Report No. 79865 - MW Malawi Public Expenditure Review November 2013 Poverty Reduction and Economic Management 4 Africa Region Document of the World Bank MALAWI – GOVERNMENT FISCAL YEAR JULY 1 – JUNE 30 CURRENCY EQUIVALENTS (EXCHANGE RATE EFFECTIVE AS OF OCTOBER 22, 2013) CURRENCY UNIT = MALAWI KWACHA (MWK) MWK 366 = US $1 US$1.0 = SDR 0.65 WEIGHTS AND MEASURES METRIC SYSTEM ABREVIATIONS AND ACRONYMS AICD Africa Infrastructure Country Diagnostic ADT Average Daily Traffic BPFP Budget Policy Framework Paper CABS Common Approach to Budget Support CAADP Comprehensive Africa Agriculture Development Program CAS Country Assistance Strategy CBE Continued Basic Education CBET Competency-Based Testing CBO Community-Based Organization CDSS Community Day Secondary School CEAR Central and East Africa Railways CEM Country Economic Memorandum CGIAR Consultative Group on International Agriculture Research CHAI Clinton Health Access Initative CHAM Christian Health Association of Malawi CIAT International Center for Tropical Agriculture CML Central Materials Laboratory COMSIP Community Savings and Investment Promotion CSR Country Status Report DAD Debt and Aid Division DCA Department of Civil Aviation DEM District Education Manager DFID Department for International Development DP Development Partner DSB Department of Statutory Bodies DSS Direct Support to Schools ECD Early Childhood Development ECF Extended Credit Facility EDM Education Division Management EFA Education for All EHP Essential Health Package EIMU Education Infrastructure Management Unit EMIS Education Management Information System ERP Economic Recovery Plan ESIP Education Sector Implementation Plan EU European Union FISP Farm Input Subsidy Program FROIP Financial Reporting and Oversight Project FY Fiscal Year GDP Gross Domestic Product GOM Government of Malawi GPE Global Partnership for Education HIPC Highly Indebted Poor Countries HR Human Resources HRH Human Resources for Health HRMIS Human Resource Management Information System HSSP Health Sector Strategic Plan ICAO International Civil Aviation Organization ICRISAT International Crops Research Institute for the Semi-Arid Tropics IEC Internal Efficiency Coefficient IFMIS Integrated Financial Management Information System IITA International Institute of Tropical Agriculture IHS Integrated Household Survey IHS3 Third Integrated Household Survey IPTE Initial Primary Teacher Education IT Information Technology IWT Inland Water Transport JFA Joint Financing Arrangement JICA Japan International Cooperation Agency KIA Kamuzu International Airport LDF Local Development Fund LGC Local Government and Councils LVSR Low Volume Sealed Road M&E Monitoring and Evaluation MASAF Malawi Social Action Fund MDAs Ministries, Departments, and Agencies MDG Millennium Development Goal MDRI Multilateral Debt Relief Initiative MEPD Ministry of Economic Planning and Development MGDS Malawi Growth and Development Strategy MGWCD Ministry of Gender, Women and Child Development MK Malawi Kwacha MLGRD Ministry of Local Government and Rural Development MoAFS Ministry of Agriculture and Food Security MoEST Ministry of Education, Science and Technology MoE Ministry of Disabilities and Elderly MoF Ministry of Finance MoGCSW Ministry of Gender, Children, and Social Welfare MoH Ministry of Health MSCE Malawi School Certificate of Education MTEF Medium Term Expenditure Framework MTFF Medium Term Fiscal Framework MTPW Ministry of Transport and Public Works MUST Malawi University of Science and Technology MZUNI Mzuzu University NHA Malawi National Health Accounts NAO National Audit Office NESP National Education Sector Plan NGO Non-governmental Organization NLGFC National Local Government Finance Committee NOCMA National Oil Company of Malawi NSSP National Social Support Program OBS Open Budget Survey ODL Open and Distance Learning OPC Office of the President and Cabinet ORT Other Recurrent Transactions PCR Pupil-Classroom Ratio PE Personal Emoluments PEA Primary Education Advisor PEFA Public Expenditure and Financial Accountability Framework Assessement PER Public Expenditures Review PFEM Public Finance and Economic Management PFEMRP Public Financial and Economic Management Reform Program PFM Public Financial Management PHC Primary Health Care POW Program of Works PPP Public-Private Partnership PQTR Pupil-Qualified Teacher Ratio PRC People's Republic of China PSIP Primary School Improvement Program PTR Pupil-Teacher Ratio PVHO Plant and Vehicle Hire Organisation R2 Coefficient of Determination RA Roads Authority RBM Reserve Bank of Malawi RFA Road Fund Administration RSP Road Sector Programme SACMEQ Southern African Consortium of Monitoring Education Quality SADC Southern African Development Community SCTP Malawi Social Cash Transfer Program SDI Malawi Staff Development Institute SMEs Small and Medium Size Enterprises SOEs State-Owned Enterprises SSA Sub-Saharan Africa SP Social Protection SWAp Sector-Wide Approach TBs Treasury Bills TEVET Technical, Entrepreneurial and Vocational Training TEVETA Technical, Entrepreneurial and Vocational Training Authority THE Total Health Expenditures TLM Teaching and Learning Materials TSIP Transport Sector Investment Programme TRANSRA Transport Regulatory Authority TTC Teacher Training College TWG Technical Working Group UNESCO United Nations Educational, Scientific and Cultural Organisation UNIMA University of Malawi WDI World Development Indicators Vice President : Makhtar Diop Country Director : Kundhavi Kadiresan Sector Director : Marcelo Giugale Sector Manager : John Panzer Task Team Leader : Tuan Minh Le TABLE OF CONTENTS EXECUTIVE SUMMARY ......................................................................................................................... i Chapter 1. Macroeconomic Developments and Public Expenditure Framework.......................... 1 1.1 Introduction ................................................................................................................................... 1 1.2 Macroeconomic Trends................................................................................................................. 1 1.3 Structure of the Budget and Fiscal Trends .................................................................................... 4 1.4 The Fiscal Framework (2012/13 - 2015/16) ................................................................................. 7 1.5 Restraining Expenditures Through Cross Sectoral Policies........................................................ 10 Chapter 2. Strengthening Planning and Budgeting Processes ....................................................... 17 2.1 Introduction ................................................................................................................................. 17 2.2 Progress with Planning and Budgeting Reforms ........................................................................ 17 2.3 Advancing the MTEF Reform .................................................................................................... 20 2.4 Strengthening Management of the planning and Budgeting Process .......................................... 29 Chapter 3. Agriculture ...................................................................................................................... 35 3.1 Introduction ................................................................................................................................. 35 3.2 Assessment of Agriculture Expenditures .................................................................................... 35 3.3. Agriculture Sector Performance and Key Issues ........................................................................ 47 3.4 Policy Priorities or Suggestions for Reform ............................................................................... 52 Chapter 4. Transport ......................................................................................................................... 55 4.1 Introduction ................................................................................................................................. 55 4.2 Assessment of Transport Expenditures ....................................................................................... 55 Chapter 5. Education ......................................................................................................................... 74 5.1 Introduction ................................................................................................................................. 74 5.2 Assessment of Education Expenditures ...................................................................................... 74 5.3 Education Sector Performance and Key Issues ........................................................................... 83 Chapter 6. Health ............................................................................................................................... 90 6.1 Introduction ................................................................................................................................. 90 6.2 Assessment of Health Expenditures ............................................................................................ 90 6.3 Health Sector Performance and Key Issues .............................................................................. 104 Chapter 7. Social Protection ........................................................................................................... 111 7.1 Introduction ............................................................................................................................... 111 7.2 Assessment of Social Protection Expenditures ......................................................................... 111 7.3 Social Protection Sector Performance and Key Issues ............................................................. 118 7.4 Policy Priorities or Suggestions for Reform ............................................................................. 122 Chapter 8. CONCLUSION ............................................................................................................. 124 ANNEXES ........................................................................................................................................ 126 REFERENCES ........................................................................................................................................ 140 ANNEXES Annex A1: Details of Generic Goods and Services .............................................................................. 126 Annex A2: Structure of Arrears (December 2012) in MK Billion ....................................................... 127 Annex A3: Progress Made in Addressing Key Budget Process Issued Identified in 2001 Public Expenditure Review .............................................................................................................................. 128 Annex A4: Key Stages in Malawi’s Budget Calendar .......................................................................... 129 Annex A5: MoF Treasury Department Organization Chart.................................................................. 130 Annex A6: Breakdown of Recurrent Expenditures (Personnel, Operations & Maintenance of Assets) - MK million: 2007 Base Year ................................................................................................................ 131 Annex A7: Public Road Network Composition .................................................................................... 131 Annex A8: Trends in overall expenditure ............................................................................................. 132 Annex A9: Recurrent Expenditures by Public Institutions and Countries ............................................ 134 Annex A10: Comparison of Tuition Fees among Countries in the SADC Region ............................... 135 Annex A11: Some Literature on the Impact of Double Shifting .......................................................... 136 Annex A12: Malawi Social Protection Programs and Expenditures .................................................... 137 Annex A13: Estimated cash transfer levels to mitigate vulnerability resulting from the Forex crisis .. 139 FIGURES Figure 3.1: Trends in approved and actual expenditures of MoAFS and the Department of Irrigation, 2000/01-2012/13, current USD million ...................................................................................................... 36 Figure 3.2: Share of actual recurrent and development expenditures in MoAFS budget, 2000/01-2011/12, % ................................................................................................................................................................. 39 Figure 3.3: Trends in MoAFS wage bill, 2000/01-2011/12, MK billion .................................................... 40 Figure 3.4: Trends in internal travel expenditures in MoAFS, 2000/01-2011/12, MWK billion ............... 40 Figure 3.5: Respective shares of focus areas in ASWAp budget, 2011/12-2014/15 (in %) ....................... 41 Figure 3.6: Functional classification of MoAFS actual expenditures, 2007/08-2011/12 (in %) ................ 42 Figure 3.7: Functional classification of total actual agricultural expenditures, 2007/08-2011/12 .............. 42 Figure 3.8: Execution rates of MoAFS expenditures by type, 2009/10-2011/12, % .................................. 43 Figure 3.9: FISP recipients by wealth and poverty status, 2011 (in %) ...................................................... 46 Figure 3.10: FISP net subsidy share, rural population, 2011, % ................................................................. 46 Figure 3.11: Agricultural households receiving FISP by total land area decile, 2011, % .......................... 46 Figure 3.12: Poverty and ultra-poverty headcount, IHS2 and IHS3 ........................................................... 47 Figure 3.13: Evolution of maize yield, area harvested and production, 1990-2011 ................................... 48 Figure 3.14: Trends in MoAFS budget allocation to research, 2000/01-2011/12, current USD million .... 49 Figure 3.15: Percentage of research in MoAFS budget, 2000/01-2011/12 (in %)...................................... 49 Figure 3.16: Sources of funding of food crop research, 2007/08-2011/12, current USD million .............. 50 Figure 3.17: Trends in MoAFS budget allocation to extension, 2000/01-2011/12, current USD million .. 51 Figure 3.18: Percentage of extension in MoAFS budget, 2000/01-2011/12 (in %).................................... 51 Figure 3.19: Proportion of agricultural households receiving extension by wealth, IHS2 and IHS3 (in %) .................................................................................................................................................................... 52 Figure 3.20: Proportion of agricultural households receiving extension by region, poverty status and gender, 2011 (in %)..................................................................................................................................... 52 Figure 4.1: Sector Expenditures (actual) for GOM and DPs - MK millions: 2007 Base Year ................... 57 Figure 4.2: Median Distance from Community to Nearest Tar/Asphalt Road (Km) Malawi, Rural and Urban, by Community Wealth .................................................................................................................... 66 Figure 5.1: Share of Education in On-Budget GoM Expenditure ............................................................... 74 Figure 5.2: Education Expenditure as % of GDP in Selected Countries (2011) ......................................... 75 Figure 5.3: Distribution of On-Budget Education Funding by Source ....................................................... 80 Figure 6.1: Trend of health spending as a share of GDP ............................................................................ 92 Figure 6.2: Sources of health financing ...................................................................................................... 92 Figure 6.3: Trend in public health spending, in millions of MKW............................................................. 95 Figure 6.4: Pool Donors’ Pledges vs. Actuals (Without Global Fund) in USD millions, 2008-2012 ........ 95 Figure 6.5: Distribution of spending for PE, ORT, and Development (MOH and LGFC) ......................... 96 Figure 6.6: ORT spending in health, 2006-07 to 2011-12 .......................................................................... 97 Figure 6.7: Distribution of ORT costs only MOH and LGC in percent ..................................................... 97 Figure 6.8: Proportion of commodities funded by Government and Donors in FY 2012-13 ..................... 98 Figure 6.9: MOH Original and Revised Budget for FY 2011-12 ............................................................... 99 Figure 6.10: Trend of expenditures financed by public health sector by sub-programs ............................. 99 Figure 6.11: Regional health expenditure patterns ................................................................................... 100 Figure 6.12: Distance to Health Clinic (w/or without Doctor) (% communities by wealth) .................... 101 Figure 6.13: Health spending as share of total household expenditure..................................................... 103 Figure 6.14: Distribution of Government Subsidy by Wealth, within Rural and Urban areas, 2011 ....... 103 Figure 6.15: Use of health services by income quintile ............................................................................ 104 Figure 6.16: Global comparison of life expectancy relative to income and health spending, 2009 ......... 105 Figure 6.17: Infant and Under-5 Mortality Rate ....................................................................................... 105 Figure 6.18: Maternal Mortality Rate ....................................................................................................... 106 Figure 6.19: Reasons why women are unable to access health care services ........................................... 107 Figure 7.1: Social Protection interventions by poverty and vulnerable categories ................................... 115 Figure 7.2: Ultra-Poverty in TCTP Districts ............................................................................................. 116 Figure 7.3: Trends for Social Protection Allocations................................................................................ 116 Figure 7.4: Farm Input Subsidy Program in Sector Budgets .................................................................... 117 Figure 7.5: Households Accessing School Feeding Programs: Malawi and Urban vs. Rural .................. 118 Figure 7.6: Resource allocation in MASAF III ......................................................................................... 121 TABLES Table 1.1: Key Economic Indicators, 2005/06-2012/13 (Percent of GDP Unless Specified Otherwise) ..... 2 Table 1.2: Changes in Revenues and Expenditures (Percent of GDP) ......................................................... 4 Table 1.3: Discretionary Expenditures as a Share of GDP and Total Expenditures ..................................... 6 Table 1.4: Projected Central Government Operations (Percent of GDP) ..................................................... 9 Table 1.5: Progress of Reforms of Expenditure Control............................................................................. 12 Table 1.6: Growth in Employment and Wages in Civil Service ................................................................. 13 Table 1.7: Average Wage (in‘000 Kwachas) and Employment (%) in Private Company Sector and the Civil Service................................................................................................................................................ 13 Table 1.8: Travel Expenditures as a Share of Total Expenditures .............................................................. 14 Table 1.9: Subsidies and Transfers in the Budget (Percent of GDP) .......................................................... 15 Table 2.1: Changes in PEFA Scores 2006-2008 and 2008-2011 ................................................................ 18 Table 2.2: Open Budget Survey Scorings 2012 .......................................................................................... 18 Table 2.3: Budgeted Development Expenditure 2012/13 and DP Off-Budget Project Financing 2012/13 24 Table 2.4: Share of Development Expenditure on Non-Capital Items (2012/13 Budget – excludes off- budget DP financing) .................................................................................................................................. 25 Table 3.1: Distribution of actual agricultural expenditure by institutional status and share in total national expenditure, 2007/08-2011/12 (USD million, current and %) .................................................................... 38 Table 4.1: GOM Transport Sector (TS) Expenditures - MK million: 2007 Base Year .............................. 56 Table 4.2: Total Sector Expenditure for GOM and DPs (actual) - MK millions: 2007 Base Year ............ 57 Table 4.3: Expenditure on Road Maintenance, Rehabilitation and Upgrading/ Construction as a Proportion of GDP ...................................................................................................................................... 58 Table 4.4: IWT Recurrent and Development Budget Allocations - MK millions: 2007 Base Year........... 59 Table 4.5: Civil Aviation Budget Allocations - MK million: 2007 Base Year (real) ................................. 60 Table 4.6: Wholly Government Funded Road Projects .............................................................................. 63 Table 4.7: Planned and Achieved Physical Road Works Outputs - km ...................................................... 68 Table 4.8: Anticipated Budget and Actual Expenditure – MK million: 2007 Base Year (real) ................. 69 Table 5.1: On-Budget Expenditure Growth Rates ...................................................................................... 75 Table 5.2: Education sector on-budget expenditures .................................................................................. 76 Table 5.3: Composition of Education Sector Recurrent Expenditures ....................................................... 76 Table 5.4: On-Budget Recurrent Expenditure by Sub-Sector: 2008/09 - 2011/12 ..................................... 78 Table 5.5: Distribution of DP Funding by Type (% of total DP funding) .................................................. 80 Table 5.6: Primary and Secondary PCRs; Primary-Secondary Transition Rate: 2007/08 – 2011/12 ......... 84 Table 6.1: Health Expenditures by Southern African Development Community countries, ...................... 91 Table 6.2: Trend analysis of public health spending between 2006-07 and 2011-12 ................................. 93 Table 6.3: Sources of Financing in the public health sector in 2012-13 in MKW billion .......................... 94 Table 6.4: Catastrophic Expenditure Headcounts in Malawi, 2010-11 .................................................... 102 Table 6.5: Policy Recommendations ........................................................................................................ 109 Table 7.1: Social Expenditures by Government of Malawi in 2012/13 (MK billions) ............................. 112 Table 7.2: Comparison of SP Expenditures with FSIP and GOM Travel and Pensions Expenditures..... 113 Table 7.3: Overhead Costs as Percentage of Total Program Expenditures ............................................... 114 Table 7.4: Access to FISP by Rural Agricultural Households .................................................................. 117 Table 7.5: Expenditures and Institutions fragmentation: Malawi SP Programs and Implementing Agencies .................................................................................................................................................................. 120 ACKNOWLEDGEMENTS The 2013 Public Expenditure Review (PER) was prepared jointly by the World Bank and the Government of Malawi (GOM). The team wishes to acknowledge the leadership and support provided by the former Secretary to the Treasury, Mr. Randson Mwadiwa (MOF), throughout the process. Under the overall leadership of the Secretary to the Treasury, the Government PER Management Team and the Technical Staff Teams were established from the beginning of the process and made excellent contributions to the success of the PER. Mr. Simbani (Director of Debt and Aid), Mr. Kumwembe (former PS Administration, MOF, and current Secretary to Treasury), Dr. Kabambe (former Budget Director, MOF, and current Planning Director, MOH), Mr. Mphwiyo (former Deputy Director, MOF and current Budget Director), Mr. Zhuwao (Director, Economic Affairs, MOF), Ms. Mabvumbe (Deputy Budget Director, MOF), and Mr. Simwaka (Director of M&E, MEPD) as part of the GOM Management Team provided invaluable advice and perspectives on the institutional and procedural aspects of the country’s planning and budgeting and the medium-term macro-fiscal framework. Other members of the Management Team--Dr. Magwira (Principal Secretary, MoEST), Dr. Mwansambo (Principal Secretary, MoH), Dr. Luhanga (Principal Secretary, MoAFS), Mr. Chitimbe (Principal Secretary, MoTPW), Mr. Sitimawina (Principal Secretary, MoEPD), Mr. Phiri (Principal Secretary – Budget, OPC), Dr. Shawa (Principal Secretary, MoGWD), and Mr. Nkondiwa (Principal Secretary, MoGWD)—also provided the management support and strategic guidance for the Technical staff Teams to complete the individual chapters of the PER. On the World Bank side, the team was led by Tuan Minh Le (Senior Economist, AFTP1). The team wishes to express their gratitude to Kundhavi Kadiresan (Country Director, AFCS3), Marcelo Giugale (Sector Director, AFTPM), John Panzer (Sector Manager, AFTP1), Sandra Bloemenkamp and Laura Kullenberg (former and current Country Managers, respectively), and Vijay Pillai (Country Program Coordinator, AFCZM) for their overall guidance, insightful inputs, management support, and encouragement throughout the various stages of the study. Praveen Kumar (Lead Economist and Sector Leader PREM) provided invaluable oversight and advice on strategy, quality management, and client engagement. The lead authors from the World Bank Team and contributors from the GOM Technical Staff Teams are acknowledged. The World Bank chapter authors are: Chapter 1: Macroeconomic Developments and Public Expenditure Review Framework: Sudhir Chitale (Consultant). Chapter 2: Strengthening Planning and Budgeting Processes: Andrew Bird (Consultant). Chapter 3: Agriculture Sector Review: Olivier Durand (Senior Agriculture Specialist, AFTA3). Chapter 4: Transport Sector Review: James Markland (Senior Transport Specialist, AFTTR). Chapter 5: Education Sector Review: Muna Meky (Senior Education Specialist, AFTEE), and Ronald Mangani (Consultant). Chapter 6: Health Sector Review: Karima Saleh (Senior Health Economist, AFTHW). Chapter 7: Social Protection Sector Review: Maniza Naqvi (Senior Social Protection Specialist, AFTSE). Rui Benfica (Consultant) prepared the technical background work on benefit incidence analysis for the five sectors (Chapters 3-7). P.K. Subramanian (Lead FM Specialist, AFTME) and Khuram Farooq (IFMIS specialist) shared the technical inputs with the World Bank team for updating chapter 2 on strengthening the implementation of IFMIS. The World Bank core team members of the PER also include: Appolenia Mbowe (Senior Economist, AFTP1), Temwa Roosevelt Gondwe (Economist, AFTP1), Ida Manjolo (Senior Social Protection Specialist), Stephanie Sweet (Consultant), Michael Vaislic (Consultant), Melissa Rice (Consultant), Manjiri Bhawalkar (Consultant), Spy Munthali (Consultant), Martha Khungwa (Consultant), Goliath Konywani (Consultant), Klaus Broersma (Consultant), Alex Mkandawire (Consultant), Charles Kaira (Consultant), Jutta Franz (Consultant), and Joel Hourticq (Consultant). Appolenia Mbowe played a critical role for the success of the PER: Since the inception, she had actively coordinated with the GOM and other development partners in Malawi to identify the scope and additional financing of the PER and contributed macro-fiscal insights to help frame the PER concept and structure. Madeleine Chungkong (Senior Program Assistant, AFTP1), Deliwe Ziyendammanja, Zione Edith Kansinde, and Esther Lozo (AFMMW) provided untiring assistance in various aspects of the administration of the PER. Zeria Banda (AFRSC) provided excellent guidance and advice on the dissemination process. The GOM Technical Staff Teams responsible for the respective chapters are: Chapter 1: Mr. Kandio (Deputy Director – Revenue, MoF), Ms. Luhanga (Economist, MoF), Mr. Nyasulu (Economist, MoF), Mr. Loti (Assistant Director – DAD, MoF), Ms. Kanthambi (Economist, MoF), Mr. Nyirenda (Principal Economist, MoEPD), and Mr. Hanjahanja (Economist, RBM). Chapter 2: Ms. Mabvumbe (Deputy Budget Director, MoF), Mr. Perekamoyo (Deputy Budget Director- Planning, MoF), Ms. Samuel (Assistant Budget Director, MoF), Mr. Edward (Budget Officer, MoF), Mr. Banda (Debt and Aid Officer, MoF), Mr. Phiri (Chief Economist, MoEPD), and Mr. Munyenyembe (Chief Accountant, Acc. General). Chapter 3: Mr. Mauwa (Principal Economist – Budget, MoAFS), Mr. Kamchiputu (Chief Accountant, MoAFS), Mr. Munyenyembe (Principal Economist, MoLGRD), Mr. Chowa (Economist, MoEPD), Mr. Nyasulu (Assistant Budget Director, MoF), Ms. Msukwa (Principal Economist, MoF), and Mr. Kondowe (Economist, MoF). Chapter 4: Mr. Phiri and Mr. Kaunda (respectively, present and former Directors of Planning, MoTPW), Ms. Soweya (Senior Economist, MoTPW), Ms. Malongo (Economist, MoTPW), Mr. Njala (Economist, MoTPW), Ms. Nyambose (Acting Director – NAO Support, MoF), Mr. Kamlaka (Principal Budget Officer, MoF), Mr. Kabaghe (Principal Budget Officer, MoF), Ms. Tsonga (Budget Officer, MoF), and Mr. Mulungu (Principal Economist, MoEPD). Inputs from other MoTPW staff and from the Roads Authority and Roads Fund Administration are also gratefully acknowledged. Chapter 5: Mr. Lungu (Director of Planning, MoEST), Mr.Kahcepa (Former Directorof Finance, MoEST), Mr. Mwamlima (Principal Budget Officer, MoEST), Ms. Dub (Budget Officer, MoEST), Mr. Sakala (Budget Officer, MoEST), Mr. Changadeya (Budget Officer, MoEST), Ms. Kawanga (Chief Accountant, MoEST), Mr. Khomba (Principal Budget Officer, MoF), and Mr. Msuku (Principal Economist, MoEDP). Chapter 6: Mr. Njati (Former Director of Planning, MoH), Mr. Kuyeri (Principal Economist, MoH), Mr. Chunga (Economist, MoEPD), Mr. Chipendo (Assistant Budget Director, MoF), and Ms. Kajawo (Budget Officer, MoF). Chapter 7: Mr. Mwalima (Director of Social Protection, MoEPD), Mr. Kachale (Principal Economist, MoEPD), Mr. Mzembe (Economist, MoEPD), Mr. Manda (Assistant Budget Director, MoF), Mr. Matewere (Budget Officer, MoF), Mr. Chaphulika (Budget Officer, MoF), and Mr. Ngwalangwa (Budget Officer, MoF). The report benefited from and is dedicated to our late friend and colleague Eduardo Ley. The Team greatly appreciated the insightful comments from peer reviewers: Dino Leonardo Merotto (Lead Economist, PRMED), Khwima Nthara (Senior Economist, EAP), Manuk Ghazanchyan (IMF), John McIntire, Gael Raballand (Senior Economist, MNSPS), Cristian Aedo (Senior Education Economist, ECSH2), Abdo S. Yazbeck (Lead Economist, AFTHW), and Michele Gragnolati (Sector Leader, LCSHD). Dr. Tench offered priceless insights into the framing of chapters 1 and 2 and perspectives on integrating and regularizing the PER in the on-going PFEM reform agenda. The team gratefully acknowledges the financial support and technical inputs provided by the other development partners, particularly the AfDB, DFID, the European Union, GiZ, and Norway. Tireless efforts by P.K. Subramanian and Mr. Simbani to organize the PFEM MDTF funding to complete the PER are greatly appreciated. EXECUTIVE SUMMARY The macro-fiscal framework indicates a tightened resource envelope over the medium term: While domestic revenue effort is expected to run on its continued rising trend, grants are to be reduced from their seemingly peak level in 2012. A tight macroeconomic and fiscal framework coupled with a high level of non-discretionary spending is seen by the MoF as leaving little fiscal space for funding new policy and strategic initiatives. Policies responses must restrain and prioritize expenditures and improve efficiency while meeting the objectives of growth and social protection. This Public Expenditure Review (PER 2013) is prepared in response to a request by the Government of Malawi. It is aligned with the fifth Country Assistance Strategy (CAS) FY13-FY16. The PER offers policies to improve public expenditure efficiency – defined in the context of Malawi as delivering similar or improved level and quality of Government services with constrained overall resource envelope as described in the new Government of Malawi (GOM) fiscal framework. It is worth noting that the PER 2013 was completed with the Government technical inputs just prior to the revelation of scandalous embezzling of public funds and defraud on a large scale through the Information Management Information System (IFMIS).1 The full fiscal impact of such events therefore cannot be gauged within the scope of this PER. It indicates, however, that the plan for further reforming PFM in general and IFMIS in particular should be reviewed in this new light at both technical and political levels and include the option for re-implementing IFMIS with strengthened management, strict oversight, and strict accountability across government agencies. The PER has four main objectives. First, it supports the government to enhance the quality and efficiency of public financial management and provide inputs to the preparation of its budget. Second, it complements the on-going Public Finance and Economic Management (PFEM) reforms. Third, it provides development partners in Malawi with analytical inputs into their operations. Fourth, the PER is expected to become a crucial component of the implementation of the fiscal framework, underpinning the new Extended Credit Facility (ECF) agreed with the IMF in July 2012. As in the past, the proposed PER was prepared jointly with the Government of Malawi. It draws on major analytical work carried out by the World Bank2, and other national and international organizations. It relies on the following major sources of data: (i) primary data collection from Government Ministries, Departments and Agencies (MDAs) and local governments for five fiscal years for which there is complete data (2006/07 through 2010/11); (ii) documentation available in the five sectors3 under the scope of the PER; and (iii) consultation with a wide range of stakeholders including major MDAs, and NGOs. The PER 2013 consists of seven chapters. The first two chapters focus on the overall macro-fiscal framework, planning and budgeting processes. Chapters 3-7 analyze the allocative, technical efficiency and equity of sectoral public expenditures in agriculture, transport, education, health, and social protection respectively. The core analysis of the PER is summarized below followed by a consolidated policy matrix. 1 Such wispread corruption led to the arrest of some government officials and the dissolvement of the cabinet in October 2013. 2 The most relevant recent Bank work includes the PER 2001, PER 2007, CEM 2010, and the World Bank May 2012 multi-sector Technical Assistance mission. 3 The five sectors under review are agriculture, transport, education, health, and social protection. i Chapter 1: Macroeconomic Developments and Public Expenditure Framework Recent economic performance in Malawi from FY 2006 can be divided into three distinct phases: (1) FY 2006 - FY 2010: steady growth and stable macro-fiscal environment; (2) FY 2011 - FY 2012: macroeconomic imbalances, falling donor inflows and growth; and (3) from May 2012 onwards: the launch of new policies to restore macroeconomic balances, donor inflows, and growth. The trends in the budget show that there was a steady increase in public expenditures during the period of high growth (FY 2006- FY 2010), which could not be scaled down during the period of economic contraction (FY 2011-FY 2012). During the high growth phase, Malawi deployed the increased fiscal resources from aid and increased tax effort mainly for increased wage bill and for domestically financed development expenditures. Within the context of tightened budget over the mid-term, the Government aims to achieve fiscal sustainability by strengthening revenue mobilization and restraining growth in expenditures. With the strong projected tax and non-tax revenue intake and declining grants, the fiscal accounts indicate that the overall expenditure would have to be reduced by about 3.5 percentage points of GDP over the period of FY 2013 – FY 2016. As the reduction in interest payments is targeted at about 1.4 percentage points of GDP, the outlay on goods and services in the recurrent and development budgets is expected to decline by about 2.1 percentage points of GDP. The projected reduction of over 2 percentage points of GDP in total non-interest expenditures is very substantial considering a large portion of the expenditures is non-discretionary and therefore cannot be easily reduced. Sector reforms are required to improve intrasectoral allocation and utilization of public resources. In addition, consistent efforts have to be made to focus on the core fiscal policies to improve PFM, manage the wage bill, reduce the travel costs, and rationalize the public transfers and subsidies, including the FISP. Efforts will have to be made to control malfeasance in implementing the IFMIS with consideration to address deep-rooted problems in policies, processes, and risk of further collusion and patronage. Chapter 2: Strengthening Planning and Budgeting Processes Progress with strengthening planning and budgeting processes has been uneven. Positive results have been achieved in developing a more robust macro-fiscal framework, in strengthening cash planning, and in introducing a stronger output focus to the Budget. However, the approach to budget planning remains incremental with limited policy focus, the major share of Development Partner (DP) project funding remains off-budget, and the planning and budgeting of recurrent and development spending are not well integrated. Measures to strengthen financial management and accounting systems involving the development of an integrated financial management information system (IFMIS) have yet to be fully implemented. Further medium-term expenditure framework (MTEF) reforms would require: (i) development of an initial strategic phase to the budgeting processes that involves reviewing and updating sector strategy frameworks and expenditure plans; (ii) bringing all DP project financing of government capital and recurrent spending on-budget; and (iii) consolidating responsibility for both recurrent and investment expenditure programming, budgeting and monitoring under transferring responsibility for the public sector investment program (PSIP) budgeting and monitoring to the Ministry of Finance (MoF). Measures should also be taken to strengthen Cabinet and Parliamentary involvement in the Budget process through the preparation of a budget policy framework paper (BPFP) and its presentation to Parliament. ii Strengthening the PFM and particularly expenditure control becomes imperative. The recent ‘cash-gate’ scandal strongly hightlights the immediate demand to address the long lasting IFMIS problems: x Delays in Communication of ceilings and budget guidelines to MDAs. x Loss of data/errors due to human intervention. x Budget virements conducted outside the normal system functionality. x Missing clear link of the budget to ministry’s plans or to MTEF. x Inaccurate budget report figures from the system. x Unstable connectivity on the budgeting function. x Lack of real-time updates at regional Central Payment Office (CPO) leading to budget figure variances between approved budgets and IFMIS budgets. x Lack of segregation of duties between user rights and system administration rights; and poor governance in safeguarding accuracy in system information4. The planned redesign of the IFMIS based on a comprehensive review of public financial management (PFM) business processes will need to be prioritized, with particular emphasis given to strengthening accountability, commitment control and accounting functions. Implementation of these measures should be focused around the introduction of an integrated planning and budgeting calendar and ensuring that key calendar deadlines are met. This would require strengthened senior management oversight and accountability for both the timeliness and quality of the budgeting process, backed up by a dedicated unit in the MoF responsible for coordinating and monitoring the budget calendar, including the preparation of budget circulars and guidelines and editing of budget documents. The institutional framework and organizational structure of the planning and budgeting system need to be reviewed in order to align structures better with business processes, reduce duplication of functions, and make more effective use of available capacities. Chapter 3: Agriculture Malawi agricultural policy orientations have produced mixed results in the past decade. On the one hand, public expenditure in agriculture was considerably increased to reach about 19% of total expenditure and the launching of FISP in 2005/06 induced a strong maize production increase allowing the country to recover food self-sufficiency at national level. On the other hand, the country now finds itself blocked in a situation in which the MoAFS has very little fiscal space: FISP absorbs the lion’s share of the Ministry’s financial and human resources (69% of MoAFS financial resources since the FISP inception and 51% of total public spending in agriculture over the 2007/08-2011/12 period). In addition, a substantial share of agricultural spending is not under MoAFS direct oversight (off-Budget expenditure and agricultural expenditure under the supervision of other ministries accounted for 31% of total agricultural spending over the 2007/08-2011/12 period). Agricultural spending is also penalized by numerous inefficiencies. First, the FISP. The FISP as a pro- poor instrument has proved ineffective and has taken a heavy toll on MoAFS staff resources, aggravated by substantial fraud, corruption and distortions. Second, weak linkage between policy framework and budget planning. The issue is compounded with the high fragmentation of aid and the high proportion of off-Budget expenditures that entail limited oversight and ownership by the Government and high transaction costs. Third, low efficiency of budget planning and implementation. Cumbersome procedures, low level of expenditure control, weak monitoring and evaluation, and low staff motivation limit both sector expenditure efficiency and equity. And fourth, high level of centralization of agricultural policy 4 Source: Accountant General’s Department report on the IFMIS Project, July 2013 . iii making and implementation. The process continued with insufficient involvement of deconcentrated administrations and non-State actors. In order to remedy these imbalances and inefficiencies and revive the sector’s capacity to produce and sustain robust growth, the critical reforms are proposed to target the four broad set of policies: (1) Improve intra-sectoral technical efficiency; (2) Operationalize the ASWAp investment framework; (3) Re-design FISP in order to serve productive farmers in a market-smart way and strengthen pro-poor safety nets; and (4) Foster the decentralization process - that would be revived in 2014 with the election of the District Assemblies - through a greater involvement of District administration, local communities, farmers’ organizations, NGOs and private operators. To be a factor of change and progress, commitment by all stakeholders must go beyond intentions and translate into changes in processes and organizational arrangements. Chapter 4: Transport The transport sector in Malawi is regarded as a key contributor to the Government’s Malawi Growth and Development Strategy, the Economic Recovery Plan and the National Export Strategy due to the importance of the sector in servicing trade and development. This chapter covers road, rail, air and water transport modes. In line with the current levels of use and past expenditure, the main focus is on the infrastructure elements of the road sub-sector. It has been prepared in close collaboration with the Ministry of Transport and Public Works (MTPW), the Ministry of Finance (MOF), the Roads Authority (RA) and the Roads Fund Administration (RFA). The findings and recommendations have been developed through a series of discussions with the principal stakeholders. The road sub-sector receives the major part of Government and development partner funding allocations, reflecting road transport’s predominant role in carrying Malawi’s freight and passenger traffic. Major challenges face the road sub-sector, where substantial contractual arrears are limiting the much needed rehabilitation and periodic maintenance of the paved road network, and the recent erosion of Road Fund income has led to reduced maintenance levels. Malawi’s road network is large by regional standards in relation to national GDP and the vehicle fleet is small, making it difficult to generate sufficient income for maintenance through road user charges. Careful project prioritization and the use of road design standards that match the levels of traffic will be key to improving the efficiency of the sector. Road network size, available funding levels and rural access requirements are linked and need to be assessed in a holistic manner in order to develop a strategy to improve rural access. The preparation of a national transport sector strategy would inform Government’s decisions on these issues, and on the optimal balance of investment across the sub-sectors. There are concessions in place in the rail and inland water transport (IWT) sub-sectors, which provide opportunities for investment from the private sector in cases where sufficient demand is present. The rail and inland water modes have faced difficulties in maintaining their market share in the face of strong competition from road transporters, much less being able to fulfil their potential in reducing transport costs. Insufficient levels of maintenance have restricted the growth of the rail and water transport sectors due to the poor condition of much of the infrastructure which has given rise to unreliable services. The development of the Nacala railway line offers an exciting opportunity to make increased use of rail transport. There has been a modest increase in the number of passengers carried by air, but the sector needs urgently to address safety concerns raised by the International Civil Aviation Organisation (ICAO). Whilst significant progress was made by Government in the implementation of recommendations from the PERs carried out in 2001 and 2007, it is noted that several of the previous findings have been highlighted in this PER. Political factors such as the challenge of increasing fuel levies which might lead to higher fuel prices and the need to respond to pressures from politicians to pave particular roads have iv proved difficult to overcome. Strategies must be devised to address these difficult issues if the benefits from the current PER are to be delivered. Chapter 5: Education With education sector expenditure increasingly accounting for the largest share of total GOM expenditure (19% in FY 2013), Malawi's commitment to education is evident. There is rapid growth in on-budget funding, especially from development partners since the introduction of a pooled fund to support the Education SWAp in an attempt to provide a proportionate response to the scale of issues faced in Malawi’s education sector. GOM resources have grown by an average of 25.0% per annum between 2008/09 and 2012/13, while pooled development partner resources increased enormously in 2010 and 2011, by 219.8% and 48% respectively. While enrolments and access at all levels of the education system in Malawi have improved dramatically over the past four decades, there is still concern. Low participation and retention rates among the poorest families and alarmingly low levels of reading and numeracy (among the worst in the region) are evidence that the public education system is not functioning effectively. A rapidly growing population and increased demand by employers and students for skills acquired in post-basic education will only intensify pressure on an already weak system and lead to increased demand on sector investments and recurrent expenditures. The weak outcomes in the education sector are partially a result of limited resources and poor quality of inputs, but also from the inefficient ways in which resources are deployed and managed. Policy recommendations to support a more efficient, effective, and equitable education system include: (1) Improve budgetary execution through enhanced decentralization; (2) Enhance implementation of measures aimed at addressing high drop out and repetition rates; (3) Enhance implementation of measures to lower PTRs; (4) Implement improvemed governance and incentive arrangements for teachers; (5) Deepen implementation of double shifting and overlapping to assist in reducing PTRs and PCRs, and evaluate the impacts on learning outcomes; (6) Reform higher education financing arrangements to enhance equity and avoid duplications; and (7) Improve equitable access and labor market responsiveness of the TEVET system. Chapter 6: Health The period under this study (FY 2007- FY 2012) witnesses three distinct phases in the health public expenditures. The period 2006/07 to 2008/09 saw the increased sector spending (with notably significant boost of external financing following the SWAp agreements). The spending became stabilized (in real terms) due to higher inflation during FY 2009 - FY 2011. In the subsequent period (FY 2012), spending fell as a result of the financial crisis. The sector has become increasingly reliant on external financing. By 2008/09, two-thirds of the sector’s total financing derived from external sources, while the public share had declined from 35% (2002/03) to 18% (2008/09). This pattern creates a high dependence and a vulnerability of the sector performance to external shocks. Malawi was far from meeting its Abuja targets (public share of total government expenditure fell from 9% in 2002-03 to 5% in 2008/09). Findings suggest the existence of both allocative and technical inefficiencies. PE share is increasing at the cost of ORT and capital investment. Geographical differentials in ORT spending are not explained by economic or health needs. Public resources are not shared equitably and according to needs across districts. They are also not necessarily going towards the more cost-effective and high impact interventions. Furthermore, Malawi faces a big challenge in health staffing, especially in lower level v facilities and among positions for physicians. Quality of services was noted to be worse in lower relative to higher level health facilities. Low bed occupancy rate and low bed turnover are significant in public hospitals, especially below the tertiary level (central hospitals). In terms of health sector outcomes, mixed results are observed with respect to meeting the millennium development goals (MDG) health targets. MDG 4 (Reduce Child Mortality) and MDG 6 (Combat HIV/AIDS, Malaria, and other diseases) are on track and likely to be met, but MDG 5 (Improve Maternal Health) is unlikely to be achieved. Key policy recommendations include: (1) Overtime, better commitment is required from public sector, by increasing “domestic” public share of spending in health. MOH could negotiate with MOF for a higher share of domestic spending on health, considering they come up with an efficiency gain strategy. MOH is already preparing a health financing strategy which should include a section on cost containment with options to improve allocative and technical efficiency and to reduce leakages and wastage in use of funds; (2) Continue to support goals for universal coverage on the essential health package (EHP); (3) A public- private partnership (PPP) policy that could improve health service performance and access, especially for the underserved and poor. Continue to partner with non-public sector, e.g. CHAM and NGOs. Strengthen community level access and provide outreach/mobile clinics. Base partnership on performance based and targeted contracts. Consider analysis on who is benefiting from MOH-CHAM service level agreement (SLA) by quintiles; (4) Implement the HRH strategy (2012-16) drafted by the MOH under the EHP priorities and within the fiscal space for health; (5) Prioritize the pharmaceutical supply management reform; and (6) Rationalize capital investment plan: (i) Revise urgently a capital investment masterplan based on needs and medium term objective linking to Universal health