Report No: AUS5649 Republic of Congo Enhancing efficiency in education and health public spending for improved quality service delivery for all A public expenditure review of the education and health sectors June 2014 AFTEW AFRICA Document of the World Bank Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. FISCAL YEAR January 1 – December 31 CURRENCY EQUIVALENTS (Exchange Rate Effective May 1, 2014) Currency Unit = Central African Francs (XAF) XAF 473 = US$ 1 Weights and Measures Metric System ABBREVIATIONS AND ACRONYMS CMR Child Mortality Rate COMEG Congo Essential Generic Drugs Agency Conférence des Ministres de l’Éducation des États et Gouvernements de CONFEMEN la Francophonie/ Conference of Ministers of Education and Governments of French Speaking Countries Centre de santé intégré avec paquet élargi d’activités de santé/ CSI-PMAE Integrated Health Center with Expanded Services Offered Centre de santé intégré avec paquet minimum d’activités de santé/ CSI-PMAS Integrated Health Center with Minimum Services Offered CSR Country Status Report Department Administrative et Financier/ Financial Administration DAF Department Direction du Contrôle et de l’Evaluation des Investissements/ Investment DCEI Control and Evaluation Directorate Direction des Études et de la Planification/ Research and Planning DEP Directorate Directions Départementaux de l'Enseignement Primaire et Secondaire DDEPSA Chargé de l'Alphabétisation/ Regional Departments of Education Primary, Secondary Education, and Literacy DDS Direction Départementale de Santé/ Departmental Health Directorate DGB Direction Générale du Budget/ Budget General Directorate Direction Générale du Plan et du Développement/ Planning and DGPD Development General Directorate Délégation Génerale de Grands Traveaux/ Large-Scale Projects General DGGT Directorate i DHS Demographic and Health Survey/ Enquête Démographique et de Santé DRC Democratic Republic of Congo ECOM Enquête Consummation Ménages/ Household Survey Enquête Démographique et de Santé du Congo/ Congo Population and EDSC Health Survey Enquête sur l’Emploi et le Secteur Informel au Congo/ Congo EESIC Employment and Informal Sector Survey EMIS Education Management and Information System GDP Gross Domestic Product GER Gross Enrolment Rate GoC Government of Congo GPE Global Partnership for Education HIPC Heavily Indebted Poor Countries IFC International Finance Corporation IMR Infant Mortality Rate Institut National de Recherche et d'Action Pédagogique/ National INRAP Institute of Research and Pedagogical Action MDG Millennium Development Goal Ministère de l’Économie, des Finances, du Plan, de l’Intégration et du MEFPIPP Portefeuille Public/ Ministry of Economy, Finance, Plan, Integration and Public Portfolio Ministère de l’Enseignement Primaire, Secondaire et Alphabétisation/ MEPSA Ministry of Primary, Secondary Education, and Literacy MES Ministère de l’Enseignement Supérieur/ Ministry of Higher Education Ministère de l’Enseignement Technique, Professionnel, de la Formation METPFQE Qualifiante et de l’Emploi/ Ministry of Technical and Professional Education, Qualifying Training and Employment MMR Maternal Mortality Rate Ministère de la Santé et la Population/ Ministry of Health and MSP Population NDP National Development Plan NER Net Enrolment Rate PAP Plan d’Action Prioritaire/ Priority Action Plan Programme d’Analyse des Systèmes Educatifs de la CONFEMEN/ PASEC CONFEMEN Education Systems Analytical Program ii PER Public Expenditure Review PFM Public Financial Management PRAEBASE Projet d’Appui à l’Education de Base/ Basic Education Support Project Projet de Renforcement des Capacités de Transparence et de PRCTG Gouvernance/ Transparency and Governance Capacity Building Project PRSP Poverty Reduction Strategy Paper PTR Pupil Teacher Ratio Recensement Général de la Population et de l’Habitation/ General RGPH Population and Housing Census Schéma de Déconcentration de l’Enseignement Supérieur au Congo/ SDES Congo Higher Education Deconcentration Plan Secrétariat Exécutif Permanent du Conseil National de Lutte Contre le SEP/CNLS SIDA/ Fight Against AIDS National Council Permanent Executive Secretariat Système Intégré des Dépenses et des Recettes de l’État/ Integrated Fiscal SIDERE Revenue and Expenditure System SSA Sub-Saharan Africa Tableau d'Operations Financières de l'Etat/ State Financial Operations TOFE Board TVET Technical and Vocational Education and Training UIS UNESCO Institute of Statistics UNMG University Marien Ngouabi UNESCO United Nations Education, Scientific and Cultural Organization UNICEF United Nations Children's Fund WB World Bank WHO World Health Organization Regional Vice President: Makhtar Diop Country Director: Eustache Ouayoro Acting Sector Director: Tawhid Nawaz Sector Manager: Peter Nicolas Materu Task Team Leader: Cristina Isabel Panasco Santos iii Table of Contents Acknowledgements xi Part I – Overview I. Introduction 2 II. Country Context 3 Economic and Fiscal Context 3 Human Development in Congo 7 III. Public Spending on Education and Health 8 Education 8 Health 16 IV. Cross-Cutting Issues in Education and Health 28 V. Recommendations 30 Part II – Sector Reports Education Public Expenditure Review 34 I. Introduction 34 II. The Education Sector in Congo 35 Objectives of the Education Sector 35 Governance and Management of the Education System 36 Access and Quality 38 III. Spending in Education 43 Sources of Financing 43 Trends in Spending and Intra-Sectoral Allocations 44 Budget Execution 46 Financial Aid and Scholarships 47 Unit Costs 49 IV. Efficiency of the Education System 51 Internal Efficiency of Congo’s Education System 51 External Efficiency of Congo’s Education System 55 V. Equity in Education 60 Affordability of Education and Equity in Access 60 Equity in Attainment 62 Equity in Human Capital Stock and Income Holding 64 Spending in Education and Equity 65 VI. Conclusions and Recommendations 68 Health Public Expenditure Review I. Introduction 74 II. Institutional Environment 75 Objectives of the Health System 75 Organization of the Government Health System 77 III. Health Outcomes and Health Risks 79 Infant Mortality Rate (IMR) and Child Mortality Rate (CMR) 79 iv Maternal Mortality Rate (MMR) 82 Fertility and Malnutrition 85 Life Expectancy at Birth (LEB) 86 HIV/AIDS 87 Malaria 87 IV. Public Resource Mobilization and Sources of Finance 88 Government Budget for Health 88 Government’s Budgeting, Spending and Resource Allocation in the Health Sector 91 Health Budget Execution 93 Health Expenditure by Source 95 Catastrophic Payments for Health Care 101 V. The Delivery and Utilization of Healthcare 104 Inventory of Resources – Manpower, Inpatient Beds and Health Care Facilities 104 Regional Disparities and Efficiency 106 Analysis of Some Performance Variables in the Health Sector 110 Equity in Access 112 Public Spending and Equity 120 VI. Conclusions and Recommendations 122 Annex A.1 129 Annex A.2 141 Annex B.1 145 Annex B.2 154 References 157 v List of Boxes Part II – Education Public Expenditure Review Box 1.1: Structure of the Education System 37 Box 1.2: Internal Efficiency 51 Box 1.3: External Efficiency 55 Part II – Health Public Expenditure Review Box 2.1: On Quality of Service Delivery 106 Box 2.2: On the Performance Analysis 110 List of Tables Part II – Education Public Expenditure Review Table 1.1: Weight of actual education expenditure in the total expenditure and in the GDP, 2008-2013 44 Table 1.2: Intra-sectoral allocation, 2008-2013 44 Table 1.3: Budget execution rates (percentage) by Ministry, 2008-2012 47 Table 1.4: Public financial aid in education (percentage of total) by Ministry, 2008-2012 48 Table 1.5: Scholarships: value and number of beneficiaries (public sector), 2008-2012 49 Table 1.6: Primary GER and NER of autochthone population, circa 2010 62 Table 1.7: Summary of Issues and Recommendations 73 Part II – Health Public Expenditure Review Table 2.1: Infant, child mortality, and neo-natal mortality rates, 2011 in a group of SSA countries with per capita income above PPP in US$1,400 81 Table 2.2: Government health budget and execution, 2007-2013 94 Table 2.3: Health expenditures by financing agents by function and by providers, 2009 and 2010 (millions of XAF of each year) 96 Table 2.4: Total health financing in absolute amount and as a share of GDP, 2009-2010 (XFA and percentage) 97 Table 2.5: Household total and health spending around 2011 according to ECOM 2011 survey (millions of XAF) 100 Table 2.6: Incidence and intensity of catastrophic health payments, using non-food expenditure 103 Table 2.7: Utilization rates in Base Hospitals, by départment, 2012 (percentage) 108 Table 2.8a: Key health indicators in urban and rural areas, 2005-2011 116 Table 2.8b: Key health indicators in urban and rural areas, 2011 117 Table 2.9: Household total health spending around 2011 according to ECOM 2011 survey (millions of XAF) 118 Table 2.10: Summary of Issues and Recommendations 128 List of Figures Part I - Overview Figure 1: GDP per capita (constant 2005 US$) 4 Figure 2: GDP Annual growth rate 4 Figure 3: Annual growth of non-oil sectors 5 Figure 4: Government revenue growth, 2003-2013 5 Figure 5: Share of total government budget by sector, 2011-2013 6 Figure 6: Current and Capital expenditure execution rates, 2009-2012 6 vi Figure 7: Enrollment rates by gender and by level of education, 2005 and 2011 9 Figure 8: Retention rate by level of education 10 Figure 9: Evolution of budgeted education expenditure (million FCFA), by ministry (2008-2013) 11 Figure 10: Enrolment and budget allocations by education level, 2005 and 2011 12 Figure 11: Enrolment per education level and quintile, 2005-2011 14 Figure 12: Income holding per quintile, 2005 and 2011 15 Figure 13: Structure of Congo’s public health system 18 Figure 14: Infant and child mortality rates, 1997-2009 (deaths per 1,000 live births) 19 Figure 15: IMR according to mother's education and household wealth quintile, 2005 and 2011 2012 (deaths per 1,000 live births) 20 Figure 16: Concentration curve for the infant mortality rate, 2005 and 2011-2012 20 Figure 17: Government, MSP budget and budget execution (millions of XAF of Dec. 2012 and percentage) 21 Figure 18: Sources of health financing, 2010 22 Figure 19: Problems identified during last visit to health facility, if any 23 Figure 20: Health payment shares 24 Figure 21: Availability of health facilities by département 25 Figure 22: Expenditure in health by Government and households 27 Part II – Education Public Expenditure Review Figure 1.1: Enrollment rates by gender and by level of education, 2005 and 2011 38 Figure 1.2: Gender parity by level, 2005-2011 39 Figure 1.3: Access rate of official school age children for SSA countries, circa 2011 (out-of-school rate, percentage) 39 Figure 1.4: Regional comparison of Net Primary Enrollment Rate and Higher Education Access Rate 40 Figure 1.5: Percentage of repeaters in primary and lower secondary education in SSA, circa 2011 41 Figure 1.6: Retention rate by level of education 42 Figure 1.7: Budget allocations for selected sectors, 2008-2010, (percentage of total state budget) 43 Figure 1.8: Evolution of budgeted education expenditure (million XAF), by ministry (2008-2013) 45 Figure 1.9: Evolution of public and education expenditure (recurrent and investment), 2008-2012 46 Figure 1.10: Unit costs per education level, 2005 and 2011 50 Figure 1.11: Enrolment and budget allocations by education level, 2005 and 2011 50 Figure 1.12: Student school satisfaction primary and lower secondary, 2005-2011 (percentage) 53 Figure 1.13: Percentage of trained teachers and PTR in primary schools in SSA countries, circa 2011 53 Figure 1.14: Secondary education PTR, regional comparison, circa 2011 54 Figure 1.15: Educational attainment of the labor force, 2005 and 2011 56 Figure 1.16: Youth and adult literacy rates, 2011 57 Figure 1.17: Unemployment rate by level of education, 2005-2011 57 Figure 1.18: Rates of return by level of education, national and female, 2005-2011 58 Figure 1.19: Percentage of the working age population living below the absolute poverty line by level of education, 2011 59 Figure 1.20: Enrolment per education level and quintile, 2005-2011 60 Figure 1.21: Reasons for being out-of-school 61 Figure 1.22: Factors affecting educational attainment 63 Figure 1.23: Regional Comparison of gap between poorest and richest quintile in terms of net primary enrollment rate (percentage) 63 Figure 1.24: Average years of education of working age population by gender, income, region and proximity to school 64 Figure 1.25: Income holding per quintile, 2005 and 2011 65 vii Figure 1.26: Distribution of enrolled students and school age population by quintile 66 Figure 1.27: Benefit incidence analysis of public expenditure on education, 2011 67 Figure 1.28: Lorenz Curve for Household consumption expenditure and public spending on education by level 67 Figure 1.29: Education benefits by level 68 Part II – Health Public Expenditure Review Figure 2.1: Structure of Congo’s public health system 78 Figure 2.2: Infant and child mortality rates, 1997-2009 (deaths per 1,000 live births) 80 Figure 2.3: Treatment of diarrhea and malaria, 2005 and 2011-2012 (percentage) 82 Figure 2.4: Maternal mortality rate - mean and confidence interval, 2005 and 2011-2012 surveys (deaths per 1,000 live births) 82 Figure 2.5: Percentage of assisted deliveries by qualified health personnel, 2005 and 2011-2012 83 Figure 2.6: C-sections as a share of all deliveries, 2005 and 2011-2012 (percentage) 83 Figure 2.7: Services provided during pre-natal visits, 2005 and 2011-2012 (percentage) 84 Figure 2.8: Ideal number of children according to women, 2005 and 2011-2012 (number) 85 Figure 2.9: Chronic child malnutrition, 2005 and 2011-2012 (percentage) 86 Figure 2.10: LEB in 2011 86 Figure 2.11: Government budget within and outside of social sectors, 2008-2013 (millions of real XAF of Dec. 2012) 88 Figure 2.12: Nominal (US dollars) and real per capita (XAF of Dec. 2012) public health budget, 2009-2013 89 Figure 2.13: Government health spending as a share of GDP and of government spending, 2011 (percentage) 90 Figure 2.14: IMR and health expenditure: deviations from estimates based on per capita income (PPP-adjusted dollars) and schooling (literacy rate in percentage), 2011 91 Figure 2.15: Structure of the Government’s budgeting and expenditures system in the health sector 92 Figure 2.16: Government, Ministry of Health budget and budget execution (millions of XAF of Dec. 2012 and percentage) 93 Figure 2.17: Executed MSP budget per capita (XAF Dec. 2012) 93 Figure 2.18: Execution of Government health budget, 2007-2011 (millions of XAF of Dec. 2012) 95 Figure 2.19: Sources of health financing, 2009 and 2010 98 Figure 2.20: Structure of Government and households health expenditures by provider category and function, 2010 (millions of XAF of each year and percentage) 99 Figure 2.21: Structure of household annual out-of-pocket health spending, around 2010-2011 (percentage) 100 Figure 2.22: Health payment shares 102 Figure 2.23: Sub-Saharan African Countries: physicians per 1,000 people, 2011 104 Figure 2.25: Availability of health facilities by département 107 Figure 2.26: Population per integrated health center (CSI) by départment, 2012 107 Figure 2.27: Estimate of Government expenditure per capita by département 109 Figure 2.28: Relation between per capita public expenditure on health by department and selected indicators of use and performance 110 Figure 2.29: Per capita public expenditure on health and number of beds in public facilities by region 111 Figure 2.30: Reporting of illness and visits to health provider 112 Figure 2.31: Visits to health facilities by type of provider and expenditure quintile 112 Figure 2.32: Expenditure in health by Government and households 113 Figure 2.33: Problem identified during last visit to health facility, if any 113 Figure 2.34: Reason why did not visit health provider when needed during last illness episode 114 Figure 2.35: Child mortality rate according to mother's education and household wealth quintile, 2005 and 2011-2012 (deaths per 1,000 live births) 115 Figure 2.36: Concentration curve for child mortality rate, 2005 and 2011-2012 115 Figure 2.37: Access to malaria treatment by expenditure quintile 115 Figure 2.38: Reporting of illness and corresponding visit to health providers 117 viii Figure 2.39: Visits to health facilities by type of provider and rural/ urban location 118 Figure 2.40: Reason why dud not use health service when needed, 2011-2012 119 Figure 2.41: Reason why unsatisfied with service, 2011-2012 119 Figure 2.42: Concentration curves of Government spending on hospital and ambulatory care 120 Annex A-1 List of Tables Table A.1: Distribution of actual public current expenditure by education levels (in percentage of total) 130 Table A.2: Distribution of education benefits 130 Table A.3: Education budget execution rates (percentage) 131 Table A.4: Education Sector Budget 132 Table A.5: Education Sector Budget Execution 132 Table A.6: Budget Execution rates (current and investment per ministry) 133 Table A.7: Budget Execution rates at constant prices (current and investment per ministry) 133 Table A.8: Weight of each Ministry in the total expenditure (execution) 134 Table A.9: Share of each Ministry on the total education expenditure (execution) 134 Table A.10: Execution rates per ministry and per category (percentage) 135 Table A.11: External financial support 136 Table A.12: Input indicators by level of education (2011/2012) 137 Table A.13: Logistic regression of determinant of schooling 138 Table A.14: Employment and education indicators of the autochthone population (2007) 140 List of Figures Figure A-1: Share of investment and recurrent expenditure in total education expenditure and in METPFQE expenditure (percentage) 129 Figure A-2: Composition of the recurrent expenditure 129 FigureA.3: Distribution of employment per education level, 2005 – 2011 137 Figure A.4: Higher education benefits 139 Annex B-1 List of Tables Table B.1: Incidence and intensity of catastrophic health payments 150 Table B.2: Number of hospital deds in Base Hospitals, by départment, 2012 151 Table B.3: Number of beds days in Base Hospitals, by départment, 2012 151 Table B.4: Base Hospitals: utilization statistics, 2012 152 Table B.5: 2 National Hospitals: bed utilization statistics 152 Table B.6: Health care subsidies, constant unit subsidy assumption 153 List of Figures Figure B.1: Child vaccinations BCG and DTP, 2005 and 2011-2012 (percentage) 145 Figure B.2: Child vaccinations polio and measles, 2005 and 2011-2012 (percentage) 145 Figure B.2: Child vaccinations yellow fever 2005 and 2011-2012 (percentage) 146 Figure B.3: Knowledge of contraceptive methods, women in union, 2005 and 2011-2012 (percentage) 146 Figure B.4: Current use of contraception by women in union, 2005 and 2011-2012 (percentage) 147 ix Figure B.5: Median number of months since the previous delivery, 2005 and 2011-2012 (percentage) 147 Figure B.6: Percentage of pregnant women who received pre-natal care by qualified health personnel, 2005 and 2011-2012 (percentage) 148 Figure B.7 CMR and health expenditure in selected countries: deviations from estimates based on per capita income (PPP-adjusted US$) and schooling (literacy rate in percentage), 2012 148 Figure B.8 MMR and health expenditure in selected countries: deviations from estimates based on per capita income (PPP-adjusted US$) and schooling (literacy rate in percentage), 2012 149 Figure B.9 LEB and health expenditure in selected countries: deviations from estimates based on per capita income (PPP-adjusted US$) and schooling (literacy rate in percentage), 2012 149 x Acknowledgements ____________________________________________________________________ This Public Expenditure Review of the Education and Health sectors in the Republic of Congo was prepared by a team led Cristina Panasco Santos (Sr. Education Specialist, AFTEE). The following team members and consultants contributed to the Review as follows: Cristina Panasco Santos wrote Part I – Overview, and Part II – Education Report, with contributions from Rita Costa (Consultant, AFTEW), Kebede Feda (Human Development Economist, AFTEW), Vitor Dionizio (Consultant, AFTEW), and Luc Laviolette (Sector Leader, AFTHD). Rita Costa and Ricardo Bitran (Consultant, AFTEW) wrote Part II – Health Report with guidance from Hadia Samaha (Sr. Operations Officer, AFTHW). Etaki Wa Dzon (Consultant, AFTP5), Kirsten Majgaard (Education Economist, AFTEW) and Marie-Yvette Sacadura (Consultant, AFTEW) provided inputs to early drafts. Fulbert Tchana Tchana (Sr. Economist, AFTP5) provided inputs and guidance to the overall Review. Peter Materu (Sector Manager, AFTEW) and Luc Laviolette provided overall guidance and advice to the Team. Technical inputs were also provided by Deon Filmer (Lead Economist, DECHD) as well as by Keiko Inoue (Sr. Education Specialist, ECSH2) and Hebatalla Elgazzar (Sr. Human Development Economist, MNSSP) who peer reviewed the report. Sylvie Dossou (Country Manager, AFMCG) and Yisgedullish Amde (Country Program Coordinator, AFCCD) provided guidance to the finalization of the Review. The Team expresses its gratitude to government officials from the Ministère de l’Economie, des Finances, du Plan, de l’Intégration et du Portefeuille Public; Ministère de l’Enseignement Primaire, Secondaire et Alphabétisation; Ministère de l’Enseignement Technique, Professionnel, de la Formation Qualifiante et de l’Emploi; Ministère de l’Enseignement Supérieur; and Ministère de la Santé et la Population, who kindly provided data and were involved in the discussions around the preparation of the Public Expenditure Review. xi Part I Overview 1 I. Introduction 1. The development of a wealthier, literate and healthy society is a fundamental goal of Congo’s National Development Plan (NDP) 2012-16 and Poverty Reduction Strategy Paper (PRSP) 2012-16. Although, human development indicators have been improving in recent years, such progress has not accompanied the impressive economic growth of the country, and close to half of the Congolese population is still poor or extremely poor. Reaching the education and health Millennium Development Goals (MDG), institutional strengthening of both education and health systems, and improving the quality of service delivery in education and health provision are thus key aims of the NDP and PRSP 2012-16. The Congolese authorities have been developing and implementing policies and programs aimed at achieving such objectives. 2. Appropriate funding allocations and efficient use of funds in education and health are fundamental for the development of the sectors. With the end of the armed conflicts in 2003, Congo has progressively been increasing funding allocations to the social sectors. This is an important trend but does not suffice. Education and health systems require well-qualified staff, distributed appropriately by the various regions of the country, and with the necessary quality inputs to respond to the needs of the population. Further, education and health funds need to be used efficiently and in an equitable manner, ensuring that the poor receive the immediate and future benefits of education and health. 3. This Public Expenditure Review (PER) of the Congolese education and health sectors aims at providing inputs to improve efficiency and equity in spending in these sectors. This Review is a sectoral follow-up of the 2009 World Bank (WB) macro PER and it takes into account the following findings of the macro PER: (i) spending on the social sectors is still low although it has increased over time; (ii) the fiscal space generated by the increased oil revenues has largely boosted investment expenditure; and (iii) budget execution is low which contributes to lower the real level of public spending. While looking at such issues in detail regarding the education and health sectors, the PER is guided by the following research questions: a. Do budget allocations (level and composition) contribute to achievement of the set strategic education and health goals? b. Is there space for re-prioritizing and improving allocative and operational efficiency in the education and health sectors? c. Does public spending contribute to improve equity in access to quality education and health services? 4. To respond to these questions, the analysis in the PER uses data from various sources and makes use of various methodological approaches that provide inputs on sectoral performance, sectoral spending, efficiency and equity. Data sources include: (i) administrative data from the Ministère de l’Enseignement Primaire, Secondaire et Alphabétisation (MEPSA), the Ministère de l’Enseignement Technique, Professionnel, de la Formation Qualifiante et de l’Emploi (METPFQE), the Ministère de l’Enseignement 2 Supérieur (MES), and the Ministère de la Santé et la Population (MSP); (ii) administrative data from the Ministère de l’Economie, des Finances, du Plan, de l’Integration et du Portefeuille Public (MEFPIPP); and (iii) survey data: household and demographic and health surveys from 2005 and 2011. Among the analytical approaches used are benefit incidence analysis, performance analysis, and education rate of returns analysis, and catastrophic payments for health care analysis. Data limitations (which are discussed in each of the sector reports) limited the scope of the analysis. 5. The PER is divided in two main parts. Part I, in which this introduction is included, constitutes an overview of the two sector reports. Thus, it presents a brief analysis of the context of the country, a summary of findings of the education and health public expenditure reviews, a discussion on cross-cutting themes on spending in the two sectors, and a summary of recommendations. Part II includes the education and health public expenditure reviews. II. Country Context Economic and Fiscal Context 6. Congo is among the richest, on a per capita basis, and least populated countries in Sub-Saharan Africa (SSA). With a population of 4.3 million people, it is the region’s second most urbanized country after Gabon, with 62 percent of the population living in urban areas, mostly in the capital city of Brazzaville and in the port city of Pointe Noire. It has one of the lowest population densities in SSA, with 12 persons per square kilometer. About one- half of the population lives below the national poverty line (46.5 percent in 2011). 7. Macroeconomic performance has improved and economic resilience increased despite the global economic downturn, as a result of improved fiscal discipline and debt management. Despite oil dominance, macroeconomic stability is relatively well established in the Congo since 2008. The country has reached lower middle-income status some years ago, and the Gross National Income (GNI) per capita in 2012 was US$2,550. During the last decade, Congo’s economic growth has been higher than the world growth, and almost equal to SSA growth, yet lower than growth in low middle-income countries (LMICs). Between 2000 and 2013, on average, the global growth rate was about 2.6 percent, while Congo’s growth was about 4.5 percent; the average growth rate for SSA was about 4.9 percent. However, Congo is still lagging behind the LMICs, which have experienced strong economic growth, with an average annual rate of 6 percent during the same period. 3 Figure 1: GDP per capita (constant 2005 Figure 2: GDP Annual growth rate US$) 2,500 10 8 2,000 6 1,500 Percent 4 1,000 2 500 0 0 -2 2004 2005 2006 2007 2008 2009 2010 2011 2012 -4 Congo, Rep. 2001 2003 2005 2007 2009 2011 2013 Sub-Saharan Africa (all income levels) World Lower middle income Lower middle income Congo, rep Sub-Saharan Africa Source: World Development Indicators, 2013. 8. Congo’s economy is dominated by the oil sector which has been the main driver of growth, but economic diversification is key for sustainable growth. In nominal terms, the share of the oil sector in the Gross Domestic Product (GDP) has remained above 60 percent over the last five years. The lowest share was recorded in 2009, with 62 percent of GDP, while the highest was recorded in 2011 (mainly due to increase in prices) in which the sector accounted for 70 percent of GDP. Since then, with the oil price stabilization, its size has slightly declined and it is estimated at 63 percent of GDP in 2013. During the last three years, the non-oil sector grew at an average rate of 9.0 percent, also a sharp increase from an average growth rate of 5.4 percent during the 2005–2009 period, with all non-oil sectors experiencing positive growth from 2011 to 2013. The least growing sector, forestry and logging, grew at an average growth rate of 2.2 percent, while the highest growing sector, manufacturing, buildings and public works grew on average, at 9.4 percent. In 2011 and 2012, agriculture and livestock, manufacturing, and buildings and public works were the top performers among the non-oil sector, with almost double-digit growth rates. Economic diversification is very important in Congo to sustain the growth of non-oil sectors. Although Congo is still one of the main oil producing countries in SSA, the maturity of Congo’s oil fields is already resulting in a decline in oil production. In fact, in real terms, the share of the oil sector in Congo’s GDP has been falling since 2000, with an annual average rate of -2.2 percent, dropping from 43.6 percent in 2000 to 23 percent in 2013. 4 Figure 3: Annual growth of non-oil sectors 14 12 10 Percent 8 6 4 2 2003 2005 2007 2009 2011 2013 Agriculture, livestock, hunting and fishing Manufacturing Buildings and Public Works Transport and telecommunications Trade, restaurants and hotels Source: World Bank, IMF and National Authorities. 9. Oil revenues account for three fourths of Congo’s total revenues; however, government revenues from taxes have been growing steady in the last ten years. In 2013, government revenues decreased by -2.5 percent and by -0.4 percent in 2012. This came as a result of the sharp decline in oil revenues consecutive to a drop in oil production, which was not compensated fully by a strong increase of non-oil revenues. In 2013, the tax administration is estimated to have collected about 96 percent of the forecasted non-oil revenue, which is a good rate as a result of significant progress made in the revenues management accounting system. Figure 4: Government revenue growth, 2003-2013 120 100 80 60 Percent 40 20 0 -20 -40 -60 -80 2003 2005 2007 2009 2011 2013 Total revenue oil revenue Non-oil revenue Sources: World Bank, IMF and National Authorities. 5 10. From 2011 to 2013 government spending has also been increasing steadily . Total expenditure increased with an annual average rate of 21.5 percent from 2011 to 2013, consisting mainly of capital expenditure (62.0 percent on average). This expenditure represented 30.4 percent of GDP on average each year. Current expenditure increased on average by 6.4 percent, partly due to the response to the large explosion of an ammunition depot in Brazzaville which cause a high number of casualties and destroyed many infrastructures. Capital expenditure has been growing at a higher pace, 39.4 percent on average, reflecting the government's commitment to provide relevant infrastructure in support to economic activities. Spending with the social sectors has risen from 20.0 percent in 2011 to 22.5 percent in 2013, mainly due to the declaration by the Head of State of 2013 as l’Année de le Formation Qualifiant – the Year of Qualifying Education and Training year1. The infrastructure sector remained stable with more than 30 percent of the total budget all over the three-year period. Also, the share of Public Finance and Economic Affairs increased from 15.1 in 2011 percent to 18.7 percent in 2013, mainly driven by the sharp increase of the budget of the Ministry of Finance. The other sectors saw a decrease of their share. 11. Nonetheless, budget execution rates continue to be low. From 2009 to 2012, the execution rate averaged 93.4 percent with a low of 83.9 percent in 2012. The execution rate has been lower for investment spending (90.6 percent) than for current spending (95.4 percent). This low execution rate is explained by difficulties in procurement and disbursement. Since the adoption of the new public procurement code, lack of procurement specialists, resistance of some stakeholders involved in the procurement process, and the adoption of a lengthy procedure to move from the preparation into a tender for public contracts have posed some challenges. Further, the centralization of the process by the Ministry of Finance led to increasing delays in awarding of public contracts. The disbursement system also has many bottlenecks. It is not transparent enough. Figure 5: Share of total government Figure 6: Current and Capital budget by sector, 2011-2013 expenditure execution rates, 2009-2012 100 120 Sovereignty 90 110 80 100 70 Public Finance 90 60 and Economic 80 50 Affairs 40 Production 70 30 sector 60 20 50 Social 2009 2010 2011 2012 10 0 2011 2012 2013 Current expenditure Capital expenditue Source: World Bank and National Authorities. 1 Further discussion on this will be developed in the next sub-section. 6 Human Development in Congo 12. Despite the macroeconomic successes, poverty remains pervasive with about half of the 3.7 million Congolese living below the poverty line. As economic growth was not associated with a diversified economy, it has not led to job creation. The majority of the Congolese earns a living in the informal sector or is underemployed. Unemployment is high among the youth. According to the 2009 Employment and Informal Sector Survey (EESIC), in urban areas, 25 percent of the age group 15-29 years is unemployed. In the country-wide survey of 2005, the unemployment rate for the same age group was more than 40 percent. The unemployment rate declines significantly with age: only 5 percent of those over 50 years are unemployed. Women are more affected than men with higher unemployment rates especially in the 30-49 years age group. 13. Further, progress has been made in human development but important challenges remain. Congo still ranks very poorly in the Human Development Index; in 2013, Congo ranked 142 out of 187 countries. Progress towards the achievement of the MDGs 2 has been continuous but uneven. Congo is far from meeting MDG 1 (Eradicate Extreme Poverty) as almost half of its population still lives below the poverty line. The country has however seen more progress in achieving universal primary education (MDG 2) and in some aspects of MDG 3 (Promote Gender Equality and Empower Women). Most school age Congolese are in school with 12.2 percent of boys and 13.1 percent of girls being out-of-school. However, in spite of improvements, not all complete primary education, which completion rate in 2011 was 88 percent. Gender parity has seen a significant improvement in all education levels. Recent data from the latest household and demographic and health surveys point out to important improvements in meeting MDGs 4 and 5 with important progress made in reducing child mortality and improving maternal health. Progress has been taking place with regards to MDG 6 however malaria is still endemic in the country and responsible for a large number of deaths every year. HIV/AIDS rates are low (3.3 percent). 14. In conclusion, in recent years Congo experienced high rates of economic growth mainly due to oil revenues. The economy is little diversified and there is need for increased diversification as oil reserves have reached maturity and oil production is dropping. Steps have been taken in such direction and improvement of growth rates have been seen in some non-oil sectors. Economic prospects for Congo are good; there is need however, for a reflection of such success in human development and poverty reduction. Almost half of the Congolese live below the poverty line and earn their living in the informal sector. Progress has been made in human development, but more is required. The Congolese authorities are aware of the need to invest in the social sectors and spending, although slowly, has been increasing, even if overall country budget execution rates are still low. The next section provides a summary of the public expenditure reviews of the education and health sectors, discussing in more detail results, spending trends, efficiency and equity in the sectors service delivery. 2 MDG 1 – Eradicate Extreme Poverty; MDG 2 – Achieve Universal Primary Education; MDG 3 – Promote Gender Equality and Empower Women; MDG 4 – Reduce Child Mortality; MDG 5 – Improve Maternal Health; MDG 6 – Combat HIV/AIDS, Malaria and Other Diseases; MDG 7 – Ensure Environmental Sustainability; MDG 8 – Global Partnership for Development. 7 III. Public Spending in Education and Health Education Good progress in access but important challenges on quality and retention 15. Education development in Congo is guided by four main objectives expressed in the NDP and PRSP 2012-2016. These are: (i) ensuring universal primary education for all by 2015 (in line with MDGs 2 and 3); (ii) improve retention in primary and secondary education while improving the flow of students through the cycles; (iii) develop technical and vocational education in line with market needs and economic diversification; and (iv) develop quality higher education in line with market demands and priority sectors growth. In order to meet these objectives primary education is free, textbooks and learning materials are distributed to schools, civil works increased the number of primary schools in the country, and teacher training programs have been implemented; 2013 was declared as l’Année de le Formation Qualifiant with the aim of supporting physical and technical upgrading of technical and vocational education and training (TVET); and new legislation on higher education is being launched to regulate quality assurance and governance mechanisms for the sub-sector.3 16. Since 2005, good progress has been made in terms of provision and gender equity; less progress has been made with regards to completion and quality. Currently, close to all school-age Congolese boys and girls enroll in primary education. In 2011, primary gross and net enrolment rates (GER and NER) were of 116 percent and 88 percent, respectively. Completion rate was 88 percent and gender parity was achieved; further, important improvements in gender parity have also taken place at other education levels with gender parity achieved in lower secondary and within reach (0.9) in upper secondary and higher education. With regards to quality, the latest results from the Program d’Analyse des Systèmes Éducatifs de la CONFEMEM4 (PASEC) (2007) position Congo as a low performer in comparison to other countries such as a Cameroon. Congo’s performance at fifth grade was slightly below the average for PASEC countries in French (34 percent), and well below the performance of Cameroon (55 percent), while slightly above the average in mathematics. 3 Congo’s education system includes several sub -sectors as follows: 3 years of pre-primary education, followed by 6 years of primary education; secondary education includes 4 years of lower secondary and 3 years of upper secondary. TVET is delivered at the secondary education level, at Centres des Métiers (2 years) and Collèges d’Enseignement Technique (2 years) at the lower secondary level; and at Lycées de Enseignement Technique (3 years) et Écoles Profissionnelles at the upper secondary level. Higher education is organized by the system Licence, Matrise, Doctorat (first degree, master and doctorate). Each sub-sector is under the mandate of a specific ministry. Pre-primary, primary, secondary and non-formal education fall under the mandate of the Ministère de l’Enseignement Primaire, Secondaire et Alphabétisation (MEPSA); technical and vocational education and training fall under the mandate of the Ministère de l’Enseignement Technique, Professionnel, de la Formation Qualifiante et de l’Emploi (METPFQE). The Ministère de l’Enseignement Supérieur (MES) has the mandate on higher education. 4 Program d’Analyse des Systèmes Éducatifs de la CONFEMEM stands for Program of Analysis of CONFEMEN Education Systems. And Confereration des Ministres de l’Éducation des États et Gouvernments de la Francophonie (CONFEMEN) stands for Confederation of the Ministers of Education from States and Governments of Francophone Countries. 8 Figure 7: Enrollment rates by gender and by level of education, 2005 and 2011 133% 2005 2011 140% 124% 118% 114% 120% 98% 94% 100% 85% 81% 80% 67% 61% 60% 43% 35% 40% 14% 12% 20% 8% 4% 0% Male Feamle Male Female Male Feamle Male Feamle Primary Lower sec upper sec Tertiary Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 17. However, student retention is very low. Repetition is still very high, close to 25 percent in primary education, which combined with the high cost of post-primary education (supported by the households), contributes to low student retention, even if enrolment rates have been increasing in time. Repetition remains high in post-primary education, reaching 18.4 percent in lower secondary (Collège) and 17.2 percent in upper secondary (Lycée). Repetition is a key factor in drop out which is the main reason for being out of school. Congo out-of-school numbers have decreased in time and the out-of-school rate (12 percent) is below the average for SSA (26 percent). Of the total out-of-school children, 5 percent were never in school and 7 percent dropped out. Eight percent of primary school age Congolese children were out-of-school in 2011, and the respective percentages for lower secondary and upper secondary age children were 10 and 26 percent. Congo is amongst the worst SSA performers with regards to repetition, comparing favorably only to Burundi, but it is a top performer with regards to out-of-school children, favorably comparing to its neighboring Cameroon and Democratic Republic of Congo. 9 Figure 8: Retention rate by level of education 100 90 92.24 2005 2011 80 70 60 51.11 50 40 30 20 23.11 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Primary Lower secondary Upper Secondary Post Secondary Source: Authors: Estimations calculated from ECOM 2005 and ECOM 2011 data. Increasing funding for education but still below the SSA average, change in intra-sectoral allocations favoring TVET and investment expenditure, extremely high unit costs for higher education 18. Funding for education has been steadily increasing and the sector is mostly funded by public budget; the private sector is important. The weight of education in the country’s budget, as a percentage of total public expenditure, has increased from 9.8 percent to 7.8 percent (executed) between 2008 and 2012, and to 12.9 percent in 2013 (programed). As a percentage of GDP this corresponds to an increase from 2.0 percent to 2.8 percent (and to 5.1 percent in 2013). The recurrent budget covers salaries, scholarships and learning materials, while the investment budget covers mostly infrastructure and equipment. Financial aid represents an important share of the total recurrent expenditure in the education sector, and most of this is used in higher education scholarships. No maintenance of capital goods is covered. Household contributions support salaries of bénevoles5, food, some furniture, school maintenance, and examination fees. External financing has focused mainly on technical assistance and support to infrastructure development. Over the period of 2008-2012, external financing contributed with an amount of approximately US$70 million, which represented about 5 percent of total financing for the education sector. Although there is no direct public subsidizing of private education provision, the salaries of teachers in private accredited schools are paid by public budget. In 2011, these teachers represented 20 percent of all primary school teachers and 4 percent of all secondary teachers paid by the state. In fact, private education plays an important role in Congo; 31 percent of primary school Congolese children are enrolled in a private school, which is a high rate when compared to 16.6 percent that is the average for Sub-Saharan Africa. Private higher education provision enrolls almost half (44 percent) of all higher education students. 5 The bénevoles are primary and lower secondary teachers hired by the communities, often with very limited capacity and qualifications. Most of them were hired during the period of the conflicts to fill the void left by public provision of education services. 