Report No. 55388-NP Nepal Public Expenditure Review June 29, 2010 Poverty Reduction and Economic Management Sector Unit South Asia Region The World Bank Document of the World Bank CURRENCY EQUIVALENTS (Exchange rate effective June 5, 2010) Currency unit = Nepalese rupee (NPR) US$1 = 74.95 NPRs GOVERNMENT FISCAL YEAR July 15–July 14 ABBREVIATIONS ADB Asian Development Bank IPSAS International Public Sector Accounting Standards AG Auditor General LBFAR Local Bodies (Financial Administration) Regulations APP Agriculture Perspective Plan LSGA Local Self-Governance Act BMIS Budget Management Information LTO Large Taxpayer Office System CIAA Commission for the Investigation of MDG Millennium Development Goal Abuse of Authority DALY Disability-adjusted life year MoHP Ministry of Health and Population DDC District Development Committee MoLD Ministry of Local Development DDF District Development Fund MPPW Ministry of Physical Planning and Works DECS District Expenditure Control System MTEF Medium-Term Expenditure Framework DfID UK Department for International NARDF National Agriculture Research and Development Development Fund DHS Demographic and Health Survey NER Net Enrolment Rate DoLIDAR Department of Local Infrastructure NGO Non-Governmental Organization Development and Agricultural Roads DoR Department of Roads NLSS Nepal Living-Standards Survey DTCO District Treasury and Controller Office NPC National Planning Commission DTMP District Transport Master Plan NRB Nepal Rastra Bank DTW Deep Tube Well OECD/ Development-Assistance Committee of the DAC Organisation for Economic Co-operation and Development EHCS Essential Health Care Services PAC Public Accounts Committee FAR Financial Administration Regulations PAF Poverty Alleviation Fund FCGO Financial Comptroller General Office PEFA Public Expenditure and Financial Accountability FMIS Financial Management Information PFM Public Finance Management System FUG Forest User Group PPIS Project Performance Information System GDP Gross Domestic Product PPMO Public Procurement Monitoring Office GFS Government Fiscal Statistics (of the PRS Poverty Reduction Strategy IMF) GoN Government of Nepal RBN Roads Board Nepal HIPC Heavily Indebted Poor Countries SMC School Management Committee HSS Health Sector Strategy SOE State-owned Enterprise IA Internal Audit SRN Strategic Road Network IAP Immediate Action Plan STW Shallow Tube Well IDA International Development Association SWAp Sector-wide Approach IDP Internally Displaced People VAT Value Added Tax IFMIS Integrated Financial Information VDC Village Development Committee Management System IMF International Monetary Fund WTO World Trade Organization Vice President Isabel M. Guerrero, SARVP Country Director Susan G. Goldmark, SACNP Sector Director Ernesto May, SASPF Sector Manager Miria A. Pigato, SASEP Task Manager Roshan Darshan Bajracharya, SASEP Table of Contents ACKNOWLEDGEMENTS ........................................................................................................................................... I EXECUTIVE SUMMARY........................................................................................................................................... II 1. STATE BUILDING AND THE ECONOMY .......................................................................................................... 1 A. Political Context ......................................................................................................................................... 1 B. Economic Profile ........................................................................................................................................ 3 C. Nepal’s Development Strategy and Reform Agenda .................................................................................. 6 2. FISCAL FRAMEWORK AND ISSUES IN AGGREGATE EXPENDITURE ................................................................. 8 A. Fiscal Framework and TYIP Implementation ................................................................................................. 8 3. SECTOR EXPENDITURES AND OUTCOMES ................................................................................................... 21 A. Allocation of Budget by Sector ...................................................................................................................... 21 B. Moving Ahead – Making Choices .................................................................................................................. 26 4. EDUCATION ................................................................................................................................................ 31 A. Education in Nepal: Strategy, Outcomes and Challenges ............................................................................ 31 B. Education Expenditures................................................................................................................................. 40 C. Planning, Implementation and Monitoring Issues ........................................................................................ 46 D: In Conclusion................................................................................................................................................ 47 5. HEALTH ...................................................................................................................................................... 48 A. Health: Outcomes and Challenges ............................................................................................................... 48 B. Public Health Expenditure ............................................................................................................................ 53 C. Planning, Implementation and Monitoring Issues ........................................................................................ 60 D. In Conclusion ................................................................................................................................................ 61 6. SOCIAL PROTECTION .................................................................................................................................. 64 A. Poverty, Risk and Vulnerability ..................................................................................................................... 64 B: Trends in Social Protection Expenditures and Composition 2005-2010 ...................................................... 66 C: Social Protection: Program Coverage, Targeting and Efficiency ................................................................ 69 D: In Conclusion................................................................................................................................................ 73 BOXES Box 1: Transfers............................................................................................................................................................. 15 Box 2: Budget Implementation................................................................................................................................... 19 Box 3: Trends in Non-Budgetary Aid and Its Impact............................................................................................. 20 Box 4: Safety Net........................................................................................................................................................... 28 Box 5: Community Managed Development.............................................................................................................. 29 Box 6: Federalism and Social Service Provision....................................................................................................... 30 Box 7: Community-Managed Schools...................................................................................................................... 37 Box 8: Teacher Relief Quota (Rahat)......................................................................................................................... 41 FIGURES Figure 1.1: Manufacturing Real Value Added ..........................................................................................4 Figure 2.1: Revenue (Percentage of Total Revenue).............................................................................. 10 Figure 2.2: Capital Expenditure and Primary Deficit (% of GDP)..................................................... 11 Figure 2.3: Transfers (Current and Capital) vs Capital Formation (Percentage of GDP).............. 14 Figure 2.4: Recurrent Expenditure and Revenue (Percentage of GDP)........................................... 16 Figure 2.5: Changes in Expenditure in 2007–2003 and 2009–2007......................................................17 Figure 3.1: Public Expenditure – Major Sectors (Percentage of GDP).............................................. 21 Figure 3.2: Social Services – Sectoral Trends (Percentage of GDP)................................................... 