coverage (come up with a best tool, e.g. marginal budgeting for bottleneck). This plan could be rationalized and directed to underserved regions, and clinics/district hospitals. (ii) Rationalize equipment investment favoring cost effective interventions: clinics, labs, district hospitals, emergency and first aid. Do inventory, replacement and service plan, institutionalize, and decentralize assessment and databases. Chapter 7: Social Protection Social protection expenditures as currently designed and implemented are insufficient and ineffective. Social Protection expenditures, excluding FISP, total only MK 12.8 billion (US 38.4 million) in 2012/13, which amounts to 2.6% of total government expenditures (MK 476 billion) or 1.1% of GDP (MK 1,183 billion). This spending on SP is half the African region’s average and a fifth of the 5.7% world average. Moreover, by comparison, the FISP represents 4.6% of GDP or 11.5% of the total budget expenditures. In addition, expenditures on pensions (MK 16 billion) for government civil servant retirees equal the entire social protection expenditures for 3 million people. The inclusion of FISP expenditures as Social Protection masks the lack of funds for social protection in the form of other benefits and safety nets. Current expenditures do not adequately cover the vulnerable and in some cases are not well targeted, leading to errors of inclusion and exclusion. SP programs are largely donor-funded and institutionally fragmented, which warrants harmonization. The main mechanisms for safety nets are marred by fragmentation: due to the mini empires or turfs of programs, a plethora of fragmented, unsustainable, insufficient, and uncoordinated cash transfers programs end up not impacting chronic poverty. The Social Cash Transfer (SCT) Program has recertified beneficiaries on the basis of the GOM’s latest household survey using criteria of reaching 10% of households (HHs) in every district who are ultra poor and labor constrained. However, such one-size-fits-all criteria may not be appropriate for different districts with varying needs. Internal efficiency is also problematic. Evidence suggests that Public Works, through cash transfers and asset creation, stimulate Malawi’s economy and have a multiplier effect. Cash transfers have assisted vi beneficiaries to address food security needs, farm inputs purchase, as well as basic health and education needs. However, the Social Cash Transfer Program has, in its set up period and due to low coverage, a high overhead administrative cost ranging from 10% (MASAF) to 19% of total costs (Social cash transfer), and is facing a financing gap that tend to prevent it from covering every district. Short term recommendations for increasing the effectiveness and efficiency of Social Protection expenditures focus on rebalancing the total budget expenditures, better targeting mechanisms, and harmonizing programs and institutions. Specifically, policies are recommended to review, reallocate and rebalance the total budget expenditures to more productive social protection programs; redesign and redefine the FISP; harmonize Social Protection expenditure mechanism through expediting the establishment of the Social Support Fund Mechanisms; and unify beneficiary data and utilize E-transfer. Medium term policies should focus on continued adjustment of total expenditures through systematic monitoring and reviewing to ensure better targeting and institutional coordination; strengthening the registry of beneficiaries on the basis of poverty profile through the unified registry of beneficiaries and linked to the National Registry of all citizens; and periodically reviewing all Social Protection Programs and their trade-offs. vii Overview of Findings and Recommendations Chapter 1 - Macroeconomic Developments and Public Expenditure Framework Institutions to be Issues to Address Actions to Take responsible/coordinated Continued fiscal consolidation required through a combination of sector and cross-sector fiscal policies. - Further implement tax policy and administration reforms to Improve tax effort. The budget is facing - Improve PFM (ref. chapter 2). tightened resources over the medium term mainly due to a - Control wage bill, targeting both the level of civil servants’ hiring and appropriate adjustments of their salaries. MOF, MEPD, Office of reduction in grants from their President and Cabinet high level in 2012. - Control travel costs on a sustainable basis. (OPC), MDAs (with support by DPs) Overall expenditures are - Rationalize transfers and subsidies (especially the FISP). projected to be reduced by 3.5 percentage points of GDP - Reduce the risk of build-up of contingent liabilities in SOEs over FY13-16. operations by following pricing policies to fully recover costs (ref. para 65 for fiscal risk resulting from SOEs). - Implement sectoral policies (ref. chapters 3-7) to improve efficiency and equity of sector public expenditures. Chapter 2 - Strengthening Planning and Budgeting Processes Institutions to be Issues to Address Actions to Take responsible/coordinated Institutional and procedural measures to advance MTEF reforms. Progress with strengthening - Reinforce the strategy-policy-budgeting linkage via the planning and budgeting development of an initial strategic phase to the budgeting processes uneven. processes that involves reviewing and updating sector strategy frameworks and expenditure plans. Approach to budget planning - Develop and implement a strategy for bringing DP project remains incremental, major financing of government capital and current spending on- share of DP project funding Budget. remains off-budget, and planning and budgeting of - Consolidate responsibility for both recurrent and MOF, MEPD, OPC, recurrent and development investment expenditure programming, budgeting and MoAFS, and other MDAs, spending not well integrated. monitoring. DPs with wide stakeholder participation Systemic weakenesses in - Strengthen the Cabinet and Parliamentary involvement in expenditure controls that led the Budget process. to fiscal defrauding involving the abuse of IFMIS. The - Re-implement IFMIS with strengthened controls, issues are manifested in all accountability, and upgraded functionalities. A set of technical, policy, reform measures will have to focus on all three areas of management, and weaknesses in IFMIS: policy and processes, accountability levels. functionalities, and capacity. viii - Introduce an integrated planning and budgeting calendar. Chapter 3 – Agriculture Institutions to be Issues to Address Actions to Take responsible/coordinated Short Term Ministry of Agriculture and Low budget planning and Food Security in implementation efficiency, Improve technical efficiency: Review implementation procedures; coordination with other especially for capital Rationalize internal agricultural sector budget allocation; Strengthen government institutions spending and donor funded M&E; Implement FISP efficiency measures. involved in agricultural activities implemented by the projects and programs, GoM. MOF, MEPD High fragmentation of aid and Initiate dialogue with all stakeholders on the restructuring of MoAFS large number of projects organization and budget as well as donor support in line with implemented off-budget ASWAp. and/or outside national Implement ASWAp effectively and inclusively with all stakeholders MoAFS, MOF, and DPs procedures, resulting in to put MoAFS in the driving seat, to increase efficiency, ownership, limited oversight and accountability and policy outcomes. ownership by GoM and high transaction costs. FISP as a productivity increase program is not Initiate debate on redesigning FISP: targeting the most efficient - Implement a differentiated approach to serve productive farmers in and productive farmers. a market-smart way MoAFS, MOF FISP as a pro-poor program is - Enhance more effective pro-poor safety nets to help the poorest not providing the adequate farming households. support to the poorest farming households. Medium Term Substantial and fast growing Re-balance spending towards currently under-funded subsectors and – but imbalanced – capital investment, especially for agriculture commercialization and agricultural budget, market development (including rural finance), land and water MoAFS, MOF dominated by FISP. sustainable management, technology research and development, and Low capital budgets and dissemination and livestock development. under funded subsectors. Review the current portfolio of demand driven projects and funding Centralized agricultural mechanisms at all levels and strengthen instruments aimed at development and limited maximizing the involvement of local communities, farmers involvement of non-State MoAFS organizations, NGOs and private promoters in the development actors in policy making and process with appropriate support by central and deconcentrated implementation. public services. ix Chapter 4 – Transport Institutions to be Issues to Address Actions to Take responsible/coordinated Short Term Progressively restore Roads Fund (RF) revenues to 2009/10 values (10% of fuel price). Ministry of Finance Funding levels are Transfer fuel levy arrears to RF, and define fuel levy as % of fuel Ministry of Transport and insufficient to meet road price. Public Works sector maintenance needs. Take measures to limit further increases in contract arrears. Roads Fund Administration Identify sources of finance and negotiate payment terms for arrears. Establish & adopt economic and social criteria for project prioritization across all modes of transport. Be more selective and focus on projects which will bring demonstrable benefits: Roads – maintain core network and improve rural accessibility, Ministry of Transport and Effectiveness of project Aviation - address safety concerns, Public Works prioritization & targeting. Railways - link to Nacala Line opportunities, Roads Authority Waterways - challenge concessionaire to develop full potential of market. Improve value for money and quality from road maintenance expenditure by introducing longer term maintenance contracts, and performance-based contracts on a pilot basis. Medium Term Road Network Sustainability: Assess the size and classification of the network in the light of demand and available funding. Ministry of Transport and Public Works Transport Sector Strategy. Roads Authority Sector Regulation: Establish either a single transport sector regulator Roads Fund Administration or regulators for each sub-sector. Quality control in design and construction (roads): Enhance technical and managerial capacities of RA, districts, consultants and Ministry of Transport and contractors to improve all stages in the project cycle. Public Works Roads Authority Sector Capacity Ministry of Local Development. Scale-up implementation of rural access and connectivity: Government and Rural Decentralize the management of rehabilitation and maintenance of Development local roads to district level, and build capacity at local level, Ministry of Education, consolidating existing initiatives. Science and Technology Chapter 5 – Education Institutions to be Issues to Address Actions to Take responsible/coordinated Decentralize personnel emoluments to address weaknesses in budget execution that lead to the accumulation of significant amounts of Improve budgetary execution salary arrears over time. MOF, MoEST (in through enhanced coordination with NLGFC decentralization. Extend implementation of the PSIP grants and other stakeholders) Extend devolution of Teaching and Learning Material funds. x Continue to use the LDF in the construction of primary school infrastructure and extend to community day secondary schools. Increase collaboration between the NLGFC and other stakeholders to deepen the effectiveness of the decentralization process. Enhance implementation of measures to address very high and costly Enhance implementation of primary school dropout and repetition rates. measures to improve internal Implement measures on automatic promotion, fewer repetition and MoEST, MOF efficiency, and assess their small class sizes—complemented with planning for double shifting performance. and overlapping to reduce PCRs. Increase the number of teachers to address very high pupil-teacher and pupil-qualified teacher ratios at all levels. Extend the ODL Enhance the implementation program to secondary education to improve secondary school teacher retention rates. MoEST of measures to lower PTRs. Ensure that the deployment of teachers is equitable across classes and schools. Implement improved Implement measures to enhance teacher performance across the governance and incentive education system in general, and for rural teachers in particular MOF/MoEST arrangements for teachers including rationalizing the rural teacher allowance scheme Adopt output-based and performance-based funding MoEST, Ministry of Reform higher education Economic Planning and financing arrangements to Develop a transparent and sustainable student financing system Development, Ministry of enhance equity and strategic Lands and Housing expansion Increase capacity for internally generated funds by universities TEVETA through research, consultancy and outreach. Promote PPPs in the provision of ancillary services. TEVETA Review the current apprentishep system which is expensive and does MoEST Improve equitable access and not allow for significant access expansion Ministry of Labor labor market responsiveness of the TEVET system. Need to mainstream non-formal skills development for currently underserved groups in TEVET system. Address the problem of costly and confusing multiple assessment through harmonization. Chapter 6 – Health Institutions to be Issues to Address Actions to Take responsible/coordinated Short Term Analyze efficiency impact of substitution between PE and ORT. Prioritize frontline workers, and incentivize distribution to deprived Within recurrent health areas. spending, higher share is Potential savings are to achieve from refining the eligibility criteria being taken up by PE. for students to receive subsidies for pre-service education and for MOH/MOF (in coosultation Efficiency gains in recurrent overseas training and from refining the eligibility for subsidies of the with MOEST). (both PE and ORT) and overseas medical treatment. development expenditures are Rationalize equipment favoring cost effective interventions: clinics, to be improved. labs, district hospitals, emergency and first aid. No new domestically financed infrastructure in the short-term while reviewing and rationalizing overall capital investment. Medium Term - Demand Side xi Patients unable to afford transport costs and have Develop public-private partnership (PPP) policy to improve health limited access to health service performance and access, especially for the underserved and MOH/MOH PPP services in rural/deprived poor. Unit/MOF/Local areas. Continue to partner with non-public sector: CHAM, NGOs. governments/and Ministry Patients unable to afford Strengthen community level access and provide outreach/mobile of Social Welfare and EPD medical treatment costs clinics. and MDAs with vested (services/drugs). Base partnership on performance based and targeted contracts. mandates in social Challenges in identification Consider analysis on who is benefiting from MOH-CHAM service protection. of the poor and the level agreement (SLA) by quintiles. vulnerable. HR issues in the sector: High MOH has drafted a HRH strategy (2012-16), that should be attrition levels, skewed implmented under the EHP prioirties and within the fiscal space for distribution in favor of urban health. areas, and subsidization with MOH in coordination with Elaborate an HRH (public, CHAM and others) assessment study for public monies of tertiary other stakeholders. a benchmarking and baseline situation analysis in which high education for all. Low vacancies, urban bias, accountability mechanisms, and productivity performance and low and performance improvements should be addressed. Continue productivity. Limited HRH strengthening HRH database including public and private sectors. information. Prioritize the pharmaceutical supply management reform. Innovative schemes can be considered by partnering with private sector (e.g. Essential drug shortages. MOH and DPs. insourcing private pharmacies into public hospitals; agreeing on lower mark-ups, etc.). Assess feasibility. Revise urgently a capital investment masterplan based on needs and medium term objective linking to UHC which could be rationalized and directed to underserved regions, and clinics/district hospitals. Infrastructure and Equipment Rationalize equipment investment favoring cost effective MOH and EPD misalignment. interventions. Do inventory, replacement and service plan; Institutionalize; Decentralize assessment and databases. Engage negotiations between MOH and MOF to increase the share of domestic spending on health, given the volatility and uncertainty The Abuja target is currently of the level of donor support for the sector over the medium-long missed. terms. Low revenue base and Hospital reform strategy may identify areas of revenue generation retention policy. MOF, MOH, LGFC, MOF and areas of efficiency gains. Inequitable distribution of Revenue retention can help cross subsidize public health programs, funding across country. etc. Establish a progressive allocative formula that takes into consideration equity and needs. Chapter 7 - Social Protection Institutions to be Issues to Address Actions to Take responsible/coordinated Short Term Rebalance budget by redistribution between sectors to increase SP MOF/MEPD Total Budget Expenditure allocations. One consideration would relate to reallocating part of MOA/MoGC&SW/MoLRD Allocations. FISP expenditures - currently ineffective as SP - to programs that and LDF xii reach the poor effectively and equitably (Specifically the amount of MK25 billion is proposed for inter-sectoral transfer between FISP and SP). SP Expenditures (minus Redesign the implementation of FISP as safety net and productivity- FISP) are inadequate to cover enhancing program. the needs of the vulnerable, while FISP expenditures as currently designed are not well targeted to those in need. Encourage FISP complementarity with other social support FISP is not effectively and programs. Better target FISP towards only rural farming households efficiently reaching the with land. Focus on the second and third quintile and producing poorest farming households Maize and other food crops. and cannot be defined as social protection in the current implementation mode. Social Protection Harmonize SP-fragmented projects of cash transfers and other expenditures are fragmented interventions and schemes into systems. Expedite work on unified across programs and registry of beneficiaries on the basis of Poverty and Income Profile institutions. In most cases, and link it to the national registry system. Invest in E-Transfers. expenditures are MOF/MEPD/ MGC&SW uncoordinated, the funds flow and MoD&E, MoLRD, and agencies involved do not LDF, DPs Merge the Ministries of Gender, Children, Social Welfare and the allow for timely, transparent, Ministry of Disabilities and Elderly into a Ministry of Social accountable delivery of funds Protection with Departments for each sub sector. with minimal risk of duplication. Benefit levels are too low to Transfer levels should be revised annually to enhance effectiveness achieve the intended of interventions. MEPD/ MGC&SW and objectives. MoD&E, MoLRD, LDF Revisions in benefit levels are KFW-EU SP expenditures on cash and non-cash benefits should be increased eroded substantially due to to adequately cover the poor and vulnerable. devaluation. Medium Term SP expenditures should be Strengthen the registry of beneficiaries based on poverty profile adjusted, monitored, and through the unified registry of beneficiaries and link it to the MoF/MEPD, OPCS reviewed to ensure they cover National Registry of all Citizens. (national bureau of registry, the needs of the poorest and Design SP programs on the basis of a review of trade-offs in National IDs) vulnerable: targeting should outcomes and benefits between direct cash transfers versus cash MOA/MoGC&SW/MoLRD be reviewed and adjusted to transfers for Public Works, school meals and other interventions. and LDF. reach those most in need. Enhance horizontal and vertical linkages. xiii CHAPTER 1. MACROECONOMIC DEVELOPMENTS AND PUBLIC EXPENDITURE FRAMEWORK 1.1 Introduction 1. The objective of this Chapter is to provide the overall macro-fiscal framework for the Public Expenditure Review. The Chapter is focused on three areas: (i) the fiscal developments and challenges in the recent past and how they have affected the overall economic performance, (ii) the Macro-Fiscal framework for the coming 3-5 years and its implication to the overall expenditure envelope, and (iii) key cross-sectorial policies that are to support the Government mid-term fiscal expenditures. 1.2 Macroeconomic Trends 2. After experiencing a period of weak macro-fiscal performance marked with large fiscal deficits, high inflation, high interest rates, and unpredictable donor inflows between 2000- 2004, Malawi gained steady growth, restoration of donor support, and a stable macroeconomic environment from 2006-2010. During this period, the economy grew by nearly 7 percent p.a., and inflation remained between 6-8 percent p.a. There was a relatively stable balance of payments with manageable current account deficits. 3. Government reforms, debt relief, and grants (including budget support) from development partners led to steady improvement in the fiscal situation. First, beginning in 2005/65, the Government implemented reforms to improve tax effort, expenditure control, and fiscal discipline. The fiscal deficit, excluding grants, declined from about 13.3 percent of GDP in 2005/06 to about 10.3 percent of GDP in 2009/10 (Table 1.1). Second, Malawi had increased access to debt relief. Malawi reached its HIPC completion point in August 2006 and subsequently qualified for Multilateral Debt Relief Initiative (MDRI). Malawi’s external debt decreased from $3.4 billion (160 percent of GDP) at the end of 2005 to only $0.8 billion (20 percent of GDP) by the end of 2006. The external debt service payments in the budget decreased from 3.7 percent of GDP in 2004/05 to 1.5 percent of GDP in 2006/7. Third, Malawi concluded a series of arrangements with the IMF and received Poverty Reduction Support Grants (PRSGs) from the Bank. The Common Approach to Budget Support (CABS) development partners also resumed budget support. From FY2006-FY2010, on average, Malawi received grantsamounting to nearly 10 percent of GDP p.a. 4. Fiscal restraint, external debt relief and steady donor inflows made it possible for Malawi to maintain an average fiscal deficit (after grants) of about 2.3 percent of GDP . These small fiscal deficits were financed mostly by concessional foreign borrowing. On average, net domestic borrowing was maintained at 0.4 percent of GDP from FY2006-FY2010, with small repayments to the banking system in FY07, FY08 and FY10 (Table 1.1)6. Thus the total stock of domestic debt fell from 24 percent of GDP in 2004 to about 14 percent of GDP by 2010. This reduction helped improve the fiscal situation by reducing domestic interest payments in the budget from 4 to 2.7 percent of GDP (High interest payments, almost 8.5 percent of GDP in FY04, were a major factor in the nation’s earlier fiscal disequilibrium). 5 Fiscal trends are analyzed on a fiscal year basis. FY10 is the year 2009/10 going from July 1 2009 to June 30 2010. 6 Nearly 95 percent of domestic debt is in the form of 91 day treasury bills and other treasury instruments of less than one year maturity. The domestic debt therefore has to be rolled over several times during the year to finance the fiscal deficits. The interest obligations in the budget are therefore vulnerable to intra-year fluctuations in the T- bill rates. 1 Table 1.1: Key Economic Indicators, 2005/06-2012/13 (Percent of GDP Unless Specified Otherwise) 2005/6 2006/7 2007/8 2008/9 2009/10 2010/11 2011/12 2012/13 Actual Actual Actual Actual Actual Actual Actual Estimate National Income and Prices (Calender year) Real GDP Growth 2.1 9.5 8.3 9.0 6.5 4.3 1.9 5.5 Nominal GDP (MK Billion) 424 511 601 710 812 880 1056 1310 Inflation (CPI Annual Average Percent p.a.) 13.9 8.0 8.7 8.4 7.4 7.6 21.3 20.2 Fiscal Trends (fiscal Year) Revenue 27.3 30.6 29.4 32.1 33.8 32.1 26.5 37.8 Tax revenue 16.0 16.5 17.3 18.2 18.6 20.8 19.4 20.6 Nontax revenue 2.0 1.5 1.5 2.3 4.9 3.8 2.7 3.3 Grants 9.4 12.5 10.5 11.6 10.3 7.6 4.4 15.1 o/w Budget support 3.5 1.9 2.3 3.0 4.5 1.8 0.0 6.5 Expenditure and net lending 31.2 31.9 30.0 37.8 33.8 35.0 34.9 40.2 Current expenditure 24.2 20.3 18.6 30.6 25.7 27.2 27.0 32.0 Wages and salaries 5.4 5.1 5.4 5.7 5.9 6.9 7.2 8.1 Of which: health SWAp 0.2 0.2 0.2 0.2 1.1 0.6 1.0 0.0 Interest payments 4.9 3.5 2.2 2.7 2.8 2.7 2.5 2.8 Domestic 4.0 3.1 2.1 2.6 2.7 2.6 2.2 2.4 Foreign 0.9 0.4 0.1 0.1 0.1 0.1 0.2 0.3 Goods and services 8.9 6.6 5.5 13.1 11.0 11.2 9.8 12.0 o/w Generic goods and services 5.0 4.7 3.1 6.7 5.8 4.8 5.5 4.7 Subsidies and other current transfers 5.1 5.1 5.5 9.0 6.0 6.4 6.1 8.3 o/w Transfers to public entities 1.5 1.6 1.4 1.9 1.7 1.7 1.8 1.9 o/w Fertilizer and seed subsidy 1.9 2.0 2.7 5.8 2.9 2.6 2.5 4.4 Arrears adjustment 0.0 0.1 0.0 0.1 0.0 0.0 1.5 0.8 Development expenditure 7.0 11.5 11.5 7.1 7.9 7.7 8.0 8.0 Part I (foreign financed) 6.0 9.0 8.1 4.9 4.5 3.7 3.6 5.0 Part II (domestically financed) 1.0 2.4 3.4 2.2 3.4 3.9 4.4 3.0 Unallocated expenditure 0.0 0.0 0.0 Net lending 0.0 0.1 0.0 0.0 0.2 0.1 0.0 0.0 Overall balance (excluding grants) -13.3 -13.9 -11.2 -17.3 -10.3 -10.5 -12.8 -16.3 Other financing needs n.a. n.a. n.a. 0.1 0.0 0.0 0.0 0.0 Overall balance (including grants) -3.9 -1.3 -0.6 -5.7 0.1 -2.9 -8.4 -1.2 Total Financing (net) 3.9 1.3 0.6 5.7 -0.1 2.9 8.4 0.4 Foreign financing (net) 1.0 1.0 2.5 2.0 0.9 1.3 1.6 1.9 Borrowing 3.8 2.1 2.7 2.2 1.1 1.5 1.9 2.5 Repayments 2.7 1.1 0.2 0.2 0.2 0.2 0.3 0.5 Domestic financing (net) 0.6 -0.3 -1.9 3.7 -0.9 1.7 6.8 -1.6 Discrepancy 1/ 2.6 0.7 0.0 -0.1 -0.1 -0.1 0.0 0.8 Net domestic debt, percent of GDP 15 12 17 18 14 16 20 17 Balance of Payments Trends (Calender Year) Real Export Growth 6.9 34.5 30.9 23.1 15.4 4.3 10.8 -12.5 Real Import Growth 14.6 -12.5 59.1 -18.9 73.4 -8.8 15.9 -9.6 Current A/C Deficit Without Grants -26.7 -16.3 -25.0 -19.3 -21.7 -16.4 -19.3 -4.5 Current A/C Deficit With Grants -11.3 1.0 -9.7 -4.8 -1.3 -5.9 -7.1 -4.3 Gross Official Reserves (Months of Imports) 1.1 1.2 1.5 0.7 1.5 1.0 1.1 Real Effective Exchange Rate Index (1998=100) -5.7 -3.0 20.4 9.5 -6.0 -3.3 -- -- Sources: Malawi Ministry of Finance and IMF and World Bankstaff estimates. 1/ The Discrepency in 2012/13 includes a carry over of revenues from FY2011/12 which was used in 2012/13 to retire unbudgeted borrowing in the last quarter of 2012/13. 2 5. Macroeconomic imbalances interrupted Malawi’s impressive economic performance beginning in 2008. In 2009 high fertilizer prices and the election-year increase in expenditures (by almost 1.2 percent of GDP) triggered a fiscal disequilibrium. Governance problems and a series of political missteps led to a withdrawal of donor support; total grant inflows fell from 10.3 percent of GDP in 2008/09 to only 4.4 percent by 2011/12. Budget support, which finances many critical expenditure items, fell from a high of 4.5 percent of GDP in 2009/10 to 0 by 2011/12. The resulting fiscal and external imbalances required a robust policy response, fiscal adjustment and monetary restraint. Unfortunately, the Government maintained a policy of fixed exchange. With uncontained expenditures and waning donor support, the Government had to borrow domestically, even from the RBM, to finance the widening fiscal deficit. Net domestic financing of the budget went from net repayment to the banking system in 2009/10, to net borrowing of 1.7 and 6.8 percent of GDP in 2010/11 and 2011/12, respectively. 6. The new administration of President Joyce Banda, which took office in May 2012, acted swiftly to arrest the economic crisis and implemented a number of critical economic reforms. x Devalued the Kwacha by 33 percent in one step and adopted a floating exchange rate x Allowed banks and foreign exchange bureaus to set their rates for buying and selling foreign exchange. x Removed the requirement to surrender tobacco export proceeds to the Reserve Bank; sellers can now transfer US dollars directly to their commercial bank accounts at the market-determined exchange rate. x Tightened monetary policy by adjusting the policy rate from 13 percent to 16 percent x Reinstated the automatic adjustment mechanism for petroleum products, which adjusts prices against import parity prices. x Restricted the public sector oil company NOCMA to managing strategic reserves, rather than engaging in regular oil importing. x Increased electricity tariffs by 77 percent so that revenue covers more of the costs of production. 7. The new Government also moved quickly to improve its relationship with donors . Government repealed a number of laws that had undermined human rights and political freedoms. It repealed section 46 of the penal code, which had allowed cabinet ministers to ban publications deemed “not in public interest,” and removed the VAT on newsprint. The USAID – Millennium Development Corporation, which had suspended aid after the July 20, 2011 riots, has resumed their program. In August 2012 the Government entered into an IMF Extended Credit Facility (ECF) program and was supported with a series of “Rapid Response” operations from the World Bank. Finally, some of the CABS group of donors also restored budget support. 8. The government reforms and restoration of aid are yielding positive results. Sales of tobacco increased through official channels, the parallel market for foreign exchange collapsed, and the private sector’s access to foreign exchange including fuel imports increased. However, a return to a growing economy has been slow. GDP has grown only by 4.3 and 1.9 percent p.a. in 2011 and 2012 respectively. By the end of December 2012 the country suffered from a sharp depreciation of the exchange rate, sporadic fuel shortages and a rise in inflation to almost 35 percent. But the Government has maintained the basic principles of its reform program: market-based exchange rate 3 and automatic adjustment of fuel prices. Further policy reform is crucial to a sustainable and stable financial situation. The challenges in achieving this, the main objective of the PER, are discussed below. 1.3 Structure of the Budget and Fiscal Trends 9. The budgets show a steady increase in public expenditures during the period of high growth (2005/06-2009/10), a trend which could not be reversed during the subsequent economic contraction (2010/11-2011/12). During the high growth phase, Malawi benefited from an increase in revenue (Table 1.2). Domestic tax and non-tax revenues rose by almost 5.6 percentage points of GDP, and grants increased by about 0.9 percent of GDP. Meanwhile domestic and foreign interest payments decreased by almost 3.3 percentage points of GDP. As Table 1.2 shows, the government allocated these resources to increased outlays on (i) wages, (ii) generic goods and services, (iii) the fertilizer and seed subsidy, part of the Farm Input Subsidy Program (FISP) (1 percentage point of GDP) and (iv) development expenditures (2.3 percentage points). Table 1.2: Changes in Revenues and Expenditures (Percent of GDP) FY06-FY10 1/ FY10-FY12 Total Revenues 6.5 -7.3 Tax and nontax revenue 5.6 -1.5 Grants 0.9 -5.9 o/w Budget support 0.9 -4.5 Project and Dedicated Grants 0.0 -1.4 Total Expenditures 2.4 1.4 Current expenditure 1.5 1.3 o/w Wages and salaries 0.5 1.3 Domestic and Foreign Interest Payments -3.3 -0.8 Generic goods and services 0.8 -0.3 Fertilizer and seed subsidy 1.0 -0.2 Development expenditure 0.9 0.1 Part I (foreign financed) -1.5 -0.9 Part II (domestically financed) 2.3 1.0 Overall Balance including Grants 4.1 -8.7 Total Financing -4.1 8.7 o/w Foreign financing (Borrowing -Repayments) -0.1 0.7 Borrowing -2.6 0.8 Repayments -2.5 0.1 Domestic Financing (net) 2/ -4.0 8.0 1/ The numbers report change over FY10-FY06 and FY 12 over FY10 2/ Includes discrepency and repayment of arrears. 10. These trends of rising expenditures could not be reversed during the subsequent period of economic contraction and waning donor inflows, from FY 2010 – FY 2012. While there were small reductions in the outlays on generic goods and services and on FISP, (the big-ticket items) the wage bill continued to increase as the Government increased civil service wages. Domestically- 4 financed development expenditures also continued to increase because the Government had entered into contracts financed by those expenditures. The rising deficit from 2010-2012 led to an increase in net domestic financing (by about 8 percentage points of GDP and ultimately, higher domestic interest payments. Expenditures on Generic Goods and Services The total expenditure on goods and services in the 2011/12 budget amounted to about 9.8 percent of GDP (about 28 percent of total expenditure). Of that total, nearly 44 percent (4.3 percent of GDP) was for nondiscretionary, donor-financed specific programs. These were expenditures for elections, emergency maize purchases, and goods and services financed by Health and Education SWaPs or procured by the National AIDS Commission. Faced with limited fiscal resources, the Government could only adjust payments for the remaining, discretionary “generic Goods and Services” (just 5.5 percent of GDP). Chart 1: Structure of Generic Goods and Services Internal Travel 15.8 16.5 Public Utilities Payments Office Supplies and Expenses 8.1 Rents Training Expenses 13.0 32.1 Acquisition of Technical Services 6.2 8.3 Other Unclassfied Goods and Services As indicated in Chart 1, generic goods and services mainly consist of internal travel, rent, training expenses, utilities such as water and electricity, and technical services. While outlays on these items could in principle be reduced, it would adversely affect the day-to-day functioning of the Government. 11. Only a small proportion of the total expenditures are discretionary and even these cannot be scaled down easily. Only the “generic goods and services” expenditures (see text box) and Part II development expenditures are financed by purely domestic resources and therefore, discretionary and controllable in the short term. Together, these discretionary sectors amounted to about 10 percent of GDP and only 28 percent of total expenditures in 2011/12 (Table 1.3). During the period of high growth, using domestic resources to finance Part II development expenditures was an important element of Government’s fiscal adjustment strategy. Even as outlays on generic goods and services remained around 5 percent of GDP, outlays on domestically-financed development expenditures increased from about 1 percent of GDP in 2005/6 to about 3.4 percent of GDP by 2009/10. These expenditures were for road projects and irrigation projects, now being implemented. 5 Thus, these expenditures have been difficult to reduce, even in the face of economic contraction and minimal donor inflows. Table 1.3: Discretionary Expenditures as a Share of GDP and Total Expenditures 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 As a Percent of GDP Total Expenditures 31.2 31.9 30.0 37.8 33.8 35.0 35.0 o/w Total Discretionary Expenditures 6.0 7.1 6.5 8.9 9.2 8.8 9.9 Generic Goods and Services 5.0 4.7 3.1 6.7 5.8 4.8 5.5 Development Expenditures Part II 1.0 2.4 3.4 2.2 3.4 3.9 4.4 As a Share of Total Expenditures Total Discretionary Expenditures 19.3 22.3 21.6 23.6 27.3 25.0 28.2 Generic Goods and Services 16.0 14.6 10.4 17.7 17.2 13.8 15.7 Development Expenditures Part II 3.3 7.7 11.2 5.9 10.0 11.2 12.4 Source: Malawi Ministry of Finance, IMFand Bank Staff Estimates 12. Several important expenditure items, partially financed by donor inflows, must be maintained even if donor funds are withdrawn. When health SWaP funding dried up, the government was still accountable for high expenditures on wages and goods and services in that sector. As indicated in Chart 1 below, the wage bill (as a percentage of GDP) increased steadily during the high growth phase, from about 5.4 percent of GDP in 2005/06 to about 6 percent in 2009/10. SWaP resources financed only 3 percent of the 2005/6 wage bill but almost 19 percent of the 2009/10 bill. In FY11, donors withdrew health sector SWaP financing after disagreements with the Government about the forensic procurement audit of the operation. But the government could not reduce current salaries or renege on commitments to the civil service union to increase salaries in the future. Consequently the total wage bill increased to 7.2 percent of GDP by 2011/12 and the shortfall had to be covered by domestic resources. In addition to wages, SWaP funds also had financed a substantial portion of goods and services. SWaP financing of health sector goods and services had grown from 1.4 percent of GDP in 2005/6 to 2.6 percent in 2009/10. But from 2009/10-2011/12 the health SWaP outlays reduced from 2.6 percent of GDP to 1.7 percent. To cover the shortfall in resources, the Government had to forgo needed goods and services, divert domestic resources, or let arrears accumulate. 6 Chart 2: Trends in Wage Bill and the Share of Wage Bill Financed by Health SWaP 7.5 20 18 7.0 16 6.5 14 6.0 12 10 5.5 8 5.0 6 4 4.5 2 4.0 0 FY06 FY07 FY08 FY09 FY10 FY11 FY12 Heath SWaP as a Share of Wage Bill Wage Bill as a Percent of GDP (Left Axis) 13. The fall in external and domestic resources during the period of contraction, together with the difficulties in scaling back expenditures, resulted in high domestic borrowing and accumulation of arrears. To finance expenditures that could not be scaled down and the resulting deficits (discussed above), the Government borrowed 1.6 percent of GDP on a net basis from the domestic market in 2010/11 and borrowed 6.6 percent of GDP in 2011/12. In addition, a part of the expenditures fell into arrears. According to the National Audit Office (NAO), the Government’s arrears amounted to about MK 72 billion (7 percent of GDP) as of May 2012. There are three types of arrears (see Annex A2 for details). About half the total arrears (MK 35 billon) were incurred by parastatals, which form a contingent liability on the Government. Another MK 30 billion were incurred by Government Ministries, Departments and Agencies (MDAs). Of these, the largest arrears were incurred by the Malawi Police Service (MK 10 billion) and the Road Sector Agencies (MK 9 billion).7 Finally, about MK 5.2 billion came from overdue pension contributions, teachers’ salaries and subscriptions. Some arrears accumulated because of poor financial management and reporting, but also because MDAs and parastatals could convince suppliers to provide goods and services on credit and persuade contractors to do construction work (outside the IFMIS) on the promise of payment at a later date. 14. As a part of planned fiscal adjustment, the Government is preparing to clear the arrears . The Government has issued promissory notes that have a maturity of 3.5 years and an interest rate of 91 day T bill plus a 2 percent premium. The first payments on these notes were due in September 2012. Of the total promissory notes, about MK 33 billion are issued to banks (several of the suppliers and parastatals took out loans and overdrafts from banks which they were not able to service because of the Government’s nonpayment for goods and services). In addition, the Government issued MK 7.7 billion in promissory notes directly to contractors for money owed to them by the MDAs. The Government has also budgeted about MK 10 billion p.a. for clearing arrears that are not covered by the promissory notes. Finally, it issued a promissory note of MK 4.9 billion to Air Malawi as a part of this parastatal’s negotiated privatization. 1.4 The Fiscal Framework (2012/13 - 2015/16) 15. The fiscal framework described below (Table 1.4) shows that the Government faces a tight resource envelope over the medium term, driven by an expected reduction in grants from their seemingly peak level in 2012. In the current fiscal year, 2012/13, the Government had access 7 These are mainly on account of extension of works, redesign of some section of roads, and fuel and foreign exchange shortages in 2011 and 12. 7 to increased resources, almost 10 percentage points of GDP, due to a quick recovery of donor pledges. The first priority before the Government was to allocate these increased resources to social safety net programs for the poorest households, which were hard-hit by the currency devaluation8. The Government increased funds for programs like labor-intensive public works, school meals, school bursary, social cash transfers, and the FISP. In its effort to manage the political and social pressures associated with the ongoing reforms, in February 2013, the Government granted wage increases to striking civil servants. These ranged from 60 percent for the lowest pay grades to 5 percent for the highest grade, with an average increase of 19 percent in nominal terms. To reduce the pressure on the 2012/13 budget, payments for January and February will be deferred to 2013/14; in other words, the payments in 2012/13 will be limited to four months March-June. The impact of these wage increases on the total wage bill in 2012/13 is estimated to be about 0.3 percent of GDP (about 3.8 percent of the wage bill), and about 0.6 percent of GDP (about 7.7 percent of the wage bill) in 2013/14. The Government also indicated that the impact of the wage increases on the overall 2012/13 budget is neutral, since it is being accompanied by subsequent equivalent cuts and savings from the Farm Input Subsidy Program FISP) and other recurrent expenditures in the Ministry of Education. The net impact of the wage increases, expenditure control measures, and increases in grants and revenue indicates progress toward fiscal consolidation. The overall deficit, including grants, is projected to decrease from 8.5 percent of GDP in 2011/12 to only 1.2 percent of GDP in 2012/13. With net external financing included, the budget contains a net repayment to the banking system of only about 1.6 percent of GDP. 8 A World Bank mission assisted the Government in the formulation of the 2012/13 budget. 8 Table 1.4: Projected Central Government Operations (Percent of GDP) 2011/12 2012/13 2013/14 2014/15 2015/16 Change Actual Budgeted Projected Projected Projected 2012/13-15/16 Revenue 26.5 39.0 37.3 36.2 35.5 -3.5 Tax and nontax revenue 22.1 23.9 24.4 24.8 25.1 1.2 Tax revenue 19.4 20.6 21.6 22.0 22.3 1.7 Nontax revenue 2.7 3.3 2.8 2.8 2.8 -0.5 Grants 4.4 15.1 13.0 11.4 10.5 -4.6 o/w Budget support 0.0 6.5 5.0 4.3 3.9 -2.6 Expenditure and net lending 35.0 40.2 38.6 37.2 36.3 -3.9 Current expenditure 27.0 32.2 30.0 29.0 28.5 -3.7 Wages and salaries 7.2 8.1 8.6 8.5 8.5 0.4 Interest payments 2.5 2.8 2.3 1.8 1.5 -1.3 Goods and services 9.8 12.0 10.9 10.7 10.6 -1.4 O/W Generic goods and services 5.5 4.7 4.1 3.8 3.7 -1.0 Subsidies and other current transfers 6.1 8.3 7.4 7.4 7.4 -0.9 O/W Fertilizer and seed subsidy 2.7 4.4 3.5 3.5 3.5 -0.9 Arrears adjustment 1.5 1.0 0.8 0.8 0.8 Development expenditure 8.0 8.0 8.7 8.3 7.6 -0.4 Part I (foreign financed) 3.6 5.0 5.5 5.0 4.6 -0.4 Part II (domestically financed) 4.4 3.0 3.2 3.2 3.2 0.2 Overall balance (excluding grants) -13.0 -16.3 -14.3 -12.4 -11.3 5.0 Overall balance (including grants) -8.5 -1.2 -1.3 -1.0 -0.8 0.4 Total financing (net) 8.5 0.4 1.3 1.0 0.8 0.4 Foreign financing (net) 1.6 1.9 1.8 1.5 1.3 -0.6 Domestic financing (net) 6.6 -1.6 -0.5 -0.5 -0.5 1.1 Discrepency 0.8 -0.8 Memorandum items: Real GDP Growth 1.9 5.5 6.1 6.5 6.7 Nominal GDP 971 1,183 1,401 1,584 1,772 Net domestic debt, percent of GDP 20 17 14 12 11 -6.0 Sources: Malawi Ministry of Finance IMF, and Bank staff estimates. 16. The Government plans to continue fiscal consolidation in 2013/14. The tentative 2013/14 budget framework recognizes that the increase in grants from 4.4 percent of GDP in 2011/12 to 15.1 percent of GDP in 2012/13, is temporary, as it includes undisbursed balances from 2011/12. Therefore the 2013/14 framework projects a decline in grants of about 2 percentage points of GDP from the 2012/13 levels. In addition, the tentative budget also reflects several Government commitments including the cost of the May 2014 tripartite elections estimated at US$60 million (about 1.5 percent of GDP9); the conversion of arrears into promissory notes; deferred wage awards and the full-year impact of wage increases; and expenditures from the privatization of Air Malawi. The overall fiscal deficit after grants remains at the same as in the 2012/13 budget. After taking into 9 The Government is seeking assistance from development partners to cover at least half of the estimated cost of the elections. They have received some indications of support but no firm commitments. 9 account net foreign borrowing, there is a projected net repayment to the banking system of 0.5 percent of GDP. 17. The medium term macro-fiscal framework10 aims at achieving fiscal sustainability by improving revenue mobilization, restraining growth in expenditures and making small repayments to the banking system. The main features of the macroeconomic framework are: x Acceleration of GDP growth from 1.9 percent in 2012 to about 6.5 percent by 2015 x Decline in inflation to single digits. x Overall fiscal deficit (excluding grants) of around 11-12 percent of GDP, in line with projected external grants, and financed by modest (1-2 percent of GDP) concessional external borrowing x Small annual repayments to the banking system, resulting in declining domestic debt and lower domestic interest payments. x Projected decline in interest payments of 1.4 percentage points of GDP x Strong revenue growth so tax and non-tax revenue increases from 22.1 percent of GDP in FY12 to almost 25.2 by FY17. x Projected reduction in grants of nearly five percentage points of GDP, from a high of 14.5 percent in FY13 to 9.2 by FY17. The overall expenditure will have to be reduced by about 3.5 percentage points of GDP over the next five years. Given the projected decline in interest payments of 1.4 percentage points, the outlay on goods and services in the recurrent and development budgets will have to be reduced by about 2.1 percentage points of GDP. 1.5 Restraining Expenditures Through Cross Sectoral Policies 18. A reduction of 2 percentage points of GDP in total expenditures is very substantial, as the majority of expenditures is non-discretionary and therefore cannot be easily reduced. Policies responses must restrain and prioritize expenditures and improve efficiency while meeting the objectives of growth and social protection. Cross sectoral solutions include improved expenditure control, wage and employment policies, policies to manage travel costs, and transfers to parastatals (Chapters 3-7 address sector-specific policies). 1.5.1 Expenditure Control through Public Financial Management (PFM) 19. The 2007 PER identified weak expenditure control as an important cause of overspending and poor fiscal outcomes. Certain reforms are being implemented (Table 1.5). The recent revelation of massive fiscal scandal has however called for forensic audits and comprehensive review of IFMIS. Further PFM and IFMIS reforms are expected to be determined by the outcome of such audits and system review. As the PER was completed just at the time when the major fiscal frauds began uncovered, specific plans for IFMIS reforms are outside of the scope of this PER. The review of the 10 The fiscal framework was developed by the Government and forms the basis of discussions on the Extended Credit Facility (ECF) with the IMF and the Malawi Development Policy Operation of the Bank. 10 Auditor General’s Office’s report on IFMIS (July 2013), however, sheds light on the long-lasting problems to be uprooted: x Loss of data/errors occurs due to human intervention. x The time taken to capture ministry budgets by Budgets Department is too long. x Budget template does not take into account the outputs of the MDAs. x Budget virements are done outside the normal system functionality. x There is no clear link of the budget to ministry’s plans or to MTEF. x Inaccurate budget report figures from system have been observed. x Budgeting/Planning Framework is not available to MDAs. x Connectivity on the budgeting function is not stable. Lack of real-time updates at regional CPOs leads to budget figure variances between approved budgets and IFMIS budgets. x Budget Department does their own system administration although they are also system users. There is no segregation of duties between user rights and system administration rights. x Incidences of MDAs and therefore GoM exceeding budgets are very high. 20. The Accountant General Office’s report also identifies the major process risks, which include the following: IFMIS does not reflect Parliamentary approvals and there is high risk of operating outside approved budgets; Inconsistencies in financial reporting; Too many changes in CoA each year leading to compromise in reporting consistency; Unbudgeted fund releases and Expenditure overruns; Inaccurate budget reports due to poor governance in safeguarding accuracy in system information. 21. Major challenges in the PFM system remain11. The 2011 PEFA assessment found very little improvement since 2007 in the average scores for six key dimensions of the PFM system12. In fact there was a slight decline in the “External Scrutiny and Audit” score. The follow-up on the recommendations of the National Audit Committee and the Parliamentary Accounts Committee (PAC) has been weak. Finally, Malawi continues to accumulate large arrears, due in part to poor expenditure management and fiscal reporting. Parastatals and Government departments frequently acquire goods and services on credit, with no repercussions. 22. The next phase of PFM reforms will make special efforts to address the chronic problem of arrears. First, as discussed in Section 1.3, the Government (National Audit Office) has already verified the current stock of arrears and issued promissory notes to settle them. Second, the Government will address some of the systemic causes of arrears. These reforms include: (i) reconfiguring and strictly enforcing the Local Purchase Order (LPO) module in the IFMIS to strengthen commitment control, (ii) conducting quarterly reviews of arrears, (iii) monitoring more closely the performance of parastatals, which account for nearly 50 percent of arrears, (iv) restructuring the IFMIS so that financial transactions occur with very little manual intervention13, (v) carrying out a media campaign about the established procurement rules, and (vi) improving the cash management system by improving the short term deficit forecasting system. 11 See more details in chapter 2. 12 These are: Budget Credibility, Comprehensiveness and Transparency, Policy Based Budgeting, Predictability and Control in Budget Execution, Accounting Recording and Reporting, and External Scrutiny and Audit. 13 This is being done with the assistance of the multi-donor funded Financial Reporting and Oversight Improvement Project (FROIP). 11 Table 1.5: Progress of Reforms of Expenditure Control Policy Actions 2007-2012 Actions to be implemented Improve budget Improved budget comprehensiveness by The Government plans to progressively extend comprehensiveness improving coverage of donor support. processes for capturing donor funded project transactions in IFMIS to all projects where bank accounts are controlled by the Government IFMIS Rolled out the Integrated Financial Configure the IFMIS purchase order module to Management System (IFMIS). support commitment control. Re-implement IFMIS on the basis of comprehensive review following the planned forensic audits late 2013. Internal Audit Strengthened Internal Audit Committees Further srengthen the Internal Audit Committees in 10 largest spending Ministries. Regularly issue minutes of the Internal Audit Committee meetings. Domestic arrears Domestic arrears built up by Government has already verified these arrears and Government Ministries and parastatals promissory notes have been issued for nearly MK 40 over the last three years amounted to billion have been issued to settle a part of the arrears. MK72 billion (7 percent of GDP). For the rest, the plan is to settle them through budgetary provisions over time. To avert the build up of arrers in the future, the Local Purchase Order Module (LPO) in the IFMIS is being strictly enforced. Payroll management Human Resource Management System HRMIS system to be extended to 12 district councils. (HRMIS) has been introduced. This will avoid the build up of arrears on teachers pay. A payroll audit of all civil servants was carried out and currently salaries are being paid directly to public employees bank accounts. Source: Compiled by Bank Staff 1.5.2 Wage Bill and Civil Service Employment 23. The wage bill is the single largest expenditure item in Malawi’s budget. Managing the wage bill is crucial to controlling expenditures. It has been increasing, rising from about 5.4 percent of GDP and 22 percent of the total expenditures in 2005/06 to about 7.2 percent of GDP and 26 percent of total expenditures in 2011/12. 