Report No: ACS2873 . Kingdom of Bhutan: Human Development Public Expenditure Review . March 2013 . Somil Nagpal and Susan Opper SASHD SOUTH ASIA . i Standard Disclaimer: . This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. . Copyright Statement: . The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978- 750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ii Contents Acknowledgments vii Abbreviations and Acronyms viii Executive Summary ix Matrix of Recommendations from the Public Expenditure Review xvii 1 Introduction, Objectives and Overview of HD PER 1 2 Macroeconomic Context 3 Introduction and Country Context 3 Macroeconomic indicators 3 Economic outlook 6 3 Public Sector Management and Financing 8 Governance Structure 8 Quality of Public Administration 8 Role of Private sector 9 Budget Process 9 Equity of Public Resource Use 10 Budget and Financial management 11 Financing the Health and Education Sectors 13 4 Health Expenditure in Bhutan 15 Introduction and Sectoral Context 15 Health Financing and Resource Flows in Bhutan 19 Household Expenditure on Health: Additional Analysis of Bhutan Living 22 Standards Survey 2007 Health Infrastructure and Access to Health Services: Analysis using 28 Geographic Information System (GIS) tools Costing of Health Services at Different Levels of Health Facilities 33 Experience with Piloting Alternative Provider Payment Systems- Special 36 Consultation Services (SCS) at Jigme Dorji Wangchuck National Referral Hospital (JDWNRH) Summing up 39 5 Education 40 Organized Public Education 40 Outcomes 41 Education Sector Financing 52 Preparing the 11FYP 58 6 In Conclusion: Policy Challenges and Options for Reforms 65 Bibliography 69 iii Figures 2.1 Economic growth rates in selected SAS countries, 1980-2010 4 2.2 Key fiscal/economic indicators in Bhutan, 2000-2017 4 4.1 Bhutan’s performance on key health outcomes, 1960-2010 17 4.2 Government Share of Health Expenditure versus Income per Capita 18 4.3 Infant mortality rates in selected comparators (1960-2008) 18 4.4 Life expectancy in selected comparators (1960-2008) 18 4.5 Health Indicators Relative to Income and Spending 19 4.6 Financing flows in the health sector in Bhutan (NHA 2009/10) 22 4.7 Elasticity of government health spending to GDP, 1995-2010 23 4.8 Per-capita household spending on health by districts in Bhutan: 2009-10 25 4.9 Composition of household expenditure on health in selected districts 26 4.10 GIS map showing health infrastructure, road network and population settlements 30 in Bhutan 4.11 GIS map showing magnitude of OPD visits in health facilities in Bhutan (2010) 31 4.12 GIS map showing maternal deaths in 2009 and 2010 vis-à-vis health facilities 32 and road network in Bhutan 4.13 GIS maps showing patterns of Infant Mortality, Under-5 Mortality, Stunting and 33 Rural Poverty in the districts of Bhutan 5.1 Structure of Education Sector in Bhutan 41 5.2 Historical Growth in Enrollments 44 5.3 Elasticity of Government Education Spending to Income, 1999-2008 54 5.4 Budget execution results for district and education. 2010/11 (Nu. millions) 58 5.5 Expenditure per student FY 2010/11 (current + capital, Nu. ‘000) 58 5.6 Key Targets for 11FYP and Indicative Financing Gaps 65 iv Tables 2.1 Central Government Operations (in percent of GDP) 7 2.2 Bhutan Medium Term Macroeconomic Framework 8 3.1 National budget: sector expenditures for fiscal year 2009/10 (millions of Nu.; % 14 GDP) 4.1 Heath Infrastructure in Bhutan (2001-2012) 16 4.2 Key Indicators of Health Expenditure in Bhutan (NHA 2009-10) 19 4.3 Public Spending on Health in Bhutan: 2009/10 21 4.4 Mean distribution of percentage of health related expenditure by consumption 23 quintiles and region 4.5 Mean distribution of share of health related expenditure by consumption quintiles 24 (expressed as % of total expenditure by households) 4.6 Mean distribution of share of health related expenditure by geographic region 24 (expressed as % of total expenditure by households) 4.7 Mean share of household expenses on rimdo and religious activities by 25 consumption quintiles and region (expressed as % of total expenditure by households) 4.8 Percentage distribution of people who suffered in last four weeks by consumption 26 quintiles 4.9 Percentage distribution of people who consulted with different sources of 27 healthcare services, by consumption quintiles 4.10 Percentage distribution of respondents’ reasons for not consulting any healthcare 27 providers, by consumption quintiles 4.11 Percentage distribution of women who received prenatal care by consumption 28 quintiles 4.12 Mean time required (in minutes) to access services, by consumption quintiles and 28 region 4.13 List of the facilities studied for the costing exercise in Bhutan, 2011 33 4.14 Average Unit Costs at different levels of health facilities in Bhutan, 2011 35 5.1 Key Education Indicators: Bhutan and other Regions of the World 42 5.2 Bhutan Enrollment Growth in Secondary and Higher Level Institutions (1997-2010) 43 v 5.3 Improvements in Internal Efficiency (%) of the Education System (2002-2011) 45 5.4 Percentage of secondary school age children attending secondary school or higher, 46 and percentage attending primary school, 2010 5.5 Structure of households’ non-food consumption by geographic area in Bhutan (%), 47 2007 5.6 Mean per capita monthly consumption expenditures, by household income 47 quintile, 2007 (in Ngultrum) 5.7 Average per capita Education Expenses (Nu.) by Households with Currently 48 School Going Children, across areas and sex, 2007 5.8 Education Expenditures by Level of Education, 2009/10 54 5.9 Breakdown by Level of Education (%) 55 5.10 Numbers of Students PP-Higher Education and per student costs, 2009/10 56 5.11 MOF Expenditure Data (actuals, current & capital) 2009/10 56 5.12 Bhutan compared with other South Asian Countries: Primary Education 59 Indicators, 2009/10 vi Acknowledgements This Public Expenditure Review of the Human Development sectors of Health and Education was undertaken by the World Bank in close collaboration with and with invaluable support from the Ministries of Health and Education, as also the Department of Public Accounts and the Gross National Happiness Commission of the Royal Government of Bhutan. The authors sincerely thank Dasho Nima Wangdi, Secretary, Ministry of Health, and Dasho Aum Sangay Zam, Secretary, Ministry of Education, along with their entire teams, for their continued support and insights. In particular, the team expresses its heartfelt gratitude to the PPD team in the Ministry of Health, including Mr. Kado Zangpo, and his team comprising of Mr. Tandin Dendup, Mr. Jayendra Sharma and Mr. Ugyen Wangchuck, and also to Ms. Sangay Wangmo who was associated with this task during the earlier stages of this report. The continued support of officials from the Ministry of Education, Dasho Pema Thinley, Vice Chancellor, and his team at the Royal University of Bhutan; Dr. Shivaraj Bhattarai, Dean, Royal Thimphu College; the team overseeing vocational training at the Department of Human Resources, Ministry of Labour & Human Resources; and Dr. Karma, General Manager, Projects, DHI-INFRA (Education City) as also of Mr Nim Dorji and Ms Chuni from the Department of Public Accounts (Ministry of Finance) is sincerely appreciated. The PER was a joint effort of many people representing various stakeholders who dedicated their time to this task. We are indeed grateful for their guidance and feedback during numerous discussions and presentations undertaken during this process. We also thank the joint review team of development partners for their technical advice during the presentation of initial findings from the study. We do not name all these persons here for want of space, but do certainly express our heartfelt gratitude. We sincerely thank our colleagues in the World Bank HD PER team, Ajay Tandon, Ananya Basu, Annika Kjellgren, Aphichoke Kotikula, Keiko Kubota, Lalita Moorty and Shinsaku Nomura for their valuable contributions to this document. Ajay Ram Dass and Julie-Anne M. Graitge provided efficient and indispensible administrative and logistics support to this task. The continued guidance and support of the World Bank leadership, in particular the Country Directors Nicholas Krafft and Robert Suam, and Mr. Amit Dar and Ms Julie McLaughlin, sector managers for education and health, is sincerely appreciated. Peer review comments at the concept note stage were received from Paolo Belli (Country Sector Coordinator, ECSH1), Rafael Rofman (Lead Social Protection Specialist, LCSHS), George Schieber (Consultant, EASHD), and Ajay Tandon (Sr. Economist, EASHH). The peer reviewers at the decision meeting stage were Daniel Cotlear (Lead Economist, HDNHE), Michelle Riboud (Consultant Economist) and Paolo Belli (Country Sector Coordinator, ECSH1). The task team is immensely grateful to these peer reviewers for their valuable insights and advice. In addition to the financial support extended by the Bhutan country office, the Health Results Innovation Trust Fund and the EPDF TF within the World Bank played a major role in supporting the costs for this study, which is duly acknowledged. vii Acronyms and Abbreviations AAR Annual Audit Report INTOSAI International Organization of Supreme Audit Institutions AASBB Accounting and Auditing Standards Board of Bhutan IPSAS International Public Sector Accounting Standard ABSD Accelerate Socio-economic development JDWNRH Jigme Dorji Wangchuk National Referral Hospital AFS Annual Financial Statement KII Key informant interview BBE Bhutan Board of Examinations LC Letter of Credit BCSEA Bhutan Council for School Examinations and Assessment M&E Monitoring & Evaluation BHMIS Bhutanese Health Management and Information System MDAs Ministries, Department and Agencies BHTF Bhutan Health Trust Fund MDG Millennium Development Goals BHUs Basic Health Units MMR Maternal Mortality Rate BLSS Bhutan Living Standards Survey MOF Ministry of Finance BLSS Bhutan Living Standards Survey MPs Members of Parliament BLSS Bhutan Living Standards Survey MYRB Multi Year Rolling Budget system BMIS Bhutan Multiple Indicator Survey NCDs Non-Communicable Diseases BMIS Bhutan Multiple Indicator Survey NEA National Education Assessment BPHC Bhutan Population and Housing Census NFE Non-formal Education CAD Central Accounts Department NHA National Health Accounts CPI Consumer Price Index OD Organizational Development CPS Community Primary Schools OOP Out-of-Pocket DANIDA Danish International Development Agency OPD Outpatient Department DNB Department of National Budget ORCs Outreach Centers DPA Department of Public Accounts PAC Public Accounts Committee DPC Development Policy Credit PEMS Public Expenditure Management System DRC Department of Revenue and Customs PHE Public health expenditure DSA Debt Sustainability Analysis PlaMS Planning and Management system DVED Drugs, Vaccines and Equipment Division PLC Project Letter of Credit ECCD Early Childhood Care and Development PM Prime Minister ECR Extended Classrooms PPP Public-Private Partnership Policy FGD Focus group discussions PS Primary Schools G2C Government to Citizen PSDC Private Sector Development Committee GCA Government Consolidated Account RAA Royal Audit Authority GDP Gross Deomestic Product RGoB The Royal Government of Bhutan GDP Gross domestic product RUB Royal University of Bhutan GFATM Global Fund for AIDS, Tuberculosis and Malaria SAS South Asian sub-continent GIS Geographical Information System SBDs Standard Bidding Documents GNH Gross National Happiness SCS Special Consultation Service GNHC Gross National Happiness Commission SCS Special Consultation Services HDI Human Development Index SRFPs Standard Request for Proposals HDPER Human Development Public Expenditure Review TGE Total government expenditure ICHA International Classification of Health Accounts THE Total health expenditure IMF International Monetary Fund UNICEF United Nations Children's Fund IMR Infant Mortality Rate WHO World Health Organization viii Executive summary Context of Bhutan Bhutan is situated between the Tibetan Plateau in the North and Indian plains in the south. The development philosophy in Bhutan is embedded in the concept of Gross National Happiness (GNH)1 that, as a public policy strategy, seeks to address a more meaningful purpose of development that goes beyond the fulfillment of material satisfaction. The concept is grounded in the four pillars of development; socio-economic, environment, culture, and good governance. Bhutan’s record on growth and development has made it a top performer in the South Asian region. The average annual growth rate of GDP over 1980-2010 in country was more than 7.6 percent, one of the highest in the South Asian sub-continent (SAS). Bhutan – with a GDP per capita of about US$2,000 – is now classified as a lower-middle income country. However, one of the most notable features about Bhutan’s macro economy is its lack of diversification, dependence on and exposure to external developments, and the high levels of year-to-year volatility in its economic growth. Sustained investment in the social sectors has enabled Bhutan to make remarkable progress. The country has achieved or is on track to achieve many of the Millennium Development Goals (MDGs). Continued pressures on aggregate demand, steeper rise in social sector expenditures in the future, economic constraints associated with debt sustainability, potential overheating of the economy, resource constraints from uncertain donor support, rising citizen expectations and implementation capacity constraints are some of the concerns around continued investments in the social sectors. Objectives of the PER The central issue for the Human Development Public Expenditure Review (PER) is whether Bhutan’s domestic revenue base will be sufficient as it continues the trajectory of development, and goes through rapid increase in demand for education and health services due in part to the demographic momentum generated by expansion of coverage in the recent past, and in part to the need to catch up with regional levels and to make Bhutan’s labor force more competitive. The PER explores whether the RGoB will need to realize significant efficiency gains to free up fiscal space, additional resources from taxes, wider participation of the private sector, or a mix of these and other measures. Given these concerns, the objectives of the PER are as follows:  To discuss the effectiveness of financial resources in Bhutan to date, and the potential impact of future targets and reform efforts on financing the health and education sectors  To provide evidence to inform policy in line with the major macroeconomic concerns, with special focus on public systems management, health and education sectors  To feed into preparations of the 11FYP  To contribute to the Development Policy Credit (DPC) series financed by the World Bank Methodology of the PER The focus of the PER was on human development sectors –health and education. Detailed document and secondary data analysis from existing studies, government monitoring information systems data and other published literature was conducted to understand the broader 1 Refer www.grossnationalhappiness.bt for overview as well as thematic discussions ix macroeconomic context of Bhutan and the governance structures and public expenditure patterns in the priority human development sectors of the country. As a part of undertaking the PER, analyses of data from three main data sources were conducted. These include the following:  Review of available data from RGoB sources and sectoral ministries.  Review of existing data from studies undertaken by RGoB and development partners.  Evidence synthesis from new studies undertaken/supported by the World Bank Public Systems Management Since 2008 with the first ever national election, Bhutan became a Democratic Constitutional Monarchy with a written Constitution and a two-house Parliament. Some aspects of public sector management are more recent, however, including the administrative decentralization of service delivery. In terms of governance structure, the country is administratively divided into twenty units called Dzongkhag with larger Dzongkhags further divided into sub-units called Dungkhag. A group of villages form a Geog. Local governance is carried out by elected local institutions called the Dzongkhag Tshogdue (at the Dzongkhag level) and Geog Tshogdue (at the Geog level). The institutions are endowed with considerable powers and authority on administrative, regulatory, financial and general aspects of development. In addition to the Dzongkhags and Geogs, around 10 ministries and 12 agencies are involved in offering government to citizen, government to business, and government to government services. Data from an organizational development exercise undertaken by RGoB in 2007 to understand quality of public administration indicate a reasonable level of policy coordination and integrity in public service. Civil service administration is merit-based, structured around open competition and equal pay for equal value of work. More recently, the RGoB created an integrated program (“Government to Citizen,” G2C)2 to reduce the time citizens need to access key services, simplify processes, create a culture of serving the public, and develop an 'optimal' staffing plan for public agencies. In terms of the process of budgeting and public resource use, the Public Finance Act (June 2007) provides for the development of a three-year Budget Policy and Fiscal Framework Statement based on the Medium Term Fiscal Framework. To improve the resource allocation mechanism, the RGoB moved to formula-based allocation of the current budget as from 2009/10. The central government allocates block grants to local government based on the criteria of: population, land area, and poverty (from poverty map).3 The Gross National Happiness Commission (functioning as Planning Commission) has a centralized M&E system to track outputs of projects and programs; to date this is partially operational. Poverty analysis has further been used to track benefit incidence. Central Government entities representing at least 75 percent of total expenditures are audited annually by the Royal Audit Authority (RAA), the constitutional body set up for the purpose. The RAA also audits state-owned and private/public listed companies, NGOs, and Revolving 2 An independent G2C body administers the program but is overseen by a Committee of Secretaries as well as the Prime Minister's Office. G2C signs an agreement with each line department under the program which is then monitored by the Committee of Secretaries and, in the last instance, the Prime Minister. 3 Household surveys are conducted regularly to make reliable poverty estimates. The National Statistics Bureau produces poverty-related information from the Bhutan Living Standards Survey (BLSS), which is conducted every five years. The latest poverty numbers are from the 2007 BLSS; the 2012 BLSS survey is in process. The next poverty mapping exercise will be conducted soon after the new poverty indicators release in 2012. x Funds whose expenditures are not captured in Government Financial Statements. The Public Accounts Committee (PAC) begins review of the Annual Audit Report (AAR) soon after it is received from the RAA in April, and presents a report to the summer session of Parliament. Public financing is the primary source of health and education expenditure in Bhutan and comprises the resources allocated to the respective ministries and Dzongkhags. Of the total government resources, nearly 30 percent of the resources for health and more than half the budget provision for education is now transferred from central level to Dzongkhags. Findings of the PER Health Despite its geographical challenges, Bhutan has made good progress in achieving universal health coverage, through a general revenue-funded National Health Service. All residents in the country are eligible for public health services, which are largely free of cost. In 2010/11, public health expenditure stood at 4 percent of GDP and 9 percent of total government spending (MOF, 2012)4, and these figures are among the highest in the South Asia region. However, several challenges confront the Bhutanese health sector. The ongoing epidemiological transition, dual burden of communicable and non-communicable diseases, cost and delivery implications of providing health services in hard to reach areas, rising expectations and economic transition of the country together demand a strategic vision for sustainability in health financing. High rates of undernutrition5 pose a challenge to the economic productivity of the coming generations, the maternal health issues have not yet been fully addressed while increasing burden of non-communicable diseases implies recurring expenditures of larger amounts that are required to be spent over large amount of time. Issues surrounding efficiency and effectiveness of expenditures of public expenditure also exist, especially when the country’s performance on health outcomes is compared with other countries having the same level of income and health expenditure. The recent uncertainty around continued assistance from GFATM, and the phaseout plans of other development partners are already looming large on the health financing space. Main barriers to health care access are equity related, and include remoteness, transport difficulty and increasingly, the health needs of the urban and rural poor. The PER includes findings from four studies undertaken in the health sector, and analysis of existing data from different surveys. The country’s first-ever National Health Accounts (NHA) estimates for 2009/10, undertaken as part of the PER exercise, reveal that the public health expenditure as a share of total government expenditure was 6.27 percent and was 3.23 percent of the country’s GDP in 2009/10. As much as 88 percent of the resources for health care came from the government (including support from external sources) and about 11 percent came from households. Household expenses on transportation are not included in this number as they do not form part of the boundaries of ‘health’ as per the international classification o f health accounts but if these were included, the household share would go up to about 20 percent, still considerably lower than other countries in South Asia. 4 Ministry of Finance, Bhutan. Annual Financial Statement 2010-11. Thimphu, 2012 5 Bhutan Multiple Indicator Survey (2010) recorded 33.5 percent stunting, 5.9 percent wasting, and 12.7 percent underweight prevalence, while low birth weight at birth was 9.9 percent. xi A costing study (MOH, 2011) undertaken in health facilities at different levels of care, supported under the PER, showed large productivity differences between facilities and indicates that efficiency gains can be made by adapting staffing to actual workload. Another key concern is that the referral system is practically non-functional, causing overload at the hospital level, especially at Jigme Dorji Wangchuck National Referral Hospital (JDWNRH). The unit costs for OPD-visits, admissions as well as disease-specific groupings of admissions were estimated and confirm that services are generally more costly at higher levels of care. An OPD- visit is generally four times as costly at a referral hospital compared to a BHU and twice as costly compared to a district hospital. Inpatients unit costs show the same pattern. The average inpatient cost of a referral hospital is about three times that of a BHU and more than one-half time that of a district hospital. This underscores the need for an effective referral system and strengthening quality of service delivery at lower levels of care. Analysis of data from the Bhutan Living Standards Survey (BLSS) 2007 shows that there is considerable variation in the geographical distribution of out-of-pocket health spending, where households in Bumthang and Lhuentse spending less than a tenth (in absolute terms) of what is being spent in Tashigang and Wangdue. Household expenditure on rimdo6 and other religious activities as a percentage of total household expenditure was also quite high, and a multiple of the household health expenditure itself, accounting for as much as 2.7 to 5 percent of household expenditure. The PER also supported a study using GIS tools which focused on selected health indicators and their dependence on social and geographical determinants to health, using different layers of GIS maps. The study reveals the situation related to health services, equity and distribution of resources across the country. To illustrate an example, the analysis of the GIS maps shows the significance of road networks from the spatial distribution of maternal deaths. The geographical areas in and around urban pockets of the country have lower maternal deaths than the areas which are remote and have poor access to roads and health facilities. Socio-economic factors, such as the poverty rate, were strongly correlated to infant mortality, as was revealed in Bhutan’s infant mortality map. IMR is clearly high in the districts with high poverty rate, in particular the eastern parts of the country. Another indicator of equity was related to access to health services - the mean time to reach a health facility for the poorest consumption quintile group was 115 minutes as opposed to 33 minutes for the richest quintile. A rapid assessment of the pilot of Special Consultation Services (SCS) - a paid clinic for consulting the publicly employed doctors after the normal working hours of the national referral hospital at JDWNRH showed that about three-fourths of the SCS patients did not find the paid care to be better than the normal (free) hours. The indicated waiting time to see a doctor, for normal hour patients, was most commonly stated as 1-2 hours, while the commonest response from the SCS respondents was a waiting time less than 15 minutes. About 8 percent of the respondents attending the SCS indicated that they were told by the doctors to come for SCS. Qualitative data from FGDs and interviews with doctors showed discontentment with regards to 6 ‘Rimdo”, also spelled as Rimdro, refers to Buddhist religious rituals in Bhutan, performed by monks and often aimed at health and well-being. xii current remuneration system, especially around the capping for their additional income. While it seems that this pilot was advised as a mechanism to accommodate the convenience of the patients as also improve the availability of clinical services through a performance-linked remuneration system, the fine line between patient convenience and a subtle or even overt promotion of the SCS service does need to be treaded carefully. In terms of future directions, immediate and concerted multi-sectoral action on complex challenges posed by undernutrition, maternal mortality and non-communicable diseases (NCDs) is now an imperative. The number and staff composition of ORCs and BHUs may also require revisiting as workload of facilities changes. It is increasingly apparent that there are significant capacity constraints which require attention in the sector. While concerns around sustainability will bring pressures to improve efficiency, the rising income and awareness levels of the general population will raise expectations around the quality of public services. The ministry will also need to gear up to undertake the role of sectoral stewardship, and not just function as financier and direct provider of services, in an environment where both the public and private sector will co-exist. A re-look at remuneration systems, and where necessary, seeking flexibility in Royal Civil Service rules, may be required to introduce focus on results, including incentives based on performance where warranted and feasible. Education Bhutan has established and built up an organized public education system since the 1960s. This diversified sector now includes over 650 schools and institutes, more than 180,000 students and nearly 9,000 teachers. The Constitution (2008) provides for the State to ensure free education to Grade 10 to all children of school going age, and to make technical and professional education generally available, and higher education equally accessible to all on the basis of merit. Bhutan is on track to achieve the Millennium Development Goals (MDGs) in education by the end of the 10th FYP (10FYP, 2008 – 2013). Notably, gender parity is achieved to Grade 10. Females are nearly 50 percent of enrolment in higher secondary education and nearly 40 percent at tertiary level. Public education expenditure stood at 7.3 percent of GDP and 16.7 percent of total government spending (MOF, 2012). Once again, these figures are amongst the highest in the South Asia region. However, as Bhutan achieves the MDGs and the country is reaching middle income status, development partners are withdrawing their