22 Figure 3.3: Economic Services – Sectoral Trends (Percentage of GDP)........................................... 23 Figure 3.4: Financing 2007 and 2008 Budgets........................................................................................ 26 Figure 4.1: Comparative of Literacy in the 15–24-year Age Groups, by % of GPI........................ 31 Figure 4.2: Survival Rate to Grade 5 and Dropout Rate to Grade 1................................................. 32 Figure 4.3: Primary level NER and GER, South Asia 2007............................................................... 32 Figure 4.4: Gender Parity Index at Primary level, South Asia.............................................................. 34 Figure 4.5: Secondary level NER and GER, South Asia 2007............................................................. 35 Figure 4.6: Gender Parity Index at Secondary level, South Asia........................................................ 35 Figure 4.7: Education Sector-Sources of Funding .............................................................................. 43 Figure 5.1: Health Indicators – Regional and International Comparison........................................ 49 Figure 5.2: Sources of Funding 2007 and 2010..................................................................................... 54 Figure 6.1: Spending on Social Protection in South Asia 2004/05 (Percentage of GDP).67 Figure 6.2: Social Protection Expenditure 2004/05–2009/10 (Total and % of GDP).................. 68 Figure 6.3: Beneficiaries of Old Age, Widow and Disability Allowances 1996–2008..................... 69 Figure 6.4: Approved Budget of Main Cash Transfer Programs(NPRs Million )1995/96 and 2009/10........................................................................................................................................................ 71 Figure 6.5: Breakdowns of Wage and Non-Wage Costs by Programs 2008/09............................... 73 TABLES Table 1.1 Constituent Assembly Election Results and Composition of the CA.................................................. 2 Table 1.2 Service-Sector Growth During and Post-Conflict – Value Added, FY01–09..................................... 5 Table 2.1: Fiscal Framework, FY07-FY10.................................................................................................................. 9 Table 2.2: Total Expenditure during the Tenth Plan and the Interim Plan ....................................................... 13 Table 2.3: Public Spending and Resource Mobilization............................................................................................14 Table 3.1: Sectoral Spending Trends (Percentage of GDP).....................................................................................25 Table 3.2: Pro-Poor Expenditure (Percentage of Total Expenditure)................................................................. 25 Table 4.1: Primary Level (Grades 1–5) Net Enrolment Rates............................................................................... 33 Table 4.2: Secondary Level (Grades 6–10) Net Enrolment Rates.......................................................................... 36 Table 4.3: Educational Sector Expenditure: Percent of Education Sector Budget (2003–2010) 40 Table 4.4: Education Spending-Economic Head...................................................................................................... 40 Table 4.5 Number of Teachers and Salaries in 2010..................................................................................................42 Table 4.6: Scholarships Budget (NPRs ‘000) ..............................................................................................................43 Table 4.7: Education Spending by Development Regions and Ecological Belts 44 Table 4.8: Primary and Secondary Education Spending by Development Regions and Ecological Belts...44 Table 5.1: MDG Achievement..................................................................................................................................... 48 Table 5.2: Total Health Expenditure in SAARC Countries (Percentage of GDP) ........................................... 50 Table: 5.3: Health Care Expenditure by Funding Source and Major Functions (Percentage of Total Health Expenditure).................................................................................................................................................................... 50 Table 5.4: Health-Related Expenditure by Funding Source and Major Functions (Percentage of Total)....... 51 Table 5.5: Status of Health Infrastructure and Personnel 2010.............................................................................. 51 Table 5.6: Status of Human Resources: State Institutions...................................................................................... 52 Table 5.7: Household Expenditure by Facility (Per Capita).................................................................................. 52 Table 5.8: Percentage of Women Who Had Skilled Birth Attendance....................................................................53 Table 5.9: Health Sector Expenditure (Percentage of Health Sector Budget 2003–2010)..................................54 Table 5.10: Health Spending (Economic Heads).......................................................................................................55 Table 5.11: EHCS – Regional Expenditure by Major Economic Heads 2007 and 2009.................................... 56 Table 5.12: Beyond EHCS – Regional Expenditure by Major Economic Heads 2007 and 2009..................... 56 Table 5.13: Public Expenditure in Health Sector by Regions and Geographical Belts (Percentage of Health Spending 2001–2009)..................................................................................................................................................... 57 Table 5.14: Public Expenditure in Essential Health Care and Beyond Essential Health Care by Regions and Geographical Belts 2004–2009.................................................................................................................................... 57 Table 5.15: Health Outcomes by Development Regions and Ecological Belts- 2006.........................................58 Table 5.16: Health Sector–Spending Performance 2004-2009................................................................................ 59 ACKNOWLEDGEMENTS 1. The Public Expenditure Review process that led to this report was initiated in November 2009 under the guidance of Dr. Pushkar Bajracharya, Hon’ble Member of the National Planning Commission. In this process, the team worked with officials of National Planning Commission, Ministry of Finance and Central Bureaus of Statistics. 2. This document also builds on earlier reports, including the World Bank’s 2000 Public Expenditure Review; Managing Public Finances for New Nepal, 2007; and, Reforms under Fiscal Stress: A Policy Note on the Priorities for Fiscal and Budget Reform in Nepal, 2006. 3. The World Bank team was led by Roshan Darshan Bajracharyan. The team was composed of Venkatesh Sundararaman, Saurav Dev Bhatta, Mohan Aryal, Albertus Voetberg, Nastu Prasad Sharma, Ceren Ozer, Prajwal Shahi, Ekaterina Georgieva Stefanova, Nimanthi A. Attapattu, Muhammad Shafiq, Sunita Kumari Yadav and consultants: Binod Bhattarai, Shreehari Dhungana, Pradeep Kumar Shrestha. The team worked under the guidance of Susan G. Goldmark, Country Director, Miria A. Pigato, Sector Manager, and Hisanobu Shishido, Policy Cluster Leader. Peer reviewers who provided comments included Dr. Bimal Prasad Koirala, former Chief Secretary; Mr. Vidhyadhar Mallik, Former Finance Secretary and Yasuhiko Matsuda, Sr. Public Sector Specialist, World Bank, Stephane Guimbert, Sr. Economist, World Bank. 3. The concept note for the work program was reviewed on December 02, 2009. Findings of the Public Expenditure Review were shared with concerned government officials and specifically with Government Budget Management and Expenditure System Review Commission. i EXECUTIVE SUMMARY Introduction The Government of Nepal (GoN) has taken significant steps in implementing its poverty reduction strategy to achieve the Millennium Development Goals (MDGs), while maintaining macroeconomic stability and prudent fiscal management. Within this framework, it has initiated a comprehensive assessment of its spending policies. The World Bank is contributing to this assessment through a programmatic and participatory Public Expenditure Review (PER), spread over two-three years, to be conducted jointly between the Ministry of Finance, the relevant sector ministries as well as donors. The work includes analytical pieces, capacity-building exercises, workshops and technical assistance. This report presents the first of several contributions to the PER. It reviews public spending in the health, education and social-protection sectors, bringing together, in a comprehensive fashion, available information on spending and intended outcomes. It assesses the links