10 19. In 2013, a shift in intra-sectoral funding allocations took place benefiting TVET and increasing significantly the overall investment expenditure in education . Until 2012, a large proportion of the education budget was allocated to MEPSA, in line with the objective of achieving the MDGs in 2015. An increasing attention to the goal of developing skills for growth sectors led to a very important increase in the funding allocation to the METPFQE. This accompanied the Head of State declaration of 2013 as l’Année de le Formation Qualifiant. As a result, MEPSA’s allocation decreased, while the allocation for the METPFQE doubled. The 2014 budget does not change this new intra-sectoral allocation pattern. Allocations to higher education have not changed significantly during the period. Further, until 2012, both the METPFQE and MES were allocated an increasing share of recurrent expenditure, whereas MEPSA benefited from a higher share of capital expenditure. In 2013, 83 percent of the 2013 budget allocated to the METPFQE was on investment6. If the budget for 2013 was implemented as planned, capital expenditure in education was higher than recurrent expenditure. Figure 9: Evolution of budgeted education expenditure (million FCFA), by ministry (2008-2013) 250,000 200,000 150,000 100,000 MES METPQE 50,000 MEPSA 0 Current Investment Current Investment Current Investment Current Investment Current Investment Current Investment 2008 2009 2010 2011 2012 2013 Source: MEFPIPP Budget Execution and other data, 2008-2012 and Loi de Finances 2008-2012. 20. The execution rates of the consolidated education budget improved noticeably over the last years, both for recurrent and investment expenditure, although the latter are systematically lower than the previous; further, they vary among the three education ministries. The education budget consolidated execution rate increased from 89.9 percent to 94.6 percent between 2008 and 2012. The execution rate for investment expenditure is lower than for recurrent expenditure, but this rate showed the most remarkable improvement, increasing from 39 percent in 2008 to 83.4 percent in 20127. The lowest budget 6 The percentage of investment expenditure budget for the METPFQE for 2014 is slightly lower, but still very high: 77 percent. 7 It should be noted that year 2012 was atypical due to the passage of a supplemental budget, following the government’s response to the immediate effects of the explosion of the ammunitions depot in Brazzaville. The supplemental budget increased the recurrent expenditure by over 46 percent and doubled the investment expenditure comparing to 2011. In practice, these expenditure levels were kept practically unchanged, in 11 execution rates are found in METPFQE and MEPSA, and in investment expenditure8. The major increase in METPFQE investment expenditure can pose additional challenges to investment and consolidated execution rates. MES seems to be facing overruns both in recurrent (5 percent in 2012) and investment expenditure (14 percent in 2011), although more significantly in the latter. 21. Except for TVET, unit costs have been decreasing, yet, higher education unit costs are disproportionately high. The main drivers for unit costs in TVET and higher education are teacher salaries and scholarships. Spending on salaries accounted for 36 percent and 56 percent on average, respectively; spending on scholarships accounted for 26 percent and 37 percent on average. In the case of higher education, the low enrolment rate combined with the large amount spent on scholarships justifies the disproportionately high unit cost for this education level. Figure 10: Enrolment and budget allocations by education level, 2005 and 2011 Enrollment 2005 Enrollment 2011 Budget 2005 Budget 2011 61% 54% 34.9% 30% 27.5% 26%25% 26% 16% 16.1% 12% 12% 10.5% 8% 11% 10.2% 2% 5% 3% 1% 2% 3% 1% 2% Preschool primary lower sec upper sec TVET Higher Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. Although financing of the sector increased, there are important sources of inefficiency preventing many Congolese to benefit from education returns which are very high in Congo 22. Costs of low student retention and high repetition are very high for both public financing and households. In 2011, grade repetition costs amounted to about 0.6 percent of GDP, close to 21 percent of the annual current education expenditure (at the 2011 prices)9. School dropout has very important implications in public and household expenditure and nominal terms in the budget for 2013, thereby setting in a new expenditure base. The impact of these changes in the education sector was basically felt on the investment side, which rose 69 percent between 2011 and 2012. 8 It was not possible to provide an analysis of the causes explaining the lower execution rates due to the unavailability of recent data on the composition of investment expenditure. 9 The of cost of grade repetition is calculated based on: (i) the direct cost of schooling that is generated from total number of repeaters based on per student annual unit cost in public and private schools, and (ii), discounted value of forgone opportunity costs of expected earnings due to lag and length of labor market engagement. The latter is estimated based on wage employment earnings by taking into account the age of labor market entry and associated unemployment rate. 12 income. The current dropout rate in Congo (7 percent, two percentage points lower than in 2005) implies an opportunity cost of 3.3 percent of GDP, and 10 percent of household total consumption expenditure10. 23. Poor management of human resources is another important source of inefficiency. With regards to share of trained teachers and pupil teacher ratio (PTR), Congo ranks in the middle of SSA countries, comparing favorably with countries such as Cameroon and Togo. In 2011, the primary school PTR for Congo was 44 to 1. In the same year, the share of primary trained teachers was 80 percent. However, the ratio teacher-administrative staff is very high in Congo: 1.5 teachers in relation to non-teaching staff in primary education; 1.7 in the Collège and 3.3 in the Lycée. Further, the number of teachers with more than one function is very high: 21 percent of the teaching staff in primary and 6.7 percent in the Collège accumulate teaching and administrative functions; 10.8 percent of the total number of primary school teachers are also school directors. The financial implications of these overlapping functions are not known, as the level of aggregation of the data from the Ministry of Finance does not allow for an accurate analysis. The latest primary education teaching census dates from 2008 and at the time it allowed for the identification of 5,148 ‘ghost’ teachers, whose salaries were frozen, and 1,672 ‘ghost’11 personnel which salaries were also suspended. Further, 2,253 staff from other ministries was identified under MEPSA’s budget. 24. Internal inefficiency prevents many Congolese to benefit from education and improve their living conditions – education rates of return are very high in Congo. Between 2005 and 2009, the contribution of education to a change in earnings was on average 91 percent; this was much higher than the average for any other associated labor market factors (35 percent). The highest impact results from holding a TVET or higher education diploma. Incomplete primary and complete primary produce much lower rates of return compared to any other education level. Holding of an upper secondary education diploma or higher education is more important for women than men. The employment profile varies accordingly with the education level, and thus those with higher education are mostly employed in wage jobs and earn more than the others. In 2011, 80 percent of those with a higher education diploma were in wage employment against are 55 percent, 56 percent, 36 percent, 21 percent, 14 percent, 10 percent, for those with a TVET diploma, secondary, lower secondary, primary education, incomplete primary, and no education, respectively. Given that currently, out of 100 Congolese children enrolled in grade 1 only 23 reach grade 10, i.e. the last grade of lower secondary education, the remaining 77 are prevented from benefiting from education and can be caught in the intergenerational cycle of poverty. Inequity is a major challenge in education: primary education is pro-poor but post-primary education benefits the well-off – post-primary education is not affordable by the poor 10 The earnings of individuals by level of education are estimated to determine the forgone opportunity costs. This is done by reviewing the earning difference between those completing the level of education and those who dropped out before completing such level. 11 ‘Ghost’ teachers and ‘ghost’ personnel are individuals who appear in wage lists, yet no longer are or have never been part of the ministry staff. 13 25. While enrolment in primary education decreases with income, the opposite is the case for enrolment in post-basic education levels; disparities of income are also present in attainment. This suggests that children from poor households benefit much less from post- basic education. While it is true that improvement has taken place from 2005 to 2011, the proportion of children from poor households in secondary and higher education is still extremely low. Further, during this period, there have been high increases in the number of children from wealthiest households in secondary, and particularly, in higher education. In terms of quality, PASEC national results hid statistically significant differences of income. Figure 11: Enrolment per education level and quintile, 2005-2011 Q1 Q2 Q3 Q4 Q5 60.0 50.0 40.0 30.0 20.0 10.0 - Higher Higher Primary TVET Primary TVET preschool preschool upper secondary upper secondary lower seconsdary lower seconsdary 2005 2011 Source: Estimate based on ECOM 2005 and ECOM 2011. 26. Disparities in educational attainment are very high in Congo compared to other SSA countries. For example, the gap between the poorest and richest quintiles in upper secondary enrollment is 78 percent in Congo compared with 59 percent which is the average for SSA. Congo is amongst the 6 worst performers in a group of 37 SSA countries. Further, it compares poorly as well with SSA countries of a similar income, for which the average is 68 percent. However it is a better performer than Gabon and Cameroon, which is last of the group. 27. Income, area of residence and gender play a role in the average years of education of the working age population in Congo, and income and rural urban disparities are very significant. While the average number of school years for working age Congolese in the top quintile is 10.2, the equivalent for those in the lowest quintile is 5.3. Different regions of the country also present different results, with Brazzaville presenting 9.9 years of schooling (highest) and Pool only 4.7 years (lowest). Given education disparities among the young Congolese and among those of working age, these seem to be replicated in an intergenerational manner. At the national level, only 17 percent of the total wealth is hold by the lowest quintile, which represents 40 percent of the Congolese population. A regional comparison of 27 SSA countries for which data is available positions Congo among the 10 14 high income high inequality countries. Further, analysis of 2005 and 2011 household surveys data shows that the income holding the bottom 40 percent slightly decreased. This implies that, the poorer seem to be getting poor in Congo. Figure 12: Income holding per quintile, 2005 and 2011 120 Q1 Q2 Q3 Q4 Q5 100 80 45 44 60 40 22 23 16 16 20 10.9 10.8 0 6.1 6.1 2005 2011 Source: Authors estimations from data from ECOM 2005 and ECOM 2011. 28. Current spending in education does not favor the poor: while it is pro-poor in primary education, it is regressive in post-primary levels. In 2011, the poorest quintile received 21.1 percent of the public benefits allocated to primary (just slightly above the population share of the quintile), while the richest received 16 percent (4 percent less than the population share of the quintile). For post-basic education, however, the situation is different with 22 percent of the budget allocated to higher education and per student spending 12 times higher than for any other level. Compared to 2005, the benefit to poorest households decreased, even in primary education, and increased for the richest households. 15 In conclusion 29. After the end of the conflicts in the 2003, the Congolese authorities have steadily been investing in the development of the education sector guided by a comprehensive set of goals, including the achievement of the MDGs for the sector, and the establishment of an education system that is capable of providing the necessary skills for a diversified economy. Good results have been obtained in access however less progress has been seen in quality, and above all, in student retention. Drop out and repetition are very costly for both the public budget and households, and an important source or inefficiency, along with poor human resources management. Financing of the sector has focused on primary education in line with the objective of achieving the MDGs, but this focus has shifted towards TVET, in line with the goal of providing skills for growth. Revamping TVET institutions has increased significantly investment expenditure. Given the lower rates of investment budget execution, this change in intra-sectoral allocations can decrease the consolidated budget execution rate of the sector. Overall, budget execution rates have been improving but are still low. Congo’s returns to education are very high and education is a fundamental means to reduce poverty. However, financing of education is progressive in primary and regressive in post-primary. Thus, issues of inequity are fundamental, and the education system in Congo is not yet paying a real contribution to poverty reduction and equality. Policies and interventions are required to reverse this situation. Health Good recent progress in major health outcomes but important challenges in terms of inequality and infectious diseases 30. In Congo, the improvement of the health conditions of the population is considered fundamental in achieving poverty-reduction and economic growth outcomes and this can be clearly seen in its Poverty Reduction Strategy Paper 2012-16. The PRSP 2012-16 and the Programme National de Développement Sanitaire12 (PNDS) – currently being updated - define key areas of intervention in the sector. Priority is given to the achievement of the Millennium Development Goals for health: child infant mortality, maternal health, fighting HIV, malaria and other infectious diseases, by guaranteeing equitable access to quality healthcare services by the population. Important attention is also given to the general strengthening of the health system in terms of governance. An important focus is given to reducing inequities in access to health services for the poor and for vulnerable pregnant women. Strengthening the supply of healthcare by promoting health services, social markets, and communication is also a priority. The plans also recognize the importance of improving the quality of services and of stronger management of essential medicine supply, including strengthening the Congo Essential Generic Drugs Agency (COMEG). 12 Programme National de Développement Sanitaire stands for National Program for Sanitary Development. 16 31. Raising public expenditure in health and ensuring an efficient use of resources is an important means for improved health outcomes, thereby contributing to poverty reduction and increased equity. Balanced intra-sectoral budget allocations can play an important role in ensuring that priority sector policies are implemented. Gains in efficiency allow for improved health outputs and outcomes. Although a middle-income country, in 2011 Congo’s poverty rate was 46.5 percent. Equity in access to public health services is crucial to guaranty that such a large section of the population is not simply excluded. 32. The government health care delivery system is formally decentralized and comprises three hierarchical levels: (i) central, (ii) departmental 13, and (iii) operational. At the bottom of the public delivery system are the health dispensaries. According to the recent Health Statistics Yearbook issued by the Ministère de la Santé et Population14 (MSP), in 2012, there were 345 dispensaries distributed across all departments, except in the two largest urban centers in the country, Brazzaville and Pointe Noire. The next referral level is composed of Centre de Santé Integré CSIs)15. The first referral facility for CSIs is the Base Hospital (Hôpital de base). In 2012, all departments except Kouilu had one or more Base Hospitals. These inpatient facilities, in turn, have as their referral, in the public system, the so- called General Hospital (Hôpital general), of which there are six in the country, including the University Hospital Center (Centre hospitalier universitaire) located in the capital city of Brazzaville. Private health care providers also play a major role in Congo ’s health system, but there is little coordination between public and private providers and little regulation of the latter. According to a World Bank/ International Finance Corporation report that analyzed Congo’s private healthcare delivery system, in 2005 there were 1,712 health care providers in the country, of which more than half (1,002) were private. The vast majority of private providers (88 percent) were for profit. 13 The territory of Congo is divided into regional areas called departments, thus, the word departmental is used in this case, in relation of each of the 12 Congo départments. 14 Ministère de la Santé et Population stands for Ministry of Health and Population. 15 Centre de Santé Integré stands for Integrated Health Center. 17 Figure 13: Structure of Congo’s public health system Minister of Health’s Office, general and central directorates, public Central level autonomous health entities, and specialized public health programs Departmental Health Directorate National National Departmental level DDS (DDS) Hospital Hospital 12 DDS 6 National Hospitals Socio-Sanitary Circumscriptions CSS (CSS) Base Hospital 27 Base Hospitals Integrated Peripheral and Health Center operational level (CSI) 306 CSI 36 CSS CSI Health Dispensaries 345 health posts Health Dispensaries Source: CNSEE (2005) and CNSEE and ICF International (2012). 33. The pharmaceutical products procurement agency, COMEG, has the responsibility of procuring and supplying government hospitals and health centers with generic drugs and consumables. Its management capabilities and performance are limited and many public providers, including Base Hospitals and CSIs prefer instead to bypass the institutional procurement system in order to purchase drugs and supplies directly from private providers, at lower prices. The lack of regulation and control of the private drugs procurement system poses patient safety problems. 34. In terms of health outcomes, up until recently Congo was lagging behind other countries in the region. Prior to the release of the most recent Demographic and Health Survey, EDCS 2011-12, Congo seemed to be outperformed by other countries in the region in most health indicators. Accordingly, the 2010 government’s Millennium Development Goals report stated that the Congo had made little or no progress toward the MDG goals of reduced infant/ child mortality (Government of Congo 2010), for example. 35. In contrast with these trends, the EDCS 2011-12 report presented an encouraging picture. By comparing the estimates of the infant mortality rate (IMR) and child mortality rate (CMR) between the previous EDCS (2005) and the most recent ones (2011-12), the report concluded that both of these had decreased, and that their improvement was statistically significant and accelerating. It estimated that in 2009, the IMR was 39 deaths per 1,000 live births, the CMR was 68, and the death rate for children ages 1-5 was 30 (Figure 15). The maternal mortality rate (MMR) has also fallen considerably from 781 deaths per 100,000 live 18 births reported in 2005, to the rate of 429 reported in the EDCS 2011-12. This improvement is likely related to the positive performance of several indicators related to maternal health, including Congo’s high rate of institutional deliveries by qualified personnel, which has improved from 2005 to 2011-12, but still shows important inequalities across wealth quintiles. Figure 14: Infant and child mortality rates, 1997-2009 (deaths per 1,000 live births) Infant mortality rate Child mortality rate Congo: Taux de mortalité infantile et infanto-juvénile, 1996-2009 140 131 Mortality rate (per 1000 live births) 117 120 118 Child mortality rate (0-5 years) 100 99 80 60 54 68 45 40 48 30 44 Child mortality rate (1-5 years) 20 - EDSC-I (2005) EDSC-II (2011-2012) Source: CNSEE (2005) and CNSEE and ICF International (2012). 36. Significant improvement can also be observed for most other health indicators presented in the EDCS 2011-12: child vaccination (except for polio); access to the treatment of diarrhea with oral rehydration salts for children under five years; rate of institutional deliveries by qualified health personnel; knowledge about contraception; use of pre-natal care; and HIV prevalence, among other indicators. 37. It is also important to keep in mind, however, that the indicators mask important inequalities in outcomes for different populations. Particularly across wealth quintiles and different areas of residence, with poorer households and those living in rural areas presenting worse outcomes. Infectious diseases like malaria also continue to pose significant challenges, despite the progress. 19 Figure 15: IMR according to mother's Figure 16: Concentration curve for education and household wealth the infant mortality rate, 2005 and Congo: Concentration curve for the infant quintile, 2005 and 2011-2012 (deaths 2011-2012 mortality rate, 2005 and 2011-2012 per 1,000 live births) 100 140 134 120 80 100 Cumulative % of infant deaths 80 IMR 63 60 60 44 40 32 20 40 0 Q1 (poorest) None Q2 Q3 Q4 Primary Secondary 1st cycle Secondary 2nd cycle Q5 (richest) 20 0 0 20 40 60 80 100 Mother's education Wealth quintiles Cumulative % of the population EDSC 2005 EDSC 2011-2012 EDSC 2005 EDSC 2011-2012 Equality line (base 100) Source: CNSEE and ICF International (2012) and World Bank’s DataBank. 38. It should also be noted that a few indicators that are critical for sustained growth are still lagging in Congo. Fertility has increased: the average number of children born per woman was 5.1 in EDCS 2011-12, an increase from 4.8 in 2005. The country also still has a quarter (24 percent) of its children under five years of age that are chronically malnourished (stunted), with two thirds (67 percent) of children in the same age group suffering from anemia. Both fertility and malnutrition rates are higher in households living in rural areas and households from the poorest income quintiles. Increasing public funding for health but still among the lowest in SSA, high dependency on households health spending which creates an access barrier against the poorest households 39. The analysis of the government expenditures shows that MSP budget remained nearly constant, in real per capita terms, between 2008 and 2011 . On one hand, the share of the government budget devoted to the social sectors dropped sharply in 2009 but subsequently experienced a mild but progressive increase. Between 2008 and 2009 the social sectors’ share of the public budget fell by one-half from 19.6 percent to 10.0 percent, but from then on, it increased progressively, albeit mildly, to reach 13.3 percent in 2012. On the other hand, within the social sectors, the MSP has been gaining importance. In 2009 the MSP budget represented about 75 percent of the budget for Education but by 2012, the MSP and Education budgets were similar. In 2011 the MSP budget represented 3.8 percent of total 20 government budget, a slight increase from 3.4 percent in 2010. This percentage increased to 5.3 percent in 2012 and decreased again to 4.2 percent in 2013. 40. Still, when seen in the regional context, in Congo both total health expenditure as a share of GDP and government health expenditure as a share of total government spending are very low, in relation to the country’s per capita income. Whereas Congo’s per capita income in international dollars is among the highest in Sub-Saharan Africa, the share of the economy that is devoted to the health sector is among the lowest, and so is the share of government spending going to health. Despite recent increases, Congo’s proportion of government expenditures allocated to health falls short of the Abuja Declaration commitment of increasing government funding for health to 15 percent of government’s total expenditure. Figure 17: Government, MSP budget and budget execution (millions of XAF of Dec. 2012 and percentage) Source: MSP budget information supplied by the MEFPIPP. 41. It is important to keep in mind however, that health system performance is influenced by many variables other than health spending. One way of analyzing the situation is to use a target outcome such as IMR and assess a country’s performance in terms of its deviation from both predicted IMR and predicted total health spending as a percentage of GDP, using linear regression techniques. A good performing country would be one with an IMR better than predicted and total health spending below prediction. Congo is a poor performer from that analysis: it spends on health less than expected, given its income level and literacy rate, and it has an IMR that is higher than expected given those two explanatory variables. 21 42. The degree of actual execution of the MSP budget has historically been weak but with an upward trend. In 2007, 2008, and 2010, the execution of the health budget was under 80 percent. In 2009, only 42.7 percent of the budget was executed. The year 2011 saw the highest level of execution, at 90 percent. Between 2007 and 2011, the total amount of executed health budget increased in real terms from 71.679 million XAF (of Dec. 2012) and 107.784 million XAF. 40. A National Health Accounts (NHA) study has been completed with support from the World Bank and the World Health Organization (WHO), with information for the years 2009 and 2010. It shows that besides government expenditure, health system financing in the Congo is heavily dependent on households’ out-of-pocket spending. According to the NHA study, in 2009, 48 percent of all financing for health care came from households; in 2010 it fell to about 37 percent. Estimates obtained by these authors using the latest ECOM (2011) household survey indicate that household health spending may have been largely underestimated in the NHA study and may actually represent more than one-half of total health pending. On the other hand, donor financing for health care is relatively low for regional standards and other sources of financing, such as community contributions and spending by enterprises are also very small overall at the national level. Figure 18: Sources of health financing, 2010 Source: Ministère de la Santé et de la Population (2013). 22 41. Strong reliance on household spending in Congo is partly the consequence of a drug revolving fund-like system that operates in all or virtually all government health facilities. Aside from autonomous public hospitals, which receive a block grant and may use part of their budget to purchase medicines, all other public providers lack budget resources to purchase medicines and other medical supplies. To have a stock of medicines they charge user fees to patients and with the revenue collected from these fees they purchase medicines from COMEG or private providers. Moreover, the fact that the government has not issued any guidance regarding the services and products that can be subject to a fee and the level of fees has resulted on each facility adopting its own fee system. This means that user fees in public facilities may be imposed not only on medicines, but also on other goods and services provided to patients and that they vary from one facility to another. All of this challenges equity in the system with poor households having less access to treatment and, especially, medicines due to financial constraints. Indeed, the most common cause of discontentment with the health system/ reason for not visiting a health facility when needed is the high price charged, a problem that is a lot more common among the poorest households (Figure 20). 42. Looking at the structure of household out of pocket spending (OOPS) by category obtained from the 2011 ECOM, medicines are the major expenditure category, accounting for more than half (56 percent) of household spending on health . Spending on actual medical services, in the form of fees paid to public and private providers, represents just 20 percent of the households OOPS. Figure 19: Problems identified during last visit to health facility, if any 40% 35% 30% 25% 20% 15% 10% 5% 0% not waiting expensive lack of lack of treatment other at least welcoming time trained medicines did not one staff work problem Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Source: ECOM and 2011. 23 43. Analysis of catastrophic payments for health care carried out by the authors indicate that OOP payments for health care in Congo absorb more than one quarter of household resources net of food costs in 0.9 percent of households, with this percentage reaching 2 percent for the lowest quintile of income. Graphically, Figure 21 highlights that the percentage of households spending more than 20 percent of their non-food income in health is reduced. Even though such numbers are low and can be considered common in developing countries, they mean that for some of the poorest families such levels of spending can only be accommodated through the diversion of considerable resources from current consumption and/ or through the accumulation of debt, or the exhaustion of savings and assets with long-term consequences for household welfare. Figure 20: Health payment shares 1.4 Health payments as % of household expenditure 1.2 1 0.8 OOP/total 0.6 exp. OOP/nonf ood exp. 0.4 0.2 0 0.0 0.2 0.4 0.6 0.8 1.0 Cumulative proportion of households ranked by decreasing health payments budget share Source: Authors calculations based on ECOM 2011. Important challenges in system management remain: regional inequalities in the deployment of staff, distribution of facilities and resources and low bed utilization rates in base hospitals; however there is some evidence of correlation between different levels of spending per capita in each of the regions and utilization of the system 43. Human resource management remains one of the main challenges in Congo’s health system. Weak management has led to a poor deployment of professional staff and functionality of facilities, especially in rural and remote areas. The number of health personnel in all categories has, however, increased significantly, from 10,000 in 2005 to about 15,000 in 2010 (of which 80 percent are in the public sector). Unfortunately, this important 24 recruitment effort has not succeeded in addressing the existing major urban-rural disparity issues in service delivery. When comparing Congo with other Sub-Saharan African countries, one can see that with 0.1 doctors per 1,000 people, the country has fewer medical doctors than would be expected given its per capita income and it also presents a deficit of hospital beds and nurses with just 0.84 nurses and 1.6 hospital beds per 1,000 people. Yet Congo has a huge asset in terms of health personnel which is to be found within its diaspora; they are estimated to be more numerous than those working inside the country. 44. There is broad variation in the availability of health facilities across départements. While the availability of dispensaries is hard to evaluate since factors like population density and the degree of rurality of the departments can influence regional needs. In the case of CSI, MSP policy calls for one CSI per 2,500 to 10,000 people in rural areas, and one CSI per 10,000-15,000 people in urban areas so it is straightforward to say that at least three départements, Brazzaville, Bouenza, and Pointe Noire do not meet the norm. Figure 21: Availability of health facilities by département 50 Facilities per 100,000 people 45 40 35 30 25 20 15 10 5 0 Dispensaries / Health posts CSI with additional services offered (CSI à PMAE) CSIwith minimum services offered (CSI à PMAS) Reference hospitals (Hôpitaux de référence) Source: MSP 2013. 45. When looking at estimates of government per capita expenditure by region we can also see strong disparities. The département where the government spends the most per person (Cuvette) on health is spending three times more than the lowest-spending département (Kouilou). This is not necessarily bad, the levels of rurality in different départements are diverse and providing health services to rural, widespread populations might result in a much higher cost per capita then in urban settings (we would need more information to evaluate). In the same way, the burden of disease as well as the exact needs of the populations might vary across départements, and that can influence spending patterns. 46. In order to assess the levels of efficiency of the system, the authors analyzed the existence of any correlation between different levels of spending per capita, in each of 25 the regions and variables that approximate the use of the system (numbers of visits) and health outcomes in each of the regions. This analysis is illustrative only, pending more regular and reliable collection of health data at the département level. The analysis reveals that there is some evidence of a positive relationship between public per capita spending and indicators of health system performance, such as number of visits to public health facilities (as reported in the ECOM 2011 survey), the use of contraceptives and the percentage of women that received pre-natal care and that took anti-malarial drugs during pregnancy. However, the positive relationship does not hold for all outcomes. Important indicators on children’s health like the percentage of fully immunized children show negative correlation with per capita spending. There is also evidence of a positive correlation between number of beds in public facilities and public spending on health per capita – this suggests that resources are located where people are looking for them the most. 47. Utilization rates of functional beds are very low in Base Hospitals, except in Brazzaville. Hospital utilization rates were as low as 7 percent in one department (Sangha), and in eight out of 11 departments they were below one-third. Considering that utilization was computed on the basis of functional hospital beds, these low rates of utilization reveal a major inefficiency in Congo’s government health system - scarce public resources are being allocated to hospitals which remain little used. 48. Low access to curative care may be the result of relatively high user fees, of user fees being charged to all patients in government facilities irrespective of their ability to pay, of poor quality of health services, and other problems in supply. A study about quality of care in government ambulatory and inpatient facilities found that while most CSIs studied had basic medicines at the time of the visit, more than half had experienced inventory stock outs in the 30 days preceding the survey. It also found that few facilities had treatment protocols and few health staff had been trained in the use of protocols (PDSS 2010). 49. In Congo, the prevalence of health problems does not vary much across quintiles, according to self-reported survey data about illness in the previous four weeks (ECOM 2011). In the same way, the percentage of respondents that reported seeking help from some type of health provider did not differ much across the wealth quintiles. This seems to be a good sign in terms of equity – poorer individuals are not excluded from the system and know where to seek health care. It can also be a sign of the high level of urbanization of the country; according to the same survey a vast majority of the population (85 percent) lives within five kilometers of a health facility. When looking at the type of health providers sought, however, it is clear that the population in richer quintiles has more access to private care and uses less traditional medicine healers. The use of public facilities, however, does not vary much across quintiles, which is also a positive sign in terms of equity of access to public services. 50. In what concerns expenditure, on the government side there is not much variation across quintiles (slightly larger for the 5th quintile), which is also a good sign in terms of equity. However, as we have seen in Figure 23 richer quintiles spend a lot more of their own money in absolute terms in obtaining health services. As discussed, there is a real possibility that households in the poorer quintiles are not accessing all the services and 26 medicines they would need for lack of financial resource. Indeed, when asked about their level of satisfaction with the service received at health facilities, the most common problem identified is the fact of service being too expensive. This complaint is, unsurprisingly, more common in the lowest quintile of income. Similarly, when analyzing the reasons why respondents did not go to a health facility when they became ill, high prices is the number one reason mentioned, particularly for the bottom quintiles of income. In fact, benefit incidence analysis carried out by these authors indicates that government spending with ambulatory care is pro-poor, while government spending in hospitals is slightly pro-rich. 51. Rural populations report a slightly higher rate of illness than urban ones (40 percent versus 36 percent) but they also report a higher use of health services for the reported illnesses. This indicates that services are available and populations know how to access them even in rural areas. In terms of providers, there is less reliance on private health services in rural areas and more reliance on informal providers such traditional healers and churches. Figure 22: Expenditure in health by Government and households 60,000 Millions CFA 40,000 20,000 - 1 2 3 4 5 Quintiles Government Household Source: ECOM 2011 and MSP. In conclusion 52. In recent years, the Congolese authorities have steadily been paying attention to the health sector guided by a comprehensive set of goals, including the achievement of the Millennium Development Goals for health: child/ infant mortality, maternal health, fighting HIV, malaria and other infectious diseases. Despite the fact that financing for the health sector is still below regional standards for a country of its level of income and that the health budget has remained nearly constant, in real per capita terms, between 2008 and 2011, recent survey data shows positive results in most, albeit not all, health outcomes. Strong progress has been recorded in reducing child and maternal mortality, but there remain high rates of fertility and of child malnutrition. HIV/AIDS prevalence has also been reduced but malaria remains a significant public health problem. 53. The limited resources (human, physical and financial) available in Congo are not distributed evenly across the country, which contributes to inequalities in access and outcomes across different populations. The excessive price of services, especially medicines, seems to create a barrier to the poorest households and those in rural areas. The low utilization 27 of beds in base hospitals and limited use of services in general suggests efficiency problems in the system. Health financing in general and particularly the purchase of medicines in health facilities heavily relies on household spending which contributes to inequalities. Finally, the system still presents serious challenges in terms of efficient management of staff and physical and financial resources. In addition to increases in public spending on health, policies and interventions focusing on strengthening system management, with an equity lenses, are crucial in the near future. IV. Cross-Cutting Issues in Education and Health 54. During the past decade, Congo has made progress in the delivery of education and health services which has been reflected in improvement of key indicators . Education and health are key priority sectors in Congo’s NDP and PRSP 2012-2016 and several well defined policies and interventions have been and are expected to be implemented to meet the defined targets, among those, the MDGs. In education the most significant progress regards access and gender parity, which position Congo well with regards to achieving the MDGs for the sector close to 2015. In health, positive trends in several indicators position Congo well with regards to MDGs 4 and 5. However, achievement of the education and health goals has been uneven. In education challenges remain with regards to student flows and appropriateness and quality of post-basic education. In health, malnutrition is still high and so is the case for fertility. It is very important for Congo to sustain the gains while improve performance in areas still lagging behind. 55. Financing for both education and health is still low but has been steadily increasing, although this is more the case for education than health; further, household contributions are very important. Financing of the education sector has increase in recent years and there have been intra-sectoral changes in line with the government decisions around the pursuit of specific education objectives. Thus, a shift towards the development of skills for growth sectors led to an increase in the budget allocation to the TVET sub-sector, and to a reduction in the allocation for primary and secondary education. The education budget increased from 9.8 percent to 7.8 percent (executed) between 2008 and 2012, and to 12.9 percent in 2013 (programed). As a percentage of GDP this corresponds to an increase from 2.0 percent to 2.8 percent (and to 5.1 percent in 2013). In 2009, total health financing amounted to about 103,000 million XAF, equivalent to 2.1 percent of GDP. Both total health financing and GDP have been going up, so total health financing as a share of GDP has been more or less constant around 2 percent between 2008 and 2012, with the highest share being recorded in 2012 at 2.8 percent. Both the education and health budgets are low for the income level of Congo. It is also worth noting that initiatives such as year of health and year of qualifying education and training lead to significant increases in sector budgets, and to changes in intra-sectoral allocations, not always associated with sound planning and with real absorptive capacity of the ministries. While such initiatives are commendable as they focus on moving forward with the improvement of human development, and give visible to the sectors, they should be accompanied of sound planning and be implemented in a sustainable manner. While donor financing is very low in both education and health budgets, household 28 contributions are very important, particularly in health, reaching almost half of the overall budget for the sector in 2008. Households pay for medicines and palliative care. In education household contributions pay fees, salaries of bénévoles, school maintenance and materials post-primary education. Recurrent spending in education and health support mostly staff and administrative services. 56. As is the general case in Congo, budget execution rates are low both for the education and health budgets, and particularly so for investment. Budget execution rates have improved but continue low. Education budget execution rates are higher than health ones, but the impressive increase in investment expenditure for the development of TVET may decrease the consolidate execution rate for the sector. The challenges encountered by other sectors, and associated with the introduction of the new procurement code and the centralization of procurement for large works, have also been seen in the education and health sectors. 57. There are important sources of inefficiency in both sectors, one of which being poor management of human resources, common to education and health . Quality of human resources is an issue for both sectors, and so is distribution. Further, in education, the high proportion of administrative staff combined with the high number of staff with multiple functions, has an important impact in the wage bill, even if it is difficult to access it due to the data limitations. In health, the availability of doctors and nurses is below the average for similar income countries, there are important geographic inequalities in distribution of those resources and there is limited information about the quality of the services delivered. A major source of inefficiency in the education system is very low retention, high repetition and dropout rates which have very important costs for both households and public budget. In health, the lack of regulation on the fees applied by healthcare providers and the challenges around procurement of drugs have important inefficiency effects. 