financial support in education, and this will have significant impact on Bhutan’s prospects for adequately financing the education sector under the 11th Five Year Plan (11FYP, 2013 – 2018). Similarly, the magnitude of the projected growth in the sector will drive up current costs, which will need to be sustained long term. In primary education, the near-universal enrolment and high completion rate demonstrate that public financing has been effective. There is high access, but this is not effective coverage in terms of learning gains. Quality concerns call for a more effective strategy to provide in-service support for teachers, incentives for performance in the public sector, and service delivery mechanisms which are more cost-effective for geographically remote areas. xiii By the end of the 10FYP, Bhutan is closing the equity gap in primary education, but marked gaps remain for secondary education. Poverty remains key among constraints on school attendance, as corroborated by data from the Poverty Analyses (National Statistics Bureau and World Bank team) demonstrating that Geogs where fewer children attend school tend to be those with comparatively higher poverty rates. This is mainly due to the high out-of-pocket expenditure for households- 26 percent, on average – the highest among all household expenditures for non-food consumption. Review of expenditures against original allocations in the Dzongkhag budgets indicates generally efficient budget execution especially for current expenditures. The PER also finds that primary school teachers’ salaries currently average two times the GDP per capita, compared with 2.6 for lower and middle secondary school teachers, and 2.7 times the GDP per capita for higher secondary school teachers. The levels are low by some international comparisons; for example, the Education for All Fast Track Initiative noted that primary school teacher salaries are 3.5 times GDP per capita in many successful countries. Based on these characteristics and policy challenges, the PER recommends several reforms for the education sector. There is a need to free up fiscal space for the ambitious 11FYP targets which call for increasing the secondary education completion rate to 95 percent (from the current average 75 percent) while also increasing the transition rate from middle to higher secondary education to 90 percent (from the current average rates of 68 percent for males and 61 percent for females). Increases of this magnitude mean higher costs to cover student places – not only in classrooms but also in boarding facilities. Improving quality, also to make secondary education more relevant and attractive to a larger share of the secondary school age population, will further increase costs. Sector wide, the majority of education service delivery is through the public sector. The exception is higher secondary education (HSS, Grades 11-12) where 50 percent of the students are in private institutions. The 11FYP expansion plan to boost student numbers will require more intensive collaboration with the private sector and attention to school mapping to attenuate the existing inequitable pattern of participation in secondary education. The growing role of private institutions and financing in the tertiary sub-sector, notably with the new Education City (under preparation) will become an important factor in expanding coverage. This may not over burden public expenditures, but it raises new issues of quality and relevance. As it will take the Education City some years to become operational, the RUB is the main reference point for the quality assurance and accreditation framework. It is critical that the expansion plans of its constituent colleges be harmonized with, and not operate in parallel to, this framework. The new policy for linking technical and vocational education and training to tertiary level programs ideally requires parallel reform in middle and higher secondary education curricula. The relevant clusters of skills will need to be created and consolidated among the program options in secondary schools to facilitate seamless transfer to tertiary. Depending upon the policy options to be chosen, there may also be implications for the vocational training institutes (currently under the mandate of MoLHR) which presently offer specific occupational training, and for which there is limited interest among the youth. A detailed analysis of the xiv vocational training institutes was not possible within the limits of the first phase of this PER and requires further research. There is some room for improvement in managing capital expenditures. In future, greater effectiveness in public financing could be achieved through more highly tailored strategies responsive to the challenges and opportunities of each Dzongkhag. With in-country migration expected to contribute to more than a four percent increase in the urban population by 2015, the relevant Dzongkhags could also play a larger role in encouraging more private sector involvement in education service delivery. In rural areas with persistent poverty, private schools are not sustainable measures to improve enrolments. Underlying causes for the high dissatisfaction of teachers need to be addressed. Teachers commonly complain about their higher workload (lesson preparations, marking tests, and other school level administrative obligations) compared to civil servants at the same pay grade in other sectors. Only 25 percent of teachers are reportedly satisfied with their profession, which is not adequate to sustain a high performing system. The 11FYP expansion plans for tertiary education are ambitious, but essential for Bhutan to build up a human resource base to support a knowledge-based economy. Student intake is to increase to 15,000 (from just over 5,000 currently). The student numbers alone will require larger outlay in public expenditures, even with the new policy for at least a third of tertiary students to be full-fee paying. At the same time, all constituent colleges of the Royal University of Bhutan (RUB) are planning additional new academic programs and such expansion will also increase costs. A vigorous equity intervention will be critical to ensure students from poor families are not disproportionately affected by the new tuition fee policy at tertiary level. It would be important to conduct a sensitivity analysis of elasticities of household ability to pay, as the RGoB introduces the new system. Such analysis may also help gauge longer term public fiscal impact of the new student loan scheme that is coming on stream. The country has been able to expand access to near universal primary education. Now the challenge is to improve quality and efficiency in service provision, as also a need for more systematic analysis of local conditions analogous to that carried out in the health section in this phase of this PER. Future directions In addition to specific recommendations for the health and education ministries included above, a greater coordination and monitoring role will be expected from the Ministries of Planning and Finance in setting expectations and in monitoring results from the sectoral ministries (Education and Health). To a large extent, the web-based financial reporting tools for budget and expenditure (MYRB and PEMS) are integrated and well co-ordinated, but the same is not the case for performance measurement (PlaMS). This integration should ideally result in an overarching monitoring and evaluation framework for the country with linkages between financing and performance of social sector ministries. xv The current phase of the PER has sought to give policy makers specific and actionable information as a point of departure, but the analytical work needs to be continued in conjunction with the 11FYP preparations, and also updated with results of the BLSS- 2012 which is currently in progress. xvi Matrix of Recommendations from the Public Expenditure Review Issue Key gaps identified in the PER Team Recommendations Time Frame PER HEALTH Sustainability of Health Financing… The ongoing epidemiological transition, A strategic vision for The formulation of a strategic vision Medium term dual burden of communicable and non- sustainability in health for sustainability in health financing communicable diseases, cost and financing Policy action to address the significant delivery implications of providing health capacity constraints in the ministry and services in hard to reach areas, rising field personnel expectations and economic transition of the country will create pressure on Defining basic public health The health ministry will also need to available resources for health services. services as enshrined in the gear up to undertake the role of sectoral constitution, which will be stewardship, and not just function as available for free. financier and direct provider of services, in an environment where both the public and private sector will co- exist and will require significant regulatory capacity in the health ministry. High rates of undernutrition… High rates of undernutrition7 pose a Nutrition has not yet received Immediate and concerted multi- Immediate challenge to the economic productivity the level of policy attention sectoral co-ordination mechanism for of the coming generations and intervention that it the complex challenges posed by under requires. nutrition, Specific action is required to An ongoing national nutrition Short-term, Medium- address stunting and iron assessment will provide specific term and Long-term deficiency anemia recommendations to address these measures, starting at nutrition challenges. the same time 7 Bhutan Multiple Indicator Survey (2010) recorded 33.5 percent stunting, 5.9 percent wasting, and 12.7 percent underweight prevalence, while low birth weight at birth was 9.9 percent. xvii Increasing Burden of Non-communicable diseases… The country needs to address the The health system needs to A reorientation of public health Short-term, Medium- increasing burden of non- gear up to the challenge of services and concerted multi-sectoral term and Long-term communicable diseases- including non-communicable diseases co-ordination mechanism for the measures, starting at prevention, early diagnosis and complex challenges posed by non- the same time effective case management communicable diseases (NCDs) Increasing efficiency and effectiveness of public health expenditure… There are concerns around the Though partly accounted for Efficiency gains can be made by Short-term efficiency and effectiveness of public by geographical and adapting staffing to actual workload health expenditure. For instance, the population density reasons, rather than standard norms. country’s performance on health there are large productivity outcomes lags other countries having differences between The number and staff composition of the same level of income and health facilities. This is ORCs and BHUs may also require expenditure compounded by applying revisiting as the workload of facilities standard norms for staffing changes with improvements in road these facilities. connectivity. A re-look at remuneration systems, and where necessary, seeking flexibility in Royal Civil Service rules, may be required to introduce focus on results, including incentives based on performance where warranted and feasible. Though the health system is Increased emphasis on due referrals, Medium-term structured with different especially when seeking care at levels of health facilities, the National and Regional hospitals. referral system is practically More effective and universal non-functional. This causes application of primary care centres or overload at higher levels of ‘filter clinics’ in urban areas, reducing care, and underutilization of the use of referral hospitals for basic more peripheral facilities. care. xviii Phase-out of international development assistance in health… The recent uncertainty around Though bilateral aid from Bhutan has used international aid Medium-term continued assistance from GFATM, India is likely to continue, primarily for capital investments, and and the phase-out plans of other the impact of reduced future needs for this may need to be development partners such as presence of development carefully assessed and the required Denmark, and the country’s likely partners can impact both investments need to be continued from transition from IDA, are all looming financing and local capacity- available domestic and other sources. large on the health financing space. building. Equity of health expenditure and barriers to access… Main barriers to health care access are Socio-economic factors, equity related, and include remoteness, such as the poverty rate, transport difficulty and increasingly, were strongly correlated to the health needs of the urban and rural infant mortality, as was poor. revealed in Bhutan’s infant mortality map The geographical areas in and around urban pockets of the country have lower maternal deaths than the areas which are remote and have poor access to roads and health facilities Another indicator of equity was related to access to health services - the mean time to reach a health facility for the poorest consumption quintile group was 115 minutes as opposed to 33 minutes for the richest quintile xix EDUCATION Increasing needs for education investments and the phase-out of international development assistance… Similar to the situation in the health The ambitious 11FYP There is a need to free up fiscal space Medium-term sector, as Bhutan achieves the MDGs targets, which call for for sustaining and indeed enhancing the and the country is reaching middle increasing the secondary required investments in the education income status, development partners education completion rate to sector. are withdrawing their financial support 95 percent (from the current in education, and this will have average 75 percent) while significant impact on Bhutan’s also increasing the transition prospects for adequately financing the rate from middle to higher education sector. secondary education to 90 percent (from the current The magnitude of the projected growth average rates of 68 percent in the sector will drive up current costs, for males and 61 percent for which will need to be sustained long females), will require huge term investments. Quality, effectiveness and learning gains… There is high access to education, Quality concerns call for a As also mentioned in the health section, Short-term and but this is not effective coverage in more effective strategy to a re-look at remuneration systems, and Medium-term, starting terms of learning gains. provide in-service support where necessary, seeking flexibility in simultaneously for teachers, incentives for Royal Civil Service rules, may be performance in the public required to introduce focus on results, sector, and service delivery including incentives based on mechanisms which are more performance where warranted and cost-effective for feasible. geographically remote areas. The Royal University of Bhutan is the main reference point for the quality assurance and accreditation framework. It is critical that the expansion plans of its constituent colleges be harmonized with, and not operate in parallel to, this xx framework. Underlying causes for the high dissatisfaction of teachers need to be addressed. Teachers commonly complain about their higher workload (lesson preparations, marking tests, and other school level administrative obligations) compared to civil servants at the same pay grade. The new policy for linking technical and vocational education and training to tertiary level programs ideally requires parallel reform in middle and higher secondary education curricula. Equity and access to education… Poverty remains key among constraints A vigorous equity The 11FYP expansion plan to boost Medium-term on school attendance intervention will be critical student numbers will require more to ensure students from poor intensive collaboration with the private families are not sector and attention to school mapping to disproportionately affected attenuate the existing inequitable pattern by the new tuition fee policy of participation in secondary education. at tertiary level. It would be important to conduct a Short-term sensitivity analysis of elasticities of household ability to pay, as the RGoB introduces the new system. Such analysis may also help gauge longer term public fiscal impact of the new student loan scheme that is coming on stream. xxi Chapter 1 Introduction, Objectives and Overview of HD PER 1.1.1 Sustained investment in the social sectors has enabled Bhutan to make remarkable progress. The country has achieved or is on track to achieve many of the Millennium Development Goals (MDGs). Primary school net enrollment is 95 percent (2011) with a 100 percent completion rate. There is gender parity from primary through middle secondary school (up to and including Grade 10). Many of the health-related MDGs have also been met – several already by 2005. Infant mortality fell to 40.1 per thousand live births by 2005, and child mortality was down to 21.4 by 2005, against the MDG targets of 41 and 30, respectively. The proportion of the population with access to safe drinking water rose to 84 percent in 2005, again exceeding the MDG target a decade in advance. Three out of four pregnant women in Bhutan make the minimum four or more antenatal visits (Bhutan Multiple Indicator Survey, 2010) while over 69 percent of pregnant women deliver safely with the help of health workers (Ministry of Health, 2011). Immunization coverage now extends to over 94 percent of all children. 1.1.2 These achievements occurred during a period of strong economic growth that averaged 8.5 percent per annum over the last decade, during implementation of the Ninth Five-Year Plan (9FYP) (2002/03 – 2007/08, extended by one year) and the 10FYP (2008/09 – 2012/13). The Royal Government of Bhutan (RGoB) is pursuing an ambitious development agenda, including reforms in the social sectors, while scaling up systems to achieve and maintain the targeted levels for the MDGs. More than a third of the total projected costs for the MDGs occur in the final three years leading up to 2015, overlapping with the onset of the 11FYP (2013-2018) that is currently under preparation. Spending pressures are high. At the same time, many donors intend to decrease or end their support after the 10FYP, as Bhutan approaches middle income status. 1.1.3 The central issue for the Human Development Public Expenditure Review (PER) is whether Bhutan’s domestic revenue base will be sufficient as it continues the trajectory of development, and goes through rapid increase in demand for education and health services due in part to the demographic momentum generated by expansion of coverage in the recent past, and in part to the need to catch up with regional levels and to make Bhutan’s labor force more competitive. The PER explores whether the RGoB will need to realize significant efficiency gains to free up fiscal space, additional resources from taxes, wider participation of the private sector, or a mix of these and other measures. The main purpose of the PER is to provide evidence to inform policy in line with these concerns. 1.1.4 The RGoB is systematically improving efficiencies to meet the 10FYP milestones,8 but less attention has been given to the more fundamental and potentially structural issues surrounding adjustments to public financing which may be necessary to support the reform agenda and sustain actions over time. The PER therefore discusses the effectiveness of financial 8 The initiative to “Accelerate Socio-economic development (ABSD) focuses on the education, health, government- to-citizen services, tourism, agriculture and construction sectors. The three main aims are: (i) improved public service delivery, (ii) transformations in governance, including re-alignment of accountabilities and commitment to results through “performance contracts,” and (iii) job creation, to help reduce the overall unemployment rate to no more than 2.5 percent by 2013. 1 resources to date, and the potential impact of future targets and reform efforts on financing the health and education sectors as Bhutan moves into the 11FYP period. The chief considerations from a macro perspective are:  continued pressures on aggregate demand, as public expenditure has risen over the 10FYP period and achievement of FYP ambitious spending targets has required significant external financing;  steeper rise in social sector expenditures in future, given the country’s youthful demographic profile and that the rights to free basic education (to Grade 10) and basic health services are enshrined in the Constitution; greater health systems resources will be needed as the population undergoes an epidemiological transition to more chronic and non-communicable diseases;  economic constraints associated with debt sustainability, potential overheating of the economy due to spillovers from the hydropower and development spending, financial sector vulnerabilities, and gaps in reserve management;  resource constraints from uncertain donor support, and the need for an exit strategy from ODA that would sustain investments for the MDGs beyond 2015;  rising citizen expectations from public services and the increasing public pressure to improve efficiency and effectiveness of services; and  implementation capacity constraints within Bhutan. 1.1.5 The PER findings are expected to feed into preparations of the 11FYP. The PER will also contribute to the Development Policy Credit (DPC) series financed by the World Bank, to give further impetus to the RGoB reform process through this vehicle. 1.1.6 The PER begins with an overview of the macro-economic context (chapter 2) and public sector management (chapter 3) in Bhutan. This is followed by an analytical summary of achievements to date in the health and education sectors (chapters 4 and 5). The PER concludes (chapter 6) with policy options and recommendations which are intended to spur discussion and analysis among policy-makers in Bhutan as they look into different possibilities to increase fiscal space in domestic resources while enhancing the quality of expenditures and improving outcomes in the health and education systems. 2 Chapter 2 Macroeconomic Context9 2.1. Introduction and Country Context 2.1.1 Bhutan is a kingdom in the eastern Himalayas, situated between the Tibetan Plateau in the North and Indian plains in the south. It covers an approximate area of 38,294 square kilometers, spanning roughly 150 km North to South and about 300 km East to West. The country’s terrain is mountainous and rugged with elevation ranging from about 180 meters above sea level in the south to more than 7,550 meters above sea level in the North. 2.1.2 The development philosophy in Bhutan is embedded in the concept of Gross National Happiness (GNH)10 that, as a public policy strategy, seeks to address a more meaningful purpose of development that goes beyond the fulfillment of material satisfaction. The concept is grounded in the four pillars of development; socio-economic, environment, culture, and good governance. Strategies for achieving GNH have been dynamic, evolving and guided by the Bhutan Vision 2020 framework.11 Of late, 72 indicators have been finalized to measure the various aspects of GNH through status indicators, demographic indicators, and various causal and correlation indicators (Wangmo and Sharma, 2010). This chapter focuses on the macroeconomic status. 2.2 Macroeconomic indicators 2.2.1 Bhutan’s record on growth and development has made it a top performer in the South Asian region. The average annual GDP growth rate over 1980-2010 was more than 7.6 percent – one of the highest in the South Asian sub-continent (SAS) (Figure 2.1) – and Bhutan’s real GDP grew at an average annual rate of 7.8 percent over 2008/09 to 2011/2012. The RGoB has recorded impressive progress implementing its medium-term program of institutional and policy reforms through the country’s recent (2008) transition to democracy. With a GDP per capita of about US$2,000, Bhutan is now classified as a lower middle income country. 9 Contributions from Ajay Tandon and Lalita Moorty in developing this chapter are gratefully acknowledged 10 Refer www.grossnationalhappiness.bt for overview as well as thematic discussions 11 Online version available at www.gnhc.gov.bt 3 2.2.2 Bhutan’s economy is highly dependent on, and exposed to, external developments, especially in India. The high levels of year-to-year volatility (see Figure 2.2) largely reflect Bhutan’s reliance on its hydropower sector – with associated spurts of growth as new plants become operational – and exports of electricity and other goods to India. Three-fourths of Bhutan’s imports are from India, and 95 percent of its exports go to India, with about half of all exports being hydroelectricity.12 Figure 2.1: Economic growth rates in selected SAS countries, 1980-2010 Annual GDP growth rate in selected SAS countries, 1980-2010 Bangladesh Bhutan India 30 20 10 Percent (%) 0 Maldives Nepal Sri Lanka 30 20 10 0 1980 1990 2000 20101980 1990 2000 20101980 1990 2000 2010 Year Source: IMF Figure 2.2: Key fiscal/economic indicators in Bhutan, 2000-2017 Key fiscal/economic indicators for Bhutan Actual:1995-2010; Projected:2011-2017 115 20 95 General government debt 15 Economic growth rate Annual percent change (%) 75 Percent of GDP (%) 55 10 Expenditures 35 Revenues 5 15 Inflation rate Unemployment rate -5 General government balance 0 1993 1997 2001 2005 2009 2013 2017 1993 1997 2001 2005 2009 2013 2017 Year Year Source: IMF 2.2.3 Bhutan faces challenges for the 11FYP (2013 – 2018). Economic volatility remains a concern; macroeconomic management is complicated by the heavy dependence on hydropower revenues and timely receipt of budgetary grants. There are economic constraints associated with 12 IMF (2011), Bhutan Article IV Consultation Staff Report, Country Report No 11/123, International Monetary Fund, Washington, DC. 4 debt sustainability and overheating of the economy from growing aggregate demand (due to expanding fiscal deficits, spillovers from rising hydropower expenditures, and sustained growth in private sector credit which worsened by end 2011 – early 2012). Development partners are planning to withdraw their support in several sectors, including the social sectors, as Bhutan attains middle income status and closes in on the MDG targets. Private sector development is constrained by the absence of scale economies, limited access to financing, skills mismatch in the labor force, and under-provision of infrastructure due to Bhutan’s rugged terrain. 2.2.4 The RGoB is maintaining macroeconomic stability through a fiscal strategy that is largely based on the use of concessional loans to finance development of the hydropower resources. Since the onset of the 10FYP, these resources have accounted for approximately 20 percent of the country’s value added by contributing to growth directly through electricity exports to India, and indirectly by stimulating the construction and transport sectors as well as energy-intensive industries. Agriculture contributes 20 percent of non-power GDP, and the service sector (tourism, finance, transport) accounts for another 30 percent. 2.2.6 Supported by fiscal prudence and significant foreign assistance, fiscal balances were at or near surpluses during 2008-2010. According to the Constitution of Bhutan, the government must ensure that the cost of current expenditures is met from the country’s internal resources. The fiscal deficit target is set at five percent of GDP. RGoB’s fiscal anchor has been to contain current spending below domestic revenue. The Constitution further requires that reserves are maintained at a level equivalent to 12 months of imports. Foreign reserves have grown at about 13 percent annually since FY2002/03 and stood at 8.4 months of import cover at end 2010/11. 2.2.7 In the first two years of the 10FYP (FY2008/09 and FY2009/10), including foreign grants averaging 14 percent of GDP, the RGoB ran fiscal surpluses of around 1.6 percent of GDP. In FY2010/11, the fiscal position including grants (excluding the hydropower sector) deteriorated. This was triggered by rising public spending, including a 20 percent increase in civil service wages and allowances. There was also a significant increase in capital expenditure, coinciding with a fall in total revenue that stemmed partly from factors connected with a takeover of Tala Hydropower Corporation by Bhutan’s Druk Green Power Corporation. 2.2.8 Inflation has been rising in recent years, partly as a result of trends (price developments) in India, and partly as a result of domestic demand pressures. Year-on-year inflation accelerated to nine percent in September 2011, compared with seven percent in September 2010. Bhutan’s inflation tracks India’s inflation closely not only because of the large amount of products imported from India, but also because Bhutan’s currency (Ngultrum) is pegged to the Indian rupee. 