24. Currently Malawi has about 162,000 employees in the civil service and 6,000-7,000 in the defense forces. About 40 percent of civil servants are teachers employed by Ministry of Education, and another 20 percent are health workers. At about 11 civil servants per 1,000 citizens, the size of the nation’s civil service is roughly in line with regional norms. 25. Given the tightened resource envelope in the fiscal framework, the burden of managing the wage bill will fall on the civil service wages. Over the past five years, civil servants have received 12 large salary increases in real terms. As indicated in Table 1.6, the number of civil servants rose by about 4.5 percent p.a. from December 07-November 12, slightly above population growth during that period. The increase in the wage bill of almost 28 percent p.a., therefore, can be attributed to a high growth of civil servants’ salaries, which increased over the same period by about 23 percent p.a. in nominal terms or nearly 11.5 percent p.a. in real terms. Table 1.6: Growth in Employment and Wages in Civil Service Dec-07 Dec-08 Dec-09 Dec-10 Dec-11 Nov-12 Growth Dec 07 -Nov 12 Percent p.a. Total Employment (No. of Persons) 1/ 124,707 131,958 131,426 148,746 149,619 155,424 4.5 Monthly Wage Bill (MK Million) 1850.1 2456.2 2984.1 4305.6 4786.2 6487.0 28.5 Average Nominal Monthly Wage (MK ) 14835 18614 22706 28946 31989 41738 23.0 Real Monthly Wage (MK in 2007 14835 15906 19197 23026 23113 25524 11.5 prices) CPI in December of the year. 1 1.17 1.18 1.26 1.38 1.64 10.3 Source: Department of Pulic Service Management and Development (DPSM) 1/ Refers to civilian employment only. In addition to these there are about 6000-7000 defense personnel in 2012. Table 1.7: Average Wage (in‘000 Kwachas) and Employment (%) in Private Company Sector and the Civil Service Highest Educational Qualifications Acquired Private Company Civil Servants (%) 1/ Employees Average Employed Average Employed Wage (%) Wage (%) None 8 37.8 8.9 12.1 PSLC 8.9 13.5 11.4 6.1 JCE 11.1 18.0 15.3 14.2 MSCE 17.8 21.1 21.5 44.3 Average for MSCE and Below 11 90.3 17.6 77.0 Non-University Diploma 70.5 5.6 32.5 15.0 University Degree 144.4 3.1 78.7 6.8 Post-Graduate Degree 440.1 0.8 146.7 1.4 Average Non-University Diploma and above 125.8 9.5 52.9 23.2 All Levels 22.1 100 25.8 100 Source: IHS3 1/ Notes: PSLC: Primary School Leaving Certificate, JCE: Junior Certificate Examination, MSCE: Malawi Secondary Certificate of Education 26. In controling civil service wage increases, the Government also needs to correct the disparity between senior civil servants’ salaries and their private sector counterparts’ salaries . Our analysis of the IHS3 data indicates that in 2011, on the whole, the average civil servants’ salaries were about 16.7 percent higher than their private sector counterparts’ salaries. However, that comparison is only valid at the junior levels. 77 percent of civil servants have below-MSCE 13 education, compared to 90% of private sector employees. These civil servants earn about 60 percent higher average salaries than comparable workers in the private sector. On the other hand nearly 23 percent of the civil servants have at least a non-university diploma (compared to only 9.5 percent of private sector employees) but they earn about 60 percent less than their counterparts. 1.5.3 Travel Costs 27. There is a Government consensus civil sevants’ travel costs are very high, and could be significantly reduced if some of the inefficiencies and malpractices are addressed. The Government, with the assistance of the CABS group of donors, commissioned a review of public expenditures on travel in 2010. The Travel PER (TPER) found the following. x Travel costs14 amount to between 4-5 percent of GDP, 12-14 percent of total expenditures, and about half of the total expenditure on goods and services. x Travel costs in Malawi are much higher than in comparable countries. For instance, in Uganda and Tanzania, travel costs amount to only 2 and 1.6 percent of GDP respectively. x Domestic travel costs are 80 percent of total travel costs; external travel and vehicle maintenance account for the remaining 20 percent. x Subsistence allowance amounts to about 31 percent of the total domestic travel costs and 22 percent of salaries. x Fuel costs amount to 23 percent of domestic travel costs; thus, travel costs are highly sensitive to fuel prices. Table 1.8: Travel Expenditures as a Share of Total Expenditures 2006/07 2007/08 2008/09 2009/10 External Travel 1.0 1.2 0.9 0.9 Domestic Travel 8.0 10.4 9.8 8.6 Vehicle running expenses 2.9 2.8 2.2 2.1 Total Travel Expenditures 1/ 11.9 14.4 12.9 11.6 Memorandum Items Expenditures on Goods and Services 20.5 18.2 34.8 32.7 Travel Expenditures as a Share of GDP 3.8 4.3 4.9 3.9 Source: Public Expenditure Review of Travel 1/ Excludes vehicle purchases expenditures amounting to between 1.6 and 2 percent of recurrent Expenditures 28. There are three reasons why travel costs are high in Malawi. Poor internal controls in Ministries and Departments leads to widespread malpractice, as shown by internal and external audits and case 14 The travel cost defined here exclude cost of vehicle acquisition which, in 2010 amounted to an additional two percent of recurrent expenditures 14 studies referenced in the TPER. Some common malpractices include: collecting allowances without travelling, collecting multiple per diems for a single day, using government fuel for private purposes, kickbacks for vehicle service, and not crediting the Government for refunds. The analysis in the TPER suggests that false and questionable claims could amount to 30-40 percent of total claims. Second, there are many common practices which are not necessarily illegal but create inefficiencies. These include unnecessary travel and needless events to collect allowances and unnecessarily large Government delegations. Third, there is a perception that travel allowances are a salary supplement. As indicated above, subsistence allowances are a significant portion (23 percent) of the salaries in Malawi. 29. Travel costs need to be brought down in order to achieve the targets in the macroeconomic framework. In December 2012, the Government announced a moratorium on government-funded international travel, with exemptions overseen by Office of the President and Cabinet (OPC). However, truly reducing travel costs requires reducing domestic costs. The TPER suggests a change in office culture in the civil service. Perhaps it will help if the Government adopts a clear policy and publicizes it widely. In addition, the improved IFMIS could be used to improve internal control and enact penalties for malfeasance. Using the MALSITCH card for fuel purchases could eliminate some reimbursements. The Government could also lease vehicles instead of buying them and sign vehicle maintenance contracts with private garages to prevent kickbacks. These policy changes will have to be evaluated for cost effectiveness before implementation. 1.5.4 Transfers and Subsidies 30. The total subsidies and transfers in the budget amount to between 5 and 6 percent of GDP (Table 1.9). The most important – which amounts to around 40 percent on average of the subsidy and transfer budget – is the fertilizer and seed subsidy under the FISP program. Other transfers go to pensions, road authorities, local governments, and to “public entities,” e.g. subvented organizations which have no independent sources of revenue such as universities, libraries, Malawi Broadcasting Corporation and the National Commission on Science and Technology. Table 1.9: Subsidies and Transfers in the Budget (Percent of GDP) 2005/6 2006/7 2007/8 2008/9 2009/10 2010/11 2011/12 2012/13 Total Subsidies and Current Transfers 5.1 5.1 5.5 9.0 6.0 6.4 6.1 8.3 Of Which: Pension and gratuities 0.9 1.0 0.8 0.8 0.8 1.4 1.1 1.3 Transfers to road and revenue authorities 0.9 0.4 0.5 0.6 0.6 0.6 0.6 0.6 Transfers to public entities 1.5 1.6 1.4 1.9 1.7 1.7 1.8 1.9 Transfers to local governments 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Fertilizer and seed subsidy 1.9 2.0 2.7 5.8 2.9 2.6 2.5 4.4 Source: MOF 31. Although there are no subsidies and transfers explicitly budgeted for the State-Owned Enterprises (SOEs), they pose a fiscal risk in Malawi through their operational losses, arrears and accumulation of contingent liabilities. The four main (and problematic) SOEs are: Air Malawi, Water Utilities, ESCOM, and ADMARC. As per the Ministry of Planning Economic Reports, the four parastatals reported a small profit in 2011/12 and required no transfers from the 15 budget. However, all of these SOEs pose a risk to the budget. Contingent liabilities arise from unpaid loans to banks and arrears to suppliers15, while the operational losses mainly arise from tariffs set below cost recovery levels. The tariff policies have generated many implicit and explicit subsidies. 32. The Government’s strategy for SOEs: x Air Malawi: The Government has found a strategic partner for Air Malawi, the Ethiopian Airlines, following its liquidation in February 2013. The Ethiopian Airlines has 49 percent stake in the new Malawi Airlines Limited, with 20 percent of the shares owned by the Government and remaining 31 percent by the public. x For both water utilities and ESCOM, the main problem is setting tariffs to cover costs with regular adjustments based on changes in exchange rate and inflation. Currently implicit subsidy still exists for ESCOM, as tariffs are inadequate to fully cover costs. x The Government converted all of ESCOM’s unsustainable on-lent loans to equity in 2011 in order to strengthen ESCOM’s balance sheets. x The National Oil Company of Malawi (NOCMA) once imported oil and sold it at a low official price, incurring major losses for the Government. By the end of 2012 the price stabilization fund had lost MK17.5 billion. To avoid these risks, the Government has recently limited NOCMA to its core activity of managing strategic reserves of fuel. The bulk of fuel import business will revert to the private sector. 15 As of August 2012, these parastatals together had accumulated payment arrears amounting to nearly US$ 37 million. 16 CHAPTER 2. STRENGTHENING PLANNING AND BUDGETING PROCESSES 2.1 Introduction 33. The chapter presents an overview of the progress and remaining challenges in planning and budgeting processes since early 2000s. It identifies measures for advancing Malawi’s medium- term expenditure framework (MTEF) reform in order to provide a more effective link between planning and budget processes. This will require: (i) development of an initial strategic phase to the budgeting processes that involves reviewing and updating sector strategy frameworks and expenditure plans: (ii) bringing all DP project financing of government capital and recurrent spending on-budget; and (iii) consolidating responsibility for both recurrent and investment expenditure programming, budgeting and monitoring under the Ministry of Finance (MoF). Measures should also be taken to strengthen Cabinet and Parliamentary involvement in the Budget process through the preparation of a budget policy framework paper (BPFP) and its presentation to Parliament. 2.2 Progress with Planning and Budgeting Reforms 34. The initial thrust of Malawi’s public expenditure management reform in the late 1990s involved the introduction of a medium-term expenditure framework (MTEF). Later, the focus broadened to include strengthening budget execution, commitment control and accounting, and developing an integrated financial management information system (IFMIS). A series of multi-year public financial management (PFM) reform plans have provided the basis for coordinating, managing and resourcing the reform16 with considerable Development Partner (DP) support made available to assist implementation. DP funding commitments for PFM reform initiatives over the period 2001-10 have been estimated at around USD 104 million, although actual expenditures are believed to have been considerably less17. 35. Recent Public Expenditure and Financial Accountability (PEFA) assessments indicate that the significant progress in strengthening PFM in the mid-2000s has not been sustained. Between 2006 and 2008 scores improved for 13 PEFA indicators and deteriorated for only 2 indicators. The subsequent 2011 PEFA assessment gave a more mixed picture with scores improving for 6 indicators and deteriorating for 8 indicators (Table 2.1). Malawi scores relatively well in the areas of budget credibility, comprehensives and transparency and policy-based budgeting (6 out of 10 indicators assessed in the 2011 PEFA were rated B or better), but significantly worse for budget execution and control, accounting and reporting, and external scrutiny and audit (4 out of 15 indicators rated B or better). 16 The four reform plans were: (i) the 2003 Malawi Financial Accountability Action Plan; (ii) the 2006-08 Public Financial and Economic Management Action Plan; (iii) the 2009-11 Revised Public Financial and Economic Management Action Plan; and (iv) the 2011-14 Public Financial and Economic Management Reform Program (PFEMRP). 17 Most of the DP assistance for PFM reform has been provided through directly financed TA operations for which comprehensive disbursement and expenditure data are not available. 17 Table 2.1: Changes in PEFA Scores 2006-2008 and 2008-2011 2006 and 2008 PEFA 2008 and 2011 PEFA Scores tha t Scores tha t Scores tha t Scores tha t Scores tha t were Deteri or- Scores tha t were Deteri or- Improved Uncha nged a ted Not-Scored Improved Uncha nged a ted Not-Scored Credibility of the Budget (PI1-PI4) 0 3 0 1 1 0 2 0 Comprehensiveness and Transparency (PI5-PI10) 2 2 1 1 3 1 1 1 Policy Based Budgeting (PI11-PI12) 1 0 1 0 1 0 1 0 Predictability in Budget Execution (PI13-PI21) 7 1 0 1 1 4 2 2 Accounting, Recording and Reporting (PI22-PI25) 2 2 0 0 0 3 1 0 External Scrutiny and Audit (PI26-28) 0 2 0 1 0 2 1 0 Donor Practices (D1-D3) 1 1 0 0 0 2 0 1 Overall Change 13 11 2 4 6 12 8 4 36. An analysis of progress with planning and budgeting reforms highlights problems in sustaining key reform measures. The 2001 Public Expenditure Review (PER) noted that, despite adoption of an MTEF, Malawi’s budget process was still characterized by a single -year fiscal framework and a dual budget. Formal rules and procedures governing budget formulation execution and reporting were not enforced. DP project financing and parastatal contingent liabilities were not fully captured in the government budget and accounts. Incomplete and inconsistent economic and financial data hampered preparation of economic forecasts and contributed to the lack of timely and comprehensive financial reporting. 37. Malawi has made steady progress in improving the transparency of Budget processes . The 2012 Open Budget Survey (OBS), carried out for the Washington based International Budget Project assessed Malawi’s OBS score for budget transparency at 52 out of 100 having increased from 28 in 2008 and 47 in 2010. Nevertheless Malawi was classified as only providing “some information” on its budget. Among the 25 Sub-Saharan African (SSA) countries surveyed Malawi’s achieved the fourth highest transparency scoring being exceeded by Namibia (55%), Uganda (65%) and South Africa (90%). Malawi scores are lower for the other OBS indicators covering public engagement in the budget process, and the strength of the roles played by the Legislature and Supreme Audit Institution (Table 2.2). Table 2.2: Open Budget Survey Scorings 2012 Ranking Malawi's No. of among SSA Indicator Group Score Indicators Countries (out of 100) (total 25) 1. Transparency (OBI)# 95 52 4 2. Public Engagement 12 19 5= 3. Strength of Legislature 11 57 12 4. Strength of SAI 4 42 17= # The final OBI scorings are classified according to five levels of transparency: (i) extensive information provided on the Budget (scores 81-100); (ii) significant information (scores 61-80); (iii)some information (40-60); (iv) minimal information; (v) scant or no information (0-20). Scores for the other indicator groups can be similarly classified. Source: Open Budget Survey 2012 18 38. An evaluation of PFM reform in Malawi carried out in 201118 emphasized the linkage between political commitment and reform progress during the preceding decade. It identified the period from 2000-04 as being characterized by limited political commitment to PFM reforms, with reform initiatives being seen as DP led and generally failing to achieve sustainable results. By contrast the following five years saw significant progress linked to a clear government commitment to address corruption through its emphasis on improvements in expenditure control, transparency and accountability. The report also found some evidence that political commitment to reform may have stalled after 2008 and that this could have contributed to the negative change in PEFA scores between 2008 and 2011. Progress on Issues Identified in the 2001 PER 39. Many of the public expenditure management issues identified in the 2001 PER have yet to be addressed effectively (Annex A3). Progress has been made in developing more robust macroeconomic and fiscal forecasts and extending these to a three year period. The MTEF reform has been relaunched on at least two occasions, most recently in 2010-11 with an improved output- based budget presentation and a new budget program classification. Nevertheless, senior officials consider budget preparation to be incremental in approach and lacking a strong policy focus. Recurrent and development expenditure budgets continue to be prepared separately with limited consideration of the recurrent costs arising from capital projects. While progress was made in bringing DP funding on budget during the mid-2000s, more recently the proportion of DP project funding classified as off-budget has increased substantially. 40. Measures to strengthen budget execution have yet to achieve expected results . Improvements have been made in cash management and monthly monitoring and reporting of recurrent expenditures and domestically financed (Part II) of development expenditures. Expenditure processing and accounting reforms have focused on the introduction of an IFMIS which has been rolled out to all MDAs and is now being introduced in local governments. However, a review of the IFMIS carried out in 2012 concluded that it had restricted functionality and operated primarily as a payments system with limited transactional coverage. Other modules including commitment control, procurement and revenue management were not fully functional19. 41. Orderliness and observance of the budget process continues to be a problem. While Malawi has a well elaborated budget calendar, there is little adherence to intermediate deadlines. As a result the time available for budget preparation by MDAs becomes highly compressed which significantly undermines both the quality of budget preparation and its strategic focus. 42. Lack of political support, sequencing and management issues were also identified in the 2001 PER as key factors affecting implementation of public expenditure management reforms . Inadequate linkages to program implementation and front-line delivery meant that reforms where implemented often had little impact on program level outputs and outcomes. The lack of a robust pay and incentive framework across government particularly affecting mid-level technical and senior management positions was identified as a major factor affecting implementation performance. The Government has taken steps to address these issues. PFM reforms plans have sought to provide a more coherent and sequenced framework for strengthening public expenditure 18 Fölscher, A. Mkandawire, A. and Faragher, R. (2012). Evaluation of Public Financial Management Reform in Malawi 2001-10. Fiscus Ltd, UK. Country Case Study Report in the Joint Evaluation of Public Management Reform managed by the African Development Bank, Denmark and Sweden. 19 Phiri, M. and Chisala, R (2012) Report on the Status Review of Integrated Financial Management Information System and High Level Operational Requirements. 19 management. Better remunerated performance-based contracts were introduced for senior civil servants, and pay levels across the civil service increased significantly in real terms 20. The adoption of sector-wide approaches (SWAps) provided for more explicit linkage between program planning, budgeting and performance in some sectors. 2.3 Advancing the MTEF Reform 43. A recent World Bank study of international experience with MTEF’s classified Malawi as having sustained a Level 2 MTEF (Box 2.1). However, the assessment in the previous section indicates that while core elements of a level 2 MTEF are in place there remain significant challenges to realizing an integrated policy led medium-term budgeting process. These challenges include further refining the macro-fiscal framework, making the budget comprehensive of all government expenditure, strengthening linkage between policy priorities and expenditure planning, achieving better integration between current and capital spending programs. These measures will need to be supported by further improvements in the transparency and accountability of the budgeting process, and improvements in downstream financial management and accounting processes. These requirements are discussed below. Box 2.1: Medium-Term Expenditure Framework – Levels and Elements Level Element Features Good Practice Debt and deficit targets based on debt Level 1 MTEF Specifies fiscal targets consistent sustainability analysis. Revenue forecasts Medium with macroeconomic stability and based on forecasting models. Independent Term Fiscal fiscal sustainability and which are macroeconomic forecasts used and fiscal Framework embedded within realistic and forecasts subject to external scrutiny. Aid (MTFF) internally consistent macroecon- commitments included in debt sustainability Level 2 MTEF omic projections. analysis and revenue forecasts. Background paper on MTFF prepared. MoF issues Budget Strategy Paper setting Level 3 MTEF out national development and budget Specifies spending agency expend- priorities. Budget circular includes MDA Medium- iture ceilings based on a comprom- expenditure ceilings. MDA budget requests Term Budget ise between top-down resource reflect strategic objectives and show cost of Framework availability from the MTFF and current and new activities. Sector strategies (MTBF) bottom-up resource priorities to and MDA spending plans are adjusted finance sector spending plans consistent with final budget ceilings and published. Shifts the focus of attention from Sector strategies discuss program outputs, Medium- spending agency to spending outcomes and performance. Budget targets Term program inputs and/or program/- based on MDA output, outcome and Performance agency outputs, outcomes and performance indicators. MDAs report on Framework performance. Focus on value for results relative to targets. Comprehensive (MTPF) money of public spending. spending reviews conducted periodically.. 2.3.1 Medium-Term Macro-Fiscal Framework 44. An uncertain economic environment continues to pose challenges for fiscal and budgetary planning and management. The quality of macroeconomic and fiscal forecasting has improved in 20 Although performance-based contracts were subsequently abandoned in Oct 2009 due to problems with their administration, the Government is currently considering reintroduction. 20 recent years and the medium-term fiscal framework (MTFF) is now updated twice yearly linked to the joint reviews with the IMF of the Extended Credit Facility (ECF). Nevertheless, dependence on agricultural production and exports, and fluctuations in donor budget support mean there is significant uncertainty over medium-term economic and fiscal forecasts. 45. The Ministry of Finance (MoF) does not currently prepare alternative fiscal forecasts based on pessimistic assumptions for agricultural production and aid receipts. The development of lower-case scenarios would help in determining where adjustments to MDA budget might be required. The nature of such adjustments is that they fall unevenly on different MDAs and categories of expenditure. Planning for and recording required adjustments would enable MDAs to better manage cut-backs and also make more explicit provision for restoring funding allocations in subsequent years when conditions improve. Recommendation 46. Linked to the further strengthening macro-fiscal analysis and forecasting capacities introduce an analysis of alternative less favourable scenarios into the MTFF. This should identify in which expenditures categories necessary adjustments and cut-backs in expenditure might be required and how they might impact on MDA resource allocations. 2.3.2 Link between Policy Frameworks and Budgeting 47. Although Malawi has a well-established policy planning process focused around the Malawi Growth and Development Strategy (MGDS), its impact on public expenditure planning has been limited. The MTEF is meant to provide a framework for a more policy-led budget process. However, the MoF considers that budget planning in MDAs continues to be incremental with limited strategic thinking over the implications of policy and program priorities for budgetary resource allocations. There are a number of factors that contribute to this situation. Limitations in the policy framework provided by the MGDS 48. From a budget planning perspective there are significant limitations in the policy framework and sector strategies provided in the MGDS. First there is little analysis of policy trade-offs or specification of policy targets and priorities. Second, the sector strategies are of variable quality and in some sectors are little more than statements of the objectives and role of government in the sector. Third, the MGDS spending plans are presented on a needs basis, rather than prioritised against a resource constraint, and therefore tend to aspirational rather than realistic. Fourth, the MGDS is overly focused on public investment and new spending initiatives and fails to address issues around the reforms needed to make better use of existing resources. Finally, although the MGDS is reviewed annually, there is no process for annual updating and rolling forward of the strategy to maintain the forward planning horizon. Inadequate strategic phase to budget planning 49. A feature of most MTEF reforms has been the introduction of a strategic phase to the Budget process during which: (i) the overall economic and fiscal strategy and priorities for the coming MTEF period are agreed; and (ii) sector strategies and MDA spending plans are reviewed, adjusted and updated. In effect the strategic phase provides an opportunity to modify and update the Government’s strategy and related action plan. Malawi’s present budget calendar provides for a two stage budgeting process with the MoF carrying out an initial update of the macro-fiscal framework and MDAs preparing their initial activity-based spending proposals during 21 October/November, with a the final update of the macro-fiscal framework and MDAs preparing their detailed budget estimates during February/March. 50. However, the initial phase of the Budget process in Malawi does not have a clear strategic focus. In MDAs the preparation of activity based spending proposals is treated as a detailed bottom- up needs-based budgeting exercise, which lacks a strong review and prioritisation focus. In many MDAs this leads to aspirational spending plans that have to be substantially cut-back during the subsequent preparation of MDA draft expenditure estimates. The situation is better in sectors with DP assisted SWAPs where formal strategy and program reviews are used to program donor funding for the coming fiscal year (FY). 51. A further limitation in budget planning is that the strategic phase does not lead to a presentation of the overall strategy, framework and priorities underlying the preparation of the Budget. The budget calendar provides for this requirement through the preparation of a budget policy framework paper (BPFP) to be submitted to Cabinet in December although this has yet to be implemented21. Limited fiscal space 52. A tight macroeconomic and fiscal framework coupled with a high level of non-discretionary spending is seen by the MoF as leaving little fiscal space for funding new policy and strategy initiatives. Limited scope for reconfiguring the Budget brings into question the feasibility of a more strategic policy-based approach to budget planning. From an annual budget perspective expenditure items such as personnel, transfers and subsidies can be considered as non-discretionary while spending on other items such as goods and services may be fully committed to support existing activity levels. However, over the medium-term there will be scope for reallocating resources from lower to higher priority programs and activities and for delivering existing levels of public services more efficiently. A feature of the budget planning processes in Malawi and many other SSA countries is the absence of systematic procedures for reviewing and reprioritising expenditure programs, eliminating redundant functions and identifying potential efficiency savings. A further issue is that a major share of discretionary spending on capital projects and service delivery initiatives is financed off-budget through donor funded projects. Recommendation 53. Measures to reinforce the link between the Government’s strategic policy and program priorities and the Budget should be centred on strengthening the initial strategic phase of an integrated planning and budgeting cycle. This would involve: x Developing and rolling out guidelines for MDAs to review and update their sector strategies and spending priorities during the initial strategic phase of the budget process. The guidelines should be developed jointly by MoF and the Ministry of Economic Planning and Development (MEPD) so that the process can be integrated with the monitoring and updating of the MGDS. x Replacing the current activity-based costing exercise by a top-down prioritisation and resource allocation exercise in MDAs that identifies: (i) required shifts in resource allocations between spending programs and sub-programs consistent with the priorities set out in the sector strategy; (ii) the scope for potential efficiency savings that would allow current levels of service provision 21 The MoF prepared a “pilot BPFP” for the 2012/13 Budget, but this was not presented to Cabinet or circulated. 22 to be delivered at lower cost; and (iii) priority initiatives to be resourced from efficiency savings and/or additional funding available within the MDA’s resource ceiling. x Providing preliminary budget ceilings to guide the initial strategic phase of budget preparation in MDAs. These could be set as a baseline ceiling to fund current activities and on-going investment projects and an additional conditional ceiling for financing new initiatives to improve service delivery. The starting point for determining the ceilings should be the outer year ceilings in the current MTEF adjusted to reflect the updated macro-fiscal framework. 2.3.3 Budget Comprehensiveness 54. A substantial share of externally financed project support provided by DPs and lending agencies is classified as off-budget and not reflected in the Budget Estimates (Table 2,3). The Debt and Aid Management Division MoF tracks off-budget support and prepares an annex to the Budget Financial Statement giving details of projected off-budget support. This is shown only for the coming year broken down by project and does not include projections for the outer two years of the MTEF or actual and revised estimated disbursements for the previous two years. For 2012/13 off-budget project assistance to Government was estimated at MK 64.3 billion for the year, representing 62% of total externally financed project support to Government. Slightly over half of estimated off-budget project assistance for 2012/13 was to be provided in the form of loan financing. 55. One reason for the high level of off-budget expenditures is that the MoF has recently tightened requirements for determining which types of external project support qualify for inclusion in the Budget. Only projects for which the Government directly manages all project activities and financing are now included in the Budget. For example, payments made directly by DPs and lending agencies to foreign suppliers are treated as off-budget even where the project is implemented by Government and the payment requested by Government. Projects for which financing is channeled through special government bank accounts held in the commercial banks rather than through the Treasury Single Account may be included in the Budget since government is responsible for their financial management. However in practice such projects are in many cases not included. Consequently, a substantial share of government development expenditure falls outside of the Budget thereby undermining budget comprehensiveness and transparency. This also raises significant accountability issues particularly in the case of loan financed projects where Government will be responsible for subsequent loan servicing and repayment. 56. A significant share of off-budget external project support is also not included in the public sector investment program (PSIP). Only around half of externally financed project expenditures appear to be reflected in the 2012/13 PSIP with the level of coverage varying significantly between sectors. For example, the 2012/13 PSIP shows no DP financing for the Ministry of Health while in the Budget Financial Statement off-budget DP project funding for the ministry is estimated at MK 12.8 billion. While some of this shortfall may reflect technical assistance projects that are not included in the PSIP, it is clear that significant externally financed project capital and operational financing is not being captured in the PSIP. By contrast the PSIP for the Roads Authority shows Part I financing totaling MK 16.5 billon, none of which is included in the Budget Estimates, while the Budget Statement projects off-budget financing at MK 17.6 billion. 23 Table 2.3: Budgeted Development Expenditure 2012/13 and DP Off-Budget Project Financing 2012/13 Estimated Total Budget Off-Budget Development Expenditure (Budgeted) DP Off- + DP Off- as % of Total Part I (DP) Part II (dom) Total Budget Budget Budget + DP (MK billion) Off-Budget 190 Agriculture and Food Security 13.2 1.1 14.3 1.9 16.2 11.9% 210 Irrigation and Water Development 2.8 1.4 4.2 13.4 17.6 76.0% 250 Education, Science and Technology 4.4 5.1 9.5 6.9 16.4 42.0% 310 Health 0.3 3.5 3.8 12.8 16.6 77.2% 420 Roads Authority 0.0 12.2 12.2 16.5 28.7 57.7% Other MDAs 17.5 14.7 32.3 12.8 45.0 28.3% Total 38.3 38.0 76.2 64.3 140.5 45.8% 57. The exclusion of substantial elements of external project financing from both the Budget and the PSIP is a major constraint to the successful implementation of initiatives, such as output-based budgeting, that aim to improve integration between planning and budgeting processes. The impact is further magnified because external project support represents a much greater share of the discretionary funding available to MDAs to fund policy initiatives, particularly in the economic and social sectors. Recommendations 58. The MoF should develop and implement a strategy for bringing DP project financing of government capital and current spending on-Budget. Key elements of this strategy should include: x Agreeing measures with the DPs to enable an increased share of project funding to be channeled through government systems. x Bringing on-budget all DP and loan funded project spending managed by Government outside Treasury systems including: (i) financing channeled through Government bank accounts held in commercial banks; and (ii) direct payments to foreign supplier at the request of Government. x Establishing a unit in the Office of the Accountant General responsible for tracking and bringing to account expenditures on DP funded projects that are implemented off-system. 2.3.4 Unified Approach to the Budgeting of Recurrent and Development Expenditure 59. Limited progress has been made towards adopting a unified approach to budgeting recurrent and development expenditure. Preparation of the PSIP follows a parallel procedure to the development of recurrent spending proposals. It is only at a later stage when the Budget Estimates are prepared that recurrent and development spending proposals are brought together. This results in limited integration in planning recurrent and development spending. Little consideration is given to trade-offs between capital and recurrent spending or to the recurrent costs arising from development projects. The PSIP document sets out a five year forecast of expenditure on each project and analyses proposed spending by MGDS II thematic areas and key policy priority areas. While it shows the total estimated cost for each project, it does not include data on expenditure in previous 24 years. Aside from off-budget allocations there can be significant differences in project level funding allocations in the PSIP and the Budget Estimates. 60. There are practical problems with treating the preparation, appraisal and approval of investment projects as a stage in the budget planning process. For larger projects these tasks typically extend over a period from several months to 2-3 years and involve significant resource commitments. For this reason the process of identification, planning and appraisal of investment projects in other countries often takes place within a much longer term investment programming funding perspective22. 61. There is a substantial non-capital element of budgeted development spending. Across the 2012/13 budget this amounts to around 55% of total development spending (Table 2.4). For the Roads Authority, routine maintenance represents 99% of planned development spending for 2012/13 with rehabilitation and construction of new roads financed almost entirely through off- budget DP project funding. In the Ministry of Agriculture and Food Security non capital items amount to 66% of budgeted development spending in 2012/13 with internal travel and vehicle running costs alone amounting to 22%. Table 2.4: Share of Development Expenditure on Non-Capital Items (2012/13 Budget – excludes off-budget DP financing) Budgeted of which Non- MDA Dev. Exp. Capital Items# MK mi l l i on MK mi l l i on % Local Government and Rural Development 6,094.5 1,665.6 27.3% Agriculture and Food Security 14,312.8 9,481.9 66.2% Irrigation and Water 4,229.2 1,920.3 45.4% Education, Science and Technology 9,498.2 1,129.9 11.9% Health 3,772.5 1,148.4 30.4% Gender, Children and Social Welfare 9,448.1 1,577.4 16.7% Roads Authority 12,150.0 11,981.0 98.6% Other MDAs 16,716.1 12,968.0 77.6% Total 76,221.4 41,872.5 54.9% # Ca pi tal Items compri s e: Item 40: Gra nts a nd Subventions a nd Item 41: Acqui s i tion of Fi xed As s ets Source: 2012/13 Output Ba s ed Budget Document 62. While some of the expenditure on current expenditure items such as professional (consultancy) services is related to the design and management of capital projects, it is clear that a significant amount of what is effectively recurrent expenditure is financed through the development budget. For the Ministry of Agriculture it was reported that up to 30% of the development expenditure could be used for financing current expenditures. It is sometimes argued that such expenditures should be classified as investment or development expenditure because they involve benefits over more than a single year. However, on this basis virtually all education expenditures could be considered as investment. In practice, the division between investment and non-investment expenditures is not clear-cut and is less operationally useful than that between capital spending on physical assets and current spending. Classifying current spending as development expenditures reduces budget transparency and provides MDAs with opportunities for 22 For example, Ireland uses a ten-year capital investment programming and indicative budgeting framework. 25 “double-dipping” to fund their recurrent expenditure programs. It also risks embedding an incremental approach to budgeting in which the recurrent expenditure budget is seen as an “overhead” budget with initiatives to improve service delivery being seen as incremental development expenditure rather than in terms making best use of existing available resources and budgetary provision. Recommendations 63. Moving to a more integrated and unified approach to expenditure planning and budgeting has a number of implications for the definition and scope of the PSIP and for the division of responsibility between the MoF and the MEPD. Specifically it is recommended that: x The PSIP is limited to the Government’s capital investment programme, with the current expenditure element of development expenditure budget transferred to the recurrent expenditure budget except where expenditure is directly linked to investment in a capital asset. x A more strategic approach is adopted for planning and management of the PSIP that focuses on the identification, appraisal, approval and evaluation of major capital investment projects and departmental investment programs within a longer-term planning perspective. x Responsibility for budgeting of both recurrent and development expenditures is further consolidated within the Budget Division of the MoF replacing the existing PSIP budgeting exercise undertaken by MEPD. x Development spending is shown in the Output-Based Budget under the new budget program and sub-program structure. 2.3.5 Budget Transparency and Accountability Publicly Available Budget Documents 64. Malawi’s budget documentation is extensive and generally of high quality, although a pre - budget statement and Citizen’s Budget are not currently prepared. The Budget documents submitted to the Parliament comprise: (i) the Minister’s Budget Statement; (ii) the Economic Report prepared by the MEPD; (iii) the Financial Statement containing the aggregated estimates of revenue and expenditure and supporting annexes; (iv) the Detailed Budget Estimates for each MDA broken down by cost center and by program, sub-program and sub-sub program; and (v) the Output-Based Budgets Estimates for each MDA broken down by budget program and sub-program. Additionally, the MEPD finalizes and issues the PSIP document in May/June prior to the presentation of the Budget in Parliament. Accessibility to budget information would be improved if the full set of budget documents were published on the MoF website. 65. A number of reports are prepared on the implementation of the current year’s Budget but are not consistently made available on the MoF website. The Minister makes a Mid-Year Budget Statement to Parliament reporting on budget implementation during the first half of the FY and supporting the submission of the Supplementary Budget Estimates to Parliament. Since 2010/11 the MoF has prepared a Quarterly Budget Performance Report. An Annual Review of the Public Debt Portfolio is also prepared. 26 66. In more advanced budgeting systems, including South Africa and Uganda, a pre-budget report forms the basis for Parliamentary review of the strategy, forecasts and priorities against which the coming medium-term budget is to be prepared. Typically this takes place 3-4 months before the draft Budget is presented. Experience has shown that earlier involvement of Parliament facilitates subsequent consideration of the draft Budget since the underlying strategic framework has already been discussed. Malawi’s 2003 Public Finance Act anticipates this requirement by obligating the MoF to prepare and submit to the National Assembly by 1st April an Economic and Fiscal Policy Analysis for the next financial year and the subsequent two years. The planned introduction of a BPFP could meet this requirement. 67. Presentation of MDA budgets has improved since the adoption with the 2011/12 Budget of a revised output based presentation of MDA Budget Estimates. This provides an overview statement of the MDA’s mission objects and strategies, a summary of recent achievements and statement of priority measures output and measures for the next three years. Recurrent budget allocations are broken down by program and sub-program23 for the full three years with sub- program level outputs shown for the next year. A major weakness is that development expenditures are not shown at program and sub-program24. 68. A Citizens Budget was introduced in the mid 2000s but its production has lapsed since 2010. A simplified presentation of the Budget in non-technical terms can play an important role in promoting understanding of the budget among the general public and in providing for broader public accountability. Involvement of Cabinet, Parliament and Civil Society 69. The role of the Cabinet in the initial strategic phase of the Budget preparation could be further strengthened. Under the existing budget calendar Cabinet is involved in the review and finalization of the BPFP in December25, and the draft PSIP and draft Budget in April. It also reviews and finalizes the mid-year review of implementation of the current year’s budget prior to the Minister’s Mid-Year Budget Statement. The strengthened strategic phase of the Budget process proposed above would benefit from an earlier consultation with Cabinet, possibly involving a Cabinet Retreat (Box 2.2) to discuss the key policy issues, options and priorities to be addressed during in the preparation of the BFPP and the development of the next Budget. Box 2.2: Cabinet Consultation at the Commencement of Budget Preparation – Uganda and South Africa In Uganda, the strategic phase of budget preparation commences with a Cabinet Retreat held in October to discuss the macro-economic outlook as well as the key policy priorities and issues for the next Budget. In South Africa the Minister’s Committee on the Budget (a sub-committee of Cabinet) meets in the first quarter of the FY (May/June) to discuss the outlook and priorities for the next Budget. This is followed by three-day Cabinet retreat (Lakota) held early in the second quarter of the FY (July) at which updated Departmental (ministry) proposals for the coming MTEF are presented and discussed. 23 Note that in Malawi’s budget program classification programs are generally defined at the level of CoFOG function and therefore most MDA budgets comprise a public administration program and their sector specific program. The main areas of the MDAs operations are therefore shown at sub-program level. 24 However, when preparing their detailed budget estimates MDAs do break-down development expenditures by program and sub-program according to the same program structure as is used for the recurrent budget. 25 As noted above the preparation and approval of the BPFP has yet to become an established part of the budget preparation cycle 27 70. Parliamentarians complain that their role in the budgetary process is perfunctory and they are unable to undertake effective scrutiny of the Government’s budget proposal . In part this reflects the legacy of the “Westminster” parliamentary model under which the draft budget is presented less than a month before the start of the new budget year. This contrasts with practice elsewhere in which the Budget is typically presented to Parliament 2-3 months before the end of the FY. The relevant PEFA good practice benchmarks are for the Legislature to have at least two months to review the Executive’s budget proposal (Performance Indicator 27) and for the Budget to be approved before the start of the new FY (Performance Indicator 11). The absence of a pre-budget report presentation further limits the role of Parliament by not providing a prior opportunity to discuss the strategy and priorities underlying the preparation of the Budget. The 2012 OBS noted that the Legislature had very limited powers to change the Executive’s budget proposal and that the public was not allowed to be present during the Parliamentary Hearings on the draft budget. Recommendations 71. Further measures should be taken to strengthen participation in the budget process and improve the range and quality of information available on the Budget. Specifically, it is recommended that: x The MoF prioritize the introduction of the BPFP linked to the proposals for strengthening the strategic phase of Budget preparation. Following Cabinet endorsement, the BPFP should be submitted to Parliament for information purposes. x Consideration be given to introducing a Cabinet Retreat in September/October to provide an opportunity for early discussion of the macro-economic outlook and the key policy priorities and issues to be addressed during the strategic phase and in the preparation of the BPFP. x Changes are made to the scope and content of budget documents in order to increase clarity and reduce repetition. As the output-based budget presentation is further developed it could replace the detailed budget estimates presentation, which could be retained in the IFMIS for budget implementation control purposes. Similarly, the BPFP will have implications for the structure and content of the Economic Report and the Minister’s Budget Statement. x All public documents on the Budget are made available on the MoF website. Consideration should be given to establishing a dedicated unit in the MoF responsible for preparing documents for publication and maintaining a comprehensive and up-to-date ministry website. 2.3.6 Accounting and Financial Management 72. Despite significant investment in the IFMIS since 2001, Malawi stills lack a robust and unified budget recording and accounting system. Serious frauds in implementation of IFMIS—just discovered in September 2013—indicate the risk of malfeasance and the systemic weaknesses in expenditure management and control. Civil servants in charge could use the IFMIS to defraud the GoM on a massive scale, using multiple loopholes: Monitoring of budget execution at the level of controlling officer of the MDAs was not evident; manual vouchers associtated with purchases were not always available for verification at the Accountant General Department; and bank reconciliation could lag behind by several months (Preliminary Security Audit Assistance Report, September 2013, Soft-Tech Consultants Limited). Forensic audit is expected to be carried out during the last few months of 2013. Also, the 2011-14 Public Financial and Economic Management Reform Program (PFEMRP) found that the IFMIS does not capture significant elements of budget 28 operations including much of the development budget, revenue, debt flows and expenditures of the organizations such as the Road Fund Administration and National Aids Commission. 73. The status review of the IFMIS carried out in 2012 noted that implementation of the IFMIS had been quite selective and that while it provided a payments system and an improved budget preparation module it lacked several core areas of functionality normally associated with an IFMIS. These included commitment control, inventory control, an Electronic Payments System, and effective interface with subsystems such as the payroll system. The incomplete implementation of the IFMIS was attributed to a number of factors including: (i) weaknesses in project management; (ii) insufficient attention to business process review, process re-engineering and change management; (iii) weak IT leadership reflecting the lack of a government wide IT implementation strategy and insufficient IT capacities in the project management unit. The experience mirrors that of other SSA countries that, despite substantial investment, have failed to establish fully operational IMFISs. 74. The particular weakneses in the system of budget execution controls that were exploited in the recent fiscal scandal involving the abuse of IFMIS indicates that efforts are to be made to actually re-implement the IFMIS. Many useful recommendations were alrady made in the various reports issued by the World Bank, IMF and SofTech, focusing on all three key areas: policy and process, IFMIS functionalities, and capacity. Many of the accounting policy and process issues are being catered through the formulation of a revised accounting policies and procedures manual under Financial Reporting and Oversight Improvement Project (FROIP). Weaknesses in MTEF design and cash management function are planned to be addressed through follow-on project after FROIP. IFMIS is planned to be re-implemented with strengthened controls and upgraded functionalities.26 MoF is developing a program for further development of the IFMIS based on a comprehensive review of business processes and related process re-engineering and change management requirements, and systems development options. In carrying out the review, attention should be given to: (i) the implications for organizational mandates, structures and staffing in both the MoF and MDAs; (iii) how to secure the necessary buy-in from the stakeholders; (iv) system sustainability and technical support requirements to maintain the system and undertake further customisation; and (iv) the on-going costs for system maintenance and technical support and hardware replacement and maintenance. Recommendations 75. The MoF should prioritise further development of a comprehensive IFMIS, including the implementation of the commitment control module, as a critical component of its plans to strengthen the planning and budgeting system. In so doing it should draw on the experience of other SSA countries in introducing IFMIS. 