58. Inequity is an issue in both education and health. Post-primary education is non- affordable to the poor, yet completing primary education does not produce sufficiently important returns. Low student retention and poor quality of learning outcomes prevent young Congolese to obtain a diploma that could allow them to benefit from wage employment. Further, financing for post-basic education is regressive and thus moving the poor away from the possibility of better income. Spending on higher education scholarships produces extremely high unit costs for this level, and favors the well-off as very few poor, if any, reach tertiary education. In health, the excessive out-of-pocket spending in care and medicines prevents the poor from reaching good quality health services. Analysis of catastrophic payments for health care highlights that for some of the poorest families the levels of spending with health can only be accommodated through the diversion of considerable resources from current consumption and/ or through the accumulation of debt, or the exhaustion of savings and assets with long-term consequences for household welfare. Further, challenges remain with regulation of fees applied by providers, with a direct impact on poor households. 59. Data issues are common for both sectors, and further analysis is required to better understand the sectors. The education sector does have a fully operational 29 information and management system. Administrative data is not always reliable, timely and complete. Further, there is limited information on human resource management, a key area that requires improvement. Budget data is also challenging. Budget categories change in time, and data aggregation often limits the extent of the analysis to be made. Data on health require further confirmation as the results of the most recent DHS, although more reliable than the administrative data, present a very important discrepancy from other data. Further, while in the education sector it is easier to analyze progress in outcomes in relation to expenditure, this is not the case in health. Although progress is not simply explained by expenditure, Congo has been a poor performer, and little expenditure increase has taken place that can justify some of the progress observed. Improvements in information and management systems in health and education are very important. V. Recommendations 60. Increase public financing for education and health in a planned and sustainable manner. Further, improving education performance and health status and financial protection in Congo’s education and health sectors necessarily calls for increased volumes of financing. This would be in line with international recommendations such as the Abuja Declaration by which African Union Members pledged to commit 15 percent of their budgets to health; and the Global Partnership for Education recommendation of 20 percent budget allocation to education. 61. Maintain and further increase the rate of budget execution through better procurement and disbursement procedures. Strong budget execution begins with how the budget is formulated. This should be a process in which a range of stakeholders are consulted. A more decentralized budget management process may increase execution. In the case of health, the block grants currently allocated to national hospitals should be evaluated to determine the efficiency consequence of such grants. In the case of education, intra-sectoral allocations should be balanced and take into account financial sustainability. Large works are procured and paid for outside the line education and health ministries. Improvements in procurement of large works and payments of large contracts overall in the country, will contribute to better execution rates in the education and health sectors, too. Such improvements are urgent. 62. Implement measures to improve quality of services and improve the allocation of human and physical resources throughout the country. In health, The PBF approach which will be scaled up nationally, has been shown to generate improvements in quality through its incentive structure which is focused not only on quantity of services delivered, but also tracks and remunerates against quality. Further, increasing financing to ambulatory health providers could be considered. A growing share of public financing should also be allocated to the poorest departments in the country. Policies to ensure a more equal geographic distribution of health staff should also be put in place. In education, further knowledge on teacher distribution, bénévoles, and ‘ghost’, and administrative staff is required to defined efficiency driven policies on human resources distribution and improve the payroll. Measures to improve teacher quality and quality education inputs are also required. 30 63. Track carefully the evolution of inequality in health and education and adopt policies, as needed, to bridge the gap between the poor and non-poor. Defining a national policy to identify the poor and vulnerable, in order to waive them from user fees in government health facilities that might help reduce inequity in access to basic health services. Ensure that schools receive allocations for maintenance and that all bénévoles are integrated in government payroll to effectively implement the fee free primary education policy. As completion of lower secondary education is fundamental, consider incentives to the poor to ensure their retention in the system, along with measures to decrease repetition and dropout rates. Allocating a growing share of public financing to the poorest departments in the country, and to urban areas with high concentration of poor residents, might help achieve these goal. 64. Carry out national surveys at regular intervals and build capacity for statistical analysis. Given the importance of data for decision making and the gaps and discrepancies in the data that is available in Congo to undertake public expenditure reviews, a new survey should be conducted in the near future to verify that the gains reported in the EDCS 2011-12 are maintained or furthered. Also, the government and its development partners should continue to support in a systematic way the initiative that led to the production of the 2012 Statistical Yearbook, as well as the improvement of the education and information management system. This involves strengthening of institutional capacities in health and education information management systems, training of staff involved in data collection and reporting, and the supply of computers and other equipment required to operate such systems. 65. Other sector specific recommendations include: Education a. Consider a trade-off between allocations to TVET and secondary education to increase the allocation to the latter. Completion of lower and upper secondary education produces the highest benefits to the Congolese. TVET unit costs are very high and will benefit a very low proportion of the population. Further, international experience shows that the most successful TVET provision occurs when within an overall framework of close partnership between the public and private sectors. Such partnership should be promoted in Congo both at the technical and financial level, so as to align supply and demand of skills and to allow for the development of secondary education. b. Revision of recurrent and investment expenditure for higher education . Budget overruns have been recurrent in higher education. This needs to be resolved. Further, the excessive expenditure with higher education scholarships leads to very high unit costs for the sub-sector. A review of the scholarship policy and implementation is recommended for new policy on the matter. c. Define and implement a teacher training program to improve the teaching skills and knowledge of all teachers (including bénévoles), and review pre-service teacher training. Quality of learning outcomes depends a great deal on the quality of the teaching and learning process. In order to improve learning outcomes, teacher 31 training programs customized to the existing teaching staff should be developed, so as to improve knowledge and teaching skills of staff in the system and bénévoles. Further, a progressive reform of pre-service teacher training could also be considered. d. Establish a planning and monitoring system for teacher distribution . This would allow for a distribution of human resources according to geographic and population needs, while also improving the efficiency of the system by identifying ‘ghost’ personnel. e. Additional studies. These include a teachers’ and administrative staff census, a study on causes of repetition; and a study on the impact of an incentives package to support the poor in attending secondary education. Health a. Introduce innovations in provider payment methods. So far, Congo has not innovated much with regards to provider payments, beyond the block grants and performance-based financing (PBF) which is planned for national scale-up. Despite anticipated implementation challenges, it is likely that PBF will result in higher output, better quality and more equitable access. b. Increase the share of the government budget that is allocated at the CSI level . The share currently allocated to the CSI level is exceedingly low and signals a major problem of allocative efficiency. It should be a priority to progressively expand the share of public resources going to health dispensaries and health centers. This would allow CSIs to waive poor and vulnerable patient fees, and it would serve to attract more qualified health staff to these facilities through higher salaries and the adoption of economic incentives. c. Define a specific benefits package for health providers in Congo. Expanding and rationalizing government health spending and moving towards Universal Health Coverage (UHC) may call for the explicit definition of a benefits package for Congo. There has been some experience (i.e., Cordaid) in Congo to define and cost a benefits package. An actuarial study of this or an alternate package will be necessary to determine the volume of public and private financing required to deliver explicit benefits. d. Focus on policies that can compensate for the heavy reliance on out-of-pocket spending. This could start with a review of the implementation of ongoing programs that offer free health services. There should also be regulation of fees to ensure that they are not abusive, that they are waived for the poor, and that they are charged only for medicines and not for other services. e. Additional studies. Carry out specific studies assessing hospital occupancy, the reasons behind low occupancy rates, and the feasibility of closing some beds and on human resources needs. 32 Part II Sector Reports 33 Education Public Expenditure Review I. Introduction 66. The education sector in general and primary education in particular, is considered fundamental in achieving the poverty-reduction and economic growth outcomes in the Congo Poverty Reduction Strategy Paper (PRSP) 2012-16. Education policy and strategy are guided by Congo’s National Development Plan (NDP) 2012-16. The PRSP 2012-16 and the Document de stratégie sectorielle de l'éducation 2008-202016 define key areas of intervention in the sector. Priority is given to the achievement of the Millennium Development Goals (MDG) for education, i.e. to completion of universal primary education and gender parity. However, important attention is also given to post-basic education in line with the need to develop education and training in an aligned manner with labor market demands and economic diversification. 67. Raising public expenditure in education and ensuring an efficient use of resources is an important means for improved educational outcomes, thereby contributing to poverty reduction and increased equity. Balanced intra-sectoral budget allocations can play an important role in ensuring that priority sector policies are implemented. Gains in efficiency allow for improved education outputs and outcomes. Although a middle income country, in 2011 Congo’s poverty rate was 46.5 percent. Equit y in access to education and retention can help improve overall equity in the country. 68. This chapter examines recent achievements in educational outcomes at various levels, the trends and composition of public expenditure in the education sector as whole and in sub-sectoral allocations for the period of 2008-2011, and provides insights on the sector internal and external efficiency; particular attention is provided to issues of equity. The chapter is guided by three main research questions. a. Do budget allocations (level and composition) contribute to achievement of the set strategic education goals? b. Is there space for re-prioritizing and improving allocative and operational efficiency in the education sector? c. Does public spending contribute to improve equity in access to quality education service? 69. The analysis uses a combination of methods to respond to the research questions : (i) key sector performance indicators were calculated to assess progress towards sector priorities and MDGs; (ii) budget data was analyzed so as to provide insights on sector and intra-sectoral allocations, composition of the budget and rate of execution; (iii) quality of 16 In the text, this document will be referred to as the education sector strategy. A review and updating of the strategy was initiated early 2014 under the leadership of the education authorities and the oversight of UNICEF. It is worth noting that implementation of the Strategy was never accessed and thus, it is not clear of how far associated interventions may have had any impact. Further, the main policy for the sector (free primary education) preceded the strategy. 34 education, retention, and use of resources were analyzed so as to provide information on the sector’s internal efficiency; (iv) external efficiency was discussed based on an analysis of rates of returns to education and educational attainment in the labor market; finally, (v) an analysis of disparaties in access to education was used to provide insights on equity along with an analysis of household contribution to education spending, education levels in the human capital stock, and a benefit incidence analysis. Annex A provides information on the various methods used in the analysis. 70. Several data sources were used for the analysis in the chapter. These include administrative data from the three ministries in charge of the education sector, namely the Ministère de l’Enseignment Primaire, Secondaire et Alphabétisation (MEPSA), the Ministère de l’Enseignement Technique, Professionnel, de la Formation Qualifiante et de l’Emploi (METPFQE), and the Ministère de l’Enseignement Supérieur (MES); administrative data from the Ministère de l’Economie, des Finances, du Plan, de l’Integration et du Portefeuille Public (MEFPIPP)17; and data from the 2005 and 2011 household surveys. Annex A also provides information on the data used and its limitations. 71. The chapter is organized in five main sections . Section II provides an overview of the sector and of sector progress; Section III focuses on sector expenditure and spending; Section IV discusses the internal and external efficiency of the education sector; Section V provides an analysis of equity and education, and finally, Section VI provides conclusions and recommendations for the next five years. II. The Education Sector in Congo Objectives of the Education Sector 72. Education is one of the priority sectors under Pillar 4 – Social Development and Inclusion of the PRSP 2012-2016. Both the PRSP and the Document de stratégie sectorielle de l'éducation 2008-2020 set the following key objectives for the sector: a. Ensure universal primary education for all by 2015 in line with the MDGs. b. Improve retention in primary and secondary education while improving the flow of students through the cycles. c. Develop technical and vocational education in line with market demands and economic diversification. d. Develop quality higher education in line with market demands and priority sectors growth. 73. In order to achieve the objectives above, several programs and interventions were proposed, and some are under implementation. Civil works for primary education 17 MEPSA stands for Ministry of Primary, Secondary Education and Literacy; METPFQE stands for Ministry of Technical and Professional Education, Qualifying Training and Employment; MES stands for Ministry of Higher Education; and MEFPIPP stands for Ministry of Economy, Finances, Plan, Integration, and Public Portfolio. 35 along with interventions aimed at improving the quality of service delivery have been under implementation. These include teacher training, distribution of textbooks, incentives for minority children to attend school (including distribution of school uniforms and learning materials), among others. School based activities, such as support to school-based management, and further teacher training are proposed to improve retention levels and improve the flow of students between cycles. These programs were originally launched to support the implementation of the 2007 free primary education policy. Updating of technical and vocational education provision through improved facilities and updated curricula is also a proposed intervention, along with a reform in higher education focusing on improving quality. In order to support the implementation of some of these reforms, 2013 was declared the Year of Qualifying Education and Training by the Head of State. During 2013 a new law regulating higher education was also submitted to Cabinet. Governance and Management of the Education System 74. The mandate on education is shared between the MEPSA, the METPFQE, and the MES. MESPA is responsible for primary and secondary education and literacy programs. Primary education is of 6 years, followed by 4 years of lower secondary and three years of upper secondary education. Technical and vocational education and training (TVET) is offered primarily at the technical secondary education level, with the first diploma reached 4 years after completed primary schooling (Brevet d’Étude Technique—BET)18. Additional diplomas build upon the BET and therefore require even more years of schooling. TVET falls under the administrative mandate of the METPFQE established in 2002 to support the development of this education level. Higher education has recently adopted the system License, Master and Doctorate and it is managed by MES. Box 1 provides a detailed summary of the structure of the education system. 75. Education management in Congo is highly centralized at the ministries level, despite some recent decentralization steps. MEPSA and METPFQE are organized by directorates at the regional level, which constitute an intermediate layer in terms of administrative and pedagogical coordination between central administration and schools. Devolution to the regional directors of education of some responsibility over pedagogical guidance and staff allocation has been taking place, however, overall staff management is centralized and schools have very limited autonomy. Each ministry individually implements its own policy, and there is no coordinating body overseeing policy making and implementation for the overall sector. The establishment of a Conseil National d' Éducation19 to play such coordinating role is included in the education sector strategy however this has not yet been established. 18 Brevet d’Étude Technique stands for certificate of technical studies. 19 Conseil National d' Éducation stands for National Education Council. 36 Box 1.1: Structure of the Education System Number Year of years Cycle Type of institution and degree of study 26 8 3rd cycle of higher 25 7 Doctorate education studies 24 6 Faculties, 23 5 2nd cycle of higher Institutes and Master 22 4 education studies Grands Ecoles 21 3 1st cycle of higher 20 2 First Degree education studies 19 1 18 terminale Lycee Professional 17 1st Upper secondary Lycee technique Schools 16 2nd 15 3rd College 14 4th technique Lower secondary College 13 5th Centre de 12 6th metier 11 CM2 10 CM1 9 CE2 Primary Primary 8 CE1 Schools 7 CP2 6 CP1 5 P3 Pre-primary 4 P2 Pre-primary centers 3 P1 Source: MEPSA, METPFQE, and MES. 76. The private sector plays an important role in education delivery in the Congo . During the armed conflicts of the 1990s, communities and private sector filled in the void in education delivery left by the public sector. Communities hired local primary and lower secondary teachers (les bénévoles), often with very limited capacity and qualifications, and paid their salaries. Private schools at all levels opened in the country. As a result, 31 percent of primary school Congolese children are enrolled in a private school, which is a high rate 37 when compared to 16.6 percent that is the average for Sub-Saharan Africa (SSA). Private higher education provision enrolls almost half (44 percent) of all higher education students. There are also accredited schools (écoles conventionées), which are basically confessional schools; nonetheless they represent a small percentage within the overall private sector provision in the country. Access and Quality 77. From 2005 to 2011, there were important improvements in access in all levels of education. Most school age Congolese children are enrolled in primary education (primary gross and net enrolments rates (GER and NER) of 116 percent and 88 percent, respectively, for 2011) and good progress has been made in all other education levels. However, the number of school age Congolese attending secondary (lower and upper) education is still low (NER of 49 percent for lower secondary and of 24 percent for upper secondary). While primary education is free (school fees were abolished in 2007, as mentioned previously) and textbooks and learning materials are distributed to schools by the MEPSA, this is not the case for post-primary education for which families are forced to pay an important financial contribution. Enrollment in higher education has more than doubled in the period of 2005- 2011, in association with an increase in provision mostly by the private sector. Data on Technical and Vocational Education and Training (TVET) is scattered, but this level has covered on average only 0.3 percent of all Congolese students. Figure 1.1: Enrollment rates by gender and by level of education, 2005 and 2011 133% 2005 2011 140% 124% 118% 114% 120% 98% 94% 100% 85% 81% 80% 67% 61% 60% 43% 35% 40% 14% 12% 20% 8% 4% 0% Male Feamle Male Female Male Feamle Male Feamle Primary Lower sec upper sec Tertiary Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 78. Access to primary education is high in Congo for boys and girls. Gender parity in access to primary education is achieved, and progress towards gender parity in post-primary education is even more significant (Figure 1.2). In recent years, a significantly higher number of girls have enrolled in secondary and higher education making gender parity attainable at these levels in the near future. In fact, with regards to access to primary education (for both boys and girls), Congo compares well with countries such as Namibia and has access rates higher than the average for SSA, and for some of Congo’s neighboring countries, such as 38 Cameroon and Gabon (Figure 1.3). Further such high access also indicates a low number of out-of-school children, both boys and girls.20 Figure 1.2: Gender parity by level, 2005-2011 2005 2011 1.5 1.0 1.1 1.0 0.9 0.9 0.9 1.0 0.7 0.6 0.5 0.0 Primary Lower secondary Upper secondary tertiary Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. Figure 1.3: Access rate of official school age children for SSA countries, circa 2011 (out- of-school rate, percentage) 60.0 50.0 Male Female 40.0 30.0 27.5 20.0 13.1 10.0 0.0 Niger Zambia DRC SSA Togo Rwanda Malawi Cameroon Chad Namibia Burundi Mozambique Senegal Guinea South Africa Sao T&P Mauritania Cote d'Ivoire Ethiopia Benin Swaziland Gabon Ghana Mali SLE South Sudan Sudan Kenya Uganda Nigeria Tanzania Burkina Faso Congo, Rep Zimbabwe Comoros Liberia Lesotho Madagascar Gambia Lower/Upper middle Source: Authors computation using ECOM 2011 for Congo Rep. and similar household surveys in other countries: Benin (2010), Burkina Faso (2010), Burundi (2010), Cameroon (2011), Chad (2011), Cote d'Ivoire (2011), Comoros (2004), DRC (2010), Ethiopia (2011), Gabon (2011), Gambia (2010), Ghana (2010), Guinea (2012), Kenya (2008), Lesotho (2011), Liberia (2010), Madagascar (2010), Malawi (2010), Mali (2010), Mauritania (2008), Mozambique (2009), Namibia (2010), Niger (2011), Nigeria (2010), Rwanda (2010), Sao T&P (2010), Sierra Leone (2011), Senegal (2011), South Africa (2012), South Sudan (2009), Sudan (2009), Swaziland (2010), Tanzania (2010), Togo (2011), Uganda (2010), and Zambia (2010), and Zimbabwe (2011). 79. Although not all young Congolese complete primary education, the number of completers has increased in the last decade. Primary completion rate increased from 72 percent in 2005 to 88 percent in 2011. Primary female completion rate also showed good progress, increasing from 69 percent to 77 percent in the same period. Thus, even if Congo will not be able to attend the MDG, it compares very favorably with most SSA countries. As Figure 1.4 highlights, by presenting a regional comparison of the net primary rate (which is a proxy for primary completion rate) and higher education access rate, Congo compares with 20 Further discussion on out-of-school children in Congo will be developed later in the chapter. 39 strong performers such as Cape Verde and performs better than its neighboring countries such as Gabon. Figure 1.4: Regional comparison of Net Primary Enrollment Rate and Higher Education Access Rate Source: Authors computation using ECOM 2011 for Congo Rep. and similar household surveys in other countries: Benin (2010), Burkina Faso (2010), Burundi (2010), Cameroon (2011), Chad (2011), Cote d'Ivoire (2011), Comoros (2004), DRC (2010), Ethiopia (2011), Gabon (2011), Gambia (2010), Ghana (2010), Guinea (2012), Kenya (2008), Lesotho (2011), Liberia (2010), Madagascar (2010), Malawi (2010), Mali (2010), Mauritania (2008), Mozambique (2009), Namibia (2010), Niger (2011), Nigeria (2010), Rwanda (2010), Sao T&P (2010), Sierra Leone (2011), Senegal (2011), South Africa (2012), South Sudan (2009), Sudan (2009), Swaziland (2010), Tanzania (2010), Togo (2011), Uganda (2010), and Zambia (2010), and Zimbabwe (2011). 80. Access to education and completion of primary education improved in recent years in Congo; however challenges remain with regards to student repetition, which on average is one of the highest in SSA. Repetition is high for all education levels, but particularly so for primary education. Close to 1 in 4 primary school Congolese children is a repeater. Repetition rates reach 18.4 percent in lower secondary (Collège) and 17.2 percent in upper secondary (Lycée). Such high rates place Congo amongst the worst performers in SSA countries, with only Burundi presenting higher repetition rates for primary and lower secondary education (33.1 percent for primary and 25.2 percent for lower secondary). Congo’s rates are similar to those of neighboring Central African Republic (CAR) (22.7 percent for primary and 15.0 percent for lower secondary) but much higher than Cameroon (12.3 percent for primary and 11.4 percent for lower secondary) or the Democratic Republic of Congo (DRC) (11.2 percent for primary and 11.9 percent for lower secondary) (Figure 1.5). 40 Figure 1.5: Percentage of repeaters in primary and lower secondary education in SSA, circa 2011 35 primary Lower secondary 30 25 20 15 10 5 0 Guinea-… Sierra… Mozamb… Equatori… Maurita… Cabo… Burkina… Madaga… Côte… Niger Chad Gambia Sudan Guinea Zambia Burundi Malawi Ghana Ethiopia Sao T&P Benin Lesotho Kenya Uganda Rwanda Namibia CAR Mauritius Eritrea Togo Tanzania Liberia Congo Senegal Swaziland Mali DRC Angola Cameroon Source: Authors’ computation using ECOM 2011 for Congo Rep. and similar household surveys in other countries: Benin (2010), Burkina Faso (2010), Burundi (2010), Cameroon (2011), Chad (2011), Cote d'Ivoire (2011), Comoros (2004), DRC (2010), Ethiopia (2011), Gabon (2011), Gambia (2010), Ghana (2010), Guinea (2012), Kenya (2008), Lesotho (2011), Liberia (2010), Madagascar (2010), Malawi (2010), Mali (2010), Mauritania (2008), Mozambique (2009), Namibia (2010), Niger (2011), Nigeria (2010), Rwanda (2010), Sao T&P (2010), Sierra Leone (2011), Senegal (2011), South Africa (2012), South Sudan (2009), Sudan (2009), Swaziland (2010), Tanzania (2010), Togo (2011), Uganda (2010), and Zambia (2010), and Zimbabwe (2011). 81. Although some small improvement has occurred, still only 50 Congolese children out of 100 enrolled in grade 1 reach grade 6 (the last grade in primary), and only 23 reach grade 10 (the last grade in lower secondary education). Further, given that only 88 percent complete primary education, and 54.9 percent complete lower basic education, many Congolese loose the opportunity of education. The education system still fails many through repetition, which combined with dropout and low retention, is responsible for this low performance. Even so, the rate of out-of-school children (aged 6-18) in Congo (12 percent) is well below the average for Sub-Saharan Africa (26 percent) and has improved 5 percent during the period of 2005-2011 (see also Figure 1.3). Of the total out-of-school, 5 percent were never in school and 7 percent dropped out. Eight percent of primary school age Congolese children were out-of-school in 2011, and the respective percentages for lower secondary and upper secondary age children were 10 and 26 percent. 41 Figure 1.6: Retention rate by level of education 100 90 92.24 2005 80 70 60 51.11 50 40 30 23.11 20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Primary Lower secondary Upper Post Secondary Secondary Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 82. Quality, as measured by learning outcomes is also still a challenge, even if improvements have taken place in recent years. The latest results from the Program d’Analyse des Systèmes Éducatifs de la CONFEMEM21 (PASEC) (2007) position Congo as a low performer in comparison to other countries such as Cameroon. Congo’s performance at fifth grade was slightly below the average for PASEC countries in French (34 percent) and well below the performance of Cameroon (55 percent), while slightly above the average in mathematics. 83. In conclusion, during the period between 2005 and 2011, with the abolishing of school fees, the Congolese education system attracted most young Congolese, both boys and girls, reducing the number of out-of-school children in the country. More students, including many girls enrolled in post-primary education and with the expansion of private higher education, many Congolese are now enrolled in tertiary education. However, there is room for improving the quality of learning outcomes, and many challenges remain with regards to repetition and to the ability of the system to retain its students. Although more Congolese are completing the various levels of education, the numbers decrease with the education level. 21 Program d’Analyse des Systèmes Éducatifs de la CONFEMEM stands for Program of Analysis of CONFEMEN Education Systems. And Confereration des Ministres de l’Éducation des États et Gouvernments de la Francophonie (CONFEMEN) stands for Confederation of the Ministers of Education from States and Governments of Francophone Countries. 42 III. Spending in Education Sources of Financing 84. Education in the Congo is mostly financed through public budget, although it also benefits from external partners and household contributions. The recurrent budget covers salaries, scholarships and learning materials, while the capital budget covers mostly infrastructure and equipment. No maintenance of capital goods is covered. Household contributions support salaries of bénévoles, food, some furniture, school maintenance, and examination fees. External financing has focused mainly on technical assistance and support to infrastructure development. Over the period of 2008-2012, external financing contributed with an amount of approximately US$70 million, which represented about 5 percent of total financing for the education sector. Among the external partners supporting the sector, the World Bank (WB) was the main financier (33 percent), followed by the World Food Program (WFP) (22.5 percent), and the United Nations Children Fund (UNICEF) (13.2 percent). Smaller financiers include the United Nations Education, Science and Culture Organization (UNESCO), the Agence Française de Developpement (AFD), the United Nations Development Program (UNDP) (which mostly implements projects from other financiers), and the International Partnership for Human Development (IPHD). Although there is no direct public subsidizing of private education provision, the salaries of teachers in private accredited schools are paid by public budget. In 2011, these teachers represented 20 percent of all primary school teachers and 4 percent of all secondary teachers paid by the state. 85. The commitment made to education development by the Congolese authorities has been reflected in increasing budget allocations to the sector. Public spending in education is the highest among the social sectors. But, it is still lower than spending with Affaires Economiques22 (mainly composed by capital expenditure for infrastructure development) and Defence et Securité Publique23. Figure 1.7: Budget allocations for selected sectors, 2008-2010, (percentage of total state budget) 60.0 2008 2009 2010 50.0 40.0 30.0 20.0 10.0 0.0 Economic Affairs Defense and Education Social Protection Health Public Security and Housing Source: MEFPIPP, Loi de Finances. 22 Affaires Économiques stands for Economic Affairs. 23 Defence et Securité Publique stands for Defense and Public Security. 43 Trends in Spending and Intra-Sectoral Allocations 86. In relative terms, the weight of education expenditure in total public expenditure increased between 2008 and 2010 from 8.5 to 11.1 percent. In 2012, due to the response to the catastrophe resulting from the explosion of an ammunitions depot in Brazzaville, the weight of education expenditure in total expenditure dropped to 7.8 percent. In nominal terms, the actual total expenditure with education increased from XAF 107,256 million in 2008 to XAF 197,214 in 2013, an overall increase of 84 percent. If the 2013 budget was fully executed, its weight in the total public spending would climb to 12.9 percent. However, these figures are still below the average for SSA, which was 20 percent in 2011. When measured as a percentage of Gross Domestic Product (GDP), expenditure with education increased from 2 percent in 2008 to 2.8 percent in 2012, and it would jump to 5.1 percent if the 2013 budget was fully executed, thereby surpassing the SSA average of 3.9 percent of the GDP. Table 1.3: Weight of actual education expenditure in the total expenditure and in the GDP, 2008-2013 2013 2008 2009 2010 2011 2012 (Prog.) Total recurrent and capital expenditure in 107 106 135 155 197 360 Education (Billion XAF) As a percentage of total public expenditure (%) 9.8 10.7 11.1 8.8 7.8 12.9 Recurrent 15.5 16.3 17.0 16.8 13.1 16.8 Capital 2.2 5.0 5.0 3.8 4.2 10.7 As a percentage of GDP 2.0 2.3 2.3 2.3 2.8 5.1 Per capita in XAF 28,300 27,348 34,027 38,017 47,214 n.a Per capita in US$ 63.2 57.9 68.7 80.6 92.5 n.a Sources: Authors estimates based on IMF, Republic of Congo: 2012 Article IV Consultation—Staff Reports (Dec 2012 and May 2013) and Loi de Finances pour chaque année (Programmed) and Exécution Budget Sectoriel (Executed). 87. From 2013 onwards, there is a change in intra-sectoral allocations benefiting the METPQE. Until 2012, a large proportion of the education budget was allocated to MEPSA, in line with the focus of achieving the MDGs in 2015. An increasing attention to the goal of developing skills for growth sectors led to a very important increase in the funding allocation to the METPQE. This accompanied the Head of State declaration of 2013 as l’Année de le Formation Qualifiant – the Year of Qualifying Education and Training. As a result, MEPSA’s allocation decreased, while the allocation for the METPQE doubled. The 2014 budget does not change this new intra-sectoral allocation pattern. Allocations to higher education have not changed significantly during the period. Table 1.2: Intra-sectoral allocation, 2008-2013 2013 2008 2009 2010 2011 2012 Prog. MEPSA 61,3 65,3 61,6 56,2 61,5 49,0 METPFQE 16,9 11,5 17,3 20,1 16,7 34,5 MES 21,8 23,2 21,1 23,8 21,8 16,5 TOTAL 100,0 100,0 100,0 100,0 100,0 100,0 Source: MEFPIPP, Budget Execution and other financial data 2008-2012 44 88. The pattern of expenditure across the three education ministries changed favoring an increase in investment expenditure. Until 2012, both the METPQE and MES were allocated an increasing share of recurrent expenditure, whereas MEPSA benefited from a higher share of capital expenditure. In 2013, METPQE’s allocation doubled and did so through an increase in investment expenditure: 83 percent of the 2013 budget allocated to this ministry was on investment24. This was to allow revamping of outdated infrastructure and equipment to meet TVET training needs. The increase of MTPFQE investment budget from 7 billion in 2012 (executed) to 100 billion XAF (programmed) in 2013, corresponded to a 52 percent of the total investment in education. This is a significant change that consolidates the upward trend: in 2008, investment in education represented 10 percent of the total expenditure; in 2012 it reached 31.9 percent. If the budget for 2013 was implemented as planned, then capital expenditure was higher than recurrent expenditure. Figure 1.8: Evolution of budgeted education expenditure (million XAF), by ministry (2008-2013) 250,000 200,000 150,000 100,000 MES METPQE 50,000 MEPSA 0 Current Investment Current Investment Current Investment Current Investment Current Investment Current Investment 2008 2009 2010 2011 2012 2013 Source: MEFPIPP, Budget Execution and other financial data, 2008-2012 and Loi de Finances 2008-2012. 89. Between 2008 and 2012 and in real terms,25 overall public expenditure in education grew at an average rate of 11.4 percent, a figure that was below the growth rate of total public expenditure for the same period – 17.9 percent. This was largely the result of an imbalance between investment and recurrent allocations, strongly favoring investment ones. During the period, education capital expenditure was multiplied by a factor of five, whereas total public capital expenditure was multiplied by a factor of 2.7. Recurrent expenditure in education followed the same pattern as overall public recurrent expenditure. 24 The percentage of investment expenditure budget for the METPFQE for 2014 is slightly lower, but still very high: 77 percent. 25 The estimate of the public education expenditure in real terms was based on the consumer price index (WB source). 45 Figure 1.9: Evolution of public and education expenditure (recurrent and investment), 2008-2012 Public Investment Expenditure Education Investment Expenditure Public Recurrent Expenditure Education Recurrent Expenditure 600.0 500.0 400.0 300.0 200.0 100.0 0.0 2008 2009 2010 2011 2012 Source: Authors estimates from MEFPIPP, Budget Execution and other financial data, 2008-2012. Budget Execution 90. The execution rates of the consolidated education budget improved noticeably over the last years, both for recurrent and investment expenditure, although the latter are systematically lower than the previous. The consolidated execution rate increased from 89.9 percent to 94.6 percent between 2008 and 2012. While it is true that the execution rate for investment expenditure is lower than for recurrent expenditure, it is important to note that it was the previous rate that showed the most remarkable improvement. The investment expenditure execution rate increased from 39 percent in 2008 to 83.4 percent in 2012 26. The introduction of a new Procurement Code in May 2009 (revised in 2011) along with the implementation of other measures from the Public Financing Management Program (PFMP) framework, such as the decentralization of budget execution and the introduction of a computerized budget chain, seem to have bared such good results. 91. In spite of the good progress made there are several important challenges to address to continue to improve budget execution rates in the education sector. The implementation of the new procurement code seems to have not yet been fully appropriated and requires sound dialogue among the various line ministries. All procurement, irrespective of the line ministry, higher than XAF 1 billion falls under the responsibility of the Délégation Génerale de Grands Traveaux27 (DGGT). In 2012, projects managed directly by the DGGT accounted to 73 percent of all projects in Congo. This centralization of large procurements can lead to different investment budget execution rates in the various line ministries. Procurement for high visibility and larger projects may precede other procurements and thus, favor the budget execution rate of one ministry over the others. Other challenges in the 26 It should be noted that year 2012 was atypical due to the passage of a supplemental budget, following the government’s response to the immediate effects of the explosion of the ammunitions depot in Brazzaville. The supplemental budget increased the recurrent expenditure by over 46 percent and doubled the investment expenditure comparing to 2011. In practice, these expenditure levels were kept practically unchanged, in nominal terms in the budget for 2013, thereby setting in a new expenditure base. The impact of these changes in the education sector was basically felt on the investment side, which rose 69 percent between 2011 and 2012. 27 Délégation Génerale de Grands Traveaux stands for General Delegation of Large Works. 46 execution of public investment relate to issues such as (i) lengthy reaction from the Treasury during the payment phase, (ii) a highly centralized decision making mechanism at the minister level, and (iii) different procedures required by external financiers. It is worth noting that feasibility and sustainability studies are not often included in the decision making process for investments, which poses risks to their sustainability. 92. The budget execution rate varies among the three education ministries. MES seems to be facing overruns both in recurrent (5 percent in 2012) and investment expenditure (14 percent in 2011), although more significantly in the latter. This suggests potential programing problems that MES needs to address. The lowest budget execution rates are found in METPQE and MEPSA, and in investment expenditure.28 In 2010, the last year for which there is available data on the composition of investment expenditure, the lower execution rate was associated with the low execution of categories (i) elaboration of studies (only 37 percent executed) and (ii) acquisition of equipment (64 percent), potentially reflecting a limited capacity of the planning departments in the preparation of the tender processes. Table 1.3: Budget execution rates (percentage) by Ministry, 2008-2012 2008 2011 2012 Current Investment Total Current Investment Total Current Investment Total Total 52.7 104.7 67.0 107.2 107.3 107.2 121.6 76.6 90.1 Budget MEPSA 100.3 29.0 83.2 90.7 95.0 91.6 97.1 84.6 92.1 METPQE 124.8 87.7 113.1 104.1 79.1 91.0 107.3 72.7 91.7 MES 107.5 21.4 96.3 111.7 114.3 112.3 106.1 99.4 105.0 Total 104.9 39.0 89.9 97.2 91.6 95.6 100.9 83.4 94.6 Education Source: MEFPIPP Budget Execution and other financial data, 2008-2012. Financial Aid and Scholarships 93. Financial aid represents an important share of the total recurrent expenditure in the education sector, and most of this is used in scholarships. For the purposes of this chapter, three categories of financial aid will be considered: (i) tuition free primary education and Collège implemented through transfers to schools of amounts based on a student per capita basis; (ii) free textbooks (French and Math for primary education and Collège); and (iii) scholarships for TVET and higher education students. The two first categories seek to improve access and quality of education by reducing household contribution to basic education, while gradually integrating les bénévoles in the public payroll29. The scholarships policy dates back to 199730 and aimed at bolstering both TVET and higher education. In 2012, a total of 26,870 million XAF were spent with financial aid in the whole education system (tuition fees, textbooks and scholarships), representing 20 percent of the total current expenditure with education, compared to 31 percent in 2008. Scholarships represented the 28 It was not possible to provide an analysis of the causes explaining the lower execution rates due to the unavailability of recent data on the composition of investment expenditure. 29 There is very little information about this process and it was not possible to access its impact on recurrent expenditure. 30 Decree 77/515 setting out the decision process for awarding a scholarship. 