2.2.9 Government debt is high and rising. Government domestic debt is negligible; nearly 98 percent of RGoB’s debt is owed to external creditors.13 All external debt is public or publicly guaranteed. Until recently, this was all long-term in maturity. In the last three years, non- concessional short-term borrowing has risen since Bhutan accessed liquidity facilities to ease 13 In FY2011, India accounted for 43 percent of the total government debt, and Asian Development Bank (21 percent), and the World Bank (16 percent) were the largest multilateral creditors. 5 rupee shortages arising from sharply increased demand for imports from India -- related to construction of hydropower projects and fuels and also consumption-related imports. 2.3 Economic outlook 2.3.1 The RGoB is taking measures to stabilize the fiscal framework in the wake of the recent rupee shortages. The impact on the domestic economy cannot be fully assessed at the time of writing the PER. Best case could be that impacts are mainly short-term, and the macroeconomic position rebounds to retain the moderate sustainability risk rating assessed by the joint IMF/World Bank Debt Sustainability Analysis (DSA) in May 2011. The DSA suggested that the accumulation of hydropower debt will be met by the commercial viability of the hydropower projects and their rupee-denominated power export earnings. However, this prognosis was modified by later analysis, including that presented in Table 2.2. 2.3.2 The 2012/13 Budget endorsed by the National Assembly in June 2012 envisages expenditure compression and rationalization to reduce the central government deficit to 1.6 percent (from 4.3 percent in 2011/12). The budget allocation for capital expenditure is lowered from 24 percent of GDP in 2011/12 to 16 percent, with priority on completion of ongoing works. No new activities are planned to extend into the 11FYP. Allocations for non-essential current expenditures are cut, to redirect savings to spending on maintenance of public infrastructure. 2.3.3 Careful expenditure prioritization and management of fiscal expansion are crucial for the 11FYP. Compounding the macroeconomic pressures are those directly associated with the social sectors, stemming from the country’s youthful demographic profile and the right to free basic education (to Grade 10) and basic health services enshrined in the Constitution. Citizen expectations are rising for improved efficiency and effectiveness of service delivery, and costs are incurred by the need to address very significant implementation capacity constraints. 2.3.4 Fiscal sustainability and medium term growth outlook: Macro-analysis shows that short-term prospects are subject to down side risks, but Bhutan’s medium term growth outlook is favorable.14 Revenue is projected to decline through 2014/15 (Table 2.1), mainly due to a projected fall in external grants. Decreases in domestic revenue are also expected because of higher maintenance costs and investment costs associated with the Tala hydropower plant. However, the fiscal and monetary policy actions, coupled with the physical controls the RGoB took in mid-2012 have helped reduce aggregate demand pressures and dampen overheating of the economy. Furthermore, although public sector debt is high and increasing,15 this is mostly for commercially viable hydropower projects. These factors contribute to the medium term macroeconomic framework (Table 2.2). While short-term prospects are subject to down side risks, Bhutan’s medium term growth outlook appears favorable. World Bank16 assessments suggest that after 2011-12, growth will be relatively strong, underpinned by the construction of 14 World Bank assessments suggest that growth will be relatively strong, at nine to ten percent for FY12-14. Bhutan Economic Update. April 2012. Poverty Reduction and Economic Management, World Bank (South Asia Region). 15 Public sector debt was estimated at 75.8 percent as of June 2012, compared with 63.9 percent a year earlier. 16 Bhutan Economic Update. World Bank, April 2012. 6 hydropower projects and donor support. Realization of this scenario, however, hinges on how the current macroeconomic challenges are addressed. If the expansionary macroeconomic policy stance were to persist on the fiscal front, overheating would likely intensify. Inflationary pressures would continue and import growth would remain strong, exerting pressure on the external account and fueling further rupee shortages. If the current (time-limited) restrictive quantitative limits on the India rupee were to dampen private sector activity, this would undermine growth and domestic revenue collection. Table 2.1: Central Government Operations (in percent of GDP) 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 Act. Act. Act. Act. Est. Proj. Proj. Proj. Total revenue, 34.9 35.6 40.0 36.6 38.4 31.1 24.3 25.7 incl. grants Total revenue, 23.7 24.2 23.4 22.9 22.9 20.7 18.8 17.8 excl. grants Internal revenue 23.7 24.2 23.4 22.9 22.9 20.7 18.8 17.8 o/w Electricity 8.8 9.5 5.7 5.0 4.0 3.3 2.9 2.5 revenue Tax 10.1 11.2 14.4 15.2 15.5 14.9 12.5 12.1 Non-tax 10.5 10.1 6.4 5.4 4.5 3.9 3.6 3.3 Grants 11.2 11.3 16.6 13.7 15.5 10.4 5.5 7.9 India 9.0 7.6 10.9 10.3 11.1 7.3 3.7 5.4 Other 2.2 3.8 5.7 3.4 4.4 3.1 1.9 2.5 Interest receipts 3.2 2.9 2.5 2.3 2.9 1.8 2.7 2.4 Expenditure & 34.6 33.3 38.0 37.2 42.7 32.7 26.8 25.5 net lending Expenditure 37.8 36.0 38.6 38.4 43.6 33.7 28.0 27.0 Current 18.7 19.1 19.3 19.2 19.9 17.8 16.9 16.2 Wages & 6.8 7.9 8.4 8.5 8.5 .. .. .. salaries Interest 3.3 3.0 2.6 2.3 3.0 2.7 2.7 2.3 Other 8.6 8.2 8.3 8.4 8.5 .. .. .. Capital 19.1 17.0 19.3 19.2 23.6 15.9 11.1 10.8 Net lending & -3.2 -2.7 -0.6 -1.2 -0.9 -1.0 -1.2 -1.5 net advances Fiscal balance, 0.3 1.9 1.6 -2.1 -3.5 -1.6 -2.5 0.3 incl. grants Fiscal balance, -10.9 -9.5 -15.0 -15.8 -19.0 -12.0 -8.0 -7.6 excl. grants Financing -0.3 -1.9 -1.6 2.1 3.5 1.6 2.5 -0.3 External -2.1 -0.7 -0.4 0.7 -1.6 2.4 -2.8 -0.8 financing (net) Domestic -0.3 -0.3 -0.3 -0.2 -0.2 0.0 0.2 1.0 financing (net) Changes in cash -2.2 0.8 1.0 -1.6 -5.3 0.9 -5.0 0.6 balance Memorandum items: Gov’t debt 59.8 59.4 55.1 63.9 75.8 96.1 102.2 119.6 Domestic debt 2.5 2.1 1.6 1.2 0.7 0.8 0.5 2.2 7 External debt 57.3 57.3 53.5 62.8 75.1 95.3 101.7 117.4 Sources: Bhutanese authorities and World Bank Staff Estimates. (Extracted from: International Development Association. Program Document for a Proposed Credit in the Amount of SDR 23.9 Million to the Kingdom of Bhutan for a Second Development Policy Credit. Poverty Reduction and Economic Management, South Asia Region, World Bank, August 20, 2012 Table 2.2: Bhutan Medium Term Macroeconomic Framework 2007/08 2010/11 2011/12 2012/13 2013/14 2014/15 Actual Actual Est Proj Proj Proj GDP growth 10.8 8.5 7.7 12.3 11.5 9.5 Hydropower GDP 38.8 11.1 9.3 34.6 30.1 20.9 Non-hydropower GDP 4.4 6.6 7.1 3.9 2.3 2.3 Consumer prices (period average) 0.0 8.6 9.1 7.6 7.6 7.6 Central government operations Total revenue & grants 34.9 36.7 39.2 32.2 24.3 25.7 Total expenditure & net lending 34.6 38.8 42.7 32.7 26.8 25.5 Overall balance, incl. grants 0.3 -2.1 -3.5 -1.6 -2.5 0.3 Overall balance, excl. grants -10.9 -15.8 -19.0 -12.0 -8.0 -7.6 Government debt 59.8 63.9 75.8 96.1 102.2 119.6 Government domestic debt 2.5 1.2 0.7 0.8 0.5 2.2 Government external debt 57.3 62.8 75.1 95.3 101.7 117.4 External sector Current account balance -2.2 -24.3 -23.5 -24.3 -32.4 -35.3 Trade balance -5.6 -30.7 -27.7 -23.7 -27.5 -32.9 Export (goods) 46.4 39.3 37.5 34.8 33.9 32.3 Imports (goods) 52.0 70.1 65.1 58.5 61.4 65.2 Grants (current transfer) 11.4 13.7 15.5 10.4 5.8 8.5 Capital account balance 4.5 27.5 28.6 25.5 30.2 44.5 Loans (net) 4.5 20.7 22.7 17.4 17.3 31.0 Overall balance 4.7 5.8 5.1 1.2 -3.9 6.6 Gross int’l reserves 646 906 897 871 791 938 (in millions of US$) Gross int’l reserves 10.1 8.4 8.5 8.1 6.7 6.7 (months of G&S imports) Sources: Bhutanese authorities and World Bank Staff Estimates. (Extracted from: International Development Association. Program Document for a Proposed Credit in the Amount of SDR 23.9 Million to the Kingdom of Bhutan for a Second Development Policy Credit. Poverty Reduction and Economic Management, South Asia Region, World Bank, August 20, 2012 2.3.5 The overall macroeconomic framework is sound despite current economic problems. As the economy does run a risk of over-heating in the near future, careful expenditure prioritization will be important going forward. Containing fiscal expansion and enhancing liquidity management to avoid economic overheating are critical for the positive medium term outlook. Longer term, the authorities are well aware that the development paradigm that 8 successfully brought Bhutan to its present level of success may not be adequate or appropriate for the future and that a solution may eventually require adjustment and structural transformation of the economy, while building on past successes. 9 Chapter 3 Public Sector Management and Financing 3.1 Governance Structure 3.1.1 Since 2008 with the first ever national election, Bhutan became a Democratic Constitutional Monarchy with a written Constitution and a two-house Parliament. The Constitution enacted in July 2008 is based on the principles and provisions of the various written laws and legislation which had already been guiding the actions of the King and the functioning of the Royal Government, the judiciary and the National Assembly. Some aspects of public sector management are more recent, however, including the administrative decentralization of service delivery. As a result, the clarification of roles and responsibilities of the local authorities vis-à-vis sector ministry central administration is still being refined at an operational level. Service delivery involves a high provision of public goods. There is an emphasis on health and education as the two sectors together make up a quarter of total budgetary spending. 3.1.2 The country is administratively divided into twenty units called Dzongkhag with larger Dzongkhags further divided into sub-units called Dungkhag. A group of villages form a Geog. A Dzongkhag or Dungkhag would have 2-18 Geogs depending on their respective size. Local governance is carried out by elected local institutions called the Dzongkhag Tshogdue (at the Dzongkhag level) and Geog Tshogdue (at the Geog level). The institutions are endowed with considerable powers and authority on administrative, regulatory, financial and general aspects of development. In addition to the Dzongkhags and Geogs, around 10 ministries and 12 agencies are involved in offering government to citizen, government to business, and government to government services. 3.2 Quality of Public Administration 3.2.1 Overall, policy coordination is reasonable. Integrity in the public service is good. About three percent of the population is in the civil service. This includes teachers which make up approximately a third of all civil servants. Pay, allowances, and other personal emoluments are equivalent to about 43 percent of current expenditure, which is higher than the average (30-33 percent) for low and low-middle income countries. 3.2.2 The RGoB initiated an Organizational Development (OD) exercise in 2007 to strengthen the Civil Service; these efforts are critical to improved governance follow-through on reforms, including decentralization. Civil service administration is merit-based, structured around open competition and equal pay for equal value of work. The rules and regulations require that every civil servant obtain an audit clearance certificate prior to requesting processing of promotion, training, post-retirement benefits, further studies, and participation in conferences and seminars. The Royal Audit Authority issues an Audit Clearance Certificate only if the Audit Information Management System (see below) does not contain any pending adverse reports against the applicant. This has encouraged personal and professional discipline in the discharge of fiduciary duties. There are also strict rules to avoid politicization of the civil service. Among these is that 10 civil servants wishing to participate in competitive party politics must resign their posts, and may not be re-hired for a period of three years. 3.2.3 To strengthen the legal framework for procurement, the RGoB revised and launched a new procurement manual in April 2007, subsequently revised in 2009, containing the rules and regulations guiding public procurement processes in Bhutan. In 2011, the RGoB issued Standard Bidding Documents (SBDs) and Standard Request for Proposals (SRFPs) which are now in use. 3.2.4 More recently, the RGoB created an integrated program (“Government to Citizen,” G2C)17 to reduce the time citizens need to access key services, simplify processes, create a culture of serving the public, and develop an 'optimal' staffing plan for public agencies. The impetus was concern over the limitation in access to services to few centralized locations, which requires citizens to travel many days to reach the service. Depending upon the nature of the service, turn- around times can vary from 7-10 days to 300 days. Seventy-five services were selected for the G2C program. To date, it has focused on computerizing the civil registry (primarily death and birth certificates), creating an on-line system for applying for security clearances (needed to get a job in Bhutan), simplifying the issuance of timber permits and trading licenses, and improving the drug regulatory process. 3.3 Role of Private sector 3.3.1 The RGoB is actively encouraging private sector development among the strategies for Bhutan to accelerate its economic growth. The Private Sector Development Committee (PSDC) was reconstituted in 2009. Aided by an Investment Climate Assessment in 2010, the PSDC is building mechanisms for consultations on cross-cutting and sector-specific policies, strategies for facilitating private sector development, and diagnostics and design of investment climate reforms. Guidelines for Private Participation in Infrastructure Projects were released in May 2010. The RGoB is drafting a Public-Private Partnership Policy (PPP) to govern institutional arrangements for government facilitation of such partnerships in Bhutan, and a proposed (2011) institutional framework for PPPs is under review. 3.4 Budget Process 3.4.1 The Budget is results focused and linked to policy priorities. The Public Finance Act (June 2007) provides for the development of a three-year Budget Policy and Fiscal Framework Statement based on the Medium Term Fiscal Framework. In this results-based planning approach, the actual planning and budgeting timeframe for implementing programs and projects is three years. The five-year framework provides the medium-term perspective with priorities drawn from the respective Five-Year Plan. Outcomes are monitored in terms of measurable goals and targets. 3.4.2 Sectors and agencies are required to ensure their inputs, products, and services contribute to clearly stated results. This practice was reinforced under the 10FYP when performance compacts 17 An independent G2C body administers the program but is overseen by a Committee of Secretaries as well as the Prime Minister's Office. G2C signs an agreement with each line department under the program which is then monitored by the Committee of Secretaries and, in the last instance, the Prime Minister. 11 were instituted; they are signed between the Prime Minister's (PM) office and several ministries and organizations (health, education, civil aviation, Tourism Council of Bhutan, labor). The compacts commit to the most important initiatives to be carried out over three years and hold those accountable to specific targets. Status of achievement is measured and monitored regularly by the Performance Facilitation Unit. The PM is briefed on a weekly basis to take decisions to expedite progress. 3.4.3 To improve the resource allocation mechanism, the RGoB moved to formula-based allocation of the current budget as from 2009/10. The Department of National Budget (DNB) has developed agency-specific formulae which are based on objective indicators including number of employees and frequency of travel, and on this basis agencies receive their allocation for current expenditures. 3.4.4 The Budget covers all government operations, including donor funded projects and decentralized levels of government. Budget formulation and budget execution are based on an administrative classification of national and local Government ministries and entities. All spending by budgetary agencies is subject to the budget, which is expressed in detailed appropriations and authorized through allotment process under the responsibility of the Ministry of Finance. Therefore, the front-line service delivery units like primary schools and primary health clinics fall under the concerned Geog budget. The Geog budgets are disaggregated by detailed activities, and this information is widely available through the budget documents to facilitate timely withdrawal of funds for use by the frontline agencies. 3.4.5 The legislature’s procedures for budget review are still evolving and are not detailed. The annual Budget is formally approved by Parliament in its summer session, and this is followed by a supplementary (revision) Budget, with changes from initial budget allocations, if any, approved by Parliament in its winter session. Generally, only a week has been provided to the Members of Parliament (MPs) before presentation of the annual Budget to the House of Parliament. Experience from the past sessions indicates that the Budget is approved as soon as the presentation is completed in both Houses of Parliament, i.e., a maximum of two days of presentation in the National Assembly followed by another day of presentation in the National Council so long as there is no controversy over the budget heads and amounts mentioned in the Annual Budget Report. 3.5 Equity of Public Resource Use 3.5.1 The Budget reflects and aligns with key poverty priorities. The overarching theme of the 10FYP is poverty reduction, and the Budget is planned accordingly. The central government allocates block grants to local government based on the criteria of: population, land area, and poverty (from poverty map).18 Efforts to reach the most vulnerable groups tend to be implemented through an area and community based targeting—such as the Kidu (or welfare program) – which identifies the poorest villages for special assistance. 18 Household surveys are conducted regularly to make reliable poverty estimates. The National Statistics Bureau produces poverty-related information from the Bhutan Living Standards Survey (BLSS), which is conducted every five years. The latest poverty numbers are from the 2007 BLSS; the 2012 BLSS survey is in process. The next poverty mapping exercise will be conducted soon after the new poverty indicators release in 2012. 12 3.5.2 Moreover, the RGoB spends heavily on social services. Both health and education services have been expanded to the most remote parts of the country in recent years, and the RGoB makes every effort to reach the poor, despite that remoteness of many poor communities sharply raises the marginal cost of providing services. 3.6 Budget and Financial management 3.6.1 Once the Budget is passed by Parliament it becomes a law and the budgetary agencies have the legal right for the approved amount. The Budget Department of the Ministry of Finance notifies all the budgetary agencies and distributes the budget booklets. Apart from this, no cash flow forecasts are made. The recurrent expenditure budgets are released on a quarterly basis (25 percent of total recurrent budget) except for Geogs where it is done on a half-yearly basis (50 percent of recurrent budget). In this way, agencies reliably know how much forward commitment they can afford to make for recurrent expenditures at least for some three months (six months for Geogs) ahead. The releases for capital expenditures are made as and when agencies request for fund releases. Since the approved budget signals the available amounts for expenditures, spending agencies, in general, can access the resources for forward commitments. Within the approved budget, the Ministries, Department and Agencies (MDAs) have the full authority to re-appropriate and make technical adjustment to their budget as per the revised delegation of Financial Powers. 3.6.2 The Ministry of Finance has only recently begun to develop Bhutan Accounting Standards. Pending implementation of such Standards, reference is made to Bhutan’s Public Finance Act (2007) and the FRR2001. The latter specifies internal control rules and procedures for processing and recording transactions; it explicitly rules out arrears. In parallel, the Department of Public Accounts (DPA) has been applying the International Public Sector Accounting Standard (IPSAS) Cash Basis of Financial Reporting as from FY 2006/07. 3.6.3 The Ministry of Finance has a strong system to track expenditures against the budget. All line items in the budget are linked to the expenditures. Moreover, the system can track location where the expenditures take place. Additionally, the Gross National Happiness Commission (functioning as Planning Commission) has a centralized M&E system to track outputs of projects and programs; to date this is partially operational. Poverty analysis has further been used to track benefit incidence. 3.6.4 In respect of aggregate fiscal discipline, the overall budget outturn is strongly controlled, with spending and revenue generally meeting original budget intentions. There is nonetheless a consistent pattern of under-spending on initial allocations, suggesting that there is room to consider whether more aggressive budget review processes would be beneficial for service delivery, to realize efficiencies by re-allocating likely under spending to other areas of emerging needs. 3.6.5 Fiscal reporting. The report on actual Budget performance covers all the budgetary agencies and is compatible with the budget estimates for the period reported. However, since the present system of accounting has no mechanism for recording commitments or obligations, 13 expenditures are captured only at the payment stage and there is no formal tracking of payment arrears. 3.6.6 Releases for recurrent expenditures are made on a quarterly or half-yearly basis, but the budgetary agencies are required to submit the accounts on receipts and expenditures of government resources on a monthly basis in respect of individual bank accounts allotted to them. All the information on government receipts and expenditures are received at the DPA on a monthly basis and can be consolidated for the whole government on a monthly basis, but neither the Public Finance Act 2007 nor the FRR2001 require an in-year budget report. 3.6.7 The reconciliation of Government Consolidated Account (GCA) involves reconciliation of the two sub-principal accounts operated by the DPA and Department of Revenue and Customs (DRC). The individual banks statements are reconciled with the statements from the Central Accounts Department (CAD) of the Central Bank’s Headquarters on a monthly basis. All the Letter of Credit (LC) and Project Letter of Credit (PLC) accounts operated by various budgetary agencies are also reconciled on a monthly basis. 3.6.8 Audits. The legal underpinning for the audit function dates to the Audit Act enacted in June 2006. Central Government entities representing at least 75 percent of total expenditures are audited annually by the Royal Audit Authority (RAA), the constitutional body set up for the purpose. The RAA also audits state-owned and private/public listed companies, NGOs, and Revolving Funds whose expenditures are not captured in Government Financial Statements. For local Governments, the RAA has a policy of carrying out audit of Dzongkhags at least once in two years; for 2009/10, 17 of the 20 Dzongkhags were audited. RAA auditing standards are adapted from the International Organization of Supreme Audit Institutions (INTOSAI) Auditing Standards. In addition, the RGoB has established an Accounting and Auditing Standards Board of Bhutan (AASBB). 3.6.9 The RAA Inspection Reports – a mix of financial audit, compliance audit, and performance audit – are addressed to the ministers/chairs concerned and are summarized in the RAA’s Annual Report. The reports include identification of the persons who should be held accountable for defects, and adverse reports are entered against their name in the RAA Audit Information Management System. The RAA monitors timely follow-up actions through its database system and tracks all pending audit issues and actions to disallow civil servants of Audit Clearance. Generally, the RAA reports on all agencies reveal that the internal control system is comprehensive and well understood, with a high level of compliance with rules and insignificant use of any simplified or emergency procedures. 3.6.10 The Public Accounts Committee (PAC) begins review of the Annual Audit Report (AAR) soon after it is received from the RAA in April, and presents its report to the summer session of Parliament. The system to conduct hearings, where necessary, is being established; there is little completed experience during the life of the new Parliament. However, in previous years, the PAC have conducted hearings and issued directives in cases of controversial observations in the audit of some Dzongkhag construction cases. It should be noted that the PAC have also questioned the under spending of budget (mentioned above) and further recommended that the government explore opportunities to adopt budgeting processes to minimize inefficiencies. 14 3.6.11 The Annual Financial Statement (AFS) of the Government covers all government budgetary agencies. It has complete information on revenues, expenditures, grants and loans. It also includes statements of the Government Equity Portfolio in corporations and financial institutions, Government Guarantees, and government outstanding debt. The four month time line for completing the audit of the AFS is being achieved, and the Audit reports are tabled in the National Assembly during the summer session. 3.7 Financing the Health and Education Sectors 3.7.1 Public financing is the primary source of health and education expenditure in Bhutan and comprises the resources allocated to the respective ministries and Dzongkhags. Of the total government resources for health, nearly 30 percent is allocated to Dzongkhags under the decentralized system and spent by them. External assistance plays a considerable role in the health sector, and accounts for about 18 percent of the total government spending. 3.7.2 The situation is similar in education, whereby more than half the budget provision for education is now transferred from central level to Dzongkhags who are responsible for teacher deployment, construction and maintenance, and (since 2010) ordering and distributing textbooks. The proportion of the education sector expenditures financed from external sources varies from one year to the next depending upon project cycles, and was over 20 percent in 2009/10. Budgeted resources are program based. Budgets to schools are based on enrollment, salaries and equity; schools have few discretionary funds. School heads consult with and inform parent councils about school performance, but parent councils have no major decision making authority. At the tertiary education level, the Royal University of Bhutan transitioned into a new modality of funding and financing starting July 2011 as the University was accorded autonomous status. The RUB now receives its operational budget as a block grant based on per-student costs. In addition the University now admits a specified percentage of students on self-finance per year; this is expected to reach 30 percent by 2013. In addition, the University is now more actively engaged in fund raising from other sources. 3.7.3 The momentum Bhutan sustained since 2000 to improve social indicators in education and health has been supported by public expenditures which ranged from five to seven percent of GDP for education, and from three to five percent of GDP for health. In both cases, the spending as proportion of GDP is quite high compared with other countries in South Asia (with the exception of Maldives). To illustrate, for FY2009/10, about 17 percent of Bhutan’s public domestic expenditures were in education – higher than any other category of government developmental expenditure – and over seven percent in health (Table 3.1). Further detail on historical trends and future projections follows, in the respective chapters on health and education. 