2.4 Strengthening Management of the planning and Budgeting Process 76. Capacity limitations are widely cited as a major constraint to the successful implementation of planning and budgeting reforms in Malawi. Capacity issues have often been viewed in terms of pay and incentives, staff shortages in key disciplines and in the provision of training. Since the 2001 PER significant progress has been made in some of these areas. First, the Government has moved ahead with appointing senior staff on fixed term performance related contracts. Second, there have been real increases in pay across the civil service, although there remain problems of staff retention particularly for middle level professional positions. Third, successive PFM reform action plans and 26 Preliminary Security Audit Assistance Report, Soft-Tech Consultants Limited (September 2013). 29 the 2011-14 PFEMRP have provided a comprehensive and sequenced framework for public expenditure management reform. 77. This section looks at management and capacity issues in terms of getting the basics right by having: (i) a well elaborated integrated planning and budgeting calendar; (ii) robust arrangements for managing the calendar; and (iii) an organizational structure that makes effective use of available capacities. Measures will also be required to build analytical and management skills, address staff and skills shortages in key areas such as IT and provide staff with on-the-job training in the new procedures. 2.4.1 Integrated Planning and Budget Calendar Existing Budget Calendar 78. Malawi has a comprehensive Budget Calendar (Annex A4). Budget preparation is framed within a realistic macro-fiscal framework that is updated in October and March linked to the semi-annual review of the ECF. The development of MDA budget proposals involves a two stage process in which MDAs initially prepare a costing of their proposed activities for the coming three years. Following review of MDA activity proposals, the MoF issues the budget circular with approved ceilings which are consistent with the updated macro-fiscal framework. Bi-annual meetings with DPs, held in October and March, are used for programming and confirming planned DP support and form an integral part of the budget planning process. The calendar also provided for the preparation of the BPFP. 79. However, the existing budget calendar has not led to a strategic policy-led approach to budget planning. Because the workstreams are not sufficiently integrated both analytically and organizationally, the budget process risks becoming a series of compartmentalized tasks and deadlines. Examples include: (i) the separate preparation of the activity based recurrent budget and the PSIP; and (ii) inclusion of off-budget financed projects as an annex to Financial Statement, rather than mainstreamed within the wider development and presentation of the Budget. The calendar also does not specify sufficiently the initial strategic phase of budget planning leading to the preparation of the BPFP. Proposed Changes to the Budget Calendar 80. Malawi should look to adopting an integrated planning and budgeting calendar that is focused on the key stages with more detailed guidance and timings provided through the Budget Circulars. The new calendar should also provide for better integration with the Government’s planning processes through a strengthened initial strategic phase (Figure 2.1). Key requirement are: x A comprehensive and integrated approach. Preparation of sector/MDA medium-term strategic expenditure plans and budget requests inclusive of (i) recurrent and development expenditure; and (ii) on-budget and off-budget financing. This would incorporate the existing PSIP programming and budgeting exercise. x A Strengthened strategic phase. This would involve: (i) MDAs reviewing and updating their MGDS sector strategies and developing strategic expenditure plans consistent with expected resource availability (replacing the existing activity-based budgeting exercise); (ii) joint sector spending reviews involving MoF, MEPD, and MDAs to review sector expenditure strategies 30 and MDA strategic spending plans; and (iii) preparation of the BPFP setting out the medium- term macro-fiscal strategy and framework and medium-term sectoral spending priorities. x Improved guidance to MDAs. ¡ A First Budget Circular: setting out the detailed calendar for the preparation of the MTEF and Budget, providing detailed guidelines to MDAs for the initial strategic phase tasks, and giving indicative sector/MDA resource ceilings for the sector strategy reviews and development of MDA strategic expenditure plans. ¡ A Second Budget Circular: providing guidance for the preparation of MDA detailed budget requests and containing preliminary budget ceilings for preparation of MDA estimates submissions. The ceilings would be based on the first macro-fiscal framework update and be approved by Cabinet concurrently with the BPFP. The final budget ceilings/allocations would be negotiated during the MDA Budget Request Reviews. x Enhanced transparency of the budget process: ¡ Documentation: (i) preparation of a Strategic Phase Issues Paper as a background paper for the Cabinet Retreat and the First Budget Circular; (ii) publication of the BPFP as the main output of the strategic phase; and (iii) reintroduction of the Citizen’s Budget. ¡ Consultation: (i) introduction of the Strategic Phase Cabinet Retreat; (ii) Cabinet endorsement of BPFP and preliminary budget ceilings; and (iii) discussion of BPFP in Parliament. Recommendations 81. The MoF together with MEPD should adopt an integrated planning and budgeting calendar. The new calendar should provide for: (i) better linkage between the MGDS sector strategies and spending plans; (ii) a unified and comprehensive approach to expenditure planning that combines the budgeting of recurrent and development expenditure and includes off-budget DP project financing of infrastructure and programs; and (iii) strengthened Cabinet involvement and Parliamentary engagement. 31 Figure 2.1: Proposed Integrated Planning and Budgeting Calendar Jul-Sep Oct-Dec Jan-Mar Apr-Jun BPFP Appropriation discussed in Bill approved Parliament by Parliament Cabinet/ Cabinet Cabinet Parliament Cabinet approves BPFP approves Retreat + Preliminary Budget and Ceilings submits to Parliament Budget Policy Macroeconomic, fiscal Second update of Budget Framework and expenditure policy Macro-Fiscal Documents review (PRGF Review) Paper (BPFP) 2nd Budget prepared Strategic Framework (PRGF prepared Circular + MoF and Phase Issues Review) Preliminary MEPD Paper Ceilings MDA Budget First update of Preliminary Estimates Macro-Fiscal Ceilings finalised Framework proposal 1st Budget Circular + Indicative Ceilings MDA MDA Budget Strategy Request Reviews Reviews MDAs Sector/MDA Sector/MDA Medium- MDA draft Medium-Term MGDS Term Strategic Expenditure Plans Budget Requests prepared review/update Jul-Sep Oct-Dec Jan-Mar Apr-Jun Strategic Phase Budget Preparation Phase Budget Approval Phase 2.4.2 Coordination and Management of the Budget Calendar 82. Failure to adhere to budget calendar deadlines results in insufficient time being available for key stages in Budget preparation. Most notably delays in issuing the budget guidelines and budget ceilings leave MDAs with minimal time within which to prepare their final budget requests. For the 2012/13 Budget MDAs were left with only one week for this task compared with a PEFA good practice benchmark of at least six weeks. Other deadlines and benchmark dates in the budget calendar are also regularly missed due to: (i) insufficient resources in the MoF being allocated to managing the budget calendar; (ii) compartmentalisation and not taking account of the impact of missed intermediate deadlines on subsequent stages of the budgeting process; (iii) not providing an updated budget calendar and specification and guidance on key tasks at the outset of budget preparation. 83. Better management of the budget calendar will be a critical factor to the success of measures to strengthen the planning and budgeting process. This will require strong leadership from senior management in MoF and MEPD in ensuring coordination of the planning and budgeting process. Although the Budget Preparation and Coordination Section in the MoF’s Budget Division has day- to-day responsibility coordination of the budgeting process, its staff are primarily involved in budget analysis and preparation functions. There is no dedicated unit responsible for managing the budget calendar including: (i) updating the budget calendar each year; (ii) preparing the Budget Circulars; (iii) overseeing revisions to the supporting guidelines and organising associated training workshops; (iv) monitoring progress at key stages of the calendar and addressing issues that may give rise to delays; (v) coordinating the preparation and publication of key budget documents; and (vi) carrying out annual ex-post reviews of the process so that lessons learnt can be fed back into planning for the next budgeting cycle. Such a unit might require only 2-3 staff full-time staff. 32 Recommendations 84. The OPC, MoF and MEDP should agree and introduce a package of measures to provide for effective management of an integrated, planning and budgeting calendar. These should include: x Making adherence to Budget Calendar deadlines and quality of budget process outputs part of the institutional and senior management performance indicators for MoF, MEPD and MDAs. x Assigning a dedicated Budget Calendar Management Unit in the MoF responsible for facilitating timely implementation of the key stages in the budget planning process. x OPC taking on an institutional coordination, monitoring and troubleshooting role to facilitate improvement in the timeliness and quality of annual planning and budgeting processes. 2.4.3 Organizational Issues 85. The organizational structure of the planning and budgeting system results in a sub-optimal utilization of available capacities. There is significant duplication in functions between the MEPD and MoF leading to fragmentation of budget processes and staffing capacities: x Macroeconomic analysis and forecasting. Responsibilities for economic policy analysis and forecasting are divided between the Economic Planning Division in MEPD and the Economic Affairs Division in MoF. In the Budget Process MoF takes lead responsibility for development of the macro-fiscal framework and related forecasting while the MEPD prepares the Economic Report that forms part of the Budget presentation. x PSIP preparation and budgeting of development expenditures. There is considerable duplication in the preparation of the PSIP overseen by MEPD and the preparation of the development expenditure component of the Budget overseen by MoF. x PSIP and budget monitoring. Monitoring of PSIP implementation is undertaken by the Monitoring Division in MEPD, while the use of funds allocated for development expenditures is monitored by the Expenditure Monitoring Section in the Budget Division of the MoF. Both organizations undertake field trips to review the progress of development projects. x Aid coordination and management. While the main responsibility for aid coordination rests with the Debt and Aid Coordination Division in MoF, the MEPD retains an International Cooperation Division whose functions include coordination of aid and technical cooperation. 86. With a relatively small civil service and significant capacity constraints, the Government should consider options for streamlining planning and budgeting functions. The present overlap and duplication raises questions over the cost effectiveness of having separate finance and planning ministries. 87. The structure of the Budget Division in the MoF does not sufficiently reflect workstream requirements. The Budget Division is organised into a Budget Preparation and Coordination Section, a Budget Expenditure Monitoring Section and a Cash Flow Management Section (Annex A5). In each section officers are responsible for particular sectors and MDAs. This structure may be contrasted with that in other countries where budget departments tend to be organised into: (i) sectoral sections with MDA desks responsible for oversight of budget preparation, execution and 33 monitoring; and (ii) small specialist units responsible for coordinating budget preparation, budget implementation monitoring, and cash flow planning and management. The advantages of such a structure is that sectoral sections can be headed by relatively senior staff who can interact at a higher level with counterparts in the MDAs while desk officers are able to focus on a narrower portfolio of MDAs. Comparisons with other countries would suggest that some reallocation of posts to the Budget Division from the other Divisions in Treasury Department may also be needed in order to resource a strengthened sector desk structure. 88. A related issue is the tendency to introduce planning and budgeting initiatives without considering capacity demands and the implications for changes to existing procedures. This results in significant process overlay and redundancy and exacerbates capacity demands. Examples include: (i) the retention of separate PSIP and development expenditure budgeting processes; (ii) the significant duplication in content between the different budget documents; and (iii) the requirement for MDAs to prepare activity based budgets which are often unrealistic and contribute little to determination of the MDA final budget settlement. In developing reforms to the planning and budgeting process careful consideration should be given not only to the feasibility of the new procedures and related capacity demands, but also the implications for changes to existing procedures and organizational demands. Addressing these requirements when reforms are introduced will be essential to their eventual success. Recommendations 89. The present organisation of planning and budgeting functions should be realigned to enable more efficient utilisation of available capacities. To achieve this it is recommended that: x Options should be investigated for consolidating and streamlining organisational structures and capacities. Specifically this should consider: ¡ transfers of functions and related posts between MEPD and MoF so that all expenditure programming and budgeting is undertaken by the Budget Division of the MoF; and ¡ as part of a wider streamlining of Malawi’s public administration, whether the remaining functions of the MEPD could be brought within a combined planning and finance ministry; x The MoF should consider restructuring its Budget Division so that: ¡ the majority of staff are assigned to sectoral desk/sections responsible for both expenditure planning and monitoring including DP financed on-budget and off-budget projects; ¡ the Budget Monitoring Section, Cash Flow Management Section and the proposed Budget Calendar Management Unit operate as smaller technical support units. x An assessment should be undertaken of the relative workloads across the MoF’s Treasury Department to determine possible requirements for reallocating posts between Divisions. 34 CHAPTER 3. AGRICULTURE 3.1 Introduction 90. Though agriculture is the backbone of Malawi’s economy, it largely remains subsistence farming plagued by low productivity and high vulnerability. This chapter, building upon the GOM’s PER in Agriculture (AgPER), will analyze the efficiency of public investments in agriculture, the level of public expenditures27, the sector’s expenditure allocative efficiency, regional distribution, and technical efficiency in delivering outputs. The budget and expenditure data analysis will shed light on the achievement of food security, diversification, commercialization, infrastructure development (including irrigation) and impact on the most vulnerable. 3.2 Assessment of Agriculture Expenditures 3.2.1 Level of Expenditures Policy environment 91. Agriculture development and food security are amongst the key priorities of the Government of Malawi (GOM) to achieve sustainable economic growth and poverty alleviation. Increased agricultural productivity, diversification and commercialization constitute key focus areas of the overarching national development framework, the Malawi Growth and Development Strategy (MGDS). This priority has been recently translated into a series of sectoral strategy documents: a Comprehensive Africa Agriculture Development Program (CAADP) compact was signed in 2010, a National Agricultural Policy (NAP) for the period 2010-2016 was developed and an Agricultural Sector Wide Approach (ASWAp), aligned with the CAADP pillars and the MGDS, was finalized in 2010 and updated in 2011. 92. The ASWAp prioritizes three areas for investment: (i) Food security and risk management; (ii) Commercial agricultural, agro-processing and market development; and (iii) Sustainable land and water management. It also identifies two key support services to achieve the sector objectives: (i) Technology generation and dissemination; and (ii) Institutional strengthening and capacity building. The ASWAp is a single comprehensive programme and budget framework that offers a formalized process to promote donor coordination and harmonization of investment and alignment of funding arrangements between GOM and donors in the agricultural sector. It promotes increased use of local procedures for programme design, implementation, financial management, planning and M&E. The ASWAp, implemented through 2011-2015, sets a growth target of 6% per annum for the agricultural sector, in line with the Maputo Declaration and CAADP. Status of Agricultural Sector Expenditure 93. Malawi is one of the few African countries that comply with the Maputo commitment of devoting at least 10% of national public spending to agriculture. Notwithstanding the fact that forestry expenditures could not be obtained and included in the calculation as they should, agricultural expenditures (recurrent and investment) consistently accounted for 18-21% of total 27 These include expenditures channeled through the national budget or off-budget and from various sources including donors and NGOs. 35 national expenditures during 2007/08-2011/12, averaging at19% (Table 3.1). Malawi therefore largely exceeded the Maputo national budget commitment objective. 94. Approved and actual expenditures executed by MoAFS and the Department of Irrigation increased about tenfold over the period, from about USD 20 million in 2000/01 to about USD 200 million in 2011/12, with a peak at USD 315 million in 2008/09 (Figure 3.1). While it had remained at about USD 20-30 million up to then, agricultural expenditure skyrocketed to USD 175 million in 2005/06 with the launching of the FISP. In that year and the following three , actual expenditures exceeded the approved budgets by 36% in 2008/09 due to the surge in fertilizer and fuel prices. Actual expenditures were again contained within approved budgets from 2009/10 to 2011/12. Figure 3.1: Trends in approved and actual expenditures of MoAFS and the Department of Irrigation, 2000/01-2012/13, current USD million28 350 300 250 USD million 200 150 Approved 100 Actual 50 0 Source: AgPER from MoAFS and MoF data 95. In addition to the expenditures executed by the MoAFS and the Department of Irrigation, other public expenditures in the agriculture sector included: i. Agricultural projects and programmes carried out by other ministries or GOM bodies: a. Agriculture-based programmes implemented by the Ministry of Local Government and Rural Development (MLGRD); b. Agricultural programmes carried out under the direct supervision of the Office of President and Cabinet (OPC): in the 2007/08-2011/12 period, this concerned only the Green Belt Initiative with a loan of USD 50 million from the Republic of India in 2010/11 used to purchase tractors and other equipment for cotton ginneries; 28 Conversion of agricultural expenditure into constant terms was not attempted because to adequately reflect the purchasing power of MoAFS budget, three different deflators should be applied: (i) an international fertilizer price deflator for the FISP component which represents about 70 percent of MoAFS expenditure; (ii) international inflation should be used for the other imported goods and services; and (iii) the local consumer price index which would prevail for local costs. Applying only one of these deflators to the total expenditure would produce distorted results. 36 c. The World Bank financed Community Based Rural Land Development Project implemented until 2011/12 by the Ministry of Lands, Housing and Urban Development; d. Agricultural interventions under the World Bank funded Malawi Social Action Fund (MASAF III) implemented under MOF. ii. Transfers to District Councils: in the decentralization process, Districts Councils appeared as votes in the national Budget effective from 2005/06 with seven sectors devolved as of 2006/07 (health, agriculture, education, water, rural housing, trade and gender and community services); fisheries started receiving an individualized allocation in 2009/10 and irrigation in 2010/11. The amounts transferred for agriculture, fisheries and irrigation remain very limited (about USD 3- 4 million, for the 28 districts) and are used exclusively for financing operational costs of agricultural services at district level; iii. Off-Budget expenditures, made of predominantly donor funded projects29: they accounted for 92% of total off-Budget agricultural expenditures over the period of study; and to a lesser extent research programmes implemented by the CCIAR centres present in Malawi (8%); 96. Over the 2007/08-2011/12 period, the expenditures executed by MoAFS and the Department of Irrigation represented 68% of total agricultural expenditures, off-Budget expenditures 25%, expenditures incurred by other ministries 6%, and the transfer to District Councils 1% (Table 3.1). Total agricultural expenditures oscillated between USD 250 million and USD 365 million over the period. Off-Budget expenditures more than doubled as of 2009/10 to stabilize at around USD 100 million per year thereafter, and as a result their share in total agricultural expenditure rose to 35-40%. 97. MoAFS has very little space to promote expenditure efficiency. This is because on one side, its major programme, FISP, absorbs the greatest share of its financial and human resources (69% of MoAFS financial resources since FISP inception in 2005/06 and 51% of total public spending in agriculture over the 2007/08-2011/12 period); and on the other side, a large share of agricultural spending is not under MoAFS direct oversight (off-Budget expenditure and agricultural expenditure under the supervision of other ministries accounted for 31% of total agricultural spending over the 2007/08-2011/12 period). MoAFS is thus left with a very little share of the resources dedicated to agriculture (19% of total agricultural spending over the 2007/08-2011/12 period) to both maintain a minimum level of activity in its traditional missions and possibly promote new high growth potential orientations called for by ASWAp and ERP (irrigation development, agriculture diversification and commercialization). 29 As per the World Bank’s “Practitioners’ Toolkit for Agriculture Public Expenditure Analysis” (Report N°60015- GLB, March 2011), donor expenditures which do not appear in fiscal accounts are considered off-budget. 37 Table 3.1: Distribution of actual agricultural expenditure by institutional status and share in total national expenditure, 2007/08-2011/12 (USD million, current and %) 2007 2008 2009 2010 2011 % in Total 2008 2009 2010 2011 2012 total MoAFS + Irrigation 186 315 167 206 180 1,054 68% Other ministries 13 6 5 59 5 88 6% Tranfers to District Agricultural 4 4 4 4 4 20 1% Councils expenditure Off-Budget 50 40 112 96 100 396 25% Total agricultural 252 364 289 365 288 1,558 100% expenditure Total GoM expenditure 1,224 1,785 1,707 1,932 1,610 8,258 - Share of agricultural in total 21% 20% 17% 19% 18% 19% - Sources: AgPER for agricultural expenditure (2012), IMF for total GoM expenditure Government vs. donors finance 98. Agricultural expenditures are increasingly donor financed. On average, over the 2007/08- 2011/12 period, agricultural expenditures were financed at 55% by local resources and 45% by external resources. However, a change in trends was observed as of 2009/10 with the share of external resources growing to become slightly bigger than that of internal resources in 2010/11 and reach just over 60% of total expenditure in 2011/12. In absolute terms donor support to agriculture kept increasing throughout the period while internal financing dropped by more than 35% in 2011/12. While the agriculture sector was affected by the freeze in donor support to Malawi in 2010 and 2011 through the reduction in general budget support, this was paradoxically more than offset by donor direct financing that had started increasing as of 2009/10, especially off-Budget. 99. On average, over 50% of donor financing was spent on off-Budget projects and programmes, while the share of donor financing that was registered in MoAFS accounts was split almost equally between contribution to FISP (15%) and part I of the Development account (17%). An increase to 28% of the share of the contribution to FISP is also observed in 2011/12, when donors that had traditionally been involved in supporting the programme more than doubled their contribution to help GOM overcome the foreign exchange shortage crisis. 3.2.2 Allocative and Technical Efficiency Allocative efficiency Recurrent vs. development/capital expenditure 100. Since its creation in 2005/06, the FISP has mobilized 69% of MoAFS budget on average, the rest being equitably split between other recurrent and development expenditures (Figure 3.2). While other recurrent expenditures were entirely financed on internal resources, donors contributed to FISP (13% on average over the 2007/08-2011/12 period with a peak at 41% in 2011/12) and to development expenditures (79% on average over the 2000/01-2011/12 period). 38 Figure 3.2: Share of actual recurrent and development expenditures in MoAFS budget, 2000/01- 2011/12, % 100% 90% 80% 70% 60% 50% FISP 40% Other recurrent expenditures 30% 20% Development expenditures 10% 0% Source: AgPER from MoF and MoAFS data 101. Elements of capital expenditure30 within development expenditures during the said period have been extremely low. As in many other Sub-Saharan African countries, Government development accounts in Malawi “hide” substantial amounts of salaries and other recurrent costs. This is an important issue as it reduces budget transparency and precludes ministries from adequately planning and monitoring both recurrent and capital expenditures and in particular, from ensuring that sufficient provision is made for investment operation and maintenance beyond the investment phase. Over the 2000/01-2011/12 period, the non-capital element of actual development expenditures has been estimated at 63% (of which 4% salaries and 59% other recurrent expenditures), leaving only 37% for the actual elements of capital expenditure. As a result, real capital expenditure was very low over the period and rarely exceeded 5% of MoAFS actual expenditures. Wage vs. non-wage expenditure 102. The wage bill in the agriculture sector administration has increased more than eleven times and three times in current and real terms respectively from 2000/01 to 2011/12 (Figure 3.3). The pattern of wage bill growth is however in line with the tendency observed for the civil service as a whole, as a result of an effort by the GOM to improve civil servants’ salaries and motivation.31 30 Defined as one-time investments that will bear fruit and be amortized over a period of time (such as physical infrastructure, equipment, training, as well as salaries and consumables required to put in place these investments), as opposed to expenses related to regular day to day administration services activities. 31 For more detailed analysis of the overall GOM wage bill during 2007-12, see Chapter 1. 39 Figure 3.3: Trends in MoAFS wage bill, 2000/01-2011/12, MK billion 4.0 3.5 3.0 MWK billion 2.5 Salaries current prices 2.0 1.5 Salaries constant 2007 1.0 prices 0.5 0.0 Source: AgPER from MoF and MoAFS data 103. Another finding consistent with observations made for the civil service as a whole is the very high cost of internal travels32. In MoAFS internal travel costs have dramatically increased since 2005/06 (Figure 3.4). They amounted to 60% of salary expenditures over the 2000/01-2011/12 period. In addition, the erosion of senior civil servants’ salaries compared to their private sector counterparts should be corrected in order to increase motivation and accountability, improve staff retention, particularly for middle level professional positions, and effectively start resolving the issue of abusively high travel costs. Figure 3.4: Trends in internal travel expenditures in MoAFS, 2000/01-2011/12, MWK billion 2.5 2.0 Internal travel current prices MWK billion 1.5 1.0 Internal travel constant 2007 prices 0.5 0.0 Source: AgPER from MoF and MoAFS data 32 Travel PER 2010 40 104. The culture to regard travel allowances as a salary supplement generates malpractices and inefficiencies and must be discontinued. The development of this culture has been favored by the lack of control over expenditures (limited implementation of IFMIS), DPs’ competition for GOM staff scarce resources, and the erosion of senior civil servants’ sal aries compared to their private sector counterparts (see below). It leads to widespread malpractices (collecting allowances without travelling, collecting multiple per-diems for a single day, etc.) and inefficiencies (unnecessary travel, large delegations or needless time-consuming workshops), aggravated by the fact that unnecessary travel also increases fuel consumption (estimated at 23% of domestic travel costs) and thus the pressure on scarce forex reserves. In an attempt to limit abuses, a ceiling of five days of travel per month per person had been established, which as a blind measure was not necessarily an optimum solution in terms of efficiency either, but that ceiling was lifted in February 2013 as part of a package to respond to civil servants’ concerns about the erosion of their purchasing power, which again confirms that travel allowances are first and foremost regarded as a salary supplement. Technical efficiency Functional composition and alignment with national strategy 105. Comparing the functional classification of MoAFS actual expenditures over the 2007/08- 2011/12 period (Figure 3.6) with ASWAp intentions for 2011/12-2014/15 (Figure 3.5) reveals significant discrepancies between what is presently being done and what is aimed at. Figure 3.5: Respective shares of focus areas in ASWAp budget, 2011/12-2014/15 (in %) Institutionnal strengthening and Food security and capacity building risk management 6% 36% Technology generation and dissemination 5% Livestock 2% Commercial Sustainable land and agriculture and water management market development 44% 7% Source: MoAFS 2011a 41 Figure 3.6: Functional classification of MoAFS actual expenditures, 2007/08-2011/12 (in %) Institutionnal Other support 6% 9% Technology generation and dissemination Irrigation 5% 2% Livestock 2% Crop production 76% Source: AgPERfrom MoAFS and MoF data Figure 3.7: Functional classification of total actual agricultural expenditures, 2007/08-2011/12 (in %) Other 14% Institutionnal support 8% Technology generation and dissemination 7% Irrigation 6% Crop production 63% Livestock 2% Source: AgPERfrom DPs and GoM data. 106. The current predominance of FISP (69% of MoAFS budget over the 2007/08-2011/12 period) does not leave room for developing the sustainable land and water management (which includes irrigation), commercial agriculture, and market development components to the levels planned in ASWAp. The situation improves slightly when off-Budget expenditures and other ministries agricultural expenditures are brought into the picture (Figure 3.7) but still, it is clear that unless additional resources are raised or shifted from FISP, some crucial components of ASWAp will not receive sufficient support and are highly likely to fail to achieve their objectives. 107. One can also question the relatively low shares of resources in ASAWp dedicated to technology generation and dissemination (5%) and to livestock development (2%). While they more or less correspond to the share these subsectors currently enjoy in agricultural expenditures, they may not be sufficient for these subsectors to express their potential in terms of growth stimulation, all the more since the current annual resources available for support to agriculture total 42 only about half of the resources that were expected when ASWAp was launched (USD 250 to 300 million instead of USD 500 to 600 million in ASWAp budget). Efficiency of Government agricultural expenditure planning and execution 108. Execution rates of on-Budget donor financed expenditures are very low, well under the execution rates for the expenditures financed on national resources, which speaks in favour of more resources to be spent by Government under national procedures. Execution rates have been highly variable for all categories of expenditure throughout 2000/01-2011/12 and no clear tendency can be brought out, except that execution rates for donor funded development activities appear to be consistently lower than those of the other categories of expenditures, all financed on local resources (Figure 3.8). This is an issue frequently encountered in Sub-Saharan Africa countries that is largely accounted for by deficiencies of communication between the donors and the local administration and the difficulty to master donor procedures. In some cases the execution rate may also turn out to be low only because planned expenditures were not correctly computed and/or actual expenditures not fully recorded. Figure 3.8: Execution rates of MoAFS expenditures by type, 2009/10-2011/12, % 100% 80% 60% 40% 20% 0% Salaries Other recurrent Dev Part I Dev Part II Total (donors) (national resources) Source: AgPER from MoAFS and MoF data 109. Also, one could question the relevance of a fiscal year from July to June in a country whose economy is largely agriculture-based with a rainy season extending from November to March. A fiscal year matching the civil year would allow for about three months after the Budget is passed to prepare all activities and then to make full use of the dry season to implement infrastructure works, etc. whereas in the current situation activities are hardly prepared that their implementation gets disrupted by the onset of the rains. In the case of FISP, a fiscal year starting in January would give nine months to procure the fertilizers at the best prices on international markets. Link between policy framework and budgeting 110. Progress remains to be made in strengthening the link between policy framework and budgeting in the agriculture sector, in spite of the adoption of an Agricultural Sector Wide Approach (ASWAp) in 2010. Although Malawi was one of the first countries in Sub-Saharan Africa to introduce a medium-term expenditure framework (MTEF) in the late 1990s and adopted an output-based budget presentation and a new budget program classification in 2010/11, the budgeting process is still very much input-based, incremental in approach and little policy-led. In the agricultural sector, there are also classic cross-cutting public finance management issues, such as the 43 limited implementation of IFMIS that does not allow policymakers to get real-time budget execution data and the separation in the elaboration of the recurrent budget and the development budget. Three inter-related factors explain the slow progress achieved in integrating policy and budget planning in the agriculture sector: i. Insufficient capacities to organize a strategic thinking phase prior to budget planning: capacities at MoAFS Department of Agricultural Planning Services (DAPS) are inadequate in all units (Policy Analysis, Programme Development, M&E) and further weakened by an important turnover; the fact that the Department had no Director for several years until recently is symptomatic of the neglect from which strategic planning suffers. A strategic thinking phase upstream from budget planning would allow MoAFS, in collaboration with all stakeholders, to take stock of the progress made in ASWAp implementation, update strategies and re-establish priorities and allocate budgetary resources accordingly. Planning and M&E capacities must be strengthened at all levels. ii. Inadequate organizational arrangements, resulting in low levels of ownership and accountability: this is a frequent weakness of policy reform attempts in Sub-Saharan African countries: new policies are prepared but the organizational aspects of their implementation are overlooked. The need for revisiting existing procedures and organizational arrangements is not assessed, and this very often results in a “business as usual” behavior amongst the various stakeholders under which the new policy is most likely to remain rhetoric. Adjustments in MoAFS organizational chart and budget are required for a greater consistency with ASWAp architecture. iii. Lack of fiscal space: with the highly politicized FISP that takes the lion’s share of MoAFS budget on one hand and highly fragmented and often off-Budget DP financed projects on the other hand, it is clear that MoAFS fiscal space to achieve a greater linkage between policy framework and budgeting is currently rather limited. Nevertheless, the discretionary funding at MoAFS disposal could be increased through the inclusion in the Budget of all DP financed activities (see next section). Finally, fiscal space can also be brought about through a change of mindset: the budgeting process as currently implemented overly focuses on new spending initiatives and fails to address the possibility of a better use of existing resources through the reorientation of those expenditures failing to produce valuable outcomes, which sends back to the issue of M&E capacities. Aid fragmentation and donor alignment 111. Donor financing in the agriculture sector in Malawi is extremely fragmented, which poses a serious challenge in terms of linkage between policy framework and expenditure33. The inclusion in the Budget of all DP financed activities would also give a greater visibility for strategic planning, and the greater use by DPs of both Government systems and common financing mechanisms (trust funds, basket funding, sectoral budget support, etc.), would also help reduce the currently exorbitant aid transaction costs. In this respect the new trust fund that major DPs34 are currently pulling together to finance the ASWAp-SP is a welcome initiative, provided it becomes a pool of resources available to stimulate a proper annual strategic planning process and not an additional project with pre-set and hardly changeable activities. 33 Multiple donors engaged in financing the agriculture include: the EU, the World Bank, Norway, DFID, AFDB, JICA, India, USAID, Ireland, CGIAR, FICA, IFAD, BADEA, FAO, and others. 34 Norway, EU, DFID, USAID, Irish Aid and FICA, for a total amount of USD 125 million. 44 112. Although DPs claim that their interventions are aligned with ASWAp, alignment remains very theoretical. The only criterion used to support DPs’ alignment claim relates to whether or not the field of intervention of their projects belongs to one of ASWAp key areas, but given that ASWAp broadly covers the whole agriculture sector it would be difficult for a project not to be declared aligned. Vetting project proposals at TWG level in view of their fit into ASWAp under-served priorities and providing orientation to project design at the conceptualization stage would be the first step of effective alignment but such mechanism is not in place yet. 113. The issues discussed above of insufficient attention paid to strategic planning, inadequate organizational arrangements, poor alignment and harmonization of donor funded activities, etc., have been recurrent in policy documents over the past decade, which calls for a better follow- up on validated recommendations of policy documents in the future. The challenge today is not so much to determine what the best options are but rather to effectively start implementing reforms that, for some of them, have been agreed upon more than a decade ago. 3.2.3 Intra-sectoral equity/Benefit Incidence Analysis 114. Although the FISP continues to be politically sensitive given that it is currently GOM’s biggest programme to reach the rural poor, there is now some consensus around the fact that it is not an effective pro-poor instrument. Amongst the various so-categorized social safety net instruments currently in place to reach the poor and vulnerable, the FISP mobilizes over 80% of the financial resources available35. According to implementation guidelines (MoAFS 2011b), FISP is clearly targeted at “resource poor Malawian farmers of all gender categories” with special attention to “vulnerable groups: elderly, HIV positive, female, child, orphan, and physically challenged headed households”. However, various recent studies (Chirwa et al. 2011, Dorward et al. 2012, Benfica 2013) have evidenced that the coupon redistribution mechanisms in place at village level, with village headmen eager to spread the programme benefits as much as possible, result in the subsidy being quite equitably spread across the rural population (Figures 3.9 and 3.10). It was also evidenced that although access to FISP coupons is relatively high among all groups, there is a higher likelihood of receiving coupons among households with larger landholdings – a 25 percentage point difference between the land poor and the land rich (Figure 3.11). In addition, there is no clear evidence that the changes to targeting processes that were introduced to enhance coupon allocation transparency (open meetings, etc.) increased the likelihood of the poor being targeted (Dorward et al. 2012). Finally, although this is not documented, it is thought that large quantities of fertilizers, after having been redeemed by poor smallholders, are resold to better-off farmers in order to generate immediate cash income particularly needed at a time that also corresponds to the lean period and the start of the school year. In total, adding the effects of coupon redistribution and fertilizer reselling, it can be reasonably assumed that the distribution of the subsidy across the rural population is relatively strongly biased towards the better-off quintiles. 35 In 2012/13, the other safety net instruments targeted to the poor and vulnerable (cash transfers, public work programmes, etc.) would mobilize MWK 12.8 billion whereas MWK 54.9 billion were earmarked for FISP (see social protection chapter). 45 Figure 3.9: FISP recipients by wealth and poverty status, 2011 (in %) 70 60 60 55 57 52 54 46 48 50 40 % 30 20 10 0 Poorest Q2 Q3 Q4 Richest Poor Non-Poor Source: Benfica 2013 Figure 3.10: FISP net subsidy share, rural population, 2011, % 25.0 19.6 20.0 20.3 20.3 19.8 20.0 15.0 % 10.0 5.0 0.0 Poorest Q2 Q3 Q4 Richest Source: Benfica 2013 Figure 3.11: Agricultural households receiving FISP by total land area decile, 2011, % 80 69 70 62 60 55 58 58 60 52 54 50 50 44 40 % 30 Log. (%) 20 10 0 Land D2 D3 D4 D5 D6 D7 D8 D9 Land poor rich Source: Benfica 2013 46 115. As a result of not being pro-poor, FISP has logically had no significant impact on rural poverty and some areas are still affected by chronic food insecurity. The latest Integrated Household Survey (IHS3, based on 2010-2011 data) established that rural poverty36 has stagnated at about 56% since IHS2 (2004-2005), while urban poverty has declined from 25.4 to 17.3% (Figure 3.12). Figure 3.12: Poverty and ultra-poverty headcount, IHS2 and IHS3 60 Poverty (%) 55.9 56.6 30 Ultra-poverty (%) 28.1 52.4 50.7 24.5 24.2 50 25 22.4 40 20 30 25.4 15 17.3 20 10 7.7 4.3 10 5 0 0 Malawi Urban Rural Malawi Urban Rural 2005 2011 2005 2011 Source: Benfica 2013 116. Nationally the poverty headcount has not varied significantly, from 52.4 to 50.7%. Ultra- poverty has increased both nationally and in rural areas. As a result, inequality has increased both in rural areas, especially in Central and Southern regions, and nationally, while it remained high but stagnant in urban areas. The IHS3 also established that only 58% of the population enjoy high food security while 33% has very low food security37. The most densely populated Southern region is most affected by chronic food insecurity, with very low food security incidence comprised between 40 and 70% in four districts and over 70% in two districts. These observations from IHS3 confirm the findings of several studies (amongst which Chirwa et al. 2011) which had evidenced that the impact of FISP on its direct beneficiaries in terms of food security, food consumption, self-assessed poverty and overall welfare was weak and that only some positive impact on primary school enrolment, under-5 illness and exposure to shocks could be demonstrated. 3.3. Agriculture Sector Performance and Key Issues 117. The effectiveness of agricultural public expenditure is analyzed through the effectiveness of FISP, which has increasingly taken the lion’s share of agricultural budgets, and through the achievements of agricultural research and extension. 36 The poverty line, comprised of a food (62%) and non-food (38%) element, was set at MK 37,002 per person per year in 2011. The population that had total consumption below MK 37,002 per person per year was deemed poor and the population with total consumption below the food component was considered ultra-poor. Source: NSO 2012. 37 High food security households do not report any concern about accessing enough food; at the opposite, very low food security households experience multiple indications of disrupted eating patterns and reduced food intake during the year. In between, marginal food security and low food security households have concerns about adequacy of the food supply and may occasionally reduce the quality and variety of the food consumed but not its quantity. Source: NSO 2012. 47 Effectiveness of FISP 118. The impact of FISP on agricultural productivity and national food self-sufficiency has been spectacular, as its introduction during the crop season 2005/06 sparked a series of bumper maize harvests induced to a limited extent (5%) by an increase in the area cultivated and to a much larger extent (95%) by a surge in yields (Figure 3.13). The average harvest during the 2005/06-2010/11 period reached 3.2 million tons, representing an increase by about 80% over the average harvest during the preceding six-year period (1.8 million tons). 95% of this increase was accounted for by an increase in yields from 1.2 tons/ha to over 2.0 tons/ha, the remaining 5% being imputable to a slight increase in the area under maize (6%). As a result, the country that had been chronically affected by food shortages in the decade preceding the introduction of FISP, with particularly dramatic episodes in 2002 and 2005, became self-sufficient. Figure 3.13: Evolution of maize yield, area harvested and production, 1990-2011 4.0 3.5 3.0 Production (million tons) 2.5 2.0 Yield (tons/ha) 1.5 Area 1.0 Harvested 0.5 (million ha) 0.0 Source: FAOSTAT http://faostat3.fao.org 119. The effectiveness of FISP as a pro-poor instrument is discussed in the benefit incidence analysis (section 3.2.3). Achievements of Government-based research and extension 120. Although agricultural research and extension are crucial elements that determine agricultural productivity growth, the AgPER review has revealed that both have during the study period received very little funding from public resources. Food crop research 121. The limited resources dedicated to food crop research in MoAFS budget (less than 2.5% for much of the period) have largely been spent on salaries and not in funding research activities . The analysis shows that consistently, food crop research has received less than 2.5% of the total 48 MoAFS budget for much of the period except during 2004/05 and 2009/10 when research exceeded 4 and 5% of MoAFS actual expenditures respectively (Figure 3.14). It is noted however, that the amount of funding to research consistently increased in nominal terms since the beginning of the study period and reached the peak during 2008/09 but has tended to decline since then (Figure 3.15). Furthermore, during most years of the period, actual expenditures have either been closely equal to or slightly lower than the approved budget for the sub-sector. Figure 3.14: Trends in MoAFS budget allocation to research, 2000/01-2011/12, current USD million 12.0 10.0 8.0 USD million 6.0 Approved 4.0 Actual 2.0 - Source: AgPER from data provided by MoF and MoAFS Figure 3.15: Percentage of research in MoAFS budget, 2000/01-2011/12 (in %) 6% 5% 4% 3% Approved 2% Actual 1% 0% Source: AgPER from data provided by MoF and MoAFS 122. External sources account for 60% of food crop research funding, which puts research continuity at risk. A study by Phiri et al. (2012) found that research activities in the Department of 49 Agricultural Research Services (DARS) still largely depended on donor funded projects. This is confirmed by the present AgPER (Figure 3.16). Over the past five years, Government funds channelled through DARS have consistently represented less than 50% of total food crop research funding and even less than 20% in 2010/11 and 2011/12 (38% in average over the period). DARS off-Budget revenues (agreements with CGIAR centres, foundations and donors outside the country, proceeds generated by contracts with seed companies for certification, etc.) have been increasing steadily but are still very limited (USD 0.6 million in 2011/12, 2% of food crop research funding in average). The bulk of funding (60% in average over the period) came from off-Budget non- Government sources, 49% from the GGIAR centres present in Malawi and 11% through off-Budget donor financed projects. 123. This is a major concern as it means that in the absence of CGIAR and donor support, there would be no research going on in public research institutions . Such discontinuity already occurred at the beginning of the 2000s when two important World Bank supported projects ended. Most of the achievements of these two projects got quickly lost due to lack of funding for subsequent operation and maintenance of the investments realized and failure to retain the researchers that had been trained (ASTI 2004). The paper by Phiri et al. argues that unless Government increases its financial contribution to R&D, research agenda in Malawi will remain donor driven, creating a precarious situation for sustainable agricultural productivity growth. Agricultural research needs long-term commitment since investment in equipment and human resources will bear fruit in the medium-long run. Research programs need time to deliver technologies ready for dissemination to farmers. The Government should ensure continued support to research programs to avoid gaps between donor-funded projects that lead to interruption in research activities, loss of results and turn-over of human capital. Figure 3.16: Sources of funding of food crop research, 2007/08-2011/12, current USD million 25 20 Donors (off-Budget) USD million 15 CGIAR centres in Malawi (off- Budget) 10 DARS other resources (off-Budget) MoAFS/DARS 5 0 2007/08 2008/09 2009/10 2010/11 2011/12 Source: AgPER from data provided by MoF, MoAFS/DARS, ICRISAT, CIAT, IITA, CIP and donors Extension 124. MoAFS funding of agricultural extension has increased in absolute terms but dramatically decreased as a share of total MoAFS budget. Data show that the allocation of MoAFS budget to 50 agricultural extension has tended to increase over the whole study period, even if actual expenditures have tended to be lower than approved in recent years (Figure 3.17). Translating these figures as a percentage of the total agricultural budget reveals that the trend reached its peak early in the series in 2003/04 and 2004/05 and drastically declined thereafter (Figure 3.18). While at its peak, the percentage of actual MoAFS expenditures allocated to agricultural extension almost reached 15%, in 2011/12, it represented less than 5%. The trend for agricultural research has in part been due to the fact that a large percentage of the total agricultural budget is allocated to FISP, thereby leaving the rest of the other components of agricultural budget to share an increasingly smaller proportion. Figure 3.17: Trends in MoAFS budget allocation to extension, 2000/01-2011/12, current USD million 12 10 USD million 8 6 Approved 4 Actual 2 0 Source: AgPER from data provided by MoF and MoAFS Figure 3.18: Percentage of extension in MoAFS budget, 2000/01-2011/12 (in %) 25% 20% 15% Approved 10% Actual 5% 0% Source: AgPER from data provided by MoF and MoAFS 51 125. As a result of greater budget allocations, the incidence of Government extension has dramatically increased since 2005 but remains low at about 23%. The incidence of Government extension has increased among all rural population quintiles (Figure 3.19) and while extension use rates are very similar across regions, they are slightly higher among relatively well-off households (Figure 3.20). Overall, the use of extension services remains low at just below 23%. Figure 3.19: Proportion of agricultural households receiving extension by wealth, IHS2 and IHS3 (in %) 30 25 26 26 25 23 22 20 17 16 14 2005 15 13 13 13 10 2011 10 5 0 Malawi Poorest Q2 Q3 Q4 Richest Source: Benfica 2013 Figure 3.20: Proportion of agricultural households receiving extension by region, poverty status and gender, 2011 (in %) 25 23 23 23 24 23 22 20 20 20 15 10 5 0 Source: Benfica 2013 3.4 Policy Priorities or Suggestions for Reform 126. As analysed, Malawi agricultural policy orientations have produced mixed results over the past decade. On one hand public expenditure in agriculture was considerably increased to reach 52 about 19% of total expenditure and the launching of FISP in 2005/06 induced a strong maize production increase allowing the country to recover food self-sufficiency at national level. On the other hand, the country now finds itself somehow blocked in a situation in which the MoAFS has very little fiscal space. In addition agricultural spending is penalized by numerous inefficiencies as summarized below. i. FISP inefficiencies. FISP appears as a cumbersome targeting process that takes a heavy toll on MoAFS staff resources and eventually proves ineffective, aggravated by fraud, corruption and distortions. It should also be pointed out that the very strong signals the current FISP conveys in favour of heavy State intervention in the sector does not contribute to a conducive environment for private sector expansion in agriculture and agribusiness as called for by ASWAp and ERP. ii. Weak linkage between policy framework and budget planning. The issue is compounded with the high fragmentation of aid and the high proportion of off-Budget expenditures that entail limited oversight and ownership by the Government and high transaction costs. iii. Low efficiency of budget planning and implementation. Cumbersome procedures, low level of expenditure control, weak monitoring and evaluation, and low staff motivation limit both sector expenditure efficiency and equity. iv. The high level of centralization of agricultural policy making and implementation , with insufficient involvement of deconcentrated administrations and non-State actors. 