47 lion share in the total of the financial aid, reaching 72.8 percent of the total aid in 201231. Scholarship attribution does not seem to be directly related to a pro-poor policy and there is no clear framework for scholarship attribution in spite of various revisions in the law 32. This issue is discussed in the education strategy, but it basically focuses on higher education, and in the specific context of the possible trade-off between the attribution of scholarships and the reinforcement of social services at the university level. Table 1.4: Public financial aid in education (percentage of total) by Ministry, 2008-2012 2008 2009 2010 2011 2012 Tuition fees 3.1 7.4 19.8 19.3 16.4 Textbooks 29.3 7.5 21.0 9.2 5.0 Scholarships 36.9 75.1 42.2 58.8 72.8 Internal (UNMG) 11.5 24.3 13.9 18.2 20.0 Internal (Technical Education) 3.6 5.9 2.9 12.1 11.1 Internal (Primary and Secondary) 3.1 3.5 1.7 1.7 1.2 Abroad (Tertiary) 17.6 37.3 21.1 18.4 30.5 Abroad (Technical Education) 1.0 4.2 2.6 8.5 10.0 Other (didactic material, etc) 30.7 10.0 17.0 12.8 5.8 TOTAL 100.0 100.0 100.0 100.0 100.0 Source: MEFPIPP Budget Execution and other data, 2008-2012. 94. Most scholarships benefit higher education students but the number of scholarships has been decreasing especially for those studying abroad. In the academic year of 2011-2012, a total of 14,990 students of the University Marien Ngouabi (57 percent of the total) benefited from a scholarship, down from 68 percent in 2007-2008. The relative number of students benefiting from an in-country scholarship has been increasing (from 73 percent in 2008 to 78 percent in 2012) as compared to those benefiting from a scholarship abroad. Furthermore, the relative weights of the scholarships paid in-country and abroad have reversed. In 2008, the value of domestic scholarships represented 40 percent of the total, whereas in 2012 it represented 57 percent. This trend seems to indicate an orientation towards a progressive reallocation of the financial aid to the domestic tertiary system. This is an important shift in financial terms as scholarships for studying abroad cost three times more than scholarships to study in Congo. Overall, scholarships in higher education make up 37 percent of the overall budget for this education level. 31 Due to the paucity of data on these spending categories, the internal composition of this type of expenditure in previous years can be biased. In some years there is a large amount classified as other didactic material, etc. 32 The Law 25/95 sets out the general conditions that should be met to benefit from a scholarship, but secondary legislation is missing to set specific criteria; the Decree 86/722 establishes the conditions for the attribution of scholarships, the different types of scholarships, and other social support measures to the beneficiary students; Decree 2012/68 of February 2012 sets the monthly values of the scholarships, to be paid for studying in Congo and abroad; the Decree 2012/17 of February 2012 lays down the budgetary procedures underlying the payment of scholarships. 48 Table 1.5: Scholarships: value and number of beneficiaries (public sector), 2008-2012 2008 2009 2010 2011 2012 UNMG Number of beneficiary students 7,835 8,490 9,453 9,886 11,717 % of total 73.1 73.5 76.6 78.5 78.2 Total value (million XAF) 2,901 3,277 3,794 4,087 7,785 Unit value per student 37.0 38.6 40.1 41.3 66.4 Abroad Number of beneficiary students 2,881 3,058 2,887 2,706 3,273 % of total 26.9 26.5 23.4 21.5 21.8 Total value (million XAF) 4,282 4,659 3,785 3,582 5,933 Unit value per student 148.6 152.4 131.1 132.4 181.3 Total Number of beneficiary students 10,716 11,548 12,340 12,592 14,990 Total value (million XAF) 7,183 7,936 7,579 7,669 13,718 Unit value per student 67.0 68.7 61.4 60.9 91.5 Number of tertiary students (public) 15,764 16,341 20,383 26,362 n.a % of students benefiting from a scholarship 68.0 70.7 60.5 47.8 n.a Source: MES, Effectif des étudiants boursiers et incidence financière annuelle; MEFPIPP, Budget Execution. Unit Costs 95. Student unit costs vary significantly among education levels. Drivers of unit costs in primary education largely relate to teacher salaries. The slow increase between 2005 and 2011 in primary unit costs can be seen as resulting from a balance between an increase in teacher salaries and an increase in the pupil teacher ratio. Between 2008 and 2010, the accumulated increase on the average salary (about 23.6 percent in nominal terms, but only 3 percent in real terms) was compensated by a higher student/ teacher ratio (almost 2 percentage points higher). Preliminary analysis on the short-term impact of the free tuition policy points out to the continuation of a strong financial contribution from households in primary education. This seems to be the case mostly for schools that still include a large number of bénévoles, which salaries are paid by the parents. 33 Such a financial contribution has also been playing an important role in the low and relatively stable primary education unit cost, which could have significantly jumped with the introduction of the free tuition policy. 33 Oulai, D. and Farba, D. (2001). La gratuité de l’éducation au Congo. Etat des lieux des frais liés à l’éducation, analyse des conséquences de la gratuité (Rapport Provisoire), UNICEF/UNESCO, 8 mai 2011. The analysis further indicated (based on a survey conducted to support the analysis) that in the school year of 2009-2010, still 41 percent of schools mentioned not having received transfers from MEPSA, thus justifying the fees charged to the parents. 49 Figure 1.10: Unit costs per education level, 2005 and 2011 2005 2011 1,600,000.0 1,400,000.0 1,200,000.0 1,000,000.0 800,000.0 600,000.0 400,000.0 200,000.0 - Preschool primary Secondary TVET Higher Total Source: Authors: Estimations calculated from ECOM 2005, ECOM 2011 data, and data from MEFPIPP 96. Unit costs for TVET are high, but unit costs for higher education are disproportionally high in comparison to all others. The main drivers for unit costs in TVET and higher education are teacher salaries and scholarships. Spending on salaries accounted for 36 percent and 56 percent on average, respectively; spending on scholarships accounted for 26 percent and 37 percent on average. In the case of higher education, the low enrolment rate combined to the large amount spent on scholarships justifies the disproportionately high unit cost for this education level. Figure 1.11: Enrolment and budget allocations by education level, 2005 and 2011 Enrollment 2005 Enrollment 2011 Budget 2005 Budget 2011 61% 54% 34.9% 30% 27.5% 26% 25% 26% 16% 16.1% 10.5% 11% 12%10.2% 12% 8% 5% 3% 2% 1% 2% 3% 1% 2% Preschool primary lower sec upper sec TVET Higher Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data, and data from MEFPIPP. 97. In conclusion, overall spending in the sector increased since 2008 even if is still below the average for SSA. The focus on developing skills for growth sectors produced a change in the pattern of intra-sectoral allocations favoring the TVET sub-sector. This change brought with itself a significant increase in overall investment expenditure in the education sector. Budget execution rates improved since 2008, but challenges remain with regards to 50 investment expenditure. Although, primary education is free, many households continue to provide financial contributions to schools, which combined with a balance between increased teacher salaries and increased pupil teacher ratios produced a small increase in unit costs in primary education from 2008 to 2012. The weight of scholarships in higher education associated with the low enrolment in this sub-sector is responsible for a very high unit cost in higher education. IV. Efficiency of the Education System Internal Efficiency of Congo’s Education System 98. Internal efficiency. Congo has made important progress in improving key indicators and thus, towards achieving its education goals; spending in education increased in line with goals and priorities. This sub-section discusses the internal efficiency of the Congolese education system, as briefly defined in Box 1.2, and makes use of results from Sections II and III. Box 1.2 Internal Efficiency The internal efficiency of an education system can be analyzed in various ways. The chosen approach in this section is the most widely used and accepted, which can be defined as the ability of an education system to educate the greatest number of students in the shortest period of time, and with the least use of financial and human resources. The following indicators/aspects are used in the analysis of internal efficiency: (i) repetition rate, (ii) dropout or retention rates, (iii) survival or completion rates by level of education, (iv) quality of education, and (v) recourse utilizations (pupil - teacher ratio, student textbook - ratio, and scholarship administration.) 99. Although Congo increased the funds allocated to education, student retention is very low and barely changed between 2005 and 2011; repetition is a key factor of inefficiency. As discussed in Section II (see Figure 1.6), of 100 Congolese children enrolled in primary school, 50 reached grade 6, 23 reached grade 10, and hardly any reached university in 2011, very similarly to what had been the case in 2005. High repetition is very costly. In 2011, the grade repetition cost in Congo was about 0.6 percent of GDP, close to 21 percent of the annual current education expenditure (at the 2011 prices).34 School dropout has very 34 The of cost of grade repetition is calculated based on: (i) the direct cost of schooling that is generated from total number of repeaters based on per student annual unit cost in public and private, and (ii), discounted value of forgone opportunity costs of expected earnings due to lag and length of labor market engagement. The latter is 51 important implications in public and household expenditure and income. The current dropout rate in Congo (7 percent, two percentage points lower than in 2005) implies an opportunity cost of 3.3 percent of GDP, and 10 percent of household total consumption expenditure.35 While improvements have taken place with regards to reducing the number of dropouts, repetition rates continued persistently high between 2005 and 2011 (as discussed in Section II), reaching more than 20 percent in primary education. On average, and at national level, it takes 7.4 years of schooling for a Congolese child to complete 6 years of primary education, and 4 years to complete 3 years of upper secondary education. Efficiency gains can be made in this area. 100. More Congolese children are completing primary education but quality is still an issue. As the analysis in Section II pointed out, although completion rate in primary education increased to 88 percent, Congo is a low performer among PASEC countries, and there is some indication that quality may decrease with grade in primary education. However, there have been important investments to improve the availability of quality inputs such as textbooks - close to 3 million French and Math textbooks were distributed between 2007 and 2012 - and some initiatives in teacher training with more than 9,000 teachers benefiting from in-service training in the same period. 36 101. Nonetheless, most students seem to be satisfied with their schools, although many complain about overcrowded classrooms. In fact, student satisfaction is high both for primary and lower secondary education, and it has slightly increased between 2005 and 2011. Nonetheless student perception of key factors for quality teaching and learning point out to challenges around teacher distribution and teaching issues – overcrowded classrooms, lack of supplies and teacher absenteeism. The fact that there are no real major changes in students perceptions (some more significant change regarding school conditions, which can result from the recent efforts in school construction) between 2005 and 2011, can indicate possible slow changes in improvement of key efficiency factors such as teacher distribution and teacher quality. estimated based on wage employment earnings by taking into account the age of labor market entry and associated unemployment rate. 35 The earnings of individuals by level of education are estimated to determine the forgone opportunity costs. This is done by reviewing the earning difference between those completing the level of education and those who dropped out before completing such level. 36 Information from Republic of Congo Support to Basic Education Project, World Bank Project Implementation Completion Report, April 2014. 52 Figure 1.12: Student school satisfaction primary and lower secondary, 2005-2011 (percentage) 35 Primary Lower secondary 30 2005 35 25 30 20 2011 25 2005 15 20 2011 10 15 5 10 0 5 Other problems Overcrowded Satisfied Lack of teachers Poor education Property in poor conditions Teachers often absent Lack of supplies 0 Other problems Overcrowded Satisfied Lack of teachers Poor education Property in poor Teachers often absent Lack of supplies conditions Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 102. With regards to share of trained teachers and pupil teacher ratio (PTR), Congo ranks in the middle of SSA countries, comparing favorably with countries such as Cameroon and Togo. PTR and share of trained teachers are two indicators that can be used to measure efficiency in use of resources and quality of service delivery. The PTR is an important indicator in education planning, and a low PTR may give a pupil a better chance of contact with the teacher, hence better (quality) teaching and learning processes. However, a low PTR also increases unit costs, since teacher salaries constitute a large proportion of the total cost of schooling. In 2011, the primary school PTR for Congo was 44 to 1. In the same year, the share of primary trained teachers was 80 percent. Figure 1.13: Percentage of trained teachers and PTR in primary schools in SSA countries, circa 2011 120 100 80.3 80 60 40 44.4 20 0 Togo Chad DRC Cameroon Niger country Senegal CAR Guinea Burundi Namibia Benin Malawi Sao T&P Mali Congo South Africa Rwanda Côte d'Ivoire Angola Swaziland Mozambique Mauritius Mauritania Ghana Sierra Leone Ethiopia Cabo Verde Eritrea Uganda Kenya Comoros Liberia Botswana Gambia Nigeria Tanzania Lesotho Burkina Faso Madagascar Equatorial Guinea Guinea-Bissau Seychelles Sudan Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 53 103. PTR in secondary education increased mostly at the lower level, indicating a possible lack of teachers to support the expansion of this level. In secondary education, the PTR raised from 24:1 in 2008 to 30:1 in 2012, as a result of the steep increase of enrolment which seems to have not been accompanied by a correspondent increase in teacher recruitment. This average ratio hides a significant difference between the Collège and the Lycée. In fact, the ratio for the Collège (35:1) contrasts markedly with that for the Lycée (only 8:1). The very low ratio in the public Lycée deserves particular attention as it suggests that there is room to substantial savings if the ratio is lifted to a more reasonable level. Overall, Congo has low secondary education PTR compared with several SSA countries. Figure 1.14: .Secondary education PTR, regional comparison, circa 2011 70 68 60 50 42 37 38 40 40 33 33 33 35 30 30 30 26 26 26 26 27 27 28 30 21 22 23 25 25 25 25 18 19 19 19 19 20 20 15 15 16 17 12 14 10 0 Equatorial… Cameroon Botswana Togo Mali Chad DRC Kenya Niger Burundi Guinea CAR Congo Rwanda Namibia South Africa Senegal Malawi Mauritius Swaziland Sao T&P Mauritania Mozambique Cabo Verde Somalia Angola Ghana Uganda Eritrea Ethiopia Tanzania Lesotho Burkina Faso Madagascar Nigeria Seychelles Guinea-Bissau Sudan Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 104. Further, the ratio teacher - administrative staff is very high in Congo, which indicates an additional inefficiency in the distribution of human resources in the education system. The excessive number of administrative staff (1.5 teachers in relation to non-teaching staff in primary education; 1.7 in the Collège and 3.3 in the Lycée), suggests that the education system is being used as an employment buffer, implying a significant financial burden that could be reallocated to other needed inputs. As the budget information provided by the Ministry of Finance does not distinguish the remuneration of the teaching and non- teaching staff it is not possible to reckon the associated financial implications. The teaching function is also often performed in accumulation with other functions. Indeed, MEPSA's education statistics reveal the functional accumulation of the school staff, namely administrative staff that also teach (21 percent of the teaching staff in primary and 6.7 percent in the Collège), as well as school directors that also carry out teaching activities (10.8 percent of total teachers in the primary level). The financial implications of these overlapping functions are also not clear, given the level of aggregation of the information from the Ministry of Finance. In fact, there may be additional sources of inefficiency for which there is no available information. The latest primary education teaching census dates from 2008 and at the time it allowed for the identification of 5,148 ‘ghost’ teachers, whose salaries were frozen, and 1,672 ‘ghost’ personnel which salaries were also suspended. Further, 2,253 staff from other ministries was identified under MEPSA’s budget. However, no follow up assessments 54 were made, as no clear information has been collected on the existing bénévoles in the system. There is also no additional detailed information on staff paid under the METPFQE and MES. This is required as the both staff costs for both ministries are very high and are one of key drivers for the unit costs for TVET and higher education. 105. In conclusion, there are important areas where efficiency gains can be made. These need to focus on an overall improvement of quality of service delivery that increases the number of quality teachers, decreases the ratio between administrative staff and teaching staff, that make a better use of teaching time, separating the teaching function from others, and that is based on improved planning of distribution of human resources. Further, quality of teaching needs to be addressed and gains are required in retention and repetition rates, as well as in drop out ones. Data collection and analysis on inefficiency factors are also required in order to better identify other sources of inefficiency and act upon them. External Efficiency of Congo’s Education System 106. External efficiency. While it is fundamental to identify areas where internal efficiency gains can occur, the efficiency of an education system cannot disregard its impact in the social and private benefits and gains it generates. This sub-section provides an analysis on the external efficiency of the Congolese education system. Box 1.3 External Efficiency The external efficiency of the system provides a perspective of the social and private benefits generated by education, as well as other intermediate benefits of education that can lead to better social and economic rates of return. In order to provide some insights on the external efficiency of the Congolese education system, a Mincer regression model was used to estimate earning increases associated with additional years of levels of schools. A logistic regression was used to estimate the role of education in choices of job based on security and return differential. Further, given that educational attainment is a critical determinant of poverty in developing countries, an estimation of poverty incidence by level of education for the working age was also carried out. 107. The educational attainment of working age37 Congolese is overall low and between 2005 and 2011, although the percentage of those with a higher education degree increased, the percentage of those with complete primary education decreased . This 37 Working age is defined as aged 15 to 64. 55 decrease from 34.8 percent to 29.2 percent can be a reflection of the impact of the armed conflicts in education. The percentage of Congolese with complete post-primary education is low indicating that for some time, the system has not been favoring progress between cycles. Further, men are more likely to attain higher levels of education than women: 4.6 percent of men concluded secondary education against 3.7 percent of women, and 7.9 percent of men concluded higher education against 5.6 percent of women. The percentage of women without instruction is more than double that of men (11.3 percent against 4.9 percent). This difference is wider in rural areas, where one woman out of five has no instruction. 38 Figure 1.15: Educational attainment of the labor force, 2005 and 2011 2005 2011 40.0 30.0 20.0 10.0 0.0 No education Incomplete Completed Lower upper TVET Higher primary primary secondary secondary education Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 108. However, Congo compares favorably in relation to many SSA countries in what regards the level of literacy of its working age population. In 2011, only about 15 percent of the working age population in Congo had no formal education, a much lower figure than the average for SSA countries (32 percent) and higher than Gabon (8 percent), but lower than Cameroon (22 percent). In the same year, the national average for youth and adult literacy rates were 91 percent and 88 percent, respectively. These figures are higher than the average for SSA (73 percent and 62.58 percent, respectively) and those from neighboring countries such as Cameroon (80 percent and 75.58 percent, respectively). Such high literacy rates point out to a system that although with some current challenges of quality, has been providing some basic skills to most of its population but not supporting progression in the education system. 38 Data from the Demographic and Health Survey of 2011-2012 was used for this analysis. The level of educational attainment is measured in this survey through the percentage of the population that has reached a certain level of education. 56 Figure 1.16: Youth and adult literacy rates, 2011 100 Youth and Aduly Literacy Rates 91 80 Youth Adult 60 40 88 20 0 Chad Nigeria Guinea Sudan Burkina Faso Gambia, The Niger Zambia Malawi Cote d'Ivoire Ghana Comoros Lesotho Ethiopia Rwanda Sierra Leone Uganda Kenya Sao T&P South Sudan Mozambique Mauritania South Africa Mali SSA Tanzania Senegal Madagascar Cameroon Congo, Rep. Source: Authors: Estimations calculated from ECOM 2011 and DHS 2011-2012 data. 109. Between 2005 and 2011, unemployment dropped for all levels of education, but particularly so for upper secondary and higher education. Given the strong participation of the informal sector in the Congolese economy, unemployment figures should be read with caution. Such a low overall unemployment rate of 8 percent (or even of 13 percent as was the case for 2005) hide the fact that more than six out 10 working age Congolese earn a living in the informal sector. The formal sector is largely circumscribed to the public sector (that employs one of out of three Congolese in Brazzaville), and the formal private sector is very limited. This can explain the low unemployment rates for working age Congolese with very limited educational attainment – most of these do not have wage jobs and earn their living in the informal sector. As for the highly educated Congolese, jobs have been found in the public sector. It is the middle level educated Congolese (secondary education) that find it more difficult to find wage jobs. However, some progress has been made as highlighted by the decrease from 25 percent to 12 percent of the unemployment rate of Congolese with a secondary education diploma. Figure 1.17: Unemployment rate by level of education, 2005-2011 30% 2005 2011 25% 25% 23% 23% 20% 20% 17% 14% 13% 15% 12% 12% 11% 9% 9% 10% 8% 5% 4% 3% 5% 0% No Incomplete Completed Lower Upper TVET Higher Total education primary primary secondary secondary Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 57 110. Education rates of return in Congo are high, decreased from 2005 to 2011, but education was the main factor justifying increases in earnings during this period; further, the higher the education level, the higher the probability of being wage employed. During the period, education was responsible for an average earning change of 91 percent, while labor market factors only contributed to a 35 percent change. The highest returns resulted from a TVET and higher education diploma, and this pattern is common to men and women although rates are slightly lower for women; the role of education in increasing earnings was more significant in 2011 than in 2005. Incomplete primary and complete primary education produce much lower rates of return when compared to any other education level. Holding of an upper secondary education diploma or higher education one is more important for women than men. The employment profile varies accordingly with the education level, and thus those with higher education are mostly employed in wage jobs and earning more than the others. In 2011, 80 percent of those with a higher education diploma were in wage employment against 55 percent, 56 percent, 36 percent, 21 percent, 14 percent, 10 percent, for those with a TVET diploma, secondary, lower secondary, primary education, incomplete primary, and no education, respectively. Figure 1.18: Rates of return by level of education, national and female, 2005-2011 2005 2011 200 184 175 176 180 164 163 160 146 132 140 125 120 100 97 106 100 89 76 80 66 70 58 60 44 40 37 40 28 9 3 15 20 3 0 Higher Higher TVET TVET Lower secondaruy Lower secondaruy upperc secondary upperc secondary complted primary complted primary Incomplte primary Incomplte primary Ntional Female Source: Estimate based on ECOM 2005 and 2011. 111. Further, education is a major factor in poverty reduction in Congo, which poverty rate is 46.5 percent and varies inversely with educational attainment. At the national level, about 47 percent of the total population and 42.1 percent of the working age population lives below the absolute poverty line (less than US$1.25 a day). The poverty incidence falls extremely with the level of educational attainment. Estimates based on data from the ECOM 2011 for the working age population (age 15-64), reveal that about 61 percent of the workforce with no education lives below the extreme poverty line compared to 58 50 percent with complete primary education, and rates improve much further for higher education levels. Completion of lower secondary education is fundamental for improving living conditions. As Figure 1.19 indicates, there is a substantial decrease in the percentage of working age Congolese living below the poverty line with complete lower secondary education when compared to those with no or complete primary education. Figure 1.19: Percentage of the working age population living below the absolute poverty line by level of education, 2011 70 61 59 60 50 50 42 38 40 30 25 22 20 17 10 0 No Incomplete Completed lower Upper TVET Higher Total education primary primary secondary secondary education Source: Estimate based on ECOM 2011. 112. In conclusion, the Congolese education system has been able to provide some basic literacy skills to the country’s population. However, the majority of the Congolese still do not reach post-basic education, which completion provides them with a better opportunity of increasing their earnings, and thus improving their living conditions. Education is not only the main factor for increasing individual earnings, although it contributes to that than labor market associated factors, but it is also crucial in reducing the very high poverty rate of the country. As many Congolese do not reach post-basic education they are sidelined from wage employment and earn a living in the informal sector. This produces an overall low unemployment rate for Congo, which should be considered with caution. 59 V. Equity in Education Affordability of Education and Equity in Access 113. Education is a determinant factor for poverty reduction in Congo; however it is not affordable to all. While enrolment in primary education decreases with income (as there are many more children from poor quintiles than from richer ones), the opposite is the case for enrolment in post-basic education levels. This suggests that children from poor households benefit much less from post-basic education. While it is true that improvement has taken place from 2005 to 2011, the proportion of children from poor households in secondary and higher education is still extremely low. Further, during the period, there have been high increases in the number of children from wealthiest households in secondary, and particularly higher education; this is a concerning trend that suggests an important inequity challenge. Given the high unit cost of higher education in Congo (as discussed in Section III), spending in post- basic education seems to be favoring the wealthier households. Figure 1.20: Enrolment per education level and quintile, 2005-2011 Q1 Q2 Q3 Q4 Q5 60.0 50.0 40.0 30.0 20.0 10.0 - Higher Higher Primary TVET Primary TVET preschool upper secondary preschool upper secondary lower seconsdary lower seconsdary 2005 2011 Source: Estimate based on ECOM 2005 and ECOM 2011 114. In fact, post-basic education is unaffordable for the poor Congolese. All levels of education are very expensive for poor Congolese households. In relative terms to per capita income, poorest households spend 41.5 percent with each primary student compared to 5.2 percent spent by the richest households. Spending increases to prohibitive figures at the higher education level for poorest households, reaching 952 percent of per capita income, and making higher education completely unaffordable to the poorest households. While poor households spend a high share of non-food expenditure in primary education, the opposite is true for non-poor households, who spend most with post-primary education. This is an indication that poor households have no capacity to invest in post-basic education. For 60 example, the richest households spend more than 57 percent of total education spending on secondary and tertiary education, compared with only 33 percent for the poorest households. Given that the richest households have better total non-food consumption expenditure, the share allocated to education is high, which implies that well-to-do families invest more on education in anticipation of future returns. 115. Issues of affordability of education are also the main reason for out-of-school children in Congo, and the situation worsened from 2005 to 2011 especially for post- basic education. Econometric analysis using data from 2011 indicate that the household social economic status is the main determinant of school attendance for children ages 6 to18, followed by the level of education of the household head. Children from the richest quintile have a probability of attending school of 117 percent; for children from the poorest quintile this probability is only 63 percent (middle quintile and rich quintile probabilities are 88 percent and 101 percent, respectively). Further, holding all other characteristics constant, primary age children from the richest households in 2005 had 73 percent more chance of attending school than their counterparts, and this figure only slightly decreased to 71 percent in 2011. However, this was not the case for secondary school age children, where the role played by wealth increased the probability of school attendance from 60 percent in 2005 to 79 percent in 2011. While the level of education of the household head is also important, the main difference resides on whether the household head is or not educated – the probability of a child from a household where the head has incomplete primary school is 40 percent compared to a probability of 79 percent for a head with complete primary education, and 88 percent for secondary education or more. A rural child has half the probability of an urban child to attend school (21 percent and 40 percent, respectively) however the impact of the area of residence on attendance nearly disappears when all other factors are accounted for. The same is the case for gender. Figure 1.21: Reasons for being out-of-school Source: Estimate based on ECOM 2005 and ECOM 2011 61 116. Approximately 8 percent of Congo’s population is included in the autochthonous population group, which faces important challenges regarding access to education . This population is distributed by several regions of the country and has a nomadic way of life. 39 It is estimated that 65 percent of autochthonous adolescents (ages between 12 to 15 years old) are out-of-school. However, access to school for this population does not depend solely on the supply of education services in their communities, but it is conditioned by traditional economic and social activities. A legal framework40 has been established to support measures to increase access to education for autochthonous children and some progress has been made in recent years. There is very limited information about the autochthonous population to allow for a detailed analysis. Table 1.6: Primary GER and NER of autochthone population, circa 201041 Autochthonous Indicator Population Gross enrolment rate (Primary) Men 76.9% Women 59.0% Total 67.9% Net enrolment rate - Primary (6 – 11 years) Men 47.8% Women 40.2% Total 44.0% Source: CNSEE, Census of 2007. Equity in Attainment 117. Income is the main factor of inequity in attainment; area of residence and education of household head also play an important role; gaps have widened from 2005 to 2011. Girls from the poorest uneducated families are the most severely deprived from education. Poor, rural children have the lowest primary enrollment rates and very high dropout rates. Primary enrolment of children from urban households of the top quintile, whose head of household has a secondary or higher education diploma, is more than 95 percent, with less than five percent dropout rate each year. In contrast, enrolment of children from rural households of the lowest quintile, whose head has no education, is 51 percent and 41 percent for boys and girls, respectively. By the end of six years of primary school, enrollment among the urban top quintile children from high education household hovers around 75 percent for boys and girls both in 2005 and in 2011, compared with that for rural lowest quintile from poorly educated households, which was below 25 percent. The disparities grow exponentially with education levels attained, and the gap widen between 39 Autochtonous population is distributed along nine departments (Likouala, Sangha, Cuvette Ouest, Plateaux, Lékoumou, Niari, Pool, Bouenza et Kouilou). 40 Stratégie Nationale d'Éducation des Populations Autochtones du Congo under implementation since 2009. 41 There is very limited data regarding autochthonous population. 62 2005 and 2011. Of 100 children from the lowest quintile only 1 completed 13 years of education compared to 60 from the top quintile (boys and girls). Figure 1.22: Factors affecting educational attainment 120 2005 Poorest rural girls --head no education 2005 Richest urban girls --head 100 secondary + 95 2005 Poorest rural boys--head no education 2005 Richest urban boys--head 80 secondary + School attendance 77 Poorest rural girls --head no education 75 71 64 Richest urban girls --head secondary + 60 61 53 40 26 20 1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Primary Lower secondary Upper Post Secondary Secondary Source: Estimate based on ECOM 2005 and ECOM 2011. 118. Disparities in educational attainment are very high in Congo compared to other SSA countries. For example, the gap between the poorest and richest quintiles in upper secondary enrollment is 78 percent in Congo compared with 59 percent which is the average for SSA. Congo is positioned thirty second out of 37 countries for which data is available, performing very poorly in terms of equity – it is amongst the 6 top worst performers. Further, it compares poorly as well with SSA countries of a similar income, for which the average is 68 percent. However it is a better performer than Gabon and Cameroon, which is last of the group. Figure 1.23: Regional Comparison of gap between poorest and richest quintile in terms of net primary enrollment rate (percentage) 100 86 94 78 78 78 78 80 80 71 72 74 66 67 67 68 58 59 60 62 62 52 55 60 46 46 48 48 49 51 39 39 33 34 35 35 40 28 28 32 20 20 14 0 Source: Same as Figure 1.3 63 119. Disparities of income and area of residence are also found in learning outcomes . In both PASEC exercises discussed in Section II, national results hid statistically significant differences. The most important differences relate to geographic and income variables. Gender variation was marginal. Both boys and girls did significantly better in urban areas than in rural ones. Children in private schools scored better (11-12 points higher) than those in public schools. Despite the higher average performance of private schools, when examining overall performance relative to school budgets, none of the schools that have above-average performances across grades and subjects while using below-average financial resources was private. In part, this may be because private schools are more expensive. However, the schools that are uniformly below-average in performance and above average in costs are mostly in urban areas (95 percent of students in such schools). Interestingly, the majority of children in under-performing schools were in public schools in Pointe Noire (39 percent), followed by public school children in Brazzaville (17 percent). Equity in the Human Capital Stock and Income Holding 120. Income, area of residence and gender play a role in the average years of education of the working age population in Congo, and income and rural urban disparities are very significant. On average, the Congolese working age population has not completed primary. At the national level, the number of years of schooling of the working age population is 8.1; however, this figure is very different in rural and urban areas: 5.0 and 9.4, respectively. Better equity is found with regards to gender, as the average number of years for women is 7.3 and for men 8.9. The most important differences are once again found at level of income. While the average number of school years for working age Congolese in the top quintile is 10.2, the equivalent for those in the lowest quintile is 5.3. Different regions of the country also present different results, with Brazzaville presenting 9.9 years of schooling (highest) and Pool only 4.7 years (lowest). Figure 1.24: Average years of education of working age population by gender, income, region and proximity to school 8.1 of res. Gender 8.9 female 7.3 Area urban 9.4 5.0 10.2 quintile income Q4 8.9 7.9 Q2 6.8 5.3 9.9 Pointe-Noire 9.2 6.4 Cuvette 6.1 Region 6.1 Bouenza 6.1 5.6 Plateaux 5.2 5.1 Sangha 5.1 5.0 Pool 4.7 Distance primary school 0-29 minutes 8.5 to 8.0 60 plus mintus 4.5 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Source: Estimate based on ECOM 2005 and ECOM 2011. 64 121. Given education disparities among the young Congolese and among those of working age, these seem to be replicated an intergenerational manner, which has extremely negative consequences – inequality in educational outcomes in Congo closely correlates with income holding; many Congolese are stuck in a poverty cycle. At the national level, only 17 percent of the total wealth is hold by the lowest quintiles, which represents 40 percent of the Congolese population. A regional comparison of 27 SSA countries for which data is available (see Annex A.1) positions Congo among the 10 high income high inequality countries. Further, analysis of 2005 and 2011 household surveys data shows that the income holding of the bottom 40 percent slightly decreased. Figure 1.25: Income holding per quintile, 2005 and 2011 120 Q1 Q2 Q3 Q4 Q5 100 80 44 45 60 22 23 40 16 16 20 10.9 10.8 0 6.1 6.1 2005 2011 Source: Estimate based on ECOM 2005 and ECOM 2011. Spending in Education and Equity 122. Population of primary school age and enrolment at primary level seemingly decrease with the wealth quintile, from the poorest to the richest quintile. This means that, in comparison to the richest households, more children from poor households are enrolled and benefit from public spending at primary level.42 In post primary levels, although the population figures show very minor differences except for the richest quintile, where there are fewer primary school age children, enrollment figures clearly show that the poorest household children benefit less from post-primary education. It should also be noted that, as stated above, primary education is free in Congo since 2007, which might have contributed to the increase in the share of primary enrollment from the poorest households from 22.8 percent to 26.8 percent between 2005 and 2011. Post primary education poses an important financial burden on households, and enrollments are thus mostly comprised of children from high income quintiles. For instance, in 2011, only 3.8 percent of upper secondary students, and 0.3 percent of higher education students came from the poorest households, while the 42 There are several factors why poor households are associated with large numbers of children: (i) more educated people have lower fertility rates and less prevalence of poverty, (ii) children contributes to household production as they grow, and poverty level of the household diminishes with increased working household members, (iii) in rural areas where poverty rates are high, people have less access to health services and less practice of family planning methods. 65 corresponding figures from the richest quintile were 27.5 percent and 56.8 percent, respectively (Figure 1.26). Compared to 2005, the share of children from the poorest households enrolled in post-primary education declined. This implies that government spending on post-primary education benefits primarily more affluent households. Figure 1.26: Distribution of enrolled students and school age population by quintile Q1 Q2 Q3 Q4 Q5 60 56.8 50 43.5 40 37.6 34.5 32.0 30 26.8 27.5 22.8 23.1 19.9 20 16.1 14.6 15.8 13.3 11.4 10 6.2 6.4 3.8 1.1 0.3 0 Preschool Primary Lower sec upper sec Higher Preschool Primary Lower sec upper sec Higher 2005 2011 Source: Estimate based on ECOM 2005 and ECOM 2011. 123. Public spending in primary education has been pro-poor, while spending in post- basic education has mostly favored the well-off. In 2011, the poorest quintile received 21.1 percent of the public benefits allocated to primary education (just slightly above the population share of the quintile), while the richest quintile received 16 percent (4 percent of less than the population share of the quintile). For post-basic education, however, the situation is different with 22 percent of the budget allocated to higher education and per student spending 12 times higher than for any other level. Compared to 2005, the benefit to poorest households decreased, even in primary education, and increased from the richest households (Figure 1.27). 124. The benefits of public expenditure on both primary and lower secondary education were relatively more biased towards the poor than was the distribution of income. In other words, when looking at the Lorenz curves (figure 1.28), expenditure on these levels of education was relatively more equitable than income, as the benefit incidence for public spending on each were both above the consumption concentration curve. Public spending on primary education was progressive, while expenditure on lower secondary education was still progressive, though to a lesser degree than primary, but was not-pro-poor, as was the case for post-basic education. 66 Figure 1.27: Benefit incidence analysis of public expenditure on education, 2011 70 Q1 Q2 Q3 Q4 Q5 62 60 50 46 40 37 31 30 25 26 25 24 23.9 23 22 21.2 19 21 19 17.3 20 17.8 18 20 16 16 12 12.3 78 7.2 10 3.6 3.6 4.2 0.5 0 Preschool Primary Lower sec upper sec Higher Preschool Primary Lower sec upper sec Higher 2005 2011 Source: Estimate based on ECOM 2011. Figure 1.28: Lorenz Curve for Household consumption expenditure and public spending on education by level 100% Perfect equality Consumption preschool 90% primary Lower secondary Upper secondary TVET 80% Higher Education 70% 60% 50% 40% 30% 20% 10% 0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Source: Estimate based on ECOM 2005 and ECOM 2011. 125. Overall spending in education favors the richest households; however, the allocation of the education resources by education levels has clearly different patterns. About 32 percent of the total education benefits accrue for the poorest quintile, comparing to 10 percent to the richest. Whereas in primary and secondary education benefits are progressive, in TVET and higher education these are regressive (Figure 1.29). 