between, and consistency of policy, planning and budgeting; updates and analyzes sectoral and subsectoral public expenditure trends; and explores what major inputs are financed by public expenditures in health and education sectors. Follow-up work will include an analysis of spending efficiency, which will be made possible by the availability of data from the National Living Standards Survey (III) – expected by 2011. An assessment of expenditures in the road sector is expected in FY11. Fiscal Outcomes Prudent fiscal management has been responsible for macroeconomic stability in the past five years, a remarkable achievement. Net domestic financing of the deficit is below 2 percent of GDP and current public debt level is sustainable, at 40 percent of GDP. Revenues rose from 12 percent of GDP in FY07 to 14.8 percent in FY09, due to reforms in revenue administration and strong customs and VAT collection. The expansion in tax and duty collections reflected higher consumption, supported by remittances from abroad. Notwithstanding recent progress, there is still considerable scope to raise the revenue to GDP ratio, a necessary step to meet Nepal’s ambitious social programs. While the overall budget deficit is low, the increasing trend in recurrent public spending raises concerns. The budget deficit (after grants) is expected to reach 2.4 percent of GDP in FY10. Nepal’s public expenditure is lower than in comparable countries but it has risen fast in recent years. Overall spending rose from an average of 15 percent of GDP during the Tenth Plan (FY 03- FY 07) to 18 percent in FY07 and 22 percent in FY09. Recurrent spending is expanding the most – from 11 percent of GDP in FY07 to an expected 15 percent in FY10 owing to higher government salaries, additional allocations for hiring new teachers, and increased public security expenses. Social spending – health, education, social safety-nets, drinking water and sanitation, and local development –has increased from 6 percent of GDP to 10 percent in the same period. While much of the increase in health and education spending is justified by the need to improve social indicators, large commitments in recurrent spending are difficult to reverse. The wage bill represents a quarter of current expenditure and is ii expected to rise with the indexation of allowances to inflation and the proposed further expansion of the civil service; the pension bill is estimated to equal that of salaries in the next four years; and, servicing of outstanding debt and restructuring of state-owned enterprises will contribute to a further rise in spending. The quality and quantity of public investment are poor. A significant increase in capital spending would be required to boost the productivity of the economy. Capital formation is low by international standards, and inadequate to ensure sustained growth rates in the medium to long run. Average public capital spending during the Tenth Plan (FY03-FY07) was 4.4 percent of GDP and rose to 6.6 percent during the first two years of the TYIP (FY08-10). However, almost half of the spending classified as ―capital expenditure‖ in the budget refers to transfers to two loss-making public corporations and other ―capital transfer payments‖. For the country as a whole, the level of fixed capital formation is extremely low (3.2 percent of GDP in 2009); much lower than that of comparable countries. Besides increasing the level of investment expenditure, the Government needs to raise the quality and efficiency of this spending, avoiding misallocation of resources while improving implementation capacity. Aid availability has increased the fiscal space. Aid has risen from 3.6 percent of GDP in FY07 to an estimated 5.2 percent of GDP in FY10 (with grants reaching 3.8 percent of GDP). However, much additional aid appears to be delivered outside the budget, a signal of donors’ concerns with the government’s implementation and oversight capacity. The presence of a large external budget makes the alignment of public expenditures with the government’s poverty strategy particularly challenging, and undermines the primacy of the budget as the principal instrument of national policy. Better alignment of aid with the Government strategy and budget is required. As budget absorption capacity increases, the donors should: (i) Share timely information on disbursements and commitments with the government; (ii) Align their objectives with national objectives and priorities; and (iii) Progressively channel more of the external aid through the budget, giving national authorities control over a larger share of resources. Inter-Sectoral Budget Allocations Pro-poor expenditure comprises almost half of total expenditure, in line with the priorities outlined in the Three-Year Interim Plan (TYIP, FY08-FY10). The Medium Term Expenditure Framework (MTEF) has further improved alignment of donor funds to the sector strategies. Pro-poor spending rose after 2008, mainly in the education and health sectors and in outlays for block grants under the Poverty Alleviation Fund. Spending on agriculture roads has increased and that on rural connectivity is improving. By contrast, spending has not grown commensurately with the overall trend for drinking water and sanitation sectors – which together received less than 1 percent of GDP. More investment in hydro power is needed to address power outage in the country. Some notable trends in social spending include:  Education spending has increased from 2.8 percent of GDP during the PRSP/Tenth Plan (FY03-FY07) to 3.5 percent of GDP in the first two years of iii the TYIP (FY08-FY10). Much of this increase went into financing the Education for All strategy, focussing on primary education.  Health spending is also on the rise; from an average of 0.9 percent of GDP during the Tenth Plan (FY03-FY07), close to about 1.5 percent of GDP in the first two years of the TYIP (FY08-FY10). The health SWAp, which provided close to half of the 2010 health-sector budget, seeks to expand the coverage of essential healthcare services, with a focus on reducing maternal and infant mortality rates.  Local development spending has increased from less than 1 percent of GDP during the Tenth Plan (FY03-FY07) to 1.5 percent of GDP in 2009; key areas of increased spending going to rural infrastructure and local government grants. The use of funds for local infrastructure, however, declined in 2009 due largely to the absence of elected local bodies and the reported capture of resources by various groups.  Spending on economic sectors is low, 4.75 percent of GDP during the first two years of the TYIP (FY08-FY10). Notable trends are: Investment in agriculture, including spending on irrigation services, has been under 2 percent of GDP for the past 10 years despite the fact that almost 70 percent of the rural population depends on farming for food security. Spending on feeder roads has doubled compared to the Tenth Plan period but the quality of these roads is below international standards. Investment on road rehabilitation and maintenance remains low. Overall, the prioritization of programs and projects needs to be carefully reviewed and revised. Since 2008, the government has categorized over 80 percent of the budget as high-priority activities. This is being undertaken under the on-going MTEF exercise which aims to ensure a more efficient resource allocation under the new Three-Year Plan (FY08-FY10). Education: Outcomes and Challenges Nepal’s literacy level and gender parity index are comparable to those of other South Asian countries and the MDG on primary education is attainable. The net enrolment rate (NER)1 reached 92 percent in 2009. Disparities across regions, poverty and ethnic groups have also narrowed. The progress in girl enrolment has been strong, with the current NER now comparable with that of boys. There has also been progress in retaining students up to grade 5. The second SWAp in education (The US$2.6 billion School Sector Reform Program, [SSRP], 2010-2014) anchors the EFA plan of action for further improving primary education. The NER in secondary schooling reached 50 percent in 2008 from 31 percent in 2001 – a considerable achievement, given the starting point. Several challenges remain, however, that may impede further expansion of enrolment to secondary schools. These include: children dropping out of school in order to support their families; overall low quality of education; and lack of options for students to receive vocational or technical education. The gains made in NER of secondary schools are 1 Expressed as a percentage of children enrolled at an appropriate level of schooling. iv much more pronounced for lower-income quintiles. Continuation to secondary schooling, especially for underserved groups, is a key issue to address. Many of the achievements in education stem from the decentralization of school management, licensing of teachers and opening up of textbook publishing to the private sector. Nepal began transferring school management to local committees in 2002 and