127. In order to remedy these imbalances and inefficiencies and revive the sector’s capacity to produce and sustain robust growth, the following areas of reform are proposed: i. Improve technical efficiency. Different dimensions of the policy set for consideration are: - Rationalizing the allocation of sector expenditures envelope. Capital spending should be increased, and the level at which it is incurred (public infrastructure, infrastructure - such as irrigation schemes or storage facilities - handed over to beneficiaries, Government or non-Government support services) should be recorded through adequate budget coding. - Correcting current imbalances in agricultural public expenditures in favor of agriculture commercialization and market development (including rural finance), land and water sustainable management, technology research and development, and dissemination and livestock development. - Adopting a new budget presentation allowing better accounting and thus better planning and monitoring of both capital and recurrent spending and adequate provisions for operation and maintenance costs. - As part of the overall PMF strengthening, ensuring full rolling out of IFMIS to generate comprehensive and real-time budget execution data. - Streamlining procurement procedures, strengthening M&E at all levels. - Correcting the erosion of civil servants’ salaries combined with enhanced performance assessment mechanisms, and discontinuing the use of travel allowances as salary supplement. - Using the calendar year as the fiscal year could also be worth-considering. 53 ii. Operationalize the ASWAp investment framework for which only the apex oversight bodies are in place at the moment, in order to increase ownership and accountability and establish a stronger linkage between policy framework and budget planning. In particular, MoAFS organizational chart and budget should be adjusted to become consistent with ASWAp architecture; all DP financed activities should be brought into MoAFS Budget in order to facilitate strategic planning and increase MoAFS fiscal space; as fiduciary capacities increase, DPs should make greater use of both Government systems and common financing mechanisms in order to further increase fiscal space and reduce aid transaction costs; finally, financial resources should be constantly reallocated from unsuccessful initiatives to more promising ones with the objective of spending better rather than spending more; iii. Re-design FISP in order to serve productive farmers in a market-smart way and concomitantly, strengthen pro-poor safety nets(cash transfer, rural pensions, public works, etc.). There is now some consensus that targeting has not been effective and has generated fraud and distortions. Attempts at tightening targeting would most probably merely exacerbate further these issues. This calls for a shift of paradigm in reforming FISP with less emphasis put on targeting and avoiding commercial sales displacement and more on eliminating fraud, corruption and distortions, and promoting private sector participation. The re-designed FISP would have to be accompanied by enhanced social safety nets as the most vulnerable may not get any longer the cash transfer they were getting through the reselling of their fertilizer allocation under the current system; iv. Foster the decentralization process that will be revived in 2014 with the election of the District Assemblies, through a greater involvement of District administration, local communities, farmers’ organizations, NGOs and private operators. Matching grants to finance demand-driven initiatives by local communities or local promoters with the technical support of the deconcentrated administration have proved powerful tools to support decentralization in other countries (e.g. Burkina Faso). 128. As a concluding remark, let us recall that to be a factor of change and progress, commitment by all stakeholders must go beyond intentions and translate into changes in processes and organizational arrangements. Most of the recommendations listed above are not new and were already formulated in previous studies and documents by the GoM itself (e.g. MPRS) more than 10 years ago. One critical recommendation is therefore to start implementing and operationalizing the agreed recommendations. A number of Malawi strategies had very valid (and still so) objectives based on solid analysis but failed to induce transformation and achieve progress due to lack of operationalization. 129. There is a tendency to set ambitious objectives and overlook the practicalities to attain them, carrying on business as usual and neglecting to adapt to changes and modify organizational arrangements. This is clearly the case for MoAFS which has elaborated and adopted with all stakeholders a sound and relevant investment framework. However, operationalization is lagging also because the Ministry’s own internal organization and budget is not aligned to the ASWAp. To foster donor alignment, MoAFS should demonstrate more ownership and accountability towards its own investment framework. 130. Agriculture will play its part in reaching the objectives of MGDS II only if the sector as a whole, MoAFS, donors, farmers and private sector, will have the resolve to change their approach and depart from a system characterized by lack of accountability and ownership to a real ASWAp whereby MoAFS, with the support of donors and the full involvement of non-State actors, would recover its central policy-making, regulatory, M&E and public good provision role. 54 CHAPTER 4. TRANSPORT 4.1 Introduction 131. The transport sector in Malawi is regarded as a key contributor to the Government’s Malawi Growth and Development Strategy, the Economic Recovery Plan and the National Export Strategy due to the importance of the sector in servicing trade and development. This chapter evaluates the efficiency, equity, and sustainability of public spending on the road, rail, air and water transport modes, with a particular focus on the infrastructure elements of the roads sub-sector. Transport services, managed primarily by the private sector, do not fall within the scope of the assessment. Detailed review of the sector can be found in the consultant’s Final Report38. 4.2 Assessment of Transport Expenditures 4.2.1 Level of Expenditures Policy Environment 132. During the period covered by the current PER, the Malawi Growth and Development Strategy (MGDS) 2006-2011 has been the overarching document for sector management and development, within the framework of the 2004 National Transport Policy which is under review by Government. Important transport sector documents were prepared later, notably the 2010-2020 Road Sector Programme (RSP), the Transport Sector Investment Programme (TSIP) of June 2012, and the Roads Authority’s (RA) Five Year Strategic and Business Plan 2011-2016. The consequences of the 2011 economic crisis and contractual over-commitments have severely constrained the implementation of a substantial part of these plans, which appear to have set overly ambitious output targets in relation to available funding and implementation capacity. 133. The Economic Recovery Plan (ERP) 2012 presents Government’s latest plans for the medium term and features transport infrastructure as one of five key sectors . The ERP prioritizes activities from the MGDS II, which defines overall goals and targets by sector for the period 2011- 2016. The ERP has introduced new and ambitious priorities given the prevailing economic situation, with a budget of over US$ 2 billion planned for the transport sector to 2020. In comparison, MGDS II allocated MK 46 billion (in 2005 constant prices, US$ 341 million equivalent), varying between MK 8.9 and 9.3 billion per year for transport infrastructure development, which represented almost 54% of the total envisaged for the six priority areas. Status of Transport Sector Expenditure 2006-2012 134. Table 4.1 shows Government’s Recurrent and Development Expenditure, expressed in real prices for 2007 (Base Year). It is characterized by a significant increase in GOM’s Development Budget (57% to 72%) for the transport sector and peaks at 3.8% of GDP in FYs 2009/10, decreasing to 2.1% of GDP in FYs 2011/12. Details on the Development Partners’ (DPs) contributions to the sector are given in Table 4.3. 38 The full report was shared with the MoTPW and is on file with the World Bank Transport Team. 55 Table 4.1: GOM Transport Sector (TS) Expenditures - MK million: 2007 Base Year Category 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Recurrent 3,387 3,657 5,692 7,819 6,841 3,892 Recurrent (% of Total) 43% 47% 47% 35% 32% 28% Development 4,555 4,186 6,330 14,807 14,562 10,175 Development (% of Total) 57% 53% 53% 65% 68% 72% Total TS GOM 7,941 7,843 12,022 22,626 21,403 14,068 GOM TS Expenditure as % of 1.7% 1.5% 2.2% 3.8% 3.3% 2.1% GDP GDP real growth % (IMF) 9.5% 8.3% 9.0% 6.5% 4.3% 1.9% Source: MOF/IFMIS (Itemized Expenditure Analysis) & RA Audited Annual Financial Statements 135. Most of the Recurrent Expenditure relates to the roads sub-sector (RA/RFA and road maintenance expenses) while the share of total Transport Sector Recurrent Expenditure as a % of total GOM Recurrent Expenditure initially increased from a 3% level to 4-5%, then decreased to 2% (Annex A6). 136. An overall picture of Recurrent and Development Expenditure is presented in Table 4.2 with the trends shown in Figure 4.1. The dominance of roads sub-sector expenditure can clearly be seen, although the budget share for other sub-sectors has risen to around 6%. Development Partners continue to contribute a substantial share, an average of 30% of total sector expenditure, and 39% of expenditure on ‘Development’ over the full period FY 2006-2012. 56 Table 4.2: Total Sector Expenditure for GOM and DPs (actual) - MK millions: 2007 Base Year Category 2006/07 2007/8 2008/09 2009/10 2010/11 2011/12 Roads Routine Maintenance 1,969 981 2,042 2,469 2,722 1,338 Periodic Maintenance * 533 994 527 3,798 3,241 1,591 Rehabilitation 198 1,615 3,209 2,467 3,040 2,862 Upgrading/Construction 4,199 6,715 6,629 16,608 15,550 14,513 Sub-total 6,899 10,305 12,407 25,342 24,552 20,304 Other Transport Modes Aviation (estimate) 35 125 537 1,784 1,672 639 Inland Water (incl. Shire) 16 5 7 8 101 173 Railways (estimate) - - - - 32 61 Sub-total 52 130 545 1,793 1,805 873 Grand Total: Transport Sector 6,950 10,435 12,952 27,135 26,358 21,177 Of which DP's contribution (estimate) DPs – Roads 1,585 5,421 4,242 4,846 4,567 7,896 DPs – Aviation (estimate) - - 387 1,359 1,265 321 Total from DPs (estimate) 1,585 5,421 4,629 6,205 5,832 8,217 DP’s Share (% of total) 23% 52% 36% 23% 22% 39% * It should be noted that the majority of periodic maintenance work is a mixture of overlaying or resealing with rehabilitation or partial reconstruction works. Source: MOF/IFMIS (Itemized Expenditure Analysis) & RA Audited Annual Financial Statements Figure 4.1: Sector Expenditures (actual) for GOM and DPs - MK millions: 2007 Base Year 25,000 20,000 15,000 10,000 5,000 - 2006/07 2007/8 2008/09 2009/10 2010/11 2011/12 Road Maint. Road Rehab. & Upgr. Other Sub-sectors DPs Total Roads Sub-sector 137. The ‘Approved Budgets’ and associated ‘Actual Expenditures’ for the roads sub -sector show significant differences in practically every year of the PER period. The Government appears to have been too optimistic about the capacity to disburse Development Partners’ contributions. Their 57 share in the Development Budget (Roads) was an average of 68% over the period from 2006 to 2012, while the actual (disbursed) contribution has been 37%. 138. GOM’s intention to increase its share of funding for periodic maintenance rehabilitation and backlog maintenance is visible at the planning stage (from almost zero in 2006/07 rising to MK 3 billion in 2011/12), with an anticipated support from DPs at about MK 4-5 billion/year. In the planning of upgrading and construction, there has been a trend for a higher growth in both anticipated DP and GOM funding, from a low level of MK 3 billion in 2006/07 to more than MK 20 billion in 2011/12, compared with MK 7 billion for backlog maintenance. 139. Comparison of budgeted and actual expenditures in real 2007 prices for the sub-sector shows that: x Over the period 2006 to 2012, GOM ‘Development’ expenditure has been MK 49.0 billion, exceeding what was planned in the Budget by 64%, while MK 22.2 billion from the Roads Fund went to maintenance (MK 11.5 billion for routine maintenance and MK 10.7 billion for periodic maintenance and other works), and x The DPs’ contribution of MK 28.6 billion disbursed over the period represented only 43% of what was anticipated by GOM in the Budget. 140. Whereas ‘Recurrent’ expenditure for routine maintenance is determined by income from the Roads Fund, any additional funding required for periodic and back-log maintenance needs to be allocated under the ‘Development’ budget. Review of ‘Development’ expenditure shows that ‘Periodic Maintenance combined with Rehabilitation’ had seriously under-performed with 49% of the budget expenditure, whilst 94% of the budget was spent for ‘Upgrading/construction’. Overall, total roads ‘Development’ expenditure was 81% of the planned budget. Table 4.3: Expenditure on Road Maintenance, Rehabilitation and Upgrading/ Construction as a Proportion of GDP FY 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Routine Maintenance (Roads 0.4% 0.2% 0.4% 0.4% 0.4% 0.2% Fund) Periodic Maintenance (RF) * 0.1% 0.2% 0.1% 0.6% 0.5% 0.2% Periodic Mtce mix 0.0% 0.3% 0.6% 0.4% 0.5% 0.4% w/Rehabilitation (DEVELOPMENT) Upgrading/Construction (DEV) 0.9% 1.3% 1.2% 2.8% 2.4% 2.2% Sector Total 1.5% 2.0% 2.2% 4.2% 3.8% 3.0% Real GDP (MK millions) base 466,282 510,539 553,112 603,090 642,492 670,002 2007 Source: Roads Authority Audited Annual Financial Statements 141. Table 4.3 shows that periodic maintenance expenditure has been very low at 0.1-0.2% of GDP, except in FYs 2009/11 when it was 0.5-0.6% of GDP. When combining Periodic Maintenance from recurrent/Roads Fund and rehabilitation from development funding, the share fluctuates between 0.5% and 1% of GDP, and between 22% and 30% of total roads expenditure, excluding 58 FYs 2006/07. Road upgrading (paving of unpaved roads) and new construction claimed 2.8% of GDP in fiscal year 2009/10 but had reduced to 2.2% by 2011/12. Rail Sub-sector 142. Railway operations were concessioned to Central and East Africa Railways (CEAR) in 1999 for a 20-year period. Since then there has been no investment in rail infrastructure until FYs 2010/11. Between 2010 and 2013, a total of MK 270 million was budgeted for emergency repairs, while expenditure in FYs 2010/12 is estimated at MK 120-125 million (both in ‘nominal’ cost terms), with the balance to be spent later on outstanding contract obligations. 143. Development partners have supported feasibility studies on railway infrastructure: the EU funded a study covering the rehabilitation of the complete network, and JICA has financed a study on the development of the SENA corridor. The major foreign investment in the Moatize-Nkaya- Nacala corridor estimated at US$ 986 million for works in Malawi financed by VALE is to be disbursed between 2012 and 2016. Inland Water Transport Sub-sector 144. The Marine Department is subdivided in six Cost Centres, and has three small regional offices. Table 4.4 presents the most recently available of approved or revised budget allocations for the period 2006-2012 (in 2007 real prices), for recurrent and development categories. Table 4.4: IWT Recurrent and Development Budget Allocations - MK millions: 2007 Base Year Category 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Marine Department – Recurrent 018 - Marine Headquarters 25.3 26.0 14.7 20.4 n/a 30.5 019 - Marine Training College 16.5 15.0 13.8 17.8 n/a 21.3 020 - Port Management 24.2 23.0 26.7 28.0 n.a 29.7 021/22/23 - Marine North/Centre/South 3.3 4.0 12.0 - n/a 9.1 034 - Development Corridors - - - 11.9 n/a - Total 69.3 68.0 67.2 78.1 n/a 90.6 Marine Department - Development 018 - Marine Headquarters 55.0 - 8.3 - 16.6 99.0 019 - Marine Training College - 5.0 7.4 - - - 020 - Port Management 30.8 - - 12.7 - - Total 85.8 5.0 15.6 12.7 16.6 99.0 Source: MOF/IFMIS – Vote 400 (MTPW Recurrent and Development Budget) 145. Most development expenditure was on navigation aids and Lake Malawi port infrastructure (Nkhata Bay jetty) with funding registered under Marine HQ, and a small amount on the Marine Training College. Furthermore, MK 217 million were spent off-Marine HQ budget on the Shire-Zambezi Waterway Feasibility Study to guide the extension of the Lake Malawi transport potential. 59 146. Costs for the Build-Operate-Transfer-implemented Nsanje World Inland Port are also believed to be ‘off-budget’, presumably pre-financed by the contractor MOTA-Engil. Recovery of the investment costs could be a challenge for the sector, and the Ports Management Concession (granted to a MOTA-Engil subsidiary) might offer opportunities to offset this public debt. Aviation Sub-sector 147. The Department of Civil Aviation (DCA) comprises five ‘Cost Centres’ (# 9 -13) under Vote 400. The approved or revised budget allocations over the period 2006-2012 for recurrent and development budgets are presented in Table 4.5. Table 4.5: Civil Aviation Budget Allocations - MK million: 2007 Base Year (real) Category 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 DCA – Recurrent 009 - Civil Aviation Headquarters 64.9 64.0 45.1 87.5 N/A 64.0 010 - Kamuzu International Airport 79.2 71.0 96.6 82.4 N/A 80.0 (KIA) 011 - Chileka International Airport 56.1 56.0 57.1 81.5 N/A 153.8 012 - Mzuzu, Karonga and Minor 18.7 17.0 16.6 22.1 N/A 27.4 Aerodromes 013 - Civil Aviation Training School 12.1 10.0 10.1 14.4 N/A 17.5 Total – Recurrent 231.0 219.0 225.5 287.9 N/A 342.7 DCA – Development 009 - Civil Aviation Headquarters - 74.0 147.3 254.8 N/A 198.0 010 - Kamuzu International Airport 920.6 574.0 508.0 97.7 N/A - 011 - Chileka International Airport - 25.0 - 22.9 N/A 243.70 012 - Mzuzu, Karonga and Minor - - - 29.7 N/A - Aerodromes 013 - Civil Aviation Training School 5.5 5.0 5.5 5.1 N/A - Total – Development 926.1 678.0 660.8 410.2 N/A 441.7 Source: MOF/IFMIS – Vote 400 (MTPW) 148. Total expenditure from the GOM Development Part II budget for ten projects undertaken during the period 2006-2012 was about MK 1.5 billion. Development partners made substantial contributions (MK 3.3 billion) to projects for airport navigation equipment and the rehabilitation of the runway at KIA. Recurrent Expenditure of MTPW Cost Centres 149. Recurrent budget allocations and expenditures for MTPW and its agencies can be broken down in ‘Cost Items’ and ‘Cost Centres’. The Salaries/Allowances cost item constitutes around 40% of total Recurrent Expenditure: the cost items for internal and external travel combined with motor vehicle running costs also take a substantial allocation share of 25%. With the proliferation of internet services (tele-conferencing) there are cheap alternatives for long distance, time consuming travel. This section of expenditure may be moved in the future to cost items such as staff training, 60 educational/ technical services, and feasibility studies, which all could raise the professionalization level of the Ministerial agencies. 150. Among the Cost Centres of the MTPW, Civil Aviation and Road Traffic Departments are leading in recurrent expenditure (close to the MK 400 million/year level), followed by Headquarters (close to the MK 200 million/year level) and the Marine Department (at half the latter expenditure level). The Road Agencies (RA and RFA) spend the larger part (56%) of total TS Recurrent Expenditure, albeit financed from the Roads Fund income. 151. There are several institutions and agencies reporting to MTPW that are not defined as separate cost centres in the Integrated Financial Management Information System (IFMIS), such as the Railways Division, the Plant and Vehicle Hire Organisation (PVHO) and the Central Materials Laboratory (CML). Transparency and accountability of public expenditure would certainly be improved if these ‘producing’ entities39 could be easily identified as cost (and revenue) centres in budgets and expenditure. Further analysis of cost item and cost centre recurrent expenditure data in combination with staffing numbers (distinguishing between professional and support staff) is recommended to help sub-sector management identify changes to achieve output- based tasks with greater efficiency. 4.2.2 Allocative and Technical Efficiency 152. During the course of the PER, various areas have been identified in which improvements should be considered to increase the efficiency of the use of the scarce resources available to the sector. 153. Prioritization: Whilst primarily an issue for the road sector - where prioritization is discussed in more detail - it is essential that proposed investments in the different sub-sectors, be they publicly or privately financed, are evaluated and selected in relation to their viability and the contribution they might make to reductions in transport costs, either on their own or as part of a multi-modal approach. The preparation of a strategy for the transport sector would assist in this regard. Roads Sub-sector 154. Road Sector Capacity: As the main channel for transport sector expenditure, the capacity of the government agencies that plan and manage the programmes and the private sector that executes them is fundamental. Planning, procurement, technical skills and contract management need to be strengthened in the implementing agencies40. Road works contractors and design and construction supervision consultants need to be held accountable for the quality of the works through proactive professional guidance and control from the contracting agencies. A comprehensive approach will be required to address this complex challenge. 155. Road Network Size and Appropriate Levels of Funding: Malawi’s classified road network asset value is estimated at MK 609 billion or US$ 3.69 billion, almost 70% of the GDP (MK 880 billion in FYs 2011/12). Approximately 94% of this value is vested in the paved road network. This compares well with findings of a regional study41 which found that Malawi’s current road asset value represented 60% of GDP. It should be noted that the average network value for the region is 39 Business plans for PVHO and CML are to be updated. 40 The absence of a current sector capacity analysis is to be addressed under support from development partners. 41 Comparative regional data taken from the Africa Infrastructure Country Diagnostic (AICD), The Burden of Maintenance: Roads in Sub-Saharan Africa, Background Paper 14, June 2008. 61 25% of GDP: Malawi’s road network is therefore large by regional standards in relation to the country’s GDP. 156. The annual maintenance input required to sustain the road network should be at least 3.3%42 of the network asset value, based on a lifetime cycle of 30 years. However, the average annual amount Malawi spends on maintenance and rehabilitation only represents 1.5% of the asset value. 157. The 3.3% benchmark would imply the need for at least US$ 110 million/year for the paved network of more than 4,300 km. An alternative estimate based on typical unit rates for maintenance leads to an annual requirement of about US$ 60 million/year for the full paved network43, probably under-estimating the backlog element. It has been estimated that the Roads Fund could contribute US$ 25 million/year for maintenance of the paved network, leaving US$ 35- 40 million/year from the total US$ 55 million/year ‘Development’ allocation for all modes of transport. This leads to the conclusion that an assumed allocation of 5.5%44 of GDP for the transport sector can barely sustain the existing paved road network and will contribute too little for unpaved roads which would require a benchmark amount of US$ 45 million/year for the network of 11,139 km. Another 1-1.5% share of GDP (i.e. an average US$ 20-30 million/year) would be needed to sustain unpaved roads, railways, IWT and aviation. There are also proposals to expand the classified network by adding over 9,000 km of unpaved district and community roads, putting even greater pressure on the annual budget requirement. 158. It is clear that the issue of optimal road network size should be revisited given the substantial finance required to sustain the current network and the need to provide access to rural areas . According to AICD data, network density per unit area is high by regional standards, but well below average per unit of population; tertiary and unclassified road density is also low, which indicates imbalances between the different classes of road. Expansion of the network will serve little purpose in the longer term unless mechanisms to fund adequate levels of maintenance are put in place. 159. Project Prioritization: GOM, MTPW and RA have established procedures for the assessment of road project viability, applying a requirement for a minimum IRR of 12%. All road projects must fulfil this criterion to be included in the annual PSIP and incorporated in the Development Budget. A review of the feasibility study for the Ntcheu - Tsangano - Neno - Mwanza road showed the traffic counts identified relatively low levels of traffic on the road while high levels of generated traffic have been assumed. Further investigation of the assumptions that were made would be instructive. Though no further reviews of feasibility studies for other projects were conducted, experience indicates that the design and construction standards adopted are high in relation to the actual levels of traffic on the roads. 160. There is a high level of expectation from politicians and the general public that roads should be paved. Due to the generally poor condition of unpaved roads, they are not seen as a viable means of providing access. It is therefore important that more effective maintenance systems are introduced. Political pressures can play an important role in the project selection process, meaning that attention turns to demonstrating the viability of a targeted project rather than focusing on its best value (rate of return) in relation to other available options. The RSP (2010) was prepared using such a programmatic process where many options were compared. It is important that it is updated 42 Benchmark for annual maintenance requirement as a proportion of replacement value based on asset life, proposed by team. 43 Assumes 10% of paved network receives periodic maintenance intervention every year. 44 4% from GOM and 1.5% contributed by DPs - Ref. Table 4.1. 62 using a similar wide stakeholder consultative process and that in the case of major investments should be based upon the results of comprehensive economic analysis. 161. An important recommendation is therefore to increase the effectiveness of project prioritization and targeting by re-establishing and enforcing economic and social criteria. This will allow project prioritization to be applied in Cost-Benefit Analysis 45 and Multi-Criteria Analysis in TSIP, RSP and PSIP preparation exercises. The Roads Authority and the consultants employed to carry out the economic analysis in the design process need to increase their capacity. MTPW staff should be trained to monitor the quality of economic analyses and feasibility studies for projects. To allow some room for political input in the selection process - removing it entirely may not be realistic - limits could be set for such ‘discriminatory’ spending. Reductions in transport costs are used as the prime justification for transport infrastructure investment although many other factors such as the competitive environment, border procedures and non-tariff barriers contribute substantially to costs. The effect of each of these factors needs to be assessed and activities focused on where most impact will be made. The Bank is discussing support for the preparation of a sector strategy and action plan with MTPW which will cover the issues of cost, demand and future infrastructure provision. 162. Construction Costs and Choice of Road Design Standards: Government has financed a series of contracts for road upgrading and rehabilitation: details are given in Table 4.6. Five of the projects are for the upgrading of Secondary and Tertiary Roads at unit costs ranging from US$ 697,000 to US$ 1,450,000 per kilometre, a similar range of costs as for the upgrading of Main Roads (US$ 876,000 to US$ 1,279,000 per km). Table 4.6: Wholly Government Funded Road Projects Road Project Details Km Initial Revised Unit Cost Remarks No. Contract Contract (US$ Sum Sum ‘000/Km) (MK Million) (MK Million) S150 Goliati-Chiperoni Upgrade from earth: (extension of Malowa- 21.1 3,637 4,651 1,450 Completed Goliati) S125 Upgrade from Bunda-Mitundu 9 795 1,021 746 gravel: Completed T309 Ekwendeni-Ezondweni- Upgrade from earth: 25 1,890 4,155 1,007 Mtwalo Completed S107 Mzimba-Mzalangwe 62 7,127 7,127 697 Upgrade from earth M7/ Lumbadzi-Dowa-Chezi Upgrade from earth - 38 3,131 4,082 707 M16 (12 km completed) 18 km S145/ Chiringa-Miseu Folo- S147/ Chiradzulu+Migowi- 90 6,448 14,952 1,007 Upgrade from gravel T411 Phalombe+Nguludi- PIM-Chiradzulu M1 Upgrade from earth: Nsanje-Bangula 48 7,415 10,129 1,279 Completed 45 WB study “Transport Prices and Costs: The need to revisit Donors’ policies in transport in Africa” demonstrates that the internal rate of return of road rehabilitation/upgrading is generally not positive in cases of low traffic, low truck utilization and old fleet. 63 Road Project Details Km Initial Revised Unit Cost Remarks No. Contract Contract (US$ Sum Sum ‘000/Km) (MK Million) (MK Million) M18 Msulira-Nkhotakota Reconstruction of 33 2,417 4,085 396 (Game Reserve Section) failed road - Lilongwe & Mzimba 1.82 356 407 1,355 Completed streets M1 Period Maintenance Dedza-Nsipe (with EU) 83.5 3,891 2,557 186 & Rehabilitation M1 Ngabu-Bangula (continuation of EU 35 9,601 9,601 876 Upgrade from earth project) Source: RA – Quarterly Construction Progress Report up to end September & end December 2012 163. The main reason for the high upgrading cost is the choice of the highest design standard (Class I Bitumen) for most of these roads, irrespective of the road class. The majority of these roads carry relatively low levels of traffic. An alternative solution would have been to adopt a Low Volume Sealed Road (LVSR) standard for these roads, with an estimated unit cost of between US$ 250,000 and US$ 400,000 per km. This method has been adopted for the Mchinji-Kawere Road (S118), a Secondary Road supported by the EU. A considerable amount of research has been done on this form of construction, and a draft design manual has been prepared for use in Malawi. Adopting this LVSR standard, the resulting 60% cost saving could have been allocated to, for example, backlog maintenance of the economically important core paved road network. 164. A programme of works on 360 km of paved road, predominantly a combination of periodic maintenance and rehabilitation, was carried out between February 2008 and November 2012 with the support of the EU. Most of the works were typical backlog maintenance, a mixture of resealing and rehabilitation, undertaken within a cost range of US$ 140-190,000 per kilometre. The Government funded Dedza-Nsipe road also fell within this range. This demonstrates that it is possible to complete these types of contracts at a reasonable price, albeit with significant time delays and some cost increases, provided that there is a professional and well-controlled project management regime in place. 165. Comparisons with sample unit costs for road works in countries in the region (Zambia, Mozambique, Uganda) demonstrate that costs in Malawi are within the expected range. In each of these countries, costs are significantly higher than they were 5-7 years ago, and with large variations. Measures to encourage greater competition for medium and large contracts will help reduce construction costs. Rail Sub-sector 166. The sum of MK 270 million was allocated for emergency repair works over the last three years aimed at maintaining a minimum level of service on the network . The works were split into five contracts, each with a planned implementation period of three months. Implementation suffered long delays and by the end of FYs 2012/13, the actual expenditure is expected to reach MK 250 million. Meanwhile, priorities were changed in response to flood damage in January 2013. Given the difficulties experienced in concluding these emergency repair contracts, MTPW is eager to find financing partners to undertake the rehabilitation and upgrading works defined in the EU- funded study of 2009 in order to reduce the line vulnerability. 64 Inland Water Transport Sub-sector 167. During the period 2006-2012, the Marine Department planned eight projects for an estimated value of MK 4.5 billion, received budget approval for MK 266 million with actual expenditure of MK 168 million for three projects. Only one was substantial (Nkhata Bay jetty), a second (navigational aids on Lake Malawi) was seriously vandalized after completion. Aviation Sub-sector 168. The implementation of the ten projects undertaken by the Department of Civil Aviation during the period 2006-2012 experienced inefficiencies such as delays in procurement and funding shortages. Actual GOM expenditures were 60% of the total Approved Budget which represented approximately 20% of the Total Estimated Cost of the projects, with the balance being funded by DPs. Projects that have suffered most were the acquisition and modernization of fire fighting equipment and upgrading of secondary aerodromes. The more successful projects were the acquisition of airport navigation equipment supported by JICA for US$ 7.1 million, and the Kamuzu runway rehabilitation financed from BADEA and OFID loans of US$ 17.5 million, with a local expenditure component of MK 229 million. Some MK 260 million was spent on the rehabilitation of Chileka terminal buildings (works still ongoing). Several of the ongoing projects are aimed at complying with the ICAO regulations, which is essential if the sub-sector is to contribute to tourism and international business travel. 4.2.3 Intra-sectoral equity/Benefit Incidence Analysis 169. The relationship between road access and welfare has been investigated based on household and community welfare data from the 2011 Third Integrated Household Survey (IHS3) . The ‘incidence analysis’ used information from the 768 communities covered by the IHS3 Survey and combined community level data with household level data. Community wealth levels were calculated as average Household Expenditure per capita, grouped in five ‘Community Wealth Quintiles’ (from 1= Poorest 20% of communities, to 5= Richest 20% of communities). 170. Information from the community surveys regarding type of roads; distance to closest tar/asphalt roads; presence of all-year roads; accessibility by bus, lorry; and presence of public/private transportation was then used to establish the relative access characteristics of each income group. Most of the results confirm in statistical terms that rural access and connectivity remain a great concern, particularly for the poorer rural households. 171. Figure 4.2 demonstrates that there is a clear direct relationship between level of wealth and distance to roads, with the poorer communities located further away from tar/ asphalt roads. As an international benchmark, acceptable rural access is often defined as when the average distance of a community to the nearest all-weather road does not exceed 2 km46. Even the richest rural communities appear to be at a median distance of 5 km from a paved road. The situation in urban areas is much better: the median distance for poorest communities is 2 km. 46 This survey measured distances to a ‘paved’ road rather than to an ‘all-weather’ road, as used by the benchmark. The median distances could be expected to reduce significantly if distances to all-weather roads were to be used. 65 Figure 4.2: Median Distance from Community to Nearest Tar/Asphalt Road (Km) Malawi, Rural and Urban, by Community Wealth 30.0 25.0 Median Distance (Km) 20.0 15.0 10.0 5.0 0.0 Malawi Urban Rural Poorest Communities 22.0 2.0 24.0 Q2 15.0 1.5 14.0 Q3 9.5 1.0 12.0 Q4 7.0 0.1 8.0 Richest Communities 0.5 0.1 5.0 Source: IHS3 Survey data 172. The analysis also shows differences between the poorer and richer households for perceived level of transport service by minibus (passenger travel) and truck (goods transport), indicating that goods transporters generally make greater efforts to get access than public transport. This seems to illustrate that there must be (i) a certain level of wealth and community size in combination with (ii) characteristics of transitability and distance to a paved road to make it profitable for interested minibus owners to operate on the route. The IHS3 Survey offers a wealth of data that could be used for further analysis of rural roads access issues at the disaggregated level of specific communities affected by the various rural Public Works Programmes or for the wider planning of rural roads. The survey also collected data on vehicle ownership (car, motorcycle, bicycle) at the household level. 4.3 Transport Sector Performance and Key Issues Regulation of the Transport Sector 173. With the growing involvement of the private sector in the supply of services in the sector, either through concessions or as transport operators, the need for effective regulation is increasingly important. The task of ensuring independent oversight of compliance with regulations, operating procedures, safety and security in each of the four transport modes could be implemented by: (i) a single Transport Sector Regulator overseeing three or four modes (excluding or including the Road Traffic Department), or (ii) four future regulators (Road Traffic Authority, Civil Aviation Authority, rail regulator and IWT regulator) each dealing with largely similar challenges (PPP/Concession management, inspection services, licensing, etc.). Clarification of the mandates of and relationship between the PPP Commission and MTPW on concession management is also needed. 4.3.1 Roads Sub-sector Network Performance 174. The public road network coverage is 15,451 km (classified) of which about 28% are paved, the rest being unpaved and the majority earth surfaced. The paved proportion of the classified 66 network (28%) is above the average for the region which is 20%. A road reclassification study carried out in 2006 identified a further 9,478 km of currently undesignated roads that serve rural communities, and are instrumental for rural access and connectivity with the higher level network. Annex - 4.2 presents the classified public road network composition at the end of FYs 2011/12. 175. The reported condition of the paved road network - 96% in a good and fair condition in 2011 - compares well with the regional average of 79%47 and is an increase from 84% in 2007. This improvement in condition reflects the substantial investments made in the sector. However the latest road condition survey might not be representative since it was carried out between October and December 2011 on 2,520 km of the paved road network, representing 58% of the total paved network. This percentage may be lower today. The figure of 17% of the paved network in ‘fair’ condition is an indicator of the asset preservation needs in the immediate future. 176. The situation on the unpaved network, important for sustaining agricultural production, is less encouraging. Its condition has fallen from 83% in good and fair condition in 2007 to 65% in 2010, while RA reported that ‘visual impressions seem to indicate an overall negative trend’ for the period 2010-11. Regular, accurate road condition surveys should be carried out. 177. Malawi’s motor vehicle fleet has grown significantly in the period 2006-2010, but remains small relative to the size of the network. The road user charges generated by this vehicle fleet are insufficient to sustain the road network. The increasing scarcity of foreign exchange since 2010 seems to have had a direct effect on the importation of new and second-hand motor vehicles, slowing down the growth of the vehicle fleet, at least temporarily. 178. Average Daily Traffic volumes are low, with levels exceeding 1,000 motor vehicles/day found on about 10% (1,500 km) of the total classified road length, according to the latest extensive traffic counts from 2010. The largest ADT volumes were found on urban roads in Lilongwe and Blantyre. Performance Against Targets 179. When performance is assessed against planned targets the level of achievement of the dominant roads sub-sector has shown a disappointing performance in terms of: (i) a reducing level of achievement of physical outputs over recent years; (ii) under-funding of maintenance (in particular periodic maintenance) of the core road network, and (iii) questionable choices in road project prioritization. The low performance on the first two points has been caused primarily by the difficult economic situation which has reduced fuel sales and revenue transfers to the Roads Fund (RF), and the shortages of foreign exchange and fuel which brought many contracts to a halt. Point three relates to the focus on paved road upgrading: more resources could have been allocated to maintenance and the unpaved network. 180. In the roads sub-sector, routine road maintenance has performed well, but most periodic maintenance and rehabilitation works remained below planned outputs. Targets for the latter were found to be unrealistic given the evolving economic situation and capacity constraints, and were not updated on year-by-year basis. 47 Comparative regional data taken from the Africa Infrastructure Country Diagnostic (AICD), The Burden of Maintenance: Roads in Sub-Saharan Africa, Background Paper 14, June 2008. 67 Table 4.7: Planned and Achieved Physical Road Works Outputs - km Road Works Activity 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Total PLANNED Routine Mtce 3,963 3,881 4,957 11,000 11,000 10,969 45,770 (RM) TOTAL RM Paved Planned n/a n/a 3,500 3,580 3,598 3,692 RM Paved Achieved n/a n/a 3,083 4,073 2,178 % Achieved n/a n/a 86% 113% 59% RM Unpaved Planned n/a n/a 1,457 7,420 7,402 7277 RM Unpaved Achieved n/a n/a 4,957 3,577 9,303 % Achieved n/a n/a 67% 48% 128% ACHIEVED Routine Mtce 4,034 3,200 4,077 8,040 7,650 11,481 38,482 TOTAL % Achieved 102% 82% 82% 73% 70% 105% 84% PLANNED Periodic Mtce 24 40 102 126 209 277 778 Paved ACHIEVED Periodic 18 20 67 110 67 52 334 Mtce Paved % Achieved 75% 50% 66% 87% 32% 19% 43% PLANNED Rehabilitation 53 164 75 186 231 259 968 Paved ACHIEVED 38 85 67 65 124 2 381 Rehabilitation Paved % Achieved 72% 52% 89% 35% 54% 1% 39% Planned & Achieved Rehabilitation Unpaved - Negligible (no separate records) PLANNED Upgrading/ 7 150 180 317 50 - 704 NC TOTAL ACHIEVED Upgrading/ 14 102 120 174 49.6 - 459.6 NC TOTAL % Achieved 200% 68% 67% 55% 99% - 65% Source: RA reports to Joint Transport Sector Review meetings 181. Table 4.7 shows an average of 84% in achieving targets (a range of 70% to 105% over the PER period) for routine maintenance, but an average achievement of only 39% - 43% for periodic maintenance and rehabilitation, with a significant decrease in recent years. Inferences on the type of activities undertaken (important in determining the effectiveness and efficiency) or the quality of the work may be drawn from the road condition data. In FYs 2011/12 grass cutting was the only routine maintenance activity that was carried out on the paved network due to the low level of Roads Fund income. Reshaping of unpaved roads, however, showed a very good performance: over 9,300 km, a 128% achievement in 2011/12. Although upgrading of unpaved roads showed an average of 65% in achieving targets, many of the on-going projects are at risk of further delays and cost increases. 68 Accumulation of Payment Arrears 182. The greatest and most pressing challenge for the roads sub-sector is the high level of contractual arrears48, estimated at between MK 22-28 billion or more than US$ 60 million at the end of FY 2012/13. The ‘Approved Budget’ (Part II Development) has in recent years been insufficient to cover contractual obligations made by the previous regime in FY2008/09. The most serious consequence is that presently more than half of the annual budget needs to be allocated to pay these arrears, curtailing implementation of the periodic maintenance and rehabilitation work that is required. At the same time, the non-completed and delayed projects have suffered delays due to foreign exchange and fuel shortages, triggering interest charges and price increase claims which have further added to the arrears. The situation is illustrated in Table 4.8. Table 4.8: Anticipated Budget and Actual Expenditure – MK million: 2007 Base Year (real) Gap between Anticipated & Actual Funding 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 All road projects (Development) Anticipated DPs funding (Part I in Dev 6,594 11,693 8,835 13,041 13,991 12,685 Budget) Actual DPs funds disbursed 1,585 5,421 4,242 4,846 4,567 7,896 Approved GOM Budget (Dev Part II) 542 871 6,548 4,704 7,149 9,254 Actual GOM Expenditure 2,811 2,909 5,597 14,229 14,022 9,479 Annual (Potential) Build Up of GOM Debts 2,269 2,038 -952 9,525 6,873 225 Accrued Contract Expenses (RA) 423 811 1,021 4,441 9,070 8,370 Total Payables (Aud. Ann. Fin. Statement) 559 916 1,384 6,531 10,678 9,516 Source: Consultant’s estimate from Roads Authority data 183. In order to limit the impact of the arrears it is recommended that: (i) as many of the on-going contracts as possible are terminated to reduce commitments, and (ii) a restructuring of the remaining contract payments over the next 2-3 years is negotiated, unless an alternative source of funding to cover the arrears is identified. This could reduce the arrears and contract commitments by 50%, burdening the next 2-3 years’ budget with a reduced ‘a-priori expenditure’ of MK 5 billion. Restoration of Roads Fund Income Levels 184. The second great challenge facing the road sector is the restoration of the Roads Fund (RF) income stream to a level at which it can make a significant contribution to the core network preservation needs. Despite the withholding of income in FYs 2011/12, the RF provides some degree of certainty about forthcoming financing, more so than GOM’s Development Budget funding. The statutory obligation of the Roads Fund is to support both routine and periodic maintenance needs. In practice, RF income has been insufficient to fund periodic maintenance and as a result repair and rehabilitation works are increased. 185. As discussed above, it is unlikely that RF income will rise to levels at which it can fully fund routine and periodic maintenance: by recognized asset management standards, maintenance of the road network is seriously underfunded. The Fuel Levy expressed as a percentage of the pump price per litre has dropped from 13% in early FYs 2009/10 to 6% during 2012, and stood at 6.5% 48 Including contractual commitments and potential claims. 69 (12 US$ cents/litre) in February 2013. In order to alleviate the maintenance funding shortage, the levy should be raised to a level of at least 10%, increasing gradually as part of the monthly fuel price adjustment process. 186. Scenarios for FYs 2012/13 to 2022/23 based on current IMF GDP growth projections, estimate the Roads Fund to increase from MK 8 billion in FYs 2012/13 to MK 30 billion in FYs 2022/23. This would imply an accumulated RF income over the period of US$ 400-450 million. Maintenance Methodology 187. The present maintenance management systems use short term contracts on both paved and unpaved roads, meaning that contractors are not continuously in place to carry out conventional routine maintenance, particularly important on unpaved roads. The Roads Authority has expressed interest in introducing output and performance based contracting methodologies, benefiting from successful experiences elsewhere in the region. Pilot contracts would be awarded for works on both the paved and unpaved networks, but would require some support from development partners. Rural Accessibility 188. Rural accessibility continues to be a constraint for agricultural development. The road network is already large by regional standards, and limited funds are available for expansion or even an adequate level of maintenance. Careful prioritization must be used to identify roads critical for agriculture or other social and administrative functions. 189. A new approach has been recommended using appropriate technology to rehabilitate the unpaved network to a maintainable condition with an effective regime of routine maintenance. This could be avhieved using output and performance based methods. Four complementary solutions would establish a management and maintenance regime for the unpaved road network: (i) rehabilitation using labour-based methods and appropriate equipment; (ii) routine maintenance system using a small tractor and light grader technology; (iii) spot improvement of problem sections where traffic levels do not justify full rehabilitation, and (iv) upgrading of selected earth road routes to a more durable all-weather surface where high socio-economic returns and sufficient levels of traffic can be demonstrated. This approach should build upon the widespread experience in Malawi of the use of labour-based methods, and support the coordination and management of such initiatives at district level. 4.3.2 Rail Sub-sector 190. The railway network in Malawi totals 797 km of mainline single Cape Gauge track. The network serves the south and centre of the country, and is part of an international network extending from the Indian Ocean port of Nacala in Mozambique to the dry port at Chipata in Zambia. The infrastructure suffers from a history of insufficient maintenance and is vulnerable to flooding, whilst rolling stock is unreliable and in short supply. Except for the relatively short section of the network that will carry coal traffic from Moatize to Nacala, the current level of traffic is not sufficient to finance the major investments necessary by the concessionaire to modernise the permanent way and rolling stock without support from Government and development partners. 70 4.3.3 Inland Water Transport Sub-sector 191. Malawi’s inland water transport system comprises Lake Malawi and the Shire-Zambezi river corridor envisaged to serve the newly constructed Nsanje World Inland Port; the development of the IWT system from Nsanje on the Lower Shire River along the Zambezi to the sea is to be the subject of a feasibility study. The improvement of the road network has contributed to the decline of water transport from 2006 to 2012. 