67 Figure 1.29: Education benefits by level Source: Estimate based on ECOM 2005 and ECOM 2011 126. In conclusion, issues of inequity are very important in education in Congo. The education system has been pro-poor in primary education and favored the mostly the well-off in post-basic education, thus contributing to the intergenerational cycle of poverty among the Congolese. Rural and urban disparities are important in the educational attainment of both young and working age Congolese; regional differences are also significant, but disparities of income are the most important ones. Education in Congo is a fundamental factor for poverty reduction but spending on education has not supported this role. VI. Conclusions and Recommendations Congo has been making progress towards the achievement of main education goals but important challenges remain 127. The Congolese education system has been guided by clearly defined goals and progress has been made towards their achievement. As a result of the armed conflicts of the 1990s, the Congolese education system suffered a negative impact which jeopardized the country’s ability to meet the MDGs by 2015. As a response, the Congolese authorities focused policy and interventions during the first decade of the century in rebuilding the physical and human resources network so as to ensure universal primary education for all by 2015, one of the main strategic goals of the system. This resulted in important investments in civil works, teacher training, and in the implementation of free primary education policy from 2007 onwards, which entailed the end of fees in primary schools, along with the distribution of free textbooks and learning materials. As a result, the country was able to bring into school most of Congolese school age children, both boys and girls, and although completion rates are 68 not yet ideal, Congo is well positioned to meet the education MDGs close to 2015. This progress has produced some improvement in access to all other levels of education for both boys and girls, making gender parity a possibility beyond primary education. 128. The good progress made towards access and completion of primary education has not been so significant in other education levels. While access to post-primary education increased, this has been much lower than desirable. High repetition rates, combined with drop out and low retention particularly at lower levels of the system, along with school fees and no financial support to poor households, which constitute close to half of the Congolese population, are blocking the access of young Congolese to post-primary education. The strategic goal of improving retention in primary and secondary education while improving the flow of students through the cycles is still far from being achieved even if positive trends have been established. Meeting such goal is fundamental in Congo given the central role of education for poverty reduction in the country. Such blockage in the system plays a contribution to the intergenerational cycle of poverty in which many Congolese are trapped. 129. The Congolese economy is highly reliant on oil; its diversification, among other things, requires a workforce that is capable of responding to the skills demands of growth sectors. The Congolese authorities see the development of technical and vocational education and training as a means to prepare such workforce, and in order to do so have the revamping of the TVET sub-sector as a goal in key strategic documents. The Head of State declared 2013 l’Année de la formation qualifiant. The strategic goal of developing technical and vocational education in line with market demands and economic diversification is a new goal, and little can still be said with regards to progress to its achievement other than the high level of attention being paid to such a goal including at the budget level. It is worth noting though that achievement of this goal will require a concerted effort from stakeholders other than the TVET sub-sector in order to prevent a mismatch between skills demand and supply, and to ensure TVET provision in a sustainable manner, given that changes in supply of skills take longer than changes in the economy of skills demands, and that unit costs for TVET are very high. 130. Although more Congolese are completing higher education, and among them many women, higher education is very expensive both for households and government . As is often the case in most countries, private provision of tertiary education in Congo increased under very little quality control and coordination. This allowed for an increase in the number of Congolese holding a tertiary education diploma, but not necessarily to an improvement in the quality and numbers of the highly skilled workforce required by the country. Further, costs of higher education are prohibitive for the poor and the public sector spend large amounts in scholarships for the well-off. The strategic goal of developing quality higher education in line with market demands and priority sectors growth has seen some progress especially at the policy level with the preparation and approval of new legislation aiming a regulating the sector, implementing mechanisms of quality control, and developing partnerships with the private sector; but there is still a long way for sound progress to be visible. 69 Increasing attention has been given to the education sector in the national budget however education expenditure is unevenly supporting the achievement of strategic education goals 131. Budget allocations to the education sector have increased and so has been the case for the budget execution rates, even if these varied between ministries. Overall, among the social sectors, education has been the one with the largest budget allocation. In relative terms, the weight of education expenditure in the total public expenditure increased in the last five years, and particularly so in the budgets of 2013 and 2014, after the declaration of 2013 as l’Année de la formation qualifiant. If the 2013 budget was fully executed, the weight of education in total spending would jump from 8.5 percent in 2008 to 12.9 percent in 2013; nonetheless, this figure is well below the average for SSA of 20 percent. Budget execution rates have also improved in the period, although with better improvements for recurrent than investment expenditure. The consolidated budget execution rate for the education sector increased from 89.9 percent to 94.6 percent between 2008 and 2012. Budget execution rates varied among ministries with MES presenting overruns in both recurrent and investment expenditure, and low execution rates for investment expenditure for MEPSA and the METPQE. 132. The progress made in access to primary education can been seen as a result of the efforts made in civil works to rebuild the school network, and in the introduction of the free primary education policy; budget allocations supported these efforts. Between 2008 and 2012, MEPSA, the main responsible for the implementation of policies supporting the goal of universal primary education for all by 2015, had on average budget allocations that were three times higher than those for MES and METPQE. Most of these (see Table A.1 Annex A.1) were allocated to primary education. Recurrent expenditure supported largely teacher salaries and school fees and textbooks and learning materials; investment expenditure supported civil works for primary education. The focus on primary education within MEPSA’s intra-sub-sectoral allocation was aligned with the international focus on meeting the MDGs by 2015, but may have prevented the implementation of policies and interventions that could have supported better results in meeting the goal of improving retention in primary and secondary education while improving the flow of students through the cycles. 133. Revamping the physical infrastructure for TVET resulted in a major increase in the investment expenditure for the sub-sector. This is in line with the aim of developing technical and vocational education in line with market demands and economic diversification. While such an increase (corresponding to more than doubling the allocation to the METPQE, of which 80 percent is allocated to investment expenditure – see Figure A.1 in Annex A.1) can be seen as a positive sign in the sense that financing is being made available to carry out interventions to meet the goal; it can also pose some stress in budget execution, especially if sound planning for the TVET sector has not been fully developed. Further, this increase in METPQE investment expenditure increases the consolidated investment expenditure for the sector (making it higher than recurrent), which may have an impact in the execution rate of the consolidated budget. 70 134. MES has presented overruns both for recurrent and investment expenditure and budget allocations to higher education have not changed significantly in time . In fact, a large amount of MES expenditure is for scholarships and although there has been a shift in the trend favoring in-country scholarships, expenditure is very high. Other than financing the development of highly skilled professionals through the attribution of scholarships, expenditure in higher education does not seem to have been fundamentally supporting interventions aimed at developing quality higher education in line with market demands and priority sectors growth. There is space to re-prioritize and improve the allocative and operational efficiency of the Congolese education system 135. While all strategic education goals are fundamental for sector development, it will be important to better align budget allocations with such goals, based on strategic planning and sound understanding of the role of public budget for key development areas. The gains obtained in primary education need to be sustained. The development of TVET is important but revamping of TVET infrastructure and equipment is not sufficient to ensure the development of the required skills. There is room for a revision of the investment allocation for the METPQE. A trade-off can be considered with secondary education, for which very limited funding is now being allocated, and mostly to finance recurrent expenditure. Expenditure with scholarships in higher education is extremely high. Given the lack of regulation and control around the attribution of such scholarships, it is not even clear whether these are providing the Congolese labor market with the most appropriate high level skills. Given the weight of scholarships in MES overall expenditure and the recurrent overruns, there is need to reconsider the distribution of funds between expenditure categories within MES budget. The scholarships category is a source of inefficiency. Both TVET and higher education have high unit costs; further economic diversification requires a dynamic skills development system that cannot exclusively rely on the public sector. Thus, there is need for the public sector, through MES and the METPQE, to create the conditions for the establishment of sustainable partnerships with the private sector that provide both technical and financial support, and decisively contribute to the achievement of the strategic sector goals. 136. Repetition and drop out are systemic sources of inefficiency in the Congolese education system. Although practically all school aged Congolese children enroll in primary school, close to half of them complete this level, and less than a fourth complete lower secondary education. Very few reach higher education. This pattern of retention has high costs for the system, for families, and for the Congolese labor market. There is a need to further understand the underlying causes of such high repetition levels, and to create the conditions to retain students in higher levels of education than primary. This will require a combination of policies and interventions that target improvement of quality of human resources and make at least the lower levels of post-basic education pro-poor. 137. Although there is insufficient data to allow for a detailed understanding of the distribution of human resources, the high ratio administrative staff – teaching staff, and 71 evidence that many parents still pay fees to cover salaries of bénévoles in primary education indicate potential systemic sources of inefficiency around human resources in the sector. Further, the last teacher census indicated a large number of ‘ghost’ personnel in the sector. There is need to further understand the profile and distribution of human resources in the Congo beyond the simple knowledge of pupil - teacher ratios, so as to allow for the definition of clear policies that can tackle the existing inefficiencies. Further, the quality of human resources is a key factor in quality of education, which is still poor in the Congo. Public spending in Congo is pro-poor in primary education, regressive in post-basic education, and overall favors the well-off, thus it does not contribute to improve equity in access and education attainment 138. Education is a major variable to improve equity in Congo, with a much larger impact than any labor market associated factor, however, the current system is favoring the well-off and leaving many Congolese caught in the poverty trap. Free primary education has supported access to education to poor Congolese, however, most of them either do not complete this level or even if they do so, do not move up in education ladder to secondary and upper levels, which are those that would allow them a significant increase in income. Post-primary education is unaffordable and higher education is prohibitively expensive for the poor Congolese, which constitute half of the country’s population. Although gender, distance to school and geography play a role in equity of access and attainment, it is income that prevent poor Congolese to obtain an education. At the same time, public spending is regressive in post-primary education reinforcing the inequities – for example, the high level of spending in scholarships for high education targets the most well-off, which are those who can reach this level. There is thus, an urgent need to create the conditions to increase equity in education. Table 1.7 summarizes some concrete recommendations to address the identified issues. 72 Table 1.7: Summary of Issues and Recommendations ISSUE ACTION OUTCOME Education expenditure is Improve intra-sectoral allocations and allocations between categories unevenly supporting the  Trade-off between allocations to TVET and Secondary Education to Increased progress in strategic achievement of strategic Policy/Intervention increase the allocation to the latter education goals education goals  Revision of investment expenditure for TVET to ensure alignment with sound medium and long term planning and actual absorptive capacity  Revision of recurrent and investment expenditure for higher education  Establishment of the regulatory framework necessary for the participation of the private sector in education, especially at the TVET and higher education level  Promote the establishment of public-private partnerships in TVET and higher education both at the technical and financial level There are systemic sources of Improve retention rates inefficiency in the Congolese  Analysis of causes of early grade repetition and definition of policies and education system: Policy/Intervention measures to decrease it considering the possibility of regulating by law Larger number of Congolese repetition rates complete primary education (i) Low retention, Improve the quality and use of human resources and secondary education of high repetition Carry out a teacher and administrative staff census for primary and secondary better quality. and drop out Policy/Intervention education and from its results: (ii) Inefficient  Remove ‘ghost’ personnel from the system More funds are made available quality and use  Define a sustainable method to incorporate still existing bénévoles from improved use of of human  Define and implement a teacher training program to improve the teaching resources. resources skills and knowledge of all teachers (including bénévoles) and review pre- (iii) Higher service training High level skills in the labor education market are better aligned with  Establish a planning and monitoring system for teacher distribution scholarships economy needs. Improve the efficiency of the scholarship program in higher education Policy/Intervention Revise the scholarship regulatory framework, allocation criteria, and numbers of scholarships attributed; and improve the alignment between skills needs and scholarships granted. Public spending in Congo is Make post-primary education affordable pro-poor in primary  Design a package of incentives to reduce the costs secondary education More Congolese improve their education, regressive in post- Policy/Intervention (lower and upper) to the poor, including the attribution of scholarships for earnings due to better basic education, and overall fees, for example. education. favors the well-off, thus it does  Consider a policy change of making basic education (primary and lower, not contribute to improve perhaps of 9 nine years) free and compulsory for all looking at financial equity in access and education impact and sustainability. attainment (these measures increase student retention and thus contribute to improved efficiency both internal and external) 73 Health Public Expenditure Review I. Introduction 139. In Congo, the improvement of the health conditions of the population is considered fundamental in achieving poverty-reduction and economic growth outcomes and this can be clearly seen in its Poverty Reduction Strategy Paper (PRSP) 2012-16. The PRSP 2012-16 and the Programme National de Développement Sanitaire (PNDS)43 – currently being updated - define key areas of intervention in the sector. Priority is given to the achievement of the Millennium Development Goals (MDG) for health: child/infant mortality, maternal health, fighting HIV, malaria and other infectious diseases. Important attention is also given to the general strengthening of the health system in terms of governance. 140. Raising public expenditure in health and ensuring an efficient use of resources is an important means for improved health outcomes, thereby contributing to poverty reduction and increased equity. Balanced intra-sectoral budget allocations can play an important role in ensuring that priority sector policies are implemented. Gains in efficiency allow for improved health outputs and outcomes. Although a middle-income country, in 2011 Congo’s poverty rate was 46.5 percent. Equity in access to public health services is crucial to guaranty that such a large section of the population is not simply excluded. 141. This chapter examines recent achievements in health outcomes, the trends and composition of public expenditure in the health sector for the period of 2008-2012, and provides insights on the sector efficiency and equity, in order to respond to the following research questions: a. Do budget allocations (level and composition) contribute to achievement of the set strategic health goals? b. Is there space for re-prioritizing and improving allocative and operational efficiency in the health sector? c. Does public spending contribute to improve equity in access to quality health services? 142. The analysis uses a combination of methods: (i) key sector performance indicators were calculated to assess progress towards sector priorities and MDGs; (ii) budget data was analyzed so as to provide insights on sector allocations, composition of the budget and rate of execution; (iii) quality of health services and use of resources were analyzed so as to provide information on the sector’s efficiency; and, finally, (iv) an analysis of disparities in access to health services was used to provide insights on equity across different quintiles of income and geographically 43 Programme National de Développement Sanitaire stands for National Program for Sanitary Development. 74 disadvantaged populations. A benefit incidence analysis and catastrophic payments to health care analysis were also used. 143. Several data sources were used for the analysis in the chapter. These included administrative data from the Ministère de la Santé et la Population (MSP); administrative data from the Ministère de l’Économie, des Finances, du Plan, de l’Integration et du Portefeuille Publique (MEFPIPP) 44; and survey data from the 2005 and 2011 Enquetes Congolaise Auprès de Ménages (ECOM) – household surveys - and Congo’s Demographic and Health Surveys (DHS), known as EDSC 2005 and EDSC 2011-2012. 144. In several instances there were data discrepancies between administrative and survey data. In such cases, a preference was given to survey data since it is generally more reliable, and in most of the present cases reflects the most recent information. The main source of discrepancy opposes the data reported by the World Bank’s World DataBank (WBWDB) for Congo and the estimates obtained through the EDSC 2011-12 survey. The survey data was selected as main reference since it is more recent, and in settings where administrative data collection is challenging (which is the case), survey data tends to be considered more reliable. For consistency, comparisons of health outcomes with other countries were also done using DHS data. 145. The chapter is organized in seven main sections. Section II introduces the objectives and organization of the health system; Sections III and IV provide an analysis of sector progress and sector spending; Section V focuses on issues of efficiency and equity in the delivery of health services; and Section VI provides conclusions and recommendations for the next five years. II. Institutional Environment Objectives of the Health System 146. Congo has adopted a National Health Policy in 2003. The general objectives of this policy were (i) promoting the health of Congolese citizens throughout the country; (ii) guarantying access to quality healthcare services by the population; and (iii) reinforcing the national capacity of management of the health system. 147. Two Programme (PNDS) have since been published to translate into actionable measures these general goals. With the support of development partners, the Government set up and implemented the most recent PNDS between 2007-2011, in an aligned manner with the aim of achieving the MDGs, and within the context of the vision laid out for the country and sector by the Head of State known as the Nouvelle 44 Ministère de la Santé et la Population stands for Ministry of Health and Population; Ministère de l’Économie, des Finances, du Plan, de l’Intégration et du Portefeuille Publique stands for Ministry of Economy, Finances, Plan, Integration and Public Portfolio. 75 Esperance (New Hope). This Plan, which is being updated, aims to improve the performance of the health system in order to reduce the burden of morbidity and mortality, and promote health by strengthening care and services at the district-level, general hospitals, specialized support services, as well as strengthen institutional capacity and partnership coordination. 148. Specific priorities for the health sector have also been identified in the most recent Poverty Reduction Strategy Paper (2012-2016) and will likely be the basis for the updated version of the PNDS. These include: a. Improving the governance and direction of the sector by developing the institutional and legal framework for health development, planning, and programming, as well as reform of health sector financing; b. Improving access to healthcare services by improving health coverage (building and equipping new health facilities, rehabilitating and equipping existing health training programs, etc.); c. Reducing inequities in access to health services by ensuring access to healthcare for the poor and improving vulnerable pregnant women’s access to prenatal care (PNC), intermittent preventive treatment (IPT), and sulfadoxine-pyrimethamine (SP)45; d. Strengthening the supply of healthcare by promoting health services, social markets, and communication and developing health action appropriation mechanisms; e. Improving the quality of services by developing leadership, to include management teams, the provision of medically assisted procreation (MAP) and appropriate patient controlled analgesia (PCA), responding to the population’s needs, particularly for water, electricity, and a system for disposing of biomedical waste; f. Managing medications by coordinating supplies, strengthening the Congo Essential Generic Drugs Agency (COMEG), streamlining the prescription of essential generic drugs (EGDs), and using appropriate guides in healthcare institutions, developing a quality assurance system, and managing auxiliary services, particularly the National Public Health Laboratory (LNSP) and the National Blood Transfusion Center (CNTS); g. Combatting communicable and non-communicable diseases, with particular emphasis on maternal and child health; h. Managing emergencies, disasters, and responses to epidemics by strengthening the health emergency management system and providing the departments with emergency kits. 45 Drug used in treatment and prophylaxis of malaria. 76 Organization of the Government Health System 149. The government health system in Congo is formally decentralized and comprises three hierarchical levels. They are (i) central, (ii) departmental46, and (iii) peripheral and operational (Figure 2.1). The country is subdivided into 12 départements (regional departments) and these, in turn, are subdivided into communes and/ or districts.  The central level is strategic and normative and is responsible for the planning, monitoring, evaluation, coordination, mobilization, and resource allocation of health system. It comprises the Minister's Office, the general and central directorates, public autonomous health entities, and specialized public health programs.  The intermediate level is in the hands of the departmental health directorates (DDS). They provide technical support to Socio-Sanitary Circumscriptions (CSS) (defined below) in the dissemination of information, the specific adaptation of national standards to local conditions, their implementation, and the supervision of staff.  The peripheral and operational level comprises operational units responsible for program planning and implementation. It is composed by CSSs, each of which is subdivided into health areas. A CSS consists of a network of ambulatory health providers, both public and private. The CSS is administered by a management team responsible for planning of health activities and local resource management. In rural areas a CSS typically has a catchment area of 50,000 to 100,000 inhabitants, while in urban areas it covers between 100,000 and 300,000 people. Each CSS, in turn, is subdivided into health areas, each of which operates an Integrated Health Center (CSI) with a catchment area of 2,500 to 10,000 people in rural settings, and 10,000-15,000 people in urban areas. 150. The government health care delivery system comprises a national network of health facilities distributed throughout the country and organized under a pyramidal referral system. At the bottom of the public delivery system are health dispensaries. According to the recent Health Statistics Yearbook issued by the MSP, in 2012, there were 345 dispensaries distributed across all départments, except the two largest urban centers in the country, i.e. Brazzaville and Pointe Noire. The next referral level is composed of CSI, of which there are two types, depending on the services provided. The Centre de Santé Intégré à Paquet Minimum d’Activités Elargi (CSI à PMAE)47 offer a broad array of preventive and curative ambulatory services, and in 2012, there were 84 of such facilities around the country; whereas the Centre de Santé Intégré à Paquet Minimum d'Activités Standard (CSI à PMAS)48 offer a narrower set of such services, and in the same year, there were 222 of them. The first referral facility for CSIs is the Hôpital de Base (Base Hospital). In 2012, all départments except Kouilu had one or more 46 The territory of Congo is divided into regional areas called departments, thus, the word departmental is used in this case, in relation of each of the 12 Congo départments. 47 In English, Integrated Health Centers with a Minimum of Services Offered. 48 In English, Integrated Health Centers with Additional Services Offered. 77 Base Hospitals. These inpatient facilities, in turn, have as their referral, in the public system, the Hôpital Général (General Hospital), of which there are six in the country, including the Centre Hospitalier Universitaire (University Hospital Center) located in the capital city of Brazzaville. Figure 2.1: Structure of Congo’s public health system Minister of Health’s Office, general and central directorates, public Central level autonomous health entities, and specialized public health programs Departmental Health Directorate National National Departmental level DDS (DDS) Hospital Hospital 12 DDS 6 National Hospitals Socio-Sanitary Circumscriptions CSS (CSS) Base Hospital 27 Base Hospitals Integrated Peripheral and Health Center operational level (CSI) 306 CSI 36 CSS CSI Health Dispensaries 345 health posts Health Dispensaries Source: Authors from information from the MSP. 151. Private health care providers play a major role in Congo’s health system, but there is little coordination between public and private providers and little regulation of the latter . The private health sector was formally institutionalized 25 years ago, through the Decree N° 88/430 of June 6, 1988, laying down the conditions for private practice of medicine and paramedical and pharmaceutical professions. Official documents (the National Health Policy of 2003, the National Health Development Plan 2007-2011, and the Health Sector Development Program 2010) confirm the Government’s commitment to collaborate with the private sector to strengthen the health system, improve the population’s health status, and protect the citizens’ fundamental right to a healthy life. According to a 2012 World Bank (WB)/ International Finance Corporation (IFC) report that analyzed Congo’s private health care delivery system, in 2005, there were 1,712 health care providers in the country, of which more than half (1,002) were private. The vast majority of private providers (88 percent) were for profit. The MSP’s Health Statistics Yearbook listed 617 private for profit providers throughout the country in 2012 (Ministry of Health and Population, 2013). The private health care delivery sector is largely unregulated and poorly organized. Officially registering as a new health care provider involves a 78 convoluted procedure, and as a consequence a large share of private providers has only temporary licenses to function (Results for Development Institute and Health Research for Action, 2011). This issue with registration is the most likely explanation for the discrepancy between the number of private facilities in the official 2012 numbers and in the WB/IFC report. 152. The pharmaceutical products procurement agency, Congo Essential Generic Drugs Agency (COMEG), has the responsibility of procuring and supplying government hospitals and health centers with generic drugs and consumables, but its management capability and performance are limited. The MSP requires that government health providers buy drugs from COMEG using their own budgetary resources or revenue from user fees. However, many public providers, including Base Hospitals and CSIs, choose not to comply, and prefer instead to bypass the institutional procurement system in order to purchase drugs and supplies directly from private providers, at lower prices. 153. The lack of regulation and control of the private drugs procurement system poses patient safety problems. The private drugs supply system comprises five wholesalers and a large number of retailers, including pharmacies and shops, which sell their products to public and private health care providers. Whereas wholesalers and retailers must obtain government authorization to function, they are not subject to any quality controls by the public health authorities. 154. In conclusion, the Congolese health sector is guided by a set of clearly defined objectives, aligned with the MDGs and focusing on improved service delivery. The sector is decentralized at three main levels, and uses a referral system. The private sector plays an important part in delivery of care however, there are important challenges associated to it: (i) there is little regulation and a heavy process of registration; and (ii) little quality control mechanisms concerning procurement of drugs. This is also an issue with regards to public health providers, as these many times bypass COMEG, and procure directly from private providers at lower prices. III. Health Outcomes and Health Risks Infant Mortality Rate (IMR) and Child Mortality Rate (CMR) 155. Up until recently Congo was lagging behind other countries in the region in terms of health indicators. Prior to the release of the most recent Demographic and Health Survey, EDCS 2011-12, Congo seemed to be outperformed by many countries in the region with respect to IMR and CMR. Accordingly, the 2010 government’s MDGs report stated that the Congo had made little or no progress toward the MDGs goals of reduced infant/ child mortality (Government of Congo, 2010). 79 156. In contrast with these trends, the EDCS 2011-2012 report presented an encouraging picture. By comparing the estimates of the IMR and CMR between the previous EDCS (2005) and the most recent one (2011-2012), the report concluded that both these had decreased (Figure 2.2), and that their improvement was statistically significant and accelerating. It estimated that in 2009, the IMR was 39 deaths per 1,000 live births, the CMR was 68, and the death rate for children ages 1-5 was 30.49 Figure 2.2: Infant and child mortality rates, 1997-2009 (deaths per 1,000 live births) Infant Mortality Rate Child Mortality Rate Congo: Taux de mortalité infantile et infanto-juvénile, 1996-2009 140 131 Mortality rate (per 1000 live births) 117 120 118 Child mortality rate (0-5 years) 100 99 80 60 54 68 45 40 48 30 44 Child mortality rate (1-5 years) 20 - EDSC-I (2005) EDSC-II (2011-2012) Source: CNSEE (2005) and CNSEE and ICF International (2012). 157. Congo performs well in comparison to neighboring countries such as Cameroon and Gabon with regards to child health outcome indicators. The Congo is the second richest country among those selected, after Gabon. When looking at the EDSC-reported IMR of 39.0, Congo stands out as the best performer in the group, even slightly outperforming the much richer Gabon. The same result emerges when using the neonatal mortality rate (24.0). For better comparison, all the mortality data comes from DHS surveys implemented in the respective countries in close years. 49 As discussed in the introduction, preference is given to survey against administrative data. However, it is important to note that there is a large discrepancy between the data reported by the WBWDB for Congo and the estimates obtained through the EDSC 2011-12 survey. The DataBank reports much flatter IMR and CMR rates over time and this tendency is felt in most health indicators with the EDSC 2011-12 showing better results than the WBWDB. 80 Table 2.1: Infant, child mortality, and neo-natal mortality rates, 2011 in a group of SSA countries with per capita income above PPP in US$1,400 Neonatal Per capita GDP Infant mortality Child mortality mortality rate 2011 (2005 PPP rate (deaths per rate (deaths per (deaths per US dollars) 1,000 live births) 1,000 live births) 100,000 births) Benin (2011-12) 1,430 42 70 23 Cote d'Ivoire (2011- 12) 1,580 68 108 38 Cameroon (2011) 2,053 62 122 31 Senegal (2010-11) 1,834 47 72 29 Congo (2011-12) 3,850 39 68 24 Gabon (2012) 13,998 43 65 26 Source: WBWDB and ICF International (2012). 158. Several factors and government initiatives seem to have contributed for the positive trends observed in IMR and CMR. For instance, according to the EDSC 2011-12, child vaccination coverage has improved overall for the Bacillus Calmette–Guérin (BCG) (from 90.0 to 93.9 percent) and diphtheria, pertussis (whooping cough), and tetanus (DPT) (from 64.8 to 71.9 percent), and measles (from 66.2 to 74.9 percent). The coverage of the yellow fever vaccine also increased in an important way, from 31.8 percent to 54.5 percent, with a marked reduction in income inequalities. An exception has been the polio vaccine where there was an overall drop in the coverage (see Annex B). 159. There was also a 50 percent increase in access to treatment of diarrhea with oral rehydration salts (ORS) for under-five years old children. Whereas the increase was by far higher among children in the richest quintile, all quintiles experienced an improvement in this indicator. In contrast, the accessibility to antimalarial medicine worsened for the poorest quintile, but improved for the other income groups and had a reduced improvement overall. (Figure 2.3) 81 Congo: Treatment of diarrhea and malaria, 2005 and 2011-2012 (percent) Figure 2.3: Treatment of diarrhea and malaria, 2005 and 2011-2012 (percentage) 60 50 40 36.6 Percent 30 24.8 25.0 22.1 20 10 0 Total Quintile 1 Quintile 2Quintile 3Quintile 4Quintile 5 Total Quintile 1 Quintile 2Quintile 3Quintile 4Quintile 5 (poorest) (richest) (poorest) (richest) Children under 5 with diarrhea who received oral Children under 5 with a fever who received antimalarials rehydration salts EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. Maternal Mortality Rate (MMR) 160. The MMR has also fallen in Congo. The EDSC 2005 reported an MMR of 781 deaths per 100,000 live births for the six years preceding the survey, whereas the EDCS 2011-2012 reported a rate of 429 for the six prior years. Congo’s MMR compares favorably with that of countries with similar per capita income, including Cameroon, Mauritania, Nigeria and Swaziland. Figure 2.4: Maternal mortality rate - mean and confidence interval, 2005 and 2011-2012 Congo: (deaths surveysrate Maternal mortality and 1,000 -meanper live births) confidence interval, 2005 and 2011-2012 surveys (deaths per 1,000 live borths) 1,200 1,000 800 MMR 600 400 200 0 0-6 years preceding EDSC 2005 0-6 years preceding EDSC 2011-2012 Upper bound Lower bound Mean Source: Authors calculations from from CNSEE (2005) and CNSEE and ICF International (2012). 82 161. This improvement can Figure 2.5: Percentage of assisted deliveries by be related to the positive Congo: Percentage of assisted deliveries by qualified health qualified health personnel, personnel, 2005 and 2005 and 2011-2012 2011-2012 (Percent) performance of several indicators associated with 100 93.60 95.00 93.70 97.60 97.60 98.30 98.10 99.50 maternal health including 90 86.10 80.80 Congo’s high rate of 80 79.60 69.80 institutional deliveries by 70 qualified health personnel that 60 Percent has improved from 2005 to 2011- 50 40 2012, although still showing 30 important inequalities across 20 quintiles (see Figure 2.5). 10 0 Total Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (poorest) (richest) 162. A policy implemented by EDCS 2005 EDCS 2011-2012 the government, whereby public Source: CNSEE (2005) and CNSEE and ICF International (2012). providers offer free cesarean- sections, seems to have resulted in a sharp increase in the cesarean rate. The intended effect of this policy is to reduce maternal deaths and to improve child health and survival, and is likely one of the factors that led to significant improvements in MMR and IMR. 163. Knowledge about Figure 2.6: C-sections as a share of all deliveries, 2005 and Congo: Cesarian deliveries as a share of all deliveries, 2005 and contraception improved in an 2011-2012 (percentage) 2011-2012 (Percent) important way between the 14 13.20 two surveys, inequalities in knowledge narrowed 12 significantly, and use of 10 contraception went up. Percent 8 Knowledge increased 7.00 everywhere and inequality in 6 5.80 5.90 6.1 knowledge among population 4.3 4 groups dropped. Use of modern 3.2 3.30 2.8 2.6 contraception by women 2 1.6 1.90 improved in an important way 0 between the two surveys, Total Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (poorest) (richest) although the gain was almost EDCS 2005 EDCS 2011-2012 negligible among women in the poorest quintile (Figure B.5). Source: CNSEE (2005) and CNSEE and ICF International (2012). 83 164. The intergenesic period (the time lapse between birth of one child and conception of the following child in mothers with at least two children) has also improved, albeit only slightly – by about 6 months overall, with higher increases among women in the higher quintiles. 165. Use of prenatal care is very high overall according to these. For all pregnant women, it increased from 88.2 percent in 2005 to 92.6 percent in 2011-2012. The increase took place in all quintiles, although as expected, women from poorer quintiles present lower usage rates. 166. Not only has coverage and utilization of services improved between the 6-7 years that elapsed between these two surveys, but also there seems to have been a reduction in inequality. As Figure 2.7 shows, in 2005 there were large differences among quintiles in the distribution of iron supplements to women and in the provision of the tetanus toxoid vaccine. The delivery of these two services increased in an important way, and the increase was higher among poorer women, such that by 2011-2012 the quintile distributions became almost flat. Figure 2.7: Services provided during pre-natal visits, 2005 and 2011-2012 Congo: Services provided during prenatal visits, 2005 and 2011-2012 (percentage) (percent) 100 90 84.1 80 70 60.6 60 Percent 52.8 50 45.6 40 30 20 10 0 Total Quintile Quintile Quintile Quintile Quintile Total Quintile Quintile Quintile Quintile Quintile 1 2 3 4 5 1 2 3 4 5 (poorest) (richest) (poorest) (richest) Iron pills or sirup Two or more injections tetanus toxoid vaccine EDCS 2005 EDCS 2011-2012 Source: Authors calculations from from CNSEE (2005) and CNSEE and ICF International (2012). 167. Indeed, when looking at Congo’s performance in terms of reproductive health indicators, one could expect an even lower level of MMR. This is a common feature in African countries and reflects the low quality of services offered. 84 Fertility and Malnutrition 168. Fertility has increased: the average number of children born per woman was 5.1 in EDCS 2011-12, an increase from 4.8 in 2005. Fertility rates are higher in households living in rural areas and households from the poorest income quintiles. Moreover women continue to show a preference for a high number of children – according to Congolese women, the ideal number of children was 5 in 2011-12, slightly lower than 5.1, which was the figure for in 2005; further, the ideal number of children is still above 4 for all the income quintiles. Figure 2.8: Ideal number of children according to women, 2005 and 2011-2012 (number) Congo: Ideal number of children according to women, 2005 and 2011-2012 (number) 7 6 5.90 5.70 5.70 5.70 5.30 5.10 5.00 5 4.80 4.70 4.60 4.50 4.60 4 Number 3 2 1 0 Total Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (poorest) (richest) EDCS 2005 EDCS 2011-2012 Source: Authors calculations from from CNSEE (2005) and CNSEE and ICF International (2012). 169. The country also still has a quarter (24 percent) of its children under five years of age that are chronically malnourished (stunted), with two thirds (67 percent) of children in the same age group suffering from anemia. Even though the prevalence of child malnutrition dropped overall in Congo during the seven year period that elapsed between the two household surveys, this reduction was very small. The 2005 survey reported a chronic malnutrition rate of 26 percent whereas the 2011-2012 survey reported a slightly lower rate of 24.4 percent. Overall, severe malnutrition fell from 10.8 percent to 8.0 percent. Whereas both moderate and severe chronic malnutrition fell for nearly all wealth quintiles, inequality in this health indicator worsened. Moreover, moderate malnutrition increased for the three poorest quintiles, and therefore the slight improvement in the overall indicator is the result of a sharp drop in malnutrition in quintiles 4 and 5. 85 Figure 2.9: Chronic child malnutrition, 2005 and 2011-2012 (percentage) 35 30 26.0 24.4 25 Percent 20 15 10.8 8.0 10 5 0 Total Total Q2 Q3 Q4 Q2 Q3 Q4 Q1 (poorest) Q5 (richest) Q1 (poorest) Q5 (richest) Moderate Severe EDCS 2005 EDCS 2011-2012 Source: Authors calculations from from CNSEE (2005) and CNSEE and ICF International (2012). Life Expectancy at Birth (LEB) 170. The LEB for Congo was reported as 57.8 years in 2011 (World Bank DataBank). However, this number may be under- estimated since the recently-released information about the country’s drop Figure 2.10: LEB in 2011 in the IMR and CMR probably has not been reflected in these numbers. Still, LEB in Congo has steadily been going up for the past decade presenting a value slightly above the average for Sub-Saharan Africa ( 171. Figure 2.10) and is in line with the values observed for African countries with similar income per capita and HIV prevalence. However, when compared to middle- income countries worldwide, it still has a lot to catch-up. Source: WBWDB. 