the policy contributed to a 50 percent reduction in the number of out-of- school children within three years of the transfer. Public- and government-supported community schools have a major share in education (85 percent of all schools; 87 percent of total enrolment), but the contribution of the private sector in education has also been growing. The improvements in education outcomes have been made possible by increases in the education budget, which reached 7 percent of GDP under the SSRP. However, as noted above, spending on teacher salaries is the largest item, amounting to 78 percent of the education budget – relatively high by international standards – while investments in expanding and improving school infrastructure have remained relatively constant. Teacher deployment remains difficult. The number of scholarships has doubled but targeting is inadequate. The decentralisation of education has reduced costs of central education administration and these savings could be reallocated in a more targeted manner in order to address regional disparities. The achievements in the education sector need to be protected and expanded. Key recommendations presented in this report are as follows:  Strengthen the management of teachers. This requires that a standardization of rules pertaining to teacher recruitment, service rules, teacher professional development be developed in close coordination with representatives of teacher unions. The complete abolition of relief quotas (or raahat teachers) and the hiring of all teachers by school management committees are also suggested.  Expedite the transfer of school management to SMCs. There is a need to protect the gains made in education by SMCs by providing more block grants to schools doing well for improving education quality. An independent, comprehensive social audit of the functioning of the SMCs can further strengthen the accountability and governance of SMCs and assist in mainstreaming SMCs to higher levels of schooling.  Revisit the scholarship program for better targeting and monitoring. Scholarships are a priority of the education and therefore they can be made more effective by the release of one-third of the funds during the first trimester of the budget cycle. Health: Outcomes and Challenges Nepal has made considerable progress in reducing maternal and under-five mortality rates. The Maternal Mortality Ratio (MMR) declined from 415 to 229 deaths per 100,000 from 2001 to 2009. Other achievements include reductions in infant and under-five mortality and fertility, and the expansion of immunization. However, there still are urban-rural disparities and differences across ecological regions. Health sector spending has doubled its share of GDP from 0.9 percent in 2006 to 1.7 percent in 2010. Donor funding contributed 48 percent of the 2010 sector budget, which is anchored by SWAp-I. In addition, substantial donor funds are also spent through NGOs. Public spending is geared to providing essential health-care services v (EHCS). These accounts for half of the total expenditure and most of it go into providing services through health posts/sub-posts. It has been bolstered by donor support under SWAp-I, and the funds made available by NGOs. Even though spending on ―beyond essential health care services‖ has also expanded, investment in EHCS remains protected by SWAp-I. Private expenditure on health has been growing. Per-capita health expenditure increased in nominal US dollar terms from US$ 12.5 in 2001 to US$ 19 in 2006, of which the private sector contributes 56 percent2. Out-of-pocket (OOP) household health expenditure makes up 90 percent of the private-sector spending, which is spent largely on curative care while public spending focuses on EHCS. In essence, the private sector leads in healthcare capital formation and investment in human resources. Almost 80 percent of public health spending is recurrent. Salaries, as a share of total health expenditure, is declining while spending on supply of medicines, equipments and fund transfers for targeted programs and health-related social security have increased. Health-sector spending on medicines, equipments and targeted programs is now 39 percent of total sector spending, a growth of 44 percent in three years between FY06 and FY09. The government is emphasizing quality and service delivery, and has begun reallocating funds from non-performing activities to those that produce results. For example, cash incentives for safe delivery and targeted free basic healthcare services were extremely effective in reaching the intended beneficiaries in the last two-three years. But there are still regional disparities in health spending, with the Central Development Region (CDR) receiving the lion’s share. More than 90 percent of medicines and three-quarters of the capital fund are used in the CDR, where about 40 percent of Nepalis reside. The targeted programs under EHCS are focused on the Terai (plains), where over 60 percent of the allocation for these is spent. It must be noticed that in Nepal, as in other countries, variations in public health expenditures alone do not explain the disparities in health outcomes across development regions and ecological belts. Many inter-dependent variables such as literacy rate, access to safe drinking water and sanitation, road and telecommunications, availability of skilled staff and medical goods, and socio-cultural behaviours and beliefs also explain the outcomes. This suggests the need for multiple efforts for addressing all the variables for attaining the desired outcomes rather than focusing on health spending alone. The SWAp has begun to address the regional disparity in spending. The challenges facing the health sector include better access and quality of services as well as equity in health service utilization. Key recommendations presented in this report are as follows:  Realign health resources. The decline in the share of Essential Health Care Services (EHCS) in total health expenditure needs to be addressed. There is also a need to recalibrate the EHCS to underserved regions and population groups. The government can also consider contracting out to private and NGO institutions 2 Nepal National Health Accounts, July 2009. vi specific non-EHCS specific services, rather than increasing block grants to tertiary hospitals and other institutions.  Outsourcing of services to medical schools. The outsourcing should not only expand quality coverage of basic health care services, but also provide an opportunity for new doctors to experience rural health care.  Preparation of the AWPB for the coming fiscal year should start in October and be accompanied by procurement plans. A complete budget package – activities backed by a procurement plan, anchored by funds against verifiable output targets – could strengthen the planning, budgeting and implementation processes. Aligning the budget with the results framework of the NHSP II could protect spending from interest and advocacy groups.  Address inequality in service provision and outcomes. A move towards allocating budget on the basis of beneficiary size (so that allocation on per-capita basis is considered), accessibility (how accessible services are to users), and service delivery cost would improve equity in access to, and use of, health services.  Strengthen the monitoring of health-sector activities. The reporting schedule should be aligned with the budget’s fund release for facilitating mid-term fund release and/or correcting variations. There is a need for in-depth analysis of the large OOP expenditures – to understand what the funds are used for – for planning financial protection measures.  Establish an inter-ministry coordination committee to facilitate the delivery of health services. Close coordination of inputs provided by other line ministries could facilitate the utilization of existing health services more effectively.  