192. The concessioning to the private sector of the shipping services since 2010 and the port operations since the end of 2012 is expected to significantly increase current transport volumes, attracting freight traffic either from new developments or from the roads. The quality of the business plans of the concession companies, Lake Malawi Shipping and Lake Malawi Ports- both subsidiaries of MOTA Engil - are therefore of crucial importance. 4.3.4 Aviation Sub-sector 193. Malawi has four airports with paved runways of which Kamuzu International Airport (KIA) and Chileka (Blantyre) International Airport are by far the most important for both international and domestic traffic. The number of daily aircraft movements in Malawi is modest, about 50 per day presently (55% domestic) with KIA handling over twice the number of international flights as Chileka, but only 1.5 times the number of domestic movements. 194. There has been a steady growth of international passengers (6% per year over the period 2001- 2011), while domestic air travel has remained at a constant level. The annual number of international aircraft movements has remained more or less constant at Chileka and increased at an average 2% per year at KIA. 4.4 Policy Priorities or Suggestions for Reform 195. The assessment of sub-sector expenditures in Section 1 and associated sub-sector performances in Section 2 have identified the following priorities for policy and reform in the sector: Transport Sector x Prepare a transport sector strategy and increase MTPW capabilities to prioritize between and within the transport sub-sectors; x Improve the National Transport Database, collecting data, processing, analysis and use of findings for planning; Increase interaction with MPs (Parliamentary Transport Committee) to discuss options on planning and technical issues. Roads Sub-sector: x Phased restoration of the real value of the Roads Fund, which has been eroded by the devaluation of the Malawi Kwacha and the non-transfer of part of the fuel levy, in order to finance a larger share of the required routine and periodic maintenance works. The levy should be fixed as a percentage of the fuel price; x Settlement of payment arrears which are consuming the majority of the sector’s budget allocation and thus preventing the financing of necessary periodic maintenance and rehabilitation works. A strategy should be developed which includes the payment of these arrears on negotiated terms and measures to limit any further increases in their value; 71 x Improved procedures for the economic and social analysis of projects, with individual projects being selected from a programmatic assessment of candidate projects, leading to the adoption of economically justified solutions for all road works and to the more cost-effective allocation of funds; x Planning of activities based upon a Medium Term Expenditure Framework which is agreed with the MoF in order to ensure the availability of funds for longer term contracts, and update these plans regularly to reflect actual progress; x Assessment of the existing road classification system in order to reduce the gap between the current network size, funding capacity and the requirements for access in rural areas; x Capacity development to improve the quality control and oversight of all actors in the project management cycle, the Roads Authority, consultants and contractors, to ensure that better value for money is obtained from the allocated funds; x Adoption of design standards based on regional research and best practice that are closely linked to expected traffic levels; x Assessment of the seriousness of axle overloading, primarily by domestic transporters, leading to appropriate axle load control actions on trunk roads; x Improvements in maintenance of paved and unpaved roads through the piloting of output and performance based contracts, and the use of long-term routine maintenance contracts; and x Improved rural accessibility based on a decentralised approach with careful prioritization and appropriate technical solutions, combined with funding, capacity building and programme coordination and consolidation targeted at support for agricultural development, a high priority of ERP 2012 and MGDS II. Rail Sub-sector: x Adopt a prioritized investment approach for rail network rehabilitation based on demand, affordability and sustainability, identifying potential bulk export freight to be carried by intermodal transport - in particular by rail and IWT - linking to the Nacala Line; x Develop the railway concession management and regulatory capacity of the Railways Division or a future Transport Regulatory Authority (TRANSRA), and x Repair the extensive permanent way and bridge damage caused by the 2012/13 floods, and work to reduce the network vulnerability to flooding. Aviation Sub-sector: x Respond to and address the ICAO recommendations for the aviation sector in Malawi - the Significant Safety Concerns, and x Regulate the sub-sector through the Malawi Airports Authority, Civil Aviation Authority, or TRANSRA. Inland Water Transport Sub-sector: x Management of PPPs special purpose companies - Lake Malawi Shipping and Lake Malawi Ports - by the Marine Department or perhaps TRANSRA, and x Assessment of future IWT potential in a ‘Products Study’ jointly with the railway sub-sector. 72 Implementation of Previous PER Recommendations 196. Several of the key issues that have been identified in the present PER were included as recommendations in the 2007 and 2001 PERs. Government did make significant progress in some of these areas but momentum has often been lost, resulting in a low level of implementation of many of the recommendations. (i) Roads Fund income: although the Fuel Levy was increased several times, it failed to reach the recommended level of 16.5 US$ cents/litre, and the increases were not made on a regular or systematic basis due to political considerations.; (ii) Funding for systematic periodic maintenance has not been prioritized; (iii) Road data management and prioritization tools were introduced by RA and consultants working for RA, used for preparation of the 2010-20 Road Sector Programme, but not consistently since then. Collection of new data has not been regular; (iv) The unpaved rural road network has received some attention, but the initiatives have been thinly spread and have not succeeded in improving network functionality, and (v) The capacity of the domestic construction industry could not be substantially increased (e.g. no policy change towards term maintenance contracts). 197. Reasons for the lower-than-expected implementation of previous PER recommendations are reported to involve a combination of the following: (i) high level directives have at times overridden the application of rational (techno-economic) planning criteria by MTPW and the Roads Authority; (ii) difficulties in achieving a common understanding among the five key decision-makers (MTPW, MEPD, MOF, MLGRD and Parliament) on programme content due to a lack of early consultation, and pressures from strong ‘political economy’ players pursuing short -term objectives; (iii) the political sensitivity of increases in the fuel price, and (iv) insufficient influence and/or capacity on the part of the Roads Authority to raise the professionalization level of the consultants and contractors that are contracted by the sector. Strategies must be devised to address these issues if the benefits from the current PER are to be delivered. 73 CHAPTER 5. EDUCATION 5.1 Introduction 198. This chapter will analyze the data on education public expenditure allocation and utilization. The goal is to identify reforms which ensure a more efficient, effective, equitable, and strategic use of funds in the sector . 5.2 Assessment of Education Expenditures 5.2.1 Level of Expenditures Trends in Overall Expenditure 199. Education sector expenditure accounts for the largest share of total GOM expenditure. In 2011/12, expenditure on education accounted for 18% of total GOM expenditure followed by Agriculture (13%) and Health (11%), as shown in Annex A8a and Figure 5.1. 200. Education sector expenditures have been increasing faster than the overall national budget, both in nominal and real terms. Overall, education sector expenditure has grown at an annual rate of 35.0% between 2008/09 and 2012/13, which is much higher than the average growth in the overall GOM expenditure of 21.4% (Table 5.1). In particular, growth in education recurrent and development expenditures averaged 32.2% and 74.2% respectively during that period, compared with growth of 20.2% and 28.4% in the corresponding components of overall GOM expenditures. The growth of over 160% in development expenditure recorded in 2009/10 reflects the implementation of the Joint Financing Arrangement (JFA) described below. Education sector expenditures have also increased in terms of 2007 constant prices, from MK 18.4 billion in 2007/08 to MK 51.3 billion in 2012/13 (Table 5.2). Figure 5.1: Share of Education in On-Budget GoM Expenditure 25.0% 20.9% 20.4% 19.4% 20.0% Share of recurrent 18.7% expenditure 17.7% 16.4% 15.8% 14.0% 14.5% 15.0% 13.6% 12.4% 10.0% 9.0% 9.2% 8.2% 4.4% 5.0% 0.0% 2008/9 2009/10 2010/11 2011/12 2012/13 74 Table 5.1: On-Budget Expenditure Growth Rates 2009/10 2010/11 2011/12 2012/13 Average Government recurrent expenditure 3.7% 17.1% 8.9% 51.1% 20.2% Government development expenditure 26.2% 42.5% 19.2% 25.5% 28.4% Total Government Expenditure 7.3% 21.9% 11.1% 45.1% 21.4% Education recurrent expenditure 16.9% 39.1% 18.4% 55.2% 32.4% Education development expenditure 160.3% 29.5% 33.9% 85.8% 77.4% Total Education Sector Expenditure 24.9% 38.0% 20.1% 58.9% 35.5% 201. Malawi's expenditure on education relative to GDP compares favourably with other countries in the region. Figure 5.2 shows that, among the selected African countries covered, Malawi's expenditure on education as a share of GDP was at 8.7%, which is above the average of 6.6% in 2011, and second to Lesotho's expenditure of 12.4% of GDP. Angola had the lowest share at 2.6% of GDP. Figure 5.2: Education Expenditure as % of GDP in Selected Countries (2011) 14.0% 12.4% 12.0% 10.0% 7.9% 8.3% 8.7% 7.8% 8.0% 6.7% 5.0% 4.9% 5.4% 6.0% 2.6% 3.2% 4.0% 2.0% 0.0% Source: UNESCO Education Statistics 2011 Recurrent and Development Expenditures 202. Recurrent expenditures significantly exceed development expenditures, but are proportionately declining due to increased donor funding. Several DPs have switched funding modalities from the off-budget project approach to a more harmonized approach that uses the government's system, as a result of the implementation of the Joint Financing Agreement (JFA) under the Education Sector Wide Approach (SWAp). On average, recurrent expenditures have constituted about 82.9% of the education sector's budget between 2007/08 and 2012/13 (Table 5.2). This is largely attributable to growth in the sector's development expenditure, whose share of the national development budget has increased from 4.4% in 2008/09 to 12.6% in 2012/13. 75 Table 5.2: Education sector on-budget expenditures 2007/08 2008/9 2009/10 2010/11 2011/12 2012/13 Recurrent expenditure (MK million) 14645.0 26513.1 30988.8 43117.0 51057.6 79223.2 Recurrent expenditure (% of total) 73.3 94.4 88.3 89.0 87.8 85.7 Real recurrent expenditure (2007=100) 13478.5 22515.7 24507.1 32836.3 32825.3 43949.6 Development expenditure (MK million) 5324.0 1573.0 4094.7 5302.6 7101.7 13191.6 Development expenditure (% of total) 26.7 5.6 11.7 11.0 12.2 14.3 Real development expenditure (2007=100) 4899.9 1335.9 3238.3 4038.2 4565.8 7318.1 Education expenditure (MK million) 19969.0 28086.1 35083.6 48419.6 58159.3 92414.7 Education expenditure (2007=100) 18378.4 23851.6 27745.4 36874.5 37391.1 51267.7 203. Within the recurrent expenditures, personal emoluments (PE) only slightly exceed expenditures on other recurrent transactions (ORT). On average, PE account for about 59.4% of recurrent transactions (Table 5.3). The education sector is by far the largest employer within the civil services. The proportionate increase in 2012/13 reflects recruitment of additional teachers and improving teacher salaries. However utilisation of ORT tends to be low as discussed further below, largely due to weaknesses in financial and procurement systems. Table 5.3: Composition of Education Sector Recurrent Expenditures 2008/09 2009/10 2010/11 2011/12 2012/13 Average PE 56.7% 54.8% 55.2% 58.4% 72% 59.4% ORT 43.3% 45.2% 44.8% 41.6% 28% 40.6% How Government Finances Education 204. Education sector public expenditure is programmed through five main channels on the Government budget. These are (i) the Ministry of Education, Science and Technology (MoEST) vote, covering all personnel emoluments (PE), on-budget development expenditure and other recurrent transactions (ORT) expenditure in most of the sub-sectors; (ii) part of the Local Assemblies votes covering ORT in primary education; (iii) subventions which are dominated by funding for public universities; (iv) part of the Local Development Fund (LDF) region vote, covering school construction; and (v) a very low allocation to the Ministry of Gender, Women and Child Development (MGWCD) vote, to cover early childhood development (ECD) coordination. Excluding the last which is quite insignificant (only about MK 30 million), it was estimated that the proportions of the 2012/13 revised education sector budget of MK 92.4 billion for 2012/13 that were allocated to each of the 4 remaining channels were 66.1%, 6.2%, 20.1%, and 5.1% respectively. Resources transferred to the MoEST are managed through cost centers of which there were 298 in 2012/13. 205. Part of the education sector budget is devolved to local councils. In 2012-13, 6% of the education budget was devolved to the district level for ORT of primary education. Local councils are only responsible for the ORT part of recurrent expenditures for primary schools. Salaries for 76 primary school teachers and key teaching and learning materials for primary education are still centrally procured. 206. While Treasury communicates a ceiling of education sector resources to be transferred to the districts, the actual allocation of the resources to the districts is implemented by the National Local Government Finance Committee (NLGFC). At the local level, education infrastructure development (i.e. construction of classrooms, teachers’ houses and associated infrastructure) is also funded under the Local Development Fund (LDF) vote. School infrastructure development is implemented under the LDF’s Community Window. The projects are supervised directly by local communities under the overall guidance of the DEMs and the local councils in general. LDF resources transferred to the education sector totalled MK 3.5 billion in 2011/12, and MK 4.7 billion in 2012/13. The LDF is now the main and most efficient means for the delivery of primary education infrastructure in Malawi. 207. Public secondary schools are cost centres and have decision making authority over how their budgets are spent. The trend in the devolution of finances to divisions and secondary schools depicts an unclear pattern. ORT expenditures managed by the divisions and secondary schools increased from MK 0.7 billion in 2008/09 to MK 1.7 billion in 2011/12. However, secondary education ORT devolution managed by headquarters also increased from MK 0.7 billion in 2008/09 to K1.0 billion in 2010/11, before declining significantly to only K0.2 billion in 2011/12. This was due to the non-exhaustion of the budgeted allocation for teaching and learning materials (TLM). 208. Public technical education is financed through the MoEST and other votes, TEVETA levy, student fees, internally generated funding, and donor support. There are seven public technical colleges, each of which is also a cost centre on the MoEST vote. A total of MK 568.9 million was allocated to these in the 2012/13 revised budget, to cater for PE and ORT expenses. A significant amount of the sector’s funding is channelled through the Technical, Entrepreneurial and Vocatio nal Training Authority (TEVETA) whose funding arrangement is discussed below. In addition to the direct funding under the MoEST vote, technical colleges also charge MK 42,000 per student per annum in boarding and tuition fees, and receive MK 36,000 per student per annum from TEVETA for teaching materials, as part of TEVETA’s user-fee subsidisation function. 209. TEVETA, an independent and autonomous body responsible for technical education and training, manages the TEVET Fund. The fund is used to finance (a) approved technical education and training programmes; (b) special programmes to support the technical education and training system; (c) user-fee subsidisation through scholarships, grants and loans; (d) incentives to employers to invest in technical education and training; (e) an endowment fund to support technical education and training; and (f) governance and management structures of the system. 210. The TEVET Levy is the main source of financing for technical and vocational training. The TEVET Levy, established under the TEVET Act 1999, is set at 1.0 % of the basic payroll for both private employers and the government. The TEVETA Levy accounted for about 88% and 99% of TEVETA incomes in 2010/11 and 2011/12 respectively. On average, the levy accounted for 92.5% of TEVETA income during the period 2007/08 – 2011/12. Levy income from the private sector is by far the most important source of income. 211. Subvented organisations, dominated by the public universities, are funded through transfers.49 Development expenditures of subvented organisations are, however, allocated to the 49 Although revised subvention allocations are used here and compared with actual education sector recurrent expenditures, actual subvention expenditures are normally the same as the revised budgets. 77 MoEST vote. Some 74 – 80% of these recurrent budget resources accrues to the public universities, and represent the most significant source of their recurrent funding. Universities also claim 19-20% of the total education recurrent budget, while the claim for all subvented organisations in the sector is around 25%. The government contribution constitutes the largest percentage of the total revenue of public institutions, ranging from 75 to 85 percent of the recurrent budget. Tuition fees are minimal representing between 4 and 14 percent of the total cost. The balance is internally generated within the system through postgraduate tuition fees, consultancies, research grants, hires and rentals and other professional services rendered. 5.2.2 Allocative and Technical Efficiency Expenditure by sub-sector 212. Primary education accounts for the largest and increasing share of the education sector expenditures. Annex A9b shows the proportions of education sector expenditures attributable to primary education for selected Africa countries, while Table 5.4 presents Malawi’s education sector recurrent expenditures between 2008/09 and 2011/12 for each level of education and for the management of the education system.50. Annex A9b, Annex A9c, and Table 5.4 show that actual expenditures remain below the Education Sector Implementation Plan (ESIP) and Global Partnership for Education (GPE) targets, despite that there is an increasing trend. Higher education accounts for the second largest proportion of recurrent expenditures, followed by secondary education, but the proportions for both have been declining over time. The increasing trend in expenditure on teacher education reflects efforts to address poor pupil-teacher ratios. Table 5.4: On-Budget Recurrent Expenditure by Sub-Sector: 2008/09 - 2011/12 2008/09 2009/10 2010/11 2011/12 Sub-Sector K million % K million % K million % K million % Primary education 12,783 48.2 14,424 46.5 20,910 48.5 26,186 51.3 Secondary education 4,708 17.8 5,338 17.2 7,241 16.8 7,572 14.8 Technical education 275 1.0 381 1.2 625 1.4 772 1.5 Teacher education 751 2.8 1,141 3.7 1,906 4.4 3,264 6.4 Higher Education 5,703 21.5 6,989 22.6 8,833 20.5 9,412 18.4 Management 2,293 8.6 2,715 8.8 3,603 8.4 3,851 7.5 Total 26,513 100.0 30,989 100.0 43,117 100.0 51,058 100.0 Source: Mambo et al (2012) 213. Malawi's recurrent expenditure on secondary education of 16.8% is lower than the average of 26.7% for selected African countries (Annex A8d). This reflects the relatively small secondary education system that affords a low transition rate from primary to secondary education of 51%. The scope for resource reallocations on the MoEST vote is quite limited. The foregoing discussion reveals thatalthough primary education recurrent allocations are a rising proportion of total education 50 Expenditure figures for primary education do not include other components of basic education. Specifically, the primary school expenditure figures exclude early childhood development (ECD), adult literacy and continued basic education (CBE) for out-of-school youth. Although management typically includes some part of the expenditures by district and division offices, this expenditure is included under primary and secondary education respectively as it is difficult to separate out expenditure on offices from schools 78 sector recurrent allocations, their proportion falls short of the GPE target of 64.0%, being 51.3% in 2011/12. Allocations to other sub-sectors are equally low, and the increasing resource needs of these other sub-sectors are equally justified. The low transition rates from primary to secondary schools and the need to lower PTRs and PQTRs in primary and secondary education (hence to spend more on teacher education) are cases in point. Secondary education recurrent expenditure was only 16.8% of total recurrent expenditure in 2011, compared with a regional average of 26.7%. In addition, a. although personal emoluments generally exceed ORT and development expenditures, and despite the inadequacy of resources for ORT expenditures (particularly TLMs as well as allocations for routine maintenance and basic procurement for individual schools), the need to reduce the currently high PTRs and PQTRs implies that resources cannot be reallocated from personal emoluments to ORTs. Education Sector Resource Mobilisation 214. There is rapid growth in on-budget funding, especially from development partners since the Joint Financing Agreement (JFA) of 2009/10. Annex A10 and Figure 5.3 show that local (GOM) resources dominate the total funding sources to education, but that this has declined from 99.1% of the sector's total funding in 2008/09 to 73.5% in 2012/13. Thus, at MK 20.2 billion in 2012/13, pooled resources are estimated to account for more than a quarter of the sector budget. In nominal terms, however, both the local and donor resources have been increasing by more than the inflation rate. GOM resources have grown by an average of 25.0% per annum between 2008/09 and 2012/13, while pooled development partner resources increased enormously in 2010 and 2011, by 219.8% and 48% respectively. 79 Figure 5.3: Distribution of On-Budget Education Funding by Source DP discrete funding DP pooled funding GoM 4.3% 2012/13 22.2% 73.5% 0.9% 2011/12 17.7% 81.4% 2.6% 2010/11 14.4% 83.0% 4.5% 2009/10 6.2% 89.3% 0.9% 2008/09 0.0% 99.1% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% Source: Mambo et al. (2012), revised 215. Significant amounts of development assistance are still off-budget, despite the JFA51. Off-budget donor funding has declined from US$71.8 million or 97.7% of development assistance in 2008/09 to US$51.5 million or 36.5% in 2012/13 (Table 5.5). On the other hand, the share of on-budget pooled funding in total assistance has increased from zero to 53.1% during the same period, while on-budget project funding has not shown significant changes. While this trend is commendable, the fact that a significant amount of donor funding is still provided off-budget suggests partial attainment of the GOM's preferred aid modality. Table 5.5: Distribution of DP Funding by Type (% of total DP funding) Type 2008/09 2009/10 2010/11 2011/12 2012/13 Off-Budget Projects 97.7 77.8 67.5 42.6 36.5 On-Budget Projects 2.3 9.4 5.0 2.7 10.4 On-Budget Pooled Funding 0.0 12.8 27.5 54.6 53.1 Total 100.0 100.0 100.0 100.0 100.0 51 The data used in this analysis may be misleading because they exclude funding from the People’s Republic of China (PRC) after 2010/11, due to data unavailability. PRC assistance is entirely off-budget. In 2010/11, PRC provided 35% of total off-budget assistance. 80 5.2.3. Intra-sectoral equity/Benefit Incidence Analysis 216. Analysis of the IHS3 survey data indicates a number of equity issues in education sector. Enrolment is progressive in public schools and regressive in private schools. Enrolment in public primary schools consistently declines as income increases regardless of the gender of the child, area and region of residence.. Secondary school enrolment is highly regressive: only 8% of public secondary school children come from the poorest households, compared with 31% for the richest households. This pattern is replicated in private secondary schools where the corresponding figures are 14% and 27%, respectively. Moreover, children from the richest households account for up to 41% of enrolment in conventional secondary schools, compared with only 5% from the poorest households. The pattern shows that the relatively under-resourced community day secondary schools (CDSS) enrol more of the poorer children, but not in excess of children from well-off families. Thus, the transition from primary to secondary education clearly favours children from rich households. Enrolment in tertiary education is highly regressive as it is concentrated amongst the richest quintile Over 90% of university students come from the richest quintile households. This pattern is sustained, albeit weakly, when training colleges are considered. In that case the proportion of students drawn from the richest households is lower at 75.3%. Chongoma and Franz (2013) further note that this social bias in favour of the wealthy is reinforced by the mechanisms of admission to subsidized formal TEVET, which is based on merit, hence favouring the better off students. Educational attainment raises with wealth (6–24 year olds) Source: http://econ.worldbank.org/projects/edattain 217. Space limitations in secondary and higher education are a key factor limiting equitable enrolment. Enrolment in public educational institutions beyond primary school is grossly limited 81 by space which has not increased to cope with demand. Only half of the students who complete primary education can be admitted into secondary school due to space limitations, and this is an improvement from much lower level, made possible through the introduction of community day secondary schools (CDSS) in 2003. Mambo et al. (2012) note that more than 50,000 applications are received annually by the technical colleges which are able to enrol only around 6,000 students due to space limitations. Against this background of limited spaces, children from richer families become favourites due to many factors such as normal growth (as opposed to stunted growth), access to information and options, and inherent pressure to meet the expectations of their educated parents. Space limitations are also the basis for the introduction of a quota system of admission in public universities, which improves regional disparities. 218. Out-of-pocket expenses are much higher for private schools than for public schools at all levels which may condition education access by children from poor households. Out-of-pocket allowances are 13 times higher in private primary schools than in public primary schools. Although these decline to 3.3 times and 1.9 times in secondary and tertiary education, the actual amounts involved at the higher levels are quite enormous. . 219. Government subsidy is progressive at lower education level, and regressive at the higher levels. In primary education, the poorest get the largest share (29%) of government subsidy, while the richest get only 9%. In secondary education, the share of the subsidy is only 10% for the poorest and 28% for the richest, suggesting that the system is regressive. This is amplified at the tertiary level, where the poorest only get 1% compared with 82% for the richest households. Since most children from poor households do not go beyond primary education and hardly any of them reach tertiary education, government funding to secondary and tertiary education sectors benefits children from higher income households. 220. Marginalized groups, such as youths from poor households, school drop-outs and people living in the rural areas face considerable problems to access relevant skills development programs. This is due to supply restrictions in rural Malawi and limited capacities of households to pay for training fees. Equitable access to TEVET is limited because publicly subsidized skills development is merit-based and is mostly restricted to formal programs at post-secondary level. 221. Government endeavors to promote secondary and higher education access by the poor and vulnerable through the provision of school bursaries and cash transfers. Although bursaries are helpful in increasing enrolments, there are not enough resources to meet targets. Some 10,962 secondary school bursaries were given in 2010/11, against a target of 12,000. Although TEVETA is implementing a bursary scheme, the number of students effectively targeted is also low due to funding shortages (only 88 bursaries were awarded in 2011/12). 222. TEVETA has started to systematically invest in non-formal and informal TEVET. This initiative seeks to expand access to skills development for youth, people in the rural areas and other groups that are underserved in the formal education and training system. This includes the establishment of community-run “satellite centres”, strengthening the capacities of informal sector master crafts people to increase and improve their training activities, and facilitating tailor-made training. Funding is from donor programs or own income derived from sources other than the levy. 82 5.3 Education Sector Performance and Key Issues Performance against Selected Indicators 223. The MoEST established the Education Management Information System (EMIS) in order to monitor education sector outcomes and to facilitate performance evaluation against the targets set in the MGDS as well as the Millennium Development Goals (MDGs). 224. Data from the EMIS and other sources show that Malawi has made some progress to improve education sector outcomes. Improvements have been made in terms of school enrolment, completion, and gender parity at various levels of the education system. However, the education sector still reports poor outcomes against many critical indicators. Low pupil-teacher ratios, poor learning outcomes, low internal efficiency coefficient, and low transition rates, at various levels of the education system. Implementing Key Recommendations of 2007 PER 225. Some effort has been made to implement several of the 2007 PER recommendations, but more needs to be done. The percentage of on-budget resources spent on primary education has increased while that spent on higher education has declined (Table 5.4), but per capita allocations to higher education are still too high (Section III.B). An open and distance learning (ODL) rural teacher recruitment program has improved equity in the number of teachers in rural versus urban areas, and teacher output has generally increased, but disparities still exist and PTRs remain too high (Section III.E). A rural teacher allowance was introduced, but it is too low and poorly targeted (Section III.F). Public universities have increased enrolment and outsourcing of some services, but are still inefficient and reflect gross inequity in the use of resources (Section III.B). 5.4 Policy Priorities or Suggestions for Reform 5.4.1 Enhance Budgetary Execution through Enhanced Decentralization 226. Increase efficiency of sector resource utilization. While actual expenditures by the education sector have been consistently below the MGDS targets, given the fiscal space limitations, improvements in the utilisation of the resources available to the sector are critical in enhancing sector performance.. 227. Decentralise personal emoluments to address weaknesses in budget execution that lead to the accumulation of significant amounts of salary arrears over time. Salary arrears reached a peak of MK 613 million by mid-2012, but about 90% of these were targeted to be cleared by December 2012. By that date, over MK 72 million (11.7%) was still not paid. Moving forward, it is pertinent to implement decentralization of personal emoluments in order to address some of the factors that accounted for the accumulation of arrears, such as lags in communicating staff promotions and new recruitments and lags in paying out allowances to ODL students. Indications by the MoEST are that IFMIS and the Human Resource Management Information System (HRMIS) were scheduled to be installed not only at the Education Division Management (EDM) offices in Lilongwe, Kasungu, Mzuzu and Blantyre for payroll processing and ORT payments across all the divisions. 228. Decentralise the procurement of TLMs. Utilisation of the TLM budget was 7.6% for primary, and only 20.4% for secondary. Hence, some of the shortages in TLMs in primary and secondary schools 83 are due to poor budget execution As stated, utilisation of the TLMs resources managed by the MoEST headquarters is low. It is a generally accepted view that decentralisation of TLMs allocations could increase utilisation levels. Moreover, some success stories associated with the implementation of both the DSS and (especially) PSIP facilities (see Mambo et al., 2012), suggest that devolution of more finances to the school level could enhance efficiency. The MoEST reports that significant progress is being made to devolve budgets for TLM budget to local levels, and completion of the process of linking education divisions to the Accountant General’s office could facilitate this further. 229. Continued prioritization of school grants. The primary school improvement program (PSIP) provides significant resources to the school level for critical school inputs. The recent review of the Education Sector Iplementation Plan aslo found that by transferring responsibility for the control and expenditure of funds to the school the PSIP system also built local and school accountability, 230. Continue to use the Local Development Fund (LDF) in the construction of primary school infrastructure. Increase school infrastructure for both primary and secondary schools as a key priority.. The primary school PCR has deteriorated from 85.7 in 2007/08 to 105.1 in 2011/12. Due attention is needed to increase primary school infrastructure since, among other effects, high PCRs contribute to poor learning outcomes. Due to limited infrastructure access is constrained as approximately half of the primary school graduates are unable to continue to secondary, (Table 5.6). The GOM constructed 458 primary school classrooms during 2008-12 through the Education Infrastracture Manangement Unit, against the target of 5,500. On the other hand, LDF delivered 1,220 primary school classrooms in 1 year during 2011/12, against a target of 1,100 classrooms. These levels of delivery are also reflected in the utilisation of resources under the two arrangements: whereas the EIMU’s development budget utilisation was only 50% of MK 5.9 billion during 2011/12, LDF development budget utilization was 100% of MK 3.5 billion. Using the LDF is, therefore, an efficient option in speedily addressing the problem of high PCRs in primary schools. The Government should also consider extending the use of the LDF in the construction of infrastructure for community day secondary schools (CDSSs). 231. Concern has been raised regarding the quality of LDF construction outputs that they may be below the required standards. LDF projects are managed by the district council’s public works coordinator who is responsible for all infrastructure provided through the district council. This coordinator in turn has a number of supervisors that are responsible for construction supervision and inspection of infrastructure projects. Normally, none of these is a qualified civil engineer. Hence, independent building consultants and architects and civil engineers should be engaged to manage LDF projects alongside the council officers. Table 5.6: Primary and Secondary PCRs; Primary-Secondary Transition Rate: 2007/08 – 2011/12 2007/08 2008/09 2009/10 2010/11 2011/12 Primary school PCR (all schools) 85.7 94.3 97.4 101.0 105.1 Secondary school PCR 46.4 48.0 49.0 Primary-Secondary transition rate 0.54 0.58 0.51 0.49 0.51 Sources: GoM, 2013; Mambo et al., 2012. 232. Deepen the effectiveness of the decentralisation process. While the devolution of resources to lower levels of government has potential to enhance efficiencies in resource utilisation and service delivery and to promote the participation of local communities, it faces some challenges. In addition 84 to low capacities for revenue generation, especially for rural councils, and the prolonged absence of elected ward councilors, challenges include the inadequacy of resources earmarked for the devolved functions, delays in transfer of funds to local levels, or delays in procurement, low absorption and loss of real value due to inflation. 5.4.2 Reform Higher Education Financing Arrangements to Enhance Equity and Avoid Duplications 233. Adopt output-based and performance-based funding for public universities. . Malawi relies on a traditional historical / negotiated allocation system to distribute the annual recurrent budget among public universities and institutions. This means that the Ministry of Finance uses the previous year’s funding amounts as the reference point from which incremental changes are made taking into account the anticipated level of government revenue and the expected enrollment growth of the higher education institutions. By and large, the level of public funding that individual higher education institutions receive is not linked to any measure of performance, such as quality or relevance. Other countries use a combination of performance-based methods, which attempt to encourage higher education institutions to be more innovative and make good use of scarce Four main types of innovative allocation mechanisms might be considered separately or in combination for this purpose: output-based formulas, performance contracts, competitive funds, and vouchers. 234. Address duplications in the delivery of higher education. Higher education institutions have yet to identify which programs can give them a competitive advantage. There is a duplication of some of the programs made to the detriment of a diversification into new areas based on the needs of the growth sectors of the economy, as indicated in the MGDS and the relative strengths of each tertiary institution. The overall distribution of enrolment by field of study has generally not been aligned with those areas which are critical for the economic development of Malawi. The research output of the public universities is small and linkages with industry are almost inexistent. 235. Develop a transparent student financing system that allows eligible students access to public and private universities. The Government contribution constitutes the largest percentage of the total revenue of public institutions, ranging from 75 to 85 percent of the recurrent budget. Tuition fees range between 4 and 14 percent of total income. Public institutions should be allowed to charge higher fees to students from well-off families, and the student loan system should operate more efficiently, especially in terms of loan recovery, to ensure adequate protection of needy students and availability of resources on a continuous basis. Indeed, higher levels of cost sharing cannot be achieved in an equitable and efficient way without a well-functioning student loan program. Therefore, to ensure higher levels of loan repayments the government needs to carefully review the lessons of experience to avoid the problems of high default levels faced when the student loan system was administered by the Public University Student Loan Trust since the Malawi Saving Bank which was expected to improve loan recovery has pulled out of the scheme. Student loans should be available to all eligible students, including students enrolled in private institutions and in parallel programs in public institutions, provided the institution / program they are enrolled in meet satisfactory quality standards. The repayment duration should be much longer than the present four years, and the monthly repayment amount should preferably be income-contingent to protect graduates who are unemployed or whose income is too low to make loan payments. 236. Need to further Public Private Partnerships (PPPs) as a possible avenue of attracting investment in higher education and especially in infrastructure development including student accommodation. The provision of catering, security and cleaning services has been outsourced in most colleges. As a result, savings have been realised in the use of utilities and facilities, the cost of perishable items, personnel costs and pension contributions. However overall there is very little 85 engagement with the private sector as possible investors in providing infrastructure in the institutions. UNIMA has developed a Build-Operate-Transfer (BOT) Framework Agreement for Construction of Building Units on University of Malawi Campuses which spells out the obligations and undertakings of the parties engaged in the BOT but this has still to be implemented. Government has since passed the PPP Act which should facilitate the implementation of such BOT frameworks. 237. Increase the capacity for internally generated funds by universities through research, consultancy and outreach. University staff undertake research, consultancy and outreach activities, but most of these are unreported and the resources generated are not accounted for, although a significant amount of university time is used. Universities have established Research and Consultancies Committees and policies in order to coordinate these activities and generate additional funding. While these initiatives are commendable, poor remuneration within the university system, lack of adequate incentives for disclosure of such activities, and rigidities in administrative structures will continue to limit the returns of these initiatives unless they are addressed. In addition, adequate funding for research activities is unavailable for most fields, both within the universities and at the national level. 238. Since every fundingthat has been set aside for secondary has been returned to the MoF unused due to limitied implementation capacity in the EIMU. 5.4.4 Enhance Implementation of Measures to Improve Internal Efficiency, and Assess their Performance 239. An estimated MK 1.97 billion is used annually to deliver primary education services to repeaters in Malawi (World Bank, 2009). The internal efficiency coefficient (IEC) at primary level is low, estimated at 35% in 2007, suggesting that 65% of resources were wasted due to dropouts and repetitions (World Bank, 2009). Although improvements have been registered since then in terms of dropout rates, repetition rates have not changed significantly. Thus, standard one dropout rates have shown an improvement from 24.6% of all students in 2007/08 to 13.9% in 2011/12. The corresponding figures for higher classes have moved on average from 12.6% to 10.2% during the same period. In terms of repetition rates, however, the change in standard one is a deterioration from 23.4% in 2007/08 to 24.6% in 2011/12. Similarly unimpressive statistics are recorded for higher classes, from 17.2% to 18.8% during 2007/08 -2011/12. International evidence shows that high repetition rates do not favour a better mastery of learning, increase the risk of dropping out, and have adverse effects on the PTRs and costs. To improve the primary school IEC, MoEST has developed guidelines for automatic promotion, and class size which were disseminated in 2011/2012 to all districts, all the DEMs, primary education advisors (PEAs) and head teachers for implementation. This was a component implemented together with remedial education, double shifting and overlapping to reduce PCRs. Concerted efforts in implementing these measures and assessing their effectiveness are required as progress is being monitored. 5.4.5 Enhance the Implementation of Measures to Lower PTRs at All Levels 240. Malawi has among the highest in Sub Saharan Africa at 92:1 making progress in early literacy and numeracy very difficult for many children. Rural primary schools, which are attended by a majority of school children, are particularly disadvantaged, with a PqTR of 99:1. 241. The MoEST is implementing commendable short-term strategies aimed at addressing this situation to produce an estimated 11,740 student teachers as follows: (a) a decision has been made to extend 86 the ODL program which recruits teachers from rural areas with slightly less qualifications to 2017 instead of terminating it in 2015, to train an additional 6000 student teachers; and (b) the anticipated opening of Phalombe Teacher Training College (TTC) in 2014 and Chiradzulu TTC in 2015 would enrol an additional 300 and 540 student, respectively. Combined with the normal output of TTCs, the conventional teacher training program will train an additional 5,740 students by 2017. However, the GOM (see GOM, 2011) reports that some 15% of the 3,347 Initial Primary Teacher Education graduates deployed in 2010/11 did not report for work. These realities need to be included in short- term programming and addressed more sustainably through a proper teacher incentive system. 242. Assess the cost-effectiveness of using the ODL teacher training arrangement in relation to other teaching methodologies. ODL students are a cheaper and potentially very useful addition to the teaching staff of rural schools. ODL students receive an allowance which was set at MK 20,000 per month in 2011/12, lower than the entry Grade L salary of MK 24,833/m in 2011/12. Moreover, the students are not entitled to the rural teacher allowance of MK 5,000 per month, and the GOM saves a lot of resources since they only stay for short periods in training colleges. Although the lower teacher training standards of ODL teachers could compromise on education quality, findings from the baseline survey of a randomized impact evaluation of this program suggest that this program improves PTRs without sacrificing quality. 243. Extend the ODL program to secondary education to address the extremely high rates. About 1,500 - 2,000 secondary school teachers are trained per annum, but the number of teachers in public secondary schools increases by only about 220 annually. The situation is worse in rural schools than in urban schools.. Moreover, a good number of those who join the teaching profession quit after a short while. The result is that secondary schools continue to rely on teachers who are trained to teach at the primary level. Official data shows that untrained teachers comprise 57.0% of the total number of secondary school teachers, hence the wide difference between PTRs and PQTRs. One explanation for this is that most fresh graduates from universities find it unattractive to work in rural areas, and the ODL arrangement presents a credible opportunity to arrest this situation. Redistribute teachers across standards in favour of the lower classes (standards). While the numbers of pupils steadily decline as class levels increase, the numbers of teachers do not change commensurately and sometimes increase with class. Particularly, there is a clear over-allocation of teachers in standard 8 (Table 5.7). The result of this pattern is that PTRs are too high in the lower classes and too low in the high classes: they decrease from 124 in standard 1 to only 24 in standard 8. A more equalised distribution of teachers across classes could improve learning outcomes at the lower level and prepare pupils better for the upper classes. Table 5.7 Distribution of Primary School PRTs per Standard Class STD1 STD2 STD3 STD4 STD5 STD6 STD7 STD8 Total Pupils 918,885 692,064 646,944 522,930 421,650 320,774 259,300 214,284 Total 7,413 7,477 8,489 8,464 7,737 7,124 7,114 9,041 Teachers PTR 124 93 76 62 54 45 36 24 Source: MoEST 5.4.6 Implement a Better Targeted Incentive Scheme for Rural Teachers 244. Implement measures to enhance teacher motivation across the education system in general, and for rural teachers in particular. Teacher motivation is critical to reduce teacher absenteeism 87 and attrition). Although Malawi does not have results from a nationally representative sample of teachers, recent impact evaluations found that collectively teachers are absent approximately 30 percent of the school year in rural areas. Some of the aspects to be considered in this regard are performance-based teacher promotion, and improved incentives for teachers to teach in remote areas (such as staff housing and transportation). Increased decentralization of teacher management could be an effective means to enhance teacher motivation. Progress is underway to implement a more objective teacher promotion system. A new teacher performance appraisal system has been adopted for primary and secondary teachers, where 30% of the appraisal score is based on supervision reports and 70% is based on aptitude test by Teacher Service Commission. In addition to using more objective criteria, the promotion system should take on board the key recommendations made in the ESIP Review 2012. 245. Rationalize the rural teacher allowance scheme. The hardship allowances for teachers teaching in rural areas was introduced in 2010/11 and accounted for approximately 10% of total personal emoluments in 2012/13. An estimated 85% of teachers were receiving the allowance. Although this is a commendable initiative to lower PTRs in rural areas, there are some critical targeting challenges in the administration of the allowance. For instance, some teachers from different schools within the same area receive the allowance, whereas some do not. At MK 5,000 since its introduction, the allowance level is also too low and requires revision. A better targeted and rationalized teacher incentives system could be more effective in achieving intended objectives. 5.4.7 Explore Further the Potential for Double Shifting and Overlapping in addressing PTRs and PCRs 246. Deepen implementation of double shifting and overlapping to assist in reducing PTRs and PCRs. As part of efforts to reduce PTRs and PCRs where they are too high, the NESP set a target of 15% for the number of schools that should implement double shifting by 2017. Data from EMIS 2010 shows that the number of primary schools practicing double shifting more than doubled from 153 (or 3% of all primary schools) in 2009 to 489 (6.3%) in 2011/12. The recent trend suggests that this target could be reached. An additional 663 (12% of primary schools) were practicing overlapping in 2010, but this was down from 13.0% in 2009. Fieldwork conducted during this review revealed that both double shifting and overlapping were associated with some practical problems including security and other community concerns. A fundamental challenge from the perspective of public expenditures was the fact that the double shifting allowance paid was considered too low by teachers. 247. Evaluate the impacts of double shifting and overlapping on learning outcomes. While addressing problems associated with PTRs and PCRs, the literature seems to suggest there could be some negative learning impacts of double shifting and overlapping (Annex A11). Given the country’s low learning outcomes, it is important to ensure that these interventions do not worsen the situation. 5.4.8 Improve Equitable Access and Labor Market Responsiveness of the TEVET System 248. With a view on equity considerations and prevalent skills needs in key economic sectors, a move towards mainstreaming non- and informal skills development for currently underserved groups in the TEVET system appears timely (Franz 2013). With additional financial resources and technical assistance to expand and develop best practice for delivery of these programs, Malawi can nurture lower level skills relevant for the development of its priority sectors, and expand options for productive employment among the poor. 88 249. A review of the current apprenticeship system appears necessary to inform policy reforms, looking at incentives, cost effectiveness and alternative delivery modes. According to Jutta 2013, apprenticeship training has always been the core delivery mode in the formal TEVET system in Malawi. As apprentiship training is dependent on cooperating companies, access to formal TEVET remains limited in Malawi, and is particularly affected in times of economic downturn.. Furthermore, the training duration is rather long by international standards and rather costly. To sustainably increase access to formal TEVET and contain costs, TEVETA and its stakeholders should initiate a thorough review and assessment of the current apprenticeship system in Malawi to identify and inform policy reforms.. 