15 Table 3.1: National budget: sector expenditures for fiscal year 2009/10 (millions of Nu.; % GDP) Sectors Expenditure % of % to Current Capital Total Total GDP Social Services 4,517.91 2,499.6 7,017.4 24.1 10.5 Health 1,364.2 787.5 2,151.8 7.4 3.2 Education 3,153.6 1,712.0 4,865.7 16.7 7.3 Economic 2,165.2 6,530.5 8,695.7 28.9 13.0 Services Renewable Natural Resources 1,374.2 1,715.0 3,089.3 10.6 4.6 Mining, Manufacturing, Industries 145.9 60.6 206.5 0.7 0.3 Transport & Communication 395.7 3,708.7 4,104.4 13.7 6.1 Energy 50.1 776.0 826.1 2.8 1.2 Other Economic Services 199.3 270.1 469.5 1.61 0.7 Public Order & Safety 875.8 429.5 1,305.4 4.5 2.0 Religion & 511.3 442.4 953.7 3.3 1.4 Culture Housing & Public Amenities 336.9 858.7 1,195.6 4.1 1.8 General Public 4,495.5 2,168.5 10,871.4 37.3 16.3 Services General Public 3,802.7 Service 2,247.3 1,555.4 13.1 5.7 National Debt 2,248.2 613.1 7,068.7 24.3 10.6 Services (includes 1,471.8 Lending, 2,735.5 Repayment) TOTAL 12,902.7 12,929.1 30,039.2 100.0 100.0 GDP as per 66,865.3 National Statistics Bureau 1 Numbers may not add up due to rounding. Source: Department of Public Accounts and Spending Agencies, quoted in Annual Financial Statements year ended June 2010, Department of Public Accounts, Ministry of Finance. 16 Chapter 4 Health Expenditure in Bhutan 4.1. Introduction and Sectoral Context 4.1.1. Despite its geographical challenges, Bhutan has made good progress in achieving universal health coverage, through a general revenue-funded National Health Service. All residents in the country are eligible for the public health services, which are largely free of cost, and there are no volume restrictions to access the public health services. The services are also open to all visitors, whether citizens or otherwise. Presently, the service continues to be free for non-nationals also, though there are plans and polices to levy charges for foreigners in the future. 4.1.2. Bhutan has sustained investment in health over the years and made remarkable progress. In 2010/11, public health expenditure stood at 4 percent of GDP and 9 percent of total government spending (MOF, 2012)19, up from a level of 3.2 percent of GDP and 7.4 percent of the total government spending in the previous year (2009-10, MOF 2011), and these figures are amongst the highest in the South Asia region. Government is the predominant source of funds for the Bhutanese health system, and has contributed between 68-80 percent of the total health expenditure of the country over the last decade (Nagpal, 2009)20. This has been funded through domestic as well as external sources. For the domestic sources, the main inputs come from the tax and non tax government revenues supplemented by yield from the innovative health financing tool, the Bhutan Health Trust Fund (BHTF). Some health-related revenues are also collected in the form of the nominal user fees. External funds are an important source for the sector, though used only for capital expenditure, and have declined in recent years as a proportion of the total public health budget. 4.1.3. Although physical access to primary health services is already fairly high and been sustained at over 90 percent,21 the Bhutanese health system has to address the unique challenge of a small population (and limited health professionals) scattered over a geographically inhospitable terrain. There are still Dzongkhags (districts) in the country which require walking for days, not hours, from the nearest road, even though the aerial distance may only be a few kilometers. Although the road network has rapidly improved and connectivity to major towns is now far better than what existed even a few years back, reaching out to the last family and individual who may need health services continues to be a daunting task. Recent sector reviews and analyses have confirmed that Bhutan’s main challenge remains reaching these currently underserved groups, particularly for maternal health care services. 4.1.4. During the 10th Five Year Plan (10FYP) the Ministry of Health (MOH) has continued to develop the country’s health services. More than a year before the end of the plan, three (of five) service targets have already been attained: the number of health facilities now ensure that 90 19 Ministry of Finance, Bhutan. Annual Financial Statement 2010-11. Thimphu, 2012 20 Nagpal S. Health Financing and Expenditure Review- Bhutan. WHO SEARO report, 2009 21 Gross National Happiness Commission (Royal Government of Bhutan). Tenth Five Year Plan Volume 2: Programme Profile 2008-2013. 17 percent of the population live within three hours walk to a health facility, there are 25 traditional medicine units22 at Basic Health Units (BHUs), and there are more than two functioning ambulances at each district hospital. Two of the plan targets have not been reached, namely to put three doctors at each district hospital and establish a super-specialized hospital under foreign direct investment. (Joint Review Mission report, 2012). 4.1.5. An overview of the development of health services in Bhutan, as reflected in the growth in the available health infrastructure during the past decade is shown in table 4.1 (Joint Review Mission report, 2012). Table 4.1: Heath Infrastructure in Bhutan (2001-2012) 2001 2005 2010 2012 Hospitals 29 29 30 31 Basic Health Units 168 176 181 181 Out-Reach Clinics 461 485 518 518 Doctors 114 145 187 181 Health Assistants 163 171 366 429 Nurses 569 538 556 723 Source: Joint Review Mission report, 2012 based on Annual Health Bulletins, Ministry of Health, various years. Figures for 2010 are the latest published figures. For 2012, these staff numbers are based on information provided to the Joint Review Mission by the Human Resource Division of the Ministry of Health. 4.1.6. Despite numerous challenges, Bhutan has made steady and significant improvements in key health outcomes over the past several decades. Life expectancy has increased steadily from just about 37 years in 1960 to about 67 years in 2009 (Figure 4.1). Bhutan’s infant mortality rate also declined steadily from 190 per 1,000 live births in 1970 to about 44 per 1,000 live births in 2009 (Figure 4.1). Similar magnitudes of declines are evident for under-five mortality rates and other key health outputs and outcomes in the country. It is notable that Bhutan is likely to meet both the Millennium Development Goals (MDGs) for child and maternal health by 2015. In addition, Bhutan’s attainment of health outcomes has been accompanied by very low levels of out-of-pocket spending. 22 Bhutan has its own system of traditional medicine which is formally recognized by the country and formal qualifications in the discipline are also offered. As mentioned above, many primary care facilities and almost all the higher level facilities have traditional medicine units that offer treatment under the traditional system of medicine. Qualified practitioners in traditional medicine system are to be contrasted from traditional healers and other informal providers which do exist in the country, particularly in the remote and rural areas. 18 Figure 4.1: Bhutan’s performance on key health outcomes, 1960-2010 Population health indicators for Bhutan, 1960-2010 350 70 Under-five mortality Life expectancy 250 60 Mortality rate per 1000 Infant mortality Life expectancy 150 50 100 40 75 45 30 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Year Source: WDI Note: y-scales logged Source: World Bank Analysis using WDI data 4.1.7. For the health sector, the epidemiological transition, dual burden of communicable and non-communicable diseases, cost and delivery implications of providing health services in hard to reach areas, rising expectations and economic transition of the country together demand a strategic vision for sustainability in health financing. Concerns around aging population, rural to urban migration, effects of climate change, emergencies and disasters and burden of mental health problems have been identified as other emerging challenges23. The small size of the country’s population as well as its geographical constraints create a dependence on import for most health care supplies, and this combined with increasing expectations from the country’s free health care services will stretch the country’s resources. Improving road access is leading to lower level health facilities being increasingly bypassed and referral facilities being overcrowded. The cost of addressing shortage of human resources for health remains a challenge for health service delivery. High rates of undernutrition24 pose a challenge to the economic productivity of the coming generations. Maternal health issues have not yet been fully addressed. At the same time, the increasing burden of non-communicable diseases implies recurring expenditures of large amounts over prolonged periods- thus implying a lot of fiscal effort if this were to be fully financed through the public budget. 4.1.8. Issues surrounding efficiency and effectiveness of expenditures of public expenditure also exist, as can be illustrated with the benefit of international comparisons. Bhutan spends a large proportion of its public resources on health (Figure 4.2) compared to other countries in the region but has not achieved equally remarkable outcomes. Bhutan has made progress on infant mortality and life expectancy, enhancing its standing among regional peers and performing better 23 Report of the World Bank supported Policymakers’ Roundtable on Health Financing in Bhutan, Thimphu, December 2011 24 Bhutan Multiple Indicator Survey (2010) recorded 33.5% stunting, 5.9% wasting, and 12.7% underweight prevalence, while low birth weight at birth was 9.9%. 19 than, for instance, India (see figures 4.3 and 4.4).25 However, Bhutan does not perform as well when compared to other countries relative to the level of its health spending on infant and maternal mortality rates (see Figure 4.5).26 Figure 4.2: Government Share of Health Expenditure versus Income per Capita Government Share of Health versus Income, 2007 Government health spending (% of THE) Government health spending (% of budget) Government health spending (% of budget) 100 30 Maldives 80 Bhutan 20 60 Bhutan 10 Sri Lanka Nepal Maldives Sri Lanka 40 Nepal Bangladesh Bangladesh 5 India India 20 Pakistan Pakistan 250 1000 5000 25000 250 1000 5000 25000 GDP per capita, US$ GDP per capita, US$ Source: World Development Indicators, WHO, & Royal Monetary Authority, 2009 Health expenditure data are preliminary as of May 2009 Note: log scale Figure 4.3: Infant mortality rates in selected Figure 4.4: Life expectancy in selected comparators (1960-2008) comparators (1960-2008) Infant Mortality Rates in Selected Comparators (1960-2008) Life Expectancy in Selected Comparators (1960-2008) 250 80 70 100 India Life expectancy (years) 60 Sri Lanka 25 50 Pakistan Maldives 40 Bangladesh Afghanistan 5 Nepal Bhutan 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Year 30 Afghanistan Bangladesh Bhutan 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 India Sri Lanka Maldives Year Nepal Pakistan Source: WDI, WHO Source: WDI Note: y-axis log scale Note: y-axis log scale 25 World Bank analysis presented at South Asia Regional Forum on Health Financing in Maldives, June 2010. 26 World Bank analysis presented at a Health Financing workshop in Bhutan, January 2010. 20 Figure 4.5: Health Indicators Relative to Income and Health Spending Health Indicators Relative to Income and Spending Infant Mortality Rate, 2007 Maternal Mortality Rate, 2005 average Worse than average spending than to health Maldives relative Worse Maldives Bhutan Nepal Bhutan India India Pakistan than average Better than average Bangladesh Performance Nepal Bangladesh Sri Lanka Pakistan Better Sri Lanka Better than average Worse than average Better than average Worse than average Performance relative to income Performance relative to income Source: World Development Indicators, WHO, & Royal Monetary Authority, 2009 Health expenditure data are preliminary as of May 2009 Note: log scale 4.1.9. The adverse impact of the recent financial crisis which could restrict the support of international development partners, could also create resource challenges. The recent uncertainty around continued assistance from GFATM, and the phaseout plans of other development partners are already looming large on the health financing space. The uncertain external support, especially in the aftermath of global financial crisis accentuates the need for an exit strategy from external assistance which would sustain investments for the MDGs beyond 2015. The draft 11th Five Year Plan (11FYP) of Bhutan does target self reliance by 2018. 4.1.10. It was in the context of these challenges and emerging issues, in January 2010, that the Ministry of Health, Royal Government of Bhutan, in association with the World Bank and with support from other development partners including WHO, UNICEF and DANIDA, organized a workshop on strategic options for sustainable health financing in Bhutan. The workshop concluded with a recommendation to strengthen the evidence base which could lead to more informed policy making for the health sector in Bhutan. A list of proposed studies which were recommended by the workshop were later accepted in the joint review mission of the health sector (2010) as priority areas for Bhutan. The present Human Development Public Expenditure Review (HD PER) for the health sector, which was taken up by the World Bank jointly with the Ministry of Health in light of the recommendations of the above workshop, has already supported several of these identified studies. This report reflects the evidence gathered, compiled and analyzed under the following studies which have been undertaken in 2010/11 and 2011/12, namely, 1. National Health Accounts (NHA) for Bhutan (2009-10) 2. GIS mapping of health facilities and key health outcomes 3. Costing of health services at different levels 21 4. Rapid assessment of the Special Consultation Service (SCS) in the Jigme Dorji Wangchuck National Referral Hospital (JDWNRH), Bhutan. 4.1.11. These studies have been supplemented by additional analysis of existing documents and reports, as well as analysis of data from the Bhutan Living Standards Survey (BLSS), 2007 and other similar sources in forming the basis for this report. The remaining sections of this chapter describe these analytical findings in greater detail. 4.2 Health Financing and Resource Flows in Bhutan 4.2.1. National Health Accounts (NHA) provide a globally recognized framework to Table 4.2: Key Indicators of Health systematically measure the sources of public Expenditure in Bhutan (NHA 2009-10) and private health expenditures and the flow of funds in a country’s health system. Input Macro Indicators 2009-10 from NHA provide an evidence base regarding resource gaps and inefficiencies PHE as % of THE 88.0 and can help in making policy decisions to reduce out-of- pocket payments borne by PHE as % of TGE 6.3 households, increase total health expenditures, and identify cost-saving PHE as % GDP 3.2 opportunities on government spending. Linking NHA data with nonfinancial THE as % GDP 3.7 information (such as output and outcome indicators) can provide a powerful means of PHE – Public health expenditure linking financial investments with THE – Total health expenditure attainments in health status and drive TGE – Total government expenditure GDP – Gross domestic product improvements in the effectiveness, efficiency, and quality of health services27. 4.2.2. The health system in Bhutan is predominantly financed from the general revenues of the Royal Government of Bhutan. Healthcare services are provided free to all the citizens and residents of the country by the Government. In this context, the Royal Government of Bhutan undertook the country’s first ever National Health Accounts study for the year 2009 -10, as a follow up to the Paro workshop mentioned earlier in this report and as part of the PER exercise in the health sector. The recent NHA study elaborated the flow of funds in the health sector in Bhutan for the year 2009/10 for which latest public expenditure accounts were available. The NHA developed in this study relied upon data on government expenditure on health, including external assistance, as reported by the Department of Public Accounts in the Ministry of Finance. Household expenditures as estimated based on the Bhutan Living Standards Survey (BLSS) 2007 and a number of other sources. Information from three national level household surveys was available in the country at the time of the NHA exercise: a. Bhutan Population and Housing Census (BPHC), 2005 b. Bhutan Living Standards Survey (BLSS), 2007, and 27 Maeda A, Harrit M, Mabuchi S, Siadat B, Nagpal S. Creating Evidence for Better Health Financing Decisions- A Strategic Guide for the Institutionalization of National Health Accounts. World Bank, Washington DC. 2012. 22 c. Bhutan Multiple Indicator Survey (BMIS), 2010. 4.2.3. Of these, only the BLSS 2007 collected information on health expenditures incurred by the households. The survey covered all the twenty districts of the country covering 9,798 households from both rural and urban areas during March – May, 2007. The survey adopted a stratified two stage sampling method for data collection. Data on any reported sickness and injury during the past four weeks of the survey and their corresponding expenditures have been elicited and recorded during the survey. The data thus collected have been used for estimating the total household expenditure on health in the country. The total health expenditures estimated as above from the survey data represent only the expenditures incurred during a four week period. This was adjusted by a factor of 52/4 to arrive at annual health expenditures incurred during 2007. These estimates were further adjusted using the consumer price index (CPI) of health of 2008, 2009 and 2010 to reflect inflation in the country and to arrive at health expenditure for the year 201028. The estimates have been corrected for population increase as well, in addition to inflation. These adjustments to the available data were made as per the standard practices followed in similar settings in order to enable cross country comparisons. 4.2.4. In summary, the NHA estimates for Bhutan reveal that the public health expenditure as a share of total government expenditure was 6.27 percent29. The same, when compared with the country’s GDP, was 3.23 percent of the GDP in 2009-10. The estimates also reveal that about 88 percent of the resources for health care are from the government (including support from external sources) and about 11 percent comes from households (household expenses on transportation are not included as they do not form part of the boundaries of ‘health’ as per the international classification of health accounts. If these were to be included, the household share would go up to about 20 percent). Of the total government resources, nearly 30 percent is allocated to districts under the decentralized system and is spent by them. External assistance plays a considerable role in the health sector of the country. This accounts for about 18 percent of the total government spending on health. 4.2.5. Public financing is the primary source of health expenditure in the country and comprises the resources allocated to ministry of health and districts. These sources of public financing amounted to Nu.1,981 million during the year 2009/10 (including external development partner support). Of this, the spending of domestic resources through the Ministry of Health accounted for 51 percent and spending through districts accounted for 31 percent, and external assistance accounted for the rest. The per-capita public spending on health is estimated to be Nu. 2,847 for the year 2009/10. 28 Statistical year book, 2010, National Statistical Bureau, Royal Government of Bhutan 29 These NHA estimates are different from the sectoral expenditure estimates in the Annual Financial Statements, as the boundaries of what constitutes health expenditure differ. In case of NHA, these do not include water, sanitation and other such expenses in line with standard NHA methodology. Thus, while the MOF 2011 estimates for health expenditure in 2009-10 is 7.4% of the total government expenditure, the number is 6.27% for the same year using standard NHA boundaries for health expenditure. 23 Figure 4.6: Financing flows in the health sector in Bhutan (NHA 2009/10) Table 4.3: Public Spending on Health in Bhutan: 2009/10 in Per- Millions % Capita Source Nu. Share (Nu.) RGOB through Ministry of Health (MOH) 1,006 50.8 1,446 RGOB through Districts 624 31.5 896 External Assistance 351 17.7 504 Total 1,981 100.0 2,847 Source: Expenditure Summary 2009-10, Department of Public Accounts, Ministry of Finance, Royal Government of Bhutan, and reproduced from Bhutan NHA 2009-10 4.2.6. Even though health services are provided free of cost, households do end up paying for incidental expenses (such as transport, which is not included in the NHA expenditure estimates above) as well as for purchase of medicines and related supplies in the country. Insurance schemes for health are in nascent stage in the country, so most household expenditure is out-of- pocket, including that incurred for treatment outside the country. However, a few organized manufacturing and service sector industries in the country offer financing/ reimbursement 24 mechanisms (both contributory as well as non-contributory) for their employees. The relatively small OOP burden in Bhutan is encouraging but is also subject to the fact that there are limited private providers of health services (and no private medical doctors at all). Any rapid and unregulated expansion of private providers may have significant impact on the existing OOP and simultaneously on financial protection. Government being the biggest financer of health in Bhutan, any change in health care costs, will directly affect the government and its ability to fund health care. 4.2.7. Even in a conducive macro-fiscal environment, there may not always be continued increases in government health spending, even if overall government expenditures rise, and especially not at the same rates as may have been observed in the past. This has already been the case for Bhutan in recent years. Although government health spending has generally risen faster than economic growth in the country in the past, this trend has been slowing. From 1995-2002, the elasticity of government health spending to GDP was 2.03 in Bhutan, implying that on average every 1% increase in GDP was associated with a 2.03% increase in government health spending. Following 2003, this trend has declined significantly with elasticity dropping down to 1.36. If this trend observed since 2004 continues, government health spending is likely to be around 5.7% of GDP by 2017 (up from about 4.2% of GDP in 2010). Figure 4.7: Elasticity of government health spending to GDP, 1995-2010 Elasticity of government health spending to income, 1995-2010 4 2010 3 2009 2008 2007 2 2002 2006 2005 2000 2001 2003 Elasticity=1.36 1 2004 1999 1998 1997 Elasticity=2.03 1996 1995 20 40 60 80 GDP per capita, ngultrum (billions) Source: WHO Source: World Bank Analysis using WHO data 4.2.8. The household survey (BLSS 2007) collected some information on spending for items such as consultation, medicines, transportation and others. Of these categories of expenditure, households spend primarily on transportation, which is more than 45 percent of their spending. Purchase of medicines accounts for about 33 percent of household spending on health. Given the geographical terrain of the country, such high share of costs on transportation by households is understandable. However, part of this high cost on transportation could be attributable to the fact that people do travel to neighboring countries for specialized medical treatments, and though this is covered by the government in a large number of cases, it is not always the case. As mentioned earlier, expenditure incurred by households on transportation for availing treatment has been excluded from the NHA framework to be consistent with the definitions of International 25 Classification of Health Accounts (ICHA). Household spending from the BLSS 2007 data has been further analyzed (beyond the analysis undertaken for Bhutan NHA 2009-10) and is discussed in section 4.3. 4.2.9. Major burden of provision of health services in the country lies on the national and regional referral hospitals as well as to some extent on the publicly-run district hospitals, as there are no private hospitals or private medical practitioners in the country. Of the total resources spent on health in the country, 52 percent is spent on public providers, while about 24 percent is spent on general health administration. It is interesting to note that about nine percent of the resources are spent on private providers which include treatments outside country (both publicly and privately funded), and care from those traditional practitioners who are not part of the public health system. 4.2.10. It should be admitted at the outset that the estimates of health expenditures in the country can further be improved, particularly the spending by households. The recent refinements in BLSS 2012 should help improve the granularity and quality of data available for the NHA estimations henceforth. Section 4.3: Household Expenditure on Health: Additional Analysis of Bhutan Living Standards Survey 2007 4.3.1. As mentioned earlier, even though health facilities are largely government-owned and services are provided free of cost, households do end up paying for incidental expenses. Their main expense, transport, is not included in the NHA expenditure estimated above. Households also spend on purchase of medicines and related supplies in the country. Additional analysis of data from BLSS 2007 on the distribution of health services by consumption quintiles and geographic regions showed interesting patterns, which have been discussed in this section. 4.3.2. Mean percentage of health expenditure as a share of all household expenditure was highest at 1.49 percent for people in the poorest consumption quintiles and was lower at 0.94 percent for the middle quintile. In terms of geographic distribution, share of health in household expenditure was highest at 1.23 percent for the Western region and lowest at 0.96 percent for Eastern region. Table 4.4: Mean distribution of percentage of health related expenditure by consumption quintiles and region Male Female Total Male Female Total Poorest 1.47384 1.499456 1.486741 Western 1.262712 1.194805 1.228296 Second 1.346103 1.239791 1.292369 Central 1.217759 1.165985 1.191235 Middle 0.931695 0.949327 0.940714 Eastern 0.97744 0.93938 0.957838 Fourth 1.231032 1.096439 1.162613 Total 1.171038 1.114323 1.142081 Richest 1.004574 0.958835 0.980823 Total 1.171038 1.114323 1.142081 Source: BLSS 2007 26 4.3.3. The geographical distribution of health spending, now in absolute per-capita terms after adjusting for inflation and population increase, is graphically depicted in Figure 4.3. There is considerable variation between districts with households in Bumthang and Lhuentse spending less than a tenth of what is being spent in Tashigang and Wangdue. Reasons for these geographical variations require further research. Figure 4.8: Per-capita household spending on health by districts in Bhutan: 2009-10 1,693 1,500 1,340 1,071 1,054 841 616 425 419 356 301 299 286 285 276 256 234 176 128 124 Pemagatshel Haa Bumthang Trongsa Punakha Zhemgang Samtse Tsirang Thimphu Paro Lhuentse Chhukha Gasa Wangdue Mongar Sarpang Tgang Tyangtse Dagana Sjongkhar Source: Compiled from district-wise Expenditure Summary 2009-10, Department of Public Accounts, Ministry of Finance, Royal Government of Bhutan. Share of different components of household expenditure 4.3.4. Of the amount spent by households on health, the share was highest for transportation followed by expenses on medicines. Though the absolute numbers differ, the data do not show significant differences in share of transportation expenditure across consumption quintiles and region. However, the mean share of expenditure on medicine was highest for the richest, which decreased down the consumption quintiles with the lowest being spent by the poorest group (see tables 4.5 and 4.6). Table 4.5: Mean distribution of share of health related expenditure by consumption quintiles (expressed as % of total expenditure by households) Consumption Other health Total quintiles Consultation Medicines Transportation expenses Poorest 0.010888 0.029979 0.104336 0.038713 1.486741 Second 0.010413 0.035451 0.169634 0.029135 1.292369 Middle 0.009307 0.045804 0.120739 0.007579 0.940714 Fourth 0.019336 0.078177 0.1308 0.017011 1.162613 Richest 0.007473 0.123218 0.104516 0.032104 0.980823 Total 0.011506 0.068493 0.125046 0.024274 1.142081 Source: BLSS Survey, 2007 (Block 1.3 Q31) 27 4.3.5. A similar pattern is also observed across the geographic regions. The mean share of expenditure on medicine was highest for the Western region and lowest for the Eastern region (see table 4.6 and Figure 4.9). Table 4.6 Mean distribution of share of health related expenditure by geographic region (expressed as % of total expenditure by households) Other Total Geographic health region Consultation Medicines Transportation expenses Western 0.012074 0.092757 0.125237 0.023696 1.228196 Central 0.008913 0.051848 0.129374 0.027766 1.191235 Eastern 0.012975 0.0449 0.120772 0.022 0.957838 Total 0.011506 0.068493 0.125046 0.024274 1.142081 Source: BLSS Survey, 2007 (Block 1.3 Q31) Figure 4.9 Composition of household expenditure on health in selected districts 100 80 60 40 20 - Wangdue Tgang Gasa Paro Sarpang Transport Others Medicines Consult. Notes: 1. Household expenditure on health as reported in BLSS 2007. 