86 HIV/AIDS 172. In Congo, the prevalence of HIV among adults ages 15-49 stands at 3.3 percent. It is below the average of 4.9 percent for the SSA region, and significantly below the arithmetic average of 8.4 percent for the 10 richest countries in the region. This prevalence rate has been dropping steadily from 3.9 percent in the year 2000. The progress made in the area of HIV prevention with the implementation of the second generation strategic framework (2009-2012) should have contributed to this reduction. Some of the measures implemented in this context were:  Purchase and start-up of the first complete Mobile HIV Screening Unit in December 2009, allowing a large number of people to learn their serological status.  Prevention activities also carried out among men who have sex with men.  Continued preventive work among sex workers.  Increased activity of the Yellow Line (HIV/AIDS information line) and appreciable increase in the number of calls.  Organization of AIDS fairs during school holidays in Brazzaville, Pointe-Noire, and Ouesso.  Continued use of the cultural approach, involving traditional mediators in the fight against AIDS.  Increasingly larger numbers of people agree to be screened, and people living with HIV being treated in health institutions are more numerous in the country’s twelve regional departments. Malaria 173. According to available hospital statistics, malaria is the first cause for seeking health services, hospitalization and death, accounting for 23,723 cases of hospitalization in 2012 (MSP, 2013). The EDCS 2011-2012 survey tested children for hemoglobin levels as an effort to identify levels of prevalence of Malaria – a level below 8.0 g/dl is considered a good indicator of malaria. At national level, 4 percent of children of 6-59 months show a level below the threshold. There is a particular high prevalence in the département of Cuvette-West (12 percent). The survey also showed that the proportion of children with level of hemoglobin below the threshold decreases from the poorer to the richer households quintiles from a maximum of 6 percent to a minimum of 2 percent. 174. Malaria is considered stable in Congo meaning that transmission is permanent and continuing. Such situation is fed by favorable socio-economic and ecological/climatic conditions. Some specific factors that contribute to a high prevalence of malaria in the country are: 87  Resistance of the plasmodium to the most usual antimalarial medicines (chloroquine: 80 to 90 percent, sulfadoxine-pyriméthanine: 15 to 30 percent) – the malaria treatment protocols changed in 2006 to respond to this.  The frequent scarcity of medicines in the health institutions.  The disrespect of the therapeutic schemes defined in the national policy against malaria.  The weak use of mosquito nets and other materials treated with insecticide. 175. In conclusion, the latest EDSC 2011-12 has revealed a positive tendency of improvement of key health status indicators in the country. Up until the release of that report, key health indicators, such as the infant, child and maternal mortality rates were much higher than expected in the SSA context for a country with such a high per capita income. Further, it was reported that these indicators were stagnant or deteriorating and that the prospects that Congo would reach the health MDGs were remote. In contrast, the latest estimates indicate that Congo has achieved considerable gains in health status. The positive trends shown by the most recent data are definitely a good sign and cannot be neglected. A necessary next step will be to continue collecting data that can further confirm this trend and help fine-tune the real magnitude of the gains. It is also important to state that endemic diseases such as malaria still pose significant challenges, and issues of child malnutrition deserve particular attention. Further, improvements have in some indicators have been less significant for the poor quintiles. IV. Public Resource Mobilization and Sources of Finance Government Budget for Health 176. Within the social sectors, the MSP has been gaining importance. In 2009 the MSP budget represented about 75 percent of the overall education budget. By 2012, the MSP and overall Education budgets were similar. Figure 2.11: Government budget within and outside of social sectors, 2008-2013 (millions of real XAF of Dec. 2012) Source: Authors calculations using budgtet data from MEFPIPP. 88 177. The MSP budget has remained nearly constant, in real per capita terms, between 2008 and 2011. The MSP’s real per capita budget was nearly the same in the years 2008, 2010, and 2011, in the range 28,000-29,000 XAF (US$60-65). Yet, as percentage of the total government budget, MSP budget nearly doubled between 2011 and 2012 and in 2013 it fell, but to a level that was considerably higher than in the years prior to 2012. In 2011 the MSP budget represented 3.8 percent of total government budget, a slight increase from 3.4 percent in 2010. This percentage increased to 5.3percent in 2012 and decreased Figure 2.12: Nominal (US dollars) and real per again to 4.2 percent in 2013. capita (XAF of Dec. 2012) public health budget, 178. In 2012 the MSP budget 2009-2013 experienced a sharp increase to reach about 45,000 XAF per citizen, or US$87. This one-year increase reflected the government’s decision to declare 2012 as the country’s “Health Year.” Budget information for the year 2013 confirmed that this increase was a one-off phenomenon, although the 2013 per capita real MSP budget, of approximately XAF 35,840 (US$70), was much higher than in preceding Source: MSP budget information supplied by the MEFPIPP. years. The trends in nominal and real MSP per capita budgets can be seen graphically in Figure 2.12. 179. When seen in the regional context, in Congo both total health expenditure as a share of GDP and government health expenditure as a share of total government spending are very low, in relation to the country’s per capita income. Whereas Congo’s per capita income in US dollars is among the highest in SSA, the share of the economy that is devoted to the health sector is among the lowest, and so is the share of government spending going to health. Congo’s pattern is similar to that of neighboring SSA countries, such as Gabon. Despite recent increases, Congo’s proportion of government expenditures allocated to health falls short of the Abuja Declaration commitment of increasing government funding for health to 15 percent of government’s total expenditure. 89 Figure 2.13: Government health spending as a share of GDP and of government spending, 2011 (percentage) 100 50 Highest 90 45 Gabon 80 40 ROC 70 35 Mauritania Cameroon 60 30 Senegal Quintile Rank Cote d'Ivoire 50 25 Benin 40 20 30 15 20 10 10 5 0 0 Lowest Per capita GDP Health expenditures as a Government health share of GDP (%) spending as a share of total government spending (%) Source: Authors calculations from World Bank DataBank. 180. It is important to keep in mind, however, the idea that health system performance is influenced by many variables other than health spending. Figure 2.1, below, was originally presented in the World Development Report (WDR) 1993 (World Bank, 1993) to illustrate this point, using the IMR as the target health outcome. The analysis and figure have been updated here using information for the year 2011 and include all World Health Organization (WHO) members with available data. The figure ranks a country’s performance in terms of its deviation from both predicted IMR and predicted total health spending as a percentage of GDP (THS%GDP). Predictions are based on two separate linear regressions one for the IMR and another for THS%GDP, where the independent variables are per capita GDP (in PPP international dollars) and the literacy rate. Countries are therefore grouped into four categories. The best performers are those with an IMR better than predicted and total health spending below prediction. These are countries that have achieved a relatively low IMR with a relatively small amount of resources dedicated to health. The Congo is not in this category, it spends on health less than expected, given its income level and literacy rate and it has an IMR that is higher than expected given those two explanatory variables. The Congo is a poor performer. In Annex B equivalent figures are shown for the CMR, MMR, and LEB.50 50 It is worth noting that although Congo is among the poor performers, it is not very clear how poor such performance is given the combination of data used to prepare Figure 12 and equivalent figures for CMR, MMR, and LEB. What seems clear is that Congo does not spend much in health and there are positive trends in key health outcome indicators. Further analysis will be required to better understand this. 90 Figure 2.14: IMR and health expenditure: deviations from estimates based on per capita Infant income mortality (PPP-adjusted rate and dollars) health expenditure and schooling (literacy rate in in selected countries: deviations from estimates based on per capita percentage), 2011 income (PPP-adjusted US$) and schooling (literacy rate in percent), 2011 70 Worse outcome, Worse outcome, lower expenditure higher expenditure 50 DRC Deviation from predicted infant mortality rate Lesotho 30 ROC-WBWDB Gabon 10 ROC - DHS -10 Eritrea Benin Senegal -30 Morocco -50 Better outcome, Better outcome, lower expenditure higher expenditure -70 -9 -7 -5 -3 -1 1 3 5 7 9 Deviation from predicted percentage of GDP spent on health Source: Authors calculations Government’s Budgeting, Spending and Resource Allocation in the Health Sector 181. Public funding for the health sector is channeled through either delegated credits or transfers, both of which are directed to institutions and health care providers.51  Delegated credits call for the execution of budget items whose nature and amount are preset by government. This allocation mechanism seeks to ensure that the public resources allocated to the sector will be spent according to government’s intent. The authorizing officer should, in principle, allow the disbursement of funds from a specific budget item only if such a disbursement is consistent with the definition of the budget line. However, such a mechanism can only be effective when the recipients have been involved or consulted during the budget preparation. Otherwise, there is a high chance that both the nature and/ or the amount of the credit lines will not match the needs of recipient institutions.  Transfers involve the allocation of a budget envelope, or block grant, to recipients who are given the freedom to allocate these resources according to their needs and priorities. Such autonomy, when exercised in the spirit of a results-based management approach, can 51 These paragraphs draw on Ministère de la Santé et de la Population (2013). Comptes nationaux de la santé (CNS) 2009-2010 Version Finale. 91 be effective for the health system. Transfers are allocated on a lump sum basis and carry no conditions with respect to how they should be spent. 181. In Congo, just under two-thirds of all government financing for health is in the form of delegated credits and over one-third is channeled to health care providers in the form of transfers of block grants. Yet according to the NHA report (Ministère de la Santé et de la Population, 2013) transfers do not seem to have had a sizable impact in terms of improvements in health indicators. 182. The health budget, formulated mostly at the central level, between the MSP and the MEFPIPP, is structured into four main budget items. As common in developing countries throughout the world, the government budget for health is broken into four categories: salaries, goods and services, transfers, and investments. Salaries are centrally allocated to central and regional directorates. Goods and services are allocated to central directorates as well, but not to the regional level. Thus, resources for procurement of goods and services must come at these levels, mostly from user fees. Goods and services are also envisioned in the budget for central health institutions, including some health care providers and, in particular, the COMEG. Transfers are directed primarily to autonomous public hospitals. Figure 2.15: Structure of the Government’s budgeting and expenditures system in the health sector Ministry of Finance Ministry of Health and Population Goods and Salaries Transfers Investments services Minister’s office Maternity Tietie Hospital Directorate of studies and planning Hospital Center Loandjili Hospital National Inspection of Public Health Hospital Brazzaville University Central Hospital Center Directorate of cooperation Directorates Medical-Social Center Adolphe Sice Hospital General Directorate of Public Health Inspection Directorate of Primary, Secondary, and Nkayi Hospital Tertiary Education COMEG (Congolese 31 Juillet d‘Owando Essential and Generic General Hospital Regional Directorate (RD) of Brazzaville Medicines Company) Dolisie General RD of Kouiliou Hospital Regional RD of Niari Other transfers Directorates RD of Lékoumou RD of Bouenza Departmental Directorate (DD) of Pool Investment project 1 DD of Plateaux Investment project 2 DD of Cuvette Departmental DD of Shanga Directorates DD of Likouala Investment project N DD of Cuvette-Ouest Source: Authors from information from MSP. 92 Health Budget Execution52 183. In Congo the degree of actual execution of the MSP budget has been variable but with an upward trend. This is shown in Figure 2.1, which has been constructed on the basis of the information presented in Table .2. In 2007, 2008 and 2010, the execution rate of the health budget was under 80 percent. In 2009, only 42.7 percent of the budget was executed. The year 2011 saw the highest level of execution, at 90 percent. Between 2007 and 2011, the total amount of executed health budget increased in real terms from 71.679 million XAF (of Dec. 2012) and 107.784 million XAF. The MSP execution rate has been lower than the overall government’s budget execution rate that averaged 93.4 percent between 2009 to 2012, with a low of 83.9 percent in 2012. Figure 2.16: Government, Ministry of Health budget and budget execution (millions of XAF of Dec. 2012 and percentage) Source: MSP budget information supplied by the MEFPIPP. 184. Between 2008 and 2011 Figure 2.17: Executed MSP budget per capita the real executed budget per (XAF Dec. 2012) citizen went up by one-third. On a per capita basis, the real executed health budget between 2008 and 2011 was as shown in the Figure 2.17 on the right. Source: MSP budget information supplied by the MEFPIPP. 52 Obtaining and analyzing government budget and budget execution information was challenging. The MEFPIPP cannot provide MSP budget and budget execution data in electronic format owing to technical limitations in its computerized budgeting system; therefore, data analysis first required a lengthy data entry process. Additionally, changes in the nomenclature of budget items from year to year make it hard to track budgets and expenditure items over time. 93 Table 2.2: Government health budget and execution, 2007-2013 2008 2009 2010 2011 2012 Execution Execution Execution Execution Execution Execution Execution Execution Execution Execution Budget Budget Budget Budget Budget rate rate rate rate rate Budget item Millions of XAF of each year Personnel 18,448 16,787 91 19,937 9,129 46 18,596 15,680 84 20,614 14,613 71 31,214 n.a. n.a. Goods and n.a. n.a. n.a. n.a. n.a. n.a. 21,938 25,528 116 22,565 22,565 100 30,642 n.a. n.a. services Transfers n.a. n.a. n.a. n.a. n.a. n.a. 22,093 20,989 95 24,093 24,120 100 36,914 n.a. n.a. Subtotal Goods and services + 47,796 50,053 105 47,939 23,714 49 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Transfers Investment n.a. n.a. n.a. n.a. n.a. n.a. 29,811 10,456 35 47,957 42,450 89 94,134 n.a. n.a. PIP (including 30,486 9,260 30 35,500 11,330 32 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. HIPC)a Total 96,730 76,100 79 103,376 44,173 43 92,438 72,653 79 115,229 103,748 90 192,903 n.a. n.a. Millions of XAF of Dec. 2012 Personnel 21,471 19,538 91 20,713 9,484 46 19,576 16,506 84 21,416 15,181 71 31,214 n.a. n.a. Goods and n.a. n.a. n.a. n.a. n.a. n.a. 23,094 26,873 116 23,443 23,443 100 30,642 n.a. n.a. services Transfers n.a. n.a. n.a. n.a. n.a. n.a. 23,257 22,095 95 25,030 25,058 100 36,914 n.a. n.a. Subtotal Goods and services + 55,628 58,255 105 49,804 24,637 49 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Transfers Investment n.a. n.a. n.a. n.a. n.a. n.a. 31,381 11,007 35 49,823 44,101 89 94,134 n.a. n.a. PIP (including 35,482 10,777 30 36,881 11,771 32 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. HIPC)a Total 112,581 88,570 79% 107,397 45,891 43% 97,307 76,480 79% 119,712 107,784 90% 192,903 n.a. n.a. a. Program Implementing Partners (including Heavily Indebted Poor Countries, HIPC). n.a. Not available Source: MSP budget information supplied by the MEFPIPP. 94 185. Budget items have changed over time what makes analyzing trends in budget execution by budget item a challenging exercise. As can be seen in Table , budget items differed between the period 2007-2009 and the period 2010-2013. In the former period the items Salaries and Goods and Services were lumped together while no information was presented for Investments. Instead, the budget item Program Implementing Partners (including Heavily Indebted Poor Countries, HIPC) showed up in the data. In the latter period, Salaries and Goods and Services were presented separately, Investments were reported, and the item Program Implementing Partners (including Heavily Indebted Poor Countries, HIPC) was removed. Given this discontinuity in budget line items, it is not possible to present a systematic analysis of trends by item. Figure 2.1 depicts the trends in execution of the government’s health budget and its components. Between 2008 and 2010, Personnel accounted for approximately one-fifth of total government health expenditure. While in 2011, it fell to 14 percent. Goods and services and Transfers represented a decreasing share of total expenditure, from 71.3 percent to 45 percent in 2011. Investments accounted for 14.4 and 40.9 percent in 2000 and 2011, respectively. In terms of execution, the rate seems to be lower for investment spending than for current spending, but in the two years with data available there was a big improvement in the execution rate of investment that went up from 25 percent in 2010 to 89 percent in 2011. Figure 2.18: Execution of Government health budget, 2007-2011 (millions of XAF of Dec. 2012) Source: MSP budget information supplied by the MEFPIPP Health Expenditure by Source 186. A National Health Accounts (NHA) study has been completed and NHA information is now available for the years 2009 and 2010. In May 2013, the MSP, with support from the World Bank and the WHO, completed Congo’s first NHA study. Key NHA information from the study is presented in Table 2. and in the figures that follow. The study obtained information about the sources of health financing in the country for the years 2009 and 2010, including government, households, and donors, and the uses of those sources. Use of health funds was classified according to function (upper part of the table), and to the type of provider (lower part). 95 Table 2.3: Health expenditures by financing agents by function and by providers, 2009 and 2010 (millions of XAF of each year) 2009 2010 Other Other Ministry of ministries Ministry of ministries Health and including Other Health and including Other Function Population prefectures Households sources Total Population prefectures Households sources Total Health expenditures by financing agents according to function Curative care 20,230 602 24,973 922 46,727 20,889 2,633 25,701 2,015 51,238 Laboratory exams 571 - 1,349 - 1,920 674 - 1,388 1 2,063 Imaging exams - 722 - 722 - - 743 - 743 Pharmaceutical products - 103 20,019 491 20,613 - 679 20,603 950 22,232 Therapeutic devices - - 2,282 - 2,282 - - 2,348 - 2,348 Maternal and child health care 271 - 129 400 881 - - 134 1,015 Fight against transmissible diseases 507 3,411 - 38 3,956 5,463 7,959 - 60 13,482 Other promotion and public health activities 2 - - 2 31 - -31 - Public management of health system 21,932 27 - 4 21,963 27,308 264 - 6 27,578 Unspecified - - 40 40 75 77 - - 152 Capital formation 3,777 - - 167 3,944 13,211 1,747 - 424 15,382 Human resource formation 60 - - - 60 66 2 - 1 69 Research and development - - - 23 23 - - - - - Sanitary inspections - - - - - - 38 - - 38 Total 47,350 4,143 49,345 1,814 102,652 68,598 13,399 50,783 3,560 136,340 Health expenditures by financing agents according to provider Hospitals 19,841 582 10,776 922 32,121 23,180 2,543 11,090 100 36,913 Medical and dental offices 0 0 14,341 0 14,341 0 0 14,759 1,906 16,665 Other health professional’s offices 0 0 3,856 0 3,856 0 0 3,969 0 3,969 Ambulatory health centers 2,239 20 2,685 1 4,945 6,091 820 2,763 172 9,846 Laboratories 571 0 0 0 571 728 0 0 0 728 Other ambulatory care providers 771 26 0 797 827 0 27 -1 853 Pharmacies 0 10 55 492 557 0 0 57 950 1,007 Optical, hearing, and other medical devices 0 0 53 0 53 0 0 55 0 55 Retailers of pharmaceutical products 0 0 16,548 0 16,548 0 0 17,030 0 17,030 Public health program 754 3,411 0 167 4,332 5,906 8,054 0 196 14,156 Public management of health system and of health insurance 23,099 119 0 28 23,246 29,980 1,236 0 6 31,222 COMEG 75 0 0 0 75 60 0 0 0 60 Training institutions 66 402 469 Other institutions delivering related health services 0 38 38 Rest of the world 0 0 173 0 173 0 228 178 3 409 Unspecified 0 0 829 0 1,037 1,762 77 854 258 2,951 Total 47,350 4,142 49,342 1,610 102,652 68,600 13,398 50,782 3,590 136,371 Source: Ministère de la Santé et de la Population (2013). 96 187. In the years 2009 and 2010, total health system financing in Congo represented a relatively small share of GDP. In 2009, total health financing amounted to about 103,000 million XAF, equivalent to 2.1 percent of GDP (Table 2). Both total health financing and GDP went up in 2010, but total health financing as a share of GDP remained rather constant, at 2.2 percent. About one-half of total financing came from government and the other half from households. Table 2.4: Total health financing in absolute amount and as a share of GDP, 2009-2010 (XFA and percentage) Financing component 2009 2010 GDP 5,006,906 6,260,496 Absolute amount (XAF of each year) Total health financing 102,652 136,369 Government financing 49,206 77,914 Household financing 49,344 50,783 Donors and NGOs 2,681 4,743 Other financing 1,421 2,929 Relative amount (% of GDP) Total health financing 2.1% 2.2% Government financing 1.0% 1.2% Household financing 1.0% 0.8% Donors and NGOs 0.1% 0.1% Other financing 0.0% 0.0% Source: Ministère de la Santé et de la Population (2013). 188. Health system financing in Congo is heavily dependent on households’ out-of-pocket spending (OOPS). According to the NHA study, in 2009, 48 percent of all financing for health care came from households; in 2010 it fell to about 37 percent. However, those estimates may under-estimate current figures since they were inferred on the basis of the 2005 ECOM household survey. When the NHA report was produced, results from the more recent 2011 ECOM survey were not yet available. An analysis of the more recent ECOM results estimates a total household spending amount of 112,352 million XAF that is more than twice higher than the equivalent figure reported in the NHA report for the year 2010 (about 50,000 million XAF). If this figure of 112,352 million XAF is the most accurate, then household OOPS on health would 97 be 30 percent higher than the government reported health spending in 2010, of 77,914 million XAF.53 189. Donor financing for health care in Congo, on the other hand, is relatively low for regional standards. Other sources of financing, such as community contributions and spending by enterprises are also very small overall at the national level. Figure 2.19: Sources of health financing, 2009 and 2010 2009 2010 Source: Ministère de la Santé et de la Population (2013). 190. Between 10 percent and 15 percent of all public spending on health care is channeled through other ministries and public institutions. Much of that spending is dedicated to the fight against AIDS, through the SEP/CNLS (National Council for the Fight Against AIDS Permanent Executive Secretariat), and benefits the entire population, while a small part of it is for the benefit of small population groups such as the staff and families of the Ministry of Defense and the National Assembly. 191. A large share of government health spending is devoted to the administration of the public health system, followed by curative hospital care. One-third of all government health spending is devoted to health system administration. The next most important spending category 53 The results obtained from the 2011 ECOM for households’ OOPS on health in the year 2011 may or may not be directly comparable with those reported in the NHA report. Comparability depends on the exact methods used by the authors of the NHA report to estimate household OOPS for the years 2009 and 2010 on the basis of the health- related OOPS amount reported in the 2005 ECOM survey, however they are further proof of the importance of household spending in health and its upward trend. 98 is the provision of curative care, with 29 percent of all spending. The third and fourth most important categories were investments (capital formation) and programs and activities for the fight of communicable diseases, such as HIV and tuberculosis. Maternal and child health services, a category which includes pre- and post-natal care, as well as vaccination and growth monitoring for children, represented only 1 percent of government health spending in 2010. Figure 2.20: Structure of Government and households health expenditures by provider category and function, 2010 (millions of XAF of each year and percentage) Congo: Structure of government health expenditures by function, Provider Congo: Structure of government health expenditures by function, Function 2010 (million of FCFA and percent) 2010 (million of FCFA and percent) Other Other Laboratory Rest of the Maternal and ambulatory Laboratories world institutions child health Pharmaceutical exams care providers 728 delivering products COMEG 228 care 674 827 1% related health 679 60 0% 881 1% 1% services 1% Other 0% 1% Unspecified 38 289 Training 0% 1,839 0% institutions 2% Fight against 468 Ambulatory 1% transmissible Public health centers diseases management of 6,911 13,422 health system 9% 16% 27,572 Public health 34% program Public 13,960 management of Capital 17% health system formation and of health 14,958 insurance 18% 31,216 38% Curative care Hospitals 23,522 25,723 29% 31% Source: Ministère de la Santé et de la Population (2013). 192. Government spending on COMEG, the public central purchaser of essential medicines, represented a negligible share of public resources devoted to health in 2010 Similarly, when looking at the structure of government, by function, as reported in the recent NHA report we find an almost negligible spending by government on medicines (see Pharmaceutical products), equal to only 1 percent of its total health spending. 193. For households, the situation was quite contrasting. Looking at the structure of household OOPS by category obtained from the 2011 ECOM, medicines (so called modern and traditional in the survey) are the major expenditure category, accounting more than one-half, or 56 percent of household spending on health. Spending on actual medical services, on the form of fees paid to public and private providers, represents just 20 percent of household health OOPS (16 percent on medical and dental offices plus 4 on auxiliary medical services). 99 Figure 2.21: Structure of household annual out-of-pocket health spending, around 2010-2011 (percentage) Source: Authors calculations from ECOM 2011 database. 194. Further analysis of the ECOM 2011 dataset shows that the share of total spending devoted to health was almost constant among the income quintiles (Table 2.). Total household OOPS on health was 112,352 million XAF, representing on average 1.5 percent of total household spending. The share of total spending devoted to health was almost constant among the income quintiles, with the households in the poorest quintile spending the highest percentage at 1.8 percent. Table 2.5: Household total and health spending around 2011 according to ECOM 2011 survey (millions of XAF) Total Total out-of-pocket household health spending spending (OOPS) (THS) OOPS/THS Quintile 1 (poorest) 10,831 618,408 1.8% Quintile 2 17,228 1,028,264 1.7% Quintile 3 21,728 1,388,241 1.6% Quintile 4 27,026 1,830,854 1.5% Quintile 5 (richest) 35,540 2,449,999 1.5% Total 112,352 7,315,765 1.5% Source: Authors calculations from ECOM 2011 data. 100 195. Strong reliance on OOPS in Congo is partly the consequence of a drug revolving fund-like system that operates in all or virtually all government health facilities. Aside from autonomous public hospitals, which receive a block grant and may use part of their budget to purchase medicines, all other public providers lack budget resources to purchase medicines and other medical supplies. To have a stock of medicines they charge user fees to patients and with the revenue collected from these fees they purchase medicines from COMEG or private providers. In Congo, the reliance on user fees for financing of medicines in government health facilities is in line with the Bamako initiative, which calls for the implementation of a drug revolving fund. 196. The fact that the government has not issued any guidance regarding the services and products that can be subject to a fee and the level of fees, has resulted on each facility adopting its own fee system. There seems to be no single policy regarding the setting of fees. This means that user fees in public facilities may be imposed not only on medicines, but also on other goods and services provided to patients. It also implies that user fees vary from one facility to another. Additionally, there is no policy for the provision of waivers or reduced fees to lower income patients. Catastrophic Payments for Health Care 197. Responding to medical needs can absorb a large share of the household budget, which may be considered catastrophic in view of the required sacrifice of current consumption and/ or the long-term consequences for household welfare of borrowing or depleting assets to pay for health care. To further assess the impact of the disruptive effect of health OOPS in household’s living standards an analysis of household’s catastrophic payments for healthcare using data from the ECOM 2011 survey was conducted following the methodology outlined in O'Donnell et al. (2008). This approach consists in defining medical spending as catastrophic if it exceeds some fraction of household total or discretionary (non- food) expenditure in one year. 198. Table 2. shows the results for the Congo presenting different thresholds of household non-food expenditure. Table B.1 in Annex B presents the same results using total expenditure instead of non-food expenditure. In terms of which threshold to consider it is important to think about the level at which the majority of households are forced to forgo other basic needs - while 10 percent of household expenditure on OOPS healthcare payments is catastrophic, 10 percent of discretionary expenditure is likely not catastrophic. Therefore the results are present considering different thresholds. 101 Figure 2.22: Health payment shares Health payments as % of household expenditure 1.4 1.2 1 0.8 OOP/total 0.6 exp. OOP/nonfoo d exp. 0.4 0.2 0 0.0 0.2 0.4 0.6 0.8 1.0 Cumulative proportion of households ranked by decreasing health payments budget share Source: Authors calculations based on ECOM 2011. 199. OOPS payments for health care absorb more than one quarter of household resources net of food costs in 0.9 percent of households with this percentage reaching 2 percent for the lowest quintile of income. Graphically we can also see in Figure 22 how the percentage of households spending more than 20 percent of their non-food income in health is reduced. Even though such numbers are low and can be considered common in developing countries, they mean that for some of the poorest families such levels of spending can only be accommodated through the diversion of considerable resources from current consumption and/ or through the accumulation of debt, or the exhaustion of savings and assets with long-term consequences for household welfare. 200. In the Congo, the average budget share for those exceeding the 25 percent of non- food expenditure threshold is 12.5 percent meaning that these households are dedicating to health a share of 37.5 percent of their non-food budget. This is given by the mean positive overshoot, also presented in Table 2.. 201. Results also show that in the Congo the poorer households tend to spend larger fractions of total consumption on health care. This can be seen by the negative concentration indexes (Table 2.). A limitation of this method is that it identifies only those households that actually acquire treatment and does not take into account households that have illness but cannot 102 afford treatment. It is likely that these households actually incur a higher opportunity cost from poor health and they tend to be a lot more common among the poorest quintiles. Table 2.6: Incidence and intensity of catastrophic health payments, using non-food expenditure Threshold budget share 5% 10% 15% 25% 30% 40% Headcount (H) Lowest quintile 31.1 12.2 6.0 2.0 1.5 0.7 2 27.9 8.8 4.8 1.4 0.9 0.4 3 20.9 7.8 2.3 0.5 0.2 0.0 4 16.0 5.2 1.6 0.4 0.3 0.2 Highest quintile 15.0 3.6 0.9 0.1 0.1 0.1 Total 22.4 7.6 3.2 0.9 0.6 0.3 Mean positive overshoot (MPO) Lowest quintile 6.7 9.1 11.1 14.5 13.7 14.3 2 5.5 8.6 9.0 10.9 11.3 11.8 3 4.8 4.5 5.9 5.3 3.9 2.9 4 4.3 5.1 7.6 9.8 10.0 3.8 Highest quintile 4.1 5.1 7.7 25.4 26.3 31.1 Total 5.4 7.1 9.1 12.1 12.4 12.8 Concentration index, -0.158 -0.222 -0.362 -0.431 -0.478 -0.445 C_E Source: Authors calculations based on ECOM 2011 202. In conclusion, except for the year of 2009 (the “Health Year”), public allocations to the health sector have been improving very mildly and in per capita terms, have remained mostly constant. Overall, budget allocations to the sector are low in comparison to countries of similar income and in comparison to the average for SSA. Donor contribution to the health budget is also low by regional standards, but this is not the case for household contributions. The latter finance mostly curative care and medicines, and in 2010 household contributions to the health budget accounted to 37 percent or higher. Analysis of catastrophic payments for health care 103 points to the fact that for some of the poorest families the levels of spending with health can only be accommodated through the diversion of considerable resources from current consumption and/ or through the accumulation of debt, or the exhaustion of savings and assets with long-term consequences for household welfare. Budget execution rates have improved. Challenges remain with regulation of fees applied by providers, as well as control on procurement of drugs both by public and private providers. Several measures introduced to improve service provision seem to be baring fruits. V. The Delivery and Utilization of Healthcare Inventory of Resources – Manpower, Inpatient Beds and Health Care Facilities 203. Human resource management is one of the Figure 2.23: Sub-Saharan African Countries: physicians per main challenges faced by 1,000 people, 2011 Sub-Saharan African countries: Physicians per 1,000 people, 2011 the Congo’s health 1.60 Seychelles system. Weak 1.40 management has led to a Physicians per 1,000 people 1.20 poor deployment of Mauritius 1.00 professional staff and R² = 0.5409 functionality of facilities, 0.80 South Africa especially in rural and 0.60 remote areas. The number Namibia 0.40 Nigeria Botswana of health personnel in all Cape verde Gabon Equatorial Guinea 0.20 categories, has however, DRC ROC significantly increased, 0.00 2.50 3.00 3.50 4.00 4.50 5.00 passing from 10,000 in Logarithm of 2011 per capita GDP (in 2005 int. PPP dollars) 2005 to about 15,000 in Source: World Bank DataBank. 2010 (of which 80 percent are in the public sector). 204. When comparing Congo with other Sub-Saharan African countries, one can see that with 0.1 doctors per 1,000 people, the country has fewer medical doctors than would be expected, given its per capita income. This is shown in Figure 2.23, which shows that richer countries in SSA tend to have a higher per capita availability of physicians. As the figure shows, all countries in the region that in 2011 had a higher per capita income than Congo54 had a greater per capita availability of physicians. Botswana, Equatorial Guinea and Gabon are examples. 54 Appears in the figure as ROC (Republic of Congo). 104 205. Not only does Congo have a relatively low endowment of physicians given its income level, it also has a deficit of hospital beds and nurses with just 0.84 nurses and 1.6 hospital beds per 1,000 people. Figure 2.24: Congo and other reference countries in SSA organized by quintile: selected health indicators, 2009 100 50 90 45 Gabon 80 40 ROC 70 35 Mauritania Cameroon 60 30 Senegal Quintile Rank Cote d'Ivoire 50 25 Benin 40 20 30 15 20 10 10 5 0 0 Per capita GDP 2011 Hospital beds per 1,000 Nurses per 1,000 people Medical doctors per 1,000 (2005 PPP int. dollars ) people people Source: World Bank DataBank. 206. Yet the Congo has a huge asset in terms of health personnel which is to be found within its diaspora, full of countless practitioners working in the West (France, Belgium, Germany, USA, etc.) and in other African countries. They are even more numerous than those actually working inside the country. Thus, a significant improvement in the terms and conditions of exercising their profession on national territory could give them the desire and the possibility of returning to exercise, and this would contribute to solve both the problem of scarcity and inadequacy of resources. 105 Box 2.1 On Quality of Service Delivery In undertaking this study, the authors were unable to find reliable data to describe the quality of health services in the country. However, the discrepancy between an outcome such as the maternal mortality ratio, which remains high in Congo at 429 maternal deaths for 100,000 live births and the very high rates of institutional deliveries assisted by qualified health personnel (94 percent overall and 81 percent in the lowest quintile) points to a deficit in quality of services. The performance-based financing (PBF) approach, which is being scaled up nationally, has shown to generate improvements in quality while increasing the quantity of basic essential services. The incentive structure of PBF is focused not only on quantity of services delivered but the approach also tracks and remunerates based on improvements in quality. All levels of the health system are incentivized to improve service quality and to plan specific actions, which are assessed regularly and independently through an extensive quantified quality check list. In the scaled-up approach, the checklist will include Vignettes (standardized medical cases) that will check provider competence on treating common potentially life-threatening conditions, and knowledge related to obstetrical danger-signs. Regional Disparities and Efficiency 207. In 2012 the MSP sponsored the production of a statistical yearbook on health facilities and health services. It is the first such effort being carried out over the past 15 years. A draft report has recently been released (Ministry of Health and Population, 2013). Figure 2.25 shows the regional distribution of health facilities of different types according to the information available in this report. 208. Dispensaries, present only in rural areas, show broad variation among départments. The highest availability occurs in Cuvette-Ouest, with 46.4 facilities per 100,000 people. This is equivalent to an average of 2,155 persons per facility. The lowest availability occurs in Niari, with 8.1 facilities per 100,000 people, equivalent to 12,345 persons per dispensary or health post. There is great disparity in availability, but this may respond in part to the degree of rurality of the départments. Additionally, appropriate interpretation of this information calls for a spatial analysis, which takes into account the area of a départment (for example in square kilometers) and the population density. 106 Figure 2.25: Availability of health facilities by département Source: MSP 2013. 209. There is also broad variation in the availability of health centers (CSIs) across départments. MSP policy calls for one CSI per 2,500 to 10,000 people in rural areas and one CSI per 10,000-15,000 people in urban areas. In some départments the availability of CSI is consistent with the national standard, while in others it is not. This can more easily be seen in Figure 2.26, where the upper bound values of the urban and rural standards are shown as a horizontal line. Brazzaville, Bouenza, and Pointe Noire do not meet the norm. Figure 2.26: Population per integrated health center (CSI) by départment, 2012 Population per integrated health center (CSI) by department, 2012 160,000 140,000 120,000 Population per CSI 100,000 80,000 60,000 40,000 20,000 - CSI with PMAS CSI with PMAE CSI with PMAS + CSI with PMAE Upper value of norm in rural areas Upper value of norm in urban areas Source: MSP 2013. 107 210. As the country’s two largest urban areas, Brazzaville and Pointe Noire exhibit the lowest number of reference hospitals per 100,000 people. They also present the highest population per hospital bed. 211. Utilization rates of functional beds are very low in Base Hospitals, except in Brazzaville. The annual statistical yearbook contains information about the number of functional beds, the number of hospitalizations, and the number of bed days in the country’s Base Hospitals and National Hospitals. That information is summarized in Table B.2 through to Table B.5 in Annex B. In 2012, Base Hospitals had a total of about 31,000 discharges while National Hospitals had 69,000 discharges, for a national total of approximately 100,000 discharges in the public sector. Whereas the private health sector is well developed, it is active mostly at the ambulatory level. Therefore the above number of discharges may be close to the total in the country. With a population of 4.3 million, the resulting annual hospitalization rate in 2012 was 2.3 percent. Hospital utilization rates were generally low everywhere in the country, varying between a low of 7 percent in Sangha and a high of 54 percent in Kouilou. The exception was Brazzaville with a much higher hospital utilization rate (98 percent). In National Hospitals utilization rates were higher, reaching about 64 percent in Brazzaville’s large CHU. Table 2.7: Utilization rates in Base Hospitals, by départment, 2012 (percentage) Obstetrics and Department Medicine Pediatrics Surgery Gynecology Total Kouilou 59 46 60 54 54 Niari 27 29 34 59 36 Lékoumou 24 43 20 12 25 Bouenza 9 92 18 3 16 Pool 37 35 24 15 27 Plateaux 28 48 25 27 31 Cuvette 19 17 20 9 16 Cuvette-Ouest 11 40 38 28 24 Sangha 6 16 6 3 7 Likouala 28 26 11 52 27 Brazzaville 34 135 69 116 98 Pointe-Noire 21 43 22 28 28 Source: MSP (2013). 108 212. Low access to curative care may be the result of relatively high user fees, of user fees being charged to all patients in government facilities irrespective of their ability to pay, of poor quality of health services, and other problems in supply . A study about quality of care in government ambulatory and inpatient facilities found that while most CSIs studied had basic medicines at the time of the visit, more than half had experienced inventory stock outs in the 30 days preceding the survey. It also found that few facilities had treatment protocols and few health staff had been trained in the use of protocols (PDSS 2010). 213. When looking at estimates of government per capita expenditure by region we can also see strong disparities. The département where the government spends the most per person (Cuvette) on health is spending three times more than the lowest-spending département (Kouilou). This is not necessarily bad, as mentioned the levels of rurality in different départements are diverse and providing health services to rural, widespread populations might result in a much higher cost per capita then in urban settings (more information would be needed to evaluate). In the same way, the burden of disease as well as the exact needs of the populations might vary across départements and that can influence spending patterns. Figure 2.27: Estimate of Government expenditure per capita by département Source: Authors estimation based on data from MSP (2013). 