Prepare and implement a private-public partnership plan to complement the public efforts to improve health service delivery. A private-public health policy could be the first step towards making health services financially sustainable. Such a partnership could also free public resources from urban investments for use in expanding basic health services to underserved regions and social groups. An umbrella regulatory mechanism, such as a private-sector regulation Act – with clear roles and responsibilities for state and non-state actors in the health sector – would facilitate implementation of this kind of partnership arrangement. Social Protection Nepal has increased expenditures on social protection to comparable regional levels but better targeting is needed. The spending has increased from about 0.5 percent of GDP in FY05 to 2 percent in FY09 and is expected to reach 2.9 percent of GDP in FY10. The largest safety nets (allowances for widows, the elderly and the disabled) add up to 0.8 percent of GDP. The challenge is to ensure that these programs are retained while the government improves targeting and administration of social protection programs. Proper targeting may be costly to put in place initially but can be more cost-effective and purposeful over the longer term. Expansion of social protection measures and safety nets needs to be based on rigorous impact evaluations. In the absence of proper planning and targeting, there is a danger of spreading social protection measures too thinly to have considerable impact. There is therefore need to review and reallocate expenditures to direct resources to fewer, more effective programs. Better alignment of programs to needs is also likely to generate substantial vii savings. Key recommendations presented in this report are aimed at increasing the overall productivity of government outlays while preserving redistributive goals:  Reviewing the eligibility criteria and benefit levels of the Old-Age and Widows’ Allowance programs to better align them to needs;  Consolidating the Old-Age and Single Women’s programs to avoid double uptake of program benefits; and  Targeting benefits specifically to the poor. For instance, if the Single Women’s Allowance program was targeted to only poor single women above the age of 60, the savings would amount to NPRs 2.17 billion annually3. To generate further fiscal space, the GoN could also consider targeting the Old- Age Allowance program to a larger section of the poor elderly population than is currently covered by the program. This would represent a logical effort towards reducing old-age poverty, as is done in other countries in South Asia, such as India and Bangladesh. This would, however, introduce all the challenges of targeting. Moving Forward Nepal has maintained fiscal and macroeconomic stability. Expenditures have gradually become more aligned to sound sector strategies and have resulted in improved social indicators. Moving forward, the Government needs to carefully reconsider the trade-offs between further increasing spending to the social sectors and addressing long-neglected needs in the infrastructure sector. In the short run, the Government could consider a number of key steps to improve public expenditure efficiency, both across and within sectors. To improve efficiency across sectors:  Bolster planning and budgeting by approving early sectoral procurement plans together with the annual budgets;  Freeze allocations for non-implemented activities and end virement of funds nine months after the start of the fiscal year; and  Strengthen monitoring by linking spending to outputs and using the feedback for deciding sectoral allocations. To improve efficiency within sectors:  Education – Review the service rules and update the teacher database and payrolls; do away with the practice of ―relief‖ teachers. Teacher deployment should be based on the student-teacher ratio, without exceptions;  Health – Align the budget with the results framework of the National Health Sector Program II to ensure allocations based on beneficiary size, accessibility and service delivery cost, and social inclusion, governance and 3 Assuming there are no costs to targeting (Isik-Dikmelik, 2009). viii accountability.4 Commission two in-depth studies to address concerns related to sustainability of health spending and efficiency gains, and out-of-pocket expenditures of the poor; and  Social Safety Nets – Conduct impact evaluations of ongoing social protection programs as a basis to reallocate expenditures to more effectively reduce poverty and vulnerability. Key issues to look into are targeting of cash transfers, benefit levels and delivery systems. 4 The Gender Equality and Social Inclusion Strategy and the Governance and Accountability Action Plan serve as reference points. ix 1. STATE BUILDING AND THE ECONOMY Key Messages  Nepal seized the opportunity provided by the People’s Movement of April 2006 and held a successful election for a Constituent Assembly (CA) to prepare a new statute.  A stable political environment and continuous efforts by all stakeholders to manage the economy are essential to attain the national vision of a peaceful, inclusive, just and prosperous ―new‖ Nepal. There are also pending structural issues that need to be addressed to help ensure broad- based, inclusive growth.  The transition to peace has not been smooth and is still underway. Political uncertainties have prevented the peace dividends from reaching citizens.  Prudent fiscal management has created space for expanding expenditure, but there are issues of implementation efficiency, transparency, and monitoring that need to be addressed. There is also an emerging concern that prudence may give way to political expediency. Summary Nepal has made remarkable progress towards establishing peace and democracy. The People’s Movement (2006) marked the beginning of political negotiations for ending Nepal’s decade-long (1996-2006) conflict. The Comprehensive Peace Agreement (CPA) was signed in November 2006, paving the way for the Constituent Assembly (CA) election of April 10, 2008. The new 601-member parliament stemming from this election functions as both a regular legislature and an assembly charged with drawing up a new constitution5. While the political process has been underway, the business environment has remained poor and growth has averaged four percent, post-conflict. The majority of the population has yet to enjoy a peace dividend. A. Political Context 1.1 The transition to peace and democracy has been slow – the transformation has been delayed by differences between the main political parties on how to manage the transition and change a unitary state into a federal republic. The elected assembly’s inability to balance the dual responsibilities of legislating and constitution-making is partly responsible for the delay6. Inherent weaknesses in the political settlement stemming from the convergence of short-term, party-political interests rather than a common vision for the reshaping of Nepal have also complicated matters7. 5 The original deadline was May 28, 2010, but when this was not met it was extended to May 28, 2011. 6 Constitution-making involves negotiations and consensus-building, while lawmaking and running a government are more competitive in nature and require implementation of government programs that may run up against political opposition. 7 International Crisis Group. February 2009. Nepal’s Faltering Peace Process. Report No 129, page 2. 1.2 The CA election returned an assembly with 25 political parties, which decided to end monarchy in Nepal. The Unified Communist Party of Nepal (UCPN)8, formerly a Maoist rebel group, emerged as the largest party with one-third of the seats. Next came the Nepali Congress (NC) and the Communist Party of Nepal (Unified Marxist-Leninist). The election also marked the emergence of new political forces, representing the country’s plains districts along Nepal’s southern borders with India (Terai). Two newly-formed Madhesi/Terai-based parties – the Madhesi Janaadhikar Forum (MJF) and the Terai-Madhesh Loktantrik Party (TMLP) – emerged as the fourth- and fifth-largest. Table 1.1: Constituent Assembly Election Results and Composition of the CA Seats: Seats: Total in the Percentage FTTP* PR** CA of seats in Political party the CA UCPN 120 100 220 36.6 NC 37 73 110 18.3 CPN (UML) 33 70 103 17.1 MJF 30 22 52 8.7 TMLP 9 11 20 3.3 Sadbhawana Party 4 5 9 1.5 Janamorcha Nepal 2 5 7 1.2 Others 5 49 80 13.3 Total 240 361 601 100 Source: Election Commission. *First-Past-The-Post; ** Proportional Representation 1.3 Nepal adopted a mixed-representation electoral model aimed at making the parliament and CA strongly inclusive: a first-past-the-post (FPTP), or direct, contest; and a party-based vote by proportional representation (PR)9. The 575 elected members include 50 Dalits (formerly considered untouchables), 204 Madhesis and 192 Adivasi Janajatis (indigenous peoples)10. Women won 29 of the 240 FPTP seats – a relatively small 12 percentage, but with additional seats obtained through proportional representation, women occupy one-third of the CA seats11. 1.4 Political instability delayed writing of the Interim Constitution, and these tensions intensified with the resignation of the UCPN-led government in May 2009, and protests led by that party thereafter. However, the constitutional committees still managed to prepare and submit concept papers and preliminary drafts for the new 8 The UCPN (M) reorganized as the United Communist Party of Nepal (UCPN), Maoist (M) after merging with the Unity Centre Mashal (Samyukta Janamorcha Nepal or NCP Unity Centre Mashal) party on 12 January 2009. 9 Under Nepal’s electoral system, 240 members were elected directly and 335 by proportional representation, including 33 percent women. The government appointed the remaining 26 members to the CA. 10 International Crisis Group. 2008. Nepal’s Election: A Peaceful Revolution? Asia Report No.155 11 Electoral law required political parties to have 50 percent women on their list of candidates for the PR contest. 