PPA and only paying pension benefits to security personnel while not transferring the required contributions, the system will fall into a deficit and become unsustainable in 2021. 4.15. The Palestinian public pension system is not sustainable as currently designed and implemented; hence the urgent need for parametric reforms.106 Regarding Scheme II, the focus should be on converting its funded DC component to a PAYG-DB system. As for Schemes I, III, and IV, urgent measures should be adopted to reverse their current and projected deficits. Additionally, it is essential to undertake further reforms related to disability and survivorship pensions. These reforms are discussed in detail in the next section. 4.3 Reform options and their expected impact 4.16. Restoring the long-term sustainability of any pension system requires a mix of reforms and adjustments that can lead to a socially optimal solution, since adjusting a single parameter would have limited effectiveness. Several reform options that can help place the Palestinian pension system on a more sustainable track are as follows: Reform 1: A gradual increase in the retirement age to 65 (between 2015 and 2020); Reform 2: In addition to increasing the retirement age, a gradual decrease in the accrual rate107 of pension benefits to 1.5 percent between 2015 and 2020; Reform 3: In addition to an increase in the retirement age and a reduction in the accrual rate, indexing benefits to annual changes in the CPI; Reform 4: In addition to retirement age increase and a reduction in the accrual rate, indexing 50 percent of benefits to annual changes in the CPI, and 50 percent to nominal wage growth; Reform 5: Eliminating early retirement (participants are not allowed to retire before the legal retirement age of 60) or introducing actuarially fair reduction factors; Reform 6: No cashing in of benefits prior to retirement age. Unlike what is proposed under Reform 5, in this case, participants would be eligible to retire early, but they would not have the right to receive pension benefits until they reach retirement age; Reform 7: Elimination of lump-sum benefits for all pensioners – this could have particular financial implications for Schemes I, III, and IV. 105 Given the complexity of this type of scheme, it is recommended that a switch is made to a PAYG-DB system. 106 Please see Annex VI for some guidelines on the financial stability of a pension system. 107 Accrual rate is the rate at which a beneficiary builds up his/her pension benefits while a member of a DB scheme. The rate is multiplied by earnings to calculate how much money the beneficiary will be entitled to eventually. It is typically expressed as a fraction, and the bigger the fraction, the more pension benefit the beneficiary willreceive. 78 4.17. Simulations carried out to assess the financial implications of the first three above-mentioned reform options on Scheme II show that the combined adjustments produce positive outcomes. Simulation results are presented in Figure 73 and Figure 74, which project the system’s annual expenditures and current balance under the three reform options and under a baseline scenario where no reforms are implemented and Scheme II falls into deficit in 2031, as discussed in the previous section. Results show that if the PA opted for Reform 1, which only requires increasing the retirement age to 65, the time Scheme II goes into deficit is initially delayed by a few years. In the long run, however, the deficit becomes even deeper than in the baseline scenario. This paradoxical conclusion is explained by the fact that raising the retirement age increases years of service, and ultimately, pension benefits.108 In Reform 2, where in addition to increasing the retirement age the accrual rate is reduced, falling into deficit is delayed by approximately eight years. Simulations show that if Reform 3 was implemented, whereby in addition to the increase in retirement age and the reduction of the accrual rate, pensions are indexed to inflation, the entry into deficit is delayed by approximately 10 years compared to the baseline scenario, specifically in 2041.109 Figure 73: Projected gross expenditures for Figure 74: Current balance of Scheme II under Scheme II under several reform options several reform options 6.0% 3.0% 2.0% 5.0% Pension Expenditures as % of GDP Pension Expenditures as % of GDP Baseline Reform (i) 1.0% Reform (iii) 4.0% Year 3.0% Baseline Reform (ii) (1.0%) Reform (ii) 2.0% Reform (iii) (2.0%) Reform (i) 1.0% (3.0%) (4.0%) (5.0%) Year Source: Authors’ estimates using PROST. 4.18. Increasing the retirement age, reducing the accrual rate, and indexing pensions to inflation can delay the onset of the deficit, but more needs to be done to ensure long-term sustainability. PROST simulations show that a combination of parametric reforms can improve the pension system’s financial outcome through delaying the onset of the deficit by 10 years. Despite this good result, a one-time change 108 Another reason that may explain this conclusion is related to the fact that the benefit formula for Scheme II uses the last three years’ final salaries to calculate pension benefits, and as workers retire later in their careers, the salary base increases, which pushes up pension expenditures. This, however, could be avoided if the benefit formula is changed to include an individual’s lifetime average salary instead of the final three salarie s, which would delay the occurrence of the deficit in Scheme II by three or four years. 109 Notably, the evolution of reserves was not considered in the analysis. This, however, can be an important source of financing for the structural imbalance of Scheme II, particularly if the MoF is expected to pay down part of the arrears it owes the PPA. Also, including investment returns in the analysis shows that the inherent deficits of the system can certainly be delayed for a few more years. 79 of parameters will not suffice to guarantee a long-term and/or permanent equilibrium in the system, particularly since demographic trends are expected to continuously change. Therefore, additional reform measures are needed to ensure long-term sustainability. 4.19. The use of actuarially fair reduction coefficients for early retirement is a recommended option to improve the long-term financial sustainability of Scheme II. As mentioned earlier, the Palestinian public pension system offers generous early retirement options. The elimination of early retirement, however, does not seem politically feasible at this stage. Thus, one possible option is to introduce “actuarially fair” reduction coefficients that depend on an individual’s age and gender and are calculated based on projected mortality rates, pension indexation rules, and a specific discount rate. Such coefficients equalize the net present value of the reduced benefit stream given at an earlier age and a regular benefit stream awarded at the statutory retirement age. The coefficients have been calculated for Scheme II; the results are presented in Table 21 and Table 22. The results show that to preserve fairness, pensions should have significantly higher reductions than the current practice when workers retire before the statutory retirement age. For instance, retirement at age 55 would require an “actuarially fair” reduction of pensions by 23 percent for men and 21 percent for women. The reduction of pensions would be more significant for those workers who decide to retire even earlier. Table 21: Actuarially fair reduction coefficients (%) for early retirement for men Age 2015 2020 2030 2040 2050 45 49% 48% 47% 46% 45% 46 47% 46% 45% 44% 44% 47 45% 45% 43% 42% 42% 48 43% 43% 41% 40% 40% 49 41% 40% 39% 38% 37% 50 38% 38% 37% 36% 35% 51 36% 35% 34% 33% 33% 52 33% 33% 32% 31% 30% 53 30% 30% 29% 28% 27% 54 27% 27% 26% 25% 24% 55 23% 23% 22% 22% 21% 56 19% 19% 19% 18% 17% 57 15% 15% 15% 14% 14% 58 11% 11% 10% 10% 9% 59 6% 6% 5% 5% 5% 80 Table 22: Actuarially fair reduction coefficients for early retirement for women Age 2015 2020 2030 2040 2050 45 45% 44% 43% 42% 42% 46 43% 42% 41% 41% 40% 47 41% 40% 39% 39% 38% 48 39% 39% 38% 37% 36% 49 37% 36% 35% 35% 34% 50 35% 34% 33% 32% 32% 51 32% 32% 31% 30% 30% 52 30% 29% 28% 28% 27% 53 27% 26% 26% 25% 24% 54 24% 24% 23% 22% 22% 55 21% 20% 20% 19% 19% 56 17% 17% 16% 16% 15% 57 13% 13% 13% 12% 12% 58 9% 9% 9% 9% 8% 59 5% 5% 5% 4% 4% 4.20. A comprehensive pension reform program should also include adjustments to survivorship and disability benefits and to the practice of purchasing years of service. The Palestinian public pension system offers generous survivorship110 and disability benefits. An in-depth assessment of these programs needs to be undertaken to come up with recommendations on how to best deal with these growing expenditures. In addition, members are allowed to purchase extra years of service to become eligible for a pension. This option creates a fiscal burden on the system and should be gradually reduced or even fully eliminated. 4.21. Although the impact of parametric reforms on Schemes I, III, and IV was not assessed for this PER, it is believed that it will be similar to that presented for Scheme II, but will manifest much earlier. The challenge related to reforming these three schemes is that adjustments will affect pension benefits of a population that is closer to retirement age. Based on experience from other countries that implemented similar reforms, a gradual adjustment formula needs to be adopted. If such a reform path proves feasible in the Palestinian context, it would reduce pressure on Scheme II, allowing for further accumulation of reserves, which will delay their eventual depletion. 4.22. The PPA has been trying to expand coverage as part of its reform efforts, but this is not recommended at this stage. The PPA has focused recent efforts on adding individuals from the private and non-governmental sectors to the public pension system. The number of contributors from this group increased 10 percent between December 2013 and September 2014, to reach 3,350. At the same time, the number of beneficiaries increased exponentially by 675 percent, mainly through the survivorship scheme.111 Expanding coverage at this stage, when no other reforms are implemented, would further increase pension spending and worsen the system’s financial situation. This could also happen if a decision is made to add 110 Please see Annex VII for an overview of survivorship programs. 111 The number of private sector and non-governmental employees covered by the pension system increased from 3,054 in December 2013 to 3,350 in September 2014. At the same time, the number of beneficiaries increased from 4 to 31, particularly because of the survivorship pensions (26 survivorship pensioners, and 5 old-age pensioners). 81 Hamas employees in Gaza to the PA’s payroll (although the effect on pension spending could perhaps be lower because the average wage for Hamas employees is lower than that of PA employees). 4.4 Potential cost savings from implementing the 2011 Pension Reform Action Plan 4.23. Implementing the Pension Reform Action Plan approved by the Cabinet in 2011112 could generate considerable savings. Table 23 presents the estimated public pension expenditure for each scheme under a baseline scenario and under the assumption that the action plan is implemented. Notably, the table also presents figures on estimated expenditures assuming that Hamas employees from Gaza113 are added to the PA’s payroll.114 Expenditure in 2020 for this group alone is estimated at about NIS 18 million.115 According to preliminary estimates, the proposed measures may lead to accumulated savings of around NIS 130 million in the first three years of implementation. Estimates also indicate that the reduction in expenditure will even be higher in the following years, surpassing NIS 100 million by 2020. Such a reduction in pension liabilities will reduce the burden on the public budget and release funds that can be used to support social programs with a high redistributive impact. Table 23: Estimated public pension expenditures (NIS million)116 for each scenario, 2013-2050 Year 2013 2014 2015 2020 2030 2040 2050 BASELINE Scheme II (CS) 18.50 22.00 22.90 26.00 1204.00 4234.00 5300.00 Scheme II (SS) 54.75 74.00 77.00 80.00 201.00 852.00 2422.00 Assuming that an additional 38,552 new employees in Gaza are also covered) 18.40 45.63 193.40 549.79 Scheme I 218.50 225.40 235.70 262.90 246.50 150.70 54.70 Scheme III 403.05 452.60 429.20 403.00 305.00 291.00 99.00 Scheme IV 394.30 378.20 382.60 413.80 360.00 153.90 29.60 IF ACTION PLAN IS IMPLEMENTED Scheme II (CS) 18.00 19.00 22.00 1025.00 3489.00 4180.00 Scheme II (SS) 73.60 76.00 79.00 160.00 732.00 1692.00 Scheme I 212.40 213.40 206.10 146.60 75.10 50.90 Scheme III 430.00 420.00 401.00 265.00 199.00 61.00 Scheme IV 372.90 372.20 368.30 262.30 92.00 17.80 COST SAVINGS IF ACTION PLAN IS IMPLEMENTED 45.30 46.80 127.70 503.23 1287.90 2453.39 4% 4% 11% 21% 22% 29% 112 Please see Box 3 for details on the 2011 Pension Reform Action Plan. 113 Data available show that 38,552 civil servants and blue uniform security personnel were hired by Hamas. Calculations in Table 23 were made based on this figure and do not include green uniform personnel, whose data are unavailable. If those were added, the financial impact could be considerably higher. 114 Hamas employees in Gaza are covered from 2015 onwards with an average monthly wage of NIS 2,628, and a monthly wage bill of NIS 101,301,812. 115 Careful consideration should be given to these estimates since data on age/gender distribution of employees, income distribution, and other relevant information were not provided. Assumptions were made based on distributions from the currently covered employees, except for the length of service. 116 Total pension expenditure includes both regular benefits and lump-sum payments. 82 4.5 Conclusions and recommendations 4.24. The Palestinian public pension system suffers from considerable challenges. Pension spending in the Palestinian territories, at 3 percent of GDP, is high in comparison to other countries with similar coverage ratios. The system is also highly fragmented, consisting of four schemes; two of which were inherited from Jordan and Egypt and another two created later. All schemes offer extremely generous benefit formulas with replacement rates that can reach up to a 100 percent of the final salary. The system also offers early retirement options, sometimes at the age of 45. The eligibility criteria are broad, hence a wide range of cases are entitled to survivorship and disability benefits, which places a significant fiscal burden on the system. Most importantly, the PA has accumulated considerable arrears to the pension system over the years, estimated at around US$1.595 billion. 4.25. The system as currently designed and implemented is unsustainable, and urgently needs reform. According to World Bank estimates, if the status quo continues, the system is expected to run out of funding in 2021. If the PA stops the practice of accumulating arrears and starts making contributions to the PPA as required by the UPL, the onset of the deficit will be delayed by 10 years, to 2031. This is not enough to deal with the system’s structural problems; hence the need for a comprehensive program that focuses on parametric reforms. 4.26. Restoring the Palestinian pension system’s sustainability requires a mix of reforms and adjustments. Proposed reforms could include raising the retirement age to 65 years, reducing the accrual rate of pension benefits to 1.5 percent, and indexing pensions to inflation, as simulations show that these three reforms together could delay the onset of the deficit to 2041. To improve the system’s long-term financial sustainability, it is highly recommended that the PPA consider the use of actuarially fair reduction coefficients for early retirement. A comprehensive pension reform program should also include adjustments to survivorship and disability benefits and should aim to limit or gradually eliminate the practice of purchasing years of service. Particularly for Scheme II, it is recommended that its funded DC component is changed to a PAYG-DB system, which will be easier to implement given the situation. 4.27. Parametric reforms need to be implemented now to make use of resources generated by Scheme II, while it is still young. Since Scheme II is still young, it provides fiscal space that grants the PA a unique opportunity to fundamentally review the current design of the pension system and improve its sustainability and efficiency. As Scheme II matures and its surplus starts to decline, it will become more difficult to reform the overall system. 4.28. Policymakers need to be aware of the fiscal, welfare, and behavioral implications of the chosen reform program. The diversity of potential reform packages is considerably high, and no single optimal solution exists. Specific measures should be considered as part of an integrated strategy that also considers social impact. In the Palestinian context, it is recommended, for instance, that reform measures be applied gradually to avoid reducing the pensions of those closer to retirement age. 83 Chapter 5: Intergovernmental Fiscal Relations 5.1 Introduction 5.1. This chapter pursues two key objectives. The first is to review in depth the existing intergovernmental fiscal arrangements between the PA and Local Government Units (LGUs), identifying the main problems and issues related to the current vertical structure of subnational governments, the expenditure responsibilities and service delivery, financing through local taxes and fees, the system of transfers, and budgeting practices. Due to constraints in the available information, Gaza LGUs’ budget data were not included in the analysis. This means that all comparisons in terms of GDP and as a share of the general government are adjusted to include only the relevant figures excluding Gaza. The second objective is to provide recommendations that would allow the PA and LGUs to adopt policies and practices that improve LGUs’ financial health and the efficiency and equity of public service provision at the local level in the medium to long term in the Palestinian territories. 5.2. Besides the PA, only one second tier of government exists in the Palestinian territories. This second tier comprises 136 municipalities (111 in the West Bank and 25 in Gaza) and 243 village councils (VCs). The history of many municipalities predates by many years the Oslo Accords of 1993 and 1995.117 These LGUs have elected councils and are responsible for local service delivery for the highly urbanized population, with three-fourths residing in municipalities. 5.3. Compared to other countries and world regions, local government in the West Bank and Gaza is rather small. Local government revenues amount to 11 percent of total revenues, while local government expenditures account for only 6 percent of expenditures, a fraction of what is observed in other countries (Figure 75). This is an indication that fiscal decentralization is still in the early stages in the Palestinian territories. The current level of decentralization can be best analyzed by looking at the share of each level of government in the general government total revenues and expenditures. As shown in Figure 76, for 2011 and 2012, the central government accounted for around 88 percent of general government revenues and around 93 percent of general government expenditures. In turn, municipalities accounted for around 10 percent of revenues and 5.5 percent of expenditures in those same years. The respective shares of VCs are much smaller: 1.5 percent for revenues and close to 1 percent for expenditures. 117 It is not uncommon to find an antagonistic view of the PA among LGUs, especially the older longer established municipalities. This may be due to the fact that many of these municipalities were in existence before the creation of the PA and also because there is a perception of not enough recognition of the role played by the LGUs by the PA and the little assistance received by the LGUs from the PA. 84 Figure 75: Fiscal decentralization in the Palestinian territories and selected countries and world regions, 2012 35.00 32.4 30.9 28.9 30.00 25.00 19.7 19.3 20.00 18.3 18.7 18.8 15.00 11.5 11.3 9.0 10.00 6.2 5.00 0.00 Palestine Turkey Eastern Asia Eastern Europe Europe Latin America and Central and Caribbean Asia Subnational revenue, % of total Subnational expenditure, % of total Sources: PA and IMF Government Finance Statistics. Figure 76: Total revenue and expenditure as a share of GDP (%) by level of government, 2010-2013 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% 2010 2011 2012 2013 Central Government Revenue Central Government Expenditure Municipalities Revenue Municipalities Expenditure Village Councils Revenue Village Councils Expenditure Note: Data for municipalities are available for 2010-2012 and for VCs for 2011-2013. The budget of VCs for 2011- 2013 is for the first 9 months for most VCs but not all, as some VCs reported the first 10 or 11 months or full year data. GDP is value added at constant prices: 2004 is the base year. 5.4. Although formally Palestinian law does not distinguish between municipalities and VCs, the main accepted definition or difference between municipalities and VCs is population size. While most municipalities have more than 5,000 inhabitants, around 70 percent of VCs have populations below 3,000, 85 with the smallest one having a population of 272 (2011-2013 average).118 Many VCs have populations under 2,000 inhabitants, which is considered very small; this seriously handicaps efficiency in the delivery of the most common local public services.119 On the other hand, some VCs are larger in population size than some municipalities. Thus no clear size-based distinction exists between the two. Other defining characteristics of VCs (vis-à-vis municipalities) are that they tend to be poorer in terms of economic base and also very often lack permanent staff. Nevertheless, the line of separation between municipalities and VCs remains unclear. Figure 77: Population distribution in VCs (Left) and municipalities (Right) .00008 .0003 2011 2010 2011 2012 .00006 2012 .0002 2013 Density Density .00004 .0001 .00002 0 0 2000 4000 6000 8000 0 0 50000 100000 150000 200000 Population Population Source: PCBS. 5.5. Beyond the need to legally clarify the distinction between municipalities and VCs, the most important issue with the current vertical structure of government is the large number and small size of VCs. This has generally been perceived as a stumbling block for the long-term sustainability of local service delivery in the Palestinian territories. To address this problem, the PA has worked on an amalgamation program on a voluntary basis, but this effort has had very little success to date as it appears that most VCs wish to keep their own identity and are worried about losing representation as part of larger jurisdictions’ councils. This is consistent with the experience of many countries in Europe and elsewhere, whereby voluntary amalgamation programs have not worked even when they have been heavily incentivized. 5.6. Forgoing the mandatory amalgamation of LGUs, which has been increasingly used in a number of countries, the PA has introduced a program of voluntary associations of VCs for the delivery of certain local services. The Local Government Act (LGA) provides for local governments to create Joint Service Councils (JSCs) for the joint provision of public services. Over 70 JSCs have reportedly been created in the West Bank, typically for one type of specific service, including solid waste management, water supply, wastewater, and road and infrastructure maintenance. Often these JSCs are created with the 118 The average population of municipalities ranged between a bit over 13,000 in 2010 and close to 16,000 in 2012 (Figure 77). However, the smallest municipality had 2,500 inhabitants and the largest close to 200,000. 119 The international experience shows that most local public services can be produced with economies of scale –or decreased unit costs of production—up to a population of 10,000. These include all local public services with the exception of large transportation systems and brownfield sites (see Lago and Martinez-Vazquez 2013). 86 support of foreign donors so there is concern about the future viability once that support ends. In addition some JSCs suffer from their own weak governance, reflected in the lack of legal agreements among members and unclear rules on fee contribution and arrears management. 5.7. In sum, the Palestinian territories lack an effective association of local governments. The current association is perceived as somewhat dysfunctional and has little or no influence on the PA, although some large municipalities appear to have direct access to the PA. This complicates and possibly diminishes the lobbying powers of the local perspective in national policy design issues and makes it more difficult to address in a concerted way the problem of local government fragmentation. Box 3: Different Approaches Have Successfully Been Tried to Address the Issue of Excessive Local Government Fragmentation Local government associations similar to the JSCs are successfully used in other countries where local fragmentation has been a problem. A different but related approach has been to rely on the services of a private company, which as in the case of the JSC can also achieve the right economies of scale with the joint provision of the services for a number of local governments. In fact, this latter modality is being used in the Palestinian territories for the provision of electricity services through the Electricity Distribution Companies (DISCOs), which directly sell services to residents. But clearly this modality could also be adopted for other services —and has been in many other countries—including solid waste management, drinking water, and sewerage services. 5.2 Functional and expenditure assignments of municipalities and VCs 5.8. As specified in the LGA of 1997, LGUs (both municipalities and VCs) are legally responsible for 27 functional responsibilities, but it is not clear if all are mandatory.120 These include most but not all of the local public services that are common in other countries at the local level. This list of functions makes no distinction between mandatory functions for LGUs (services that need to be provided) and voluntary functions (those that may be provided if funds are available and the service is deemed convenient by the local council. The LGA does not distinguish between “delegated” responsibilities and “own” responsibilities. In the case of delegated responsibilities, the PA would ultimately be responsible for the regulation and financing of those functions but they would be implemented by LGUs.121 In contrast, for own responsibilities, LGUs would be generally responsible for the services.122 The absence of conditional grants from the PA to LGUs in the current system of intergovernmental finance (actually of any kind of transfers but for emergency grants and intercepted partial revenue sharing) indicates that the LGA of 1997 had no intention to introduce delegated responsibilities. Thus, the prevailing understanding that the lack of sufficient funding has left LGUs with many “unfunded mandates” is not entirely correct.123 5.9. While it is clear that LGUs do not fulfill all 27 functional responsibilities, it is less clear which ones they do fulfill. The current budget format does not provide a completely clear functional classification of expenditures, which would have enabled clear insights into the functions fulfilled by municipalities and 120 These are augmented by 41 “tasks” according to the MoLG’s guidance. See World Bank (2014). 121 Examples of those in other countries are education, health, and social welfare services. 122 Examples include street lighting, local roads, parks, and so on. 123 Strictly speaking, it would be necessary to know which of those 27 functions the legislators of the LGA intended to be mandatory for LGUs. Based on interviews with MoLG staff, past analyses have assumed that all 27 functions are mandatory until the MoLG clarifies it to be otherwise. There is some added confusion because the assigned local functions are concurrent with similar functions at the PA level, and the LGA does not specify the division of labor between line ministries and LGUs. These problems are clearly apparent, for example, in the cases of health and education. 87 expenditures emanating from the fulfillment of specific functions. The three budget components (operational budget, enterprise budget, and development budget) only hint about the kind of public services provided by LGUs. The enterprise budget, which has information about the provision of public utilities (i.e., electricity services and water and sanitation), shows that utility services are most consistently provided by both municipalities and VCs. This is at odds with most other countries, in which public or private utility corporations provide these services. The operational budgets have functional categories, such as health services or education and culture services, but for example in the case of VCs, lots of expenditures are grouped in non-descriptive categories such as “general services”. Therefore, what these services are is not always clear.124 5.10. According to a World Bank survey from 2010, 80 percent of the municipalities provided 12 or fewer of the 27 functions and the number is significantly smaller for VCs . This is to be expected, since VCs are generally much smaller and poorer in fiscal resources per capita. The World Bank (2014) report found that smaller VCs may perform only about four functions while the larger VCs perform eight or 10 functions.125 Some over reporting may happen because purely administrative activities such as “budgeting” are included in the list of functions. During the field visit to the Kober VC, the Bank team preparing this PER found an extreme situation whereby the only service provided, garbage collection, was stopped for lack of funding.126 Significant deprivation of services across VCs appears to be the norm; many VCs do not report even public market or cemetery functions. 5.11. Both actual and approved LGU budgets show the dominance of the enterprise fund budget in the LGUs’ finances and show that these budgets are relatively well funded.127 Somewhat surprisingly, this is especially the case for VCs. For both municipalities and VCs, the approved development budgets are quite a bit larger than the actual ones, especially for municipalities. As opposed to other expenditure categories, differences in enterprise budget expenditures per capita are not as large between VCs and municipalities. For example, in 2012 VCs spent on average NIS 212 per capita and municipalities, NIS 289.128 These figures remained fairly stable between 2011 and 2013. The most important item in enterprise budgets for all LGUs is electricity services, followed by water and sanitation services as a distant second. Revenues per capita are consistently higher, sometimes much higher, than expenditures per capita, indicating that on average enterprise operations are used to cross-subsidize other activities by both VCs and municipalities. 