109 Analysis of Performance of Some Variables in the Health Sector Box 2.2 On the Performance Analysis In order to assess the levels of efficiency of the system, the authors tried to analyze the existence of any correlation between different levels of spending per capita, in each of the regions and variables that approximate the use of the system (numbers of visits) and health outcomes in each of the regions. This analysis is illustrative only, pending more regular and reliable collection of health data at the département level. First of all, there is a time consistency problem, the expenditure data refers to 2011 and the survey data on health use/ outcomes was obtained from surveys that stretch around the period of 2011/2012. Also some of the survey questions asked for information relative to the previous five years. Finally, government budget data was not complete at regional levels and the numbers for some categories of spending had to be imputed. Going forward, regular data collection at département level should be encouraged if more reliable analysis of these relationships is to be undertaken. 214. There is some evidence of a positive relationship between public per capita spending and indicators of health system performance, such as number of visits to public health facilities (as reported in the ECOM 2011 survey), the use of contraceptives and the percentage of women that received pre-natal care and that took anti-malarial drugs during pregnancy. However, the positive relationship does not hold for all outcomes. Important indicators on children’s health like the percentage of fully immunized children show negative correlation with per capita spending. Figure 2.28: Relation between per capita public expenditure on health by départment and selected indicators of use and performance 110 Source: Authors estimation based on data from: Expenditure: MSP 2013; Visits to health facilities: ECOM 2011; Health outcomes: ECSD 2011-2012. 215. There is also evidence of a positive correlation between number of beds in public facilities and public spending on health per capita – this suggests that resources are located where people are looking for them. Brazzaville is an outlier with a much larger number of visits than any of the other regions. Figure 2.29: Per capita public expenditure on health and number of beds in public facilities by region Source: Authors estimation based on data from MSP (2013) 111 Equity in Access 216. By income. In Congo, the prevalence of health problems does not vary much across quintiles – according to self-reported survey data that asks about illness in the previous four weeks (ECOM 2011). In the same way, the percentage of respondents that reported seeking help from some type of health provider for the above-mentioned problem does not differ much across the quintiles of wealth (Figure 2.30). This seems to be a good sign in terms of equity – poorer individuals are not excluded from the system and know where to seek health care. It can also be a sign of the high level of urbanization of the country; according to the same survey a vast majority of population (85 percent) lives within 5km of a health facility. Figure 2.30: Reporting of illness and visits to health provider Source: ECOM 2011. 217. When looking at the type of health provider sought it is clear, as expected, that the population in richer quintiles has more access to private care and uses less traditional medicine healers. The use of public facilities, however, does not vary much across quintiles, which is also a positive sign in terms of equity of access to public services. Figure 2.31: Visits to health facilities by type of provider and expenditure quintile 100% 90% Other 80% Pharmacy 70% 60% Church 50% Traditional healer 40% Private doctor/dentist 30% CSI 20% Public hospital 10% Private clinic / hospital 0% 1 2 3 4 5 Source: ECOM 2011 112 218. In what concerns expenditure, on the government side there is not much variation across quintiles (slightly larger for the 5th quintile), which is also a good sign in terms of equity. However, as we have seen in Table 2. richer quintiles spend a lot more of their own money in absolute terms in obtaining health services. As discussed, there is a real possibility that households in the poorer quintiles are not accessing all the services and medicines they would need for lack of financial resource. Figure 2.32: Expenditure in health by Government and households 60,000 50,000 Millions CFA 40,000 30,000 20,000 10,000 - 1 2 3 4 5 Quintiles Government Household Source: ECOM 2011 and MSP. 219. Indeed, when asked about their level of satisfaction with the service received at health facilities, the most common problem identified is the fact of service being too expensive. This complaint is, unsurprisingly, more common in the lowest quintile of income. This problem seems to have intensified for the bottom quintiles from 2005 to 2011. Similarly, when analyzing the reasons why respondents did not go to a health facility when they became ill, high prices is the number one reason mentioned, particularly for the bottom quintiles of income. In this case, however, the situation seems to be improving from 2005 to 2011. Figure 2.33: Problem identified during last visit to health facility, if any Source: ECOM 2005 and 2011. 113 Figure 2.34: Reason why did not visit health provider when needed during last illness episode Source: ECOM 2005 and 2011. 220. All these differences in access and ability to afford services and medicines are reflected in different health outcomes across quintiles. Obviously, not all the differences in health outcomes across quintiles can be attributed to inequities in access to the health system, several socio-economic characteristics and behaviors that are correlated to income (education, type of housing, etc.) play an extremely important role as well. As an example, the following graphs depict the situation across quintiles for infant and child mortality and access to malaria treatment. 221. Children living in better off households exhibit a much lower IMR than children residing in poor households. The inequality in the IMR among wealth quintiles has narrowed over time. The reduction in IMR inequality can also be seen Figure 2.36, which shows the IMR concentration curves for the two periods. The CMR has also dropped in an important way since 2005, but CMR inequality among wealth groups has remained rather constant, if not increased slightly between the two surveys. The same inequalities across quintiles can be seen when looking at malaria treatment for children who show malaria symptoms and mothers receiving antimalarial drugs during pregnancy. 114 Figure 2.35: Child mortality rate Figure 2.36: Concentration curve for Congo: Concentration curve for the child according to mother's education and child mortality rate, 2005 and 2011-2012 mortality rate, 2005 and 2011-2012 household wealth quintile, 2005 and 100 2011-2012 (deaths per 1,000 live births) 250 80 202 Cumulative % of child deaths 200 150 60 CMR 106 100 68 57 40 50 0 Q2 Q3 Q4 None Primary Secondary 2nd cycle Q1 (poorest) Q5 (richest) Secondary 1st cycle 20 0 0 20 40 60 80 100 Mother's education Wealth quintiles Cumulative % of the population EDSC 2005 EDSC 2011-2012 EDSC 2005 EDSC 2011-2012 Equality line (base 100) Source: CNSEE and ICF International (2012) and World Bank’s DataBank. Figure 2.37: Access to malaria treatment by expenditure quintile 100 80 60 Percent 40 20 0 1 2 3 4 5 Quintiles % children being treated for fever % woman that received anti-malarial drugs during pregnancy Source: EDSC 2011-2012. 115 222. Minorities. Most of Congo’s autochthonous populations belong to the Baaka ethnic group and traditionally live in the forests. Autochthonous groups live on the margins of the dominant culture and maintain a distinct identity shaped by their environment and history. Their remoteness adds to their fragility. Living in the forest, away from roads, autochthonous populations do not have very often access to education or health services. General lack of awareness on health issues increases the risk of disease and malnutrition so they are a group of concern when discussing the country’s health system and in particular equity of access to the health system. 223. In Congo, less than 18 percent of autochthonous women receive ante-natal care and only one in four women gives birth in a health facility. Most give birth in the forest or in their village, sometimes alone, which increases risks for mother and child. With 781 deaths per 100,000 live births, maternal mortality is high throughout the country. Although no data are available on the subject the risks associated with childbirth are likely to be much higher for the Baaka. 224. Since data on autochthonous groups’ access to health is close to inexistent, an analysis of rural versus urban results for certain indicators was carried out, under the assumption that “rural” data can give us an approximation of the situation of autochthonous minorities in Congo, particularly in the départments Likouala, Lékoumou and Sangha, which present the highest number of autochthonous population. It is important to keep in mind, however, the fact that a large share of these populations lives in remote areas in the forest and will likely be missed by any survey. Comparing urban and rural populations on a few selected health outcome indicators, it is not surprising to see that results are worse for rural areas in most indicators, IMR and CMR, use of contraceptives, pre-natal care and birth assistance. In terms of IMR and CMR it is positive to see that the improvements identified at the country level, also hold for rural areas. Table 2.8a: Key health indicators in urban and rural areas, 2005-2011 Neo-natal Post-neo-natal Infant mortality Child mortality mortality mortality 2005 2011-2012 2005 2011-2012 2005 2011-2012 2005 2011-2012 Urban 36 26 31 18 66 45 108 77 Rural 35 21 58 29 93 51 136 88 Source: EDSC 2005 and 2011-2012. 116 Table 2.8b: Key health indicators in urban and rural areas, 2011 Uses one Received Used modern Gave birth in contraceptive professional pre- contraceptive health facility method natal care Urban 46.3 24.6 95.9 97.4 Rural 41.9 11.7 86.8 82.4 Source: EDSC 2011-2012. 225. Rural populations report a higher rate of illness than urban ones (40 percent versus 36 percent), but also report a higher use of health services for the reported illness . This at least gives some hope that services are available and populations know how to use them including in rural areas (Figure 2.38). Figure 2.38: Reporting of illness and corresponding visit to health providers 100% 80% 60% 40% 20% 0% urban rural Yes No Visited health provider Source: ECOM 2011 226. When looking at the type of provider sought, it is clear that in rural areas there is less reliance on private health services and more reliance on informal providers like traditional healers and churches. The use of public services, however, does not vary much from urban to rural, with rural tending slightly more to CSIs than hospitals. These facts can be just a reflection of the different availability of services in rural areas. 117 Figure 2.39: Visits to health facilities by type of provider and rural/ urban location Source: ECOM 2011 227. Consistently, the amount spent by rural households in health expenditures is much lower than that spent by urban households. However, in terms of the proportion of total spending that health represents there is not much different between urban and rural households, with the rural ones even spending a slightly higher percentage. Table 2.9: Household total health spending around 2011 according to ECOM 2011 survey (millions of XAF) Total out-of-pocket health spending % of total household spending (OOPS) dedicated to health Urban 83,888 1.4% Rural 22,215 1.6% Source: ECOM 2011. 118 228. Looking at reasons for not accessing a health provider in case of illness, it is clear that surveyed populations in rural areas report more problems both in terms of the price of services and distance. In the same way, rural households seem to have more reasons for complaining about the service recently received than the urban ones. With the “too expensive” complaint leading the way, followed by “lack of medicines”. These results seem consistent with the ones observed for the different levels of income quintiles, and price again seems to be a barrier of access to low income populations. Figure 2.40: Reason why dud not use health service when needed, 2011-2012 9% 8% 7% 6% 5% urban 4% rural 3% 2% 1% 0% too expensive too far other Source: ECOM 2011 Figure 2.41: Reason why unsatisfied with service, 2011-2012 40% 35% 30% 25% 20% 15% 10% 5% 0% not waiting expensive lack of lack of treatment other at least one welcoming time trained staff medicines did not problem work urban rural Source: ECOM 2011 119 Public Spending and Equity 229. In the Congo, government spending with ambulatory care is pro-poor, while government spending in hospitals is slightly pro-rich. Benefit Incidence Analysis using data from the ECOM 2011 and from the NHA study (2010) allowed us to estimate the amount of government subsidy by visit to public providers, separately by hospital and ambulatory centers assuming constant unit subsidies. 230. Subsidies per visit are higher for hospitals than ambulatory centers, as expected . As seen before, hospitals receive more visits than ambulatory centers but in general ambulatory centers are more used by the poorest segments of the population. Table B.6 in Annex B shows the detailed breakdown of government spending by type of provider and income quintile. 231. Taking total spending together gives a picture of a system that is not markedly pro- poor or pro-rich. This can be seen by the total subsidies curve in Figure 2.42 that closely mirrors the equity line. To get a more accurate picture of government’s spending by population segments it would have been useful to have a breakdown of spending by in-patient and outpatient at the hospital level, since the costs associated with each can be quite different, but that was not possible due to data limitations. Figure 2.42: Concentration curves of Government spending on hospital and ambulatory care 1.00 0.90 0.80 0.70 Cumulative % of subsidies Line of 0.60 equality 0.50 "Subsidy for visits to hospital" 0.40 "Subsidy for visits to amb. 0.30 center" Total subsidies 0.20 0.10 0.00 0.00 0.20 0.40 0.60 0.80 1.00 Cumulative % of population, ranked from poorest to richest Source: Authors calculation based on ECOM 2011 and MSP 2013. 120 232. In conclusion, the limited resources (human, physical and financial) available in Congo are not distributed evenly across the country what contributes to inequalities in access and outcomes across different populations. The excessive price of services seems to create a barrier to the poorest households and the ones living in rural areas. The low utilization rates of beds in base hospital and limited use of services in general, suggests efficiency problems in the system. However, there is some evidence of a positive relationship between public per capita spending and indicators of health system performance, such as number of visits to public health facilities (as reported in the ECOM 2011 survey), the use of contraceptives and the percentage of women that received pre-natal care and that took anti-malarial drugs during pregnancy that can suggest that services are located where they are the most needed. However, this positive relationship does not hold for all health outcomes. Government spending in health is not markedly pro-poor or pro-rich. VI. Conclusions and Recommendations Congo is showing a positive trend in health status indicators, but important data gaps remain to allow for a more detailed understanding of improvement and challenges in the sector 233. The latest EDSC 2011-12 has revealed that key strategic health status indicators have improved in an important way in the Congo. Up until the release of that report, key health indicators, such as the infant, child and maternal mortality rates were much higher than expected in the SSA context for a country with such a high per capita income. Further, it was reported that these indicators were stagnant or deteriorating and that the prospects that Congo would reach the health MDGs were remote. In contrast, the latest estimates indicate that Congo has achieved impressive gains in health status. According to that report, the IMR now stands at 39 deaths per 1,000 live births, the CMR at 68 per 1,000 and the MMR at 429 deaths per 100,000 live births. 234. The EDSC 2011-12 also reports major improvements in knowledge and in access to some health services. For example, knowledge about modern contraception improved and gaps in knowledge among socioeconomic groups narrowed in an important way between 2005 and 2011-12. Likewise, use of modern contraceptive methods went up by 80 percent overall but inequalities in access remained large - women in the richest income quintile were three times as likely to use modern contraception as women in the poorest income quintile. The coverage of pre-natal care went up from an already high figure of 88 percent in 2005 to 93 percent in 2011- 12; during the same period of time the rate of C-sections increased, from 3.2 percent to 5.8 percent, in part owing to a new government policy that promotes the provision of free C- sections; child vaccination coverage went up as well; and there was a 50 percent increase in the treatment of child diarrhea with ORS. 121 235. There are discrepancies between the data reported in the EDSC 2011-12 and in the newly-released MSP Statistical Yearbook. In 2013 the MSP produced its first Statistical Yearbook after more than 12 years. The document reports the inventory of public and private health care facilities in the country, as well as detailed utilization statistics for a broad range of preventive and curative services, both in outpatient and in inpatient facilities. This publication is in itself a great achievement. Yet some of the figures reported in the Yearbook contradict some reported in the EDSC. For example, the Yearbook reports a considerably lower number of institutional deliveries than the EDSC, which indicates that institutional deliveries account for about 94 percent of all deliveries in Congo. 236. In the same way, there are discrepancies between the most recent EDSC 2011-12 and the official data reported by the World Bank and other international agencies.  Since health status improvement is a central objective of any health system, it is indispensable that a new survey be conducted in the near future, to verify that the gains reported in the EDSC are maintained or furthered, and also to confirm that results are coherent with the EDSC 2011-12.  Additionally, government and the development community should continue to support in a systematic way the initiative that led to the production of the 2012 Statistical Yearbook. That involves the strengthening of institutional capacities in health management information system, both at the central and decentralized levels. 237. Child and maternal health status improved nationally, according to EDSC 2011-12, but inequality in health status did not drop in all cases. For example, inequality among socioeconomic groups in the CMR increased slightly over the seven-year period that elapsed between the previous survey, EDSC 2005, and the more recent one. In the case of the IMR, in contrast, not only did this rate drop as a national average, but inequality among socioeconomic groups dropped as well.  Achieving nationally better health status indicators in Congo is commendable, but policymakers in the country should pay more attention to the evolution of inequality in health status, and should adopt policies that seek to bridge the gaps in health status between the poor and the non-poor. For example, allocating a growing share of public financing to the poorest departments in the country might help to achieve this goal.  Likewise, defining a national policy to identify the poor and vulnerable, in order to waive them from user fees in government health facilities might also help to reduce inequity in access to basic health services and nutrition supplements, and inequality in health status. A policy of targeted public subsidies for the poor should comprise also a mechanism to compensate public providers for the user-fee income forgone as a result of their adoption of waivers for the poor and vulnerable. 122 Increasing attention has been given to the health sector in the national budget however public health spending still remains low and reliance on household expenditure is very high 238. Real government health spending has increased since 2009 both in absolute terms and as a share of GDP. In real terms, per capita government health spending in 2009 was 28,563 XFA. Assuming that in 2012 and 2013 the MSP maintained the budget execution rate of 90 percent seen in 2011 (only budgets but not actual execution rates are known for the last two years), then real per capita government health spending would have increased to 44,478 XFA in 2012 and 35,840 XFA in 2013. As a share of actual GDP, government health spending was 1.7 percent in 2011, and it is estimated to have been 2.8 percent in 2012, and 2.1 percent in 2013. Government declared the year 2012 Health Year and, accordingly, it significantly expanded health spending. 239. Still, in a regional perspective, in 2009 and 2010 the Congo devoted a relatively low total amount of resources to the health sector. As a share of GDP, total health financing was 2.0 percent in 2009 and 2.1 percent in 2010. Whereas the Congo is among the 20 percent richest countries in SSA, its share of GDP spent on health is was among the smallest in the region. The nine richest countries in SSA (which represent the top quintile out of 45 countries in the region) currently devote, on average, 6.9 percent of GDP to the health sector. That is more than three times Congo’s figure in 2010.  Further improving health status and financial protection in Congo’s health sector necessarily calls for increased volumes of financing. This would be in line with the Abuja Declaration by which African Union members pledged to commit 15 percent of their budges to health. For example, Rwanda, which is aggressively seeking to expand health coverage, devotes 10.8 percent of GDP to health and 57 percent of that amount comes from government. Ghana, which is also actively promoting health coverage expansion, devotes 4.8 percent of GDP to health and government contributes 56 percent of that amount. 240. The allocation of health financing between government and households reported through the NHA study for Congo in 2009 is not atypical for countries in SSA, although the authors’ estimates suggest that households may be contributing a greater share than reported. Among the top 20 percent richest countries in SSA, government financing represents 3.5 percent of GDP, or almost exactly one-half of total financing. According to the NHA study, in 2009 government and household financing for health were about equally split, but in 2010 government financing went up in relative terms, to reach 57 percent of total health financing. Yet estimates obtained by these authors using the latest ECOM data indicate that household health spending may have been largely underestimated in the NHA study and may represent more than one-half of total health spending. 241. The heavy reliance on households spending results in a large part from the fact that all public providers (aside from autonomous public hospitals) lack budget resources to 123 purchase medicines and other medical supplies together with a total absence of government’s guidance regarding the setting of fees in public health providers. This means that user fees in public facilities may be imposed not only on medicines, but also on other goods and services provided to patients. It also implies that user fees vary from one facility to another. All of this has resulted in a barrier of access to poorer households particularly in the ability to buy medicines that represent the largest share of household spending. 242. Starting in 2009 the government of Congo announced the implementation of some programs that would offer free health care to all citizens, in order to boost access. They included, for example, free malaria medicines and free C-sections. Yet excessive centralization has also been blamed for the limited impact of these programs. Health care inputs (such as medicines) acquired by the central level are typically routed to the department’s capital but in the DDS budgets there is no line allocated to the distribution of these goods. Consequently, the inputs reach only the sites near the department’s capital and benefit peripheral health facilities only when some of their staff happens to be in the capital town when the goods arrive. Solving these logistic problems is crucial to make sure government’s resources are being spent where they are most needed and the vulnerable populations that need free services the most can benefit from the programs. 243. Expanding and rationalizing government health spending and moving toward universal health coverage may call for the explicit definition of a benefits package in Congo. Government budget allocations to ambulatory health facilities, both dispensaries and CSIs, are a small share of their total budget. It is enough to cover the salaries of existing staff, however, it does not cover drugs and other medical supplies, and there is information that indicates that the number and type of health staff currently in place in these facilities is insufficient to deliver appropriate levels of quality services. Up until now the government has not formulated an explicit benefits package for the population, and therefore financing for ambulatory and hospital facilities is not in accordance with explicit delivery targets. The Dutch NGO Cordaid has been engaged in a project in Congo which involved the definition, costing, and delivery of a benefits package, both at the ambulatory and hospital levels. Extending this package could be considered. Progress and Challenges in the Government’s Budgeting and Expenditure System in Health 244. The formulation of the government’s budget for the health sector was described as a highly centralized process by various health sector stakeholders consulted during the data gathering phase of this PER. The just completed NHA report concludes that A low level of stakeholders involvement in the programming of the government’s health budget often leads to inconsistencies between the health system’s funding needs (both in terms of the kinds and levels of resources) and the actual budget allocations. (p.2) The NHA report also criticizes a lack of coordination among various administrative instances (such as departmental authorities and 124 community representatives) with regard to the planning of investment projects, resulting in a lack of financing for some investments and in the double allocation of financing for others. 245. The highly centralized management of public resources for health does not promote efficient spending by the recipients at the peripheral level (Ministère de la Santé et de la Population 2013). Decentralization in the health sector seems to be limited only to the ability of departmental health directorates (DDS) to fulfill their mission of supervising all health care institutions within their jurisdiction. Whereas DDS formulate annual action plans, reportedly these are not taken into account at the central level during the budget preparation process. Additionally, DDS have said to have no control whatsoever over the actual implementation of health activities in their jurisdiction. Only three percent of the health budget is managed at the departmental level by DDS, while programming of spending is the responsibility of the Prefecture (the local representative of the national government). Although reference hospitals operate within departments, the transfer and allocation of budgetary resources are entirely under central control. 246. Moreover, government budget formulation for health lacks transparency and budget criteria are basic. The main criterion for the formulation of the budget for specific providers and for departments seems to be the replication of the previous year’s budget plus any small adjustments. Improving allocative efficiency of government health spending also calls for the simultaneous adoption of several improved budgeting criteria.  National hospitals, all of which currently receive block grants from government, should be evaluated to determine the efficiency consequence of such grants, and to make any necessary changes in this financing mechanism in order to improve efficiency.  Also, to date government has not introduced innovation in provider payment methods, beyond the block grants to National Hospitals, whose efficiency remains to be determined. The experimental adoption of performance based financing (PBF) in a few departments under the Health Sector Strengthening Project I, and the planned national generalization of PBF under the Health Sector Strengthening Project II, are expected to change in a drastic way the prevailing incentives among public providers. Despite implementation challenges, it is likely that they will result in higher output, better quality, and more equitable access. 247. The historically low rate of execution of the government budgets is common to all sectors and not just health. It has been a concern of government for some time now and led to a government initiative aimed at improving budgetary and execution procedures, with the support of the IMF. However, problems still remain in procurement and disbursement. Since the adoption of the new public procurement code, lack of procurement specialists, resistance of some stakeholders involved in the procurement process, and the adoption of a lengthy procedure to move from the conception into a tender for public contracts have posed challenges. 125 Important challenges in system management remain 248. Government health spending for personal health services is mostly allocated to hospital services, either to National Hospitals or to Base Hospitals. In 2010, nearly 40 percent of all government health spending went to pay for the public system’s administrative costs, mostly at the central level. Just 10 percent of its health spending was allocated to ambulatory heath facilities, while almost one-third of its executed budget went to public hospitals. It seems that the share of the government’s budget allocated to health services at the CSI level is exceedingly low and signals a major problem of allocative inefficiency.  Progressively expanding the share of public resources going to health dispensaries and health centers should be a public policy priority in Congo. It would allow CSIs to waive poor and vulnerable patients from user fees and it would also serve to attract more qualified health staff to these facilities through higher salaries and the adoption of economic incentives. More public subsidies would also serve to subsidize an expanded set of basic medicines, a policy that government has started, but which has met logistical problems owing to a faulty implementation. 249. The authors found limited information with which to judge efficiency in the production of health services in the public sector. A noteworthy finding, however, is the low rate of utilization of hospital beds around the country, except in Brazzaville. Hospital utilization rates were as low as 7 percent in one départment, and in eight out of 11 départments they were below one-third. Considering that utilization was computed on the basis of functional hospital beds, these low rates of utilization reveal an important inefficiency in Congo’s government health system.  Assessing hospital occupancy, the reasons behind low occupancy rates is a research priority. 250. Moreover, access to some services, such as curative ambulatory care, remains very low for regional standards. For example, according to the MSP Statistical Yearbook 2012, the average Congolese make 0.13 annual curative visits to a government provider. This is equivalent to one visit every seven years. This low figure is in stark contrast with the known high frequency of infectious illness episodes among children and adults in Congo. Whereas there is an active private health care delivery sector, it is likely that the visits that occur there would improve in a significant way this low indicator of accessibility to curative care. 251. Low access to curative care may be the result of relatively high user fees, of user fees being charged to all patients in government facilities irrespective of their ability to pay, of poor quality of health services, and other problems in supply. A study about quality of care in government ambulatory and inpatient facilities found that while most CSIs studied had basic medicines at the time of the visit, more than half had experienced inventory stock outs in the 30 days preceding the survey. It also found that few facilities had treatment protocols and few health 126 staff had been trained in the use of protocols (PDSS 2010). Improving access to health services in Congo calls for a strategy with interventions in several fronts.  One front includes actions aimed at improving the quality of health care in government health facilities, for example by making treatment protocols available to health staff and by training them in their use. - The PBF approach which will be scaled up nationally has been shown to generate improvements in quality through its incentive structure which is focused not only on quantity of services delivered but also tracks and remunerates against quality.  Another front is a revision of the user fee policy of government providers. Outright abolition of fees is not recommended because it is likely to lead to greater access problems than in the current scenario, such as widespread stock outs of drugs in facilities. However, government may want to regulate the fees to ensure that they are not abusive, that they are waived for the poor and the provider is compensated accordingly, and that they are charged only for medicines and not for other services.  Improving access also requires that qualified health staff be hired by government and present in government facilities throughout the country. Currently there is a lack of doctors and nurses in CSIs, particularly in the more rural and remote locations. All countries face the challenge of endowing rural health facilities with appropriate staffing, and many countries have adopted a system of economic and other incentives to achieve this objective. Congo would benefit from learning about such approaches, to identify those that might work in the country. 127 Table 2.10: Summary of Issues and Recommendations ISSUE ACTION OUTCOME Increasing attention has been Increase health financing given to the health sector in the  Increase public financing for health Increased progress in strategic national budget however Policy  Maintain and further increase the rate of budget execution through better education goals health expenditure is still and/or procurement and disbursement procedures below average given Congo’s Intervention  Introduce innovations in provider payment methods. per capita income level  Increase alignment of financing with health desired outcomes  Improve drug regulation and functioning of COMEG Focus on pro-poor policies that can compensate for the heavy reliance on out-of-pocket expenditure Inequality in access and outcomes  Review of the implementation of ongoing programs that offer free health Better health outcomes for the Policy services poorest segments of the and/or  Regulate the fees to ensure that they are not abusive, that they are waived for Congolese population and those Intervention the poor, and that they are charged only for medicines and not for other services living in disadvantaged areas  Define a specific benefits package for health providers Better displacement of human and physical resources throughout the country  Increase the share of the government budget that is allocated at the CSI level. Policy  Increase financing to ambulatory health providers and/or  Allocate a growing share of public financing to the poorest departments in the Intervention country  Appropriate geographic distribution of health staff  Implement measures to improve quality of services Make post-primary education affordable Lack of reliable data and  Conduct a new survey to verify that the gains reported in the EDSC 2011-12 are More targeted and planned- information on the key aspects Policy maintained or furthered based policy decisions of the sector such as human and/or  Continue support to the initiative that led to the production of the 2012 resources Intervention Statistical Yearbook, including strengthening of institutional capacities in health management information systems, the training of staff involved in data collection and reporting, and the supply of computers and other equipment required to operate such system  Specific studies assessing hospital occupancy, the reasons behind low occupancy rates, the feasibility of closing some beds and human resources needs 128 Annex A.1 Figure A.1: Share of investment and recurrent expenditure in total education expenditure and in METPFQE expenditure (percentage) TOTAL EDUCATION METPFQE 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 2008 2009 2010 2011 2012 2013 2008 2009 2010 2011 2012 2013 Prog. Prog. Current Capital Current Capital Source: MEFPIPP, Budget Execution and other financial data, 2008-2012. Figure A.2: Composition of the recurrent expenditure MEPSA METPFQE 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Personnel Materials Personnel Materials Transfers Source: MEFPIPP. 129 Table A.1: Distribution of actual public current expenditure by education levels (in percentage of total) 2013 2008 2009 2010 2011 2012 Prog. MEPSA 62.3 63.3 63.3 59.7 57.3 57.0 Pre-primary 0.6 0.8 0.6 0.7 0.7 n.a Primary 40.0 37.2 36.2 32.2 32.0 n.a. Secondary 19.1 19.1 19.8 19.1 19.4 n.a Alphabetization 0.1 0.1 0.1 0.2 0.2 n.a. Transversal to all education levels 2.3 6.1 6.6 7.6 5.0 n.a. METPFQE 14.2 9.3 12.5 14.8 15.7 14.6 MES 23.5 27.4 24.2 25.4 27.0 28.4 TOTAL 100.0 100.0 100.0 100.0 100.0 100.0 Source: MEFPIPP Budget Execution and other financial data, 2008-2012. Table A.2: Distribution of education benefits Primary level Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 All households Education subsidy (Million 264 173 96 53 21 607 FCFA) % of education subsidy 43.5 28.5 15.8 8.8 3.5 100.0 Per capita education subsidy 37,263 24,422 13,519 7,523 3,004 85,732 (FCFA) Secondary level Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 All households Education subsidy (Million 117 99 82 58 33 389 FCFA) % of education subsidy 30.1 25.4 21.2 14.9 8.5 100.0 Per capita education subsidy 36,648 30,858 25,740 18,094 10,290 121,633 (FCFA) Technical and professional Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 All households Education subsidy (Million 26 37 25 34 26 147 FCFA) % of education subsidy 17.5 25.1 16.8 22.9 17.7 100.0 Per capita education subsidy 74,073 106,004 71,277 96,817 75,029 423,089 (FCFA) Tertiary Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 All households Education subsidy (Million 8 23 32 30 47 139 FCFA) % of education subsidy 5.7 16.4 23.0 21.3 33.6 100.0 Per capita education subsidy 65,477 187,078 261,909 243,201 383,510 1,141,175 (FCFA) Source: MEFPIPP Budget Execution and other financial data, 2008-2012. 130 Table A.3: Education budget execution rates (percentage) 2008 2009 2010 2011 2012 BUDGET TOTAL Recurrent budget 52.7 84.2 105.3 107.2 121.6 Wages 99.5 100.9 95.5 97.7 100.0 Goods and Services+Commons charges 107.3 90.7 112.3 125.1 188.7 Transfers 30.4 66.2 107.2 100.2 91.9 Investment 104.7 95.2 88.4 107.3 76.6 Equipment n.a. n.a. n.a. n.a. n.a. Studies n.a. n.a. n.a. n.a. n.a. Rehabilitation n.a. n.a. n.a. n.a. n.a. Construction n.a. n.a. n.a. n.a. n.a. EDUCATION TOTAL Recurrent budget 104.9 81.8 101.7 97.2 100.9 Wages 101.6 99.0 95.6 100.6 104.4 Goods and Services 107.4 30.9 116.6 80.6 87.9 Transfers 120.4 92.0 106.2 109.1 101.7 Investment 39.0 70.8 98.2 91.6 83.4 Equipment 40.5 61.8 51.6 n.a. n.a. Studies 53.0 47.0 73.0 n.a. n.a. Rehabilitation 81.5 24.6 n.a. n.a. n.a. Construction (and Rehabilitation no METPFQE) 24.4 35.2 76.2 n.a. n.a. MEPSA Recurrent budget 100.3 81.6 102.6 90.7 97.1 Wages 98.0 100.8 96.8 95.1 104.6 Goods and Services 107.7 29.2 119.5 77.4 71.2 Transfers 89.0 95.0 98.2 80.6 66.2 Investment 29.0 28.5 98.3 95.0 84.6 Equipment 22.3 18.3 n.a. n.a. n.a. Studies 57.0 n.a. n.a. n.a. n.a. Rehabilitation n.a. n.a. n.a. n.a. n.a. Construction 11.2 30.5 n.a. n.a. n.a. METPFQE Recurrent budget 124.8 60.6 92.8 104.1 107.3 Wages 148.4 86.5 74.1 72.9 80.5 Goods and Services 109.5 19.4 114.7 91.7 143.4 Transfers 95.2 86.0 100.0 238.4 114.1 Investment 87.7 53.1 98.2 79.1 72.7 Equipment 71.6 82.3 64.4 n.a. n.a. Studies 83.3 76.4 37.1 n.a. n.a. Rehabilitation n.a. n.a. n.a. n.a. n.a. Construction and Rehabilitation 67.1 65.4 90.5 n.a. n.a. MES Recurrent budget 107.5 93.5 104.5 111.7 106.1 Wages 95.7 98.1 102.7 142.5 116.1 Goods and Services 100.6 69.0 100.0 78.9 77.9 Transfers 131.9 93.1 108.0 82.2 98.8 Investment 21.4 45.2 97.9 114.3 99.4 Equipment 16.7 39.8 3.2 n.a. n.a. Studies n.a. 56.0 n.a. n.a. n.a. Rehabilitation 81.5 24.6 n.a. n.a. n.a. Construction 33.0 14.3 100.0 n.a. n.a. Sources: Plan National de Développement - Congo 2012-2016 ; Loi de Finances pour chaque année t Annuaire statistique 2010-2011. 131 Table A.4: Education Sector Budget Millions FCFA 2008 2009 2010 2011 2012 2013 Invest Invest Invest Invest Invest Invest Current Total Current Total Current Total Current Total Current Total Current Total ment ment ment ment ment ment Total 1,184,580 450,000 1,634,580 596,161 514,450 1,110,611 591,000 674,257 1,265,257 634,443 1,010,612 1,645,055 843,438 1,961,772 2,805,210 1,002,630 1,796,440 2,799,070 Budget MEPSA 60,008 18,977 78,985 63,351 20,883 84,234 65,314 16,689 82,003 75,127 19,935 95,062 79,170 52,420 131,590 95,807 80,707 176,514 METPQE 11,012 5,030 16,042 12,517 8,746 21,263 14,282 10,434 24,716 16,260 17,918 34,178 19,678 16,200 35,878 24,485 100,000 124,485 MES 21,131 3,160 24,291 23,922 5,100 29,022 24,533 2,962 27,495 25,943 6,869 32,812 34,177 6,856 41,033 47,685 11,750 59,435 Total 92,152 27,167 119,319 99,790 34,729 134,519 104,129 30,085 134,214 117,329 44,722 162,051 133,025 75,476 208,501 167,977 192,457 360,434 Education Sources: Plan National de Développement - Congo 2012-2016 - Livre 2 Loi de Finances pour chaque année (Programed) (LoiDeFinances_2008(2) 15Février) plus Execution Budget 2010, 2011 et 2012 (Executed) Budget de l'État, exercice 2013 (Loi 41-2012 du 29.12.2012) Data from "Budget 2009-2013_Loi de finances", revised in April of 2012, for budget 2012; it replaces data from source 'Loi de Finance pour chaque année. Table A.5: Education Sector Budget Execution Millions FCFA 2008 2009 2010 2011 2012 2013 Invest Invest Invest Invest Invest Invest Current Total Current Total Current Total Current Total Current Total Current Total ment ment ment ment ment ment Total 623,883 471,000 1,094,883 502,127 490,000 992,127 622,028 596,000 1,218,028 680,124 1,084,000 1,764,124 1,025,716 1,502,736 2,528,452 n.a. n.a. n.a. Budget MEPSA 60,210 5,501 65,711 51,671 17,643 69,314 67,036 16,398 83,434 68,115 18,931 87,046 76,863 44,365 121,228 n.a. n.a. n.a. METPQE 13,741 4,410 18,151 7,583 4,647 12,230 13,254 10,241 23,495 16,932 14,182 31,114 21,115 11,781 32,896 n.a. n.a. n.a. MES 22,717 676 23,393 22,368 2,307 24,675 25,633 2,900 28,533 28,984 7,851 36,835 36,278 6,812 43,090 n.a. n.a. n.a. Total 96,669 10,587 107,256 81,622 24,597 106,219 105,923 29,539 135,462 114,031 40,964 154,995 134,256 62,958 197,214 n.a. n.a. n.a. Education Sources: Plan National de Développement - Congo 2012-2016 - Livre 2 Loi de Finances pour chaque année (Programmed) (LoiDeFinances_2008(2)_15Février) plus Execution Budget 2010, 2011 e 2012 (Executed) IMF (2012), Republic of Congo: 2012 Article IV Consultation - Staff Report, p.20 (2011 - estimate; 2012 - projection) MEPSA, Annuaire statistique 2010-2011. 