2 constitution. But, when the CA failed to promulgate a constitution by the allotted date of May 28, 2010, the parliament extended the assembly’s term by one year. 1.5 Political issues remain that must be resolved to bring stability and peace to Nepal. Many of the commitments in the peace agreement and Interim Constitution are still unfulfilled, including the agreement on the Management of Arms and Armies 12. The latter issue relates mainly to the ―integration‖ and ―rehabilitation‖ of former Maoist combatants and the establishment of commissions on peace and rehabilitation and truth and reconciliation. The unfulfilled commitments on the UCPN side include returning property seized during the conflict and dismantling parallel structures formed during the conflict years. 1.6 The unsettled political issues have affected economic management. Poor public security, deterioration of law and order and the rule of law, and widespread impunity have marked the Nepali transition. Public security remains poor in the 20 Terai districts bordering India and in the eastern Hill districts. Despite these outstanding issues, the start of the peace process itself an ―open moment‖ of history with few parallels. It also raised public expectations, and meeting them is a challenge because the unstable political situation has adversely affected the business climate and constrained growth. B. Economic Profile 1.7 Real GDP growth averaged 4.1 percent during the conflict and the three years since the peace process began. The economy grew by 3.9 percent during the conflict years (1996-2006) and by 4.5 percent during the post-conflict period (2006-09). If FY10 growth projection of three percent is included in the post-conflict period, there is no significant difference in the rates of growth rate before and after peace. This shows that the economy was resilient during the conflict but that peace dividends after the conflict have been limited. 1.8 The ―resilience‖ during the conflict can be largely attributed to prudent fiscal management. Expenditures, including security spending, had been prioritized within the Medium Term Expenditure Framework (MTEF) and net domestic borrowing was kept at sustainable levels (below 2 percent of GDP). This helped maintain macroeconomic stability, keep inflation low, and make some growth possible despite the difficult security situation. 1.9 The ―peace dividends‖ still remain elusive. Growth recovered to 5.3 percent in FY08 assisted by good weather and increased tourist arrivals. But the positive impact of peace soon began to taper off. Growth decelerated in FY09 to 4.7 percent and, as mentioned above, is projected to be around three percent in FY10. Slower growth is attributed to protracted political uncertainties, law and order problems, and the associated deterioration in the business environment – together with unseasonal rains and prolonged draught. Some key structural reform issues remain unaddressed. Low growth, in turn, continues to constrain the government’s ability to meet its transition targets, which is vital for establishing and sustaining peace. 12 Agreed in November 2006. 3 Sector Developments 1.10 Agriculture remains subject to the vagaries of weather – as much land remains un-irrigated. Agricultural growth was 3.5 percent during the conflict and 2.6 percent post-conflict, largely because of the weather conditions. Agriculture employs about three-quarters of the Nepali labor force and its GDP share is 32 percent, down from around 40 percent at the beginning of the conflict (1996). Even though agriculture is still the largest contributor to GDP (and the largest employer) there has been little or no new investment for mitigating the effects of weather. The investment in agriculture and irrigation averaged 0.52 percent and 0.55 percent of GDP between FY07 and FY09. 1.11 One significant change in the agriculture sector is the reduction in labor supply through outmigration. About one-third of working males are estimated to be abroad at any time, and most are from rural areas. In many communities women, children and the elderly carry out farm work. As a result, rural wages are increasing rapidly, especially those of working-age males. It is not clear yet how this would affect the technologies employed, and what the pattern of optimal public sector intervention could be. 1.12 The industrial sector, particularly manufacturing, stagnated during the conflict and has not recovered. Industry as a whole has grown by 3.5 percent in the past 15 years and now contributes about 15 percent of GDP. There was a clear break in the growth rate in FY01, when the conflict intensified and growth slowed to 2.5 percent from 5.2 percent. The Figure 1.1: Manufacturing Real Value Added major culprit was 10.65 manufacturing, which 10.6 stagnated throughout 10.55 10.5 FY01–FY09, after 10.45 having grown at 5 10.4 10.35 percent per year during 10.3 FY96-01. The 10.25 manufacturing share of 10.2 10.15 GDP kept declining, 10.1 from 9 percent in FY96 FY97 FY98 FY99 FY00 FY01 FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY01 to the current 6.8 Log of Manufacturing Value Added percent. The sub-sector has tended to be the first casualty of political instability. Production has been disrupted by acute power shortages, other infrastructural problems, frequent strikes (bandhs), transport blockages, extortion, and labor disputes. The appreciation of the real exchange rate in FY10 has also hurt manufacturing. With regard to other sub- sectors, construction grew by 3.7 percent during FY01-09, and kept its GDP share constant at about 6 percent. Electricity, gas, and water, driven by government investment, grew by 7 percent on average. 1.13 The services sector has been the main engine of economic growth. In the past 15 years, the sector has grown by 4.8 percent a year – 5.5 percent before the conflict intensified and 4.3 percent thereafter. Its contribution to GDP has risen to 52 percent from 46 percent 10 years ago. Between FY01 and FY09, (earlier comparison is not possible, due to reclassification) the subsectors of finance, education, health, and public administration and defense grew fastest. Tourism and the retail and wholesale 4 subsectors recovered after the conflict ended. A more spectacular recovery was seen in the financial and real estate sectors, starting one or two years before the end of the conflict – probably reflecting rising remittances at that time. The education and health subsectors continued to grow throughout the conflict and post-conflict periods while other community services grew rapidly after the peace process began, reflecting the expenditures in local government grants discussed in other chapters of this review. Defense spending has declined but post-conflict public security spending has risen. Table 1.2: Service-Sector Value-Added Growth During and Post-Conflict (FY01-09) During Post- Conflict* Conflict* Average Subsector (Percent) (Percent) (Percent) Retail and Wholesale -0.5 2.1 0.5 Hotels and Restaurants -1.2 5.8 1.4 Transport and Communications 6.0 6.5 6.2 Financial Sector 3.9 15.1 10.8 Real Estate -3.7 7.4 3.1 Public Administration and Defense 11.6 1.7 7.7 Education 10.5 6.5 9.0 Health 9.1 8.9 9.0 Other Community Services 1.6 12.0 5.4 * Calculations for Financial and Real Estate sectors separate “conflict� and “post-conflict� periods in FY04, others in FY06. Large-Scale Migration, Remittances and their Impact 1.14 The considerable migration abroad of Nepali workers and large remittances they send back home are a new feature of the Nepali economy. As discussed above, one-third of the male working population is estimated to be outside the country and they send home at least 25 percent of GDP every year (official inflows in FY09 amounted to US$2.7 billion, or 22 percent of GDP, according to the Nepal Rastra Bank. But this figure excludes inflows from India and through the informal hundi system). The remittances helped reduce the poverty incidence from 42 percent in FY96 to 31 percent in FY04, and higher remittances since must have reduced poverty further. 1.15 Remittance flows have gone into consumption or real estate in the absence of an investor-friendly business environment. The high consumption is boosting imports, and together with the associated real exchange appreciation that has constrained 5 exports, the trade gap has widened. As a result, Nepal has suffered foreign reserve losses for the first time in several years (from 6 months of imported goods and services to 5 months). Part of the foreign reserve loss is attributed to rising capital flight. 1.16 High consumption and associated imports helped raise government revenues though VAT and customs, but the trend could be reversed if remittances slow. High investment in real estate has raised real asset prices, attracting speculators and creating a real-estate bubble. The speculation seems to be largely funded by financial institutions, raising concerns about deteriorating financial-sector assets. 