2. The five districts presented here account for nearly 55 percent of the total health expenditure incurred by households in the country. Per-capita expenditure on health by households is also high in these districts. 3. Reproduced from Bhutan NHA 2009-10 Expenses on Rimdo and religious activities aimed at promoting health or preventing illness 4.3.6. Though not within the internationally accepted boundaries of health expenditure, household expenditure on rimdo and other religious activities as a percentage of total household expenditure was generally high. There was mixed and slight variation in the expense across the consumption quintiles. This was highest (4.96 percent) in the West, and was lower for Central 28 and Eastern regions with mean share of 3.7 and 2.7 percent of total household expenditure respectively (see table 4.7). Table 4.7: Mean share of household expenses on rimdo and religious activities by consumption quintiles and region (expressed as % of total expenditure by households) Consumption quintiles Mean Region Mean Poorest 3.329812 Western 4.969131 Second 3.963849 Central 3.704492 Middle 4.421755 Eastern 2.659194 Fourth 4.072484 Total 3.923705 Richest 3.693762 Total 3.923705 Source: BLSS 2007 (Block9. Q.2&3) Illnesses / injuries and Work Absence 4.3.7. About 57 percent of all those respondents who reported suffering from any illness or injury during the last four weeks were women. The reported frequency of sickness and injury among the people in the poorest quintile was lowest (13.23 percent), and this increased up the consumption ladder with the highest (26.82 percent) being reported by the richest group. (see table 4.8) Table 4.8: Percentage distribution of people who suffered in last four weeks by consumption quintiles % of people who Consumption suffered from quintiles sickness/injury Poorest 13.23 Second 17.85 Middle 19.39 Fourth 22.7 Richest 26.82 Total 100 Source: BLSS 2007, (Block 1.3 Q26) 4.3.8. The subjectivity inherent in the above response can be seen by comparing the results with another query on being away from work due to sickness. It is interesting to note that while the richest group reported suffering from sickness more than the lower quintiles, the same group was away from work for the lowest number of days on an average (5.6 days), which increased down the consumption ladder with 8.3 days for people in the poorest category. Once again, causes for these variations require further research, such as whether they are due to longer, more severe or untreated episodes in the poorer quintiles, or due to a subjective difference in what is considered ‘illness’ or not. 29 Access to medical care 4.3.9. In a similar vein, the poorest quintiles were more likely to access care from traditional religious practitioners and less likely to consult doctors or qualified nurses (table 4.9). Amongst those who did not seek health care services at all, about 48 percent among the poorest group mentioned that they thought it was not necessary to consult anybody and 21 percent stated transportation problems to get to health centers. About 14 percent did not consult any services. The reasons for not consulting any services are summarized in Table 4.10. Table 4.9: Percentage distribution of people who consulted with different sources of healthcare services, by consumption quintiles Hospital/ Pharm Indigenous Traditional Consump. Quintiles No one Private doctor BHU -acist Dentist Centers practitioner Other Total Poorest 18.84 0.41 66.59 0.49 0.3 0.3 9.24 3.83 100 Second 15.5 0.98 71.32 0.75 0.15 0.28 8.47 2.55 100 Middle 13.71 1.33 74.85 1.67 0.27 0.26 6.08 1.83 100 Fourth 12.42 1.45 76.19 2.36 0.24 0.42 5.59 1.34 100 Richest 12.21 1.23 76.64 4.62 0.48 1.05 2.58 1.19 100 Total 14.01 1.15 73.91 2.3 0.3 0.52 5.87 1.94 100 Source: BLSS 2007 (Block 1.3Q28) Table 4.10: Percentage distribution of respondents’ reasons for not consulting any healthcare providers, by consumption quintiles Consumption No No No transp- Does not quintiles need No time money ortation trust Other Total Poorest 48.25 13.53 2.23 20.92 3.22 11.9 100 Second 51.75 11.57 2.82 18.42 2.48 13 100 Middle 48.19 18.93 2.96 14.46 1.49 14 100 Fourth 57.1 13.73 1.86 10.1 0.47 16.8 100 Richest 65.98 9.91 0.74 4.87 1.54 17 100 Total 54.85 13.36 2.06 13.27 1.8 14.7 100 Source: BLSS 2007 (Block 1.3.29a) Childbirth and Maternity care 4.3.10. The number of women in the sample who gave birth in last four weeks of the survey was 1,150. Among them about 87.6 percent reported receiving prenatal care from a doctor or qualified nurse. Of these women who received professional prenatal care, 18.2 , 18.5, 19.5, 20.6, and 23.4 percent belonged to poorest, second, middle, fourth, and richest quintiles, respectively. The share of deliveries in hospitals was lowest (28.9 percent) for the poorest group, which went up with the consumption quintiles and stood at 66.4 percent for richest group. In contrast, the percentage of home deliveries without specialized assistance was highest (42 percent) in the lowest quintile group and lowest (6.5 percent) in the highest quintile. 30 Table 4.11: Percentage distribution of women who received prenatal care by consumption quintiles At home At home with At home without Consumption medical with specialized Quintiles Hospital Maternity assistance midwife assistance Other Total Poorest 28.85 4.47 6.92 15.52 42 2.22 100 Second 32.93 5.32 7.16 15.33 37.59 1.67 100 Middle 38.84 8.42 8.35 11.28 30.59 2.52 100 Fourth 52.69 9.42 6.95 10.19 17.16 3.59 100 Richest 66.38 18.03 1.11 6.75 6.54 1.19 100 Total 44.41 9.33 6 11.7 26.35 2.22 100 Source: BLSS 2007 (Block 1.3Q33) Access to health facilities 4.3.11. The mean time to reach a health facility for the poorest consumption quintile group was 115 minutes as opposed to 33 minutes for the richest quintile. The average time in the Western region was relatively lower at 54 minutes as opposed to 88 minutes in Eastern region (see table 4.12). This aspect of access to health services was also examined using Geographical Information System (GIS) tools, as discussed in the next section. Table 4.12: Mean time required (in minutes) to access services, by consumption quintiles and region Consumption Region Mean quintiles Mean Poorest 115 Western 54 Second 96 Central 72 Middle 73 Fourth 52 Eastern 88 Richest 33 Total 68 Total 68 Source: BLSS 2007 (Block 4Q14) Source: BLSS 2007 (Block 4Q14) 4.4 Health Infrastructure and Access to Health Services: Analysis using Geographic Information System (GIS) tools 4.4.1. The GIS-based study on access to health services30 involved the use of GIS tools to indicate the presence of health infrastructure across the country, in relation to population settlements, road access and health outcomes such as Infant Mortality Rate (IMR) & Maternal 30 MOH, 2011. Mapping of Health Infrastructure. Thimphu, Bhutan, 2011 31 Mortality Rate (MMR). The findings from the GIS maps focus on selected health indicators and their dependence on social and geographical determinants to health and through the use of different layers of these maps, attempt to describe the situation related to health services, equity and distribution of resources across districts. 4.4.2. Bhutan shares its border with two countries, namely China to the north and India to the south, east, and west. Much of Bhutan's terrain is extremely rugged, and the country is divided into three regions that are distinguished by their altitude. These include the Himalayan region in the north that consists of mountain peaks, the Central uplands on the slopes and valleys of the Himalayas which are divided by large rivers, and the third is the Duars plain that opens out towards India from the Himalayan foothills. All the administrative systems operate through the structures at district level which also manages the units below that level. 4.4.3. Health infrastructure, road connectivity and service uptake: About 30 hospitals, 168 Basic Health Units (BHUs) and more than 400 Outreach Centers (ORCs) and sub-centers are available across the country’s 20 districts. These centers are often strategically located in areas with high density of population and human settlements. The GIS map of health infrastructure overlaid with the layers for road network and population settlements clearly indicates this spatial distribution of health facilities and its correlation with the population settlements, and also how these health facilities are partially connected with roadways. It is evident that, health facilities in various parts of districts from south west, north-west and extreme east are poorly connected with road ways. Figure 4.10: GIS map showing health infrastructure, road network and population settlements in Bhutan Source: MOH, 2011. Mapping of Health Infrastructure. Thimphu, Bhutan, 2011 32 4.4.4. The number of outpatient visits varies across districts in Bhutan, and the location of the health facility may be a factor in determining the inflow of patients to some extent. The outpatient department (OPD) visits are higher in hospitals compared to BHU and ORCs. Interestingly, the pockets in eastern part of the country with higher density of health facilities and higher number of population settlements also have lower OPD visits than the pockets of western part with lesser number of population settlements and health facilities. Moreover, good road network connectivity, where applicable, further complements the patient flow to health facility. 4.4.5. The hospitals in the districts of Trongsa, Lhuentse, Mongar, Pema Gatshel and Riserboo hospitals from Trashigang have reported low OPD which could be due to reasons such as the number of settlements around these hospitals, connectivity of roads and transportation or other reasons, which need to be further studied. The fact that these districts also have a higher number of BHUs and ORCs and filter the patients, reducing the load at district hospitals could also be a reason for the lower OPD visits in these hospitals. (Figure 4.11) Figure 4.11: GIS map showing magnitude of OPD visits in health facilities in Bhutan (2010) Source: MOH, 2011. Mapping of Health Infrastructure. Thimphu, Bhutan, 2011 4.4.6. The patient flow for inpatient department (IPD) admissions is also likely to be influenced by similar factors such as settlement density, road network and availability of other health facilities in the vicinity. The patient flow among the health facilities located in close vicinity to towns and cities have higher OPD and IPD numbers. This could explain the higher patient flows in the western part of the country. The seasonal variation and inflow of patients during the local market, festivals etc., though not forming part of the current exercise, can also be explored for planning appropriate allocation of resources and man power during peak times of workload. 33 4.4.7. Emergency transportation services: Emergency transportation services (known as ‘112’ in accordance with the toll free number designated for the service) have recently been introduced in the country and were also mapped as part of the GIS exercise. These ambulance services attend to emergencies and connect the settlements located in remote areas with the nearest health facilities. Superimposing the layer of population settlements onto the location of ambulance services indicates that even now some of the low density settlements and areas with poor road network would have lesser access to ambulance services. However, the location and distribution of ambulances is dynamic and is likely to have been refined based on demand experience and utilization of the service. The key point is that a successful emergency transportation service requires the use of such GIS tools and past demand data for evidence based planning of the location of ambulances, in view of geographical requirements and demand for services for optimal use of resources. 4.4.8. Maternal Deaths: On a small denominator of the number of births in the country, the maternal mortality rate in the country is relatively high at about 200 per 100,000 population. The significance of road networks for maternal death does emerge from the spatial distribution of maternal deaths in the districts with poor access to the country’s road network. The comparative analysis of maternal deaths reported in 2009 and 2010 indicates occurrence of the cases in different geographies for these two years. Very few of the maternal deaths occurred in the same or nearby geographical location. It is also interesting to note that, the geographical pockets with towns and cities have lower maternal deaths than the areas which are remote and have poor access to road and health facility. The reported cases of maternal mortality are, expectedly, more in areas with higher density of settlements, but are also high in areas with health facilities appearing to be not too far on the map. This poses a question for further research to find out more on the same. Figure 4.12 shows this GIS map with maternal mortality for two calendar years overlaid on the layers for road network and health facilities. Figure 4.12: GIS map showing maternal deaths in 2009 and 2010 vis-à-vis health facilities and road network in Bhutan Source: MOH, 2011. Mapping of Health Infrastructure. Thimphu, Bhutan, 2011 34 4.4.9. Infant mortality: The fact that socio-economic factors have high influence on infant mortality is also depicted in Bhutan’s infant mortality map, when correlated with the maps indicating the poverty rate. IMR is clearly high in the districts with high poverty rate, in particular the eastern parts of the country. 4.4.10. The mortality rate among the children who are under 5 years is also closely correlated and Under-5 mortality is also more in the pockets with higher infant mortality rate. A correlation of IMR, U5MR, Stunting and Poverty rates in different districts of the country in Figure 4.13 shows very similar geographical patterns. Figure 4.13: GIS maps showing patterns of Infant Mortality, Under-5 Mortality, Stunting and Rural Poverty in the districts of Bhutan Source: MOH, 2011. Mapping of Health Infrastructure. Thimphu, Bhutan, 2011 4.4.11. Efforts are underway to make these GIS tools available online with updated data on a web-enabled portal, which would make easy access to GIS information possible for the planning and policy functions of the MOH. 35 4.5 Costing of Health Services at Different Levels of Health Facilities 4.5.1. As part of the ongoing efforts of the Royal Government of Bhutan in examining alternatives for a sustainability strategy and to improve the effectiveness and efficiency of its health spending, a study on costing of health services31 was undertaken by the MOH in select facilities representing 60 percent and 57 percent of the OPD and IPD caseload in the country, respectively, funded under the World Bank project. The study reviewed the costs and activities of health facilities located in nine districts covering a total of 13 facilities including the National Referral Hospital, the two Regional Referral Hospitals, four district hospitals and six basic health units (three Grade-1 and three Grade-2 BHUs). A list of the studied facilities is given in Table 4.13. Table 4.13: List of the facilities studied for the costing exercise in Bhutan, 2011 Facility Name Facility Type Dzongkhag/District JDWNRH National Referral Hospital Thimphu Mongar RRH Regional Referral Mongar Hospital Gelephu RRH Regional Referral Sarpang Hospital Paro DH District Hospital Paro Wangdi Choling District Hospital Bumthang DH Damphu DH District Hospital Tsirang Punakha DH District Hospital Punakha Gyelposhing BHU I Basic Health Unit, Grade Mongar I Bajo BHU I Basic Health Unit, Grade Wangdi Phodrang I Bali BHU I Basic Health Unit, Grade Haa I Genekha BHU II Basic Health Unit, Grade Thimphu II Mendelgang BHU Basic Health Unit, Grade Tsirang II II Thinleygang BHU Basic Health Unit, Grade Punakha II II Source: MOH, 2011. The cost of your healthcare- a costing of health services in Bhutan. Thimphu, Bhutan, 2011 31 MOH, 2011. The cost of your healthcare- a costing of health services in Bhutan. Thimphu, Bhutan, 2011 36 Objective of the Costing Study 4.5.2. The key objective of this study was to obtain information about the costs of delivering health services at different levels in Bhutan, in order to:  Inform MOH of the cost of delivering various types of services at different levels;  Increase cost-awareness and knowledge;  Inform the policy process and serve as an input to decision making Methodology 4.5.3. The study used a replicable methodology to assess the cost of resources used to provide services. It was based on data compiled for the period July 2009 to June 2010, thus also corresponding to expenditure during the financial year 2009/10. For the purpose of the study, a standard costing model was developed. The model classified health facility services into three types: overhead, intermediate and final. Overhead services included all administrative services. Intermediate services covered various support function like investigative services, kitchen, etc. The final type included outpatient and inpatient services. 4.5.4. As a first step of the study, all costs for the facility were collected – both recurrent and capital costs. The capital items had its cost annualized to an annual depreciation cost. Then all costs were assigned to the three types of cost centers. According to various rules the costs of overhead and intermediate services were allocated to final or direct service providing departments. Based on the final costs and activity of these departments, the unit costs for services could be calculated. Key findings and recommendations of the study 4.5.5. It is important to note that the results and facility comparisons of the study should be interpreted with caution. There are many factors influencing the final unit costs of facilities, with quality of data, demand for services, case-mix of patients, geographical features, staff mix etc. being only some of them. However, with due caution, the data from the study has been interpreted to bring forth valuable information for policymakers in the country.  Costs and cost-structures at facilities. The study gives a detailed insight into total costs, cost-structures and composition of costs of facilities at various levels.  Unit costs. The costs of OPD-visits, admissions in general as well as disease- specific groupings of admissions have been calculated and show – as would be expected – that services are generally more costly at higher levels. An OPD-visit is generally four times as costly at a referral hospital compared to a BHU and twice as costly compared to a district hospital. Inpatients unit costs show the same pattern. The average inpatient cost of a referral hospital is about three times that of a BHU and more than one-half time that of a district hospital. 37 Table 4.14: Average Unit Costs at different levels of health facilities in Bhutan, 2011 Referral Hosp District Hosp BHU I BHU II Unit Cost Unit Cost Unit Cost Unit Cost OUTPATIENT DEPARTMENT - Cost per OPD-visit 635 307 163 161 INPATIENT DEPARTMENT - Cost per Admission 17,354 10,116 5,657 NA - Cost per Bedday 2,795 NA NA NA Inpatients - medical - Cost per Admission 18,007 9,157 5,756 NA - Cost per Bedday 2,345 NA NA NA Inpatients - surgical & medical - Cost per Admission 17,170 10,667 5,581 NA - Cost per Bedday 2,963 NA NA NA Source: MOH, 2011. The cost of your healthcare- a costing of health services in Bhutan. Thimphu, Bhutan, 2011  Cost-effectiveness. The study presents evidence to the effect that providing services at the lowest effective level of service delivery (health facilities) is the most cost-effective way of providing services in the long run. It is a challenge to control demand for services and channel the patients to the right facilities through a well functioning referral system. If this is managed, it will in the longer run, result in optimal utilization of the resources of the health system.  Cost-efficiency. The study suggests that more optimal and cost efficient options for assigning manpower to facilities are available and if pursued, these could result in containing costs as well as reductions in waiting time for patients.  Data issues. In general a huge amount of health related data is available in Bhutan. The costing study reviewed a lot of these sources extensively and the review and analysis conducted as a part of the study especially uncovered two areas that would benefit greatly from renewed focus and improvements. These were the Bhutanese Health Management and Information System (BHMIS) and the data management of equipments and drugs at the Drugs, Vaccines and Equipment Division (DVED) of the MOH. 4.5.6. The analytical work points towards possibilities of efficiency gains and cost cutting in two areas: a revision – or development - of the referral system and the need to further analyze, and optimize productivity at the facilities. The study would also help the ministry in its pending decision on whether and how much to charge non-nationals for health services. It also suggested that cost effectiveness and cost efficiency of services need to be further areas of research. 38 4.6 Experience with Piloting Alternative Provider Payment Systems- Special Consultation Services (SCS) at Jigme Dorji Wangchuck National Referral Hospital (JDWNRH) 4.6.1. The pilot of Special Consultation Services (SCS) at JDWNRH was initiated in October 2010 by the JDWNRH and the Ministry of Health (MoH). The pilot had a stated goal to address the mismatch between available services in the hospital and the demand of the population, thereby aiming to provide choice to the patient, reducing the waiting time and increasing the doctor patient consultation time. 4.6.2. The SCS pilot envisaged a paid clinic for consulting the publicly employed doctors after the normal working hours of the national referral hospital. In this pilot, the patient has choice of the medical practitioner, and the fee charged from the patient is used to incentivize the personnel involved in rendering services in these additional hours. MOH’s initial plans contained in the proposal for the SCS pilot included a decision to assess the pilot of the SCS after one year of its implementation. This rapid assessment of the SCS serves two main objectives:  To review the performance of SCS vis-a-vis the originally stated objectives for the pilot and in light of international experiences with similar services  Based on this rapid assessment, provide policymakers with the requisite information and evidence base to facilitate decisions such as continuation of the SCS, replication thereof in other hospitals and modifications therein. Methodology and Scope: 4.6.3. The methodology of the rapid assessment included the following approaches and techniques:  Analysis of the secondary data (output and financial data from the JDWNRH pertaining to SCS as well as the normal hours, and other published reports)  A quick survey of JDWNRH patients attending the SCS and normal hours, using self- administered and surveyor-administrated questionnaires  Qualitative approaches including Focus group discussions (FGD) and Key informant interview (KII) with the stakeholders in JDWNRH and the MOH  Literature review to learn from international best practices and experiences from similar efforts elsewhere in the world  Study visit to hospitals in Thailand which have implemented similar off-hour paid services in their outpatient departments Key Findings and Observations: 4.6.4. Normal hour OPD Data: The annual OPD workload for all the major departments in the hospital (filter clinic32, medicine, ophthalmology, ENT, dermatology, surgery, gynecology, orthopedics, psychiatry, dental and pediatric) over the past 5 years (2007-2011) has been in the range of 300,000 visits per year, of which the highest number of cases were in 2010 at 347,705 patients and the lowest in 2008 at 276,112. Though the number of patients has increased in 2010 and 2011 by about 14-15 percent 32 Filter Clinic is a triage system at JDWNRH operated by health assistants (HA) 39 over the 2007 baseline, the numbers in the earlier 3 years remained more or less static, with a dip in 2008. Reasons for these trends in OPD service utilization need further research. 4.6.5. Utilization data from the Special Consultation Service: Total number of patients served for all the three services provided under SCS: medical consultations, medical certificates (medical certificates for employment, visa, driving licenses etc) and diagnostics and procedures, during October 2010-December 2011 was 33,980. Of the total, 71 percent (24,144) cases were for medical certification followed by doctor consultation at 19 percent (6,517) and diagnostics and procedures at 10 percent (3,319). 4.6.6. In case of certification services, 81.2 percent of cases were for basic medical certificates followed by driving license purposes at 16 percent and those for visa purposes were 2.8 percent. 4.6.7. For the doctor consultation service, the total number of patients for all departments taken together was 6,475 with the highest number of cases for medical consultation at 1,910 (29.5 percent) followed by surgery at 1,643 (25.4 percent) and gynecology at 1,050 (16.2 percent). Of the total doctor consultation, a predominant 94 percent of the cases were for specialist consultations and only six percent (n=401) were for general doctor consultations. In the diagnostics and procedures group, 54 percent of cases were for X-rays followed by scaling at 20 percent and ultrasound at 14.5 percent. Higher end diagnostics such as MRI and CT scan accounted for only 6.7 percent of all cases 4.6.8. Financials of the SCS pilot: Total revenue generated by the special consultation services from October 2010 till December 2011 was approximately Nu. 8,805,954. Of the total revenue, non-clinical services (i.e. all medical certificates) accounted for 58 percent of the total revenue followed by specialist consultation at 29 percent and diagnostics/procedures at 9 percent. Within the specialist consultations, the top three revenue generators were: medical, surgery and gynecology departments at Nu. 740,500, Nu. 667,000, and Nu. 394,000 respectively. 4.6.9. Total additional expenditure on providing the SCS service from October 2010 till December 2011 (not counting the costs of medicines dispensed, cost of lab reagents or any overhead costs of the hospital infrastructure) was Nu. 4,367,835 of which payments to doctors accounted for 51% (Nu. 2,223,235), payment to nurses and technician was 37% (Nu. 1,636,600) and administrative cost were about 12% (Nu. 508000). 4.6.10. Patient Experience and Feedback A survey of 345 patients attending services in the JDWNRH was undertaken as part of the study, of which 243 respondents had attended the normal hours services while there were 102 respondents who availed the special consultation services. Of these, approximately 62% of the total participants were from Thimphu and 38 percent were from other parts of the country. Almost all were Bhutanese (98.8 percent) with non-nationals accounting for only 1.2 percent of the studied patients. 40 4.6.11. Approximately 59 percent of the normal hour participants were aware of SCS, and 21 percent of these had already used the SCS services. Similarly, 55.4 percent of the SCS patients had used the normal hour services. Interestingly, and of relevance to policymakers, is that only 25.6 percent of the SCS patients – or about half of those who had also used normal hours- found the SCS care to be better compared to the normal hours. 4.6.12. Contrary to the expectation, cost was not stated to be a deciding factor for those not using the SCS: a high proportion of the normal hour respondents cited “normal hour was convenient” as a reason for not using the SCS while those not using it for reason of it being “expensive” was smaller at 15.4 percent. In case of fee structure, majority (88.7 percent) of SCS respondents seemed to have accepted the current fee structure, and 11% were in favor of changing it to a lower amount. 