124 The budget documents do not allow clear discernment of what are operating expenditures and capital expenditures under each function. But it appears that for the most part, expenditures of a capital nature appear in the development budget and those for operations appear in the operating budget. 125 But it must be noted that the report had a very small sample size, which may invalidate the conclusions. 126 The Kober VC bills the 700 households NIS 200 per year, which covers the costs of garbage collection and also other charges such as the ceiling tax. But the VC actually collects the billed amount from less than half of the 700 households. The VC claims that it only irregularly gets part or nothing at all from the transportation fee transfer. 127 The enterprise fund budget is similar for VCs and municipalities in that electricity and water services play a dominant role. But while in the case of municipalities other explicit categories are recorded –slaughterhouse, fruit and vegetable market, and parks, gardens, rest areas and gym—for VCs those are lumped into the “others” category. 128 Given that electricity services for municipalities and VCs in many cases, and also in some cases water services, are provided by distribution companies, it is hard to make much of the average expenditure figures per capita for the two groups of LGUs. Of course, the more cases of commercial distribution decrease the average amounts, but this does not mean necessarily that fewer services are being provided in the case of the enterprise budget. 88 Figure 78: Municipalities actual expenditure Figure 79: VC's actual expenditure allocation allocation by budget type, 2010-2012 by budget type, 2011-2013 80,000,000 500,000,000 400,000,000 60,000,000 300,000,000 40,000,000 200,000,000 100,000,000 20,000,000 0 0 2010 2011 2012 2011 2012 2013 Muni. Operating Budget Expenditure VC Operating Budget Expenditure Muni. Enterprise Fund Expenditure VC Enterprise Fund Expenditure Muni. Development Budget Expenditure VC Development Budget Expenditure Note: Actual expenditures are extrapolations based on 9-month expenditure data. Source: Palestinian Authority 5.12. Significant differences exist between VCs and municipalities in terms of the shares of different expenditure components within the operating budget, but both contain essential LGU services. For VCs, the three major categories are general administration, health services, and general services. General administration expenses, which represent around 30 percent of total VC operational expenditures, are mostly related to covering public employees’ salaries. Presumably the expenditures corresponding to health services, which represent close to 30 percent of total actual operating expenditures, correspond to the health function responsibility in the LGA, an amount that is quite significant.129 The third category of general services (representing 5 percent of the total) encompasses all services other than health, education, and culture. Within the composition of the operating (actual) budget expenditures for municipalities for 2010-2012, the three major categories are general administration (20-25 percent, approximately), health services (15-25 percent, approximately), and public works (25-35 percent, approximately).130 5.13. The next largest category is “expenses not related to a certain department,” which decreased significantly from 2010 to 2013. This likely shows an improved effort to allocate expenses to different functions over time. Other much more minor categories include: public safety and firefighting services, educational and cultural services, cultural activities, social activities, and community planning and development. Although relatively small in comparison to the enterprise (public utilities) budgets, operational budgets reveal that both municipalities and VCs seem to engage in meaningful public services likely to enhance local residents’ wellbeing. This suggests some foundation for future expansion of basic 129 Looking in more detail at a sample of LGUs’ budgets, the category of “health services” includes: salaries, cleanups, expenses for solid waste management, maintenance and rents, veterinary services, combating epidemics and diseases, clinics, and waste dumps. 130 The category of public works includes many activities such as maintenance of roads and building and facilities, street lighting, etc., and tools and equipment used by the public works department. 89 (non-utility related) local public services by VCs if additional financing for these services were made available. This suggests a better picture of LGU functions than has been regularly portrayed in the recent past. Figure 80: VC's operating budget expenditure Figure 81: Municipalities’ operating budget categories' share in total operating expenditure categories' share in total operating expenditure, actual 2011-2013 expenditure, actual 2010-2012 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Actual 2011 Actual 2012 Actual 2013 Actual 2010 Actual 2011 Actual 2012 Expenses not related to a certain department General expenses- Public Administration Community Planning and Development Health Services Social activities Cultural activities Knowledge/ educational and cultural services Security &firefighting Loan repayment Public work Health Services General Services General expenses- Public Administration Source: Palestinian Authority. Note: Actual budget for 2011-2013 is for the first 9 months for most VCs but not all, as some VCs reported the first 10 or 11 months or full year data. 5.14. Unfortunately, the ability to finance large operational expenditures has not improved much in recent years. Operational average expenditures per capita (in NIS) for VCs barely moved from 2011 to 2012 –with around NIS 54 and increased some to NIS 64.6 in 2013. On the other hand, operational average expenditures per capita for municipalities steadily decreased from NIS 195 in 2010 to NIS 166 in 2012.131 Note that operational expenditures per capita on average are almost three times larger in municipalities than in VCs, implying very different levels of service provision. 5.15. Development budget expenditures are relatively small and exhibit significant intertemporal variability.132 Generally speaking, these are capital expenditures corresponding to the functional categories appearing in the operating budget. It is interesting that on average per capita development expenditures for 131 Although there was a decrease in the category of “public works” of NIS 8, most of the decrease was in the obscure category of “expenses not related to a certain department.” 132 Development expenditures are categorized into infrastructure projects—always the largest item, education and culture projects, the second largest, and also sewage, health, and environment, entertainment projects and other. 90 VCs in 2011 were almost twice as large, NIS 177, as those for municipalities at NIS 98. However, while development expenditures per capita for municipalities increased slightly over 2010-2012, for VCs, they plummeted to NIS 41 per capita in 2013.133 5.3 Key revenue assignment issues 5.16. The LGA assigns 16 revenue sources to municipalities, of which four are local taxes: the property tax, the professional tax, the transportation tax, and the education tax. Of these four, only the property tax and the professional tax are properly local taxes and even then have many limitations as discussed below. For the transportation tax, the figure is one of a shared tax (or revenue sharing) as opposed to a local tax. And the education tax has its revenues earmarked for school maintenance and renovations managed outside the municipal budgets. The lion’s share of municipalities’ own revenues comes from a variety of fees and charges. 5.17. Local revenues are relatively small and inadequate to fund LGU expenditure assignments. As Table 24 shows, total local government revenues are below 5 percent of GDP, shown earlier to be sufficient to cover only a small minority of LGU functions and corresponding expenditure requirements. Palestinian LGU revenues are considered very low compared to local government revenue levels of countries in the region and around the world. Table 24: Revenues as % of GDP for different government units Total revenue, NIS (% of GDP) 2010 2011 2012 2013 Central government 35.9% 33.3% 34.3% 35.5% Municipalities 4.0% 4.0% 3.9% n/a VCs n/a 0.7% 0.6% 0.5% 5.18. The fact that enterprise revenues represent the largest share of LGU revenues is problematic. About two-thirds or more of VCs’ budget revenues (depending on the year) come from the enterprise budget (i.e., bills paid for public utility services such as water and electricity). The largest share of enterprise revenues pertains to electricity bills, followed by water charges and very small amounts for fees related to slaughterhouse and public markets. Municipalities’ enterprise budget revenues average about 50 percent of total revenues. In countries with well-managed public finances, one would not normally find enterprise revenues in municipal budgets at all because public utilities are either commercialized or privatized. So, the administration of these revenues is outside of the purview of local governments. In the West Bank and Gaza, not only do LGUs collect these revenues (with the exception of some municipalities where electricity revenue collection was transferred to distribution companies), but they also use them to cover expenditures unrelated to the operations of public utilities, even though those utilities are failing to meet their payment obligations and are accumulating large debts (see Section 5.5 on net lending). 133 Since municipality budgets for 2013 were not available, the possibility that the same sharp drop happened for municipalities in that year cannot be excluded. One reason for the higher development expenditures per capita in VCs vis-à-vis municipalities and also their higher volatility might be that the PA and donors try to target support to VCs, although not in a programmatic way. 91 Figure 82: Revenues by Type in VCs (Left) and municipalities (Right) 100% 100% 90% 90% 80% Operating Budget 80% Operating Budget 70% Revenue 70% Revenue 60% 60% 50% Enterprise Fund 50% Enterprise Fund 40% Revenue 40% Revenue 30% 30% Development Development 20% Budget Revenue 20% 10% 10% Budget Revenue 0% 0% Actual Actual Actual Actual Actual Actual 2011 2012 2013 2010 2011 2012 Source: Palestinian Authority 5.19. Property tax revenue collection is significantly below potential. On average, developing countries collect around a half percent of GDP in property tax revenues. The amounts reported in the municipal budgets and those intercepted by the MoF fall short of their revenue potential, which could easily be at least twice the current collections. The main reason is the lack of effectiveness in property tax administration. The reasons behind this are complex. The property tax has a long history of limiting its scope to a reduced number of municipalities (traditionally 29 out of the 136), and is not levied in VCs. This long history is associated with poor and very infrequent value assessment methods and many local residents’ unwillingness to pay the tax because of their perception of ineffectiveness on the expenditure side of municipal budgets, given low quality or nonexistent services. Another reason is absence of property valuations between 1994 and 2009. Valuations resumed in 2009, and collections increased from US$10 million in 2006 to US$30 million in 2010. But the current property assessment method is not very transparent and involves a measure of replacement value understood to be the expected or potential value of the property, which in turn seems to translate into using comparable market values for land and rental value for buildings.134 Beyond inadequate property tax collection, the tax is not applied to all 136 existing municipalities, for two main reasons: (i) capacity within the MoF to carry out value assessments in all municipalities; and (ii) a legal barrier − prior to the introduction of the property tax, the law required the existence of an urban municipal plan approved by the Ministry of Local Government (MoLG), but the MoLG has been unable to develop all the required municipal plans. 5.20. Notably, municipalities seem to be more effective at collecting property taxes than the PA. Property tax is collected by the PA and then transferred to municipalities. The education tax is levied by municipalities on the very same base used to collect property taxes, and should amount to 7 percent of the property taxes collected. The analysis conducted for this PER shows that education tax collection efforts by municipalities are consistently more effective than property tax collection by the PA. Decentralization of tax collection normally increases collection effectiveness, which could explain the inferior tax collection by the PA compared to municipalities. The difference in collection rate efficiency may not be due entirely 134 The new assessment methodology to be introduced with JICA support in 2016 will be based in the more practical “area assessment” methodology but with a large number of indicators (10 main features for eac h property as reflected in about 10 elements or indicators for each feature.) The new formula and approach will need to be approved by the Cabinet of Ministers. But with 100 items to incorporate in the assessments procedure, this may prove cumbersome. 92 to effectiveness in tax administration. It might be because the education tax is earmarked for school renovation and maintenance, which is quite visible, and because the amounts involved in the education tax are much smaller, increasing residents’ willingness to pay the education tax vis-à-vis the property tax. Nevertheless, the World Bank’s recommendation is to decentralize property tax collection to LGUs, which is in line with good practice from other countries. 5.21. On average the per capita operating revenues of VCs are less than 40 percent of those of municipalities, largely because the property tax is not levied in VCs. For example, for 2012 VCs’ per capita operating revenues were NIS 63 and NIS 165 for municipalities. The largest difference in revenue sources, as noted, is that municipalities (although not all of them) have a property tax and VCs do not. If, for example, VCs were to have the per capita property tax revenues typically received by the average municipality of around NIS 52 in 2012, their total operating revenues per capita would increase by 78 percent on average and their gap with municipalities would be reduced to only one-third of per capita operating revenues. Although these figures may not be good approximations, the property tax can and needs to play a much more important role in the operating revenues of all municipalities and VCs. 5.22. VCs actually get more transfer funding per capita than municipalities due to the allocation of revenues from the transportation tax, but this is not nearly enough to compensate for the larger revenues per capita obtained on average by the municipalities from other local taxes and fees. Municipalities also benefit from Municipal Development and Lending Fund (MDLF) grants, but VCs do not, although it would be desirable to extend this practice to them. 5.23. The fact that development revenues in the approved budgets, consistently and by far, exceed those executed in the actual budgets is a sign of weakness in budget planning. The difference between approved and actual development revenues is more pronounced for municipalities. Overestimation of development revenue may indicate some strategic gaming by LGUs in the hope of actually receiving those funds from outside sources, but they also reflect inadequate budgeting and revenue management practices, an issue revisited later. 5.24. Finally, it is noteworthy that the collection fees the PA charges LGUs for taxes it collects on their behalf are very excessive by international standards. For property taxes in the West Bank, the MoF sets the base and rate and collects the tax in a limited number of municipalities (traditionally 29 municipalities but this number is expanding). In theory, after collecting the tax, the MoF would retain 10 percent of the revenues and allocate the other 90 percent to the municipality where they were collected. A retention rate of 10 percent for supposedly administration costs is high by international standards. Central governments often do not charge for these services; if they do, the amount is closer to 3 percent of revenues. In practice, much of the property tax revenues get fully or partially intercepted by the MoF because of municipalities’ arrears in payments, mostly for electricity supplier charges and water charge deductions. The data and methodology used for the intercepts remain unclear and appear arbitrary because they are not substantiated by information on the origin and size of the arrears. These issues surrounding tax revenue intercepts are discussed further below. In Gaza, on the other hand, municipalities are fully in charge of administering the tax – setting the rate, base, and collections—and they keep 100 percent of the revenues. 93 Table 25: Revenue items’ share in overall revenue in VCs (Actual budget 2012, first 9 months) Operating Budget Part I: Revenues by PA (central government transfers) 12.15% Fees for transportation on roads 11.50% Contingency budget 0.00% Government donations, grants and aid 0.64% Part II: Revenues by VC 4.20% Part III: Services revenues (user charges) 5.33% Part IV: Loans 0.03% Part V: Miscellaneous revenues 2.37% Total Operating Budget Revenue 24.09% Enterprise Fund Water and sanitation 16.16% Electricity 45.85% Others 0.52% Total Enterprise Fund Revenue 62.54% Development Budget Allocations for development from PA Treasury 1.02% International grants and assistance 4.89% Local donations and contributions 0.08% Credits 0.00% Others 0.00% Allocations for development from municipal budget 7.38% Rotating allocation for development from previous years 0.02% Total Development Budget Revenues 13.38% Note: Enterprise fund revenues from water and sanitation include the collection of previous debt and revenues from electricity and also include repayment of previous arrears. Actual budget for 2011-2013 is for the first 9 months for most VCs but not all, as some VCs reported the first 10 or 11 months or full year data. 94 Table 26: Revenue items’ share in overall revenue in municipalities (Actual budget 2012) Operating Budget Enterprise Fund Part I: Property Taxes 9.66% Water and sanitation 14.51% Property tax 8.42% Electricity 33.04% Property tax arrears 0.61% Slaughterhouse 0.16% Others 0.63% Fruit and vegetable market 1.48% Part II: Revenues by municipality 8.49% Fish market 0.00% Part III: Services revenues (user charges) 6.01% (Parks, gardens, rest areas and gym) 0.64% Part IV: Other taxes and local fees 6.45% Others 1.33% Fees for transportation on roads 5.51% Total Enterprise Fund Revenue 51.17% Professional licenses 0.72% Professional licenses arrears 0.11% Development Budget Others 0.12% Allocations for development from PA Treasury 0.07% Part V: Revenues by PA (central 1.36% 0.95% International grants and assistance government transfers) Contingency budget 0.00% Local donations and contributions 0.06% Government donations, grants and aid 0.74% MDLF grants 0.33% International Grants 0.50% Others 0.03% Other grants 0.13% Allocations for development from municipal budget 2.65% 1.54% Rotating allocation for development from previous 0.00% Part VI: Fines years Part VII: Investment returns 4.41% Total Development Budget Revenues 4.09% Part VIII: Miscellaneous revenues 4.73% Part IV: Loans 2.09% Total Operating Budget Revenue 44.74% Note: Enterprise fund revenues from water and sanitation include the collection of previous debt and the revenues from electricity include the repayment of previous arrears. 5.4 The intergovernmental transfer system 5.25. Compared to most intergovernmental finance systems around the world, the system of transfers in the Palestinian territories is very limited, irregular, and inadequate to meet key policy objectives. At present, no formalized and regular grants or transfers are available from the PA to supplement the shortage of LGUs’ own-source revenues. Three types of transfers are currently being implemented from the central level: (i) the transportation fee; (ii) emergency allocations; and (iii) capital transfers only for municipalities (not VCs) through the MDLF. The current system is not adequately performing on any of the three main objectives of fiscal decentralization: (i) using revenue sharing or general transfers to address vertical imbalances that exist when own local revenue sources fall short of overall local expenditure needs, which clearly applies to the Palestinian territories, as shown in the previous two sections; (ii) addressing horizontal disparities across LGUs in revenue capacity and/or expenditure needs by using unconditional equalization grants—horizontal disparities, as shown below, are very pronounced in the Palestinian territories; and (iii) helping achieve central government sectoral policy objectives though implementation of conditional or specific grants—which are also lacking in the Palestinian territories. 95 5.26. The criteria applied to decide on the distribution of transfers are not always clear. For instance, the transportation fee is allocated based on a number of criteria, population being the most important one (55-75 percent of transfers are distributed according to population). The specific criteria and decision-making process to arrive at the formula remain quite opaque and the MoLG only releases the formula after it has been applied.135 This approach leaves LGUs with no means to anticipate and plan for the revenues from this transfer.136 Furthermore, emergency transfers to LGUs are allocated ad hoc by the Cabinet of Ministers; no information on the allocation criteria and decision-making process is made available on a routine basis. Finally, the MDLF represents a bright spot in the system of transfers because of the transparency and objectivity of its fund allocation process. MDLF uses an allocation formula with three factors (population, need, and performance) to allocate grant funds to municipalities for infrastructure development. But while municipalities have access to development transfer funds through the Municipal Development Project 1 (MDP1) implemented by the MDLF, no systematic transparent funding mechanism exists for the VCs, although one is being proposed. Another issue with MDP1 is its relatively small size vis-à-vis municipalities’ capital infrastructure spending needs.137 5.27. In addition to the vertical imbalances discussed earlier, the horizontal disparities among LGUs are large. First, large fiscal disparities exist between municipalities and VCs. As indicated above, in 2012 the operating budget expenditures per capita in municipalities were three times larger than those in VCs. The actual differences are likely to be smaller as many VCs only report the first 9 months of executed budget—with others reporting 10, 11, and 12 months. In addition, large horizontal disparities exist separately across municipalities and across VCs. As shown in Figure 83, expenditure dispersion is quite pronounced, with many VCs having expenditures per capita between NIS 500-1,000 when the median is well under NIS 250 per capita. The fact that some VCs are distributors of electricity and some are not is no doubt a contributing factor to that dispersion. Surprisingly, having the status of electricity distributor does not seem to impact municipal expenditures (although this impact may be disguised by the fact that some municipalities do not distribute electricity but still get “dividends” from electricity distribution companies, even if the latter are making losses). In any case, differences in expenditure levels across municipalities are very large, with relatively richer municipalities having more than double the cash expenditures per capita of relatively poorer municipalities.138 It is also noteworthy that both operating and development budgets of municipalities and VCs show a significant degree of dispersion.139 135 Besides population, other factors that have been used include: financial and administrative reform, support of VCs, support of the merging of VCs and JSCs, marginalized areas and Bedouins, support of Jerusalem, etc. 136 The operation of this transfer gets complicated by the fact that the revenues are first collected by the Ministry of Transportation, deposited in the Treasury, with the MoF transferring to the MoLG the existing pool of funds for allocations, the MoLG deciding on the formula, and finally the MoF typically intercepting the funds to be appropriated to the different LGUs because of their arrears for water and electricity. 137 By incorporating the “needs” factor, the fund is likely to have an equalization effect across municipalities, but this effect is likely to be at least partially offset by the incorporation of the “performance” factor—otherwise very desirable—because poorer municipalities are likely to be worse performers in terms of overall budget requirements. 138 For the most recent available figures for municipalities in 2012, the operating budget’s mean value for municipalities was NIS 166, with a maximum value of NIS 962, and a coefficient of variation of 0.8, indicating less dispersion than in the case of VCs but still high. For the development budget in 2012, the dispersion is more pronounced. The mean per capita expenditure was NIS 130 with a maximum value of NIS 871, and a high coefficient of variation of 1.53. 139 For the most recent available figures in 2013, the operating budget’s mean value for VCs was NIS 65, with a maximum value of NIS 842, and high coefficient of variation of 1.39. For the development budget in 2013, the dispersion is also quite pronounced. The mean per capita expenditure was NIS 45, with a maximum value of NIS 453, and a relatively high coefficient of variation of 1.53. 96 Figure 83: Distribution of total expenditure per capita in VCs (Left) and municipalities (Right) .0015 .001 2010 2011 2012 2010 2011 2012 .0008 .001 .0006 Density Density .0004 .0005 .0002 0 0 0 1000 2000 3000 0 500 1000 1500 2000 2500 Panel B. Total Expenditure per capita by year Panel B. Total Expenditure per capita by year Note: Actual budget for 2011-2013 is for the first 9 months for most VCs but not all, as some VCs reported the first 10 or 11 months or full year data. 5.28. A strong inverse correlation exists between per capita revenues and expenditures at LGU level and the incidence of poverty. Thus, if municipalities are grouped in three categories based on the incidence of poverty i.e., poorest, average and richest, one would find that overall expenditure per capita in the poorest municipalities is NIS157 less than in the richest, while operational and development expenditures per capita are NIS88 less. As for revenues, they are NIS30 less per capita in the poorest municipalities compared to the richest ones. Notably, the availability of infrastructure, including water and sewage networks, signals significant increases in per capita expenditures and revenues. 5.29. Consequently, the incidence of poverty across LGUs may depend to a certain extent on the quality of service provision. A considerable number of VCs do not provide even basic services. As shown in Table 27, in 2013 183 VCs lacked a sewage network while only 18 had one. Thirty VCs had no water network availability or only partially so; for garbage collection services, 21 VCs had none or only partial services. For municipalities, the availability of basic utility services is more general. As shown in Table 28, however, in 2013 the large majority of municipalities (83 of the 109 surveyed) still lacked a sewage network and 8 municipalities still completely lacked a garbage collection service. Road infrastructure is of poor quality in most municipalities. In the vast majority of VCs, less than 10 percent of the roads are paved. Table 27: VCs’ Availability of utility services, 2013 Availability of Availability of Availability Availability of garbage water network electricity of sewage collection service network network Not available 19 0 183 10 Partially available 11 5 2 11 Available 173 198 18 183 Source: PCBS. Table 28: Municipalities’ availability of utility services, 2013 Availability of Availability of Availability of Availability of water network electricity network sewage network garbage collection service Not available 2 2 83 8 Partially available 14 5 9 5 Available 93 102 17 96 Source: PCBS. 97 5.30. Reducing horizontal disparities in revenues and expenditures between LGUs will require ambitious reforms to mobilize additional revenues and change the allocation formulas to introduce equalization criteria, but some sort of amalgamation of smaller VCs would help. For instance, larger population VCs are able to spend less per capita, potentially signifying economies of scale. Regression analysis shows that substantial economies of scale are only present in terms of the services entering the enterprise budget, meaning that larger VCs are able to spend less per capita to deliver utility services recorded in the enterprise budget than smaller sized VCs. Analysis shows that an increase in size of 100 residents would lead to savings of approximately NIS3 per capita. 5.5 The net lending issue 5.31. “Net lending” is a serious problem facing Palestinian intergovernmental fiscal relations. Net lending arises when those LGUs that function as distributors of electricity and water services collect service fees from residents but rather than using those revenues to pay water and electricity suppliers, make only partial payment or no payment at all, thus accumulating arrears. These LGUs see collected fees as a necessary source of revenue and clearly use them to cross-subsidize other activities, including their operating and development budgets. The GoI intervenes on behalf of the Israeli Electric Corporation (IEC) and other Israeli utilities and settles electricity- and water-related payment arrears by Palestinian LGUs140 by deducting a part of the value-added tax and customs revenues that GoI collects and is obliged to transfer to the PA on a monthly basis. Besides the amounts due for electricity and water, Israeli authorities charge a late payment penalty of 11 percent. The MoF attempts to recover those losses from LGUs by intercepting local revenues from the municipal property tax, the professional permit fee, and the transportation tax that otherwise would have been transferred to them.141 Because MoF intercepts of local tax revenues are often insufficient to offset local lack of payment, arrears accumulate, and disputes arise due to the lack of information. Getting a clear picture of the extent of net lending and accumulated arrears has been historically difficult because of the lack of data and opaque procedures.142 5.32. One reason for the persistence of the net lending problem is that it has all the features of a “perverse” equilibrium. First, Israeli authorities eventually get their funds plus a fine, which means higher revenues. Many LGUs prefer to keep the “cash in hand” from electricity fees, as opposed to the promised transferred funds from local taxes, which are likely to be smaller in size anyway as some Bank staff computations also seem to confirm. The PA gets sandwiched in the middle but gets to retain the property tax, professional permit fee, and transportation tax, thus at least temporarily minimizing the blow. 5.33. A key to resolving the net lending problem is understanding what difference it makes for LGUs to be in charge of electricity distribution.143 Answering this question is important both for understanding the behavior of LGUs entangled in the net lending problem and for getting an idea of the potential financing shortfalls these LGUs would suffer if their responsibilities for electricity distribution were taken away. 140 Palestinian LGUs also have major debt arrears with the Israeli water authority (Mekorot). Both the IEC and Mekorot are primarily owned by the GoI. 141 For water arrears, the MoF appears to intercept only transport tax revenues. 