250. Address the problem of costly and confusing multiple assessment in TEVET through harmonization. Although TEVETA was set up to reform the technical education system and reorient it to the needs of the labor market, there have been issues with the new TEVETA system for testing skills, namely the Competence Based Testing (CBET) system. As such, the system still runs parallel to two old systems, namely the Trade Testing system under the Ministry of Labor, and the Malawi (Advanced) Craft system under the Ministry of Education. This multiple assessment arrangement is costly, confusing. It is a key impediment to sustainable quality improvement since it prevents development of a unified employer-involving quality assurance system, and forces teachers to train on the basis of parallel curricula. A process of harmonization requires a thorough review and clarification on the structure of the TEVET Qualifications Framework, institutional responsibilities, assessment methods, financial implications and sustainability strategies, and capacity building needs. The process also has to recognize that the different systems presently operating represent different paradigms of TEVET under the domains of labor and education. 89 CHAPTER 6. HEALTH 6.1 Introduction 251. This public health expenditure review is based on an analysis of revised health budget figures for the period between 2006-07 and 2011-1252. It provides a comprehensive situation of financing health, covering public and private sectors and external partners. This chapter’s focus is to assess public spending in the health sector through the lens of efficiency, equity, and sustainability. 6.2 Assessment of Health Expenditures 6.2.1 Level of Expenditures International Comparison (2008-09) 252. When comparing Malawi to other countries of similar income and to its neighbors in 2008-09, Malawi fares well in most cases but one. Abuja targets are not met, and domestic public health spending is relatively low. When only domestic public health spending is considered, Malawi’s share of total government spending has been showing a declining trend from 9% (2002-03) to 4% (2006-07) and settling at about 5% (2008-09). 253. Malawi’s GDP health spending share (11.6% in 2008-09) fares well when compared to the other countries of similar income and to its neighbors (Table 6.1). Malawi’s per capita health spending amount ($38.50 in 2008-09) also fares well in comparison to countries in the sample53. Total health expenditures (THE) grew at a higher pace over the years than the population’s growth rate, and THE per capita is getting closer to the World Health Organizations recommended level for low income countries. However, Malawi’s public health share of government spending (5.2% in 2008 - 09) does not fare well in comparison to countries in the sample. While the United Republic of Tanzania, Zambia and Mozambique are either meeting or close to meeting the Abuja target (i.e. 15% health share of total government spending), Malawi, with 5.20%, is not. 52 In this chapter, expenditures are used interchangeably with revised budget. This chapter does not use actual expenditure data, but relies in most cases on revised budget data. 53 The section benefits from past data on the national health accounts (NHA, 2002-03 to 2008-09). 90 Table 6.1: Health Expenditures by Southern African Development Community countries, in USD 2009 General Government GNI/Capita THE as % 2 Spending on Health Country 2 THE/Capita (WB Atlas of GDP as % of Total Method – Government 1 2 2008-2012) Expenditure Malawi 360 11.60% 38.5 5.20% Mozambique 460 6.20% 27.06 14.20% Tanzania (United Republic of) 540 5.10% 25.31 18.10% Zambia 1,160 6.20% 60.61 15.70% Lesotho 1,210 8.20% 70.05 8.20% Swaziland 3,470 6.30% 155.78 9.30% Angola 3,830 4.70% 203.18 11.30% Mauritius 8,040 5.60% 377.5 8.00% Average 2,383 4.80% 163.6 9.10% Source: 1http://data.worldbank.org/indicator/NY.GNP.PCAP.CD/countries?display=default 2 Ministry of Health, 2012. Malawi National Heath Accounts with subaccounts for HIV/AIDS, Malaria, Reproductive Health, and Child Health for Financial Years 2006/07, 2007/08, and 2008/09. Department of Health Planning and Policy Development, Lilongwe, Malawi Trend Analysis on Total Health Expenditures (2002-03 to 2008-09) 254. Malawi’s public spending in health was adequate, primarily as a significant contribution came from external financing. This has also made Malawi more vulnerable to external shocks. Malawi experienced improved overall financing of the health sector between 2002-03 and 2008-09. The increase in the magnitude from MKW 14 to MKW 72 billion was due to the sector wide approach (SWAp) partnership formed to receive great support from external financiers. Total health spending as a share of GDP showed a positive trend during this period (from 10% to 12%), which is mainly due to external financing. Public health spending as a share of GDP has been declining since 2002- 03 (from around 4 % in 2002-03 to around 2 % in 2008-09) due to resources diversion caused by external financing (Figure 6.1). 91 Figure 6.1: Trend of health spending as a share of GDP 15.00% 10.00% 5.00% 0.00% 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 THE as share of GDP Govt. health exp. as share of GDP Source: Ministry of Health, 2012. Malawi National Heath Accounts with subaccounts for HIV/AIDS, Malaria, Reproductive Health, and Child Health for Financial Years 2006/07, 2007/08, and 2008/09. Department of Health Planning and Policy Development, Lilongwe, Malawi. 255. Total per capita health spending grew in nominal and real terms. It rose from $15 in 2002-03 to $38.50 in 2008-09 in nominal terms, and from $15 in 2002-03 to $20.43 in 2008-09 in real terms (NHA 2010). 256. Health sector is increasingly reliant on external financing as public sector domestic contribution declines. External financing rose from 46 % to 66% of THE between 2002-03 and 2008-0954, while public sector domestic financing decreased from 35% to 18% of THE (Figures 6.2). Figure 6.2: Sources of health financing Sources of health financing 100% 7% 7% 7% 10% 6% 5% 5% 46% 62% 60% 60% 69% 62% 66% 50% 12% 10% 9% 9% 11% 12% 35% 12% 21% 24% 22% 14% 22% 18% 0% 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 Public share of THE Hhld share of THE External Financing share of THE Pvt. Sector share of THE Source: Ministry of Health, 2012. Malawi National Heath Accounts with subaccounts for HIV/AIDS, Malaria, Reproductive Health, and Child Health for Financial Years 2006/07, 2007/08, and 2008/09. Department of Health Planning and Policy Development, Lilongwe, Malawi. 257. Domestic spending on health has been declining over time, substituted by external financing; it is still low, and raises critical concerns for future sustainability of health programs. The injection of external financing had an immediate and positive effect on household health expenditure shares in total health expenditures, which declined from 12.2% in 2002-03 to a 9% range from 2003-04 to 2005-06 and stabilized at about 11% of THE thereafter. It had, however, a negative effect on public 54 NHA 2010, NHA 2006, NHA 2002 92 health spending shares in total health expenditures. Public sector contribution bottomed out in 2006- 07, when external financing share was at its peak. Nevertheless, improved increments in allocations were shown: they reached 18% of THE in 2008-09 (Figure 6.2). Public health expenditures (2006-07 to 2011-12) Table 6.2 provides a snapshot of the trend in public health spending during FY 2007 – FY 2012. Sources of financing in the public health sector 258. The public health sector has mixed sources of financing, such as external financing, tax financing, and limited resources earned from user fees, with the former taking up the larger chunk. Table 6.2: Trend analysis of public health spending between 2006-07 and 2011-12 Indicators 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 Total Public Health Expenditure, MKW million 47,543 64,290 72,275 -- -- -- Total Public Health Expenditures (vote 310 and 15,300 25,100 31,100 32,465 34,535 -- LGC only), MKW million Public health expenditure, per capita, MKW 1,159 1,845 2,221 2,254 2,317 2,006 (nominal) Public health expenditures per capita, MKW (real) 1,159 1,698 1,886 1,782 1,765 1,290 Public health expenditure per capita, USD 8.25 13.12 15.21 14.9 14.1 6.26 (nominal) Public health expenditure as share of GDP 4% 5% 5% 5% 4% 3% (domestic and pooled) Public health share of total government spending 12% 14% 12% 12% 11% 9% (domestic and pooled) MOH share of total public health expenditure 71% 70% 84% 71% 75% 70% (includes MOH headquarter; central hospitals; regional and district hospitals) Central hospital share of total public health 16% 15% 13% 14% -- 15% expenditure LGC share of total public health expenditure 29% 30% 16% 29% 25% 30% External pooled financing as share of public health 14% 14% 22% 7% 6% -- expenditure Source: MOF IFMIS System, WB/PER team estimates Note: “—“ indicates missing information. Notes: Due to the differences in the NHA and PER methodology the figures in Tables 6.2 and 6.3 are not comparable. * Nominal exchange rate is used (source for exchange rate between 2006 and 2011 is World Bank/Malawi) Source for 2012 end of year exchange rate is http://www.imf.org/external/np/loi/2012/mwi/120512.pdf (accessed on May 28.2013) - Malawi: Letter of Intent, Memorandum of Economic and Financial Policies, and Technical Memorandum of Understanding. December 5, 2012 259. External financing is injected into the health sector through pooled financing and discreet financing. A small proportion of external financing is pooled into government budget: in 2011-12, of the $91 million of external financing, only 20% was pooled under the government budget, while the rest 93 came as discreet projects managed by development partners. In 2012-13, pooled financing made up about 37% of public health sector financing55 (Table 6.3). Table 6.3: Sources of Financing in the public health sector in 2012-13 in MKW billion Total Budget Details, MK billion Sources of Financing and Proposed Budgets MK, Dev-1 billion % PE ORT (discreet) Dev-2 Sources of financing Ministry of Finance 22.7 63% 13.8 6.3 2.6 External Pooled Financing 13.1 37% 11.9 1.2 Total Financing, MK billion 35.8 100% 13.8 18.2 1.2 2.6 Total Financing , % 39% 51% 3% 7% Source: Department of Budget at MOF, and Department of Budget and Planning at MOH (May 2012). Note: MOF estimates have been taken for external pooled financing. ORT is mission estimates, and also includes within this the procurement of medical equipment in the amount of MK 1.2 billion. 260. The situation in the public health sector continues to remain constrained post-2008-09. After the initial injection of public resources between 2006-07 and 2008-09: (a) public health spending (in nominal and real terms) starts to stabilize from 2008-09 onwards, (b) public health spending as share of total government spending is more stable during this period compared to the volatility it experienced in the previous 5-years (2002-03 and 2006-07), and (c) the shares of personnel emoluments (PE) in public health spending are growing at the cost of other budget items. There are three distinct features during this period of analysis: (i) Phase 1: 2006-07 to 2008-09 which saw increased public spending for health (from MKW 15 billion in 2006-07 to MKW 30 billion in 2008- 09 in nominal terms), (ii) Phase 2: 2008-09 to 2010-11 which saw stabilizing public spending for health in real terms due to a higher inflation56, and (iii) Phase 3: 2011-12 where public spending for health fell as a result of the financial crisis. In 2011-12, the public sector revised budgets for health amounted to MKW 30 billion, down from MKW 35 billion in the previous year. In per capita real terms, it declined to almost 2006-07 levels. The 2012-13 health budgets suggest that allocations in nominal terms are starting to increase (Figures 6.3). 55 World Bank estimates. Budget support mission. May 2012. 56 The inflation in consumer prices between 2008-09 and 2010-11 ranged from 8.7 percent to 7.6 percent. The global financial crisis in 2008-09, and the resulting devaluation of the (real) Malawian Kwacha from MKW 98.2 to 105.8 to 1 USD, has contributed to a decline in public commitment to health. 94 Figure 6.3: Trend in public health spending, in millions of MKW (a) Nominal and real terms (b) In millions of MKW Trend in public health spending real and nominal in Trend in public health spending (MOH + million MKW LGFC) in millions of MKW 40,000 40,000 30,000 30,000 20,000 20,000 10,000 10,000 - 2006-07 2008-09 2010-2011 - Nominal health spending in millions of MKW (310, LGFC) Real health spending in millions of MKW Total LGFC spending (900 series) (310, LGFC) Total MOH spending (310) Source: MOF IFMIS system. Notes: These figures are taken from revised budgets, and they are not final audited expenditures. It includes vote 310 for MOH and vote series 900 for Local Government and Councils (LGC), as well as pooled external basket funding through the Health SWAp partners via the MOF to MOH. 261. Execution of donor pledges was much affected following the global financial crisis. In 2011-12 only 25% of pledges were released, down from half of it being released the year before. This adversely impacted MOH budget post 2009-10, especially in the areas of capital investments. Mid-year is often a challenge to re-allocate resources within sectors, and MOF has primarily responded to resource needs for PE but not much for others (Figure 6.4). Figure 6.4: Pool Donors’ Pledges vs. Actuals (Without Global Fund) in USD millions, 2008-2012 Pool Donors' Pledges vs. Actuals (Without Global Fund) 2008-2012 Pledges Actuals $70 -7% $60 -15% $50 $ Millions $40 -46% $30 -72% $20 $10 $0 2008-09 2009-10 2010-11 2011-12 Fiscal Year Source: MOH and CHAI. 2012. Resource Mapping Exercise Spending patterns 262. Most of the public funds - tax financed and external pooled financing - flow through the Ministry of Health (75%), followed by the local government counsel (25%). Generally, there has been a consistent pattern of financing budget items between domestic and pooled external sources. For 95 example, domestic financing sources pay towards PE, ORT and capital investment; while external pooled financing pay primarily towards ORT and capital investment. In some years before 2012-13, external financing also supported SWAp allowance under PE. In 2012-13, it also supported drugs emergency kits, which was traditionally financed by domestic financing through MOH and LGC. 263. There is clear separation on what budget items are covered between the MOH headquarters and the other agencies and levels. MOH takes on the stewardship and management role. MOH headquarters finances PE for all of public health sector, and also finances some public health goods, such as vaccines. Additionally, MOH headquarters manages the larger capital investment projects, both externally and domestically financed. The responsibility for ORT is shared equally between MOH and its agencies and the LGC. Over time, there is a shift of responsibility for operations to LGC, which use about 35% of this ORT for drugs and supplies, and 65% for operations costs. 264. Public health spending shares have been shifting in favor of PE. In 2006-07, personnel emoluments (PE) was 1/4 of public health spending, while in 2011-12 it had increased to over 1/3 of public health spending and creeping to almost 1/2 of public health spending by 2012-13. Moreover, in 2012-13, among personnel, 30% of PE went towards CHAM staff and new recruits, and 35% went towards SWAp allowance. The increase in share for PE, which was mainly a means to retain staff and recruit new personnel, puts pressures on the budgets for ORT and capital investment. There was an effort in 2007-08 to increase ORT shares but this was at the cost of capital investments. Over time, while ORT shares have more or less remained constant, capital investment shares have declined significantly from almost one-fifth of public health spending in 2006-07 to one-tenth in 2011-12. This is primarily resulting from a declining contribution from Part 1 development spending in pooled external financing. However, this is picked up as discreet projects by development partners, i.e. the overall capital investment financing in the country may have not declined, but moved away from pooled financing through budget. Further investigation might substantiate the situation for capital investment needs and spending in the health sector (Figure 6.5 and 6.6). Figure 6.5: Distribution of spending for PE, ORT, and Development (MOH and LGFC) (a) Spending in MKW million (b) Spending in shares, % 25,000 100% 5% 1% 1% 5% 8% 14% 9% 6% 14% 22% 7% 20,000 80% 60% 57% 52% 15,000 55% 66% 62% 56% 40% 10,000 20% 32% 40% 5,000 26% 19% 21% 22% 0% - 2006-07 2008-09 2010-2011 PE ORT Total Dev PE ORT Part 1 Dev Part 2 Dev Source: MOF IFMIS System, WB/PER team estimates. Notes: The development expenditures are broken down as Part 1 and Part 2. Part 1 is financed by the pooled external basket funding through the SWAp, while Part 2 is domestically financed. 96 Figure 6.6: ORT spending in health, 2006-07 to 2011-12 (a) In MKW millions (b) in USD millions ORT expenditures (310+LGC) in millions MKW ORT Expenditure (310+LGC) in millions USD 25,000 200.00 180.00 20,000 160.00 140.00 15,000 120.00 100.00 10,000 80.00 60.00 5,000 40.00 20.00 - - Nominal ORT (MKW) Real ORT (MKW) Nominal ORT (USD) Real ORT (USD) Source: MOF IFMIS System, WB/PER team estimates. 265. Operations costs are among the largest shares within ORT. Overtime, a significant increase in operations cost and a steady decline in drugs costs are observed. The increase in operation costs may be attributed to procurement of small medical equipment being categorized under operations. Travel budget shares have increased over time while training budget shares have declined (Figure 6.7). Figure 6.7: Distribution of ORT costs only MOH and LGC in percent 70% 58% 60% 46% 47% 48% 49% 50% 40% 37% 36% 37% 40% 31% 31% 30% 15% 17% 18% 20% 14% 13% 12% 12% 7%10% 10%8% 10% 4% 2% 0% 2006-07 2007-08 2008-09 2009-2010 2010-2011 2011-2012 Drugs Operations Travel Training Source: MOF IFMIS system, WB/PER team estimates. 266. As most public health commodities are externally financed, Malawi is exposed to high dependence and vulnerability to external shocks. Donors have paid special attention during the financial crisis to ensuring that public health commodities were not adversely affected. The Figure below displays the extent to which Malawi depends on its donor partners for commodities (Figure 6.8). 97 Figure 6.8: Proportion of commodities funded by Government and Donors in FY 2012-13 Neglected Tropical Diseases* $0.005 M $0.004 M Tuberculosis $0.3 M $1.1 M Nutrition $0.4 M $6.5 M Malaria $0.9 M $22.1 M HIV Including STIs $0.2 M $83.0 M Vaccines $0.005 M $27.4 M Family Planning $0 $6.0 M 0% 20% 40% 60% 80% 100% Government (excluding pool) % of total commodities funding Donor Source: MOH and CHAI. 2012. Resource Mapping Exercise Notes: * Some additional donated commodities for Neglected Tropical Diseases were not captured in Resource Mapping Notes: Nutrition funding also includes Nutrition associated with HIV funding; TB funding also includes TB HIV 267. Over time, donors are moving capital investment projects from pooled under government budgets to discreet forms managed outside government budgets. Development expenditures are showing a downward trend after having peaked at slightly over MKW 7 billion in 2008-09, and have declined to around similar levels as in 2006-07. Since 2009-10, part 1 or donor funded capital projects have moved away from pooled funds under the SWAp to discreet projects. In 2009-10, The Global Fund, one of the major AIDS financier, resigned itself from the SWAp pool and did not sign the second SWAp arrangement. It now grants money directly to the National AIDS Commission, its principal recipient. 268. Limited release of externally financed pledges in 2011-12 affected MOH budget releases, and especially hurt capital investment projects. Development partners also limited funding capital projects through MOH following an inquiry into alleged misappropriation of funding57 (Figure 6.9). 57 http://www.theglobalfund.org/en/oig/reports/ (Audit of Global Fund Grants to the Republic of Malawi – GF OIG – 10-009, August 3, 2012 – paragraph 31 and 32). 98 Figure 6.9: MOH Original and Revised Budget for FY 2011-12 $100 Approved Revised -18% $50 6% $ Millions -56% $0 ORT PE Development Source: CHAI and MOH Resource Mapping Exercise 6.2.2 Allocative and Technical Efficiency Allocative Efficiency 269. There is evidence of allocative inefficiencies in Malawi’s health system. ORT budget has been volatile between curative versus preventive health care. While the earlier years saw a significant support for preventive care (i.e. an increase from less than 10% of MOH budget in 2006-07 to over 20% by 2008-09), the later years boosted spending for curative care. It is noteworthy that despite the renewed focus on Essential Health Package (EHP) through the Health Sector Strategic Plan (HSSP) - which extensively incorporates preventive interventions - its funding for preventive care declined after 2008-09 and is plateauing out in more recent years (Figure 6.10). Figure 6.10: Trend of expenditures financed by public health sector by sub-programs58 40% 20% 0% 2006-07 2007-08 2008-09 2009-2010 2011-2012 Other services Preventive Health Services Curative Health Services Source: MOF IFMIS System – WB/PER team estimates Note: The figure includes expenditures incurred at different cost centers (including central hospitals and DHOs). 270. Geographical differentials in ORT spending are not explained by economic or health needs, and they could be better allocated based on needs. Investment in the districts varies widely and does not necessarily reflect the burden of diseases or the epidemiological profile. The latest NHA report flags the increasing divergent health expenditures per capita by region. (Figure 6.11). 58 Other services in figure below include environmental health, ambulatory services, nutrition and HIV/AIDS services financed under the health sector, through MOH and LGC. Although note that, nutrition and HIV/AIDS services are financed by the MOF through a separate vote 094. The resources for these however go to other ministries and agencies (e.g. Malawi AIDS Commission). 99 Figure 6.11: Regional health expenditure patterns (a) Per capita health spending by regions (b) Absolute health spending by districts (2008-09) (2012-13) $9.00 North $8.00 Center $7.00 South $6.00 $5.00 $4.00 $3.00 $2.00 $1.00 $0.00 1995/96 1998/99 2004/05 2006/07 2008/09 Source (a): Ministry of Health, 2012. Malawi National Heath Accounts with subaccounts for HIV/AIDS, Malaria, Reproductive Health, and Child Health for Financial Years 2006/07, 2007/08, and 2008/09. Department of Health Planning and Policy Development, Lilongwe, Malawi. Source (b): MOH-CHAI Resource Mapping Exercise. 271. The poor have limited access to health facilities relative to the non-poor. Malawi has a comprehensive network of health facilities, including 5 central hospitals that function as tertiary care facilities. However, these facilities, particularly at the community level are often poorly equipped, lack appropriate staffing levels, and drugs. While one in four rural community has a facility within 1 km, it is poorly served by health facilities with doctors. Over half of all communities in Malawi are located more than 3 km from the closest health Clinic. A clear inverse relationship is observed between the level of community wealth and distance to the closest health facilities, implying wealthier communities are better served (Figure 6.12). Besides, private outpatient care is mostly used by individuals in better-off households than by those in poorer households, who use relatively more of public outpatient care. 100 Figure 6.12: Distance to Health Clinic (w/or without Doctor) (% communities by wealth) 80 % of Communities 70 60 50 40 30 20 10 0 More than 3 Km from the Within the Community Within 3 Km of the Community Community Poorest Communities 18.3 11.1 70.6 Q2 20.8 13 66.2 Q3 26.1 9.8 64.1 Q4 35.1 9.1 55.8 Richest Communities 56.9 19 24.1 Source: Integrated Household Survey 3 272. Significant by-passes and inappropriate referrals from clinics/hospitals to higher level hospitals add to the inefficiencies in the system. Self-referrals seem to be a common practice in Malawi, and constituted close to 80 % of cases seen at the Kamuzu Central Hospital (KCH in Lilongwe). Such high level of self-referrals could be attributed to lack of a streamlined system and perceived low quality of the lower level facilities. Furthermore, as low as 60% of cases were appropriately referred to KCH for care, suggesting poor diagnostics, or limited access to equipment and/or drugs, or limited incentives to retain patients at lower level facilities, or several of the above (Chfundo Rose Kachiza, 2005). Technical Efficiency 273. There is evidence of technical inefficiencies in Malawi’s health system. High staff vacancy is reported in lower level facilities and among positions for physicians. Though the production rate for nurses seems reasonable, the one for physicians is very low at 18 new graduated physicians per year. Staff performance, staff distribution to rural areas, and incentives to improve staff retention in the public sector need to be addressed. While many public hospitals have the appropriate basic maternal and child health package to provide pregnancy care and deliveries, many clinics (public and/or private) do not. 274. Quality of services was noted to be worse in lower relative to higher level health facilities. Limited access to drugs especially at lower facilities continues to plague the health facilities, as this was also highlighted in the previous PER for Malawi (2007). This has detrimental effects as patients, who perceive access to drugs as a quality indicator, bypass lower level facilities in favor of higher level ones (e.g. hospitals). This creates unnecessary burden at the hospital levels for services that can be cheaper and easily provided at clinic level. 275. Low bed occupancy rate and low bed turnover are significant in public hospitals, especially below the tertiary level (central hospitals). The findings clearly indicate the presence of excess bed capacity given the current level of utilization. This however does not imply an excessive capacity relative to need, since the bed density in Malawi is far lower than that recommended for the size of the population. There may be demand-side barriers of any type (e.g. financial, 101 geographical, or cultural) that negatively influence the utilization of hospital services. Evidence indicates that capacity utilization is better in public primary level hospitals compared with the non- public ones because public hospitals have more resources in terms of staff and beds. Given the low bed density in the country, it is not desirable to reduce the number of hospital beds. However, until the demand-creating interventions bear the desired behavioral change, innovative ways of using the existing relative excess capacity need to be explored (Ayoub et. al., 2012). 6.2.3 Intra-sectoral equity/Benefit Incidence Analysis 276. Malawi households do not show high burden from catastrophic health expenditures. Overall health spending is reasonable, and about 2.5% households spend over 10% of their household expenditures on health. This figure is relatively low compared to Bangladesh and Vietnam, where about 15% are above this threshold (Van Doorslaer et al., 2007), but higher than Ghana where only 1.4% spends more than 10% of the household total expenditure (World Bank, 2012) (Table 6.4). This could be explained by two factors: (i) health service utilization is low, and (ii) the essential health care package that provided free health care has helped reduce household expenditures. Table 6.4: Catastrophic Expenditure Headcounts in Malawi, 2010-11 Consumption Percent of total household spending allocated to health Quintile 5% 10% 15% 25% 40% Households spending above threshold (%) Lowest 7.0 2.4 1.0 0.3 0.2 Second 6.6 2.0 0.7 0.3 0.0 Middle 7.8 2.6 1.0 0.4 0.1 Fourth 6.8 2.9 1.5 0.5 0.0 Highest 6.2 2.5 1.1 0.5 0.0 Total 6.9 2.5 1.1 0.4 0.1 Source: Malawi IHS3 Survey 277. Many poor remain vulnerable to health spending shocks. In 2010-11, 25% of the population fell below the poverty line due to illness shocks. More ultra-poor households were vulnerable to health expenditure shocks than the poor households. Household spending patterns differ by income quintile, and the poor are much more burdened when faced with hospitalization and illness requiring drugs (Figure 6.13). 102 Figure 6.13: Health spending as share of total household expenditure (a) Spending in Drugs (b) Spending in Hospitalization Source: Malawi IHS3 Survey 278. Public subsidies slightly benefit the poor over the non-poor. Government subsidy in health (curative care) is progressive in each area - the poorest households capture a greater share. Poorest urban individuals (bottom 20%) capture the greatest share, while the richest 20% (that also use private care and put down more out of pocket) get only 10% of the net subsidy. There is a more balance distribution in rural areas. The Southern rural region (the poorest) exhibits the more pronounced progressive pattern of distribution of health expenditures benefits. It is likely that as observed in the PER 2007, the equitable distribution of benefits is due to the increased emphasis on the free essential health services package rolled out by the MOH (Figure 6.14). Figure 6.14: Distribution of Government Subsidy by Wealth, within Rural and Urban areas, 2011 30 Share of Subsidy (%) 25 20 15 10 5 0 Poor Riche Poor Riche Q2 Q3 Q4 Q2 Q3 Q4 est st est st Urban Areas Rural Areas Share of Subsidy (%) 28 26 18 19 9 0 21 22 20 20 17 Source: Malawi IHS3 Survey 279. There is a relatively well balanced utilization of outpatient care across income groups, and poorer and middle income households seem to use more public outpatient care than the richest. Private outpatient care is mostly used by individuals in better-off households than by those in poorer households, who use relatively more of public outpatient care. The Central region (the poorest of the three regions) exhibits a clear progressive benefit incidence path (Figure 6.15a). 103 280. Utilization of inpatient care at public facilities is progressive in urban areas and slightly biased towards the richer in rural areas. More females in poorer households use inpatient care compared to those in richer households. Besides, utilization of inpatient care among males is regressive. There is also greater disparity in access between the poorest and the richest in Central Hospitals than in District Hospitals (Figure 6.15b). Figure 6.15: Use of health services by income quintile (a) Where do poor go for outpatient care (b) Where do poor go for inpatient care 30 35 25 30 20 25 % among users 20 15 15 % among users 10 10 5 5 0 0 Ric Ric Poo Rich Poo Rich Poo Poo Q2 Q3 Q4 Q2 Q3 Q4 Q2 Q3 Q4 hes Q2 Q3 Q4 hes rest est rest est rest rest t t Government Private Providers District Hospital Central Hospital Faciities % among users 9 15 22 27 28 20 24 21 20 15 % among users 15 24 17 22 21 10 23 24 14 29 Quintiles of per capita expenditure Quintiles of per capita expenditure Source: Malawi IHS3 Survey, 2011 6.3 Health Sector Performance and Key Issues Health Outcomes 281. Mixed results are observed with respect to meeting the millennium development goals (MDG) health targets. Contrary to MDG 5 (Improve Maternal Health) 59, MDG 4 (Reduce Child Mortality) and MDG 6 (Combat HIV/AIDS, Malaria, and other diseases) are on track and likely to be met. In a global comparison between other countries with similar income and health spending, Malawi fares poorly except for infant mortality where it fares better than average. 282. Malawi’s life expectancy at birth is lower than those of countries with similar income and health spending. After concerted effort to controlling HIV/AIDS and large influx of investments by donors like the Global Fund, World Bank, and PEPFAR60, life expectancy at birth increased from 39 years in 2000 to 57.5 years in 200961 (Figure 6.16). 59 http://www.undp.org/content/dam/undp/library/MDG/english/MDG%20Country%20Reports/Malawi/MDGsProgr essAtaGrance-2010.pdf. Accessed on 3.1.13 60 The US President’s Emergency Plan for AIDS relief (PEPFAR). http://www.pepfar.gov/about/ 61 http://www.who.int/countries/mwi/en/, accessed on 4.8.2013 104 Figure 6.16: Global comparison of life expectancy relative to income and health spending, 2009 Global Comparisons of Life Expectancy Relative to Income and Spending, 2009 Better than average Mauritius Tanzania Malawi Mozambique Worse than average Zambia Worse than average Better than average Performance relative to income per capita Source: World Development Indicators (2010), WHO (2010), & Royal Monetary Authority (2009) Note: both axes log scale 283. Malawi’s infant and under-5 mortality rates are above average of those countries with similar income and health spending. Infant mortality rates show an impressive improvement from 134 per 1,000 live births in 1992 to only 66 in 201062. Under-5 mortality rates have equally declined from 234 per 1,000 live births to 112 in the same time period, notably thanks to a strong immunization program. Therefore, Malawi is well on its track to meet MDG 4 (Figure 6.17). Figure 6.17: Infant and Under-5 Mortality Rate (a) International comparison, 20009 (b) Under-5 Mortality rate over time 300 Deaths per 1000 live births Worse than average 250 Swaziland 200 Mozambique Malawi Mauritius 150 Tanzania Better than average Tanzania 100 50 0 Better than average Worse than average Performance relative to income 1992 2000 2004 2010 Source (b): MDHS, 2010 284. Malawi’s maternal mortality rate is below average of those countries with similar income and similar health spending. Maternal mortality ratio has improved over time (1120 in 2000, 984 in 2004, and is 687 per 100,000 live births in 201063), and is closer to average of those countries with similar income and health spending. It is still among one of the highest rates in the world (Figure 6.18). 62 Malawi DHS 2010 63 Malawi DHS 2010 105 Figure 6.18: Maternal Mortality Rate (a) International comparison, 2009 (b) Birth deliveries at health facilities over time 80 Worse than average 70 Swaziland 60 Mozambique 50 Malawi Mauritius Tanzania 40 Better than average Tanzania 30 20 10 Better than average Worse than average 0 Performance relative to income 1992 2000 2004 2010 Source (b): MDHS, 2010. 285. Malawi has shown significant traction in controlling the spread of infectious diseases, but more is required for HIV, TB, and Malaria. Number of Malaria deaths declined very little between 2008 and 2010, from 8206 to 804864. However, there is a reduction of deaths by Malaria in children under 5 (the number halved from 14,000 in 2000 to 7,600 in 2010). The HIV epidemic has one of the highest prevalence rate, with 10 % among adults and down from 13.8% in 2001.65 The TB incidence per 100,000 population per year is down by almost a third from 301 in 2007 to 191 in 2011.66 Health Outputs 286. Enhanced spending and programmatic focus on the most cost-effective interventions likely led Malawi to improve its child health outcomes. Good outcomes for child health can be partly attributed to Malawi’s robust immunization program. The recent 2010 DHS report estimates the national coverage rate at 81% for children aged 12-23 months. Only 8 out of the 27 districts are below 80% coverage. 287. Improved information and access to affordable basic services have led children to seek care quicker and more appropriately. About 62 % of the children sought some treatment for fever (possibly Malaria) and diarrhea, compared to 20% and 36% respectively in 2004 67. 288. Despite universal coverage in antenatal care, and significant birth deliveries in health facilities, maternal mortality remains high. This is a consequence of limited access to services and the poor quality of care provided at health facilities. 64 www.who.int/data - Global Health Observatory Data Repository, accessed on 2.27.13 65 www.who.int/data - Global Health Observatory Data Repository, accessed on 2.27.13 66 www.who.int/data - Global Health Observatory Data Repository, accessed on 2.27.13 67 Malawi DHS 2004 106 289. Limited geographical and financial access to quality health services are among the key reasons why women are unable to use services. These problems are more pronounced for those in lower income quintiles. In order to meet the MDG 5 targets, Malawi needs to address these challenges using both supply side and demand side mechanisms (Figure 6.19). Figure 6.19: Reasons why women are unable to access health care services (a) Top reasons women are constraint (b) Women facing challenges by quintiles No provider available 100 80 Percentage Getting money for treatment 60 Access to transport to health 40 facility 20 Distrance to health facility 0 No drugs available 0 20 40 60 80 Source: MDHS 2011 290. The following summarizes the complex picture of the key sector expenditure and performance issues. x To meet the Abuja target and achieve the health outcomes highlighted in the Health Sector Strategic Plan (HSSP), innovative approach to increase domestic spending on health is necessary. As the MOF structural restrictions on public spending will further limit spending, innovative public-private partnerships for service delivery and an improvement in efficiency at every step from planning to implementation will be needed to increase the fiscal space in the health sector. x Excessive reliance on donor support will continue for the next 5 years. Almost 50% of HSSP is financed through the SWAp, however in the longer term it is not a sustainable option. The country should already start considering options to increase public financing of health and strategize within its medium term development plan with projections to meet the Abuja target. x Inequities in access to health care services and health outputs and outcomes persist depending on gender, urban rural, region, or level of income, and in the absence of any risk pooling mechanisms. About 13 districts fall below the national average for deliveries in health facilities. x The quality and capacity of the service delivery also vary significantly across the country. This is a function of varying degrees of investment in each region. This maldistribution of inputs into the system contributes to the inequity in health outcomes. x Allocative efficiency is low when districts are not financed based on the burden of diseases and their epidemiological profile. Also, despite the increase in burden of disease from non-communicable diseases, they continue to be underfunded. There is very limited initiative to prevent and/or diagnose these diseases at early stage as well as prevent them from becoming a chronic high cost issue. 107 x Extremely fragmented and parallel pharmaceutical supply management systems add a substantial cost to procurement of drugs. x Very limited capacity for pre-service training to increase the production of health workers, especially for physicians, is an issue. Training on the job is not sufficient and interrupts service delivery substantially. Performance orientation may need to be incentivized. 6.4 Policy Priorities or Suggestions for Reform 291. Some key priorities for Health can be framed and more specific policy recommendations are presented in Table 6.5. x Continue to support goals for universal coverage on the Essential Health Package (EHP). The overall goal of the Health Sector Strategic Plan (HSSP, 2011-2016) is to improve the quality of life of all the people of Malawi by reducing the risk of ill health and the occurrence of premature deaths. Malawi has introduced a program on universal coverage on health and identified an Essential Health Package (EHP). Some of the critical areas of interventions in EHP include (a) maternal and child health, including family planning, (b) communicable disease control, including EPI, HIV/AIDS, Malaria and TB, (c) nutrition: therapeutic feeding and community based interventions, and (d) some key non-communicable diseases, reflecting burden of disease. EHP should remain the priority in 2012/13. The HSSP (2011-2016) estimated the EHP/package costs at about $3.2 billion over 5 years (i.e. $43 per capita per year for a population of 14.9 million) , which are significant, and MOH currently does not have sufficient funds to ensure EHP for all. MOH therefore plans to prioritize certain activities such as that for maternal and child health and communicable disease control. x Consideration should be given to targeted interventions focusing on the poor and the underserved. As budgets are limited, targeted interventions to ensure that the disparities in health outputs do not grow should be considered. As households are likely to be acutely affected by the economic crisis, it will be necessary to ensure resources are used in an efficient and effective manner towards high value interventions, such as highly cost-effective ones. Also, outreach services including mobile clinics to reach out to the population, especially to those who have difficulty accessing services, should be supported. x Public subsidies should be re-directed to ensure the benefits are more towards the poor and the underserved. 108 Table 6.5: Policy Recommendations Institutions to be Issues to Address Actions to Take responsible/coordinated Short Term Analyze efficiency impact of substitution between PE and ORT. Prioritize frontline workers, and incentivize distribution to deprived Within recurrent health areas. spending, higher share is Potential savings are to achieve from refining the eligibility criteria being taken up by PE. for students to receive subsidies for pre-service education and for MOH/MOF (in coosultation Efficiency gains in recurrent overseas training and from refining the eligibility for subsidies of the with MOEST). (both PE and ORT) and overseas medical treatment. development expenditures are Rationalize equipment favoring cost effective interventions: clinics, to be improved. labs, district hospitals, emergency and first aid. No new domestically financed infrastructure in the short-term while reviewing and rationalizing overall capital investment. Medium Term Patients unable to afford transport costs and have Develop public-private partnership (PPP) policy to improve health limited access to health service performance and access, especially for the underserved and MOH/MOH PPP services in rural/deprived poor. Unit/MOF/Local areas. Continue to partner with non-public sector: CHAM, NGOs. governments/and Ministry Patients unable to afford Strengthen community level access and provide outreach/mobile of Social Welfare and EPD medical treatment costs clinics. and MDAs with vested (services/drugs). Base partnership on performance based and targeted contracts. mandates in social Challenges in identification Consider analysis on who is benefiting from MOH-CHAM service protection. of the poor and the level agreement (SLA) by quintiles. vulnerable. HR issues in the sector: High MOH has drafted a HRH strategy (2012-16), that should be attrition levels, skewed implmented under the EHP prioirties and within the fiscal space for distribution in favor of urban health. areas, and subsidization with MOH in coordination with Elaborate an HRH (public, CHAM and others) assessment study for public monies of tertiary other stakeholders. a benchmarking and baseline situation analysis in which high education for all. Low vacancies, urban bias, accountability mechanisms, and productivity performance and low and performance improvements should be addressed. Continue productivity. Limited HRH strengthening HRH database including public and private sectors. information. Prioritize the pharmaceutical supply management reform. Innovative schemes can be considered by partnering with private sector (e.g. Essential drug shortages. MOH and DPs. insourcing private pharmacies into public hospitals; agreeing on lower mark-ups, etc.). Assess feasibility. Revise urgently a capital investment masterplan based on needs and medium term objective linking to UHC which could be rationalized and directed to underserved regions, and clinics/district hospitals. Infrastructure and Equipment Rationalize equipment investment favoring cost effective MOH and EPD misalignment. interventions. Do inventory, replacement and service plan; Institutionalize; Decentralize assessment and databases. 109 Engage negotiations between MOH and MOF to increase the share of domestic spending on health, given the volatility and uncertainty The Abuja target is currently of the level of donor support for the sector over the medium-long missed. terms. Low revenue base and Hospital reform strategy may identify areas of revenue generation retention policy. MOF, MOH, LGFC, MOF and areas of efficiency gains. Inequitable distribution of Revenue retention can help cross subsidize public health programs, funding across country. etc. Establish a progressive allocative formula that takes into consideration equity and needs. 110 CHAPTER 7. SOCIAL PROTECTION 7.1 Introduction 292. This chapter reviews the Public Expenditures for Social Protection (SP) in Malawi in order to assist policy makers in decision making and evaluating whether social protection expenditures are sufficient and appropriate and delivery costs justified for achieving productive human development outcomes. The chapter examines whether these expenditures are: (i) sufficient to cover the needs of the vulnerable; (ii) targeted in order to reach those most in need; (iii) effective and productive so that they improve lives and livelihoods of the recipients; (iv) efficient so that the costs of delivery and funds flow are justified for the end outcomes; and (v) coordinated so that the funds flow and agencies involved allow for timely, transparent, accountable delivery of funds with minimal risk of duplication. 7.2 Assessment of Social Protection Expenditures 7.2.1 Level of Expenditures 293. Social protection is critically important area of the public policy, given that 84% of the Malawi population is poor and vulnerable. This chapter focuses on those expenditures considered as safety net benefits and earmarked as social protection (Table 7.1 and Annex A12). 294. Social Protection expenditures, excluding FISP, total only MK 12.8 billion (US 38.4 million) in 2012/13, which amounts to 2.6% of total government expenditures (MK 476 billion) or 1.1% of GDP (MK 1,183 billion). By comparison, the Farm Input Subsidy Program (FISP) represents 4.6% of GDP or 11.5% of the total budget expenditures68. 295. Agriculture expenditures on the FISP are defined as an itemized line in the category of Social Expenditures in the Budget documents of the Government of Malawi. Indeed, over half the population lives in poverty on one of lowest GDP per capita income (around US $268) in the world. Of the 3.4 million ultra-poor in Malawi, 96% live and depend on agriculture, and more specifically on tobacco and maize farming. Hence, government interventions in agriculture are often defined as social protection. A burning issue lies in whether expenditures on fertilizer subsidies to the poorest and landless farmers are justified as productive social protection measures or whether these expenditures are ineffective and disproportionate to the needs of human development investments for the poorest segments of the population via other direct social protection transfers. 296. The FISP amounts to MK 54,904.5 billion in 2012/13, which represents more than 60% of all agriculture expenditures. By contrast, in 2012/13 the total expenditures of the Ministry of Gender, Children and Social Welfare (MoGCSW), the Ministry of Disabilities and Elderly (MOE) and the Local Development Fund (LDF) taken together for the same period totaled MK 15.06 billion (US$38.4 million) or only 27% of FISP. Out of these MK 15.06 billion, less than half was spent on social protection (safety net transfers). It should be noted that the LDF was tasked to construct teachers’ houses and school blocks because of their comparative advantage in working directly with Councils and using decentralized structures when putting in place such infrastructure (Tables 7.1 and 7.2). 68 Another glaring contrast is the GOM’s travel expenditures amounted to 3.9% of GDP or 11.4% of the 2010 national budget (Reference in Chapter 1). 111 Table 7.1: Social Expenditures by Government of Malawi in 2012/13 (MK billions) Expenditures Institutions Salaries Recurrent Development Total 2012/13 Costs Ministry of 3.0 9.0 54.904 FISP Agriculture Cash Transfers 1.04 12.74 13.784 Local for PWs, Development Village Savings Fund and Loans. Ministry of 0.518. 0.169 0.295 0.983 SCT, Village Gender, savings and Children and loans. Social Welfare Ministry of 0.5340 0.158 0.81 0.293. Cash Transfers Disabilities Government 16.0 Public Pension GOM Scheme 297. Social Protection expenditures also involve pensions, which in 2012/13 amount to MK 16 billion for about 30,000 government civil servant retirees (Tables 7.1 and 7.2) . This is by contrast more than the total expenditures on safety nets for the poor and vulnerable in the form of social cash transfers, cash transfers for Public Works (PWs), and school meals for the same period (MK 15.059 billion). Pensions are projected to reach MK 21.0 billion in 2015. In Malawi, pensions in public expenditures involve civil servants and other Government retirees. Eligibility for the Government Pension fund is based on reaching the mandatory retirement age of 60 years with a minimum service of 10 years, or by voluntary retirement after 20 years of pensionable service, or with the consent of the Minister of the Public Service with a minimum of 10 years of service at age 45. Early retirement is also granted on special terms for those who retire early on grounds of redundancy or ill health.69 Pensions are paid out of a consolidated fund, from recurrent expenditure. 70 Information on the proportion of recurrent expenditure that goes to the Government Public Pension Scheme was not available. 7.2.2 Allocative and Technical Efficiency 298. Social protection expenditures resulting in benefits to the poorest population are insufficient and limited. The cash transfer in Public works amounts to US$48 annually to approximately 500,000 households and the social cash transfer which covers 30,000 households amounts to US$108 annually. The average benefit per year for the poorest is therefore approximately US$75. The Third Integrated Household Survey (IHS3) shows that, while 14.8% of the population in 69 Evaluation of retirement systems of countries within the Southern African Development Community, Country Profile: Malawi. Oxford Policy Management 2010 70 Ibid. 112 Malawi benefit from school feeding program, food or cash for work programs directly engage approximately 2% of the population in Malawi71. 299. Malawi’s public spending on SP, excluding FISP, represents 1.1% of GDP, whereas overall in Africa, it averages 2.8% percent of GDP (excluding health expenditures)72. This spending of 1.1% of GDP on SP is half the African region’s average and a fifth of the 5.7% world average73. However, there is an increase from 2.6% to 14.2% of total expenditures (equivalent to 5.7% of GDP) when the FISP allocation of MK 54.904 billion is included74. 300. The inclusion of FISP expenditures as Social Protection masks the lack of funds for social protection in the form of other benefits and safety nets. As discussed in I.1, the facts that (i) expenditures on fertilizer crowd out other agriculture expenditures, (ii) FISP is defined as a line item under Social Expenditures in the Budget of GOM, (iii) FISP expenditures are dwarfing those for SP, and (iv) pensions for government civil servant retirees equal the entire social protection expenditures for 3 million people, all compound the SP expenditures misallocation. Table 7.2: Comparison of SP Expenditures with FSIP and GOM Travel and Pensions Expenditures Expenditure (MK billions) % of Budget % of GDP FISP 54.904 11.5% 4.6% Travel Costs of 46.137 11.4% 3.9% GOM (2010 data) Pensions (2012/13 16.0 3.36% 1.35% data) Safety Nets Targeted to the Poor and Vulnerable(Cash 12.8 2.6% 1.0% Transfers, Public Works) 301. Social Protection expenditures analysis shows that the sector is inefficient. While non FISP safety net programs reach the poor and even though 50% of the population is below the poverty line, the expenditures only reach 19% of the poor and 17% of non-poor. 302. The costs of delivery range from 8% for MASAF75 to 19% for the Social Cash Transfers (SCT)76. These costs tend to be high during program set up (sunk costs) and may reduce over time thanks to 71 While MASAF public works reach 1.6 million people - which alone represents over 11% of the population -, the IH3 indicates 2% of the population. This is probably due to the differences between the seasonality of the work programs and the data collection period. 72 It should however be noted that many health outcomes relate to SP interventions, e.g. the ability to purchase food, medicine, and have access to clean water and other infrastructures or livelihoods. 