1.17 Inflation has remained high, in double digits, since mid-2008 because of increasing liquidity infusion from remittances and an accommodative monetary policy. C. Nepal’s Development Strategy and Reform Agenda 1.18 Nepal’s strategy for managing post-conflict development challenges – the Three-Year Interim Plan (TYIP, 2007-2010) – is aimed at building a foundation for sustained growth. Its core cross-cutting policies are:  Reconstruction, rehabilitation and reintegration;  Social inclusion, poverty eradication and employment generation, development and application of information technology;  Women-empowerment and gender equality, balanced regional development and human rights (safety nets for children, women, the disabled, Dalits, indigenous nationalities, Madhesis and senior citizens)13. 1.19 Another strategic aim is infrastructure development; national as well as that identified by local populations and built with their participation. The Tenth Five-Year Plan/Poverty Reduction Strategy (PRS, 2003-07),, and the Millennium Development Goals (MDGs) provided guidance for the budget allocation. The interim plan assumed that GDP would grow by an average of 5.5 percent annually (agriculture by 3.6 percent and non-agricultural sector by 6.5 percent), but this was not attained. Financing the plan required revenue to grow by more than 15 percent annually and foreign assistance by 50 percent, neither of which materialized. Implementation remained an issue throughout the plan period; constrained by long-standing issues, such as poor infrastructure, weak institutions, slow reforms, and political instability. 1.20 Inclusive development is a major thrust of the TYIP: Targeted poverty- reduction programs are expanded and deepened, and its Gender Mainstreaming and Inclusion chapters have special programs for excluded groups. The Inclusion policies include:  New programs to eradicate discrimination, untouchability, exploitation, violence and misbehaviour based on caste, ethnicity, gender, region, language, religion and sect; and 13 National Planning Commission 2007: Approach paper for Interim Plan (FY07/08-09/10) 6  Plans to ensure basic human rights of marginalized communities such as women, children, Dalits, Adibasi Janajati, Madhesis, the disabled, senior citizens, and the poor. 1.21 Nepal’s reform agenda aims to improve public-sector efficiency and stimulate the private investment climate. Significant progress has been made in reforming the economy in areas such as revenue administration and expenditure management. The Medium Term Expenditure Framework (MTEF) has been under parliamentary discussion since FY09. It also anchors the forthcoming Three-Year Plan (FY11-13) and has been adopted as a budgeting tool. The business environment, however, has continued to deteriorate owing to the unresolved political issues and the inability of the government to ensure public security and maintain law and order. 1.22 The emerging structural challenges of the proposed ―new‖ Nepal also remain largely unaddressed. There are both political and constitutional commitments to restructure the state into federal units, but modalities of taxation, fiscal decentralization, center-local transfers, civil services, and resource-sharing have yet to take shape. □ 7 2. FISCAL FRAMEWORK AND ISSUES IN AGGREGATE EXPENDITURE Key Messages  The prudent fiscal stance, strong revenue efforts, and generous aid have allowed expansion of public expenditures.  Concerns are emerging, however, about increasing recurrent spending and inadequate quality and size of fixed capital formation.  Transparency appears jeopardized by the rising allocation to transfers, both capital and recurrent.  Addressing the fiscal risks can help strengthen public finance management, which otherwise, could undermine the prudence achieved so far and set off higher borrowing. Summary This chapter reviews the role of public expenditure in various national plans, particularly the Three-Year Interim Plan (TYIP). Section A reviews the expenditure in the plans; Section B outlines the key fiscal trends; and Section C analyses the potential fiscal risks. A. Fiscal Framework and TYIP Implementation 2.1 Rising recurrent spending has emerged as a major concern on the fiscal radar. Sound public finance management is key to managing resources needed for Nepal’s transition, state restructuring, and development. The TYIP (FY08-10) strategy builds on and gives continuity to the strategic pillars of the PRSP/Tenth Plan (FY03-FY07): attaining broad-based growth, improving service delivery, enhancing inclusion and improving governance. Fiscal management has largely been prudent, but current expenditure rose by 48 percent to NPRs 114.1 billion between FY07 and FY09. It is projected to grow by another 48 percent, this time in one year, to NPRs 168.8 billion in FY10. Wage payments and pensions are emerging as a major concern, as is the size and quality of capital expenditures. 8 Table 2.1: Fiscal Framework FY07-FY10* FY07 FY08 FY09 FY10 Budget Projection (Billions of Rupees) Total Revenue and Grants 102.0 125.3 178.1 229.5 225.5 Total Revenue 86.2 104.9 139.8 172.6 179.8 Tax Revenue 72.0 86.2 119.0 150.1 157.6 Non-tax revenue 14.2 18.7 20.8 22.4 22.2 Grants 15.8 20.4 38.3 57.0 42.7 Expenditure 115.8 142.3 191.8 263.1 249.5 Current 77.1 91.5 122.1 160.6 168.8 Of which: Interest Payments 6.2 6.4 9.8 9.7 9.7 Capital and net lending 38.7 50.8 69.7 102.5 80.7 Excluding Shares and loans 30.8 35.5 58.3 86.9 68.2 Overall Balance After Grants -13.9 -17.0 -13.7 -33.6 -26.9 Financing 13.9 17.0 13.7 33.6 26.9 Net Foreign loans 2.5 1.1 -3.7 10.3 3.8 Net Domestic financing 11.3 15.9 17.4 23.3 23.1 Memorandum item: Nominal GDP 728.2 818.4 969.8 1106.9 1116.1 Total Revenue Growth (percent) 21.7 33.3 28.6 Total Spending Growth (percent) 22.9 34.8 30.1 Current Spending Growth (percent) 18.7 33.4 38.2 Sources: The Government of Nepal, IMF, Staff Estimates * Fiscal Year ends in mid-July. Expenditures are net of repayments. 9 Fiscal Sector – Positive Developments 2.2 Prudent fiscal management enabled Nepal to keep net domestic borrowing to less than 2 percent of GDP. The rising expenditures after 2007 have been funded through revenue and foreign aid. Outstanding public and publicly guaranteed debt has, as a result, declined from 47 percent of GDP in 2007 to 40 percent in 2010. 2.3 Available resources have increased with good revenue performance and high aid disbursements. Revenues increased from 12.0 percent of GDP in FY07 to 14.8 percent in FY09. Reforms in revenue administration and strong customs and VAT collection, due to higher consumption from remittances, contributed to this performance. Government policy that allowed taxpayers Figure: 2.1 Revenue - Percent of Total Revenue to declare previously untaxed Year 2008 2009 2010 income (combined with an amnesty) added to the Tax revenue 79.1 82.3 85.2 Non tax revenue 18.4 15.0 12.6 collection in FY09. Income tax Principal repayment by corporations 2.5 2.7 2.2 and VAT have been the focus of reforms, and the two Income tax 17.7 19.4 20.6 accounted for 45.4 percent of Vat 27.7 28.1 29.2 the total collection in FY08, Sub Total 45.4 47.5 49.8 and their share is expected to Excise 10.4 10.8 11.1 reach 50 percent in FY10. The Trade tax 19.6 18.2 18.8 rise in revenue (2.8 percentage Duty and fees 5.5 6.4 4.2 points of GDP from FY07 to Dividends 4.7 4.5 4.5 FY09) was higher than that of current spending (less than 2 percentage points of GDP), resulting in a domestic revenue surplus. However, actual current spending may have grown faster, as much of the increased capital spending is suspected to be of a recurrent nature. Aid inflows also rose after 2007. 2.4 Revenue data for eight months of FY10 (mid-April 2010) suggested that actual collection could exceed the budget target, for the fourth consecutive year. Income tax surpassed trade tax as the dominant source in 2009. Income tax collection in 2009 was 19.4 percent of revenue, higher than trade tax (18.2 percent), which had previously been the largest revenue source. 2.5 Aid has increased and its composition has shifted from loans to grants. In the Ninth Plan period (1998-2002) Nepal received more loans for financing capital expenditures on power and other infrastructures. Aid composition changed during the Tenth Plan (PRSP) period, with grants comprising 2.2 percent of GDP and loans, 0.2 percent. Donor willingness to assist the peace process after 2007 resulted in an increase in grants to 3 percent of GDP. Much of the funds went to the social sector and peace-related activities. The share of loans in total aid declined further to 0.1 percent of GDP in 2009. One reason for the decline in loans is the poor implementation environment and the inability of the government to design and implement large projects. While a large portion of donor support is spent through the national budget, about 47 percent of the total aid was spent outside the budget in FY08 (see Box 3). 10 2.6 Interest payments are declining. Nepal’s interest payments obligations stood at 0.8 percent of GDP in 2009 compared to 1.3 percent in 2003. The obligations are lower because of the policy of maintaining a prudent fiscal stance and the appreciation of Nepali currency against the US dollar at the macro level as well as slow implementation and, as a consequence, low loan disbursement at the project level. 