142 Net lending got significantly muddled because for many years no data were available on the amounts billed by the Israeli electricity and water suppliers to Palestinian LGUs. Data were not available either on the revenue reductions by the GoI from the PA’s VAT and customs revenues. In addition, no data have been available o n the intercepts by the MoF of property tax, professional permit fees, and transportation taxes otherwise owed to LGUs. Some of these data only began to be disclosed in 2013. 143 A similar question could be asked about water distribution, but the scale of water arrears is very small in size in comparison to those corresponding to electricity, so the focus here is on the latter. 98 5.34. For VCs, the responsibility for electricity distribution is an important factor. Not only do those that distribute electricity have higher general revenues per capita, as would be logically expected given the electricity fee collections, but they also have considerably higher expenditures per capita for the operating and development budgets –leaving out the enterprise fund budget, which includes electricity revenues and expenditures. As shown in Figure 84, total revenues per capita in 2011-2013 for VCs in charge of electricity distribution can be up to four times higher than for those VCs without that responsibility. This is the case for actual and approved budgets for the average of the three years observed. VCs with electricity distribution functions were able to spend each year in the 2011-2013 period over twice as much in per capita operating and development expenditures than VCs not in charge of electricity distribution. Figure 84: Total revenue per capita for VCs with and without electricity revenues, actual 2011-2013 500.0 450.0 400.0 350.0 300.0 Actual With electricity revenue 250.0 200.0 Actual Without electricity revenue 150.0 100.0 50.0 0.0 2011 2012 2013 Average Figure 85: Total revenue per capita of municipalities with and without electricity revenues, actual 2010-2012 700.0 600.0 500.0 400.0 Actual With electricity revenue 300.0 Actual Without electricity 200.0 revenue 100.0 0.0 2010 2011 2012 Average Note: Actual budget for 2011-2013 is for the first 9 months for most VCs but not all, as some VCs reported the first 10 or 11 months or full year data. 99 Figure 86: Total operating and development expenditure per capita for VCs with and without electricity revenues, actual 2011-2013 200.0 180.0 160.0 140.0 120.0 Actual With electricity revenue 100.0 80.0 Actual Without electricity revenue 60.0 40.0 20.0 0.0 2011 2012 2013 Average Figure 87: Total operating and development expenditure per capita of municipalities with and without electricity revenues, actual 2010-2012 250.0 200.0 150.0 Actual With electricity revenue 100.0 Actual Without electricity revenue 50.0 0.0 2010 2011 2012 Average Note: Actual budget for 2011-2013 is for the first 9 months for most VCs but not all, as some VCs reported the first 10 or 11 months or full year data. 5.35. As opposed to VCs, it is not evident that municipalities benefit from being electricity distributors. As expected, municipalities in charge of electricity distribution have higher overall revenues per capita due to the presence of electricity fee collections in the enterprise budget. The difference in total revenues per capita is almost 100 percent between the two groups of municipalities on average and for every year in the 2010-2012 period (Figure 85). On average, however, being an electricity distributor does not provide municipalities the advantage of higher per capita operational and development expenditures. As shown in Figure 87, actual per capita expenditures in the operations plus development budgets in the 100 years 2010-2012 are lower, if not by much, for those municipalities involved in electricity distribution.144 Only the per capita figures for approved budgets in operational plus development expenditure are slightly higher for municipalities that are electricity distributors. But approved budgets are of little consequence in the end. It is unclear why VCs and municipalities differ in this regard. It is notable that big municipalities like Hebron, Nablus, and Ramallah do not play an important role in the “net-lending-arrears-intercept” process. This is partially due to the fact that those bigger municipalities have joined DISCOs (electricity distribution companies). The relationship between municipalities and DISCOs, on the other hand, could explain why it appears to make no budgetary difference for municipalities to be electricity distributors: municipalities that do not directly collect electricity revenues may still indirectly benefit from these revenues as DISCOs pay them “dividends”—even though profit has not been made and no real dividends exist to be paid out. Box 4: Methodology Used for Arriving at Payments Made by Each Municipality in 2012 and Total Municipality Arrears as of December 31, 2012  Electricity Deductions by IEC from MoF 2012 are extracted from a comprehensive report for all Israeli electricity deductions from clearance revenues for the years 2010-2013. Note that these deductions are not fully recorded in the accounting system at the MoF on a timely and regular basis because of the delay in providing such data by the Israeli authorities. This delay in recording electricity deductions in the accounting system at MoF indicates that calculating total arrears for each municipality is not necessarily comprehensive. (Deductions for previous years are not fully completed by the MoF.)  Payments by Municipality to MoF 2012 are calculated by adding up the following items: 1. Payments by municipality: All payments made by the municipality for the year 2012. (Cash Payments by Municipality) 2. Property Tax Intercepted 2012: All property taxes which belong to the year 2012, even if they were deducted in early 2013. (Property Tax Intercepted 2012) 3. Profession license Intercepted 2012: All Profession license fees which belong to the year 2012, even if they were deducted in early 2013. (Profession license Intercepted in 2012) 4. Transportation fees Intercepted 2012: Transportation tax for the year 2012 which is deducted later in 2014 because of the delays in getting and approving municipalities’ share at the MoLG. (Transportation fees Intercepted in 2012 )  Total Municipality Arrears to MoF until 31/12/2012 (Municipality Arrears as of 31/12/2012) are calculated by adding the balance of each municipality from the system and IEC deductions for 2012 from the IEC report. Total arrears balance until 31/12/2012 is not comprehensive for all municipalities. This is the result of the delay in recording IEC's deductions for all municipalities in the system on a regular basis; IEC's deductions were manually added to the reports. 5.36. To explain this puzzle, further analysis showed that municipalities do benefit financially if they collect electricity bills and keep some or all electricity revenues. The analysis assessed how much municipalities that are in charge of electricity distribution may come out ahead by incurring arrears –by not paying IEC—and keeping all (or part) of the revenues collected from electricity users even after the MoF proceeds to intercept property taxes, transportation fees, and professional permit fees. Two specific calculation methods were used to measure whether municipalities benefit from electricity service provisions: 144 This finding qualifies and generally does not support the previous understanding that—see, for example, World Bank (2010)—municipalities that have electricity services spend significantly more in per capita terms than those that do not. They appear to do so only when electricity charges and expenditures are included. 101 Net revenue position from electricity I= Revenues from electricity minus payments to IEC for supply of electricity minus intercepts from the MoF for all three taxes. Net revenue position from electricity II= Net revenue position from electricity I plus payments (allocations) received from the MoF for all three taxes. 5.37. The second measure takes into account that, despite the fact that the MoF has intercepted tax revenues for municipalities in arrears, at times it still makes a partial transfer of local taxes to municipalities. The reasons are not clear, but obviously when that happens those municipalities come out ahead. As shown in Figure 88, and leaving aside payment arrears accumulated by municipalities, electricity distribution clearly results in positive net revenues for municipalities. Figure 88. Net revenue position from electricity distribution for municipalities providing electricity services 45,000,000 40,000,000 35,000,000 30,000,000 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 Qalqilya A'qraba Deirastia Bidya Arrabeh Kharas Almutaheda Qafein Atteil Marj Ben Amer Qabalan Tafouh Alyaseriya Zeita Bruqin Jaba' Alzawiyah Kafl Hares Allar Nuba Beita Alchiokh Seir Ethna Sabastia Alsamou Bala' Azzoun Tarqomia Doura Salfit Qabatyeh Al Kafriyat Sourif Beit Oula Deir Ballout Kufr Aldeik Deir Alghusoon Bani Naim Beit Umar Kufr al-labad Tulkarm Net revenue position from electricity (I) Net revenue position from electricity (II) 5.38. As illustrated in Figure 88, one clear conclusion is that if municipalities decide not to pay for their electricity costs and accumulate debt arrears, practically all of them come out ahead. The MoF intercepts are not enough to leave them without revenues from collected electricity fees. 5.39. Some final questions to ask are: What if LGUs actually had to pay all their electricity and water provider costs? Will they be viable? And will they want to continue to be in charge of distributing those services as opposed to commercializing them? These questions actually have two distinct parts. The first is whether the average LGU functioning as a distributor of electricity and water services collect enough revenues on an annual basis to reach full cost recovery by breaking even or generating some profits. The second part is whether these LGUs are able to repay their accumulated debt arrears because of their lack of payment to electricity and water providers in past years. A World Bank (2014) study concluded that full cost recovery for municipalities’ electricity and water operations is simply not possible in the short term, especially considering the burden of accumulated debt arrears.145 The accumulated arrears of many municipalities run into the millions of NIS —as high as NIS93 million for Jenin municipality. Far from 145 However, the possibility of a final settlement clearly depends on the size of the accumulated arrears. The Treasury department states that 16 municipalities so far have restructured their arrears, allowing them to function with regularity with the PA agencies. One difficulty for reaching a settlement is that the claims of arrears are bidirectional. For example, Salfeet municipality claims that the PA has considerable arrears with the municipality for electricity and water charges (up to NIS 18 million) and that the PA often times has not paid its share of public infrastructure projects. 102 being able to repay these debts, it appears that municipalities would be unable to break even or generate a surplus. As reported in the World Bank (2014) study, the reason is that even after assuming a very optimistic fee payment compliance rate of around 93 percent, the current LGUs’ electricity tariff rate falls at least 10 percent short of the cost recovery rate.146 5.40. Some LGUs continue to provide electricity services instead of commercializing them via DISCOs147 because they financially benefit from being service providers. Lacking other sufficient sources of revenues, these LGUs find it convenient –if not completely unavoidable—to incur arrears and use the collection of utility revenues to finance other local expenditure responsibilities. Perhaps if LGUs were to have other (sufficient) sources of revenues, this behavior of incurring into arrears would stop. The important point is that withdrawing these improperly raised revenues –and therefore the implicit transfers from the PA --involved in the net lending operations would likely lead to a service provision crisis for many LGUs. They would simply lack alternative sources of revenue. Solving the net lending problem thus depends largely on implementing the necessary reforms to provide LGUs with sufficient alternative revenues.148 5.41. A corollary conclusion is that if electricity and water services cannot be operated at full cost recovery under current conditions, commercialization of these services may be unfeasible. Fully addressing this question will require studying how DISCOs are currently performing and what reforms could increase their viability. The final solution may require tariff rate increases or explicit direct subsidies from the PA to distributors. It must be noted that a considerable subsidy already exists via net lending,149 as the tax and fee revenue intercepts by the MoF are not nearly as much as the sums deducted by Israel authorities from the PA tax revenues. The big difference is that the net recipients of the subsidy are determined quite arbitrarily by the multilateral dynamics involved in the net lending process. 5.6 Significant budgeting issues 5.42. This area has seen recent reform efforts, including development of a unified chart of accounts and standard budget guidelines. In addition, all municipalities now report their budgets yearly to the MoLG. But even though a unified budget framework has been approved since 2008, integrating the development budget into the general budget (until then the operating budget), the reality is that the budgets of municipalities and VCs are still reported (up to 2013 at least) in the old format. In addition, several other important problems remain with the planning, execution, review and audit of LGUs’ budgets. This section highlights some of those problems. 146 Note that some LGUs, such as Salfeet municipality, achieve near 100 percent collection rates by using prepaid meters for electricity and attaching payment for other services and even taxes and fees to prepaid authorization of electricity services. Part of the problem with cost recovery for electricity is the technical losses due to wastage in the distribution network and also losses due to electricity theft; these losses tend to be larger in “special areas” such as camps, Area C, and Old Cities (see World Bank (2014). 147 Until now, 28 municipalities have joined electricity distribution companies, mostly NEDCO in the Nablus and Jenin governorates. 148 Historically, other issues have made compliance by the LGUs more difficult than should have been the case and it would be necessary to correct them too. For example, no good procedures exist for the invoicing of electricity from the IEC to the distributors and often some distributors, especially VCs, have claimed that they did not receive invoices on a regular basis. Also historically, the PA has introduced subsidies in the electricity tariff to keep social peace but has reimbursed distributors for only a fraction of the costs represented by the subsidy. 149 Net lending reduced the PA’s available revenues by 13.5 percent in 2012. The unpaid accumulated debt was estimated to be 16 percent of the PA’s total revenues at the start of 2014 (World Bank, 2014). 103 5.43. An important problem is that VCs’ and municipalities’ planned budgets are not being followed in the process of budget execution. Very often, municipalities and VCs plan for much larger budgets than they are able to execute. The mismatch between planned and executed budgets is more acute with VCs, but also significant among municipalities. The reasons why LGUs behave this way are not clear. Larger planned/approved budgets may provide LGUs with the necessary cushion to have and spend larger amounts of funds if they become available. Considerable uncertainty and unpredictability exist in the funding because basically all sources have been unreliable in the past. They may also signal their unmet financial needs. LGUs may also believe that planned budgets need to be taken seriously because at the end of the day, many LGUs have to function –as is the case for the PA—under cash rationing. Lastly, there are no serious consequences from the MoLG for getting it wrong on a consistent basis. 5.44. The systematic mismatch of planned budget and executed budget tends to nullify the usefulness of budget planning and the prioritization of expenditures. Budget execution on a sequestering basis – depending on cash availability—may negatively affect the efficiency and fairness of actual expenditure allocations. Figure 89: VCs' operating budget per capita averages, approved 2011-2013; 3/4 approved 2011-2013; and actual 2011-2013 100.0 80.0 60.0 40.0 2011 2012 2013 Operating OperatingRevenue RevenueAppoved Approved Operating OperatingRevenue RevenueApproved Approved (3/4) (3/4) Operating OperatingRevenue RevenueActual Actual Operating OperatingExpenditure Appoved Expenditure Approved Operating OperatingExpenditure Approved Expenditure (3/4) Approved (3/4) Operating OperatingExpenditure Actual Expenditure Actual 104 Figure 90: Municipalities' operating budget per capita average, approved and actual 2010-2012 200.0 190.0 180.0 170.0 160.0 150.0 140.0 2010 2011 2012 Revenue Approved Operating Revenue Operating Appoved Operating Revenue Operating Actual Revenue Actual Expenditure Approved Operating Expenditure Operating Appoved Operating Expenditure Operating Actual Expenditure Actual Note: Actual budgets for 2011-2013 are reported only for the first 9 months for most VCs but some VCs reported data for the first 10 months, 11 months, or the full year. Figure 91: VCs' enterprise fund per capita averages (approved 2011-2013; 3/4 of approved 2011- 2013; and actual 2011-2013) 340.0 240.0 140.0 2011 2012 2013 EnterpriseFund Enterprise Fund Revenue Appoved Revenue Approved EnterpriseFund Enterprise Fund Revenue Approved Revenue Approved (3/4) (3/4) EnterpriseFund Enterprise Fund Revenue Actual Revenue Actual EnterpriseFund Enterprise Fund Expenditure Appoved Expenditure Approved Fund Expenditure EnterpriseFund Enterprise Expenditure Approved Approved(3/4) (3/4) Fund Expenditure EnterpriseFund Enterprise Expenditure Actual Actual 105 Figure 92: Municipalities' enterprise fund per capita average (approved and actual 2010-2012) 490.0 440.0 390.0 340.0 290.0 240.0 2010 2011 2012 Enterprise Fund Enterprise Revenue Fund Revenue Appoved Approved Enterprise Fund Enterprise Revenue Fund Revenue Actual Actual Enterprise Fund Enterprise Expenditure Fund Expenditure Appoved Approved Enterprise Fund Enterprise Expenditure Fund Expenditure Actual Actual Note: Actual budgets for 2011-2013 are reported only for the first 9 months for most VCs but some VCs reported data for the first 10 months, 11 months, or the full year. Figure 93: VCs' development budget per capita averages (approved 2011-2013; 3/4 of approved 2011-2013; and actual 2011-2013) 240.0 190.0 140.0 90.0 40.0 2011 2012 2013 Development Revenue Development Revenue Appoved Approved Development Revenue Development Revenue Approved Approved (3/4) (3/4) Development Revenue Development Revenue Actual Actual Development Expenditure Development Expenditure Appoved Approved Development Expenditure Development Expenditure Approved Approved(3/4) (3/4) Expenditure Actual Development Expenditure Development Actual 106 Figure 94: Municipalities' development budget per capita average (approved and actual 2010-2012) 280.0 230.0 180.0 130.0 80.0 2010 2011 2012 Development Appoved RevenueApproved Development Revenue Development Actual RevenueActual Development Revenue Development Expenditure Development Expenditure Approved Appoved Development Expenditure Development Expenditure Actual Actual Note: Actual budgets for 2011-2013 are reported only for the first 9 months for most VCs but some VCs reported data for the first 10 months, 11 months, or the full year. 5.45. Both VCs and municipalities exhibit rather uniform behavior in the budget balances by budget type. VCs quite consistently show deficits in the operating and development budgets and surpluses in the enterprise fund budget (Table 29). Municipalities’ budgets show very similar patterns. This is a long- standing trend. 5.46. The average overall budget surpluses in Table 29 and Table 31 could give the misleading impression of an overall situation of fiscal health among LGUs, but this would be very far from the truth for two fundamental reasons. First, the overall budget surpluses hide the fact that many local public services for which LGUs are responsible are not provided at all or are underprovided. Second, the source of the surpluses in the enterprise budget fund is quite fictitious given that many LGUs in charge of electricity and water distribution do not fully pay their providers and incur large accumulated payment arrears. 107 Table 29: VCs’ average budget balance by budget type (Approved 2011-2013 and actual 2011-2013, first 9 months)* Actual budget Year Operational budget Enterprise Fund Development budget Overall budget balance balance balance balance 2011 -8,283 126,239 -572,560 96,042 2012 29,691 92,975 -772,597 102,217 2013 -20,113 150,752 -655,259 104,319 Approved budget Year Operational budget Enterprise Fund Development budget Overall budget balance balance balance balance 2011 9,974 189,885 -809,082 167,928 2012 43,326 181,135 -951,491 213,335 2013 34,643 167,367 -1,062,696 169,938 Note: Actual budget for 2011-2013 is for first 9 months for most VCs, but some VCs reported data for the first 10 months, 11 months, and the full year. * The figures are the average of budget balances across VCs because the overall budget balances do not match the sum of the three types of budgets. 5.47. Despite the recent advances and efforts to modernize LGUs’ budget formulation and reporting, serious limitations remain. The first important limitation is that it appears that LGUs continue to use cash accounting as opposed to accrual accounting. This allows LGUs to show budget surpluses that after accounting for payment arrears should have been significant budget deficits instead. New procedures to be introduced by the MoLG and the MoF should correct this important problem. However,150 the budget data received continue to show the use of cash accounting. 150 A separate sheet on “accounts receivable and payable” was created and mandated by the MoLG to track LGUs’ electricity and water arrears. But this separate sheet was not part of the budget documents received. According to the Local Government Budgets Preparation Manual, budgets should be prepared based on full or modified accrual accounting, but budgets are allowed to be prepared on a cash accounting basis until the switch into accrual accounting basis takes place in “future” budgets. 108 Table 30: Municipalities’ average budget balance by budget type (Approved and actual 2010-2012)* Actual budget Year Operational budget Enterprise fund Development budget Overall budget balance balance balance balance 2010 -237614 1540460 -6807262 1150275 2011 -278332 3010768 -7956031 2291003 2012 216349 1068842 -7768331 1184392 Approved budget Year Operational budget Enterprise fund Development budget Overall budget balance balance balance balance 2010 -576993 2195099 -10820388 1359537 2011 237029 1599286 -8401269 1680190 2012 549282 1355774 -9094705 1737157 5.48. Beyond the misclassification and reporting errors that are often not corrected at the MoLG, LGUs’ budgets still do not convey enough useful information for performance monitoring and decision making.151 For instance, having three separate budget documents breaks the most basic principle of budgetary unity. All potential uses of funds need to compete with each other in budget allocation decisions to take into account the opportunity cost of funds. In addition, the level of detail and transparency is especially low in operational and development budgets. Sometimes exceptional local transactions lead to awkward reporting. Budgets should be elaborated and reported according to the three standard format classifications: functional, economic, and administrative.152 Each of these classifications is informative in a different way and all three are needed to perform efficiency analysis of LGU operations, particularly to arrive at estimates of the costs per service.153 An important aspect for the relative efficiency of service provision is the current lack of administrative decentralization. Furthermore, municipalities lack autonomy on the spending side of the budgets; in particular, wages and salaries, which appear to represent around 80 percent of LGUs’ expenditures, are exclusively set by the MoLG. Lack of administrative autonomy limits efficiency and LGUs’ accountability. At any rate, pursuing greater efficiency will require much more information on the input costs associated with each service and on the measurement of service outputs. The information currently provided in the budget is insufficient to carry out any cost efficiency analysis. 151 Examples of flawed reporting practices: the education tax collected by LGUs (and earmarked for school maintenance) and supposedly reported to the MoE is not regularly reported in local budgets; garbage collection fees sometimes are reported as operating revenues when expenditures appear in the enterprise fund budget; operating revenues are recorded under different categories; the development budgets of municipalities contain only non- descriptive project titles, etc. 152 This would be achieved by following the format in the IFM Government Finance Statistics (GFS/IMF). 153 In terms of overall efficiency, no evidence exists of systematic overstaffing. World Bank (2010) reports that some municipalities are overstaffed but others are understaffed. The average number of employees per thousand inhabitants was 2.7, well within the international norms. However, no information was available to update this issue. 109 5.49. Finally, the use of ex-post audit and evaluation to measure budget performance is limited and there are no formal controls on LGU borrowing. Even though some financial audit is performed -- the General Control Office (State Audit Office) annually reviews a sample of municipalities—little or no performance evaluation of LGUs’ budgets has been conducted to understand to what extent local programs achieve their intended goals and at what cost. This is an area for expansion and initiative by the MoLG. Lastly, many LGUs borrow funds for a variety of purposes and although there appears to be no run away for borrowing, a system for information about and control of local borrowing would be desirable. This type of control is carried out by the MoF in many decentralized countries. Box 5: Nablus FY 2010 Budget Execution performance--The Case of an Outlier After analyzing budget execution performance for Nablus Municipality for the period 2010-2012, we agreed to exclude Nablus’s Budget Execution figures for the FY 2010 from our analysis due to the exceptional transactions that were recorded that year. Those transactions are attributed to the decision made by Nablus Municipality to transfer electricity services to North Electricity Distribution Company (NEDCO) in mid-2010. Following the participation of Nablus Municipality in NEDCO, the enterprise budget of Nablus Municipality lost a vital source of revenues (Electricity Project revenues). Both revenues and expenses of Electricity project were transferred from Nablus Municipality to NEDCO after 2010. The transfer of the electricity services from Nablus Municipality to NEDCO only came after an agreement between Nablus Municipality, Palestinian Electricity Regulatory Council (PERC), the MoLG, and the MoF to transfer Nablus electricity services to NEDCO in exchange for a monthly compensation from the MoF equivalent to 20 percent of its sales. This agreement was reflected in FY 2010 budget figures in an item named "MoF's contribution to compensate the Municipality for taking away Electricity collection" with an amount of NIS3,000,000 under the "Miscellaneous revenues" section in the operating budget. No information is available on the compensation scheme outcomes. This transfer agreement also stipulated that debts from customers to these municipalities would be collected by NEDCO and later transferred to the municipalities after deducting NEDCO’s collection expenses. Thus, Nablus Municipality had to extract customers’ electricity debt fro m the billing software and record it in the accounting software to confirm and emphasize previous debts due from customers. This was reflected in an item called “Revenues for Settling electricity Debt” with a value of NIS237,562,160.75 under the "Revenues by PA (Central Government Transfers)" section in the operating budget. In addition, there was another transaction in the FY 2010 budget to settle electricity debt with the MoF called "Electricity Debt settlement with MoF” of NIS91,456,831.77 under "Miscell aneous revenues" in the operating budget. On the expenditure side, an amount ofNIS 127,342,881.23 under “EXPENSES NOT RELATED TO A CERTAIN DEPARTMENT" in the Operating Expenses section was recorded in the FY 2010 budget that is related to the electricity transfer decision to NEDCO. It is very important to note that the above mentioned figures that occurred in FY 2010 represent accounting transactions that were booked in the accounting system after transferring electricity service to NEDCO in mid-2010. These accounting transactions have no financial impact. 5.7 Conclusions and recommendations for the short and medium term 5.7.1 Disentangle and solve the net lending issue 5.50. To regain transparency, certainty, and stability in the system of intergovernmental finances in the Palestinian territories, the “net lending” issue needs to be fully addressed. The problem of net lending negatively affects all components of the system. This is especially damaging because practically all components are also institutionally weak and underdeveloped. 5.51. In terms of expenditure assignments, the presence of net lending focuses the attention of many LGUs on utility services provision, which is commercialized in most decentralized countries. As a 110 consequence, the provision of many other local public services, more proper for local government, is unattended or shows substandard performance.154 For revenue assignments, the presence of net lending leads to the malfunctioning of the very few own tax revenue sources (the property tax and the professional permit fee) of many LGUs because of the necessary intercepts by the MoF due to electricity and water arrears. There is not only the unpredictability—timing of the tax funds transfer and the amount—but also the fact that it makes it difficult for LGUs to deliver services for which they cannot charge user fees—such as local roads. 