92 percent of the SCS respondents wanted the SCS service to continue and of that 81 percent wanted it to be extended to weekends and government holidays. However, as this is a self selected group, those finding this fee high may not be forming part of the SCS respondents and so these findings need to be interpreted with caution. 4.6.13. The indicated waiting time to see a doctor, for normal hour patients, was most commonly stated as 1-2 hours, while the commonest response from the SCS respondents was a waiting time less than 15 minutes, which was significantly lower. 4.6.14. In response to a question that may need to be continuously monitored in the future, 8.4 percent of the respondents attending the SCS indicated that they were told by the doctors to come for SCS. While it seems that this was advised as a mechanism to accommodate the convenience of the patient, the fine line between patient convenience and a subtle or even overt promotion of the SCS service does need to be treaded carefully. 4.6.15. For both the groups, the respondents’ perception on the attitude of the staff and doctors had the highest average ratings for patient satisfaction, though there was a somewhat higher score in case of SCS respondents. This difference was not statistically significant. This far, it does seem that the perceptions of service quality in normal hours are comparable to SCS, despite the clear differences in objective criteria such as waiting time. Hospital Staff Responses 4.6.16. During the Focus Group Discussion and interviews with hospital specialists and other staff, discontentment with regards to current remuneration packages, especially around the ceiling of Nu. 12000/month that was capped for their additional income, was evident. It was also insisted that participation in the service should be optional for the departments and for the individual specialists. Study Visit to Hospitals in Thailand offering similar services 4.6.17. A study team from the MOH and JDWNRH visited hospitals in Thailand to understand the Thai experience with similar evening, off-hour, paid clinics in public hospitals. It was observed by the team that the clinics in teaching hospitals of Thailand had been initiated several years back, and had served to prevent brain drain of public doctors to better paying private 41 facilities and promoted increased utilization of the facilities and infrastructure of the public hospitals. For the patients also, these clinics had a lower fee structure than private hospitals and yet offered similar convenience in terms of after-office timing, reduced waiting time and convenience. 4.7 Summing up 4.7.1. Bhutan has made remarkable progress in the health sector and has, arguably, achieved universal health coverage. The country has sustained investment in health over the years. A health financing and expenditure review conducted in 2009 demonstrated that spending by the government on health has been in the range of 7.4 to 11.4 percent of total government expenditure (Nagpal, 2009). In 2010-11, about nine percent of the government budget was allocated to health (MOF, 2012). As a proportion of GDP, public health spending alone (not counting private health expenditure) was 3.23 percent in 2010 (NHA 2009-10). These figures are amongst the highest in the South Asia region. Government is the predominant source of funds for the Bhutanese health system, with 88 percent of the total health expenditure in 2009/10 coming from public sources. 4.7.2. Main barriers to health care access are equity related, and include remoteness, transport difficulty and increasingly, the health needs of the urban and rural poor (including migrants). The costing study (MOH, 2011) shows that there are big productivity differences between facilities and that efficiency gains can be made by adapting staffing to actual workload. The key concerns regarding services are that the referral system is non-functional. This causes overload at the hospital level, especially at Jigme Dorji Wangchuck National Referral Hospital (JDWNRH). 4.7.3. There is a need for a strategic vision for sustainability in health financing in view of the epidemiological transition (dual burden of communicable and non-communicable diseases), the cost and delivery implications of providing health services in hard to reach areas, the rising expectations, and the economic transition. Besides, high rates of undernutrition pose a challenge to future economic productivity. Chapter 6 revisits some of these challenges faced by the Bhutanese health system, and puts forth certain options for further policy reform. 42 Chapter 5 EDUCATION 5.1 Organized Public Education 5.1.1. Education in Bhutan has changed profoundly since the early 1960s. Organized public education was introduced in 1961 with the First Five-Year Plan (1FYP). Prior to this, education was limited to the monasteries and the domain of the monks. Bhutan has developed a diversified system (Figure 5.1) that now includes over 650 schools and institutes, more than 180,000 students and 8759 teachers.33 The Constitution provides for the State to ensure free education to all children of school going age to Grade ten, and to make technical and professional education generally available, and higher education equally accessible to all on the basis of merit. Figure 5.1: Structure of Education Sector in Bhutan 33 These numbers include public and private sector institutions, but exclude centers for Early Childhood Care and Development (ECCD) and Non-formal Education (NFE). The latter mainly shares sites with primary schools or similar public premises. Total student numbers exceed 195,000 if ECCD (483 children in public centers, 1037 children in private centers) and NFE (12,968 learners are added. 43 5.1.2. Free basic education was initially established up to Grade 6,34 raised to Grade 8 in 1996, and to Grade 10 under the 9FYP (2002/03-2007/08, extended by one year). Additional efforts to improve education system delivery were implemented in the form of discrete and incremental policy enhancements through the various five year plans. The hallmark has been increased intake capacity of primary and secondary education and achievements in gender equity. More recently, efforts give prominence to improving the quality of education. 5.1.3. In 2010, the Royal Government launched a critical Tertiary Education Policy (“Policy”) as the first ever in Bhutan to focus on higher education as a sub-sector. The Policy reflects the Royal Government’s intensified interest in formal education when trying to assess the human capital available in the economy and, specifically, to realign the supply of skills with the needs of the emerging knowledge economy. The Policy provides for students to transition to tertiary level not only from upper secondary education but also from vocational training; albeit in both cases, on the basis of merit. Additionally, the Policy provides that continuing education is to be accessible to all irrespective of age, including people re-entering tertiary education after a period in employment. 5.2 Outcomes Accelerating access 5.2.1. By present international comparisons, Bhutan has a high level of participation in basic education. Enrollment rates in basic education are in the range of Europe and the Americas, surpassing the average level for South and West Asia (see Table 5.1). Bhutan’s enrollments in upper secondary and tertiary education resemble average regional rates for the Arab Countries. By contrast, Bhutan’s adult literacy rate (53 percent in 2005, latest year available) is lower than all regional averages, revealing the lingering vestige of the limited access to education prior to the 1960s. 5.2.2. Lying behind the steep enrollment growth that Bhutan has achieved in just fifty years (illustrated in Figure 5.2) are policies to accelerate school construction. These began at primary level in the 1980s and then broadened to include secondary schools (particularly under the 9FYP and 10FYP) to accommodate the “bulge” of primary graduates. In addition, for primary level there was a systematic effort to reduce the risk of spatial disparities – lack of access to schooling – due to the country’s hilly terrain and dispersed population. Primary Schools (PS) were complemented with “Community Primary Schools” (CPS) and “Extended Classrooms” (ECR). The CPS were built under the 9FYP in collaboration with distinct communities. The ECRs were built under the 10FYP35 to provide access within an hour’s walking distance in places where it is not feasible to establish a larger PS or CPS. Presently, just over a quarter of primary students are enrolled in CPS and less than five percent are in ECRs. Nearly 20 percent of primary students are in PS and the remaining 50 percent are in schools which offer primary and secondary education. 34 Free basic education means the Royal Government provides infrastructure, teachers and required teaching materials including textbooks, and most stationery for school administration. 35 An ECR offers primary education only up to Grade 3. The small cohort sizes often require a multi-grade teaching approach. ECRs are in rented premises, or structures such as out-reach clinics and community centers. ECRs are administratively and technically affiliated with the nearest CPS or PS. 44 With the exception of upper secondary and tertiary levels, the role of the private sector in service delivery in education is very limited. Table 5.1: Key Education Indicators: Bhutan and other Regions of the World Indicators Adult Primary Survival Secondary Secondary Tertiary Literacy NER to last 1 GER 2 GER GER (%) (%) year of (%) (%) (%) primary (%) Bhutan 53 95 931 882 49 183 Sub-Saharan Africa 62 76 70 41 27 6 South and West Asia 62 86 66 71 40 10 Central Asia 99 90 99 98 95 25 East Asia and Pacific 94 94 92 90 63 26 Arab States 72 84 97 83 53 20 South America and Caribbean 91 94 86 101 74 38 Central and Eastern Europe 98 93 97 92 84 60 North America and Western Europe 99 95 99 103 98 70 1 Survival to Grade 7 2 GER for Grades 7-8 is 98 percent; for Grades 9-10 it is 78 percent 3 Students studying in Bhutan as well as outside the country either on private funding or RGoB scholarships. Source : EFA Global Monitoring Report 2011 5.2.3. The last decade has seen marked growth in secondary and tertiary education enrollments. From the late 1990s, enrollments in lower and middle secondary more than doubled, and they increased by over 400 percent in higher level institutions (last two years of secondary, plus tertiary education; Table 5.2). At tertiary level, 6448 students were reported to be studying in Bhutan or abroad, on publicly funded scholarships in 2011. This represents 997 tertiary students per 100,000 habitants of the country.36 5.2.4. Bhutan launched its first University – the Royal University of Bhutan (RUB) – in 2003, bringing together ten tertiary institutes which had been established in the late 1960s and early 1970s. These affiliated Colleges are dispersed across the country, each with their distinctive specializations. The first private college – Royal Thimphu College – was established in 2009 and is affiliated with the RUB which confers all degrees. 36 This indicator is used with caution for international comparisons as enrollment in higher education at a given point in time does not reflect the accumulation of human capital formation on the level of higher education in society. When higher education is a relative late comer to a country, this indicator underestimates the contrast with countries where there is a deeper reservoir of human capital formed over many more years by higher education institutions. Furthermore, differences between countries in age distribution of the population – greater proportion of youth in less developed countries – tend to reduce the gap between advanced and less developed countries in the spread of higher education. 45 Figure 5.2: Historical Growth in Enrollments Growth of Students Enrolment (1996-2008) Enrolment at PP Enrolment in Primary Enrolment in Secondary 120,000 100,000 Number of students 80,000 60,000 40,000 20,000 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Years Source: General Statistics, 2008, PPD, MoE, RGoB Table 5.2: Bhutan Enrollment Growth in Secondary and Higher Level Institutions (1997-2010) Growth since Grade\year 1997 1998 1999 2000 2001 2002 2010 2002(%) Lower Secondary (Grades 7-8) 10206 11939 13563 14429 15128 15988 24904 56% Middle Secondary (Grades 9- 4534 5435 6560 8872 10508 11423 21818 91% 10) Higher level institutions (including Higher Secondary 1853 1934 2094 2434 2719 4477 18240 407% Grades 11-12) Source : Royal Government of Bhutan, Ministry of Education. Annual Statistical Reports. 5.2.5. Expanded access to primary and secondary education has been accompanied by a major advance in the participation of girls. Bhutan is on track to achieve the education Millennium Development Goals by the end of the 10FYP. Primary education (NER 95 percent37) has a 100 percent completion rate (2011). Gender parity is achieved at primary through middle secondary education (Grades 9-10). Females are 46 percent of enrollments in higher secondary and 37 percent in tertiary (beyond Grade 12). 37 On the basis of administrative and survey data, the Ministry of Education concludes that only 2 percent of the relevant age cohort is not in school. Approximately 3 percent are enrolled in the traditional monastic institutions, are abroad, or are enrolled in secondary education. 46 Internal efficiency 5.2.6. The internal efficiency of the education system has also improved. Ministry of Education data show the average study time for completing twelve years of education is 12.6 years (2011). The rate is marginally better in the upper grades (1.0 year per grade for Grades 8- 12) than in Grades PP-7 (1.1 year per grade). Drop-out rates have trended down in both the primary and secondary sub-sectors (Table 5.3). The transition rate from Grade 6 to lower secondary is high, at 97 percent in 2011 (compared to 94 percent in 2010 and 87 percent in 2006). The transition rate from lower to middle secondary is a stable 96 percent (2011), having been 95 and 94 percent in 2010 and 2006, respectively. The 64 percent transition rate from middle to higher secondary in 2011 exceeds the 10FYP target. Students who leave school after Grade 10 are considered “graduates” and “school leavers,” not drop-outs, since the system is structured to admit only the top 40 percent of Grade 10 graduates to public higher secondary schools. The remainder go on to study in private higher secondary schools, vocational training institutes (under the mandate of the Ministry of Labor and Human Relations), or end their studies to seek employment. 5.2.7. The repetition rate for primary has also improved over time (Table 5.3), but it still averages around six percent despite the Ministry of Education guidelines which favor non- detention. Patterns which have persisted over many years are that Grade 4 has the highest repetition rate and Grade 7 the highest drop-out rate. These systemic characteristics indicate some low levels of learning gain which have not been satisfactorily addressed over time. Moreover, the majority of students from PP to Grade 12 are over age for their grade. The Bhutan Multiple Indicator Survey (BMIS)38 conducted in 2010 shows, on average, 60 percent of primary school pupils are over age (30 percent are right age) and three-quarters of Grade 7-12 students are over age (20 percent are right age for grade). Ministry of Education statistics report an NER for lower, middle and higher secondary schools at 35, 27 and 17 percent, respectively. Table 5.3: Improvements in Internal Efficiency (%) of the Education System (2002-2011) Level of 2002/03 2005/06 2006/07 2009/10 2010/11 education\year Repetition rate Primary 11.6 7.0 6.7 6.4 5.8 Grade 7 13.4 8.6 7.8 6.1 6.1 Grade 8 4.3 3.1 4.9 3.9 3.3 Grade 9 13.4 6.7 6.1 5.7 4.2 Grade 10 1.8 0.4 1.1 0.5 0.8 Drop-out rate Primary 4.5 2.6 1.9 2.0 0.5 Grade 7 8.9 7.3 5.2 5.4 2.0 Grade 8 6.6 3.8 2.6 0.7 0.3 Grade 9 7.8 5.0 2.9 * 6.3  Ministry of Education statistics do not report any drop-outs. Source : Royal Government of Bhutan, Ministry of Education. Annual Statistical Reports. 38 BMIS is the customized version of the UNICEF Multiple Indicator Cluster Survey. 47 Equity of access 5.2.8. Nearing the end of the 10FYP, Bhutan is closing the equity gap in primary education, but for secondary education marked gaps remain. The 2010 BMIS reports the net attendance rate39 at primary was 85 percent for children of the poorest households and 97 percent for those from the richest. This is an equity gain over 2008, when the Poverty Analysis 40 showed 60 percent of the children of poor households and 80 percent of non-poor children attended primary school. The disparities emerge at secondary level, not only by wealth quintile but also by urban/rural dichotomy and geographic region. As presented in Table 5.4, in the poorest wealth index quintile, the percentage of secondary school age children attending secondary school is around 30 percent for both male and female, compared with 86 percent of males and nearly 80 percent of females from the richest wealth quintile. Secondary school attendance is lower in rural (46.7 percent) than in urban (75 percent) areas, and generally lower in the Eastern and Central regions than in the West. Overall, only around half of the children of secondary school age were attending secondary school or higher in 2010. A quarter was not attending school at all, while one in five was attending primary school when they should have been in secondary. 5.2.9. From another perspective, the 2007 Poverty Analysis analyzed school attendance among the 6-18 year age group against the rural poverty map at Geog level and found that Geogs where fewer children attended school tended to be those with comparatively higher poverty rates. 41 When the poverty incidence across Dzongkhags was compared with a “Human Development Index” (HDI) that combined gross enrollment rates for primary and secondary education, and adult literacy rates,42 the five Dzongkhags with the lowest HDI ratings (Dagana, Samtse, Mongar, Lhuentse, Zhemgang) were also those with the highest poverty incidence. Conversely, the top five Dzongkhags with the highest HDI ratings (Thimphu, Paro, Bumthang, Punakha, Haa) were those with the lowest incidence of poverty. 5.2.10. Poverty remains key among constraints on school attendance. These findings illustrate how, in a system where basic education is considered “free” up to Grade 10, the remaining costs to be borne by students can still militate against participation of the poorer segments of society. While there are no fees up to Grade 10, students are expected to pay for mandatory school uniforms, the school development fund, boarding fees where relevant, and for any extra private tutoring. Parents in urban schools pay for stationery. There are additional costs at the point a student transitions to lower secondary education. These include transport, since secondary schools draw students from a larger catchment area than primary. The difficulty disadvantaged households can have meeting these costs is a likely factor in the higher drop-out rate for Grade 7, noted earlier in this document. 39 Net attendance rate is drawn from household surveys and reflects whether respondent “attended” school at any time in the current year (rather than simply being registered). 40 Royal Government of Bhutan, National Statistics Bureau, Poverty Analysis Report 2008. 41 Royal Government of Bhutan, National Statistics Bureau and World Bank South Asia Region Economic Policy and Poverty. Small Area Estimation of Poverty in Rural Bhutan. August 2010. 42 Royal Government of Bhutan, Gross National Happiness Commission. Eleventh Round Table Meeting: Turning Vision into Reality: The Development Challenges Confronting Bhutan. Thimphu, 1-2 September 2011, page 44. 48 Table 5.4: Percentage of secondary school age children attending secondary school or higher, and percentage attending primary school, 2010 Male Female Total Net Net Net secondary Percent secondary Percent secondary Percent school attending school attending school attending attendance primary attendance primary attendance primary ratio school ratio school ratio school (adjusted) (adjusted) (adjusted) Wealth quintile Poorest 30.8 26.8 30.6 29.7 30.7 28.2 Second 38.7 26.8 38.6 29.0 38.6 27.8 Middle 48.3 23.8 53.2 25.3 50.8 24.6 Fourth 66.3 18.1 69.8 16.5 68.1 17.3 Richest 86.1 7.2 77.5 4.8 81.3 5.8 Residence Urban 76.7 12.9 73.5 10.1 75.0 11.4 Rural 45.4 23.4 48.0 24.3 46.7 23.9 Region Western 59.9 18.0 62.0 17.2 61.0 17.6 Central 49.9 21.0 52.8 22.0 51.4 21.5 Eastern 48.3 24.0 47.6 23.1 48.0 23.5 Source : Royal Government of Bhutan, National Statistics Bureau. BMIS 2010 5.2.11. The pressures education spending can cause for households is evident from the living standards surveys which show out-of-pocket spending to be higher – 26 percent on average – for education than for any other category of households’ non-food consumption (Table 5.5). Education represents a higher proportion of non-food consumption in rural households than in urban, even though the actual amount of spending in categories of primary and secondary school expenditures is lower in rural than in urban areas (Table 5.7).43 The household self-reported overall monthly per capita spending on education, which includes all costs from primary to tertiary, and any other forms of education, is markedly higher in the richest income quintile index than the poorest (see Table 5.6). That the richest spend over three times more than the next lowest quintile suggests that the ability to afford education, especially if it is fee paying and includes tertiary level studies, is likely beyond reach of the majority of Bhutanese households. This merits additional research. 43 These findings can also be viewed against results of a UNICEF survey in 2008 which found families bore expenses of Nu. 1,729 per pupil for six months of primary school. 49 Table 5.5: Structure of households’ non-food consumption by geographic area in Bhutan (%), 2007 Non-food Consumption Major Urban Rural Bhutan Item Tobacco and Doma 1.2 2.3 1.8 Clothing 10.7 10.8 10.8 Transportation and Communications 12.8 7.9 10.4 Recreation 3.0 1.3 2.2 Furnishing and Equipment 9.3 6.2 7.8 Miscellaneous 11.7 18.6 15.1 Educational Expenses 23.4 27.8 25.7 Health Expenses 1.5 3.5 2.5 Rent/Energy for Home/Household 26.3 21.5 24.0 Operations All Items 100.0 100.0 100.0 Estimated Value (ngultrum million)* 529 525 1,054 Source: Bhutan Living Standard Survey 2007 Table 5.6: Mean per capita monthly consumption expenditures, by household income quintile, 2007 (in Ngultrum) Mean per capita monthly Consumption quintiles expenditure on education (Nu/capita/month) 1: Poorest 95 2: Second 172 3: Middle 274 4: Fourth 381 5: Richest 1,224 All Bhutan 429 Source: Bhutan Living Standard Survey 2007 Table 5.7: Average per capita Education Expenses (Nu.) by Households with Currently School Going Children, across areas and sex, 2007 Urban Rural Bhutan Item Male Female Total Male Female Total Male Female Total School Development Fund 1,400 1,200 1,300 300 300 300 600 600 600 Boarding fees 200 200 200 100 100 100 200 100 100 Books, supplies 500 600 500 200 200 200 300 300 300 Private tutoring 100 100 100 - - - - - - Public transport to/from school 100 100 100 100 100 100 100 100 100 Other (uniform etc.) 1,200 1,100 1,200 1,200 1,100 1,200 1,200 1,100 1,200 Total 3,500 3,300 3,400 2,000 1,800 1,900 2,400 2,300 2,400 Source: Bhutan Living Standard Survey 2007 5.2.12. Participation in primary and secondary school is weaker in some remote areas. Of the five Dzongkhags identified (above) as lowest HDI and highest incidence of poverty, four (Dagana, Mongar, Lhuentse and Zhemgang) are among the eight Dzongkhags where the RGoB has made considerable effort to deliver services to the remotest areas. Over 75 percent of the schools (PP-Grade 12) in these eight Dzongkhags are located in areas the Ministry of Education 50 formally categorizes as “difficult”, “very remote” or “remote” (D/VR/R) By contrast, out of the five Dzongkhags with the highest HDI and lowest incidence of poverty, four (Haa, Paro, Punakha, Thimphu) have only 50 percent or less of their schools in D/VR/R areas. For the remaining Dzongkhags, the PER assessments reveal no clear pattern. 5.2.13. The 2010 BMIS data on primary school net attendance rate by Dzongkhag show that primary attendance is below the overall country mean in five of the eight Dzongkhags where 75 percent or more of the schools are situated in D/VR/R. However, the correlation is not always so pronounced, as the remaining three Dzongkhags (Dagana, Pemagatshel and Trashiyangtse) where 75 percent or more of schools are in D/VR/R areas report levels of primary school attendance above the country mean. Furthermore, attendance was below the mean in two Dzongkhag (Chukha and Punakha) where no more than 53 percent of schools were in D/VR/R areas. 5.2.14. One can conclude that beyond some general patterns across Dzongkhags, the dynamics are complex at the Dzongkhag level. This warrants more rigorous research into cost efficiencies of service provision, especially in remote areas, viewed against their educational outcomes. Urgent questions for 11FYP preparations include: Are Extended Classrooms and/or Community Primary Schools working better in some areas than others, and if so, what are the reasons? Since the Ministry of Education aims to consolidate scattered schools into a smaller number of larger schools, what impact would this have on unit costs and education outcomes, if a large share of primary students were shifted to schools requiring them to live in boarding facilities, away from their families? Quality of Education 5.2.15. Trends indicate that the quality of education lags considerably behind access gains. Successful learning by all is the mark of a quality education system and is an objective for the Royal Government’s education strategy. In Bhutan, there is widespread endorsement of the need to accelerate achievements in quality, and concomitantly strengthen the base to sustain results over time. Rates of learning are low; many students perform below the expected grade level in both basic and advanced academic skills and lack basic communication and analytical skills. 5.2.16. The institutional and political commitment to monitor quality levels and learning over time is evidenced by the National Education Assessment (NEA) that Bhutan piloted in 2003, as well as in the Royal Government’s conversion of the Bhutan Board of Examinations (BBE) into an autonomous “Bhutan Council for School Examinations and Assessment” (BCSEA) in July 2011. The BCSEA has a core mandate for secondary school examinations and certification, assessment and monitoring as well as training, research and publication in the areas of examination, assessment, and quality of learning. 5.2.17. To date, however, neither the institutional sources of data nor available survey research – which has consisted of occasional and largely unrelated efforts – constitute an adequate empirical base for a meta-analysis of learning levels, the driving factors behind them, and the implications for policy. This should improve over time as implementation of the NEA stabilizes. A national assessment is administered to a sample of students to assess achievements in a 51 curriculum area, to provide an estimate of the achievement in the education system as a whole at a particular age or grade level. Assessments are therefore designed to provide feedback to policy makers about system learning levels and related factors. The distinction between assessments and examinations is that the latter are intended to certify and select students for further studies. 5.2.18. The NEA in Bhutan uses curriculum-based and subject-oriented assessments, generally covering the national language (Dzongkha), English and numeracy. The 2003 NEA established a benchmark for Grade 6 in English and Mathematics. A second round of the NEA – expected to focus on Grades 6 and 10, in English and Mathematics – was put on hold until 2012, so results are not available to the PER. As for examinations, the Ministry of Education’s annual statistical reports provide a historical overview of results, for the exam at the end of Grade 10 (Bhutan Certificate of Secondary Education, BCSE) and Grade 12. Strictly speaking, these data are less useful for monitoring trends since the exam questions vary across the years, in part following changes and reforms in curricula. Furthermore, the thresholds set for minimum passing marks can also be different across the years, depending upon the system’s capacity to absorb the students selected. Despite the differences in objectives between assessments and examinations, since these provided some of the only data available in Bhutan concerning academic achievement, the PER draws on both sources in the following paragraphs. 