73 WB: Africa Social Protection Strategy 2012 74 These figures exclude an allocation of MK 1 billion to the Malawi Rural Development Fund (MARDEF) and the Youth Enterprise Development Fund (YERDEF), whose targeting criteria are considered out of line with social protection principles. 75 Malawi Social Action Fund, which is the largest nationwide safety nets public works program 113 the piloting of e-transfers, proposed introduction of unified registries, ID cards and monitoring systems. FISP delivery costs in transportation represent MK 9 billion and staff costs are MK 3 billion out of the total costs of MK 68 billion for FISP in 2013/14. This amounts to a cost of 17.6% of total FISP expenditures (Table 7.3).77 Table 7.3: Overhead Costs as Percentage of Total Program Expenditures78 Overhead Costs as Percentage of Total Program Program Expenditures FISP 17.6% SCT 10%-19% MASAF 8% 303. SP benefits targeting is ineffective. The IHS3 indicates what Malawians are most vulnerable to. Food dominates Malawian household expenditures in both urban and rural areas and across income levels. In urban areas, housing represents a high cost that increases with income. SP targeting however remains elusive. While 14.8% of Malawi’s population benefit from school feeding program, progress remain to be made in terms of targeting equity. The SCT program notably targets 10% in each district regardless of district poverty profile. Much of the targeting is based on methods of community-based identification and the Malawi Vulnerability Assessment Committee’s (MVAC) report is used to determine how many beneficiaries per district should be identified. Improvements in targeting would need (i) an enhanced method of identifying and segmenting the poor and the vulnerable; (ii) to match social protection benefits with needs; (iii) to reduce administrative costs of management and delivery; (iii) to coordinate programs to avoid overlap; and (iv) to address the challenges of political interference. The poor and vulnerable are defined into 3 categories for proposed social protection interventions according to the draft National Social Support Program (Figure 7.1). 76 MASAF III Project Documents; SCT Program, KFW and Inception Report 77 IMF and GOM authorities 78 The data for other programs is unavailable. Current expenditures do not adequately cover the vulnerable 114 Figure 7.1: Social Protection interventions by poverty and vulnerable categories79 50.7% MALAWI POVERTY LINE CATEGORIES AND POTENTIAL SOCIAL THEIR SOCIAL PROTECTION PROTECTION NEEDS PROGRAMMES/ INTERVENTIONS PROTECTION AND PROMOTION x Employment Inputs subsidy Moderately x Skill building Public works programmes 26.2% Poor x Capital x Insurance programmes x Productive Assets (Social, Crop & Livestock) x Protection from x Village savings loans asset/capital erosion x Micro-credit / Micro-finance x School feeding 24.5% Ultra Poor PROMOTION x Survival Public works programmes Ultra Poor with School Feeding x Productive Assets 15.5% Labour x Employment x Cash and food for assets combined with skills building Capacity and cash for consumption/ Adult literacy training PROVISION Ultra Poor & x Survival Social cash transfers 10% x Investment in human Incapacitated School Feeding Programe capital * To be funded out of the Government of Malawi Pool Fund for Social Support 304. Social Protection programs may not be achieving their intended objectives fast enough. Data from the Malawi Integrated Household Surveys (2004/05 IHS2 and 2010/11 IHS3) show that Ultra- Poverty incidence among households remains high, while poverty and vulnerability increased in the majority of districts due to the economic crisis (Figure 7.2). 79 National Social Support Policy 115 Figure 7.2: Ultra-Poverty in TCTP Districts Ultra-Poverty Incidence in SCTP Pilot Districts Mchinji 32 30 SCTP Pilot Districts Machinga 39 38 17 Salima 25 Mangochi 44 29 Chitipa 44 30 Phalombe 42 27 0 10 20 30 40 50 Ultra-Poverty (%) IHS3 (2010/11) IHS2 (2004/5) 305. FISP dwarfs SP spending and crowds out Social Protection and agriculture expenditures. There are relatively low budgetary allocations to Social Support Programs within the overall government budget, as they range from 5.7% in 2005/06 to 13.4% in 2008/09, before falling to 7.2% in 2010/11. The FISP share of total budget is very close to these figures since it is the dominant expenditure in SP. FISP is therefore dominant in both Social Protection and Agricultural Sector budgets, representing 70% of Social Protection expenditures and 60% of the agriculture ones. Hence, FISP crowds out other more important and effective productive interventions in both sectors (Figures 7.3 and 7.4). Figure 7.3: Trends for Social Protection Allocations 16 Trends for Social Protection Allocations 14 13.4 Share of total budget (%) 12 10 8.9 8.5 8.7 8.5 8.4 8 6.9 7.2 6.9 5.7 5.6 6.1 6 4.7 4 2 0.8 0.1 0.4 0.1 0.3 0 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 Social Protection as % of Total Budget FISP as % of Total Budget Other Social Support as % of Total Budget 116 Figure 7.4: Farm Input Subsidy Program in Sector Budgets FISP in Sector Budgets Share of sector budget(%) (Agriculture and Social Protection) 100.0 95.7 96.7 65.0 99.6 80.0 98.6 89.5 67.5 60.0 74.8 68.0 58.0 50.8 40.0 47.4 20.0 0.0 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 FISP as % of Social Protection Budget FISP as % of Agriculture Budget 7.2.3 Intra-sectoral equity/Benefit Incidence Analysis 306. The main beneficiaries of FISP are not just the poorest farming households. First, according to the IHS3 data analysis for rural areas, over half of rural agricultural households in Malawi accessed FISP coupons. This has also been well balanced across rural regions, which underlines the relatively high reach of the overall program and its balanced geographical coverage (Table 7.4). Table 7.4: Access to FISP by Rural Agricultural Households Access to FISP by Sex of Head Access to FISP by Poverty Status All Share of Households Share of Households Difference Difference Households (SE) (SE) [p-value] [p-value] Male Female Diff. p-value Non-poor Poor Diff. p-value Rural Malawi 54.7 54.4 55.7 -1.3 0.37 57.2 52.3 4.9** 0.00 Rural Region North 55.2 55.6 53.6 2.0 0.61 54.3 56.0 -1.6 0.62 Center 54.4 54.5 53.9 0.6 0.81 55.4 53.2 2.2 0.29 South 54.8 53.8 57.2 -3.4 0.12 60.0 50.8 9.2** 0.00 Note: Significance level of the difference: 1% (**), 5% (*), and 10% (+). Source: Malawi IHS2 and IHS3 307. Second, there is no statistical significant difference in the rates of access to FISP by male versus female-headed households. The program therefore appears to be neutral in terms of favoring particular gender of headship status. Third, while the program is aimed at targeting the poorest rural households, this analysis indicates that in the North and Central regions there is no significant difference in the proportion of poor versus non-poor households accessing the program, i.e. poor and non-poor households are almost equally likely to receive FISP coupons. In the South region, Table 7.4 actually indicates that non-poor households are almost 10% more likely to having received FISP coupons. 117 308. By definition, for expenditures to be considered as Social Protection, they must focus on the poorest and the most vulnerable. FISP subsidies, particularly for fertilizers, would be more effective if they were targeted to farmers who have land and more agricultural potential, while other resources supporting increasing livelihoods could be targeted to the resource poor. 309. Utilization of non-FISP Social protection programs is extremely limited among households in both urban and rural areas. This level of use does not allow for a sufficiently strong assessment of equity in access. For instance, only 2.1% of households access free maize (1.3% in urban and 2.3% in rural areas). However, there is some evidence that school feeding programs - the most common non-FISP Safety net, accessed by 16% of urban and 12% of rural households - is pro-poor, particularly in rural areas. Figures show that there is a much bigger access to school feeding programs by the bottom 20% in all areas when compared to the richest quintile (Figure 7.5). Figure 7.5: Households Accessing School Feeding Programs: Malawi and Urban vs. Rural Share of Households Accessing School Feeding Programs Malawi and Urban/Rural, 2011 30 24 Share of Households 25 18 19 20 17 15 14 15 15 13 13 13 11 10 10 10 8 4 5 0 Poorest Poorest Poorest Q2 Q3 Q4 Q2 Q3 Q4 Q2 Q3 Q4 Richest Richest Richest Malawi Urban Rural Source: IHS3 Data 310. The results of the Third Malawi Social Action Fund (MASAF III) based on the Beneficiary Incidence analysis for IH3 indicate the tangible benefits received by the MASAF recipient communities in terms of access to road construction and public transports . Indeed: (i) Significantly higher proportion of villages within 2 km from paved roads among MASAF recipient communities; (ii) clearly a higher proportion of maintained roads (better access) among MASAF recipient communities, which is particularly evident and stronger for distant communities(i.e. those beyond 2 km); (iii) distant communities that receive MASAF also show a greater number of months when mini-buses and lorries can transit; and (iv) finally, distant communities that receive MASAF also have a higher likelihood of having main community roads passable all year long. 7.3 Social Protection Sector Performance and Key Issues 311. Social protection expenditures are insufficient and ineffective despite the critical needs. Malawi is ranked 171st out of 187 countries surveyed in the United Nations Human Development Index of 2011. Poverty remains widespread and concentrated in rural areas while income also remains unevenly distributed, reflecting inequalities in the access to assets, services and opportunities across the population. Vulnerability and poverty affect over 84% of the 15 million Malawians who live in the rural areas and are dependent on agriculture for their livelihood. Of the 118 3.4 million ultra-poor, 96% live and depend on agriculture, and mostly tobacco and maize farming for their livelihood. 312. The Government of Malawi has recently approved the National Social Support Policy (NSSP) and National Social Support Program Plan for implementation in March and April 2013 respectively. It is anticipated that this development will unlock resources from development partners. This may also enhance the sector’s ability to secure parliamentary support for budgetary allocations. The NSSP provides a holistic framework which supports social and economic human rights and freedoms for designing, implementing, coordinating, monitoring and evaluating social support interventions. Its goal is to reduce poverty and enable the poor to move out of poverty and vulnerability by: (i) providing welfare support to those unable to construct a viable livelihood; (ii) protecting the assets and improving the resilience of poor and vulnerable households; (iii) increasing the productive capacity and asset base of poor and vulnerable households; and (iv) establishing coherent and progressive social support synergies by ensuring strong positive linkages to influence economic and social policies and disaster risk reduction. 313. Most of the Social Protection resources are provided by multiple Development Partners (DPs) . While the GOM has made efforts to reach the poorest 10% of the population through safety nets, expenditures for social protection are almost entirely funded by the donors through the Government budget. This implies government agreement and implementation of an administration, an oversight structure, as well as bilateral programs which would also require government approval, oversight, and management. It should be noted the National Social Support Technical Committee membership involves GOM Ministries, The World Bank, Bilateral Development Partners and International NGOs.80 314. The main mechanisms for safety nets consist of: (i) Cash transfers for Public Works for labor able households, (ii) Social Cash Transfers for the Ultra Poor labor-constrained households, and (iii) School Meals Program for children. These programs are marred by fragmentation: due to the mini empires or turfs of programs, a plethora of fragmented, unsustainable, insufficient, and uncoordinated cash transfers programs end up not impacting chronic poverty. 315. Institutional fragmentation continues to impede effectiveness and efficiency in the delivery of SP benefits (Table 7.5). Social Protection in Malawi is the responsibility of the Ministry of Economic Planning and Development through its Division of Poverty Reduction and Social Protection, The Ministry of Gender, Children and Social Welfare, The Ministry for the Disabled and the Elderly and the Local Development Fund and its Technical Support Team. Current expenditures and transfer levels do not adequately cover the vulnerable. These institutions supervise and monitor the budget transfers from the Ministry of Finance to the various line agencies who are responsible for implementing the programs. These include the Ministry of Agriculture (for FISP), Malawi 80 1.Ministry of Economic Planning and Development (Chair and Secretariat); 2. Office of the President and Cabinet; 3. Ministry of Finance;4. Ministry of Agriculture and Food Security; 5. Ministry of Education; 6. Ministry of Transport and Public Works;7. Ministry of Water Development and irrigation; 8. Ministry of Local Government and Rural Development;9. Ministry of Gender, Children and Social Welfare;10. Ministry of Labour;11. Ministry of Health;12. Ministry of Information;13. Ministry of Youth and Sports;14. Ministry of Disabilities and the Elderly Affairs;15. Department of Nutrition, HIV and AIDS;16. Department of Disaster Management Affairs; 17. Malawi Social Action Fund;18. Council for Non-Governmental Organisations in Malawi;19. The World Bank;20. Department for International Development Malawi National Social Support Programme, 2012/13 - 2015/16;21. UNICEF; 22. World Food Programme;23. Malawi Vulnerability Assessment Committee;24. Elderly People Association;25. Network of Organisations working with Vulnerable and Orphaned Children; 26. Chairperson, NGO Gender Network;27. Centre for Social Research ;28. National AIDS Commission;29. Action Aid;30. OXFAM;31. Plan Malawi; 32.DfiD;(Irish Aid and Norway may join). 119 Social Action Fund, and the Ministry of Gender, Children and Community Development. These Ministries in turn channel funds to local authorities, NGOs, CBOs, and community associations. Currently, many of the social support interventions in Malawi are fragmented, not well coordinated, systematic, or integrated. Table 7.5: Expenditures and Institutions fragmentation: Malawi SP Programs and Implementing Agencies Social Expenditures Responsible Institutions Public Works MoLG&RD, LDF, DAs, MoI, MoAFS, MoT, Department of Forestry, NGOs and CBOs, Beneficiaries Village Savings and MoGCSW, MoLGRD, MoF, RBM, MFIs, NGOs and Loans CBOs, Village Agents, Beneficiaries Social Cash Transfers MoGCSW, MoLGRD, UN Agencies, PPPs, NGOs, CBOs, Volunteers, Beneficiaries School Meals MoE, UN Agencies, NGOs, MoLGRD, MoAFS, Department of Forestry, Volunteers, Beneficiaries Farm Inputs Subsidies Ministry of Agriculture 316. The Malawi Social Action Fund (MASAF) is the largest nationwide safety nets public works program and provides conditional cash transfers on a day labor supply basis for productive public works and direct cash transfers to the most vulnerable. MASAF is supervised and supported by the Ministry of Finance’s Local Development Fund Technical Support Team (LDF- TST) under the Ministry of Local Government and Rural Development. 317. Public Works Programs under the MASAF program have reached a total 9,023 subprojects that have been implemented to address various sector gaps such as reforestation (3,499 ha leading to improvements in environment), roads (28,023 km facilitating access to socio- economic services), and irrigation (1,339.6 ha contributing towards raised food security and income). On the other hand, 12,202,644 person days have been provided in labor intensive works during the program. 318. Evidence shows that the cash transfers have assisted beneficiaries to address food security needs (62% of PWSP beneficiaries81); farm inputs purchase (35% of PWSP beneficiaries), as well as basic health and education needs (10.6% and 12.6% of PWSP beneficiaries respectively). The 2012 PW Tracking Study82 indicated that 52% of beneficiaries reported to have an increased income as a result of their participation in the PWs. A total of 1,016,887 beneficiaries had been reached with cash transfers through the Public Works program, i.e. 112.9% of the planned target with additional financing. Of these beneficiaries, 47.5% were women while 10% were under direct support (i.e. households with the elderly, orphans, disabled and chronically ill among others). 319. The MASAF program, however, does not primarily focus on safety nets. It has over the years been directed to support infrastructure building in the form of 3-bedroom teachers staff houses and school blocks. This infrastructure support involves three of the five programs in the Community Window: the Primary School Staff Housing Project; the Crisis Response Program and Community 81 2011 PW Tracking Study 82 MASAF III 120 Savings, and the Investment Promotion. In terms of infrastructure-related subprojects in the Window, the Education SWAp program has contributed the highest amount of resources, followed by the Primary School Staff Housing Program. The least funded program is the Community Open Menu, which is solely financed by GOM. Its resources have been reducing each year, even in the face of a large and increasing demand from communities to address their priorities for livelihoods and service gaps. 320. Figure 7.6 below shows the resource allocation in the MASAF III APL II, with the largest share going to the education sector. Figure 7.6: Resource allocation in MASAF III Resource Allocation in MASAF III Education Sector Wide 4% Program Community Savings and Investment Groups 30% CDD Open Menu 56% Primary School Staff 6% Housing Project Crisis Response 4% 321. Community Savings and Investment Promotion (COMSIP) is an excellent by-product of the MASAF public works cash transfers. The clients of the public works have been assisted in forming savings groups by extension workers of the Ministry of Gender, Children and Social Welfare. A total of 99,153 people receiving cash transfers from Pubic works, of which 61% are women, have formed 4,457 groups which are members of the COMSIP. The groups have mobilized MK 434.6 million in savings and invested in different income generating livelihood activities. 79%83 of group members reported to have generated income above 50% of their initial cash transfer received. 322. The Malawi Social Cash Transfer Program (SCTP), implemented by the Ministry of Gender, Children and Social Welfare (MoGCSW) currently operates in 7 districts. It was designed to alleviate poverty, reduce malnutrition, and improve school enrolment and attendance by delivering regular and reliable cash transfers to ultra poor and labor-constrained households. The pilot program started in 2006 and has grown to reach close to 30,000 beneficiary households in early 2012. 83 2012 Tracking Study 121 323. Several donors have been contributing to the SCTP since the start of the program. From 2007 till 2012, the SCTP has largely been funded by the Global Fund to Fight AIDS, Tuberculosis and Malaria (GFATM) through the National Aids Commission. Technical assistance capacity building, financial and infrastructural support is being supported by United Nations Children Fund (UNICEF) and program implementation largely by KfW and the EU based on agreed and approved annual work plans since 2012. In 2010/2011, Irish Aid provided financial support to cover a funding gap and has provided additional funding for capacity building in 2012. The European Union (EU) has topped the donor contribution through KfW to enable full coverage in the existing seven districts as well as scale-ups in additional 8 districts. KfW is supporting the Social Cash Transfer Program through significant technical support and investments in technology, including the development of an MIS and financial management software, suitable for the Malawian context. 324. By December, 2013, the SCT Program will be scaled up to 16 districts and by 2014 will cover the country and reach the most vulnerable and labor-constrained households. The SCT Program has, in the set up period and due to low coverage, a high overhead administrative cost ranging from 10% to 19%, and is facing a financing gap to be able to scale up to the remaining districts. 325. E-Cash Transfers: The SCT Program, with support from the EU, is piloting e-transfers in a few districts. By 2014 e-transfers under the program should be in all districts. WFP has also made e-transfers to 125,000 individuals in disaster prone areas. Going forward e-transfers would be a cost effective mechanism used in cash transfers for a social protection program. Once the E-Transfers come on line the cost of delivery and monitoring is expected to reduce to a fraction of what it is now. 7.4 Policy Priorities or Suggestions for Reform 326. Short term recommendations for increasing the effectiveness and efficiency of Social Protection expenditures focus on increasing the benefit levels in cash transfers, rebalancing the total budget expenditures, better targeting mechanisms, and harmonizing programs and institutions. Specific recommendations are to: x Reallocate and rebalance the total budget expenditure allocations to other more productive social protection programs. SP expenditures excluding FISP are inadequate to cover the needs of the vulnerable. Moreover, current FISP expenditures are ineffective and unproductive since they are not well targeted to those most in need. This leads to errors of inclusion and exclusion.84 x Review, redesign and redefine the FISP. The inclusion of FISP expenditures as Social Protection masks the lack of funds for social protection in the form of other benefits and safety nets. Furthermore, FISP should be redesigned with the objective of raising productivity and food security. Some social protection aspects of FISP should be redesigned as safety net and productivity enhancing programs for the poorest agriculture-dependent households with access to land. Moreover, FISP should have complementarity with other social support programs. As it is currently designed and implemented, FISP is not effectively and efficiently reaching the poorest farming households and cannot be defined as social protection. Those FISP expenditures - currently benefiting urban and rural landless poor as well 84 As part of recommendations, the Team propose that the amount of around MK25 billion be considered for inter - sectoral transfer between FISP and SP. This reallocation of the overall budget for poverty targeting could be done through rebalancing the budget from the current FISP expenditure to cash transfers, though SCTs or MoAFS packages to better target the poor, or through MASF IV public works in coordination with MoAFS. 122 as the richest urban and rural quintile - are ineffective as SP and should be reallocated to Programs that reach the poor effectively and equitably. The program should be better targeted towards only rural farming households with land and should focus on the second and third quintile producing Maize and other food crops. x Harmonize institutional arrangements. Present institutional fragmentation and the existence of a plethora of uncoordinated social expenditures do not allow for timely, transparent, and accountable delivery of funds with minimal risk of duplication. Enhancing efficiency of the limited resources for social protection calls for expediting the establishment of the Social Support Fund Mechanisms and rationalizing institutional setting (particularly, considering merging the Ministries of Gender, Children, Social Welfare and the Ministry of Disabilities and Elderly into a single Ministry for Social Protection). x Unify Beneficiary Data and utilize E-transfer mechanisms. Social Protection cash transfers should use E-transfers based on a Unified Registry systems linked to the Poverty and Income Profile and the mapping of all citizens. Work on the proposed unified registry system and harmonized targeting mechanism should be expedited and completed. 327. Medium term recommendations to improve the effectiveness and efficiency of Social Protection expenditures should focus on continued expenditures adjustments through systematic monitoring and reviewing to ensure that they cover the needs of the poorest and vulnerable. Hence the following recommendations: x Review systematically targeting and adjust it to reach those most in need (progressively). x Coordinate well expenditures with continued institutional harmonization (progressively). x Strengthen the registry of beneficiaries based on poverty profile through the unified registry of beneficiaries and link it to the National Registry of all citizens. x Review Social Protection Programs and adjust the designs on the basis of a review of trade-offs in outcomes and benefits between direct cash transfers versus cash transfers for Public Works, school meals and other interventions. 123 CHAPTER 8. CONCLUSION 328. Although the GOM has made important economic reforms, the budget faces a tightened resource envelope over the medium term. The GOM plans to enhance revenue mobilization, restrain growth in expenditures, and ensure small repayments to the banking system. But these efforts are complicated by the recent decline in growth and donor inflows and by existing government commitments to certain large expenditures. This PER, a joint effort of WBG and GOM, recommends steps to improve public expenditure efficiency, with a focus on PFM and expenditures on agriculture, transport, education, health, and social protection. It recommends, in broad terms, that GOM enact politically and institutionally compatible sector reforms, improve PFM especially in strengthening expenditure management and control, rationalize public transfers and subsidies, and control expenditures like the wage bill and employees’ travel costs. Key challenges in various areas of planning, budgeting, and sector performance are highlighted as followed to justify the specific recommendations presented in the policy matrix in the Executive Summary. 329. Progress in PFM reforms has been uneven. Positive changes include a more robust macro- fiscal framework, better cash planning, and an output-focused Budget. However, most funding from development partners remains off-budget and there has been little policy-focused budgeting. Important steps would be: integrating updates of sector strategy and expenditure plans into the budgeting process, including all DP financing in the budget, and transferring responsibility for PSIP to the Ministry of Finance. Issues in implementation of IFMIS and plan for rollout to districts need to be addressed, focusing on commitment control and accounting. 330. An estimated 85 percent of Malawi’s population depends on agriculture for its livelihood, but agricultural policy has produced mixed results. Ostensibly a pro-poor subsidy, FISP accounts for most government expenditure in the sector but is plagued by corruption and distortions. In addition, cumbersome budgeting procedures have led to the exclusion of local authorities and nongovernment parties, the fragmentation of aid, and the placement of many expenditures off-budget (where they are not subject to oversight). 331. Roads dominate Malawi’s transport sector and GOM transport expenditures. Maintenance is a challenge for this relatively large road network, especially as some contractors are still owed arrears. Prioritizing road projects based on expected traffic levels, rural road access, and available funding will help improve efficiency. At the same time, poor maintenance and competition from the road network have limited the market share of inland water and rail transit. The Nacala rail project could reinvigorate this area. Finally, the country urgently needs to address the safety concerns raised by the International Civil Aviation Organisation, especially as air traffic becomes more popular. 332. Malawi’s commitment to education is evident in its increasing share of government spending (19 percent in FY 2013). More students are pursuing post-basic education, and more employers are requiring those skills. Enrollment rates have risen dramatically in the last forty years. However retention and participation rates remain low for the poorest children and the country’s numeracy and literacy levels are among the lowest in the region. 333. By 2009, two thirds of Malawi’s health financing came from external sources, making the sector vulnerable to external shocks. The country is on track to meet MDG 4 (Reduce Child Mortality) and MDG 6 (Combat HIV/AIDS, Malaria, and other diseases) but not MDG 5 (Improve Maternal Health). Public resources are not shared equitably across the districts and PE share is increasing at the cost of ORT and capital investment. Over time the government share of health spending should increase, perhaps through a negotiated deal between MOH and MOF. GOM should continue to pursue universal 124 coverage under the essential health package, partnerships with nongovernment actors, and outreach and mobile clinics. A PPP policy, based on performance targets and contracts, could improve healthcare quality and access, especially for the underserved. 334. Social Protection expenditures are insufficient and ineffective. Including FISP in the Social Protection category masks the lack of funding for other benefits. FISP accounted for 11.5 percent of total GOM expenditures in 2012-2013, while all other SP funding accounted for just 2.6 percent – half the African average. SP programs are largely donor-funded. Cash transfers have helped the poor with food security, farm inputs, and basic healthcare and education. But these programs are highly uncoordinated and fragmented, and so they make little impact on chronic poverty. Furthermore, the Social Cash Transfer Program has high overhead and is facing a financing gap. 335. It is worth noting that a number of key issues and policy recommendations presented in this PER are not new. They were already analyzed in a series of previous PERs—in 2001 and 2007. While improvements have been made in certain policy areas, some critical reforms remain lagging. Instituting and implementing transparent and rigorous policies to enhance the sense of ownership and accountability among spending agencies and to align effective support and engagement by donors as well as non-state actors is fundamental to ensure the chance of success of further reforms and the planned medium term fiscal framework. Amongst all the challenges the country currently is facing, it is imperative that sooner than later, the GOM resolve to strengthen expenditure control at all levels—from upstream design of effective planning and policy to downstream establishment of accountability and fiduciary discipline imposed upon all MDAs. Last but not least, such interventions can succeed only if addressed at both technical and political levels. 125 ANNEXES Annex A1: Details of Generic Goods and Services Annex 1 Details of Generic Goods and Services 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Internal Travel 4.4 6.4 9.9 7.4 8.05 8.75 Public Utilities Payments 3.5 3 3.3 3.4 3.8 4.3 Office Supplies and Expenses 5.6 6.9 9.1 6.4 7.5 6.9 Rents 1.6 1.4 2.5 3.4 2.6 3.3 Training Expenses 4.2 6.6 4.5 4.2 4.9 4.4 Acquisition of Technical Services 3.9 7 11.4 12.3 14 17 Other Unclassfied Goods and Services 0.6 0.7 3.4 7.2 0.15 8.35 Total Generic Goods and Services 21.8 17.4 43.8 44.3 41 53 Source: MOF and Bank Staff Calculations. 126 Annex A2: Structure of Arrears (December 2012) in MK Billion Annex 2 Structure of Arrears (December 2012) MK Billion Parastatals SFFRFM 16.2 Air Malawi 5.0 MBC 4.3 ADMARC 4.9 National Food Reserve Agency 2.2 Malawi Rural Finance Company 1.9 Demat/Medi/Sedom 17.0 Total 35.2 Procurement of Goods and Services by MDAs Malawi Police Service 10.0 Road Sector Projects 9.0 Ministry of Health 2.0 Malawi Housing Corporation (MHC) 2.2 Malawi Defense Force 1.6 Malawi Prison Service 1.3 Immigration Department 1.2 Malawi Electoral commission 0.4 Ministry of Housing 2.1 OPC 0.6 Total 30.4 Pensions, Salaries, Utilities and Subscriptions Pension arrears 3.7 Muzuzu University 0.6 Ombudsman 0.2 Salary Arrears Ministry of Education 0.3 Other Ministries and Departments 0.5 Total 5.2 Compension for Court Cases 1.7 Grand Total 72.6 Source: Government of Malawi 127 Annex A3: Progress Made in Addressing Key Budget Process Issued Identified in 2001 Public Expenditure Review Issue Identified Status 2001 Status 2013 (2001 PER) 1. Poor coordination ƒ Data inconsistencies and poor 9 Twice yearly updates of macro-fiscal among key agencies quality fiscal forecasts. framework undertaken with IMF. undermines fiscal ƒ No medium-term macro-fiscal forecasting forecasting. 2. Single year fiscal and ƒ MTEF outer year spending 9 MTEF included in 2011/12 Financial budget framework. forecasts developed for MDAs Statement but not in 2012/13 Statement. budget not included in Budget Estimates. 9 Output Based Estimates now include outer year MTEF expenditure forecasts at program and sub-program level. 3. Budget not ƒ Substantial amounts of donor 9 Budget support has increased in comprehensive of all funding remain off-budget. importance. Government spending. 8 Off-Budget DP project financing twice level of on-budget financing. 4. Budget preparation ƒ Parallel processes for and little 8 Budget calendar has separate processes processes not integrated integration in for RB and DB for budgeting PSIP and recurrent preparation. expenses. 5. Poor adherence to ƒ Late issuance of budget ceilings 8 Key deadlines missed including for budget calendar. embeds traditional line item issuing budget ceilings – limits progress budgeting. in developing more strategically focused of budget process. 6. Weak budget execut- ƒ Cash budgeting procedures 9 Strengthened cash management ion and monitoring affected by poor compliance and procedures introduced. systems. information flows. ƒ Financial management affected by 8 IFMIS rolled out but operates primarily as compliance and control payments system, no commitment weaknesses. control. ƒ Deficiencies in timeliness, 9 Improved procedures for monthly fiscal coverage and quality of in-year reporting adopted, but data problems fiscal data. remain. 7. Lack of sustained ƒ Weak political commitment to 8 Uneven political commitment continued political support. PFM reform objectives underlies to affect progress with implementing budget discipline and account- reforms. ability problems. 8. Absence of a ƒ Reforms not sufficiently focused 8 PEFA scores indicate that budget sequencing strategy for on budget execution. execution and accounting reforms have budgetary reform had limited impact. 9. Weak pay and incent- ƒ Civil servants significantly 9 Civil service salaries significantly ive framework. underpaid compared to private increased but still underpaid compared to sector. the private sector ƒ Relatively less favourable reward 8 Difficulties persist in retaining middle structure for middle level staff. level professional staff. 128 Annex A4: Key Stages in Malawi’s Budget Calendar Cabinet/ Month MoF MEPD Line Ministries Parliament ƒ MDAs instructed to prepare planned ƒ Initial-PSIP circular Sep activity submissions. and calendar issued. ƒ Initial National Accounts Forecasts. ƒ Appraisal of PSIP ƒ PSIP project ƒ DP meeting on planned aid proposals. proposals submitted Oct commitments to MEPD. ƒ ECF semi-annual review with IMF ƒ Initial revenue forecasts ƒ BPFP prepared.# ƒ Planned activities Nov submitted to MoF. ƒ Cabinet ƒ PSIP meetings with Dec reviews BPFP# MDAs. ƒ Pre-Budget Consultations with ƒ Draft PSIP public. prepared. Jan ƒ Budget Guidelines issued. ƒ Treasury Circular + budget ceilings issued. ƒ Cabinet ƒ Draft Revenue Bill prepared reviews draft ƒ Cabinet informed of sector ceilings Feb PSIP. ƒ Economic and Fiscal Policy Statement prepared. ƒ Semi-Annual Review of ECF with ƒ PSIP Circular ƒ Draft budget IMF issued with estimates submitted ƒ Final national accounts forecasts. approved to MoF Mar allocations. ƒ PSIP cash flow fore- ƒ DP meeting confirms aid commitments casts sent to MEPD ƒ Budget hearing with MDAs ƒ Cabinet ƒ Consolidation of draft Budget. ƒ PSIP finalised and ƒ Recurrent reviews draft ƒ Budget documents prepared. issued expenditure work Apr Budget plans and cash ƒ Summary of extra-budgetary support prepared. forecasts sent to MoF May ƒ National Budget Documents printed. ƒ Budget Speech Jun ƒ Budget Session # Not yet introduced. Source: Based on the revised budget calendar provided to the PER team by MoF. 129 Annex A5: MoF Treasury Department Organization Chart Treasury Department Debt and Aid Economic Affairs Revenue Policy General Administr- Budget Division. Management Division Division Division ative Support Division Budget Preparation and Resource Mobilization Macroeconomic Policy Finance and Accounts Tax Policy Section Coordination Section Section Section Section (6 posts) (10 posts) (16 posts) (11 posts) (31 posts) Budget Expenditure Debt and Aid Planning Sectoral Policy Analysis Human Resources Non-Tax Policy Section Monitoring Section Section Section Management Section (6 posts) (20 posts) (7 posts) (7 posts) (21 posts) Cash Flow Management Disbursement and Debt Public Enterprise Reform Administration and Section Servicing Section and Monitoring Unit Office Services Section (8 posts) (6 posts) (5 posts) Development Assistance Financial Policy Section Coordination Unit. (5 posts) (6 posts) Research, Planning, Monitoring and Source: Functional Review Report Ministry of Evaluation Section Note: Excludes posts for personal secretaries Finance (Treasury Department), 2010 (5 posts) and typists. 130 Annex A6: Breakdown of Recurrent Expenditures (Personnel, Operations & Maintenance of Assets) - MK million: 2007 Base Year Category 2006/07 2007/8 2008/09 2009/10 2010/11 2011/12 MTPW Recurrent expenditures 884 1,682 3,124 1,552 878 963 (Vote 400) RA/RFA Recurrent expenditures 2,502 1,975 2,569 6,267 5,963 2,929 (Roads Fund) Total Recurrent Transport Sector 3,387 3,657 5,692 7,819 6,841 3,892 Total GOM Recurrent 109,352 121,284 178,086 165,988 176,075 190,914 TS Recurrent as % of Total 3.1% 3.0% 3.2% 4.7% 3.9% 2.0% GOM Recurrent Real GDP (2007 prices) 466,282 510,539 553,112 603,090 642,492 670,002 Source: MOF/IFMIS (Itemized Expenditure Analysis) & RA Audited Annual Financial Statements Annex A7: Public Road Network Composition Road Type Paved Unpaved Total Network Classification Km % Share Km % Share Km % Share Main 2,976 69 381 3 3,357 22 Secondary 513 12 2,612 23 3,125 20 Tertiary 44 1 4,077 37 4,121 27 District 8 0.2 3,492 31 3,500 23 Urban 771 18 577 5 1,348 9 Total designated 4,312 100 11,139 100 15,451 100 Share (%) 28 72 100 Source: Roads Authority (RA) 131 Annex A8: Trends in overall expenditure a) Sectoral Distribution of the GoM MTEF: 2011/12 – 2013:14 Budgetary Distribution (% of Total Budget) Sector 2011/12 2012/13 2013/14 Outturn Approved Projection Education 17.7 19.3 16.2 Agriculture 12.5 16.3 14.6 Health 11.4 12.0 7.5 Transport Infrastructure 9.3 8.6 8.8 Rural Development 5.1 3.7 1.6 Energy & Industrial Development 2.7 2.3 2.7 Irrigation & Water Development 2.4 1.2 1.0 Child & Youth Development 1.3 2.9 0.6 Climate Change, & Environment 0.9 0.4 0.6 Other 36.7 33.3 46.4 Total 100.0 100.0 100.0 Data Source: Draft 2012/2013 Financial Statement and others b) Regional Trend of primary education expenditure Primary Education Expenditure as % of Total Education Expenditure in Selected African Countries 70.0% 56.2% 58.3% 60.0% 54.7% 50.8% 48.5% 50.0% 45.3% 40.5% 40.0% 35.7% 35.0% 27.6% 30.0% 19.2% 20.0% 10.0% 0.0% c) Benchmarks for Public Resource Allocations to Education Benchmarks for Public Resource Allocations to Education x MGDS II: the proportion of the GoM budget allocated to education, science and technology should be 19.1%, 21.1% and 21.2% during 2010/11, 2011/12 and 2012/13 respectively. x MGDS II: education expenditure proportion shall be about 20% of the national budget, in line with the UNESCO Education for All (EFA) Dakar Declaration Framework for Action, 2000. x GPE: education sector recurrent budget should be at least 20% of the total GoM recurrent budget. x ESIP: all expenditure on primary education should rise from 56.3% in 2008/09 to 62.1% in 2011/12 132 d) Regional Secondary Education Recurrent Expenditure as % of Total Recurrent Expenditure Secondary Education Recurrent Expenditure as % of Total Education Recurrent Expenditure in Selected African Countries: 2011 45.0% 42.7% 40.0% 35.0% 31.3% 31.4% 30.2% 28.5% 30.0% 25.1% 23.8% 25.0% 22.8% 20.8% 19.8% 20.0% 16.8% 15.0% 10.0% 5.0% 0.0% Source: UNESCO Education Statistics 2011 133 Annex A9: Recurrent Expenditures by Public Institutions and Countries Ratio of Per Capital Recurrent Expenditure on Primary Per Capita Recurrent Expenditure in Public to Tertiary Education: 2011 Institutions: 2011/12 1,000,000 MWK 886,765 900,000 Uganda 14.8 800,000 700,000 Rwanda 27.5 600,000 500,000 Malawi 169.5 400,000 300,000 MWK 239,233 Lesotho 49.1 200,000 MWK 124,567 100,000 MWK 5,232 MWK 34,390 Burundi 28.2 0 Botswana 34.1 Angola 20.6 0 50 100 150 200 Source: Mambo et al. (2012) 134 Annex A10: Comparison of Tuition Fees among Countries in the SADC Region GDP/Capita Country Agriculture Arts Business Science Engineering Medical Current US$ 2011 Botswana 2,519 2,519 3,332 3,332 3,672 9,481 Kenya (Kenyatta) 1,924 1,455 1,924 5,376 851 Malawi 150 - 330 150 - 330 150 - 330 150 - 330 150 - 330 150 - 1800 351 Namibia 1,075 1,075 1,075 1,075 1,075 5,828 South Africa (UCT) 5,784 4,665 5,224 5,846 8,066 Tanzania 635 825 952 553 Uganda (Makerere) 930 550 695 1,427 1,070 477 Zambia 2,244- 1,1574 2,772 - 1,418 2,660 - 1,444 1,414 Zimbabwe (NUST) 700 800 800 800 741 135 Annex A11: Some Literature on the Impact of Double Shifting Through regression analysis using Southern African Consortium of Monitoring Education Quality (SACMEQ) and PASEC samples, Michaelowa and Wechtler (2006) record the following findings on the impact of double-shifting: x Typically, reduced effective teaching time and unsuitable class hours lead to considerable losses in student learning, especially after the 3rd grade x There are positive financial implications as compared to two separate classes, in particular if there is a shortage of classrooms. However, there are negative financial implications as compared to a single larger class, because even if there is only one teacher for two shifts, he or she must be compensated for overtime, often at higher rates (Michaelowa and Wechtler 2006: 18) x Students’ achievements decrease by 6% (PASEC) to 10% (SACMEQ) of a standard deviation in double-shift schools. Michaelowa (2001) also find a significant negative impact of double-shifts on students’ achievements (even when controlling for knowledge before school enrolment). She also finds that “below a class size of about 100 students, double shift classes are generally inefficient” (p.25). This means that the negative effect of double-shifts is still bigger than the negative effect of big classes up to 100 students. She notes that one reason for the negative impact of double-shifts on students’ achievements is the reduced time of the teacher for each class/student as well as for the preparation and reflection of lessons and assignments. The UNESCO/ IIEP finds that double-shifts have no impacts on student achievements in Brazil, Chile (both of which have a tradition of double-shifts) and in Senegal. On the other hand, UNESCO /IIEP also describe cases from Niger and Guinea where double-shifts do have a negative impact, especially in the long run. Regarding equity and access, double shifts seem to be desirable. Due to the number of students double- shift schools are often in urban areas, and often serve students from poor and middle-income families and disadvantaged pupils. In contrast, parents who have the means to do so, try to send their kids to single- shift institutions (which might be private). Furthermore, double-shifts allow children and young adults who are working to go to school. This also counts for girls who often have to support their mothers at home. Different shifts allow more flexibility. This is not only relevant for children who are working (at home or to earn money) but also in regard to crop growing and harvesting. During these times of the year children can often not go to school as they do have to help their parents. Being more flexible in terms of time, double-shifts can potentially increase enrolment rates. The literature makes these following recommendations: x “Optimize time schedules in cooperation with parents and make sure that official teaching hours are effectively respected. x Avoid double shifts as long as students can be put in a single class (only with more than 80 students, the negative effect of class size becomes so strong that double shifts usually become preferable).” (Michaelowa and Wechtler 2006: 18) 136 Annex A12: Malawi Social Protection Programs and Expenditures All Social Source Distric Intended Targeting # HHs # Benefit Cost GOM Protection of Funds ts Target Method person per HH of Budget Expenditures Group s in kind Delive Expendit or MK ry % ure of 2012/13 Total MKs (000) Farm Input GOM, 28 Fertilizer Local Leaders/ 300,000 1,600, 500 54,904. 50 Subsidy (MoA), Subsidy MoA Extension 000 annually Program DPs (i.e. coupons Workers. DADO DfID) to Poor officers brief their household EPA staff on who farming to choose who families: later brief Extension Workers. Open community forum: Villagers/Chief/P olice/Reps, NGOs, attend decide School GOM 13 Primary Schools based on 630,00 Daily 170 Feeding (MoE), School enrolment, 0 Program WFP, going attendance/ drop- Mary children out rates, food Meals, security situation Social Cash GOM 8 Direct Recertifying 10% 28,000 100,00 2700 10%- 100 (and a Transfer (MoGCS Cash poorest in each 0 mthly 19% 67 million Scheme W), EU, transfers district. Objecti Euros KfW, to. Ultra community ve: commitme Irish Aid, poor/labor Targeting, IH3, 318,000 nt over the UNICEF constraine and PMT HHs in next year d HHs all from districts. EU/KFW) Income The 15 Poor HH Community NA NA NA NA GoM to Generating European /vulnerabl Targeting provide Public Works Union e persons Program (The (MLRD) or with a program is vulnerable now referred person as to as Rural a member. Infrastructur PWs that e create Development short term Program employme (RIDP nt. MARDEF/ NA: NA NA NA NA NA NA NA 1,000 YEDEF MASAF The 28 Poor with Geographical, 586,000 2,900, 14,400 8% 13,783.87 Public Works World labor Poverty 000 annually Program Bank, capacity: Targeting; 137 (MASAF is GOM Labor MVAC now LDF) intensive Assessment report PWs that to determine # create beneficiaries per short term district employme nt. Ministry of KfW, 28 Poor, sick MoGSCW NA NA NA NA 982.81 Gender UNICEF, and Secretariat for Children and Irish Aid, vulnerable SCT program Social MoF, children, recertifying and Welfare Dfid, men and targeting using EU. women PMT and Community Targeting. Ministry of 28 Disabled Extension and NA NA NA NA 293.09 Disabilities and social workers. and Elderly. Elderly Community targeting. Government GOM: GOM GOM GoM payrolls and NA 30,000 NA NA 16,000 Public Governm Retiree Retirees records of staff Pension ent s Scheme Public (GPPS) Pension Scheme (GPPS) 138 Annex A13: Estimated cash transfer levels to mitigate vulnerability resulting from the Forex crisis The estimates present an idea of the scale required for a social protection expenditure increase if all urban and rural poor in the first 3 quintiles – i.e. more than half the country’s population - were covered without any further targeting at 33% and 50% levels of compensation.85 85 Malawi: A Comprehensive Package for Competitiveness, Growth and Poverty Reduction Presentation to Government by the World Bank, March 2012 139 REFERENCES Asbu E., Walker O., Kirigia J., Zawaira F., Magombo F., Zimpita P., Manthalu G., Nkhoma D., Eldridge C., Kataika E., Assessing the efficiency hospitals in Malawi: An application of the Pabón Lasso technique, Publication of the WHO African region, African health monitor, issue 14, March 2012 Chifundo Rose Kachiza, Analysis of Patient Care Services at Kamuzu Central Hospital: A Critical Look at the Hospital’s Challenges and Opportunities in Meeting the Need . Submitted in Partial Fulfilment of the Requirement of Master of Public Health Degree, University of Malawi, 2005 Chongoma A, and Franz J., The Skills Development System in Malawi, The World Bank, April 2013 Comparative regional data taken from the Africa Infrastructure Country Diagnostic (AICD), The Burden of Maintenance: Roads in Sub-Saharan Africa, Background Paper 14, June 2008 Department of Budget at MOF, and Department of Budget and Planning at MOH, May 2012 Evaluation of retirement systems of countries within the Southern African Development Community, Country Profile: Malawi, Oxford Policy Management, 2010 Five year impact assessment: Multi-country impact evaluation of the scale up to fight AIDS, Tuberculosis and Malaria: Malawi National Health Accounts with Sub-accounts for HIV/AIDS, TB and Malaria, May 2008 Fölscher, A. Mkandawire, A. and Faragher, R. Evaluation of Public Financial Management Reform in Malawi 2001-10. Fiscus Ltd, UK. Country Case Study Report in the Joint Evaluation of Public Management Reform managed by the African Development Bank, Denmark and Sweden, 2012 Integrated Household Survey 2010-2011: Household Socio-Economic Characteristics Report, September 2012 Malawi: A Comprehensive Package for Competitiveness, Growth and Poverty Reduction Presentation to Government by the World Bank, March 2012 Malawi National Health Accounts 2002/2003 – 2004/2005 Summary Report, Lilongwe, February 2007 Mambo M., Cole P., and Ndala K., Education Sector Plan Review Report, September 2012 Michaelowa K., Primary Education Quality in Francophone Sub-Saharan Africa: Determinants of Learning Achievement and Efficiency Considerations Hamburg Institute of International Economics (HWWA), October 2001 Michaelowa K. and Wechtler A., The Cost-Effectiveness of Inputs in Primary Education: Insights from the Literature and Recent Student Surveys for Sub-Saharan Africa, Association for the Development of Education in Africa, March 2006 Ministry of Education, Science and Technology, Education Statistics 2011, 2011 140 Ministry of Education, Science and Technology, Education Sector Performance Report 2010-11, 2011 Ministry of Health, Malawi National Health Accounts (NHA), Planning & Policy Development, Lilongwe, Malawi, 2002 Ministry of Health, Malawi National Health Accounts (NHA), Planning & Policy Development, Lilongwe, Malawi, 2006 Ministry of Health, Malawi National Heath Accounts with subaccounts for HIV/AIDS, Malaria, Reproductive Health, and Child Health for Financial Years 2006/07, 2007/08, and 2008/09, Department of Health Planning and Policy Development, Lilongwe, Malawi, 2012 Ministry of Health and Clinton Health Access Initiative, Health sector resources in Malawi: Key findings from resource mapping Exercise, 2012 Ministry of Health, Health Sector Strategic Plan (2011-2016), Malawi 2010 Ministry of Health, Human Resources Department, May 2012 Ministry of Finance, Draft 2012/13 Financial Statement, Government of Malawi, 2012 National Statistical Office (NSO) and ICF Macro, Malawi Demographic and Health Survey 2010, Zomba, Malawi, and Calverton, Maryland, USA: NSO and ICF Macro, 2011 National Statistical Office (NSO) [Malawi], and ORC Macro, Malawi Demographic and Health Survey 2004. Calverton, Maryland: NSO and ORC Macro, 2005 Phiri, M. and Chisala, R Report on the Status Review of Integrated Financial Management Information System and High Level Operational Requirements, 2012 TalmonǦl’Armee, A., Technical review of primary school infrastructure construction in Malawi, September 2012 UNESCO, Malawi EFA Profile, United Nations Educational, Scientific and Cultural Organisation, Dakar, 2012 World Bank, Africa Social Protection Strategy 2012-2022, June 2012 World Bank, Malawi Country Economic Memorandum, January 2010 World Bank, Malawi Education Country Status Report (CSR 2008/09), April 2009 World Bank, Malawi Public Expenditures: Issues and Options, September 2001 World Bank, Malawi Public Expenditures Review, September 2007 World Bank, Multi-sector Technical Assistance Mission, May 2012 World Bank, Malawi Third Social Action Fund Project Documents, 2013 141 World Bank, Practitioners’ Toolkit for Agriculture Public Expenditure Analysis (Report N°60015-GLB), March 2011 http://data.worldbank.org/indicator/NY.GNP.PCAP.CD/countries?display=default http://www.pepfar.gov/about/ (The US President’s Emergency Plan for AIDS relief (PEPFAR)) http://www.theglobalfund.org/en/oig/reports/ (Audit of Global Fund Grants to the Republic of Malawi – GF OIG – 10-009, August 3, 2012 – paragraph 31 and 32) http://www.undp.org/content/dam/undp/library/MDG/english/MDG%20Country%20Reports/Malawi/MD GsProgressAtaGrance-2010.pdf. Accessed on 3.1.13 http://www.who.int/countries/mwi/en/, accessed on 4.8.2013 www.who.int/data - Global Health Observatory Data Repository, accessed on 2.27.13 142