5.52. The presence of net lending distorts the smooth functioning of tax revenue sharing for the transport tax. Even the functioning of what may be considered the current lone star of the system of intergovernmental finance, the program implemented by the MDLF, is affected because in some cases LGUs in arrears from net lending supposedly do not get MDLF funds disbursed. In terms of overall governance, the presence of net lending has led to strained relations between the PA and LGUs. Where there is now suspicion and disputes there should be cooperation and information exchange between the different levels of government. The net financial impact is that even though some VCs benefit from the status quo with net lending, on average, municipalities that are still in charge of electricity distribution do not. 5.53. The permanent solution to the “net lending” issue is commercialization (on a regulated basis) of all electricity distribution services and possibly water services. International experience clearly shows the benefits of commercializing utility services in terms of expenditure efficiency and service quality. Only after solving the net lending issue will revenue management and planning become less of a challenge for LGUs, especially in the West Bank. 5.54. But fully disentangling the net lending problem will require several steps before commercialization. Implementation of clearance of the arrears program with municipalities and VCs (including individual arrear reduction plans and targeted support) needs to continue. The magnitude of arrears is so large that it will take a long time for the LGUs to repay it, particularly given the slim revenue sources assigned to LGUs. To prevent continuation of the current dysfunctional system of intergovernmental finance, the PA needs to consider a partial write-off of those arrears. 5.55. The performance of DISCOs, the natural recipients of utility distribution services, needs to be further regulated and strengthened. This may require a revision of service tariffs, which may be possible if generation costs are reduced with lower international oil prices. It will not produce much progress in the end if DISCOs remain interconnected with municipal finances through ownership links and transfer flows.155 The solution to net lending will require a clean break between LGUs’ budgets and the ad hoc operation financing of distribution companies. The commercialization of utility services will also require other reform in the intergovernmental finance system to provide LGUs with alternative sources of revenues. Because of potentially significant revenue losses to those LGUs now distributing electricity, it may be desirable to reintroduce the policy of having a phase-in period of several years where those LGUs may receive a declining balance of foregone funds; this phased-in approach should facilitate acceptance of the commercialization policy, but there will be a need to apply the lessons learned from the previous 154 This list of services with substandard performance would practically include all 27 services listed in the LGA other than public utilities such as electricity and water. 155 On the basis of best international practice, it would be desirable that LGUs sell their ownership in DISCOs. The current arrangement of DISCOs distributing “profits” to LGUs perpetuates the net lending problem. 111 experience with this approach.156 Commercialization will save resources for the PA from the deduction by Israeli authorities of VAT and customs tariff revenues and from the elimination of the significant penalty charges for late payments by Israeli producers. Those saved resources could be later set in a pool of transfer funds to municipalities and VCs transitioning out of electricity distribution services. Eventually those funds will revert to financing and equalization grants with a formula taking into account population needs and LGUs’ fiscal capacity. 5.7.2 Address the problem of fragmentation and suboptimal scale of VCs 5.56. Barring the possibility of mandated consolidation of smaller VCs to address the problem of excessive local fragmentation, it is desirable to improve the governance of existing JSCs. It will also be desirable to facilitate the entry of private companies offering contracts for service delivery to groups of VCs. These measures will facilitate economies of scale among VCs and help reduce horizontal disparities. 5.57. A vibrant and more effective local government sector will be served by strengthening the Association of Municipalities and VCs. This strengthened association should serve as a counterpart to be consulted with the PA for all matters of legislative and policy significance concerning LGUs. 5.7.3 Strengthen the pillars of fiscal decentralization 5.58. Expenditure assignments: The main functional responsibilities of LGUs need to be refocused. As argued above, all public utility services (in priority order: electricity, water, sewerage, and solid waste collection) should be commercialized or privatized. The role of LGUs should be just oversight and regulation of those companies. 5.59. The rest of the local services now in the LGA will need to be revised and updated following several basic guidelines that are standard in global practice: (i) introducing a differentiation or asymmetry in functional assignments between municipalities and VCs; (ii) differentiating clearly between obligatory and voluntary services and between delegated and “own” services; and (iii) clarifying in each case for delegated and own services who is responsible for the different attributes of concurrent responsibilities (regulation, financing, and actual delivery of the service). By concentrating budget funds on “non-utility services,” it is expected that LGUs will improve their quantity and quality. 5.60. Revenue assignments: LGUs lack sufficient own tax and non-tax revenues to provide the public services they have been assigned. Failure to deliver many of these services or the reality that other services are only partially fulfilled reduces the standard of living and quality of life of local residents in most locations in the Palestinian territories. It will be important that the increased financing of LGUs comes in large part from own revenue sources as opposed to transfers. The theory and best international practice in fiscal decentralization clearly show many advantages to own revenue financing including increased accountability and fiscal responsibility of local officials. 157 Increasing local revenue collections will require that all LGUs address local residents’ lack of willingness to pay, as residents perceive no clear link between their tax and fee payments and the local services they receive. This calls for increased transparency and improved service delivery. 156 A similar approach was indeed tried with a 4 year program. However, the LGUs never received the full compensation fund largely because the funding source for the compensations was never clearly identified prior to launching the program. 157 Many other “virtues” are associated with increased local revenue autonomy, including lower levels of corrup tion and enhanced macro stability at the national level. 112 5.61. The amount of revenues available will depend on the taxes assigned, the efficiency with which they are collected, and the certainty and regularity with which those funds enter LGUs’ budgets. Advances are due on all three fronts: additional tax revenue sources should be allocated to LGUs; those now available need to be more efficiently collected and that obviously applies to the MoF and the LGUs themselves. Even when revenue potential is maximized, including the property tax, it will be clearly difficult for LGUs to cover their expenditure needs for non-utility services; many services are still underprovided or not provided at all. This situation thus requires strategic thinking by the PA on how revenue adequacy can be addressed beyond rightly pressuring LGUs to raise their own revenue collections. In addition, transfers of collected tax funds from the MoF to LGUs should be much more predictable and certain. In terms of tax autonomy, the most preferred option is to allow LGUs to set tax rates within maximum and minimum rates set by the PA. Other forms of autonomy such as setting tax bases or introducing new taxes are not desirable. It should be the PA that establishes what taxes can be used by LGUs in a legislated closed list and what the tax bases are and how they are quantified. 5.62. The best choice of local tax continues to be the property tax. But its revenue potential needs to be significantly expanded by: applying it to all municipalities and also to VCs; revamping the valuation/assessment methodology used by the MoF; and introducing “area-based assessment” methodology, although in a somewhat simplified form; and exploring the actual collection of the tax in large municipalities (a potential advantage indicated by the higher efficiency in the collection of the education tax-essentially also a property tax—by municipalities). Property value assessments should remain centralized in all cases for the time being. 5.63. The profession tax currently assigned to LGUs also needs to be modernized in its coverage and unified in its administration. This tax is currently split into the profession permit fee, which is collected by the MoF, and the crafts and industrial fees, which are collected by LGUs. The modernization of this tax requires updating its base with the professionals and businesses subject to it and revising the rates charged. In terms of administration, this is one of the few taxes where LGUs have information advantages over central authorities, so its administration and collection should be left to them. 5.64. The current list of fees assigned to LGUs should be reviewed using a cost-benefit approach. This is needed to make sure that some of these have not fallen into the category of “nuisance levies” that are more costly to collect and administer than what they really report in additional revenues. 5.65. Meaningfully raising LGUs’ revenue capacity will require assigning new taxes to local governments. Although this is likely to take place only in the medium term, some tax instruments could be good candidates. In the first place, all LGUs could be assigned “betterment levies” with rates similar to those for the property tax. Betterment levies, used in many countries at the local level, are one-time charges on the increased value of properties associated with urban improvements such as the introduction of street lighting, sidewalks, or the construction of new roads, drainage, etc., in newly developed areas. Second, the land and property transfer tax now raised by the Palestinian Land Authority for the central government could be assigned on a derivation basis to LGUs. Although this is not an ideal tax, at low rates it could become a reasonable source of revenue. Third, many countries have assigned an annual car license levy to local governments; this could be an attractive revenue source in the Palestinian territories. Fourth, consideration should be given to using a local flat-rate, piggyback personal income tax. This could take the form of a surtax with a flat rate, for example between 1-2 percent, levied by the PA at the same time as the central personal income tax and paid to the local government where the taxpayer has residence. 5.66. Transfer system: Narrowing the vertical imbalance gap between the PA and local governments will likely require improving and expanding revenue-sharing arrangements. Currently, the only form of revenue sharing is the 50/50 split of “transport tax” revenues between the PA and local governments, the latter in accordance with a changing distribution formula as annually determined by the 113 MoLG. Here one possibility will be to use the entirety of these funds to finance an equalization grant as discussed next; after all, it would seem that the spirit of the methodology used by the MoLG for the allocation of the funds has more in common with an equalization grant system than with a simple revenue sharing system. The other possibility is to keep it as revenue sharing but to increase its impact potential by allocating 100 percent of the revenues to local governments and introducing a stable-over-time formula for the distribution of the funds. Another potential candidate for revenue sharing may be the personal income tax collected by the PA. But a piggyback arrangement will be generally preferable because it will bring more accountability. 5.67. The key centerpiece of the new transfer system needs to be a conventional equalization grant across municipalities and VCs that uses an objective, stable, and explicit formula based on the expenditure needs and revenue capacity of local governments. Because it is supposed to equalize, only those LGUs with horizontal gaps would receive funds. Although arriving at an acceptable formula will not necessarily be easy, the major difficulty for the introduction of these much needed transfers will be generating a sizable pool of funds on an annually stable and predictable fashion. As already discussed, one possibility would be to dedicate 100 percent of the transportation tax funds to the equalization pool (see also World Bank (2014)). But this most likely would not be sufficient. The practice followed in many countries is to dedicate a certain percentage of central tax revenues, often lagged one year to add predictability to the funding of the equalization grant. 5.68. To complete the new system of transfers, conditional grants should be used by the PA to fund delegated responsibilities and to incentivize LGUs to complement the PA’s sectoral objectives as well as to address interjurisdictional externalities regarding the environment. For the most part, this work will need to start from scratch. In the area of capital conditional grants, however, the work of the MDLF with municipalities should be continued and expanded to VCs by introducing a VC Development Project. One way to possibly improve the performance of capital development projects is to continue to introduce more flexibility, especially for larger municipalities, and to set priorities for the use of the funds. The average municipality perceives that the amount of increase in its investment grant—as it improves its performance ranking—is disproportionately small in relation to the degree of effort required to improve its ranking. 5.69. Budgeting practices: The budget document is the most important tool for managing the efficient and equitable delivery of public services by LGUs. But the effectiveness of budgets has been compromised now for years because of the consistently unrealistic approved budgets. This has negatively affected LGUs’ service delivery capability. The MoLG needs to review the fundamental causes behind the mismatch between LGUs’ planned and actual budgets. Is it poor forecasting ability or uncertainty in revenue sources? Or is it due to built-in incentives by LGUs to provide and get approved overly optimistic planned budgets? And if so, what are feasible ways to get rid of those incentives? One measure that could help in this regard is to introduce budget performance (gap between approved and actual budgets) as a criterion in the municipal and VC development funds, punishing “large” unjustified deviations. 5.70. Although it would not by itself fix the net lending problem, increasing the transparency and information value of the budget will require that LGUs adopt accrual accounting, thus recognizing existing payment arrears in the budgets, especially in the enterprise fund budget. 5.71. Finally, budget preparation and reporting can be significantly improved by introducing functional, economic, and organizational classifications of local expenditures. On the revenue side, sources need to be classified into more useful revealing categories, such as own taxes, shared taxes, general fees (without service exchange), service fees, and the different types of transfers. In addition, loan proceeds should not be part of regular operating revenues. The budget documents also need to be unified into a single document by consolidating the three documents currently used as “operating,” “enterprise,” and 114 “development” budgets. Those changes will facilitate strengthened budget evaluation and performance practices. 115 Chapter 6: Public Investment Management158 6.1 Introduction 6.1. Public investment in the West Bank and Gaza has until recently been funded almost entirely by international development partners (IDPs). Implementation of capital investment projects has either been directly by the IDPs or through PA ministries. Direct IDP funding for public investment has been managed outside of the Treasury system and for the most part has not been reflected in the PA’s budget. Since 2009, the PA has funded a significant capital investment program that in 2013 totaled US$73 million. 6.2. This situation has resulted in an incoherent approach to public investment management (PIM), the absence of consolidated data on public investment commitments and spending, and fragmented institutional responsibilities. This chapter begins by considering the definition and scope of public investment and listing the key features of a PIM system. It then looks at the available information on the scale, composition, and performance of public investment, the current institutional framework for PIM in the PA and at local government level. This is followed by a more detailed analysis of the requirements for putting in place the key features of an integrated PIM system for the PA. The final section outlines a program of next steps actions for addressing the recommendations made in the preceding sections. 6.2 What is public investment? 6.2.1 Definition of public investment 6.3. Public investment can be defined as expenditure on fixed capital assets and related equipment costs undertaken by the general government sector, but excluding investment by commercial public corporations. It includes investment undertaken by both central and local governments. Public investment projects involve investment in the expansion or rehabilitation of public infrastructure facilities. Public investment may not be completely synonymous with capital spending, however, as expenditure on routine equipment replacement and minor capital works below a certain threshold are commonly treated as part of a government’s operational budget rather than its capital investment program. 6.2.2 PIM and the budget process 6.4. Systems and procedures for PIM are generally distinct from, but linked to the annual budget process. This is because public investment projects typically require significant lead-in times between the initial specification of a proposed capital investment project and securing financing and its inclusion in the annual budget. Furthermore, implementation of public investment projects usually extends beyond a single budget year and therefore requires procedures for managing implementation and financing over the lifetime of the project. 158 This chapter was drafted before the MoF and the MoPAD were merged into a single ministry in September 2015, hence references are made to them as two separate entities. 116 6.2.3 Key requirements of a PIM system 6.5. The key requirements of a PIM system can be broken down into eight “must-have” features. These comprise:  Strategic prioritization, project development, and initial screening – strategic guidance links project identification to national and sectoral priorities; procedures for developing project concept; criteria for initial screening to eliminate poorly conceived projects.  Project preparation and appraisal – prefeasibility study to investigate alternatives and determine whether a project is feasible; feasibility study to develop the detailed project specification and subject it to financial, economic, social, and environmental analysis.  Independent review of appraisal – appraisals submitted by ministries, departments, and agencies (MDAs) subjected to independent review by the MoF or by a specialist agency or consultancy.  Project selection and inclusion in the budget – procedures for project approval, securing financing for the total project cost and making annual allocations through the budget.  Project implementation – managing project implementation within the total approved cost and implementation plan; annual work planning; procurement management.  Project monitoring and adjustment – regular progress reporting; procedures for monitoring investment program at MDA and MoF level; requirements for justifying and approving changes in project design and costs.  Facility handover and operation – planning for project completion and reporting; inclusion of the facility investment in asset register; making financing available to cover operational costs.  Completion review and evaluation – requirements for MDAs to undertake completion reviews for all completed projects; ex-post evaluations carried out for selected projects. 6.6. The extent to which these key features exist in the West Bank and Gaza and the requirements to put them in place are discussed in Section 6.6. 6.2.4 Public investment and development expenditure 6.7. The PA budget distinguishes between recurrent expenditure and development expenditure and equates the latter with public investment. This mirrors the practice that used to be common in many developing countries that maintained separate recurrent and development budgets. While initially limited to capital investment projects, development budgets came to include significant levels of current expenditures as donor projects increasingly funded institutional reforms and operational costs linked to public service improvements. As a result, the development budget in many countries became an additional source for funding recurrent expenditure. 6.8. In countries that have adopted program-based budgeting (PBB), distinguishing between recurrent and development expenditure is largely redundant. Instead the focus of the budgeting is on allocation of the available funding to achieve planned program outputs and outcomes. Within the PBB framework, however, the concept of a capital investment program remains relevant, as an important element of expenditure policy is to manage the balance between spending on the operation of existing infrastructure and services and the capital investment required to meet future infrastructure and service delivery demands. 117 6.3 Public investment in the West Bank and Gaza – recent trends and performance 6.9. Analysis of public investment in the West Bank and Gaza is undermined by the significant level of current spending classified as development expenditure and by the substantial share of IDP- funded capital investment spending that is not included in the PA budget. This makes it impossible to determine with any degree of accuracy either the total size of the capital investment program or the distribution of public investment expenditures between different sectors. 6.3.1 Comprehensiveness of data on public investment 6.10. Consolidated data on public investment in the West Bank and Gaza are not available because many IDP-financed projects are not reflected in the PA budget. The two main sources of data on development expenditures are: (i) the PA budget and expenditure reports generated through the IFMIS database, which cover all development expenditures funded through the PA and expenditures on some IDP- funded projects; and (ii) the Development Assistance and Reform Platform (DARP) database maintained by the Ministry of Planning and Administrative Development (MoPAD), which records disbursements only on IDP-funded projects (both on-budget and off-budget projects). 6.11. No consistently applied set of criteria are used to determine which IDP-funded projects are included in the budget. As a result, the extent to which IDP-funded projects are captured in their budgets varies considerably between MDAs. Projects for which the MoF has implementation oversight responsibility are routinely recorded in the budget, mainly multi-sector and multi-agency projects overseen by the MoF’s International Relations and Projects Directorate (IRPD). Projects managed by line ministries are included in the budget if arrangements have been put in place to record expenditures in the IFMIS. There appears to be little consistency in determining whether to record IDP projects in the IFMIS, however. As a result, a number of major IDP-financed projects that are managed by MDAs and well-integrated into government operations are not recorded in the IFMIS. 6.12. Because IDP-funded projects are not implemented through the PA’s Treasury systems, the recording of IDP-funded expenditures in the IFMIS is done manually. This raises questions over the comprehensive and accuracy of the expenditure data entered into the IFMIS. The policy of the MoF and the MoPAD is to encourage IDPs to channel their funding through the Treasury system and make use of government procedures wherever this is considered feasible to ensure that IDP expenditures are automatically recorded in the IFMIS. 6.3.2 Aid-financed public investment 6.13. The MoPAD records aid disbursements, relying primarily on data provided by IDPs. All IDP funding commitments are registered in the DARP database when a funding agreement is concluded. IDPs are then required to provide disbursement details quarterly and are able to enter this data directly into DARP through a web interface. The MoPAD reviews the data entered and identifies and follows up on any discrepancies and late or incomplete entries. Since the introduction of DARP in 2010, the recording of aid commitments and disbursements has improved. Significant challenges remain nevertheless, particularly in ensuring timely and accurate reporting of disbursements.159 159 The experience of other aid recipient countries is that systemic difficulties occur in obtaining consistent and timely project disbursement data from IDPs given the different ways in which projects are reported within IDP management and accounting systems as well differences in reporting periods and fiscal years. 118 6.14. Development support recorded in the DARP database, which includes the funding of public investment projects, accounted for around 17 percent of total development assistance received during 2011-12 (Table 31). In 2012, the level of recorded development support (US$314 million) was close to three times the level of aid-funded on-budget development expenditure (US$108 million). This suggests that the major share of IDP-financed public investment remains off-budget, although the figures are not directly comparable. Receipts of development assistance are recorded on a disbursement basis whereas on- budget public investment is recorded on an expenditure basis, which can occur significantly after the aid funding has been disbursed. Furthermore, the aid disbursement data can include expenditures on technical assistance and IDP overheads that would not be classified as public investment. Reporting and classification issues also mean concerns continue over the consistency and accuracy of data in the DARP database. Table 31: Official Development Assistance receipts, 2011-12 USD Million 2011 2012 Budget Support 823 761 Development Support 336 314 o/w Economic 45 23 Governance 102 40 Infrastructure 83 137 Social 106 114 Humanitarian Support 711 656 Total 1,870 1,731 Source: MoPAD (Report to Ad Hoc Li a i s on Commi ttee, Ma rch 2013) 6.3.3 On-budget public investment – trends and composition 6.3.3.1 Recent trends 6.15. On-budget development expenditure has averaged around 7.5 percent of total PA budgetary expenditure since 2009 and around 2.9 percent of GDP (Table 32). These figures are low by international standards160 and reflect both the limited scale of public investment and the existence of significant aid- funded development expenditure not recorded in the PA budget. In 2009 and 2010, around 44 percent of on-budget development expenditure was financed through IDPs. The share of aid-funded expenditures increased sharply to 60 percent in 2013 as PA-financed development spending was cut back as part of a wider program of fiscal consolidation measures. 160 Gross public investment in Sub-Saharan Africa in 2011 averaged 7.6 percent of GDP (World Bank 2014). In OECD countries, public investment over the period 1990-2009 averaged around 12 percent of GDP (OECD 2011). 119 Table 32: PA expenditures, 2009-2013 (commitment basis) USD million 2009 2010 2011 2012 2013 Total Expenditure and Net Lending 3,405.3 3,373.5 3,693.6 3,774.1 3,134.6 Current Expenditure 2,815.6 2,839.0 3,183.6 3,252.2 2,651.5 Net Lending 374.2 236.0 139.9 278.4 296.6 Development Expenditure 215.5 298.5 370.1 243.4 186.6 o/w PA Financed Expenditure N/A N/A 145.8 129.0 73.2 DP Financed Expenditure N/A N/A 161.2 108.1 112.5 Unassigned 63.0 6.3 0.9 Development Expenditure as % of Total Expenditure 6.3% 8.8% 10.0% 6.4% 6.0% as % of GDP 3.2% 3.6% 3.8% 2.4% 1.6% (1) Commi tment = expendi ture a pproved for pa yment. Memo Item: GDP (US$ mi l l i on) 6,720 8,344 9,775 10,255 11,377 Source: (i ) MoF Fi s ca l Reports ; (i i ) IMF (GDP da ta) 6.3.3.2 Sectoral breakdown of development expenditure 6.16. General public services and public order and safety represent close to two-thirds of on-budget development expenditure, although the share fell from 75 percent in 2011 to 56 percent in 2013 (Table 33). The very high share of spending in these sectors reflects the priority given by the PA to strengthening governance and security. A further factor is that IDP funding in these sectors tends to be channeled through PA institutions either as project assistance or indirectly through budget support. By contrast, development spending in the economic affairs sector and the housing and community affairs sector, which cover the major infrastructure investments, accounted for only 14 percent of on-budget development expenditure during 2011-2013. Because these sectors receive significant off-budget investment project financing, it is difficult to determine the extent to which this reflects insufficient prioritization of public infrastructure spending in the budget planning process. This highlights the importance of having comprehensive data on all public investment expenditures to facilitate planning and prioritization of the PA’s public investment program. 