5.2.19. Below the Grade 10 and 12 exams, other end-of-year exams are introduced in Grade 7, alongside continuous assessment. These latter two responsibilities are devolved to the schools,44 and results are not fully available at Ministry of Education level. For example, the ministry’s performance management system only now (2011) requires the BCSEA to monitor the top 10 and bottom 10 performers in Grade 6 examinations. 5.2.20. Existing data highlight themes which persist over time. A 2009 assessment of the quality of teaching and learning45 found that achievement levels declined as pupils moved from PP to Grade 4.46 Learning gains were slow, and the difficulties students had in mastering the curriculum within the prescribed time accumulated as students moved from grade to grade. The difficulties became the worst in Grade 4. This largely explains the high repetition level for Grade 4, as noted above. There is a parallel trend in the declining pass rates for the Grade 12 exam. The proportion of students who took and passed this exam fell steadily from 93 percent in 2002 to 85 percent in 2010. Since 92 percent of the total number of students who were enrolled in Grade 12 sat for the exam in 2010, this means only 77 percent out of the total number of students enrolled passed the exam. The cut-off point for a minimum satisfactory performance had been set at 40 44 The responsibility for primary exams was transferred to the schools in 1999; and responsibility for conducting the all Bhutan Grade 8 exam was transferred to schools in 2006. BCSEA (as the former BBE) developed modular-based assessments which are available for grades 5, 7, and 9 to guide teachers in formative assessments of learning in the classroom; these are used at the discretion of teachers and school heads. 45 World Bank. Findings from the Bhutan Learning Quality Survey. Discussion Paper Series, South Asia Human Development Unit, January 2009. 46 For Grade 2 learning ability, pupils mastered basic competencies expected of that grade within the context of the national curriculum. The average learning ability in Grade 4 was higher than the expected learning competency in Grade 2, but only by half a standard deviation, implying that it would take the average student in Grade 2 another year to reach the average competency for that grade. 52 percent that year.47 In 2010, of the 91 percent of Grade 10 students who appeared for the BCSE, 96 percent passed. This corresponds to 87 percent of the total number enrolled in Grade 10. The threshold for a passing score that year was set at only 35 percent. 5.2.21. In the new dynamics of globalization, Bhutan’s pursuit of a knowledge based economy implies that the national workforce will need strong skills in literacy, mathematics, science and technology. Evidence points to the need for improvements in these domains. The benchmark that the 2003 NEA established for Grade 6 numeracy was a mean score of 23.08 out of 50.48 Achievement was exceptionally poor in sub-tests for geometry and algebra. In geometry, 38 percent scored “0” out of 9 marks; another 40 percent scored only 1 mark out of 9. In algebra, 31 percent of students scored “0”out of 5 marks; another 36 percent scored only 1 mark out of 5. Nine years later, a certain parallel emerges in the 2012 BCSE. BCSEA’s draft report notes the 2012 BCSE results continue the declining trend in achievement in business mathematics over the years. In this domain, the majority of questions require higher order reasoning skills. 5.2.22. In science education, Bhutan spends considerable financing on laboratories, science equipment and in-service training for teachers and laboratory assistants. Bhutan was well ahead in this respect in 2008, as shown in a cross-national study49 that indicated Bhutan achieved a similar number of hours in the teaching of science as in Thailand and Japan (although less time than Botswana, Chile and Kenya). The study observed, however, that the number of Grade 12 students in Bhutan who sat for the final examination in science in 2006 fell short of the participation rates of Japan, Malaysia, and Korea. In Bhutan, of 4038 sitting for the final examination in upper secondary school, only 835 opted for the physics and chemistry paper; even fewer (545) took biology. The mean scores for physics, chemistry and biology were 58.54 percent, 60.37 percent and 50.53 percent, respectively, compared with the higher mean score of 64.42 in commerce. 5.2.23. Value in the context of the labor market is another indicator of the quality of education. Several recent surveys in Bhutan have highlighted a concern over skills shortages, to be able to respond adequately to needs of the local labor market and especially to the rapid growth of the private sector. On average, thirteen percent of Bhutanese firms report that they lack appropriate skills, and this proportion increases by firm size. 5.2.24. In terms of overall employment of secondary and university graduates, however, the trend is positive. The unemployment rate for secondary and university graduates, combined, was approximately 12 percent in 2002. By 2011, rates had fallen to eight percent for secondary graduates and five percent for university graduates. Both of these rates are significantly below the average overall 10 percent unemployment rate among the 15-24 age group. 47 For comparison purposes, a minimum score of 60 percent is needed for admission to many tertiary institutes abroad, as indicated in the requirements set out for scholarships to Sri Lanka and Cuba, open to Bhutanese students. 48 By comparison, the mean score for literacy was 26 out of 50. 49 David Johnson, Ann Childs, Kiran Ramachandran, Wangpo Tenzin (2008) A needs assessment of science education in Bhutan. UNESCO. 53 5.2.25. Key factors for quality of education outcomes. Evidence from the 2009 study mentioned above points to within-school characteristics, and particularly teachers, as having the greatest impact on students’ academic performance. Teacher quality – and consequently the quality of education at school level – was correlated with a maximum of approximately 50 percent of the variation in students’ test scores, even after controlling for child and family background characteristics.50 In situations where teachers had adequate training, this had a large and significantly positive impact on students’ scores in the grades targeted for the 2009 study (Grades 2 and 4). For Grade 2, female teachers especially had a large and significantly positive impact on pupils’ test scores. The study concluded that once school quality was taken into account, student background characteristics (household resources, including parental literacy) were only marginally correlated with any observed variations in learning achievement across the students surveyed. Apart from this study, but in a similar vein, it may be pertinent to note that the 2003 NEA showed urban students outperformed their peers in semi-urban, rural and remote areas but that this pattern is not as apparent in the 2012 BCSE. For the latter, the Eastern parts of Bhutan showed the strongest performance; that is, stronger than in the highest urban area – Thimphu. While this deserves more systematic study aided by data which are more closely comparable and detailed, the above findings suggest that variations between schools in terms of their internal characteristics, continue to be the most significant influences on students’ academic performance. 5.2.26. Teacher policy. For an expenditure review, the situation of teachers is a key investment consideration not only because of teachers’ role in ensuring the quality of education, but also because their salaries comprise a large proportion of sector expenditures. In terms of teachers’ capabilities, investments to date have brought the proportion of Bhutanese teachers trained to 87 percent of the entire work force: 57 percent have Bachelors Degrees, 6 percent Post-Graduate Diplomas, and less than 25 percent have only Grade 10 or Grade 12 certificates. Enrollments at Bhutan’s two Colleges of Education are increasing: the 1506 enrolled in 2011 is a 34 percent increase over 2004. In addition, from 2010, an upgraded entrance exam given collaboratively by the Ministry of Education and the Royal Civil Service Commission is required of future teachers. Distance education is also organized to upgrade existing teachers’ qualifications, especially focusing on those who entered service after Grades 10 and 12. As Bhutan moves towards self- sufficiency, the dependence on (higher cost) expatriate teachers has decreased. Expatriate teachers are currently around eight percent of the teaching force, and are especially needed at secondary level, in mathematics and sciences.51 5.2.27. Teacher recruitment is merit-based, and evidence points to the on-time disbursement of teacher salaries 7-9 months during the year. Salaries are differentiated according to subject, education level and years of service, but not by geographic area or teacher performance. Numerous surveys report that a very low 25 percent of teachers are satisfied with their profession. The RGoB recognizes the need for uplifting the morale and capability of teachers through stronger incentives and improved professional development opportunities. The Ministry of Education proposes an 11FYP target to lift teacher satisfaction level to 75 percent of all teachers and appears mainly to favor in-service support to achieve this goal. 50 The study commented that this pattern was not seen in neighboring India or Pakistan. 51 The RGoB has arrangements with India, Japan (JICA) and Canada to bring in teachers on a contract basis. 54 5.2.28. The Ministry of Education initiated “School Level Monitoring and Support Services” in 2011 to encourage Department and School Heads to give low-performing teachers and schools mentoring assistance from Department Heads and School Heads. However, schools – and to some extent even the Dzongkhag authorities – still have limited autonomy and resources to undertake projects or innovations of their own design to improve teaching and learning. The Ministry seeks to revamp in-service training under a more comprehensive and coherent National Master Plan. A proposal was developed by an internationally reputed firm in 2011, but was not costed at the time of this PER. 5.2.29. Quality at tertiary level. The management of tertiary education as a sub-sector is new in Bhutan and is under dynamic development. Upon its establishment in 2003, the Royal University of Bhutan (RUB) was mandated to cover “all education of Level 5 and above” (ICSED52, 1997) in the country. A common academic framework for all tertiary education programs was established for quality assurance, and all educational programs deemed “tertiary” have been accredited and qualifications recognized by RUB53. There is traditionally, however, a large number of Bhutanese students pursuing higher education studies outside the country, which do not fall within the RUB quality framework. The Tertiary Education Policy (2010) has called for the establishment of a more comprehensive quality assurance and accreditation framework to take into account all forms of home-based programs of tertiary education as well as those of foreign institutions. In addition, Cabinet approved (2010) a policy framework linking technical and vocational education and training to tertiary education programs. A Bhutan Qualifications Framework is to be developed by the Ministry of Education, Ministry of Labor and Human Resources and the Royal University of Bhutan to set national standards and harmonize all existing frameworks. Finally, Bhutan faces a much larger challenge of linking in the private sector along with quality assurance in the tertiary sub-sector. Over 2011/12, Druk Holding & Investments has formed a public private partnership with a private developer to establish an ambitious “Bhutan Education City” campus infrastructure, utilizing a turnkey design, build, finance, own, operate and transfer model. The Education City will predominantly focus on university level studies and research, but foresees some delivery at basic education level as well. 5.2.30. With the Tertiary Education Policy, the RGoB is also making financing reforms in favor of greater cost sharing. As mentioned previously, the RUB must now generate a larger portion of its own revenue. In addition, the RUB and the Ministry of Education are working out a new scheme for student financing. A higher proportion of students will be expected to pay full tuition, and a student loan mechanism is to be introduced in addition to government scholarship grants which to date have been the main source of funding students. 5.3 Education Sector Financing 5.3.1. Overall, Bhutan is making important efforts to finance its education system. A high budget allocation to education, coupled with high rates of economic growth over 1980-2010 enabled the Royal Government to absorb current cost implications of the enrollment boom. External support, particularly from the World Bank (starting under the 6FYP, FY1988-1992) and 52 ICSED = International Standard of Classirication of Education, associated with the United Nations. 53 Prior to the establishment of RUB, almost all tertiary education institutions had arrangements to offer their own awards, and few institutions were working closely with external partners. 55 the Government of India (ongoing), has helped meet capital costs of school expansion and quality enhancements including extensive curricular reform. 5.3.2. From 1960, the share of the government budget allocated to education was consistently kept at around 10 percent. From 2000, government education spending generally rose faster on an annual basis than the annual increase in economic growth. The elasticity of government education spending to GDP was approximately 1.35 between 1999 and 2008, implying that for this particular period, on average every one percent increase in GDP was associated with a 1.35 percent increase in government education spending (Figure 5.3). Figure 5.3: Elasticity of Government Education Spending to Income, 1999-2008 5.3.3. During the 10FYP, 16 to 17 percent of public domestic expenditures have been allocated annually to education.54 These spending levels are quite high compared with other countries in South Asia with the exception of the Maldives. Table 5.8 summarizes the 2009/10 expenditure levels in education in Bhutan, also indicating the percentage of gross domestic product (GDP) represented by the respective levels. Total education expenditures for 2009/10 were 3153.6 million Nu. Two-thirds of capital expenditures were financed by development partners. Education total share of GDP was 7.3 percent counting current and capital spending. Education spending only counting domestic resources is estimated at 5.7 percent of GDP. 5.3.4. Expenditure by level of education. It is estimated that primary education currently represents about forty percent of government expenditures on education. Lower and middle secondary education represent 20 percent, higher secondary 4 percent, and tertiary education 54 This is an increase from 12.9 percent in 2000 and an increase as well over the 15 percent that the 10FYP had initially earmarked for education. 56 (i.e. Royal University of Bhutan) is 14 percent of government expenditures on education (Table 5.9). Table 5.8: Education Expenditures by Level of Education, 2009/10 Actual Actual Actual Actual Actual Actual Total Breakdown by Level of Education Current Capital Total Current Capital Total % of (millions of Nu, US$; % GDP) (Nu) (Nu) (Nu) (US$) (US$) (US$) GDP1 TOTAL Educational Expenditures (excluding vocational training) 3153.62 1712.0 4865.6 63.1 34.2 97.3 7.3 of which PRIMARY 1404.0 487.0 1894.1 28.1 9.7 37.9 2.8 of which Lower and Middle Secondary School 782.3 188.9 973.8 15.6 3.8 19.5 1.5 of which LSS(Grade VII-X) 365.9 100.1 467.2 7.3 2.0 9.3 of which MSS (Grade XI-XII) 416.3 88.7 506.6 8.3 1.8 10.1 of which Higher Secondary School (Grades XI-XII) 159.2 25.1 184.9 3.2 0.5 3.7 0.3 of which Non-formal Education 0.0 0.0 1.0 0.0 0.0 0.0 0.0 of which Higher Education3 323.2 381.6 702.5 6.5 7.6 14.0 1.1 of which Central Administration 484.0 629.5 1109.4 9.7 12.6 22.2 1.7 1 Includes current + capital expenditures; GDP 2010 2 Numbers may not add up due to rounding. Also, NFE expenditures included partly under spending for Primary, partly under Central Administration. 3 Royal University of Bhutan expenditures only, as Royal Thimphu College in private sector is not part of RGoB budget. Source: Ministry of Finance, Dept. Public Accounts. Annual Financial Statement for year ended June 30, 2010; and World Bank calculations of expenditures for Primary, LSS, MSS, HSS on basis of Ministry of Education data on percent allocations of hours taught per level of education, broken down by MOE/MOF (subtracting current cost for 700 NFE teachers included in primary and central administration expenditures) 57 Table 5.9: Breakdown by Level of Education (%) Actual Actual Actual Breakdown by Level: Current Capital Total (% of Total Educational Expenditures) (%) (%) (%) TOTAL Educational Expenditures 100% 100% 100% of which Primary School (Grade PP-VI) 45% 28% 39% of which Lower and Middle Secondary School (Grade 7-10) 25% 11% 20% of which LSS (Grades 7-8) 12% 6% 10% of which MSS (Grades 9-10) 13% 5% 10% of which Higher Secondary School (Grades 11-12) 5% 1% 4% 1 of which Non-Formal Education 0% 0% 0% 2 of which Higher Education 10% 22% 14% of which Central Administration 15% 37% 23% 1 Expenditures included partly under primary education, partly under central administration 2 Royal University of Bhutan only. Source: World Bank estimates based on Ministry of Finance expenditure reports and Ministry of Education data on percentage of hours taught at primary and each level of secondary education. 5.3.5. Per capita expenditures. Based on the PER calculations, current expenditure per public student in 2009/10 ranged from US$260 at primary to US$1,162 at higher education (Table 5.10). Per-student spending on secondary as a ratio of per-student spending on primary students is 1.3. Per student expenditure on tertiary students as a ratio of per-student spending on secondary is 3.4. These ratios are in the range of what research has shown to be typical per- student cost patterns in countries succeeding in expanding secondary enrollment.55 However, the ratio between secondary and primary unit costs, at 1.3, implies that spending may not be adequate for the magnitude of secondary level expansion envisaged by Bhutan, while the country also needs to address important quality concerns. 5.3.6. Within the primary level, the PER’s assessment of allocation of instructional hours at school level (Ministry of Education database) suggests that unit costs per instructional hour are the lowest when primary education is delivered through only the primary school entities (including Community Primary Schools and Extended Classrooms). The unit costs to deliver primary in entities called “secondary schools” are successively higher, progressing from entities called “lower secondary” schools, to “middle secondary” and then “higher secondary” schools, respectively. The level of teacher salary may be the main driver of these differences in unit expenditure on primary instruction due to the higher teacher grades in secondary. 5.3.7. Expressed in terms of per capita GDP, primary school teacher salaries are on average two times the GDP per capita, compared with 2.6 for lower and middle secondary school teachers, and 2.7 times the GDP per capita for higher secondary school teachers. By some international comparisons, these levels are low. For example, the “Indicative Framework” that was initiated by 55 World Bank. Expanding Opportunities and Building Competencies for Young People: A New Agenda for Secondary Education. 2005. 58 the Education For All Fast Track Initiative noted in 2009 that the average annual salary of primary school teachers as a multiple of GDP per capita was 3.5 for many successful countries. Table 5.10: Numbers of Students PP-Higher Education and per student costs, 2009/10 Current Current Current education education education expenditure expenditure expenditure per public Public Private Total per public per public student as Students Students Students student student % of GDP (#) (#) (#) (Nu) (US$) per capita Primary 108,048 3,125 111,173 12,994 $ 260 0.13 Lower & Middle Secondary 48,452 382 48,834 16,145 $ 323 0.16 Higher Secondary 7,231 6,709 13,940 22,013 $ 440 0.21 * Average of LSS, MSS, HSS - - - 16,907 $ 338 0.16 Higher Education 5,562 886 6,448 58,106 $ 1,162 0.56 * * Total students 169,293 11,102 180,395 15,763 $ 315 0.15* * Weighted average Source: World Bank estimates based on Ministry of Finance expenditure reports; Ministry of Education Annual Statistics 2010. 5.3.8. Current expenditures (salaries, materials, maintenance) comprise approximately 65 percent of total education expenditures overall, sector wide. In Bhutan salaries are centrally determined and paid, but the financing for primary and secondary education teachers is transferred to the Dzongkhag. In the Dzongkhag education budgets, an average 68 percent is spent on salaries. Personnel costs take the largest share of current spending on education in Bhutan, but the proportion is relatively low compared to other South Asian countries. In Bangladesh, for example, 98 percent of current spending in education is devoted to salaries. 5.3.9. Education expenditures by major functional category are summarized in Table 5.11. Table 5.11: MOF Expenditure Data (actuals, current & capital) 2009/10 2009/2010 Current Capital Total Current Capital Total (millions) – Actuals Nu Nu Nu US$ US$ US$ Dzongkhag Education Budgets 2333.4 702.3 3035.7 46.7 14.0 60.7 only with salary 1585.4 - - 31.7 - - non-salary 748.0 - - 15.0 - - RUB 328.4 381.1 709.5 6.6 7.6 14.2 Royal Education Council 12.1 20.4 32.5 0.2 0.4 0.7 Education Services 260.9 28.6 289.5 5.2 0.6 5.8 MOE 218.8 579.7 798.5 4.4 11.6 16.0 TOTAL EDUCATION Actuals 3153.6 1712.0 4865.6 63.1 34.2 97.3 Source: Ministry of Finance budget execution data by Dzongkhag; Ministry of Finance , Public Accounts: Annual Financial Statement for year ended June 30, 2010. 59 5.3.10. The education sector on average represents almost half of the Dzongkhag budget. This varies greatly between only about a third in Gasa and close to two-thirds in Chukha. The education budget is highly correlated with the number of students enrolled. The range across Dzongkhags is illustrated in Figure 5.4, with budget execution results for 2010/11. Figure 5.4: Budget execution results for district and education. 2010/11 (Nu. millions) 900 750 600 450 300 150 district total 0 of which education Haa Trongsa Punakha Sarpang Chukha Gasa Thimphu Zhemgang Tsirang Pemagatshel Dagana Paro Trashigang Samtse Mongar Bumthang Lhuentse Trashiyangtse Wangdue Phodrang Sandrup Jongkhar Source: Ministry of Finance budget execution data, by Dzongkhag. 5.3.11. For the Dzongkhag education budgets, there is a negative correlation between number of students enrolled and expenditure (current + capital) per student. The fewer the number of students, the larger the expenditure per student tended to be (Figure 5.5). Figure 5.5: Expenditure per student FY 2010/11 (current + capital, Nu. ‘000) 50 40 30 20 10 0 Source: Ministry of Finance and Ministry of Education Annual Statistics 2011. 5.3.12. Analysis of the instructional costs by remoteness of school location confirmed that unit costs are generally higher in more remote regions. The highest expenditure per student is seen in 60 the Dzongkhags which have over 80 percent of the total number of their schools in “difficult,” “very remote,” or “remote” (D/VR/R) areas, the three most remote categories used in the Ministry of Education data. In Figure 5.5, this is the group at the right of the continuum, from Mongar to Gasa.56 This contrasts with the six Dzongkhags (in Figure 5.5, at the left of the continuum, from Samtse to Chukha) which have the lowest expenditure per student, and only 50 percent of their schools in D/VR/R areas. For the remaining six Dzongkhags (Sandrup Jongkhar to Bumthang) at the midrange of expenditures per student, an average 60 percent of schools are in D/VR/R areas, but this average masks great diversity. A low 45 percent of schools in Haa are in D/VR/R areas compared with Wangdue Phodrang that has a high 75 percent in D/VR/R areas. 5.3.13. Dzongkhag budget execution rates suggest a need for improvement in efficiencies of spending on capital items. For the budget years reviewed by the PER, the Dzongkhag budget execution rates for capital expenditures were an average 72 percent in the education portion of the budget and 68 percent for the Dzongkhag (capital) budget as a whole. For current expenditures, actual compared with appropriations show a higher execution rate of 97 percent for both the overall Dzongkhag budget and the education portion. The lower execution rates on the capital side are mainly procurement-related, connected with capacity constraints in managing on- time completion of civil works. 5.3.14. In conclusion, one of the main findings of the PER in the education sector is that despite some general patterns across Dzongkhags, at that level, realities are complex. There is a need for more systematic analysis of local conditions, analogous to that carried out in the health section of the PER. As Bhutan moves into consolidating the education sector under the 11FYP, each Dzongkhag will need a carefully crafted plan. There would be a need to tailor the plans, including the financial projections, closely around the distinctive challenges and opportunities of the respective Dzongkhag. 5.3.15. In doing so, this may enhance effectiveness of education sector investments longer term, for example, to take adequate account of projected population changes in the school age cohorts. In-country migration is expected to contribute to over a four percent increase in the urban population by 2015, whereas rural areas are expected to see an increase of only up to one percent. Highest growth is foreseen for Dagana, Samdrup Jongkhar, Sarpang, Thimphu and Wangdue. The lowest growth is expected for Haa, Lhuentse, Pemagatshel, Trashigang, Tsirang and Zhemgang. These population projections should help to inform decisions about building new, and repairing existing, education infrastructure; whether and how to close some schools in the remotest areas and how much to invest in enlarging schools to take in students from a broader catchment area, as well as where to promote a greater share of service delivery to be provided by the private sector. 5.4 Preparing the 11FYP 5.4.1. The characteristics and challenges of educational expenditure are distinctive at each level of the educational system. The following briefly identifies the most critical challenges for each level and implications for public financing. This section concludes with an analysis of the financial implications of various policy objectives for developments under the 11FYP and 56 Zhemgang, Lhuentse, Trashiyangtse and Gasa each has over 85 percent of their schools in such areas. 61 suggests there is an overall risk that plans in the education sector will exceed resources. These pressures accentuate the likelihood that the RGoB will need to manage trade-offs in financing within and across sectors, if total resource availability will not be sufficient to meet all needs. There may also be consequences for the country’s ability to sustain progress to date on the education MDGs. 5.4.2. Several patterns emerge:  In primary education, the near-universal enrollment and high completion rate demonstrate that RGoB financing has been effective. Cross-country comparisons (Table 5.12) with South Asian neighbors further indicate that Bhutan’s higher level of GDP for education has yielded stronger outcomes in terms of primary net enrollment rate, persistence to last grade of primary, and pupil:teacher ratio in primary. Table 5.12: Bhutan compared with other South Asian Countries: Primary Education Indicators, 2009/10 Education Primary net Persistence to last Primary Country % of GDP enrollment rate grade of primary pupil:teacher ratio (%) (%) Bhutan 7.3 95 100 28 Maldives 8.4 96 94 13 Nepal 3.7 n.a. 62 33 India 3.2 90 66 n.a. Pakistan 2.9 66 n.a. 41 Bangladesh 2.2 85 55 44 Source: EFA Global Monitoring Report, 2011 Data are for latest year available 2000-2009/10.  