132 Table A.6: Budget Execution rates (current and investment per ministry) % 2008 2009 2010 2011 2012 2013 Invest Invest Invest Invest Invest Invest Current Total Current Total Current Total Current Total Current Total Current Total ment ment ment ment ment ment Total 52.7 104.7 67.0 84.2 95.2 89.3 105.3 88.4 96.3 107.2 107.3 107.2 121.6 76.6 90.1 n.a. n.a. n.a. Budget MEPSA 100.3 29.0 83.2 81.6 84.5 82.3 102.6 98.3 101.7 90.7 95.0 91.6 97.1 84.6 92.1 n.a. n.a. n.a. METPQE 124.8 87.7 113.1 60.6 53.1 57.5 92.8 98.2 95.1 104.1 79.1 91.0 107.3 72.7 91.7 n.a. n.a. n.a. MES 107.5 21.4 96.3 93.5 45.2 85.0 104.5 97.9 103.8 111.7 114.3 112.3 106.1 99.4 105.0 n.a. n.a. n.a. Total 104.9 39.0 89.9 81.8 70.8 79.0 101.7 98.2 100.9 97.2 91.6 95.6 100.9 83.4 94.6 n.a. n.a. n.a. Education Sources: Plan National de Développement - Congo 2012-2016 - Livre 2 Loi de Finances pour chaque année (Programmed) (LoiDeFinances_2008(2)_15Février) plus Execution Budget 2010, 2011 e 2012 (Executed) Budget de l'État, exercice 2013 (Loi 41-2012 du 29.12.2012) Data from "Budget 2009-2013_Loi de finances", revised in April of 2012, for budget 2012; it replaces data from source 'Loi de Finance pour chaque année. IMF (2012), Republic of Congo: 2012 Article IV Consultation - Staff Report, p.20 (2011 - estimate; 2012 - projection) MEPSA, Annuaire statistique 2010-2011. Table A.7: Budget Execution rates at constant prices (current and investment per ministry) Millions FCFA 2008 2009 2010 2011 2012 Invest Invest Invest Invest Invest Current Total Current Total Current Total Current Total Current Total ment ment ment ment ment Total 581,252 438,816 1,020,069 444,273 433,543 877,816 524,159 502,226 1,026,385 565,616 901,493 1,467,109 799,006 1,170,592 1,969,598 Budget Total 90,064 9,864 99,927 72,218 21,763 93,981 89,258 24,891 114,149 94,832 34,067 128,899 104,582 49,043 153,625 Education MEPSA 56,096 5,125 61,221 45,717 15,610 61,328 56,489 13,818 70,307 56,647 15,744 72,390 59,874 34,559 94,433 METPQE 12,802 4,109 16,911 6,709 4,112 10,821 11,169 8,630 19,799 14,081 11,794 25,876 16,448 9,177 25,625 MES 21,165 630 21,795 19,791 2,041 21,832 21,600 2,444 24,044 24,104 6,529 30,633 28,260 5,306 33,566 Sources: Plan National de Développement - Congo 2012-2016 - Livre 2 Loi de Finances pour chaque année (Programmed) (LoiDeFinances_2008(2)_15Février) plus Execution Budget 2010, 2011 e 2012 (Executed) IMF (2012), Republic of Congo: 2012 Article IV Consultation - Staff Report, p.20 (2011 - estimate; 2012 - projection) MEPSA, Annuaire statistique 2010-2011. 133 Table A.8: Weight of each Ministry in the total expenditure (execution) % 2008 2009 2010 2011 2012 2013 (prog.) Invest Invest Invest Invest Invest Invest Current Total Current Total Current Total Current Total Current Total Current Total ment ment ment ment ment ment Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Budget MEPSA 9.7 1.2 6.0 10.3 3.6 7.0 10.8 2.8 6.8 10.0 1.7 4.9 7.5 3.0 4.8 9.6 4.5 6.3 METPQE 2.2 0.9 1.7 1.5 0.9 1.2 2.1 1.7 1.9 2.5 1.3 1.8 2.1 0.8 1.3 2.4 5.6 4.4 MES 3.6 0.1 2.1 4.5 0.5 2.5 4.1 0.5 2.3 4.3 0.7 2.1 3.5 0.5 1.7 4.8 0.7 2.1 Total 15.5 2.2 9.8 16.3 5.0 10.7 17.0 5.0 11.1 16.8 3.8 8.8 13.1 4.2 7.8 16.8 10.7 12.9 Education Sources: Plan National de Développement - Congo 2012-2016 - Livre 2 Loi de Finances pour chaque année (Programmed) (LoiDeFinances_2008(2)_15Février) + Execution Budget 2010, 2011 e 2012 (Executed) Ficheiro "REAL 2008_2009_2010_2011_2012" IMF (2012), Republic of Congo: 2012 Article IV Consultation - Staff Report, p.20 (2011 - estimate; 2012 - projection) Annuaire statistique 2010-2011 Table A.9: Share of each Ministry on the total education expenditure (execution) 2008 2009 2010 2011 2012 2013 (prog.) Current Investment Total Current Investment Total Current Investment Total Current Investment Total Current Investment Total Current Investment Total MEPSA 62.3 52.0 61.3 63.3 71.7 65.3 63.3 55.5 61.6 59.7 46.2 56.2 57.3 70.5 61.5 57.0 41.9 49.0 METPQE 14.2 41.7 16.9 9.3 18.9 11.5 12.5 34.7 17.3 14.8 34.6 20.1 15.7 18.7 16.7 14.6 52.0 34.5 MES 23.5 6.4 21.8 27.4 9.4 23.2 24.2 9.8 21.1 25.4 19.2 23.8 27.0 10.8 21.8 28.4 6.1 16.5 Total Education 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Sources: Plan National de Développement - Congo 2012-2016 - Livre 2 Loi de Finances pour chaque année (Programmed) (LoiDeFinances_2008(2)_15Février) + Execution Budget 2010, 2011 e 2012 (Executed) IMF (2012), Republic of Congo: 2012 Article IV Consultation - Staff Report, p.20 (2011 - estimate; 2012 - projection) MEPSA, Annuaire statistique 2010-2011 134 Table A.10: Execution rates per ministry and per category (percentage) 2008 2009 2010 2011 2012 BUDGET TOTAL Recurrent budget 52.7 84.2 105.3 107.2 121.6 Wages 99.5 100.9 95.5 97.7 100.0 Goods and Services+Commons charges 107.3 90.7 112.3 125.1 188.7 Transfers 30.4 66.2 107.2 100.2 91.9 Investment 104.7 95.2 88.4 107.3 76.6 Equipment n.a. n.a. n.a. n.a. n.a. Studies n.a. n.a. n.a. n.a. n.a. Rehabilitation n.a. n.a. n.a. n.a. n.a. Construction n.a. n.a. n.a. n.a. n.a. EDUCATION TOTAL Recurrent budget 104.9 81.8 101.7 97.2 100.9 Wages 101.6 99.0 95.6 100.6 104.4 Goods and Services 107.4 30.9 116.6 80.6 87.9 Transfers 120.4 92.0 106.2 109.1 101.7 Investment 39.0 70.8 98.2 91.6 83.4 Equipment 40.5 61.8 51.6 n.a. n.a. Studies 53.0 47.0 73.0 n.a. n.a. Rehabilitation 81.5 24.6 n.a. n.a. n.a. Construction (and Rehabilitation no METPFQE) 24.4 35.2 76.2 n.a. n.a. MEPSA Recurrent budget 100.3 81.6 102.6 90.7 97.1 Wages 98.0 100.8 96.8 95.1 104.6 Goods and Services 107.7 29.2 119.5 77.4 71.2 Transfers 89.0 95.0 98.2 80.6 66.2 Investment 29.0 28.5 98.3 95.0 84.6 Equipment 22.3 18.3 n.a. n.a. n.a. Studies 57.0 n.a. n.a. n.a. n.a. Rehabilitation n.a. n.a. n.a. n.a. n.a. Construction 11.2 30.5 n.a. n.a. n.a. METPFQE Recurrent budget 124.8 60.6 92.8 104.1 107.3 Wages 148.4 86.5 74.1 72.9 80.5 Goods and Services 109.5 19.4 114.7 91.7 143.4 Transfers 95.2 86.0 100.0 238.4 114.1 Investment 87.7 53.1 98.2 79.1 72.7 Equipment 71.6 82.3 64.4 n.a. n.a. Studies 83.3 76.4 37.1 n.a. n.a. Rehabilitation n.a. n.a. n.a. n.a. n.a. Construction and Rehabilitation 67.1 65.4 90.5 n.a. n.a. MES Recurrent budget 107.5 93.5 104.5 111.7 106.1 Wages 95.7 98.1 102.7 142.5 116.1 Goods and Services 100.6 69.0 100.0 78.9 77.9 Transfers 131.9 93.1 108.0 82.2 98.8 Investment 21.4 45.2 97.9 114.3 99.4 Equipment 16.7 39.8 3.2 n.a. n.a. Studies n.a. 56.0 n.a. n.a. n.a. Rehabilitation 81.5 24.6 n.a. n.a. n.a. Construction 33.0 14.3 100.0 n.a. n.a. Sources: Plan National de Développement - Congo 2012-2016 - Livre Loi de Finances pour chaque année Loi de Finances pour chaque année e Annuaire statistique 2010-2011 135 Table A.11: External financial support US$ 2008 2009 2010 2011 2012 2013 World Bank On-budget 19,000,000 15,000,000 10,000,000 Off-budget African Development Bank On-budget 23,348,716 Off-budget World Food Program On-budget 3,889,807 8,760,717 9,894,570 0 Off-budget UNICEF On-budget 1,721,493 2,861,061 2,449,429 1,104,445 5,037,000 Off-budget UNESCO On-budget 482,330 411,053 249252 Off-budget French cooperation (AFD) On-budget 7,871,484 Off-budget TOTAL 5,611,300 22,343,391 11,210,146 1,515,498 29,931,570 33,597,968 On-budget Off-budget NB: 1. World Bank, 19 Millions de Dollars de 2004 à 2009 (appui au PRAEBASE 1) 15 Millions de Dollars de 2009 à 2013 (appui au PRAEBASE 2) 10 Millions de Dollars de 20013 à 2018 (Appui au Projet sur le Développement des compétences professionnelles pour l'employabilité) 2. African Development Bank, 11 674 358 000 FCFA en appui au Projet d'appui à la réinsertion socio-économique des groupes défavorisés (PARSEGD) de 2007 à 2013 3.French Cooperation (via AFD), 6 Millions d'Euros au Projet d'appuià la refondation du système d'éducation et de formation (PARSEF) de 2007 à 2014 3. Pour les montants en FCA, le taux de change en dollars utilisé est de 1US$=500FCFA (cas de African Development Bank ) et 1€=655,957FCFA 4. Certains financements sont sur 1, deux ans et Plus (cf. les différentes couleurs des cases) 136 Table A.12: Input indicators by level of education (2011/2012) Teachers/ Ratio Administrative Pupils/ Class Pupil/teacher staff PRESCHOOL Private 25.9 2.6 32.0 Public 24.7 0.8 35.9 Total 25.1 1.1 34.5 PRIMARY Private 29.2 1.5 30.7 Public 59.6 1.5 58.0 Total 45.0 1.5 45.5 COLLEGE Private 10.9 2.5 20.8 Public 35.6 1.7 49.1 Total 20.1 2.1 33.5 LYCEE Private 1.3 6.0 7.4 Public 8.5 3.3 25.4 Total 5.9 3.9 21.3 Source: MEPSA, Annuaire statistique, 2011/2012. FigureA.3: Distribution of employment per education level, 2005 – 2011 Farm None-farm Wage 100% 5 13 9 13 90% 19 20 32 26 32 80% 19 34 34 52 32 55 54 70% 42 74 72 60% 46 38 78 38 31 50% 40 40 40% 30% 31 32 34 20% 14 21 10% 15 0% Incomplete primary Incomplete primary TVET TVET No education lower secondary Higher education lower secondary Higher education Completed primary Total No education Completed primary Total Upper secondary Upper secondary 2005 2011 Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data 137 Table A.13: Logistic regression of determinant of schooling 2005 2011 Primary Secondary Primary Secondary Student Sex (female) 0.992 1.503 1.164 1.315 (0.05) (3.46)*** (1.38) (2.64)*** Age in years 0.616 1.631 0.764 1.427 (8.03)*** (17.10)*** (7.16)*** (12.74)*** Areas of Res. (Rural) 0.532 0.411 1.730 1.059 (1.82)* (3.53)*** (3.61)*** (0.49) Consumption quintile (Ref. Poorest) Poor 0.662 0.704 0.879 0.740 (1.83)* (2.16)** (0.84) (2.00)** Middle 0.836 0.558 0.551 0.693 (0.70) (3.79)*** (2.69)*** (1.85)* Rich 0.604 0.499 0.854 0.363 (2.03)** (4.10)*** (0.65) (4.55)*** Richest 0.277 0.401 0.286 0.213 (4.14)*** (4.58)*** (3.14)*** (5.37)*** Head education (Ref. no education) Incomplete primary 0.876 0.997 0.860 0.935 (0.54) (0.01) (0.78) (0.35) Completed primary 0.618 0.527 0.586 0.567 (2.03)** (4.27)*** (3.53)*** (3.72)*** Secondary plus 0.346 0.238 0.549 0.334 (3.78)*** (7.43)*** (2.66)*** (6.01)*** Household background Head sex(female) 0.961 0.872 1.252 0.831 (0.12) (0.61) (0.84) (1.15) Head marital status(married) 0.999 0.952 0.967 1.029 (0.01) (0.94) (0.61) (0.73) Head age(in years) 0.993 0.989 0.996 0.986 (1.04) (2.17)** (0.66) (3.04)*** Household size 0.980 0.980 0.975 0.866 (0.70) (1.16) (0.71) (4.25)*** F 15.891 25.869 10.539 19.618 N 3,941 4,472 7,600 6,461 138 Figure A.4: Higher education benefits Higher Education Higher education 2005 Perfect equality Consumption 2005 1.0 1.0 1.0 Consumption 0.9 0.8 0.8 0.7 0.6 0.6 0.6 0.5 0.4 0.4 0.4 0.3 0.3 0.2 0.1 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Source: Authors estimations calculated from ECOM 2005 and ECOM 2011 data. 139 Table A.14: Employment and education indicators of the autochthone population (2007) Ensemble de Populations Indicator la population autochtones Congolaise Taux spécifique de scolarisation au primaire et au secondaire (6 – 16 ans) Sexe masculin 24.7% 80.7% Sexe féminin 20.7% 79.3% Les deux sexes 22.7% 80.0% Taux brut de scolarisation (au primaire) Sexe masculin 76.9% 117.0% Sexe féminin 59.0% 113.7% Les deux sexes 67.9% 115.3% Taux net de scolarisation au primaire (6 – 11 ans) Sexe masculin 47.8% 82.7% Sexe féminin 40.2% 80.0% Les deux sexes 44.0% 81.3% Taux spécifique épuré de scolarisation au primaire (6 – 14 ans) Sexe masculin 26.0% Sexe féminin 21.1% Les deux sexes 23.6% 70.0% Taux spécifique épuré de scolarisation au primaire (6 – 16 ans) Sexe masculin 19.4% Sexe féminin 18.0% Les deux sexes 18.7% 61.1% Taux brut d’activité Sexe masculin 52.8% Sexe féminin 46.4% Les deux sexes 49.5% Taux spécifique d’activité Sexe masculin 76.1% Sexe féminin 64.7% Les deux sexes 70.1% Taux d’emploi Sexe masculin 91.8% Sexe féminin 96.5% Les deux sexes 94.1% Source: CNSEE, Census of 2007. 140 Annex A.2 1. Data used in the preparation of the chapter and limitations of the analysis The preparation of the chapter made use of data from two household surveys - Enquetes Congolaise Auprès de Ménages (ECOM) from 2005 and 2011, the latest Congo Demographic and Health Survey (DHS) (2011), administrative data from the various ministries in charge of education (MEPSA, METPFQE and MES and data from the Ministère de l’Economie, des Finances, du Plan, de l’Integration et du Portefeuille Publique. Administrative data was very limited due to the existing limitations of the education information and management system in Congo. As such, it was not possible to carry a detailed analysis on human resources including looking at the percentage of bénévoles still in the system, teaching qualifications and detailed teacher distribution. The data did not allow for a detailed analysis on minorities either. Thus, the chapter relied significantly on the surveys data to produce as much information as possible. 2. Methodologies used in the analysis Grade Repetition and Drop-Out Costing Methodology. The estimation of loss associated with grade repetition was based on: (i) the direct cost of schooling, and (ii) the discounted value of the forgone opportunity costs of expected earnings. The direct cost of schooling was generated using the total number of children who repeated a grade by frequency of repetition, and was based on annual public and private unit costs per student. The discounted value of the forgone opportunity cost of expected earnings was estimated based on wage employment earnings, which took into account both the age of labor market entry as well as the associated unemployment rate. The opportunity cost of children who dropped out of school was calculated using the number of dropouts by level of education alongside calculated public and private unit costs. Earnings of individuals were estimated by level of education, and foregone income was computed by analyzing the earning difference between those completing levels of education and those who dropped out before completion. To account for cost differences of completing the level and dropping out, actual costs were estimated based on the average years of schooling by level for dropouts and the full cost of completion of the level. 141 Determinants of Out-of-School In addition to descriptive statistics, multivariate logit regression analysis was used to explore how different factors affect different age groups. The logit regression model followed the standard form: where The dependent variable, Y, was out-of-school, with 1 being coded if the child was not in school, α is a constant term, X is a vector of relevant household and individual characteristics, and is the error term. The included individual characteristics were gender, orphan status (defined as 1 if the child lived in a household without either parent, and 0 otherwise), and urban residence. The household level variables included were the consumption quintile, the education of the household head, the sex of the household head, an indicator variable for the household head not being married (coded 1 if the head was not married and 0 otherwise), household size, and the distance of the household to primary education. The results are presented as an odds ratio. Benefit Incidence Analysis Benefit incidence analysis (BIA) illustrates how public expenditure on services is distributed among population sub-groups, utilizing both the service provision costs and participation or usage rates of a specific service (Heltberg, Simler, and Tarp 2003). Benefit incidence studies are particularly useful in determining the extent to which public spending on social sectors - for the present chapter, education - benefits the poorest strata and therefore creates a well-targeted instrument for poverty reduction.55 BIA can likewise analyze expenditure by different groups or regional locations, though this analysis requires greater disaggregation in spending data which was not available for this analysis. This chapter has been therefore limited to the income group (denoted by expenditure quintile). Benefit incidence analysis requires three elements: household-level survey data which gathers (i) information from which to construct a proper welfare indicator (i.e. per capita household consumption expenditures, appropriately adjusted) and (ii) utilization of or participation in the public service of interest (enrollment in school), as well as administrative or budget data that provides (iii) unit costs to the government for the provision of those same services (e.g. the cost of one year of schooling per student). SLIHS 2011 is an adequate instrument for which to conduct a BIA with as it gathers appropriate information on both enrollment figures as well as consumption measures for constructing accurate welfare indicators. Welfare, in this case, is measured by aggregate household consumption over the last twelve months, after incorporating food consumption, non-food consumption, housing, and benefits derived from durable goods. The unit costs of education is derived from figures for public spending on education reported by 55 The concept of benefit incidence analysis (BIA) originally pioneered by studies by Gillespie on Canada 1965, and extended to developing countries context by Meerman (1979) on Columbia, and Seloswski (1979) on Malaysia and in its modern stage by Need (1995), Selden and Wasylenko (1992), Sahn and Yonger (1999) on Africa, Demery (2000). 142 Ministry of Finance for Public Spending on Education. By utilizing government expenditure sources in addition to household expenditure on education, a more accurate unit cost can be calculated. Individuals (or households) must first be ranked by their measure of welfare according to the household survey, and then aggregated into population groups in order to compare how the subsidy itself is distributed across these groups. These groups are typically quintiles or deciles. This analysis utilizes expenditure quintiles, in which the first quintile holds the poorest 20 percent of the population, and so on. Next, using the data provided in the household survey, the total number of individuals who participated in or used the publicly provided service in question (those who were enrolled in school) must be identified. Each user (or household) is then be multiplied by the unit cost of service provision and finally, these beneficiaries are aggregated into their appropriate population groups (consumption quintiles). It is the distribution of this in-kind transfer of the population that constitutes a benefit incidence analysis. In the case of education, the BIA model at hand can be expressed as: 4 4 E Si X j   Eij   ij Si i 1 Ei i 1 Ei where Xj is the value of the total education subsidy imputed to consumption quintile j. Eij represents the number of school enrollments of consumption quintile j at education level i, and Ei the total number of enrollments (across all consumption quintile) at that level. Si is government spending on education level i and i (=1,..,4) denotes the level of education (primary, lower secondary, upper secondary, and tertiary). Note that Si/Ei is the unit subsidy of providing a school place at level I (Demery 2000). The resulting profile illustrates the distribution of public spending on education that is allocated to each welfare group (expenditure quintile), or the “benefit incidence”. Concentration curves can then be plotted that show the cumulative distribution of these benefits across households, and can be compared to the cumulative distribution of total consumption (what is typically referred to as the Lorenz curve). The Lorenz curve is a graphical interpretation of the cumulative distribution of income on the vertical axis against the cumulative distribution of population on the horizontal axis. The progressivity of spending is pro-poor if the poor receive more of the service’s benefits than the non-poor, as well as a share greater than their share of the population; graphically this line appears above the diagonal line as this is the line indicating that each quintile in the distribution is receiving the same share, in this case, 20 percent of spending. Pro-poor spending is an indication of the successful targeting of public service benefits towards poorer households (Heltberg, Simler, and Tarp 2003). “Not-pro-poor but progressive” refers to if the non-poor receive more than the poor, but the poor still receive a share larger than their share of consumption; graphically this line appears below the diagonal but above the Lorenz. “Not-pro- poor and regressive” occurs if the non-poor receive more than the poor, and the share of the poor 143 is less than their share of consumption; graphically this line appears below the diagonal and below the Lorenz. When determining enrollment as an element of BIA, its distribution can be interpreted in one of two ways: (1) net enrollment (the share of children of school-age groups attending the corresponding school level) or (2) gross enrollment (the share of all children regardless of their age who are attending a specific school level). The differences in these two can add depth to further interpretations of the benefit incidence analysis, particularly in the case of Congo where overages and older children still enrolled in primary school contribute to differing enrollment rates. 144 Annex B.1 Congo: Child vaccinations BCG and DTP, 2005 and 2011-2012 (percent) Figure B.1: Child vaccinations BCG and DTP, 2005 and 2011-2012 (percentage) 100 93.9 90.0 90 80 71.9 68.4 70 60 Percent 50 40 30 20 10 0 Total Quintile Quintile Quintile Quintile Quintile Total Quintile Quintile Quintile Quintile Quintile 1 2 3 4 5 1 2 3 4 5 (poorest) (richest) (poorest) (richest) BCG DTP (3) EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. Congo: Child vaccinations Polio and Measles, 2005 and 2011-2012 (percent) Figure B.2: Child vaccinations polio and measles, 2005 and 2011-2012 (percentage) 100 90 80 74.9 69.1 70 66.2 60 57.2 Percent 50 40 30 20 10 0 Total Quintile 1Quintile 2Quintile 3Quintile 4Quintile 5 Total Quintile 1Quintile 2Quintile 3Quintile 4Quintile 5 (poorest) (richest) (poorest) (richest) Polio (3) Measles EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. 145 Congo: Child vaccinations Yellow Fever 2005 and 2011-2012 yellow fever 2005 and 2011-2012 (percentage) Figure B.3: Child vaccinations(Percent) 70 60 54.5 50 40 Percent 31.8 30 20 10 0 Total Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (poorest) (richest) Yellow fever EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. Figure B.4: Congo: Knowledge Knowledge of of contraceptive contraceptive methods, methods, women women in union, corrently in 2005 and 2011- union, 2012 (percentage) 2005 and 2011-2012 (percent) 100 99.3 98.6 98.6 98 96.3 96 Percent 94 92 90 88 Total Quintile Quintile Quintile Quintile Quintile Total Quintile Quintile Quintile Quintile Quintile 1 2 3 4 5 1 2 3 4 5 (poorest) (richest) (poorest) (richest) Any method Modern method EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. 146 Figure B.5: Current use of contraception by women in union, 2005 and 2011-2012 Congo: Current use of contraception by women in union, 2005 and (percentage) 2011-2012 (percent) 40 35 31.6 30 24.7 25 Percent 20.0 20 15 12.7 10 5 0 Total Quintile 1 Quintile 2Quintile 3Quintile 4Quintile 5 Total Quintile 1 Quintile 2Quintile 3Quintile 4Quintile 5 (poorest) (richest) (poorest) (richest) Modern method Traditional method EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. Congo: Median number of months since the previous delivery, number Figure B.6: Median2005 and of since the previous delivery, 2005 and 2011- months (percent) 2011-2012 2012 (percentage) 50 45.20 46.00 44.90 41.00 38.80 39.40 39.80 39.30 40 37.50 37.90 36.70 35.00 Percent 30 20 10 0 Total Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (poorest) (richest) EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. 147 Figure Congo:B.7: Percentage Percentage of pregnant of pregnant women women whowho received received pre-natal prenatal carecare by qualified by health qualified health personnel, personnel, 2005 and 2005 2011-2012 and 2011-2012 (Percent) (percentage) 100 97.10 98.40 99.20 94.80 94.80 92.60 88.20 89.80 89.90 90 83.50 84.40 80 76.90 70 60 Percent 50 40 30 20 10 0 Total Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (poorest) (richest) EDCS 2005 EDCS 2011-2012 Source: Authors calculations with data from EDCS 2005 and EDCS 2011-2012. Figure B.8: CMR and health expenditure in selected countries: deviations from CMR and health expenditure in selected countries: deviations from estimates estimates based based on per on per capita capita income incomeUS$) (PPP-adjusted (PPP-adjusted US$)rate and schooling (literacy and schooling (literacy rate in percentage), 2012 in percent), 2012 120 Worse outcome, Worse outcome, lower expenditure higher expenditure 100 Deviation from predicted infant mortality under 5 rate 80 60 ROC - WB 40 20 ROC - DHS 0 -20 -40 Better outcome, Better outcome, lower expenditure higher expenditure -60 -9 -7 -5 -3 -1 1 3 5 7 9 Deviations from predicted percentage of GDP spent on health Source: Authors calculations with data from WBWDB. 148 Figure B.9: MMR and health expenditure in selected countries: deviations from estimates based on Maternal rate andincome per capita mortality (PPP-adjusted health expenditure US$) and schooling (literacy in selected countries: deviations from estimates based on per capita income (PPP-adjusted US$) rate in percentage), and 2012 schooling (literacy rate in percent), 2012 1,200 Worse outcome, Worse outcome, lower expenditure higher expenditure 1,000 Deviation from predicted maternal mortality rate 800 600 400 200 ROC WB y DHS 0 -200 -400 Better outcome, Better outcome, lower expenditure higher expenditure -600 -9 -7 -5 -3 -1 1 3 5 7 9 Deviations from predicted percentage of GDP spent on health Source: Authors calculations with data from WBWDB. Life expectancy rate and health expenditure in selected Figure B.10: LEB and health expenditure in selected countries: deviations from countries: deviations from estimates based on per capita estimates based income on per capita (PPP-adjusted income US$) (PPP-adjusted and schooling US$) (literacy rate in and schooling (literacy rate in percentage), 2012 percent), 2012 25 Better outcome, Better outcome, lower expenditure higher expenditure 20 15 Deviation from predicted life expectancy rate 10 5 0 -5 ROC WB -10 -15 -20 Worse outcome, Worse outcome, lower expenditure higher expenditure -25 -9 -7 -5 -3 -1 1 3 5 7 9 Deviations from predicted percentage of GDP spent on health Source: Authors calculations with data from WBWDB. 149 Table B.1: Incidence and intensity of catastrophic health payments Threshold budget share 5% 10% 15% 25% 30% 40% Headcount (H) Lowest quintile 6.9 2.0 0.3 0.0 0.0 0.0 2 6.7 1.8 1.0 0.1 0.0 0.0 3 7.1 1.4 0.1 0.0 0.0 0.0 4 5.0 0.6 0.1 0.0 0.0 0.0 Highest quintile 6.3 0.2 0.0 0.0 0.0 0.0 Total 6.4 1.2 0.3 0.0 0.0 0.0 Mean positive overshoot (MPO) Quintile 2 4.4 5.6 4.1 17.2 33.3 23.3 Concentration -0.044 -0.304 -0.383 -0.451 -0.523 -0.521 index, C_E Source: Authors calculations based on ECOM 2011. 150 Table B.2: Number of hospital deds in Base Hospitals, by départment, 2012 Number of hospital beds Obstetrics & Départment Medicine Pediatrics Surgery Gynecology Total Kouiloua 0 0 0 0 0 Niari 15 26 24 19 84 Lékoumou 15 16 14 12 57 Bouenza 116 83 90 84 373 Pool 140 17 40 34 231 Plateaux 29 35 21 37 122 Cuvette 80 58 51 91 280 Cuvette-Ouest 48 41 35 41 165 Sangha 40 20 6 36 102 Likouala 65 60 65 63 253 Brazzaville 97 69 60 34 260 Pointe-Noire 23 35 16 37 111 Country 668 460 422 488 2,038 a. Kouilu has no Base Hospitals. Source: Ministry of Health and Population (2013). Table B.3: Number of beds days in Base Hospitals, by départment, 2012 Total number of bed days Obstetrics & Départment Medicine Pediatrics Surgery Gynecology Total Kouilou Niari 3,229 4,329 5,297 3,771 16,626 Lékoumou 1,492 1,668 1,718 2,605 7,483 Bouenza 10,181 13,091 6,414 3,829 33,515 Pool 4,670 5,689 2,647 412 13,418 Plateaux 3,870 4,430 1,831 1,959 12,090 Cuvette 8,248 10,174 4,725 8,998 32,145 Cuvette-Ouest 3,333 2,616 2,530 1,350 9,829 Sangha 1,570 2,897 839 3,638 8,944 Likouala 1,483 3,434 1,336 589 6,842 Brazzaville 9,891 6,528 2,416 6,468 25,303 Pointe-Noire 2,880 17,256 4,049 15,638 39,823 Country 50,847 72,112 33,802 49,257 206,018 Source: Ministry of Health and Population (2013). 151 Table B.4: Base Hospitals: utilization statistics, 2012 Medicine Pediatrics Surgery Obstetrics & Gynecology Total Hospital Bed Averag Discharge Hospital Bed Averag Discharge Hospital Bed Averag Discharge Hospital Bed Averag Discharge Hospital Bed Averag Discharge -izations days e length s -izations days e length s -izations days e length s -izations days e length s -izations days e length s Department of stay of stay of stay of stay of stay Kouilou - - - - Niari 586 3,229 5.5 568 986 4,329 4.4 972 232 5,297 22.8 220 1,038 3,771 3.6 1,015 2,842 16,626 14 2,775 Lékoumou 492 1,492 3.0 - 1,197 1,668 1.4 - 284 1,718 6.0 - 337 2,605 7.7 - 2,310 7,483 12 - 10,18 13,09 Bouenza 1,986 1 5.1 1,906 4,193 1 3.1 3,430 1,090 6,414 5.9 991 1,254 3,829 3.1 1,116 8,523 33,515 11 6,452 Pool 832 4,670 5.6 687 1,103 5,689 5.2 717 866 2,647 3.1 787 120 412 3.4 120 2,921 13,418 14 1,524 Plateaux 819 3,870 4.7 884 1,051 4,430 4.2 1,594 280 1,831 6.5 304 622 1,959 3.1 932 1,953 12,090 12 3,410 10,17 Cuvette 2,132 8,248 3.9 1,651 3,121 4 3.3 2,196 1,012 4,725 4.7 1,337 4,264 8,998 2.1 2,488 10,529 32,145 9 6,335 Cuvette- Ouest 671 3,333 5.0 654 2,616 542 4.8 519 440 2,530 5.8 435 248 1,350 5.4 144 1,901 9,829 15 1,317 Sangha 265 1,570 5.9 242 2,897 890 3.3 868 92 839 9.1 90 1,220 3,638 3.0 2,467 8,944 12 1,110 Likouala 293 1,483 5.1 267 3,434 802 4.3 802 146 1,336 9.2 124 118 589 5.0 1,359 6,842 14 1,069 Brazzaville 2,123 9,891 4.7 1,279 6,528 2,021 3.2 1,514 492 2,416 4.9 279 1,184 6,468 5.5 5,820 25,303 13 2,793 17,25 15,63 Pointe-Noire 612 2,880 4.7 604 3,733 6 4.6 3,698 525 4,049 7.7 501 4,441 8 3.5 9,311 39,823 13 4,302 50,84 72,11 33,80 49,25 206,01 Country 9,992 7 4.7 8,742 19,639 2 3.7 16,310 5,459 2 6.2 5,068 14,846 7 3.3 13,150 49,936 8 140 31,087 Source: Ministry of Health and Population (2013). Table B.5: National Hospitals: bed utilization statistics National Hospital Hospitalizations Bed days Average length of Beds Bed utilization stay rate A. Sice 13,467 18,946 1.4 NA NA Centre Hospitalier Universitaire (CHU) 29,638 169,839 5.7 723 64% Owando 3,153 14,066 4.5 145 27% Dolisie 7,877 28,210 3.6 NA NA HCA 7,734 NA NA 300 NA Loandjili 7,596 27,928 3.7 NA NA Total 69,465 258,989 3.7 1,168 61% NA. Not available Source: Ministry of Health and Population (2013). 152 Table B.6: Health care subsidies, constant unit subsidy assumption Subsidy for visits to Subsidy for visits to Total hospital ambulatory center subsidies Mean subsidy Lowest quintile 5826.81 2465.76 8292.57 2 6599.38 1836.75 8436.13 3 5722.01 1429.01 7151.02 4 6375.09 1279.64 7654.73 Highest quintile 6806.58 1406.67 8213.25 Total 6266.23 1683.55 7949.77 Shares Lowest quintile 18.6 29.3 20.9 2 21.1 21.8 21.2 3 18.3 17.0 18.0 4 20.3 15.2 19.2 Highest quintile 21.8 16.7 20.7 Total 100.0 100.0 100.0 Share in the total subsidy 78.8 21.2 100.0 Concentration Index 0.0214 -0.1428 -0.0135 Source: Authors calculation based on ECOM 2011 and MSP (2013) 153 Annex B.2 1. Data used in the preparation of the chapter and limitations of the analysis The preparation of the chapter made use of data from two household surveys - Enquetes Congolaise Auprès de Ménages (ECOM) from 2005 and 2011, the two most recent Congo Demographic and Health Survey (DHS) - known as EDCS 2005 and EDSC 2011-12, administrative data from the Ministère de la Santé et de la Population (MSP) and data from the Ministère de l’Economie, des Finances, du Plan, de l’Integration et du Portefeuille Publique. Administrative data was very limited due to the existing limitations of the health information and management system in Congo. As such, it was not possible to carry a detailed analysis on human resources, budget items changed over time what made analyzing trends in budget execution by budget item a challenging exercise and regional expenditure analysis were very limited. The data did not allow for a detailed analysis on minorities either. Thus, the chapter relied significantly on the surveys data to produce as much information as possible. 2. Methodologies used in the analysis Catastrophic Payments for Health Care56 To assess the impact of the disruptive effect of health Out of Pocket Spending (OOPS) in household’s living standards, a popular approach has been to define medical spending as “catastrophic” if it exceeds some fraction of household income or total expenditure in a given period, usually one year. The idea is that spending a large fraction of the household budget on health care must be at the expense of the consumption of other goods and services. This opportunity cost may be incurred in the short term if health care is financed by cutting back on current consumption or in the long term if it is financed through savings, the sale of assets, or credit. The two key variables underlying the approach are total household out-of-pocket payments for health care and a measure of household resources. Income, expenditure, or consumption could be used for the latter. In this particular case, we have used aggregate household consumption and household out-of-pocket payments for health care obtained from SLIHS 2011-12. A potential problem is that this budget share may be low for poor households in low-income countries. The severity of the budget constraint means that most resources are absorbed by items essential to sustenance, such as food, leaving little to spend on health care. A partial solution is to define 56 Follows the methods outlined in Analyzing Health Equity Using Household Survey Data: A Guide to Techniques and Their Implementation (Washington DC, World Bank, 2008) by Owen O'Donnell, Eddy van Doorslaer, Adam Wagstaff and Magnus Lindelow. 154 catastrophic payments not with respect to the health payments budget share but with respect to health payments as a share of expenditure net of spending on basic necessities. The latter has been referred to as “nondiscretionary expenditure”. A common approach is to use household expenditure net of food spending as an indicator of living standards. In this report we shows results for both total and non-food expenditure. If we let T be OOP payments for health care, x be total household expenditure, and f(x) be food expenditure, or nondiscretionary expenditure more generally. Then, a household is said to have incurred catastrophic payments if T/x, or T/[x-f(x)], exceeds a specified threshold, z. The value of z represents the point at which the absorption of household resources by spending on health care is considered to impose a severe disruption to living standards. That is obviously a matter of judgment and it will depend on whether the denominator is total expenditure or nondiscretionary expenditure. Spending 10 percent of total expenditure on health care might be considered catastrophic, but 10 percent of nondiscretionary expenditure probably would not. Thus, we present results considering different thresholds - from 5 to 10 percent of expenditure. Benefit Incidence Analysis57 Benefit incidence analysis (BIA) illustrates how public expenditure on services is distributed among population sub-groups, utilizing both the service provision costs and participation or usage rates of a specific service (Heltberg, Simler, and Tarp 2003). Benefit incidence studies are particularly useful in determining the extent to which public spending on social sectors - for the present chapter, health - benefits the poorest strata and therefore creates a well-targeted instrument for poverty reduction.58 BIA can likewise analyze expenditure by different groups or regional locations, though this analysis requires greater disaggregation in spending data, which was not available for this analysis. This chapter has been therefore limited to the income group (denoted by expenditure quintile). Benefit incidence analysis requires three elements: household-level survey data which gathers (i) information from which to construct a proper welfare indicator (i.e. per capita household consumption expenditures, appropriately adjusted) and (ii) utilization of or participation in the public service of interest (visits to a certain type of health facility), as well as administrative or budget data that provides (iii) unit costs to the government for the provision of those same services (e.g. the total amount of government spent on each type of provider). SLIHS 2011-12 is an adequate instrument for which to conduct a BIA with as it gathers appropriate information on 57 Follows the methods outlined in Analyzing Health Equity Using Household Survey Data: A Guide to Techniques and Their Implementation (Washington DC, World Bank, 2008) by Owen O'Donnell, Eddy van Doorslaer, Adam Wagstaff and Magnus Lindelow. 58 The concept of benefit incidence analysis (BIA) originally pioneered by studies by Gillespie on Canada 1965, and extended to developing countries context by Meerman (1979) on Columbia, and Seloswski (1979) on Malaysia and in its modern stage by Need (1995), Selden and Wasylenko (1992), Sahn and Yonger (1999) on Africa, Demery (2000). 155 both health facility use figures as well as consumption measures for constructing accurate welfare indicators. Welfare, in this case, is measured by aggregate household consumption over the last twelve months, after incorporating food consumption, non-food consumption, housing, and benefits derived from durable goods. The unit costs of health are derived from figures for public spending on health reported in the National Health Accounts study completed by the MSP in 2013. Because accounts are not sufficiently detailed to allow net public expenditure to be identified by region and facility, then all units of a given service must be weighted by the same unit subsidy estimated, thus this particular analysis assumes constant unit subsidies. Moreover, user fees were not considered since there are not official numbers for these. Individuals (or households) must first be ranked by their measure of welfare according to the household survey, and then aggregated into population groups in order to compare how the subsidy itself is distributed across these groups. These groups are typically quintiles or deciles. This analysis utilizes expenditure quintiles, in which the first quintile holds the poorest 20 percent of the population, and so on. Next, using the data provided in the household survey, the total number of individuals who participated in or used the publicly provided service in question (those who visited health providers during the year in analysis) must be identified. Each user (or household) is then be multiplied by the unit cost of service provision and finally, these beneficiaries are aggregated into their appropriate population groups (consumption quintiles). It is the distribution of this in- kind transfer of the population that constitutes a benefit incidence analysis. The resulting profile illustrates the distribution of public spending on education that is allocated to each welfare group (expenditure quintile), or the “benefit incidence”. Concentration curves can then be plotted that show the cumulative distribution of these benefits across households. The progressivity of spending is pro-poor if the poor receive more of the service’s benefits than the non-poor, as well as a share greater than their share of the population; graphically this line appears above the diagonal line as this is the line indicating that each quintile in the distribution is receiving the same share, in this case, 20 percent of spending. 156 References CNSEE (2005). Congo Enquete Démographique et de Santé 2005. Brazzaville, Centre National de la Statistique et des Études Économiques (CNSEE). CNSEE (2012). Extrait du profil de la pauvreté au Congo en 2011. CNSEE and ICF International (2012). Enquête Démographique et de Santé du Congo (EDSC-II) 2011- 2012. Brazzaville. Cadre de dépenses de moyen terme (CDMT) central (2009). (Excel file) CONFEMEN/MEPSA (2009). Rapport PASEC Congo-Brazzaville 2009 - L'enseignement primaire au Congo: à la recherche de la qualité et de l'équité. Conseil des Ministres (2012a), Compte-rendu du Conseil des Ministres du 23 juillet 2012. Decree nr. 2010-41, 28 January 2010. Décret nº 2009-233, du 14 Août 2009 Décret 2008-886, 31 décembre 2008. Government of Congo (2010). Rapport national des progres vers l’atteinte des Objectifs du millenaire pour le developpement. Brazzaville. IDA/IMF (2012). Staff Advisory Note on the PRSP 2012-2016. IMF (2010a). Status report on the implementation of the PRSP 2008-2011. IMF (2010b). Rapport sur la réforme de la chaîne de la dépense publique Congo Brazza. IMF (2013). Aide memoire Mission March 16. Institut national de la statistique (2013). Indice harmonisé des prix a la consommation des ménages (IHPC, Base 100: 2005). International Finance Corporation and The World Bank (2011). Doing Business 2011. Making a Difference for Entrepreneurs. Washington, DC. International Monetary Fund (2012). Republic of Congo: 2012 Article IV Consultation—Staff Report; Public Information Notice on the Executive Board Discussion; and Statement by the Executive Director for the Republic of Congo. IMF Country Report International Monetary Fund. No. 12/283. International Monetary Fund (2013). Republic of Congo, 2013 Article IV Consultation. IMF Country Report No. 13/282. Washington, DC, International Monetary Fund. International Monetary Fund (2013). Republic of Congo. 2013 Article IV Consultation. Loi no 5-2011, 25 février 2011. Loi 20/2012, 3 septembre 2012. Loi 12/2009, 29 décembre 2009. Loi de finances 2/2008, 15 février 2008. Mathis, J. (2006) Mesure de la Performance de la Gestion des Finances Publiques en République du Congo selon la Méthodologie PEFA. MEPSA (2008). Le Développement de l'Éducation - Rapport National de la République du Congo, Brazzaville, Septembre 2008. 157 MEPSA (2009). Stratégie Nationale d'Éducation des Populations Autochtones du Congo, PRAEBASE. MEPSA (2010), Stratégie sectorielle de l'éducation. Ministère de l'Économie, du Plan, de l'Aménagement du Territoire et de l'Intégration (2012a). Deuxième Enquête Congolaise Auprès des Ménages pour le Suivi et l'Évaluation de la Pauvreté (ECOM 2011), Rapport d'Analyse du Volet QUIBB, février 2012. Ministère de l'Économie, du Plan, de l'Aménagement du Territoire et de l'Intégration (2012b). Deuxième Enquête Congolaise Auprès des Ménages pour le Suivi et l'Évaluation de la Pauvreté (ECOM 2011) - Extrait du Profil de la Pauvreté au Congo en 2011. Ministère de l’Économie, du Plan, de l’Aménagement du Territoire et de l’Intégration (2010c). Bilan de l’exécution en fin 2010 Du Plan d’Actions pour l’Amélioration de la gestion des Investissements Publics (PAAGIP) – 2009-2010 Et Proposition d’un second Plan d’Actions Pour l’Amélioration de la Gestion des Investissements Publics (PAAGIP II) - 2011-2013. Ministère des Finances, du Budget et du Portefeuille Public (2011). Instruction d'application des dispositions fiscales de la loi n. 20 du 29 décembre 2010, portant loi de finances pour l'année 2012. Ministère des Finances, du Budget et du Portefeuille Public (2012a). Instruction d'application des dispositions fiscales de la loi n. 36 du 29 décembre 2011, portant loi de finances pour l'année 2011. Ministère de l'Enseignement Primaire, Secondaire, chargé de l'Alphabétisation (2009). Program Budget. Ministère de l'Enseignement Technique, Professionnel, de la Formation Qualifiante et de l'Emploi (Novembre 2009). Program Budget. Ministère de la Santé et de la Population (2013). Comptes Nationaux de la Santé 2009-2010 Rapport Final. Brazzaville. Ministère de la Santé et de la Population (2013). Comptes nationaux de la sante (CNS) 2009-2010 Version Finale. Ministry of Health and Population (2013). Annuaire des statistiques sanitaires du Congo 2012. O'Donnell, O., van Doorslaer, E., Wagstaff. A. and Lindelow, M. (2008). Analyzing Health Equity Using Household Survey Data: A Guide to Techniques and Their Implementation. Washington DC, World Bank, 2008. Oulai, D, and Farba, D. (2001). La gratuité de l’éducation au Congo. Etat des lieux des frais liés à l’éducation, analyse des conséquences de la gratuité (Rapport Provisoire), UNICEF/UNESCO, 8 mai 2011. PDSS, P. d. D. d. S. D. S. (2010). Etude Sur Les Etablissements Sanitaires Du Congo. 2010. Results for Development Institute and Health Research for Action (2011). Étude sur le secteur privé de la santé en République du Congo. Volume I. Présidence de la République (2008). Lettre de cadrage budgétaire pour 2009. Présidence de la République (2009). Lettre de cadrage budgétaire pour 2010. Programme d'actions prioritaire et cadre de dépenses à moyen terme 2010-2012 (Université). Republic of Congo (2012). National Development Plan 2012-2016 - Book 1. République du Congo (2010). Rapport National des Progrès vers l'Atteinte des Objectifs du Millénaire pour le Développement, avril 2010. 158 République du Congo (2010). Le RGPH - 2007 en quelques chiffres. TOFE 1997-2012. UNFPA. (2011). Protection des populations autochtones de la République du Congo. La faible utilisation des services de santé de la reproduction menace leur existence. UNESCO/IBE (2010). World Data on Education - Congo, VII Edition 2010/11. US Department of State (2013). Investment Climate Statement - Congo 2012. van Doorslaer, E., O'Donnell, O., Rannan-Eliya, R. P., Somanathan, A., Adhikari, S. R., Garg, C. C., Harbianto, D., Herrin, A. N., Huq, M. N., Ibragimova, S., Karan, A., Lee, T.-J., Leung, G. M., Lu, J.-F. R., Ng, C. W., Pande, B. R., Racelis, R., Tao, S., Tin, K., Tisayaticom, K., Trisnantoro, L., Vasavid, C. and Zhao, Y. (2007), Catastrophic payments for health care in Asia. Health Econ., 16: 1159–1184. doi: 10.1002/hec.1209 Wang, L. (2002). "Determinants of Child Mortality in Low-Income Countries: Empirical Findings from Demographic and Health Surveys." Health Policy 65(3): 277-299. World Bank (Circa 2006). Preparing PERs for Human Development. Specific Guidance for Health. WB (2010a). Country Partnership Strategy, 2010-2012. WB (2010b). Revue des Dépenses Publiques 2004-2009, April 2010. World Bank (2010c). Le système Éducatif Congolais - Diagnostic pour une revitalisation dans un contexte macroéconomique plus favorable, Document de travail n. 183. World Health Organization (2010). Congo. Factsheets of Health Statistics 2010, World Health Organization. 159