120 Table 33: Development expenditure by function (cash basis), 2011-2013 2011 2012 2013 USD mill. % of Total USD mill. % of Total USD mill. % of Total 1. General Public Services 171.9 58.4% 89.2 39.3% 49.4 30.3% 2. Defence 0.0 0.0% 0.0 0.0% 0.0 0.0% 3. Public Order and Safety 49.9 17.0% 47.3 20.8% 41.3 25.4% 4. Economic Affairs 7.5 2.5% 25.6 11.3% 15.9 9.8% o/w Ministry of Agriculture 3.1 3.5 6.6 Ministry of Energy and Natural Resources 2.0 20.9 7.8 5. Environmental Protection 0.0 0.0% 0.0 0.0% 0.0 0.0% 6. Housing and Community Amenities 26.8 9.1% 21.9 9.7% 13.4 8.2% o/w Ministry of Public Works and Housing 17.9 8.7 7.0 Palestinian Water Authority 8.7 13.0 5.9 7. Health 7.3 2.5% 6.3 2.8% 4.0 2.5% 8. Recreation, Culture and Religion 13.7 4.7% 6.1 2.7% 2.6 1.6% 9. Education 11.8 4.0% 22.2 9.8% 30.5 18.8% 10. Social Protection 5.6 1.9% 8.4 3.7% 5.5 3.4% Total Development Expenditure 294.5 100.0% 227.0 100.0% 162.7 100.0% Source: MoF - Fi s ca l Reports 6.3.3.3 Economic classification of on-budget public investment 6.17. A significant share of on-budget development spending is classified as current expenditure, bringing into question the extent to which development expenditure represents public investment (Table 34). In 2013, current expenditure amounted to 42 percent of total development spending, having fallen from 54 percent in 2011. While it is common for public investment projects to involve some level of current spending, for example on the costs of project management units, the share of current spending in the West Bank and Gaza appears unusually high. This suggests that IDP-financed development projects in the West Bank and Gaza include significant levels of public service delivery expenditure. While these expenditures support the improvement of public services they would not normally be classified as public investment. This undermines the integrity and transparency of the budget by overstating the level of public investment. It can also result in “double dipping” whereby MDAs are able to game the budget process by funding service delivery from multiple sources. 121 Table 34: Development expenditure by major economic group (commitment basis), 2011-2013 USD million 2011 2012 2013 Current Expenditure 138.1 108.5 78.8 Wages and Salaries 0.2 0.0 0.0 Use of Goods and Services 85.6 74.0 57.8 Interest 0.0 0.0 0.0 Social Assistance 3.1 3.2 3.3 Subsidies 0.0 0.0 0.0 Other 49.1 31.3 17.7 Capital Expenditure 169.0 128.7 106.9 Unassigned Expenditure 63.0 6.3 0.9 Total Development Expenditure 370.1 243.4 186.6 Capital as % of Total Development 45.7% 52.9% 57.3% Source: Mi ni s try of Fi na nce - Fi s ca l Reports 6.3.3.4 Implementation performance 6.18. Implementation performance for on-budget development expenditures in 2013 was 57 percent of budgeted allocations, although this figure was significantly higher for IDP-funded expenditures (Table 35). By international standards, this is a poor score even allowing for the generally lower implementation scores associated with capital spending due to project implementation delay factors. Unrealistic budgeting is the major factor, at 49 percent for PA-funded development expenditure.161 This reflects an underfunded budget that necessitated in-year expenditure cutbacks that particularly affected development spending. For externally financed projects, unrealistic expectations about the speed of implementation, coupled with delays in procurement, have been significant factors underlying poor implementation performance. Table 35: Development expenditure implementation performance (commitment basis), 2011-2013 USD Million 2011 2012 2013 Budget Actual Impl. % Budget Actual Impl. % Budget Actual Impl. % PA Funded 280.8 161.2 57.4% 195.5 108.1 55.3% 150.9 73.2 48.5% DP Funded 222.1 145.8 65.6% 233.6 129.0 55.2% 174.1 112.5 64.6% Unassigned 63.0 6.3 0.9 Total 503.0 370.1 73.6% 429.1 243.4 56.7% 325.0 186.6 57.4% Source: Mi ni s try of Fi na nce - Fi s ca l Reports 161 By comparison, the budget implementation percentage for current expenditure in 2013 was 96.7 percent. 122 6.4 Public investment in the West Bank and Gaza – legislative and institutional framework 6.4.1 Legal and regulatory framework for PIM 6.4.1.1 Organic Budget Law 6.19. The 1998 Organic Budget Law (OBL) provides the umbrella legal framework for PIM.162 Procedures for preparing and presenting the annual budget are set out in the OBL and make reference to treatment of capital projects. These require that proposed budget allocations for ongoing capital projects are accompanied by an evaluation of progress to date and estimates of expenditure for coming years through to the completion of the project. For new projects, additional justifications and details of the project are also required. Box 6: Organic Budget Law (OBL) Provisions Relating to Public Investment The OBL makes reference to capital investment in relation to the requirements for the preparation, submission and approval of the General Budget. Taken together these provide a broad framework for PIM covering the role and responsibilities of the Budget Directorate (BD), the responsibilities of MDAs for planning and preparing their capital investment programs, and the treatment of capital investment projects in the budget presentation as the basis for their approval by the Legislative Council. Responsibilities of the BD: “Studying and evaluating all requests, programs, works and projects for which appropriations are requested following an assessment of their economic feasibility and consistency with the approved financial policies.” Article 20(5). “Participating, with other competent bodies, in the preparation of development plans.” Article 20(10) “Formulating the standards for measuring performance in implementing all p rojects and programs to which appropriations are allocated by the law.” Article 20(12). The General Budget Circular to require MDAs to include in their budget requests: “A Statement of long-term capital financing requirements. Proposals for new capital projects shall be accompanied by a full report and the necessary documents. Requirements for financing of capital projects execution shall be in accordance with the capital expenditure table specified by the BD. This table shall be updated with former work, expenditures and price increases”. Article 28(5). In the presentation of the draft budget: “Proposals in the Draft Budget Law which pertain to capital projects under implementation must include an evaluation of actual progress in the light of planned objectives and a statement of the financial requirements for the coming fiscal years. The Draft Budget Law shall contain, in the case of new capital projects, justifications and details of implementation to enable the Legislative Council to make the appropriate decisions pertaining thereto.” Article 33. 6.20. The provisions relating to capital projects in the OBL have yet to be further specified in regulations setting out the detailed specification and procedures of the PIM system. Such regulations typically cover the requirements relating to the key features of the PIM system including: (i) project 162 Law of the Organisation of the General Budget and Public Finances, No. 7 of 1998. 123 identification and initial screening; (ii) project preparation and appraisal; (iii) independent appraisal review; (iii) project approval, financing, and budgeting; (iv) project implementation; (v) capital project portfolio review and adjustment; (vi) facility handover and operation; and (v) completion review and ex-post evaluation.163 Without procedures and guidelines covering these features, there is no comprehensive framework for prioritizing and managing the capital investment program. 6.4.1.2 Procurement legislation and regulation 6.21. Procurement legislation and regulation also forms an important part of the PIM legislative framework. Considerable progress has been made towards establishing a modern procurement system with a new public procurement law (PPL) enacted in 2011164 and amended in April 2014. A Higher Council for Public Procurement Policy (HCPPP) was established, and is the entity mandated by the Law for oversight of all public procurement activity, in addition to the development of the procurement system, including policy setting, institution building, procurement documentation, guidelines and manuals, training, and public awareness campaigns. Other responsibilities include the development of procurement-related human resources, data collection and analysis of procurement performance, and establishment of a single procurement portal for the procurement system. Standard bidding documents and a national procurement manual were due to be finalized and issued by the HCPPP in 2014. Revisions to the PPL were enacted by the President in April 2014 and a comprehensive set of implementing regulations was approved by Cabinet immediately thereafter. 163 Section 6.4 discusses the key features of each of these core elements of a PIM system and the extent to which they already exist in the West Bank and Gaza. 164 Public Procurement Law, No 15 of 2011. 124 Box 7: Key of Elements of a Legal Framework for Procurement The 2013 PEFA assessment considered the PA’s new public procurement law to be consistent with all six key requirements for an effective legal and regulatory framework for public procurement. Framework Requirement The West Bank and Gaza Compliance 1. The framework is organised hierarchically and The law specifies an institutional and organizational set-up, precedence is clearly established. comprehensive procedural provisions, and transparency and accountability requirements. 2. The framework is freely and easily accessible to the The law was published in the National Gazette. public. 3. The framework provides to all procurement under- The law is applicable to all procurement activities and all taken with government funds. procuring entities using public funds. 4. The framework makes open competitive procurement The law sets a default requirement for open competitive the default and defines situations where other methods bidding and defines other permissible methods and the can be used and the justification required. thresholds and circumstance under which they may be used. 5. The framework requires public access to information The law requires preparation of annual procurement plans, on government procurement plans, bidding the publication of bidding opportunities and contract awards. oopportunities, contract awards and resolution of All procurement plans, notices and award decisions and complaints. other procurement documentation (legislation, regulations, standardized bidding documents, procurement manual, etc.) would be published at a single procurement portal. Also the law provides for a complaint mechanism and dispute review body. 6. The framework provides for an independent The establishment of the dispute review unit is one of the key administrative procurement review process for handling innovations of the new Law which puts in place a mechanism complaints prior to contract signature. for the independent aadministrative review of complaints from aggrieved bidders concerning alleged non-compliance by procuring entities in conducting procurement proceedings. The Dispute Review Unit will essentially be a panel from which individual committees will be constituted to consider particular complaints. The DRU is to be provided secretariat services by the HCPPP. Source: The West Bank and Gaza PEFA Public Financial Management Performance Report (2013) and 2014 update. 6.4.2 Institutional mandates for PIM165 6.22. Although the OBL provides the MoF with a comprehensive PIM mandate, in practice—until the 2015 merger of MoF and MoPAD--the PIM function has involved departments in the MoF and the MoPAD. This arises primarily because of the significant IDP funding of capital projects that is currently 165 The Ministry of Finance and the Ministry of Planning have recently been merged. The new organizational structure for the merged Ministry is not clear yet, but many issues discussed herein and related to the coordination between the two Ministries should be resolved through the merger. 125 outside of the budget, the existence of a separate set of procedures overseen by the MoPAD for identifying, appraising, and approving aid-funded projects, and the absence of a dedicated PIM capability in the MoF. 6.4.2.3 Ministry of Finance 6.23. Within the MoF, responsibility for PIM is divided between the Budget Directorate (BD) and the International Relations and Projects Directorate (IRPD). The BD’s responsibilities for managing the capital investment program are set out in the OBL. Currently this role is carried out primarily in relation to PA-financed projects and on-budget IDP-funded projects and is focused around the presentation, approval, and implementation of the annual budget. There is no procedure for approving the total estimated cost and earmarking multi-year financing for projects that are scheduled to be implemented over more than a single fiscal year. While the BD is responsible for assessing project funding requests for economic feasibility and consistency with government policies, the capacity to carry out this function has not yet been put in place. Procedures and standards are similarly not in place for measuring and monitoring project implementation performance. 6.24. The MoF recently established a Public Investment Management Unit (PIMU) located within the BD, with one budget analyst assigned to the unit. Detailed terms of reference for the unit have yet to be developed. Technical assistance to support establishment of the unit and development of PIM procedures and guidelines is currently being funded by DfID. 6.25. The IRPD is responsible for administering major IDP-funded multi-sector projects that fall under the management responsibility of the MoF and for monitoring other IDP-funded on-budget projects. To carry out these functions, IRPD staff include engineers and procurement specialists. The IRPD through its Macro-Fiscal Unit is responsible for preparing the monthly fiscal development reports that include details of budget allocations and expenditures for a list of major projects. 6.4.2.4 Ministry of Planning and Administrative Development (MoPAD) 6.26. The MoPAD’s engagement in PIM derives from its strategic planning and aid coordination functions. The policy and research, sectoral, and monitoring and evaluation directorates are responsible for the development and monitoring of the Palestinian National Development Plan (PNDP) and for coordinating and overseeing sector-level strategic planning undertaken by the major sector MDAs. The 2014-16 PNDP was approved by the Cabinet in March 2014 and sets out the PA’s overall policies and priorities for the three-year plan period. The MoPAD considered that there was little consistency in the quality of the sectoral strategies prepared for the draft 2014-16 PNDP, reflecting MDAs’ limited planning capabilities. 6.27. The Aid Management and Coordination Directorate (AMCD) provides the gateway for IDPs in their dealings with the PA. It is responsible for negotiating agreements with IDPs and for ensuring that program and project agreements are in accordance with the PA’s development priorities. It provides guidance on the preparation and management of aid-financed projects, and participates in reviews of the execution of IDP program and project reviews. Additionally, it manages the DARP in which all IDP-funded projects must be registered and IDP project disbursements recorded. 6.4.2.5 MDAs 6.28. PIM responsibilities in MDAs are spread between the planning directorate, the budget and finance directorate, and the technical directorates and dedicated project implementation units (PIUs) responsible for project implementation. The planning directorate plays the lead role in coordinating the preparation and implementation of the sector strategy and often acts as point of contact with IDPs. While 126 planning directorates may be involved in the preparation of IDP- and PA-funded projects, they generally lack significant project appraisal capabilities. 6.29. The involvement of budget and finance directorates is generally limited to on-budget projects. The directorate is responsible for managing the preparation of the MDA’s budget submission, dealing with the MoF during budget execution, and for processing payments through the IFMIS. Significant progress has been made in recent years in strengthening basic budgeting and financial management procedures, linked to the development and commissioning of the IFMIS. 6.30. Project implementation is carried out by the technical departments in MDAs in collaboration with the Ministry of Public Works and Housing (MoPWH). Significant capabilities for PIM exist both in the MoPWH and in some of the PIUs set up within MDAs to manage IDP-funded projects. Both the MoPWH and the PIUs have been able to draw on the ready availability of civil engineering skills within the West Bank and Gaza, and on the capabilities within the private sector to undertake construction contracts. Box 8: Capabilities in the Construction Sector Palestinians have long provided a significant source of civil engineering skills in the region. These capabilities facilitated the rapid expansion of construction activity in the West Bank and Gaza in the late 1990s. Currently these capabilities are not being fully utilised with, for example, housing construction running at less than half the 30,000 units that were being constructed in the early 2000s. The construction of the new Rawabi City complex is an example of the capabilities within the sector to undertake major construction projects in a challenging environment. Most Palestinian companies are relatively small and therefore limited in the size of contract that they can undertake. 6.4.2.6 Institutional mapping to key PIM system features 6.31. The fragmented institutional framework for PIM reflects differences in the way projects are financed and whether they are on-budget or off-budget. For off-budget projects, the MoF is not involved and it is the MoPAD that oversees the process of project identification and appraisal carried out primarily by IDPs, and the approval of the project for financing by the IDPs. For on-budget projects, the MoF is involved with projects that are to be managed by the IRPD, but has a very limited role for other projects. The requirements of IDPs add further complexity to the current PIM procedures. As a result, no unified PA process exists for project identification, appraisal, and approval that applies to on-budget projects, whether financed by IDPs or by the PA. 127 Box 9: Oversight Responsibilities for Key Stages in Capital Project Cycle IDP-funded - on-budget IDP-funded – off-budget Stage PA-funded projects projects projects MoPAD – PNDP sets out MoPAD – PNDP sets out overall MoPAD – PNDP sets out overall overall and intersectoral and intersectoral priorities. and intersectoral priorities. 1. Strategic priorities. Prioritization, MDAs – preparation of MDAs – preparation of sector MDAs – preparation of sector Project sector strategies strategies strategies Development and Initial MDAs – project MDAs and/or IDPs – identificat- IDPs – identification and scoping. Screening identification and scoping. ion and scoping MoF – screening through MoPAD – screening prior to MoPAD – screening prior to budget process detailed preparation/appraisal. detailed preparation/appraisal. 2. Project MDAs – linked to budget IDPs – follows IDP procedures IDPs – follows IDP procedures Preparation process, often limited in with MDA input. often with MDA input. and Appraisal scope. 3. Independent MoF – but not currently IDPs - following IDP procedures. IDPs - following IDP procedures, Review of carried out. Appraisal MoF – approval through IDPs – approval following IDP IDPs – approval following IDP inclusion in budget. procedures procedures 4. Project MoPAD/MoF – ratification. MoPAD – ratification Selection and Inclusion in the MoF - annual allocation MoF – annual estimates included IDP – multi-year funding approval budget through budget process. in budget. for total project cost IDP – multi-year funding approval for total project cost. MDAs MDAs – through IDP-funded MDAs through IDP-funded PIUs 5. Project PIUs (currently no IDP-funded or by Implementation projects use Treasury systems) IDPs directly. MoF – but currently limited IDPs – monitoring follows IDP IDPs following IDP procedures. 6. Project to financial monitoring; procedures MoPAD through DARP system. Monitoring and MoF – financial monitoring. Adjustment. MoPAD through DARP system. 7. Facility MDAs MDAs IDP – handover on IDP Handover and implemented projects Operation MDAs MoF – but not currently MoF – but not currently carried MoPAD – but currently relies on 8. Completion carried out out IDPs. Review and MoPAD – but currently relies on IDPs – follows IDP procedures. Evaluation IDPs. IDPs – follows IDP procedures. 128 6.32. The development of an integrated PIM framework will require consolidation of the institutional responsibility for PIM oversight and coordination. The key requirements are:  Clarifying the respective responsibilities of the MoF and the MoPAD. Specifically: o The MoF oversees the key PIM procedures including project identification, initial screening, appraisal, approval, implementation, expenditure reporting, and monitoring and evaluation at project level. The MoPAD has an important role within these procedures in relation to its responsibilities for the planning and monitoring of government strategies and for aid management and effectiveness. This would include identifying opportunities for IDP financing and securing such financing, participating in the project screening and selection process, and assisting in bringing IDP financing on-budget where this is feasible. o The MoPAD oversees: (i) the PNDP and sectoral planning processes that provide the strategic policy framework within which capital investment priorities are determined; and (ii) identifying and negotiating IDP funding for capital investment projects, and monitoring IDP disbursements as part of its wider aid coordination and management mandate.  Determining the distribution of PIM related functions within the MoF between the BD/PIMU and the IRPD. Assuming that the existing departmental structure of the MoF is retained, this should distinguish between: o The role of the PIMU in the oversight of the PIM framework and processes covering: (i) project identification, appraisal, financing, and approval; (ii) review of budget requests for capital investment projects; (iii) implementation of capital investment program; (iv) project completion, handover, and evaluation; and (v) analysis of issues relating to capital investment program prioritization and performance. o The role of IRPD in facilitating the implementation of IDP-funded projects and in bringing them on-budget. This would cover: (i) establishing specific arrangements and procedures for bringing individual IDP-funded projects on-budget; (ii) ensuring that PIUs for IDP- funded projects are aware of the requirements and procedures for recording project expenditures in the IFMIS; and (iii) conducting periodic checks to ensure that expenditures are being recorded. The IRPD would also continue its existing coordination and oversight responsibilities on IDP-funded projects for which the MoF has implementation responsibility. 6.5 PIM at local government level 6.33. This section discusses public investment at local government level in the context of its role within the wider PIM framework in the West Bank and Gaza. It is primarily concerned with the local investment coordination, oversight, and consolidation roles of the MoLG. PIM procedures at the LGU level are discussed briefly but were not analyzed and assessed in detail. 6.5.1 Public investment by LGUs 6.34. Although LGUs are required to submit their budgets for ratification by the MoLG and to provide regular expenditure reports, local government budgetary allocations and expenditures are not consolidated across all LGUs. An exercise undertaken by the MDLF estimated total development expenditure allocations across the 135 municipalities in 2012 at around US$240 million, of which 17 percent was to be financed through the MDLF. 129 6.35. LGU development expenditures are financed from a number of sources. These include: (i) the PA budget; (ii) the MDLF; (iii) IDP grants provided to LGUs; (iv) projects directly implemented and funded by IDPs; and (v) LGU own resources. In some of the larger and longer established municipalities, these different funding sources are included in the municipality budget, which provides comprehensive coverage of municipal-level public investment. Elsewhere, coverage is less complete, particularly where the projects are implemented directly either by a PA ministry or by a contractor or NGO directly contracted by an IDP. 6.36. The division of responsibility between the PA and LGUs for funding local-level infrastructure is not always clear cut. This applies particularly in the health and education sectors, where in theory LGUs are responsible for construction of schools and primary health facilities, while the relevant PA ministry funds staff and other current costs. In practice, construction may be funded either through PA-level projects or through LGU projects financed by an IDP. In the latter case, problems can arise where facilities are designed and constructed without regard to the appropriate level of PA service provision and related facility design standards, resulting in facilities that are overdesigned and underutilized. Box 10: Deir Al-Hatab Health Centre Deir al-Hatab is a village in northern West Bank with a population of around 2,200. It is located 2 km from the Nablus urban boundary and 2 km from the larger village of Salim, which has a population of 5,100. A major project recently undertaken by the VC involved the construction of a Village Health Centre. Funding was provided by the Sultanate of Oman, a firm based in East Jerusalem was directly contracted to implement the project. The health center, which cost around US$700,000, was not reflected in the budget of the VC. The center is staffed by the MoH with three full-time nurses/auxiliaries and a doctor who attends twice a week. As such, it operates as a health clinic rather than a larger health center. Consequently, a major part of the facilities (which include a laboratory, a maternity room, and a small number of inpatient beds) is either currently not used or is substantially underutilized. With other health facilities close by, the MoH has no plans to increase the level of staffing in Deir al-Hatab. 6.5.2 Local governments 6.37. In some respects, the municipality level of local government is further ahead putting in place a PIM framework than the national level authorities. This reflects both the long history of local government in some municipalities as well as the work of the MDLF in putting in place basic PIM procedures for LGUs. As a result key elements of the capital investment cycle are in place in many major municipalities. While these procedures are mandatory for accessing MDLF funding, LGUs have applied them more widely in the implementation of capital investment programs. The MDLF has also developed a procurement manual for use by LGUs. 6.38. To date, the capital investment financing provided through the MDLF has been provided as grants. This reflects the MDLF’s initial focus on strengthening planning, administrative, and financial procedures and capabilities in LGUs and the limited creditworthiness of municipalities. The development of a lending window is part of the MDLF’s strategic plan. Further strengthening institutional capacities in LGUs and building their credit-worthiness are likely to continue to be a more immediate priority. 130 Box 11: MDLF - Capital Investment Cycle Management The procedures developed for MDLF-funded projects provide for a rigorous capital investment project cycle that could be applied more generally to the funding of LGU projects from the PA budget. The procedures, which are linked to the annual budget of municipalities, comprise the following stages: 1: Mobilization Phase  A ranking exercise that assesses municipality systems capabilities to manage a capital investment program. The ranking is a factor in determining the resource envelope for each municipality.  Determination of the respective funding envelopes for each municipality. This takes into account population, needs, ranking, and performance criteria.  Developing documentation (circulars, forms etc.) and holding workshops which groups of municipalities. 2: Preliminary (Application and Approval) Phase  Two stage process for municipalities to apply for capital project funding:  The initial application provides descriptions and justifications for the projects proposed for MDLF funding, but does not include technical documentation.  Following conditional approval, the municipality prepared a detailed technical proposal. MDLF engineering advisors may provide technical support to municipalities during this phase and assist with engaging consultants where needed.  The technical proposal should provide a full justification for the projects covering strategic justification and priority; (ii) social, environmental, participation, economic and financial analysis; and (iii) procurement milestones and operations and maintenances plans.  Evaluation and review of municipal proposals by MDLF Technical Departments which make recommendations for rejection or approval of the project. Municipalities informed and invited to sign the grant implementation agreement (GIA). 3. Pre-Implementation Phase  Following signature of the GIA, the municipality prepares the bid package, which after endorsement, is reviewed and approved by the MDLF.  The municipality procurement committee initiates an open bidding process, receives the bids and prepares the bid evaluation report, which is forwarded to the MDLF.  Following receipt of a non-objection from MDLF, a contract is prepared and signed with the contractor. 4. Implementation Phase  The municipality issues a project start order to the contractor.  During contract implementation the municipality undertakes: (i) supervision in accordance with the contract agreement; (ii) monitoring of contractor compliance and implementation; (iii) verification and approval of payments; (iv) preparation of milestones progress reports; (v) issuing of variation orders; and (vi) issuing of completion and handover certificates. 5. Project Evaluation Phase  Evaluations undertaken by the MDLF for achievement of project objectives with assessment report made available on the MDLF website. 6. Project Operations and Maintenance Phase.  During the first operational year of the project, the municipality shall prepare semi-annual reports showing covering compatibility with operations and maintenance plan, problems and solutions and recommendations for future operation. Source: MDLF (2009). 131 6.5.3 Oversight and consolidation of public investment in LGUs 6.39. The 1997 Local Government Law (LGL)166 requires LGUs to submit their budgets for ratification by the MoLG and to provide regular reports on budget implementation. The MoLG also specifies the format in which budgets should be presented and expenditures reported. This should allow for the MoLG to provide a consolidated analysis of LGU budgets and expenditure and also to separate out capital expenditures. The LGL requires the MoLG to approve all projects to be implemented by LGUs. While this includes the projects to be financed by LGU borrowing and guarantees to be approved, there is no specific provision for appraisal of the fiscal risks related to such funding. In practice, this has not been an issue to date, since few instances of LGUs financing public investment through these mechanisms have arisen. 