At a minimum, Bhutan will need to consolidate these gains under the 11FYP, keeping up with demographic changes. There is high access, but this is not effective coverage in terms of learning gains. Quality concerns call for a more effective strategy to improve performance of the teaching-learning process. This has policy implications, and depending upon the options chosen, will likely have further cost implications as well. The main factors are pedagogical support for teachers, promoting incentives for performance in the public sector, and arriving at a service delivery mechanism that adequately sustains high enrollment levels of the relevant cohorts from the more geographically remote areas.  Among the main facilitating factors instrumental for enrolling and retaining children in school, especially in rural areas, are location of schools closer to communities (CPS, ECR) and provision of day meals in schools. The impact of these incentives needs more careful scrutiny, in view of the RGoB’s interest in consolidating schools, since this would mean a greater proportion of students would be required to live in boarding facilities even for primary education. Would this act as a disincentive and potentially risk Bhutan’s losing momentum or even back-sliding on the education MDGs? In view of withdrawal of the World Food Program-financed school feeding program, what is the potential impact on 62 incentives for families to send their children to school, and would this justify domestic resources to fill the WFP gap, in order to sustain enrollments and retention at MDG target levels?  At the same time, generally low student-teacher ratios across the country suggest there may be space for a policy adjustment to consolidate schools in some urban and semi-urban areas. This would be compatible with the Ministry of Education’s interest to consolidate schools to create a critical mass of teachers for more cost-effective and intensive delivery of in-service support.  The proposed Master Plan for In-service training of Teachers can be expected to increase resource requirements. These requirements need to be quantified (estimations of cost).  At secondary level, the differences in participation between urban and rural (and differences among wealth quintiles) indicate that the level ground that primary education is providing is far from evolving toward equitable participation in secondary school.  There is a need to free up fiscal space for greater attention to secondary. The targets in view for the 11FYP are ambitious, calling for the secondary education completion rate to increase to 95 percent (from 75 percent) and the transition rate from middle to higher secondary education to increase to 90 percent (from 68 for males and 61 percent for females, respectively). Increases of this magnitude mean higher costs to cover student places – in boarding facilities as well as in classrooms. Improving quality, also to make secondary education more relevant and attractive to a larger share of the secondary school age population, will further increase costs. These are major systemic changes and the measures to be taken need to be more clearly articulated.  The new policy for linking technical and vocational education and training to tertiary level programs also implies some curricular reform in middle and higher secondary education. The relevant clusters of skills will need to be created and consolidated among the program options in secondary schools to facilitate seamless transfer to tertiary. Depending upon the policy options to be chosen, there may also be implications for the vocational training institutes (currently under the mandate of MoLHR) which presently offer specific occupational training, and for which there is limited interest among the youth.  Financial planning and management at Dzongkhag level needs greater attention to discern whether strategies and financing plans more tailored to the respective challenges and opportunities of each Dzongkhag would promote efficiency and improve outcomes. Proactivity will be important to adjust service delivery in geographic regions where there are falling student numbers and, where feasible, to encourage a greater role of the private sector in 63 service delivery. With persistent poverty in rural areas, however, private schools are not sustainable measures to improve enrollment.  Underlying causes for the high dissatisfaction of teachers need to be addressed. The civil service salary structure presents some negative incentives to quality performance in that teachers commonly complain about their higher workload (lesson preparations, marking tests, and other school level administrative obligations) compared to civil servants at the same pay grade in other sectors. Other countries facing similar challenges in teachers’ motivation have considered introducing bonus pay for top performance. If on the other hand, the salary level is low, piloting an across-the-board increase (say, ten percent) might be an option to explore57.  The expansion plans for tertiary education are ambitious for the 11FYP. Student intake is to increase to 15,000 (from just over 5,000 currently). The student numbers alone will increase total public expenditures even though at least a third of tertiary students are expected to be full-fee paying. At the same time, all constituent colleges of the Royal University of Bhutan (RUB) are planning additional new academic programs, in many cases more than doubling the current number of programs on offer, and such expansion will also increase costs.  The growing role of private institutions and financing in the tertiary sub- sector, notably with the new Education City (under preparation) will become an important factor in expanding coverage. This may not over burden public expenditures, but it raises new issues of quality and relevance. As it will take the Education City some years to become operational, the RUB is the main reference point for the quality assurance and accreditation framework. It is critical that the expansion plans of its constituent colleges be harmonized with, and not operate in parallel to, this framework.  A vigorous equity intervention will be critical to ensure students from poor families are not disproportionately affected by the new tuition fee policy at tertiary level. It would be important to conduct a sensitivity analysis of elasticities of household ability to pay, as the RGoB introduces the new system. Such analysis may also help gauge longer term public fiscal impact of the new student loan scheme that is coming on stream. Simulating financial implications of 11 FYP 5.4.3. The education sector financing strategy will need to accommodate a greatly changed landscape in terms of availability and management of resources. Development partners are 57 There is some precedent for salary enhancement, at least for a sub-set of the teaching corps. A policy that is in practice accords teachers an allowance of 10 to 15 percent, based on whether the teachers have worked more than five and ten years, respectively. The criteria for receiving this allowance include placement (of teachers) in remote and rural areas of the country where there is limited access to roads, electricity, housing and social amenities. 64 scaling down or ending support to education as Bhutan approaches middle income status. Also in the near term, the absolute amounts available to education may only grow as fast as government revenues, and these are likely to be impacted near-term by the projected deficit. Momentum for development will remain strong, as the policy enhancements in the previous sections have underscored, and a more sustainable education system financing will be needed to meet recurrent costs associated with ever expanding service delivery. 5.4.4. Decisions the RGoB may face in terms of managing trade-offs in financing: The PER team developed a model to simulate financial implications of various development objectives stipulated in Bhutan’s education strategy, with a particular focus on the main aims for the 11FYP which have been articulated to date. The simulation model consists of macroeconomic variables plus primary, secondary, and higher education modules. Each sub-sector module contains enrollment, teacher, and other recurrent expenditure variables. At this stage, capital expenditure was not included. The application of the model is still limited because of difficulties in collecting consistent expenditure data. Projections depend on a number of assumptions, including macroeconomic growth rates and allocation of public expenditure to the education sector, enrollment growth rate, and future unit cost per student. The model is also predicated on an optimistic assumption of domestically generated revenues as 25 percent of GDP. Nonetheless, the model provides an indication of relative orders of magnitude of likely resource gaps. 5.4.5. Five scenarios have been produced to illustrate the potential financial outcomes of strategic options. The scenarios are derived from combinations of three strategic variables, which are: (i) achievement of key strategic objectives stipulated by the strategy, (ii) continuation of support by development partners, and (iii) a policy of raising salary by 10 percent in order to increase teachers’ satisfaction. The variable, achievement of strategic objectives, tests a financial implication of achieving selective targets, including achievement of 95 percent secondary completion rate by 2016, increasing the rate of transition from middle to higher secondary school to 90 percent by 2016, and increasing enrollment at tertiary education to 15,000 by 2018. The second variable, financial support by development partners, produces financial flow simulations with donor support and without donor support during the next eight years. The third policy option, increasing teacher salary, is estimated to demonstrate potential financial implications if financial means were used for increasing teachers’ satisfaction. The results of five scenarios are shown in a form of financing gap (the difference between resources and expenditures) in Figure 5.6, below. 5.4.6. The Scenario 1 assumes that development partners will continue financial support as they did in the base year 2010. In this scenario, the ratio of public expenditure and development partners’ support remains the same for the entire period until 2020. In other words, development partners’ support will grow in the same rate as the growth rate of public education expenditure. Compared to other scenarios that assume no development partner support, the resource envelope is much richer in this scenario. This scenario also assumes that the key strategic targets are met on time. However, increasing completion rates or enrollment rates means that there are more students in the system because there are fewer students who dropped out. Due to increased numbers of students at upper levels of education, more teachers and resources are required; thus the cost stream becomes larger. As a result of these two parameters, Scenario 1 sees moderate 65 financial gaps. The financial gap grows annually by USD 20 million by around 2017 due primarily to increased enrollment and larger cost of education service provision. 5.4.7. The Scenario 2 examines the impact of development partners’ withdrawing from the education sector. While RGoB aims at self-sustainable educational service provision, the gap of approximately USD20 million per year will be seen immediately after development partners’ withdrawal.58 This scenario examines only the impact of development partner withdrawal, so enrollment and completion rates are assumed to remain at the same level as in the base year. The result of this scenario shows that RGoB will face an annual USD20-USD25 million financial gap to sustain the status quo if the development partners withdraw. 5.4.8. The Scenario 3 includes an additional policy change to Scenario 2, which is increased teacher salary by 10 percent for increasing teacher satisfaction. As a result of increased teacher salary by 10 percent (assumed to begin in 2014), the financial gap becomes wider than in Scenario 2. By 2020, an annual financing gap exceeds USD30 million. 5.4.9. The Scenario 4 assumes no development partners, and achieving 11 FYP targets. This scenario shows a much wider financing gap due to the smaller resource envelope and larger recurrent cost due to more students and teachers in the system. The financing gap continues to widen every year and reaches over USD 60 million by 2020. The cumulative financing gap between 2012 and 2020 is about USD 400 million. As mentioned earlier, these scenarios do not assume any measures of improving efficiency of resource use. If the government embarks on reforms to improve efficiency of the system, such as consolidating schools, using double shifting, or other measures, the widening financing gap may be somewhat alleviated. 5.4.10. The Scenario 5 derives from Scenario 4, which assumes no development partners’ support and achievement of 11 FYP targets. In addition, this scenario also assumes additional costs to achieve the 11 FYP targets. Increasing secondary completion rates from 75 to 95 percent between 2011 and 2016 does not happen automatically without any interventions. To achieve such a goal, the RGoB will be required to introduce interventions such as, improving the quality of education, alleviating financial burdens on students by providing stipends, and potentially additional measures. Since not all means to achieve strategic targets are yet clearly articulated in the 11 FYP, this scenario makes a rough assumption that a lump sum of 500 million Nu (USD 10 million) is annually required to bring up secondary completion rate, improve the transition rate to higher secondary, and increase tertiary enrollment. The projected financing gap is the largest among all scenarios, and it exceeds annual USD 70 million by 2020. The cumulative financing gap is USD 480 million between 2012 and 2020. 5.4.11. The implication from these five scenarios is that a solid financial base is required for achieving 11FYP targets. Even the “best case” scenario (Scenario 1) implies a resource gap. Achieving higher enrollment and completion rates means more students and teachers in the system. Therefore, the education system will have to bear a larger recurrent cost. In addition, increasing enrollment and completion rates will not happen naturally without any interventions. Costs of interventions to realize accomplishments of strategic objectives should be seriously 58 In Scenarios 2-5, it is assumed that development partners withdraw in 2012. 66 considered. The scenarios also imply that reform measures may be required in order to improve the financial flows. For the education system to be sustainable when it grows, the country needs to ensure reliable resource flows or reduce the recurrent costs by system reform. The system needs to be more efficient to sustain a larger and more complex education sector with a limited resource base. Figure 5.6 Key Targets for 11FYP and Indicative Financing Gaps Financial Flow Simulation 2011-2020 0 (10) (20) S1: Donor+Increased secondary completion Financing Gap (million USD) (30) S2: No Donor+No secondary change S3: No Donor+No secondary change+ Increased (40) salary S4: No Donor+Increased secondary completion (50) S5: No Donor+Increased secondary completion+intervention (60) (70) (80) 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 67 Chapter 6 In Conclusion: Policy Challenges and Options for Reforms 6.1.1. Chapters 4 and 5 discussed the situation and current status of health and education expenditure respectively, and also pointed to the emerging evidence surrounding the challenges these sectors face and will need to address in the near future. 6.1.2. It is increasingly apparent that there are significant capacity constraints which require attention in both sectors. While concerns around sustainability will bring pressures to improve efficiency, the rising income and awareness levels of the general population will raise expectations around the quality of public services. The ministries will also need to gear up to undertake the role of sectoral stewardship, and not just function as financiers and direct providers of services, in an environment where both the public and private sector co-exist. These emerging functions require revisiting the role and capacity of the respective ministries and undertaking the necessary policy reforms to achieve effective realignment. 6.1.3. Both social sector ministries will need to look at changes to their remuneration systems, and where necessary, seek flexibility in rigid Royal Civil Service rules, to introduce focus on results, including incentives based on performance where warranted and feasible. This will be increasingly important for the sectors’ own effectiveness and attaining policy objectives and development targets, as also to address human resource constraints which already exist. These imperatives will only sharpen when the private sector starts becoming an attractive option for new recruits as well as existing incumbents in the public sector. 6.1.4. A greater coordination and monitoring role will be expected from the Ministries of Planning and Finance in setting expectations and in monitoring results from the sectoral ministries (Education and Health). However, both Planning and Finance have their own capacity constraints and will not find it easy to take this on. Unless the importance of this coordination and monitoring role is well understood and adequate accountability mechanisms are institutionalized, this aspect is likely to be missed or will not be given its due importance. There is moreover a strong need to move towards integration of planning, budgeting, expenditure and performance measurement – this integration does not exist today. It could facilitate synergies as well as the management of trade-offs between sectors when priorities compete for a reduced resource base. 6.1.5. A new generation of web-enabled monitoring systems is now functional – namely, the Planning and Management system (PlaMS), the Multi Year Rolling Budget system (MYRB), and the Public Expenditure Management System (PEMS) – but these systems need to be integrated such that the integration extends to linkages between financing and performance of social sector ministries. To a large extent, the web-based financial reporting tools for budget and expenditure (MYRB and PEMS) are integrated and well co-ordinated, but the same is not the case for performance measurement (PlaMS). This integration should ideally result in an overarching monitoring and evaluation framework for the country with linkages between financing and performance of social sector ministries. The Department of National Budget and the Gross National Happiness Commission (GNHC) are the key stakeholders who could conduct joint semi-annual reviews of performance vis-à-vis financing in the Ministries of Health and 68 Education as against the current budget review exercise which only looks at the pace of spending. Options Specific to the Health Sector 6.1.6. The improvements in road connectivity to district hospitals, as also the reduction in rural population due to ongoing migration, may mean that some BHUs are now facing reduced workload and could be downgraded following a close study of the facts. However, there exists a huge difference between the norms for a BHU-II equipped with only three staff and a BHU I with 35-50 staff. This needs to be reconsidered and made more flexible in line with the workload in different facilities. At the same time, the need for as many as 560 ORCs and 218 BHU IIs in the country may also require revisiting. In this context, the recent GIS study, and the planned online GIS tool for the Ministry of Health which could map population centers and health facilities including their utilization rates, could become a very useful tool. The Ministry of Health could learn from the experience of the Ministry of Education, which has been through a similar exercise for the same reasons. The Ministry of Education found that the involvement of village heads or gups, Geog Yargey Tshogchung or local administrative council and the community, is imperative if such a process is to succeed.59 6.1.7. A performance-based remuneration mechanism for future increments to pay of health professionals and workers has been recently suggested as a policy option (Joint Review Mission report, 2012). A sound performance management system, especially for voluntary health workers would be a priority, as these workers do not receive any remuneration presently, and they could be incentivized for performance on NCD prevention, undernutrition, maternal and child health and other priority public health areas. Special allowances to retain and motivate trained staff in specialized positions (such as Emergency Obstetric Care allowance) are, likewise, another suggested area for flexibility and pragmatism. 6.1.8. Immediate and concerted multi-sectoral action on complex challenges posed by undernutrition and non-communicable diseases (NCDs) is recommended to ensure efficiency and effectiveness of social sector spending and to retain the economic productivity of the country’s future generations. 6.1.9. Recent analytical work also underlines that the establishment of JDWNRH as a teaching hospital with real autonomy will raise many challenges and call for attention from top management. Observations Specific to the Education Sector 6.1.10. The current framework of revenue assignment and expenditure tracking is highly centralized to the Ministries of Education and Finance, lacks clarity in distinguishing primary from secondary education expenditures, and has resulted in sub-optimal outcomes in service delivery notably regarding the quality of education. Furthermore, the current mix of revenue may not be adequately exploiting public finance efficiency gains due to the limited analysis of preferences in service delivery in terms of ability and willingness to pay. 59 Joint Review Mission Report, 2012. 69 6.1.11. The central government has expanded access to near universal primary education. Now the challenge is to improve quality and efficiency in service provision, which the Ministry of Education realizes is more difficult to accomplish in a heavily centralized framework. The recent decentralization that has reformed local government institutions and intergovernmental fiscal relations can be a critical element supporting more effective and efficient public spending in education. One of the main findings of the PER in terms of the education sector is that despite some general patterns across Dzongkhags, their realities are complex, and there is a need for more systematic analysis of local conditions analogous to that carried out in the health section of the PER. As Bhutan advances on consolidating and building upon past achievements, each Dzongkhag will need a carefully crafted plan and financing framework closely tailored to the distinctive challenges and opportunities of the respective Dzongkhag. 6.1.12. Only a few key headlines for 11FYP targets were available at the time of this first phase of the PER. Yet, given the weight of some of the policy challenges already in view – which include ambitious expansion in the secondary and tertiary sub-sectors – even the best case scenario projects significant funding gaps. Withdrawal of development partners is an instrumental factor, but so, too, is the magnitude of the projected growth in the sector in terms of student numbers. This will drive up current costs which will need to be sustained long term. Phase 1 of the PER has endeavoured to give policy makers specific and actionable information as a point of departure. The analysis needs to be taken further in conjunction with the 11FYP preparations and updated with results of the 2012 BLSS, that is currently in progress. 70 Bibliography EXECUTIVE SUMMARY Royal Government of Bhutan. 2011. Mapping of Health Infrastructure. Ministry of Health, Thimphu, Bhutan. __________. 2011. Mapping of Health Infrastructure. Ministry of Health, Thimphu, Bhutan. __________. 2011. The cost of your healthcare: A costing of health services in Bhutan. Ministry of Health, Thimphu, Bhutan. __________. 2001. Financial Rules and Regulations. Ministry of Health, Thimphu, Bhutan. Royal Government of Bhutan. 2007. Public Finance Act. Ministry of Finance, Department of Public Accounts. Thimphu Bhutan. World Bank. 2012. Bhutan Economic Update: April 2012. Poverty Reduction and Economic Management, South Asia Region, World Bank, Washington, D.C. CHAPTER 2 IMF (International Monetary Fund). . Bhutan Article IV Consultation Staff Report. Country Report No. 11/123. International Monetary Fund, Washington, DC. World Bank. 2012. Bhutan Economic Update: April 2012. Poverty Reduction and Economic Management, South Asia Region, World Bank, Washington, D.C. __________. 2011. Bhutan Economic Update: April 2011. Poverty Reduction and Economic Management, South Asia Region, World Bank, Washington, D.C. CHAPTER 4 Royal Government of Bhutan. 2011. Mapping of Health Infrastructure. Ministry of Health, Thimphu Bhutan. __________. 2011. Bhutan National Health Accounts: 2009-10. Ministry of Health, Thimphu, Bhutan. __________. 2011. The cost of your healthcare: A costing of health services in Bhutan. Ministry of Health, Thimphu, Bhutan. Royal Government of Bhutan. 2011. Bhutan Multiple Indicator Survey 2010. National Statistical Bureau. Thimphu, Bhutan. __________. 2007. Bhutan Living Standards Survey. National Statistical Bureau. Thimphu, Bhutan. __________.2010. Statistical Yearbook 2010. National Statistical Bureau. Thimphu, Bhutan, 2010. Royal Government of Bhutan. Tenth Five Year Plan Volume 2: Programme Profile 2008- 2013. Gross National Happiness Commission. Thimphu, Bhutan. Royal Government of Bhutan. 2011 Annual Financial Statement 2011-12. Ministry of Finance, Department of Public Accounts. Thimphu Bhutan. __________. 2011. Expenditure Summary 2009-10. Reproduced from Bhutan NHA 2009-10. 71 Ministry of Finance, Department of Public Accounts. Thimphu Bhutan, 2011. Maeda, A, M N Harrit M, S Mabuchi, B Siadat, and Somil Nagpal. 2012. Creating Evidence for Better Health Financing Decisions- A Strategic Guide for the Institutionalization of National Health Accounts. World Bank, Washington DC. Nagpal, Somil. 2009. Health Financing and Expenditure Review- Bhutan. WHO SEARO, New Delhi. World Bank. 2011. Report of the Policymakers’ Roundtable on Health Financing in Bhutan, Thimphu. Washington, DC: World Bank. __________. 2010. Analysis presented at South Asia Regional Forum on Health Financing in Maldives. Washington, DC: World Bank. CHAPTER 5 Johnson, David, Ann Childs, Kiran Ramachandran, and Wangpo Tenzin. 2008. A needs assessment of science education in Bhutan. Department of Education, Oxford University, United Kingdom; Curriculum and Professional Support Division, Ministry of Education, Bhutan. UNESCO, New Delhi, July 2, 2008. Royal Government of Bhutan. 2010. Annual financial Statements of the Royal Government of Bhutan for the Year Ended 30 June 2010. Ministry of Finance, Department of Public Accounts. Thimphu Bhutan. Royal Government of Bhutan. 2011. Eleventh Round Table Meeting: Turning Vision into Reality: The Development Challenges Confronting Bhutan. 1-2 September 2011, Gross National Happiness Commission. Thimphu, Bhutan. Royal Government of Bhutan. 2007. National Statistics Bureau. Bhutan Living Standard Survey 2007 Report. National Statistical Bureau. Thimphu, Bhutan. __________. 2008. Poverty Analysis Report 2008. National Statistical Bureau. Thimphu, Bhutan. __________. 2010. Small Area Estimation of Poverty in Rural Bhutan. National Statistics Bureau and World Bank. Economic Policy and Poverty, South Asia Region, World Bank, Washington, D.C. August 2010. __________. 2011. Bhutan: Monitoring the situation of children and women. Bhutan Multiple Indicator survey 2010. National Statistics Bureau, United Nations Children’s Fund, and United Nations Population Fund, Thimphu, Bhutan, 2011. Royal Government of Bhutan. 2011. Annual Education Statistics 2011. Policy and Planning Division, Ministry of Education, Thimphu, Bhutan. __________. 2010. Annual Education Statistics 2010. Policy and Planning Division, Ministry of Education, Thimphu, Bhutan. __________. 2010. Tertiary Education Policy of the Kingdom of Bhutan, 2010. Ministry of Education, Thimphu, Bhutan. __________. 2007. General Statistics 2007. Policy and Planning Division, Ministry of Education, Thimphu, Bhutan. __________. 2006. General Statistics 2006. Policy and Planning Division, Ministry of Education, Thimphu, Bhutan. __________. 2005. General Statistics 2005. Policy and Planning Division, Ministry of 72 Education, Thimphu, Bhutan. World Bank. 2005. Expanding Opportunities and Building Competencies for Young People: A New Agenda for Secondary Education 2005. Washington, DC: World Bank. __________. Findings from the Bhutan Learning Quality Survey. Report No. 21. South Asia: Human Develop United Nations Educational, Scientific and Cultural Organization. 2011. Education for All Global Monitoring Report 2011. UNESCO, Paris. 73