6.40. A further set of issues relates to the categorization in the PA budget of financing provided from the PA for capital projects implemented by LGUs. This should be classified in the PA budget as a transfer to the LGU for development/capital spending purposes. The funding should then be shown in the LGU’s budget as an offsetting revenue entry and the planned spending on the project as expenditure. This mechanism permits the netting out of the transfer in the consolidation of the PA and LGU budgets and prevents the expenditure from being double counted in the consolidation of general government expenditures. The PA budget does not currently show allocations for transfers for development/capital projects, however. Furthermore, the overall local government support transfer included the PA budget is quite limited, suggesting that it does not include significant transfers for capital projects.167 This suggests that investment allocations from the PA budget to LGUs are either recorded as capital expenditure rather than transfers, or are off-budget in cases where they are funded by IDPs. For example, spending of the MDLF is not recorded in the PA budget despite it being a PA institution and the most important source of capital investment funding for LGUs. 6.41. The MoLG is proposing to establish a separate public investment unit to carry out its responsibilities relating to the approval of LGU capital investment projects and the funding and monitoring of LGU capital investment programs. This is an important initiative, but should be part a broader program in the MoLG to put in place systems and procedures to support its policy development and LGU oversight roles. This should include consolidated financial reporting and monitoring of the local government sector. 6.42. The MDLF has played an extremely positive role both in strengthening PIM procedures in LGUs and in providing a vehicle for channeling IDP funding to LGU budgets. This has been facilitated by high-quality management and a strong professional staff. At the same time, the support being provided by the MDLF to LGUs contrasts with the limited capacities within the MoLG to undertake its mandated policy and oversight for LGU budgetary operations and public investment. Consideration should, therefore, be given to either extending the MDLF’s mandate to include support to the MoLG to develop its own budgetary and public investment monitoring and oversight capabilities, or to funding this requirement from another source. 166 Law on Local Government, No. (1) of 1997. 167 In 2013, transfers from the PA budget for local government support amounted to US$36.9 million (commitment basis). 132 6.6 Putting in place the key features and elements of a PIM system 6.43. This section looks at each of the eight key features of a PIM system, the extent to which they are in place in the West Bank and Gaza and the relative priorities in developing the required systems and capabilities. The analysis highlights the absence of a well-mapped out set of PIM procedures and the dependence on IDP systems for project identification, preparation, and appraisal, and in some cases, in managing project implementation. 6.44. Building a framework of PIM procedures and the capacity to implement these procedures is a major task that will take considerable time. While a large number of recommendations are made, relatively few are critically important in the short to medium term. In other less critical areas, a start can be made in introducing basic procedures, recognizing that these will need to be further developed over the medium and long term. 6.6.1 Feature 1 -- Strategic priorities, project development, and preliminary screening 6.6.1.1 Key elements 6.45. Clear and transparent procedures for the identification and initial screening of capital investment projects should be considered the foundation of an effective PIM system . Eliminating unsuitable and unsound projects at this stage helps to ensure that resources are not wasted on subsequent detailed preparation and appraisal. It is also easier to eliminate projects through an initial screening process rather than later when they may have built up a considerable political and bureaucratic momentum, which can be difficult to stop. The key are requirements at this stage are:  The existence of credible strategic plans at national and sectoral level to guide public investment strategy and prioritization between and within major infrastructure and service delivery sectors.  A formal process through which MDAs identify and develop initial specification for proposed capital investments. Typically this involves the development of a project concept paper (PCP) setting out: (i) the strategic case for the project and the alternative options considered; (ii) a brief description covering the main components of the project, and related activities and outputs; (iii) a preliminary financial analysis; (iv) a preliminary economic and social analysis; and (v) the key issues to be further investigated during detailed preparation and appraisal.  An initial screening process to evaluate whether the project concept: (i) is compatible with the strategic goals of government; (ii) reflects the MDA’s priorities and mandates; (iii) represents the most promising of the available options; (iv) is appropriate in scope and scale; and (v) stands a realistic prospect of being funded. 6.6.1.2 Existing practice in the West Bank and Gaza 6.46. The three-year PNDP sets out the government’s strategic policies and priorities while MDA sector strategies are intended to identify public investment priorities at sector level. The MoPAD reviews projects proposed for aid financing for consistency with the PNDP and sector strategies. The MoF similarly emphasizes in the budget circular that proposed capital investment projects should address PNDP and sector strategy priorities. The link between sector strategies and budget allocations is further reinforced in the new budget calendar introduced for PBB. This requires MDAs to prepare and update annually policy statements at budget program level, which are then used in determining spending priorities and in the evaluation of spending plans. Significant weaknesses exist in the current set of sector strategies, however, reflecting capability limitations for policy development and strategic planning in MDAs. 133 6.47. No procedures are yet in place for the initial scoping of capital investment projects leading to the preparation of a PCP and for the initial screening of projects prior to detailed preparation. Such procedures should operate independently of the budget preparation cycle since it is only at a later stage, following detailed preparation and appraisal, that a project would be ready for financing and inclusion in the budget. PCPs should follow a common set of instructions and be required for all investment projects, whether financed by the PA or IDPs. 6.48. The initial screening at project concept stage is typically carried out by a capital investment project screening committee that may be chaired either by the Finance Minister or by a senior official in the MoF.168 The committee meets three or four times during the year to consider a batch of proposed projects at each meeting. For the West Bank and Gaza, a public investment screening committee could include representatives of: (i) the MoPAD to provide an opinion on the project’s consistency with PA policies and priorities and on its suitability for IDP financing; (ii) the MoPWH to comment on likely technical feasibility and cost issues on infrastructure projects; and (iii) the MoLG. MDAs should attend meetings at which projects they have submitted are being considered, or if particular issues are being discussed that impact their sectoral mandates. 168 In countries with strong PIM systems, responsibility for screening projects below a certain threshold may be delegated to the relevant sector MDA. 134 Box 12: Typical Format for a Project Concept Paper (PCP)# A PCP should be a concise document perhaps 4-6 pages in length to which additional detail and justification, if available, may be attached. The format of a PCP varies from country to country but typically follows a structure similar to that shown below. 1. Administrative Information and Description Project name MDA and subordinate unit responsible for proposing project Budget program to which the project belongs 2. Strategic Case for the Project To which strategic objectives and priorities of the government will the project contribute and how The specific objectives of the project The alternative options considered for achieving this objective and justification for selection of the preferred option 3. Project Description Location of the project The main components of the project The main activities to be undertaken under each component Proposed implementation arrangements (organization responsible for implementation, oversight arrangements by MDA, etc.) The expected outputs/results of the project 4. Financial Analysis of the Project The total estimated cost of the project and the basis on which it has been determined The duration of the project and expected phasing of expenditure over the life of the project Potential sources of IDP financing for the project (if known) The estimated downstream recurrent costs arising from the project and their impact on the MDA’s budget 5. Preliminary Economic and Social Analysis The expected economic benefits arising of the project - description and scale (if known) Possible social, gender, and environmental impacts 6. Key Issues to be Further Investigated during Formal Preparation and Appraisal Main issues to be considered during project preparation Whether both prefeasibility and feasibility studies will be required – duration and costs Proposed time-frame for preparation # Other alternative titles for a PCP include Project Profile and Project Identification Document. 6.6.2 Feature 2 -- Project preparation and appraisal 6.6.2.1 Key elements 6.49. Project preparation and appraisal should be coordinated and overseen by the sponsoring MDA even though the detailed work may be undertaken by specialist consultants often funded directly by an IDP. The MDA is responsible for ensuring that projects have been adequately prepared. Typically project preparation involves carrying out: (i) a prefeasibility study to identify and test project design alternatives and determine whether the project is feasible; and (ii) a feasibility study to further develop the selected design alternatives, develop detailed costing, and carry out financial, economic, social, and environmental analyses. For smaller and relatively simple projects, it may be possible to combine the prefeasibility and feasibility analysis within a single study. 135 6.6.2.2 Existing practice in the West Bank and Gaza 6.50. In the West Bank and Gaza, IDPs have until recently taken the primary responsibility for capital investment project preparation and appraisal. The role of MDAs in the process has varied, reflecting differences in IDP approach and the extent to which the MDA has relevant capabilities and experience. PA oversight responsibility of the preparation and appraisal of IDP-funded has been exercised through the AMCD in the MoPAD. The MoF, although having formal responsibility for PIM, has had little involvement in the preparation of IDP-funded capital project preparation and appraisal. 6.51. The introduction of a limited PA-funded capital investment program has resulted in MDAs becoming more directly involved in formal project preparation and appraisal. It has also resulted in a parallel set of procedures linked to the annual budget process that operates under the oversight responsibility of the MoF. No formal guidelines have been issued by the MoF covering specific requirements for the preparation and appraisal of capital investment projects. 136 Box 13: Prefeasibility and Feasibility Studies The key features of prefeasibility and feasibility studies are shown below. Note that there is no fixed dividing line between a prefeasibility and feasibility study. In general, a feasibility study is more detailed and expensive to carry out and the prefeasibility helps to ensure that only projects that are likely to be technically and economically sound move forward to the feasibility stage. In some cases, for example, a project is using standard building designs and where site conditions are not a major factor, it may be possible to combine prefeasibility and feasibility analysis within a single study. Prefeasibility Study Feasibility Study The prefeasibility study further analyses and develops The feasibility study investigates in greater depth the the proposals contained in the PCP. It should include: proposed project solution. Typically, it concludes some The justification of the proposed investment: or all of the following: the existing (negative) situation and why action is Evaluation of alternatives technological solutions. required; Site surveys, particularly where site conditions have a the cause factors underlying the existing situation; major impact on engineering and technological consistency of the proposed investment with sectoral choices. investment priorities. Preliminary architectural and engineering design (final Technical and economic solutions considered: designs are often considered to be part of project description and of the alternative options implementation). investigated; Identification of regulatory and permit requirements analysis of alternatives – technical, economic and and their implications for project phasing. social; Detailed costing based on preliminary design. justification for the selection of the preferred Schedule for project implementation with physical and alternative. financial progress indicators. Basic technical details relating to the proposed investment: Detailed risk and sustainability assessments. site information – area, ownership, access to services, Full cost-benefit analysis. geological and hydrological characteristics, etc. Environmental and social impact assessments. outline specification of proposed physical investment including whether standard designs are to be followed. proposed implementation arrangements including for monitoring of project progress and performance. Estimates of project costs and benefits. Assessment of major risks (including institutional and budgetary) Identification of further information and analytical requirements for the feasibility study. 6.52. Few MDAs are familiar with managing or carrying out prefeasibility and feasibility studies. An exception is the MoPWH, which has significant experience in project preparation and in managing feasibility studies for major infrastructure projects. However, its expertise primarily relates to the technical aspects of public works projects and not to their wider social and economic appraisal, which should be a central element in public investment decision-making. 6.53. Once procedures for the screening at the project concept stage are in place, the MoF will need to focus on establishing standards for project appraisal and develop capabilities within MDAs for managing the appraisal process. The initial emphasis should be on providing guidelines and training to MDAs on managing the appraisal process. The guidelines should cover: (i) key appraisal concepts; (ii) stages in the appraisal process (particularly prefeasibility and feasibility analysis); (iii) role and 137 responsibilities of MDAs in managing the appraisal process; and (iv) use of IDPs to support the appraisal process. The relatively high skills demands mean that over the medium term, detailed project preparation and appraisal will continue to be carried out by IDPs and consultants. Establishing oversight by and accountability to MDAs in the appraisal process should be the initial aims. Securing dedicated funding from IDPs for supporting MDAs in carrying out project appraisals, including for proposed PA-funded investment projects, could help in building capabilities for managing and carrying out appraisals. 6.6.3 Feature 3 -- Independent review of appraisal 6.6.3.1 Key elements 6.54. A characteristic of strong PIM systems is the existence of an appraisal challenge function for major investment projects that is independent of the sponsoring sector MDA. The authority of the UK Treasury to “call in” larger projects for review prior to funding is one example. Apart from verifying the quality of the feasibility study, the appraisal challenge also allows the project to be assessed from a broader outlook than that of the sponsoring sector MDA. The knowledge that the appraisal might be challenged also helps to prevent the rigor of a feasibility study becoming subordinated to the internal politics and bureaucratic inertia of an MDA. 6.55. The appraisal challenge function is typically located within the PIMU in the MoF. While it requires professional capabilities within the unit to manage the process, consultants are commonly engaged to carry out appraisal reviews, particularly for the largest projects for which multi-disciplinary appraisal teams may be required. 6.6.3.2 Existing practice in the West Bank and Gaza 6.56. Little capability currently exists in the PA for reviewing investment project appraisal studies. One of the functions of the newly established PIMU should be to assess appraisal studies to determine whether they have been carried out to an adequate standard and provide a realistic assessment of the project’s technical feasibility and its costs and benefits. Where this is not the case, a project should be referred back to its sponsoring MDA for further analysis and justification. Establishing the capability in PIMU to carry out appraisal reviews should be seen as a medium- to long-term objective. In the short term, the MoF should consider engaging independent consultants to carry out appraisal reviews of large and complex projects that carry significant risks. It is recommended that the MoF seek IDP funding to carry out appraisal reviews of the largest and most complex projects that carry significant risks. 6.6.4 Feature 4 -- Project selection and inclusion in the budget 6.6.4.1 Key elements 6.57. Project selection involves the approval of a project for financing and its eventual inclusion in the annual budget. Typically approval takes place two to three times during the year when a group of projects that have passed through the appraisal process are prioritized and considered for financing against the available investment resource envelope. Projects to be funded by an IDP often come to the selection stage with the source of funding already identified. In this case, the role of the selection process is to endorse (or reject) the project appraisal and confirm that the project remains a priority for funding from the IDPs ’ resources. The project selection process is typically overseen by a high-level committee, often chaired by the Minister of Finance. This can be the same committee that reviews PCPs at the earlier initial screening stage. 138 6.58. Following approval of the project, the financing arrangements are finalized and the project is included in the budget proposal. For IDP-funded projects this involves negotiating a financing agreement with the IDP. For domestically financed projects, it requires earmarking funding for the full investment cost and over the lifetime of the project. Inclusion in the next annual budget can then follow. While the budget approves funding for a single year, the initiation of an investment project also implies a multi-year funding commitment. For this reason, it is considered good practice for the annual budget to include an annex providing more detailed information on the capital investment projects being funded, including their total estimated cost and phasing of expenditure. Box 14: Capital Investment Project Annex to the Budget The key information that would need to be maintained in the capital investment project annex to the annual budget comprises: A brief narrative summarizing the objectives of the project, the investment to be undertaken, the location of the proposed investment, and the body responsible for implementation of the project. The total estimated cost of the project (as approved by the Public Investment Screening and Selection Committee). The sources of financing for the project. A table showing actual expenditures incurred in previous years, a revised estimated of expenditure for the current year, budgeted expenditure for the coming fiscal year, and forecasted expenditure for a further two years, and the balance of expenditure required to complete the project. It can be useful for this table to show a breakdown of expenditure by funding source. Estimated annual recurrent costs arising on completion of the project. 6.6.4.2 Existing practice in the West Bank and Gaza 6.59. Existing PA procedures for the selection of capital investment projects and their inclusion in the budget are incomplete. No formal selection process provides for approval of a project and its total estimated cost before it is included in the budget. PA-funded projects are listed in the budget against an overall PA-funded allocation but without individual project budgets being shown. The allocation of financing takes place subsequently following approval of the budget, and may be further adjusted through the fund release mechanism. As a result, MDAs have little certainty over which projects will be financed, making it impossible to initiate procurement early in the fiscal year and making it difficult to manage commitments on ongoing projects. The arrangements for financing IDP-funded projects, including making provisions for any PA-funded counterpart contribution, are linked to project financing agreements and as a result are better established and more orderly. To improve the system, it is recommended that the MoF develops and introduces procedures for project selection and approval. The proposed Capital Investment Project Screening and Selection Committee should be responsible for project approval. The MoF should start showing in the detailed budget presentation allocations for each capital investment project that is included in the PA budget. This should include the total estimated cost of the project, expenditure in previous years, budget expenditure for the coming year, spending in the following years through to completion of the project, and subsequent annual operational costs. 6.6.5 Feature 5 -- Project implementation 6.6.5.1 Key elements 6.60. Arrangements for project management must be in place to ensure that implementation proceeds without delay. The MoF should provide guidelines on project management arrangements covering: (i) roles of project managers and PIUs; (ii) the delegated authorities of the project management 139 team; (iii) annual work plan requirements and contents; (iv) overview of procurement policies and procedures; (v) financial management and reporting requirements; (vi) arrangements for monitoring of and reporting on implementation progress; (vii) requests and justifications for revisions to project scheduling and changes in design; and (viii) arrangements for project completion reporting and review. 6.61. Predictable and timely funding releases and payments to suppliers are prerequisites for the efficient implementation of projects. This can be facilitated by project managers including estimates of monthly cash flow requirements in annual work plans, and should be updated at least quarterly. It also requires that the MoF manage fiscal operations and government cash flow so that financial resources are available when needed. 6.62. A sound and efficiently operating procurement system should facilitate project implementation and ensure value for money. This requires an effective legal and regulatory framework to be in place that: (i) sets out clearly the hierarchy, responsibilities, and practices in the procurement system; (ii) is transparent and accessible to the public; (iii) is applied to all transactions using government funds; (iv) makes competitive procurement practices the default and specifies the circumstances under which other procurement methods may be used; (v) provides free public access to government procurement plans, bidding opportunities, contract awards, and information on complaints resolutions; and (vi) provides for an independent procedure for handling complaints prior to contract signature. 6.63. A project accounting system captures and records all project costs against the total approved cost of the project over the lifetime of the project. Project implementation needs to be managed, controlled, and monitored against both the total estimated project cost and the budget allocation for the current fiscal year. 6.6.5.2 Existing practice in the West Bank and Gaza 6.64. While the PA has made considerable progress in recent years in strengthening its core public finance management systems, a robust and comprehensive framework for the implementation of capital investment projects is not yet in place. No general guidelines have been issued by the PA to managers of capital projects. In the case of IDP-funded projects implemented by dedicated PIUs, procedures tend to follow the requirements and procedures of the IDP. 6.65. Uncertainty over funding allocations and cash releases is a major constraint to the implementation of PA-financed projects. This reflects a number of factors, including overly optimistic assumptions in setting of the budget frameworks, the lack of specified project-level allocations in the budget documentation, and uncertainties over the realization of government revenue forecasts. 6.66. As noted in the preceding section, significant progress has been made in putting in place a legislative and institutional framework for public procurement that reflects international standards. The challenge moving forward will be to ensure the efficient implementation of the new framework and reduce the significant procurement delays that currently occur. 6.67. PA procedures only provide for accounting of project expenditures against annual budget allocations and not against total estimated project cost. In the case of IDP-funded projects, the basis of the total estimated cost provides the basis for project management and accounting. To improve the system, the MoF should develop general guidelines for project managers/PIUs that apply both to IDP and PA-funded projects. It should also monitor project implementation across capital investment programs to identify and address systemic issues and causes of delay. Finally, a project management module should be introduced into the IFMIS to track expenditures over the life of a project against total approved project cost. 140 6.6.6 Feature 6 -- Project monitoring and adjustment 6.6.6.1 Key elements 6.68. Periodic capital investment portfolio performance reviews that are integrated into the budget preparation and management cycle should be carried out. The purpose of these reviews is to identify requirements and justifications for: (i) significant adjustments to project design and to implementation schedules; (ii) changes to the total approved cost of the project and planned phasing of expenditures; and (iii) cases where changes in circumstance or poor performance require curtailment or closure of a project. Portfolio reviews are typically organized by the MoF and involve a series of sectoral review meetings at which MDAs report on their respective investment programs.169 A performance review meeting held at the beginning of the fiscal year can be used to review project work plans and forecasts of cash requirements. A subsequent meeting prior to MDAs preparing their budget requests can be used to identify those investment projects on which there are particular issues to be addressed during budget preparation. 6.69. Procedures are required for making adjustments to project specifications, implementation schedules, and total estimated capital costs. These should identify the limits above which approval of the Legislature is required. For example, increases in the total estimated costs of projects could be permitted up to the level of 20 percent without requiring the project to be referred back to the Capital Investment Project Screening and Selection Committee. 6.6.6.2 Existing practice in the West Bank and Gaza 6.70. No PA procedures are in place to undertake regular capital investment portfolio reviews. The MoPAD organizes periodic meetings with IDPs to review progress on the implementation of their assistance programs, but this is not specifically focused on capital investment projects and is not linked to the budgeting process. Some IDPs undertake periodic reviews of the capital investment projects that they are funding. The BD is generally not involved in these reviews. There is no periodic review process for PA- financed capital projects. The budget only shows funding allocations for a single year and the total estimated cost of capital projects does not form part of the budget review and approval process. To improve the system, and once a comprehensive register of public investment projects has been established, the MoF should introduce sector-level investment program portfolio reviews linked to the annual budget planning and management cycle. Furthermore, rules for approving changes to the total estimated cost of capital investment projects should be introduced, identifying limits above which approval of the Cabinet and the Legislature is required. 6.6.7 Feature 7 -- Facility handover and operation 6.6.7.1 Key elements 6.71. Procedures should be in place to ensure the smooth transition to facility operation on completion of a project. These should cover: (i) handover of the project on completion and ensuring that the necessary capabilities and resources for facility operation are in place; (ii) addressing faults during the initial operation period; (iii) updating asset registers; and (iv) active monitoring of service delivery over time and the benefits derived from the investment. 169 In larger states with highly delegated administrative structures, investment portfolio performance reviews may be carried out at the sector MDA level and the review findings reported to the MoF. 141 6.6.7.2 Existing practice in the West Bank and Gaza 6.72. The extent to which MDAs prepare for the facility operation on completion and handover of new facility varies considerably. Experience from a number of IDP-funded projects, where facilities have remained idle for lengthy periods following project completion, has highlighted the importance of estimating downstream operating costs during project design and assessing whether they will be affordable. At least 12 months prior to project completions, MDAs should prepare for facility operation to ensure that management, staffing, and operational resources are in place. The MoF should develop general guidelines for MDAs on preparing for facility handover and operation. Requirements will vary significantly according to the type of project – and tend to be greater for social infrastructure projects such schools, training facilities, and hospitals. Monitoring of facility performance should be done as part of the wider program- level monitoring of service delivery operations. The MoF should monitor preparations for facility handover and operation, including elaboration of required current financing, through the capital investment portfolio review process. 6.6.8 Feature 8 -- Basic completion review and evaluation 6.6.8.1 Key elements 6.73. All projects should be subject to basic completion review. Completion reviews assess the efficiency with which the projects outputs have been achieved. They typically focus on issues relating to the technical design, implementation experiences, and management arrangements. They look at actual project costs compared with the original estimates and compare these with costs of other similar projects. An effective program of completion reviews elicits key lessons for the design and management of new projects. It also helps to identify requirements for strengthening program-level management and financial procedures. Project completion reviews are typically carried out or commissioned by the PIU responsible for the project, often within guidelines set by the responsible MDA or by the MoF. 6.74. Ex-post evaluations should be carried out for a sample of major projects and capital investment spending on major service delivery programs.170 They should focus on whether envisaged outcomes are likely to be realized and the extent to which the outputs of the project have contributed to these outcomes. An ex-post evaluation is usually carried out three to five years after project completion to capture information on the extent to which outcomes are being realized. Terms of reference for evaluations should emphasize the independence of the evaluation. For this reason, they should not be carried out by the managing PIU responsible for the project or program. While evaluations of small projects may be undertaken by the responsible sector MDA, it is common for major evaluations to be commissioned by the MoF, particularly if the findings are considered likely to have wider relevance for other sectors or MDAs. 6.6.8.2 Existing practice in the West Bank and Gaza 6.75. No procedures are in place in either the MoF or the MoPAD for undertaking project completion reviews or more detailed ex-post evaluations. Currently project completion reviews and ex- post evaluations are only carried out by the funding IDP where required by its own internal procedures. The results may not always be shared with the MDA, the MoF, and the MoPAD. Steps should be taken by the MoF to put in place arrangements for carrying out project completion reviews on all projects and ex-post evaluations on selected major projects. The MoF should require project completion reports to be prepared 170 School construction is an example where it can be more appropriate to undertake evaluations at program rather than individual project level. 142 by MDAs for all capital investment projects included in the PA budget. It should also develop standard format specifications and guidelines for preparing these reports and provide training to MDAs in their application. Initially, project completion reviews would need to rely on the resources available from the IDP financing the project and provide for the formal presentation and discussion of the review findings with MDA, the MoF, and the MoPAD. 6.7 Priority next steps 6.76. In the short term, the MoF should prioritize putting in place those critical elements of a PIM framework that can be expected to have immediate impact. In this context, the immediate priorities are to:  Define the scope and composition of public investment. The MoF and the MoPAD should establish a clear definition of public investment and the circumstances under which non-capital spending may be included. One implication is that where IDP-funded projects are funding substantial current spending on service delivery, these costs should in future be treated as recurrent expenditure and no longer considered part of the capital investment program (Timing – immediate).  Elaborate a comprehensive and integrated PIM framework. The MoF, in consultation with the MoPAD, should prepare a Cabinet Paper setting out the main elements of the proposed comprehensive and integrated PIM framework and the respective institutional mandates for PIM consistent with the OBL. The paper should also set out the steps and associated timetable for progressively implementing the framework. Linked to this, the MoF should finalize the terms of reference for the PIMU and secure support for the development of its work program and technical capacities (Timing – immediate).  Improve coverage of public investment in the PA budget. The MoPAD, the MoF, and IDPs should agree to introduce additional measures to bring all capital investment projects financed IDPs on- budget to the extent that this is feasible. The MoF should at the same time improve procedures for and compliance in recording these expenditures in the IFMIS. To do that, the authorities need to develop capabilities for analyzing public investment trends and expenditures. Implementation of such measures will result in more comprehensive data becoming available on public investment allocations and expenditures. Making use of this data to analyze public investment trends, composition and implementation performance should be an important task in the MoF’s monitoring of the implementation of the capital investment program (Timing – immediate).  Rationalize and further elaborate the institutional framework for PIM. An immediate priority is to consolidate responsibility for development and oversight of a unified PIM framework within the MoF. It is recommended that the MoF prepare a Cabinet Paper settelucidating the main elements of a comprehensive and integrated PIM framework. The paper should clarify institutional responsibilities for PIM consistent with the provision of the OBL covering the roles and responsibilities of: the MoPAD and the MoF with respect to IDP-funded projects so that IDP- funded capital investment projects can be handled within the integrated PIM framework; and BD/PIMU and IRPD in managing and supporting the implementation of a comprehensive and integrated capital investment program (Timing – short term). The establishment of the PIMU is a promising first step in this process, but must be followed up by the establishment of basic professional capacities within the PIMU and the preparation of detailed terms of reference and work plans for the unit. The PIMU should remain a relatively small unit within the BD, perhaps building up to a staff of 4-5 analysts over the medium term. Requirements for related support in developing PIMU capabilities should be determined (Timing – short term). This should be followed by preparation of regulations setting out the core elements of the PIM system and associated 143 responsibilities and preparation of more detailed technical and operational guidelines (Timing – short to medium term).  Develop procedures for specification and initial screening of capital investment projects. The MoF, in consultation with the MoPAD and following Cabinet approval, should introduce procedures for capital project identification, development, and initial screening. This should involve: o Requiring MDAs to prepare PCPs for all capital investment projects, whether to be financed by the PA or IDPs. The MoF should prepare guidelines for PCP preparation and provide training and backstopping to MDAs in their application. (Timing – prepare guidelines in 2016 so new procedures to be introduced in 2017.) o Establishing a Capital Investment Project Screening and Selection Committee chaired by the Minister of Finance to review and approve PCPs prior to project projects proceeding to detailed preparation and appraisal. (Timing – committee operational by mid-2017.)  Introduce project selection and approval and budgeting procedures. The MoF, in collaboration with the MoPAD, should require that: (i) all appraised projects be reviewed and prioritized by the Project Screening and Selection Committee; and (ii) for the Committee to approve projects that have been shown to be technically sound and of high priority and which have reasonable prospects of being financed. (Timing – from mid-2017.)  Produce budget and record of public investment expenditures. The MoF should introduce budgeting and expenditure recording procedures for capital investment projects that take account of multi- year implementation schedules. Specifically: o A capital projects annex should be introduced into the budget presentation that provides for each project: (i) a brief narrative description of the objectives and main components of the project; and (ii) a table showing the total approved cost of the project, actual expenditure in previous expenditure, and planned expenditure by year to project completion. (Timing – if possible starting with the 2017 Budget, otherwise the 2018 Budget.) o A project management module should be introduced into the IFMIS that provides for expenditures to be tracked over the life of a capital project against the total approved project costs. (Timing – if possible for the 2017 Budget, otherwise the 2018 Budget.)  Consolidate public investment at local government level into the broader national PIM frameworkWhile good progress is being made with developing basic PIM capacities in municipalities, the wider policy, oversight, and consolidation functions of the MoLG are weak. Strengthening these functions should be seen as an integral element in establishing a national PIM framework at PA level and is therefore included in this PER’s recommendations. In terms of actions, a Public Investment Unit should be established in the MoLG to exercise its policy and oversight roles. The MoLG capabilities should be developed to analyse LGU budgetary operations and prepare and publish quarterly consolidated local government budget implementation reports. Also, PA funding of LGU-implemented capital investment projects should be shown as a capital project transfer in PA and LGU budgets. While these tasks should be seen as high priority, they are not immediately critical to the establishment of a PA-level PIM system. 6.77. These critical quick-win initiatives will help to establish a more ordered and comprehensive approach to PIM. Other high-priority measures identified in the previous sections will be required to develop the underlying processes and required capacities. Particularly important will be: (i) development of capacities for project preparation and appraisal in MDAs; (ii) strengthened procedures and capacities to 144 support project implementation, particularly in the areas of work planning and procurement; (iii) establishment of a basic monitoring system, focusing initially on financial monitoring and sector capital investment program portfolio reviews; and (iv) introduction of project completion reviews. These measures should be accompanied by preparation of PIM regulations and detailed guidelines. Annex VIII provides a summary matrix of the assessment made in Section 6.7 of key PIM system features and associated recommendations. 145 References AbdulRahim, et al., Maternal and child health in the occupied Palestinian Territory. Lancet, 2009. 373(9667): p. 967-977. Bloom, D.E., et al., The Global Economic Burden of Noncommunicable Diseases, 2011, World Economic Forum: Geneva. Center, E.N.I. €13 million to East Jerusalem Hospitals to provide specialized health services to Palestinians. 2013 [cited 2014 September 18]; Available from: http://www.enpi- info.eu/medportal/news/latest/33739/%E2%82%AC13-million-to-East-Jerusalem-Hospitals-to- provide-specialized-health-services-to-Palestinians Chingos M.C, The false promise of class size reduction, 2011, Center for American Progress: Washington, DC. Chingos M.C, The Impact of a Universal Class-Size Reduction Policy: Evidence from Florida’s Statewide Mandate, August 2010, Program on Education Policy and Governance Working Papers Series, Harvard University: Cambridge, Massachusetts. Gilbert, M., Brief report to UNRWA: The Gaza Health Sector as of June 2014, 2014, United Nations Relief and Works Agency: Gaza City, Palestinian territories. IHME. Data Visualisations: GBD Arraw Diagram. 2014 [cited 2014 January 10]; Available from: http://vizhub.healthdata.org/irank/arrow.php. IMF, Report on macroeconomic developments and outlook, I.M. Fund, Editor 2014, International Monetary Fund: Ramallah, Palestinian territories. Lago-Peñas, S. and Martinez-Vazquez, J., The Challenge of Local Government Size: Theoretical Perspectives, International Experience and Policy Reform, 2013, Edward Elgar Pub: Northampton, Massachusetts. Martinez-Vazquez, J., et al., The Impact of Fiscal Decentralization: A survey. International Center for Public Policy Working Paper 15-02, 2015. Mataria, A., et al., Catastrophic Healthcare Payments and Impoverishment in the Occupied Palestinian Territory. Applied Economic Health Policy, 2010. 8(6): p. 393-405. MoF, Data files received from MoF on wage bill, 2014, Ministry of Finance: Ramallah, Palestinian territories. MoH, Annual Health Report 2003, 2003, Ministry of Health: Ramallah, Palestinian territories. MoH, Annual Health Report 2009, 2010, Ministry of Health Ramallah, Palestinian territories. 146 MoH, National Nutrition Surveillance System 2010 report, 2010, Ministry of Health: Ramallah, Ministry of Health. MoH, Annual Health Report 2010, 2011, Ministry of Health Ramallah, Palestinian territories. MoH, Annual Health Report 2011, 2012, Ministry of Health Ramallah, Palestinian territories. MoH, Annual Health Report 2012, 2013, Ministry of Health Ramallah,Palestinian territories. MoH, Annual Mid-Year Health Report 2013, 2013, Ministry of Health Ramallah, Palestinian territories. MoH, Health Annual Report Palestina 2013, June, 2014, Palestinian Health Information Center: Ramallah, Palestinian territories. MoH, MoH budget files, 2014, Ministry of Health Ramallah, Palestinian territories. MoH, Referral patient and expenditure files, 2014, MoH: Ramallah, Palestinian territories. MoH, Essential Drug List and Cost of Essential Drugs, 2014, Ministry of Health: Ramallah, Palestinian territories. MoH, Medical Specialists Personnel Files, 2014, MoH: Ramallah, Palestinian territories. Municipal Development and Lending Fund, MDLF Technical Manual, December 2009. OECD, Regions at a Glance, 2011, OECD: Paris. Ozaltin, E. and J. Alaref, Managing Health Sector Dual Practice in the Palestinian territories: A policy note, 2014, World Bank: Washington, DC. Palestinian National Authority, Palestinian National Development Plan 2011–13, April 2011. PCBS, Palestinians at the end of year 2006, 2006, PCBS: Ramallah, Palestinian territories. PCBS and MoH, National Health Accounts 2000-2008 Main Findings, 2011, Palestinian Central Bureau of Statistics: Ramallah, Palestinian territories. PCBS, Labor Force Survey: Annual Report 2010, 2011, Palestinian Central Bureau of Statistics: Ramallah, Palestinian territories. PCBS and MoH, National Health Accounts 2009-2010, 2012, Palestinian Central Bureau of Statistics: Ramallah, Palestinian territories. PCBS, On the Eve of the International Population Day, 2012, PCBS: Ramallah, Palestinian territories. 147 PCBS, Palestinians at the end of 2012, PCBS, Editor 2012, PCBS: Ramallah, Palestinian territories. PCBS, UNICEF, and UNFPA, Palestinian Family Health Survey, 2013, Palestinian Central Bureau of Statistics: Ramallah, Palestinian territories. PCBS and MoH, National Health Accounts 2011-2012 Main Findings, 2014, Palestinian Central Bureau of Statistics: Ramallah, Palestinian territories. PCBS, National Accounts, 2014, Palestinian Central Bureau of Statistics: Ramallah, Palestinian territories. PCBS, Palestine in Figures, 2014, Palestinian Central Bureau of Statistics: Ramallah, Palestinian territories. PCBS, Maternal and Child Health Indicators, in Palestinian Central Bureau of Statistics 2014. . Schoenbaum, M., A.K. Afifi, and R.J. Deckelbaum, Strengthening the Palestinian Health System. 2005 Arlington, VA: RAND. Tandon, A. and C. Cashin, Assessing Public Expenditure on Health from a Fiscal Space Perspective, in HNP Discussion Paper 2010, World Bank: Washington, DC. WHO and MSH, International Drug Price Indicator Guide, 2010, Management Sciences for Health: Cambridge, MA. WHO, STEPwide Approach to Chronic Disease Risk Factor Surveillance, 2011, World Health Organization: Ramallah. WHO, Regional Data Observatory - Eastern Mediterranean Region, 2014, WHO-EMRO: WHO- EMRO. World Bank, “Government Employment and Pay in Global Perspective: A Selective Synthesis of International Facts, Policies and Experience”, 1999, World Bank: Washington, DC. World Bank, Reforming Prudently Under Pressure 2008, World Bank Washington, DC. World Bank, Technical Assistance on Health Financing Component I: Outside Referrals Report on the Analysis of Referral Data (Internal Report), 2013, World Bank: Ramallah, Palestinian territories. World Bank, The West Bank and Gaza PEFA Public Financial Management Performance Report, 2013, World Bank: Washington, DC. World Bank, World Development Indicators, 2014, World Bank Washington, DC. 148 World Bank, Population Projection Tables by Country and Group, in HNP Stats 2014, World Bank: Washington, DC. World Bank, The West Bank and Gaza PEFA Public Financial Management Performance Report, 2014 update, World Bank: Washington, DC. World Bank, Report on Referral Data Analysis, 2014, World Bank: Washington, DC. World Bank, United Arab Emirates Public Health Expenditure Review (Health and Education), 2014, World Bank: Washington, DC. World Bank, Net Lending Study for West Bank and Gaza, 2014, World Bank: Washington, DC. World Bank, A Background Note on Villages in the West Bank, Village Governance and Service Delivery, March, 2014. World Economic Forum/World Health Organization, From Burden to "Best Buys": Reducing the Economic Impact of NCDs in Low- and Middle-Income Countries, 2011. 149 Annex I: Health Status Figure AI.1: Life expectancy at birth, international comparisons, 1990-2012 West Bank and Gaza 75.0 Middle East & North Africa (developing only) Lower middle income 70.0 65.0 60.0 55.0 Source: World Bank (2014). Figure AI.2: Trends in child mortality per 1,000 live births, Palestinian territories, 1990-2009 35.0 33.3 33.0 IMR U5MR Linear (IMR) Linear (U5MR) 31.0 29.2 29.0 28.4 27.0 26.1 25.0 23.0 23.4 23.6 21.0 22.4 19.0 18.9 17.0 15.0 1994-1990 1999-1995 2004-2000 2009-2005 Sources: PCBS, UNICEF, and UNFPA (2013). Note: IMR = Infant mortality rate and U5MR = Under-5 mortality rate. 150 Figure AI.3: Child mortality per 1,000 live births, West Bank versus Gaza, 2005-2009 West Bank Gaza 30 25 20 15 10 5 0 Infant Mortality Rate Under-5 mortality rate Sources: PCBS (2013), UNICEF (2013), and UNFPA (2013). Figure AI.4: Child mortality per 1,000 live births by type of residence, Palestinian territories, 2005-2009 30 Urban Rural Camps 25 20 15 10 5 0 Infant Mortality Rate Under-5 mortality rate Sources: PCBS (2013), UNICEF (2013), and UNFPA (2013). 151 Figure AI.5: Maternal mortality ratio per 100,000 live births, international comparisons, 2013 300 Palestinian Territories Egypt Jordan MENA LMIC 250 per 100,000 live births 200 150 100 50 0 Palestinian Egypt Jordan MENA LMIC Territories Source: World Bank (2014). Figure AI.6: Reported cancer cases by governorate andgender, West Bank, 2012 250 Males Females 200 Reported Cancer Cases 150 100 50 0 Tubas Jenin Tulkarm Nablus Qalqiliya Salfit Ramallah - Jericho - Al Jerusalem Bethlehem Hebron Al Bireh Aghwar Source: MoH (2013). 152 Table AI.1: Communicable diseases in the Palestinian territories, 2013 Indicator Incidence rate per 100,000 population AFP/100,000 child < 15y 2.1 Poliomyelitis 0 Rubella 0 Diphtheria 0 Plague 0 Cholera 0 Rabies 0 Yellow Fewer 0 West Nile Fever 0 Malaria 0 Neonatal tetanus 0 AIDS 0.1 HIV Infection 0 Hepatitis A 31.5 Hepatitis B Cases 0.5 Hepatitis C cases 0 Pulmonary TB 0.7 Extra pulmonary TB 0.1 Cutaneous Leishmaniasis 4.9 Visceral Leishmaniasis 0.1 Meningococcal Disease 2 Aseptic meningitis 65.7 Other bacterial meningitis 10.6 Homophilus influenza meningitis 0.1 Brucellosis 5.4 Source: MoH (2014) 153 Annex II: Health Expenditure Figure AII.1: Trends in gross domestic product and total health expenditure, Palestinian territories, 2000-2012 12,000 GDP THE Linear (GDP) Linear (THE) 10,000 in US dollars (millions) 8,000 6,000 4,000 2,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.2: Gross domestic product, total health expenditures and government health expenditures, Palestinian territories, 2000-2012 GDP THE PHE as % of GDP (right axis) 12,000.0 6.0 10,000.0 5.0 in US dollars (millions) 8,000.0 4.0 percentage 6,000.0 3.0 4,000.0 2.0 2,000.0 1.0 - 0.0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 154 Figure AII.3: Trends in health financing indicators, Palestinian territories, 2000-2012 GHE as % of GDP OOP as % GDP PvtHE as % of GDP NPISH as % of GDP External as % of GDP THE as % of GDP (right axis) 16 14 12 percentage 10 8 6 4 2 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.4: Trends in public expenditures on health, Palestinian territories, 2000-2012 600 in US dollars (millions) 500 400 300 200 100 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 155 Figure AII.5: Trends in out of pocket expenditures, Palestinian territories, 2000-2012 140 120 in US dollars (millions) 100 80 60 40 20 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.6: Health financing indicators, international comparisons, 2012 Palestinian Territories MENA LMICs 60 50 40 percentage 30 20 10 0 PHE as % THE THE as % GDP OOP as % THE Sources: Data for the Palestinian territories are retrieved from PCBS (2014), PCBS (2012), and PCBS (2009); others from World Bank (2014). 156 Figure AII.7: Elements of total health expenditure, Palestinian territories, 2012 1% 18% PHE as % of THE 2% 39% OOP as % of THE PvtHE as % of THE NPISH as %THE External Sources as % of THE 40% Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.8: Elements of total health expenditure, Palestinian territories, 2000-2012 PHE as % of THE OOP as % of THE PvtHE as % of THE NPIS as % of THE 120 External Sources as % of THE 100 80 percentage 60 40 20 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 157 Figure AII.9: Total health expenditures by type of provider, Palestinian territories, 2012 Hospitals 11% Providers of Ambulatory Health care 15% 35% Retail Sales and Other Providers of Medical Goods General Health Administration and Insurance 15% 24% Other* Source: PCBS (2014), PCBS (2012), and PCBS (2009). Notes: Other* category includes expenditures on (i) nursing and residential care services, (ii) rest of the world, (iii) other industries in the economy, (iv) medical and diagnostics laboratories, (v)other providers of ambulatory health care, and (vi) provision and administration of public health programs. Figure AII.10: Hospital expenditures as share of total health expenditures, Palestinian territories, 2000-2012 General hospitals as % THE Mental health and substance abuse hospitals as % THE Specialty hospitals as % THE Hospital Expenditures as % THE 45 40 35 30 percentage 25 20 15 10 5 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 158 Figure AII.11: Expenditures on providers of ambulatory health care as share of total health expenditures, Palestinian territories, 2000-2012 Offices for physicians as % THE Offices for dentists as % THE Offices for other health practitioners Out-patient care centers Providers of Ambulatory Health Care as % of THE 35 30 25 20 15 10 5 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.12: Expenditures on outpatient clinics as share of total health expenditures, 2000-2012 Public Outpatient Clinics as % THE UNRWA Outpatient Clinics as % THE NPISH Outpatient Clinics as % THE Private Outpatient Clinics as % THE Out-patient care centers as % of THE 30 25 20 percentage 15 10 5 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 159 Figure AII.13: Total health expenditures by function of care, Palestinian territories, 2000-2012 120 Health Administration and Health Insurance 100 Prevention and Public Health Services 80 percentage Medical goods dispensed to out- 60 patient Ancillary services to health care 40 In-Patient long-term nursing care 20 Rehabilitative Care 0 Curative Care Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.14: Composition of expenditures on curative care as share of total health expenditures, 2000-2012 In-patient Curative Care Outpatient Curative Care Non-Classified Services of curative care Curative Care % of THE 80 70 60 50 40 30 20 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 160 Figure AII.15: Trends in hospital expenditures, 2000-2012 500.0 450.0 400.0 in US dollars (millions) 350.0 300.0 250.0 200.0 150.0 100.0 50.0 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.16: General health administration and insurance expenditures, 2000-2012 250.0 200.0 in US dollars (millions) 150.0 100.0 50.0 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 161 Figure AII.17: Expenditures on retail sales and other providers of medical goods, 2000-2012 250.0 200.0 in US dollars (millions) 150.0 100.0 50.0 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.18: Distribution of public expenditures on health, 2000-2012 600,000 Health Administration and Health Insurance Prevention and Public Health Services Medical Goods Dispensed to Out-Patient Ancillary Services to Health Care Inpatient Long-Term Nursing Care Rehabilitative Care 500,000 Non-Class Curative Care Outpatient Curative Care Inpatient Curative Care in US dollars (thousands) Total 400,000 300,000 200,000 100,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 162 Figure AII.19: Public expenditures on prevention and public health services, 2000-2012 80,000 70,000 in US dollars (thousands) 60,000 50,000 40,000 30,000 20,000 10,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.20: Public expenditures on outpatient curative care, 2000-2012 160,000 140,000 in US dollars (thousands) 120,000 100,000 80,000 60,000 40,000 20,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 163 Figure AII.21: Public expenditures on inpatient curative care, 2000-2012 300,000 250,000 in US dollars (thousands) 200,000 150,000 100,000 50,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (20140, PCBS (2012), and PCBS (2009). Figure AII.22: Public expenditure on general hospitals, 2000-2012 250,000 200,000 in US dollars (thousands) 150,000 100,000 50,000 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 164 Figure AII.23: Public expenditures on mental health and substance abuse hospitals, Palestinian territories, 2000-2012 18,000 16,000 in US dollars (thousands) 14,000 12,000 10,000 8,000 6,000 4,000 2,000 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.24: Public spending on MoH outpatient clinics, Palestinian territories, 2000-2012 100,000 90,000 80,000 in US dolars (thousands) 70,000 60,000 50,000 40,000 30,000 20,000 10,000 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 165 Figure AII.25: Out of pocket expenditures, 2000-2012 600,000 500,000 in US dollars (thousands) 400,000 300,000 200,000 100,000 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.26: Composition of out of pocket expenditures, 2000-2012 Health Administration and Health Insurance Prevention and Public Health Services Medical Goods Dispensed to Out-Patient Ancillary Services to Health Care Inpatient Long-Term Nursing Care Non-Class Curative Care Outpatient Curative Care Inpatient Curative Care Total 600,000 in US dollars (thousands) 500,000 400,000 300,000 200,000 100,000 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 166 Figure AII.27: Composition of out of pocket expenditures, 2012 1% 14% 22% Health Administration and Health Insurance 1% Prevention and Public Health Services Medical Goods Dispensed to Out-Patient Ancillary Services to Health Care Inpatient Long-Term Nursing Care 16% Non-Class Curative Care 39% Outpatient Curative Care 0% 7% Inpatient Curative Care Source: PCBS (2014), PCBS (2012), and PCBS (2009). Figure AII.28: Out of pocket expenditures on general hospitals, 2000-2012 120,000 100,000 in US dollars (thousands) 80,000 60,000 40,000 20,000 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: PCBS (2014), PCBS (2012), and PCBS (2009). 167 Annex III: Workforce and Infrastructure Figure AIII.1: Distribution of health workforce by employer, Palestinian territories, 2009-2012 Source: MoH (2010), (2011), (2012), and (2013). 168 Figure AIII.2: Breakdown of type of physicians (general practitioner versus specialists), Palestinian territories, 2007-2012 3500 Specialist Physicians 3000 General Practitioners 2500 2000 1500 1000 500 0 2007 2008 2009 2010 2011 2012 Source: MoH (2008), (2009), (2010), (2011), (2012), and (2013). 169 Annex IV: Basic Concepts of Pension Systems Among different classification criteria of pension systems worldwide, the World Bank system proposed in 2005 differentiates pension system components into five pillars: 1) a non-contributory “zero pillar”; 2) a mandatory earnings-based “first pillar”; 3) a mandatory saving-based individual accounts “second pillar”; 4) a complementary voluntary “third pillar”; and 5) a non-financial “fourth pillar” which includes access to informal support, other formal social programs (such as health and housing), as well as other individual financial and non-financial assets (such as home ownership and reverse mortgages where available). Broad architecture of mandatory pension systems An extensive variety of mandatory pension systems exists in the world. The systems can be classified by different criteria (how benefits are calculated, how benefits are financed, or who manages the system) or countries can be classified by pension modalities (by how many pension pillars they have or whether they have integrated or fragmented pension systems). More than 70 percent of all mandatory national pension systems worldwide are DB systems (defined- benefit, which means that the benefit/pension is based on a prescribed formula), mostly still financed on a PAYG basis (pay-as-you-go is a method of financing whereby current pension expenditures are paid out of current revenues from contributions). More than 70 percent of all mandatory national pension systems worldwide are also publicly managed. Thirty-two countries worldwide have implemented second pillar schemes, which are mandatory fully-funded DC (defined-contribution, a pension plan in which the periodic contribution is prescribed and the benefit, or pension, depends on the contribution plus the investment returns), privately managed, individual accounts. The prevalence of this type of scheme has been increasing worldwide. Another 30 percent of schemes are partially-funded (have accumulated reserves) and more than 30 percent are unfunded (with no accumulated reserves). Most countries in the high-income OECD have some type of pillar zero (social pensions), and their number is also growing in all regions. The pension system in the West Bank and Gaza is DB-PAYG with a DC component (not in practice although), partially- funded, and publicly mandated. Countries worldwide have been moving towards multi-pillar pension systems. During the 1980s and 1990s, the number of countries with mandatory privately managed DC schemes (second pillar) increased from one to more than 30 (although Argentina and Bolivia closed their second pillar in 2008 and 2010, respectively, as did Hungary). A few publicly managed DC (also called PF, provident funds) systems have been reforming towards PAYG-DB systems. During the 2000s, the greatest focus on new developments was shifting towards “zero pillars” (social pensions) and “third pillars” (DC-voluntary pensions). The pension system in the West Bank and Gaza has only one pillar, the “first pillar” (a “second” pillar was passed by law but never implemented in practice), and practically no “zero,” “third,” and-or “fourth” pillars (some unregulated private pension funds probably operate but the number of funds, people covered, size of assets, etc., are not known). There is currently no regulatory framework for private pensions in the West Bank and Gaza. Separate pension schemes for civil servants (and other special groups) operate in about half of the world’s countries. The increasing tendency is towards integrating pension systems (special schemes with national schemes). Currently more than 20 countries worldwide have partially integrated systems (in the process of full integration). Basically, many countries are increasingly integrating their fragmented pension systems. Strong arguments can be made for integration and covering all employees under the same scheme. The long-term goal for most countries seems to be a single national scheme for reasons of equity, 170 administrative efficiency, and labor market flexibility. Until recently, in the West Bank and Gaza only civil and security service personnel were covered by a mandatory pension system, which was fragmented but in the process of integration. The pension system institution intends to expand coverage to private sector workers and ultimately have an integrated pension system for all employees in the West Bank and Gaza. Expansion of coverage, however, is not recommended until the current system is reformed and financially sustainable. Pension scheme design principles While there is no single best way of structuring a public pension system, the choice is by no means arbitrary, and any proposed reforms can – and should – be: (i) considered with respect for a country’s enabling conditions and capacities; and (ii) evaluated using generally accepted principles of pension reform developed from international best practices. These principles include sustainability, affordability, adequacy, fairness, predictability, economic and administrative efficiency, and accessibility (i.e., coverage) and are discussed below. These principles are intended to help policymakers rule out bad policy choices, thereby freeing them to design pension systems that are consistent with international best practices while still having considerable latitude to craft solutions that are appropriate for their country’s social preferences and country-specific conditions:  Sustainability: All too often, policymakers mistakenly conclude that a pension system is financially healthy simply because it is generating cash surpluses (i.e., contributions to the scheme exceed benefits). This typically happens when populations are young and the number of persons contributing to the scheme is far larger than the number of its beneficiaries. As populations inevitably age, however, this favorable situation will slowly but inexorably reverse. Public pension schemes are terribly vulnerable to very gradual changes in the age structure of a country’s population, to almost imperceptible shifts in the relative sizes of the formal and informal economies, and to the incremental evolution of macroeconomic indicators. Because of this vulnerability, pension systems must be designed to be sustainable. A pension system is sustainable only when it has the capacity to pay current – and future – benefits over a long horizon under reasonable assumptions without shifting substantial burdens to future generations and without having to cut benefits, increase contributions, or change qualifying conditions. This can only be accomplished if pension systems are designed and periodically tweaked using careful actuarial modeling.  Affordability: A pension system is affordable only if it can be financed by contributions that do not compromise other social or economic objectives. Affordability refers to the burden placed by pension provisions on labor market efficiency, enterprise competitiveness (as a result of wage taxes and other social insurances levies), and the government (or other sources of financing) to ensure that pension promises are met. The assessment of affordability should consider: (i) current and projected contribution rates to maintain benefit levels; (ii) total pension costs; and (iii) the fiscal costs of any subsidies required to meet benefit obligations. Contributory pension schemes with high wage taxes create a strong incentive for evasion. Moreover, pension systems that consume a substantial proportion of GDP can impact a country’s competitiveness and growth.  Adequacy: Benefits are adequate when they: (i) provide sufficient income to protect participants from falling into poverty in the event that they become disabled or after they have retired; and (ii) provide a reliable mechanism for smoothing consumption between a participant’s working years and his or her old age. Moreover, benefits must be adjusted over time so that their real value is not eroded by inflation. Two criteria against which the adequacy of benefit levels can be judged are: some stated minimum income level (for example assuring an absolute or relative level of income either for the lifetime poor or for those who fall into poverty in old age); or some minimum level of 171 income replacement (i.e., achieving a particular level of income replacement to sustain a target level of consumption in one’s old age relative to that enjoyed during one’s working years). International experience varies considerably.  Fairness: In general, the benefits provided by systems of social insurance should be related to an individual’s contributions (and those made on his or her behalf by employers). This is only fair, and fairness is a core tenant of a market economy. Policymakers must bear in mind, however, that systems of social insurance (including pensions) are not savings schemes. By design, systems of social insurance are intended to protect participants against specific forms of risk, including the risk of becoming disabled and the risk of outliving one’s savings. Thus, systems of social insurance must necessarily effect wealth transfers from some groups to others within the covered population. Persons who die soon after reaching retirement, for example, typically subsidize benefits for those who live much longer. All participants who reach retirement subsidize those who become disabled during their working years. In addition to these sorts of insurance-based transfers, social insurance schemes are also designed to effect transfers from the comparatively well-off to the less well-off (a design feature known as progressivity). What is important is that these transfers be only as large as is required to attain their underlying social policy objectives and that transfers be transparent and subject to scrutiny to be sure that their underlying objectives are met as efficiently as possible.  Predictability: A predictable pension system provides benefits that are: (i) specified by law and not subject to the discretion of policymakers or administrators; (ii) adjusted over time (i.e., indexed) to protect beneficiaries from an erosion in their real value as a result of inflation; and (iii) as much as possible insulated from longevity risk (i.e., the risk that someone will outlive his or her resources in retirement).  Efficiency: An efficient pension system: (i) is designed to achieve social protection objectives without creating adverse behavioral incentives for participants or employers; and (ii) meets reasonable standards for service delivery and enforcement at reasonable administrative costs.  Accessibility: An accessible pension system is designed to gradually target wider and wider coverage of a country’s labor force through better enforcement of contribution compliance, improved behavioral incentives, and other mechanisms. It must be noted that the existence of the above principles does not imply that they are always in harmony with one another. To the contrary, these principles often exist in tension: steps to raise levels of income replacement in pursuit of benefit adequacy, for example, may reduce affordability and undermine a scheme’s sustainability over the long term. The resolution of these sorts of trade-offs is the central – and inherently difficult – challenge facing policymakers in the West Bank and Gaza. 172 Annex V: Notional Defined Contribution Plans as a Pension Reform Strategy Notional accounts are designed to mimic a defined contribution plan, where the pension depends on contributions and returns. Pension contributions are tracked in accounts that earn a rate of return. However, in notional accounts, the return that contributions earn is a notional one, set by the government, not the product of investment returns in the markets. Like traditional social insurance schemes, they are publicly provided. However, the pension formula differs somewhat from the ‘traditional’ earnings-related model, with the benefit based on the accumulation in one’s account at the time of retirement. Pension accounts in this system are called ‘notional’ because there is no pot of pension fund money per se, just a series of individual claims on the future public budget. They are pay-as-you-go-financed—current contributions pay for current benefits—just like most defined-benefit public schemes. When the individual reaches pension age, accumulated contributions and notional returns— notional capital—are converted to an annuity. Notional defined contribution (NDC) plans are sometimes an attractive option because they preserve the connection between contributions and pensions (and, therefore, encourage retirement savings and income reporting), and in addition they are more easily managed than FDCs (Funded defined contribution) plans, and the administrative costs are much lower. Further, they are not exposed to financial market fluctuations and do not impose the transition costs associated with the shift to a funded scheme. On the negative side, however, NDCs provide a weaker stimulus for saving and the long-term returns are lower than is the case with funded schemes. 173 Annex VI: Financial Sustainability of a Pension System Financial sustainability is one of the most important principles to be considered when designing or reforming a public pension system. While there is no single best way of structuring a pension system, the choice is by no means arbitrary, and any proposed design or reforms can – and should – be: (i) considered with respect to a country’s enabling conditions and capacities; and (ii) evaluated using generally accepted principles of pension design developed from international best practices, financial sustainability being one of the main principles, together with adequacy, affordability, fairness, predictability, robustness, economic and administrative efficiency, and of course accessibility (i.e., coverage). Pension systems involve long-term financial commitments. The promise to pay a benefit during retirement to today’s workers covers a period that can span many decades. The capacity to meet these promises is one of the most important issues in the design of retirement-income systems. All too often, policymakers mistakenly conclude that a pension system is financially healthy simply because it is generating cash surpluses (i.e., contributions to the scheme exceed pension payments). This typically happens when populations are young and the number of persons contributing to the scheme is far larger than the number of its beneficiaries (such is the case of Scheme II in the West Bank and Gaza). As a system’s demographics inevitably age, however, this favorable situation will slowly but inexorably reverse. Public pension schemes are highly vulnerable to very gradual changes in the age structure of the system’s demographics. Pension schemes are also vulnerable to almost imperceptible shifts in the relative sizes of the formal and informal economies, and to the incremental evolution of macroeconomic indicators. Because of these vulnerabilities, pension systems must be designed to be sustainable. Issues of financial sustainability are mainly relevant to earnings-related (ER) schemes, such as the case of the pension system in the West Bank and Gaza. The majority of mandatory public pension systems in the world still involve an ER scheme with “pay-as-you-go” (PAYG) financing mechanism, where current pensions are financed essentially out of current contributions. Many of the systems can have reserves (as in Scheme II of the West Bank and Gaza), but these act rather as a “buffer stock” to smooth adjustments to changes in benefits or contribution rates that become necessary as a result of unexpected macroeconomic and/or demographic shocks or because of the gradual maturation of the system. Surprisingly, in most cases, ER systems do not meet basic principles in terms of design to ensure financial sustainability, minimize distortions in labor supply and savings decisions, and avoid arbitrary -- often regressive -- redistributions of income (such as the case of the system in the West Bank and Gaza). Basically, a pension system is sustainable only when it has the capacity to pay current—and future— promised benefits over a long-term horizon under reasonable assumptions without shifting substantial burdens to future generations and without having to cut benefits, increase contributions, or change qualifying conditions. This can only be accomplished if pension systems are designed and periodically tweaked using careful actuarial modeling. Projections of pension-system finances need an analytical or modeling tool. The sample results presented in Chapter 4 are derived from projections using the World Bank’s PROST model (Pension Reform Options Simulation Toolkit). It is important to emphasize that spotting problems of financial sustainability early is vital to enable corrective actions that avoid more painful adjustments later on. Carrying out long-term financial projections for pension systems is not an easy task. It requires a wide range of administrative data from national authorities. Detailed current data on expenditures provide a baseline for the projections. Information on coverage patterns, the income level of members, and a variety of other characteristics are also required. Looking forward, the calculations rely on assumptions about the future path of the population, economy, and individual behavior. These include changes in the numbers of people of working age and pension age, 174 coverage of the pension system, retirement decisions, etc. Despite these challenging assumptions and the uncertainties involved, it is strongly recommended that projections are made for the next 50-75 years, since policy decisions today affect future liabilities over this long horizon. There are often long periods before an imbalance in the design of a pension system will show up, particularly when the system is still immature (such as in Scheme II). A pension system is only mature after the moment when every single retiree has contributed to the scheme from the beginning of his/her career. 175 Annex VII: Survivorship Pensions Survivors’ benefit programs are designed differently in many countries and regions throughout the world. These benefits are usually a supplement to old-age social security programs, so the structure of the latter inevitably affects the way the former is provided. Many countries have reformed their pension programs to make them fiscally sustainable and to eliminate their negative economic impact. At the same time, many of these countries have reexamined their survivors’ benefit programs, with the same goals. Countries have begun to ask: Is this program a good use of funds? Do net benefits go to the right people? Are the right behaviors encouraged? Although survivor benefit programs vary by country, broad regional patterns are observed. This may be due to common demographic and labor market forces or it may reflect the likely transmission of policy ideas within a given region that has a common language and culture. MENA countries exemplify the most traditional family structures and survivors’ benefit programs. Benefit rates are provide d solely through Pillar 1, publicly managed plans, to contributors and their survivors. As in the West Bank and Gaza, in all MENA countries, a broad extended family group is eligible, including widows at any age, children, parents, etc. However, survivors of non-contributors (which are the vast majority of the population) receive no protection. While the definition of the covered group in these programs reflects cultural values and family linkages, a generous list of beneficiaries will necessarily translate into high costs for the system. If the sustainability of the pension system is at risk, the definition of the insured family group should be considered by reform efforts, particularly if the potential cost savings are assessed as significant. A cost assessment of survivorship pensions under different alternative scenarios (definitions of family members covered) is recommended to understand the extent to which this component alone affects the financial situation of the current system, and to identify possible actions to reduce costs without affecting the basic coverage provided by pension schemes. Survivorship pensions also create considerable incentives that affect behavior, often in a way that is suboptimal for the economy as a whole. For example, if widows, unmarried, widowed, or divorced daughters, and sisters expect survivors’ benefits, this becomes a part of their implicit pension wealth when young. This enables them to work less if so desired. The system may also inefficiently influence marital decisions. It has been observed in many countries that pension system rules often discourage widows from remarrying, or sisters or daughters from marrying and/or remarrying. Parents are generally expected to fend for themselves in Europe but are eligible for survivor benefits in MENA and Latin America regions, where people usually have strong extended families. Survivors of non-contributors (the vast majority of the population) receive no protection. None of these systems are actuarially fair, so cross-subsidies are inevitably involved, especially when many different classes of survivors are covered. 176 Annex VIII: Key Features of a PIM System – Existing Situation and Recommendations (Importance: C = Critical; H = High, M = Medium. Timing: ST = Short-Term (next 1-2 year); MT = Medium-Term (2-4 years); LT = Long-Term (4-6 years)  development/introduction  ongoing application) Existing Situation in the Priority Timing Feature/Key Element Recommendations Responsibility West Bank and Gaza ST MT LT Feature 1: Strategic Prioritization, Project Development and Initial Screening MoPAD prepares 3 year National Development Use the PNDP process 1.1. Credible strategic plans and PBB initiative to Plan (PNDP) setting out at national and sectoral level strengthen strategic strategic policies and guide public investment planning and priorities that are used to MoPAD strategy and prioritization guide project selection. determine public H + MoF   between and within major spending and capital infrastructure and service Little capability for strategic investment priorities at delivery sectors. planning and sector and MDA prioritization in MDAs to levels. guide investment choices. 1.2. Formal process, specified by the MoF, through which MDAs identify and scope proposed Establish procedures and projects. Typically this develop a format and involves the development of guidelines for the No procedures defined or preparation of PCPs. a PCP setting out – (i) the specified requirements in Require that PCPs are strategic case for the project; place. submitted for all (ii) a brief description public investment MoF covering the main Different process and projects whether C (BD/PIMU)   components of the project, oversight responsibility and related activities and for identification of IDP financed by IDPs or outputs; (iii) a preliminary and PA-financed capital the PA. Roll out new financial analysis; (iv) a investment projects. procedures to MDAs preliminary economic and and provide training social analysis; and (v) the and backstopping key issues to be further advice. investigated during preparation and appraisal. 177 Existing Situation in the Priority Timing Feature/Key Element Recommendations Responsibility West Bank and Gaza ST MT LT 1.3. Initial screening to evaluate if the project concept is: (i) compatible No formal procedures in Prepare terms of with strategic goals of place: reference for and government; (ii) consistent MoPAD undertakes some establish the Capital with MDA priorities and screening of proposed Investment Project mandates; (iii) represents areas for IDP funding Screening and most appropriate of but this tends to be Selection Committee MoF available options; reactive to proposals to review PCPs C (BD/PIMU)   (iv) appropriate in scope and coming from the IDP. submitted by sector scale; (v) stands a realistic PA-funded projects – MDAs and agree prospect of being funded. either identified at which projects should Screening out inappropriate political level or proceed to detailed projects at this stage through annual budget preparation and prevents time and resources process. appraisal. being wasted on subsequent preparation and appraisal. Feature 2: Project Preparation and Appraisal 2.1. Procedures in place for feasibility analysis to determine technical, financial and economic No specified PA-wide viability of projects: procedures issued for prefeasibility studies to project preparation and identify and evaluate appraisal. Except for relevant alternatives and MoPWH limited skills Specify project select the most and capacities for project preparation and promising; preparation and appraisal appraisal procedures. in sector MDAs. Develop guidelines feasibility studies to covering: (i) key provide a detailed For IDP-funded projects appraisal concepts; MoF (PIMU) assessment of the prefeasibility and (ii) stages in the preferred alternatives feasibility studies appraisal process; H + MoPWH   including background typically contracted by + MoPAD (iii) role of MDAs in analysis such as sponsoring IDP. managing the appraisal environment and social Capability and experience process; and (iv) use impact assessments and exists in MoPWH in: of IDP assistance to rigorous financial and (i) managing technical support the appraisal economic (cost-benefit) feasibility studies carried process. analysis; out by consultants; and for smaller and relatively (ii) undertaking simple projects it may preparation of be possible to combine construction projects. prefeasibility and feasibility analysis within a single study. 178 Existing Situation in the Priority Timing Feature/Key Element Recommendations Responsibility West Bank and Gaza ST MT LT 2.2. Provision for training staff in project appraisal skills and procedures for ensuring that appraisal capacities are effectively Use IDP support in deployed. Typically involves carrying out feasibility MoF MoF in process of studies etc. to help having a central PIMU that establishing PIMU build MDA capacities H (BD/PIMU)   sets standards, develops + MoPWH guidelines and provides for managing project oversight with dedicated appraisals. sector units in major MDAs responsible for carrying out appraisals. 2.3. Existence of a portfolio Does not exist at present, of appraised projects can be but should occur as new helpful in providing a PIM procedures are competition at subsequent introduced and project stage of selection for appraisal capacities built funding from the budget. up. Feature 3: Independent Review of Appraisal Independent review 3.1. Provision for requirements not independent review of elaborated and developing project appraisals. For small capacities in the MoF to projects submitted by a carry out appraisal should Seek IDP funding to sector MDA this may be be regarded as a long- carry out appraisal reviews of the largest MoF done by staff in the central term objective. and most complex H (BD/PIMU)    PIMU. For more complex Nevertheless the MoF need projects this task may by projects that carry to be ensure that large significant risks. contracted to a competent projects that carry independent organization or significant risks have consultancy. been appraised to an adequate standard. Feature 4: Project Selection and Inclusion in the budget Develop and introduce 4.1. Procedures for selecting procedures for project projects for funding within selection and approval. the overall budgetary Make the proposed MoF No formal procedures in C  resource envelop, securing Capital Investment (BD/PIMU). place financing from the selected Project Screening and source (IDP or PA). Selection Committee responsible for project approval. 179 Existing Situation in the Priority Timing Feature/Key Element Recommendations Responsibility West Bank and Gaza ST MT LT IDP financing earmarked at 4.2. The financing and project concept stage for budgeting process should multi-year Show in the detailed recognize that implementation. budget presentation implementation of capital allocations for each PA financing approved for a investment projects typically capital investment single year rather than for takes longer than 12 months projects that is full project implementat- and therefore, involve a included in the PA ion period. No systems in multi-year funding budget. This should place for managing and MoF commitment and that include the total cost subsequent financial monitoring of the project, C (BD/PIMU)   financing/expenditure provision will be required to expenditure in over the full project meet recurrent/operational previous years, budget implementation period. costs following project expenditure for the completion. No procedures in place for coming year, spending identifying downstream in the ffollowing years operational costs and and subsequent annual incorporating these into operational costs. the budget planning process. Feature 5: Project Implementation 5.1. Effective arrangements for project management must be in place to ensure that implementation No general guidelines have proceeds without delay. been issued to MDAs, but these should cover: (i) roles established n MoPWH for of project managers/PIUs; managing and monitoring Develop general (ii) delegations of authority; implementation of guidelines for project MoF (iii)procurement policies and construction projects. managers/PIUs that H  + MoPWH procedures; (iv) financial In other MDAs apply both to IDP and management and reporting; implementation managed PA-funded projects. (v) monitoring of through PIUs, often implementation progress; established with IDP (vi) revisions to project support. schedules and design; and (vii) project completion reporting and review. Uncertainty over funding allocations and cash releases affects 5.2. Operation of public implementation of PA- financial management Monitor project financed projects and systems facilitates timely implementation across gives rise to expenditure implementation of capital the capital investment arrears. MoF investment projects. Key program to identify H (BD/PIMU)   requirements include: Regulatory and institutional and address systemic (i) timely funding releases; framework for public issues and causes of and (ii) efficiently managed procurement being put in delay. procurement procedures. place. Significant procurement delays can affect project implementation. 180 Existing Situation in the Priority Timing Feature/Key Element Recommendations Responsibility West Bank and Gaza ST MT LT 5.3. Budgeting and Projects budgeted and accounting systems able to Introduce a project accounted for annually. capture and report costs over management module PA systems does not MoF the life of a project against into IFMIS to track H  provide for managing (BD/PIMU) its total estimated/approved expenditures over life project against total cost (and not just against an of project against total approved project cost annual appropriation). approved project cost. (IDP/PIU systems do). Feature 6: Project Monitoring and Adjustment 6.1. Procedures for capital investment portfolio performance reviews to identify requirements for: No PA procedures for Once a comprehensive (i) adjustments to project carrying out capital register of public design and implementation investment project investment projects schedules; (ii) changes to portfolio reviews. has been established MoF total approved project cost the MoF should H (BD/PIMU)   Performance reviews and and planned phasing of adjustments for IDP- introduce sector level expenditures; and (iii) where funded project tend to investment program changes in circumstance or follow IDP procedures. portfolios reviews. poor performance justify curtailment or closure of a project. 6.2. Procedures for making Rules for approving adjustments to capital changes to the total projectspecifications, For PA-funded projects estimated cost of implementation schedules financing adjustments capital investment MoF and total estimated cost. made through annual H  projects should be (BD/PIMU) Procedures should identify budget process and not introduced, identifying the limits above which based on wider limits above which approval of the Legislature performance review. approval of the is required. Legislature is required Feature 7: Facility Handover and Operation 7.1. Procedures for: Develop general MoF (i) handover of the project Procedures not in place guidelines for MDAs ( BD/PIMU) on completion and ensuring leading to inconsistencies on preparing for M + MDAs   that the necessary capacities in handover process and facility handover and + MoPWH. and resources for facility in service monitoring. operation. operation are in place; Failure to consider (ii) addressing faults during Monitor preparations for downstream recurrent facility handover and the initial operation period; costs in design in some (iii) updating asset registers; operation, including IDP-funded projects has elaboration of required MoF and (iv) active monitoring of resulted in PA being current financing, M (BD/PIMU)   service delivery over time unable to fund operating and the benefits deriving through the capital costs. investment portfolio from the investment. review process. 181 Existing Situation in the Priority Timing Feature/Key Element Recommendations Responsibility West Bank and Gaza ST MT LT Feature 8 Basic Completion Review and Evaluation Require project 8.1. Basic completion completion reports for reviews required for all all capital investment projects so that lessons projects included in No requirements for learned during the PA budget. MoF completion reviews, implementation can be fed except where specified in Develop a standard M (BD/PIMU).   back into and format and guidelines IDP procedures. implementation of future for completion reports. projects. Provide training to MDAs in their application. Continue to make use of IDP resources for ex- 8.2. For selected projects ex- post evaluations of post evaluations to be major capital projects carried at least 2-3 years and ensure that the following completion and No requirements for results are presented to completion reviews, key stakeholders. MoF handover to determine the except where specified in Bring IDP support M + MoPAD  extent to which planned project outputs and IDP procedures. within an annual outcomes are being program of capital achieved. project evaluations agreed with and overseen by the MoF and MoPAD. 182