2.7 The largely sound fiscal management has widened the fiscal space, allowing the government to increase public spending. But this achievement hides many fundamental issues: low spending efficiency, rapidly rising recurrent expenditure, and chronically low fixed capital formation. Further, remittances play a large role in revenue mobilization which could thus be affected by variations in remittance inflows. A World Bank estimate suggests that a fall in annual remittance growth to below 10 percent could stretch the government’s capacity to pay for growing current expenses and principal repayments. Lower remittance growth could tilt the balance and set off a sudden increase in domestic borrowing. The FY10 revenue target is 16.3 percent of GDP. So far, foreign aid has also been readily available, having risen from 3.6 percent of GDP in FY07 to 4.7 percent of GDP in FY09. 2.8 Public expenditure has increased by about 2 percentage points of GDP annually from FY07 to FY09. Overall spending rose from an average of 15 percent of GDP during the Tenth Plan from 18 percent in FY07 to 22 percent in FY0914. The increase is a result of the focus on inclusion in public services provided by the government. But recurrent spending is also expanding fast – from 11 percent of GDP in 2006/07 to 13 percent in FY09 and 15 percent in FY10, due to higher government salaries, additional allocations for hiring new teachers, and increased public security expenses. Social spending has also increased from 6 percent of GDP to 10 percent in the same period. Besides education and health, public funds have also been spent on expanding safety nets (support for widows, single women, the disabled and marginalized ethnic groups). Spending on these safety nets doubled during FY07-09, from 0.6 percent of GDP to 1.2 percent. Emerging Issues 2.9 Although the availability of resources and widened fiscal space for higher spending is positive, there are issues that could increase future Figure 2.2: Capital Expenditure and Primary Deficit (% of GDP) risks and jeopardize fiscal management. These include the Capital Expenditures and Primary Deficit ( % of GDP) 12.0 relaxation of prudence in fiscal 10.0 management, higher wage and 8.0 pension payments, a rapid rise 6.0 in non-salary recurrent 4.0 spending, such as transfers (see 2.0 Box 1); eroding size and 0.0 quality of capital spending; and 2002/03 2003/04 2005/06 2006/07 2007/08 2009/10 2004/05 2008/09 Budget -2.0 an increase in solely- -4.0 government-funded projects of Capital expenditure Primary Balance, incl. grants questionable quality. The 14 The government’s reporting format, which includes debt repayments as spending. 11 issues related to budgeting and implementation processes remain largely unchanged since 200715. These include concerns over budget credibility, especially the expenditure outturn compared to the approved budget, monitoring of expenditure payment arrears and poor oversight of fiscal risks, among others (see Box 2). 2.10 Nepal has started to relax its conservative fiscal stance. Both the capital expenditure and primary deficit widened during the PRSP and TYIP periods. The primary deficit, including grants, widened from 0.2 percent of GDP in the Tenth Plan to 1.4 percent of GDP, during the first two years of TYIP (2008 to 2009). The gap resulted from the increased public spending and high growth in current spending. The relaxation of the previously conservative fiscal stance has resulted in increased net domestic borrowing since FY06. Net domestic borrowing was 1.5 percent of the budget in FY09 compared to 0.2 percent in FY05 and is budgeted to increase to 2.2 percent in FY10. Both deficit and borrowing levels are within the ―cautious‖ fiscal framework16. But there also are reasons for concern, owing to the accelerating actual and de-facto current spending and the eroding quality of expenditures because, for example, the less transparent ―transfers‖ are rising. 2.11 The wage bill amounts to a quarter of current expenditure and is the main cause for the increase in spending. At 3.7 percent of GDP, it is not high in regional or even international terms. What is of concern, though, is the growth rate. The government policy (2009) of indexing allowances has contributed to the wage bill increasing from 3.1 percent of GDP in 2007 to 3.7 percent of GDP in 2009 (67 percent of total consumption expenditure)17. The wages, however, are still underestimated by 0.5 percentage points of GDP and do not include teacher salaries18 under the rahat (relief teacher) quota and/or ―funding per child‖ (teacher salaries distributed through the operational subsidy). Further, spending on wages is expected to increase with implementation of the Second School Reform Program (SSRP). Besides wages, civil service pensions contributed 1.1 percent of GDP in 2009. Even though the fiscal burden of wages is increasing, Nepal has not been able to attract and retain high-calibre staff due to the low salary compression ratio (3.76) – compared to the international norm of 5.0 – and relatively low average salaries compared to what the private sector offers at higher levels. 15 World Bank: 2007 Managing Public Finances for a New Nepal: Public Finance Management Review, page 17. 16 Domestic borrowing up to 2.5 percent is the limit agreed with the IMF. 17 The Civil Service Act allows salaries and allowances to be reviewed every three years. 18 Teachers make up 32 percent of the total civil service headcount. 12 Table 2.2: Total Expenditure during the Tenth Plan and the Interim Plan Current Expenditure Trends: Tenth Plan Interim Plan 2009/10 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 Budget Total Current Expenditure 10.6 10.3 10.5 10.2 10.6 11.2 13.2 15.0 Consumption expenditure 4.7 4.5 4.6 4.6 4.8 5.2 6.0 5.7 o/w Total Salary 3.1 2.9 3.0 3.1 3.1 3.4 3.5 3.7 Service and Production Expenses 3.1 3.1 3.2 3.2 3.3 3.5 4.6 5.9 Current Transfers (Grant and Subsides) 0.6 0.6 0.8 0.7 0.9 0.9 1.0 1.4 Interest Payments 1.3 1.2 1.1 0.9 0.8 0.8 0.8 0.9 Others (refunds, office operations) 2.5 2.5 2.4 2.3 2.5 2.6 3.2 3.1 2.12 Funding civil service pensions will remain a medium-term issue. The spending on pensions has increased from 0.5 percent of GDP in 2006 to 1.1 percent in 2009, and is projected to reach around 7 percent of GDP in 2015 – equalling salary expenses19. The pension plan covers government employees only and was rolled out, initially, for army veterans about 70 years ago. This ―defined benefit‖ (pay-as-you-go) scheme added up to 31 percent of the total salary expenses in 2009. Key reasons for such a rapid projected increase include: (i) the 14,103 voluntary retirements in 2009; (ii) indexation of pensions to inflation; and (iii) longer life expectancy of retirees. Also adding to the cost are the relatively early retirement ages: for the army, after 15 years of service; civil servants, police and armed police personnel and teachers, at age 58; and for civil servants generally, encouragement to leave after 20 years of service. The government introduced ―defined contributions‖ to the pension scheme in July 2005, but the cost has kept increasing. There are no other serious efforts to address the increase in pension obligations, which could reduce the fiscal space for financing development. 2.13 Non-salary expenditures also have increased, especially current transfers. Non-salary expenditures remained relatively stable at 7.5 percent of GDP during the Tenth Plan and up to FY08. After that, the policy emphasis shifted toward making the state more inclusive and responsive to underserved regions and populations. Resource flows under various safety-net schemes and for community-managed activities increased in FY09, as current transfers (see Box 1). ―Service and production expenses‖, especially ―consultancy fees‖, also contributed to the increase in public expenditure. 19 Informal Ministry of Finance estimates. 13 Table 2.3: Public Spending and Resource Mobilization Table : Fiscal Trends Tenth Plan Interim Three Year Plan 2009/10 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 Budget Total Revenue (GoN) 11.4 11.6 11.9 11.1 12.0 13.2 14.7 16.5 Revenue 11.1 11.3 11.7 10.8 11.9 12.8 14.3 16.1 Tax revenue 8.7 9.0 9.2 8.8 9.8 10.4 12.1 14.0 Non tax Revenue 2.5 2.3 2.5 2.0 2.1 2.4 2.2 2.1 Grant 2.3 2.1 2.4 2.1 2.2 2.5 3.6 5.3 Fiscal Deficit -3.7 -3.0 -3.2 -3.7 -4.0 -4.6 -5.9 -8.4 Deficit after grants -1.4 -0.9 -0.8 -1.6 -1.8 -2.1 -2.3 -3.1 Loan -0.2 0.3 0.6 0.2 0.3 0.1 0.1 1.0 Net Domestic Borrowing 1.6 0.6 0.2 1.4 1.5 1.9 2.3 2.2 Total Expenditure (GoN) 17.1 16.7 17.4 17.0 18.3 19.7 22.4 26.7 Total Expenditure 14.8 14.3 14.9 14.5 15.9 17.4 20.2 24.5 Current Expenditure 10.6 10.3 10.5 10.2 10.6 11.2 13.2 15.0 o/w interest payment 1.3 1.2 1.1 0.9 0.8 0.8 0.8 0.9 Capital expenditure 4.2 4.0 4.4 4.3 5.3 6.2 7.0 9.5 excluding shares and loans 2.9 2.9 3.2 3.1 4.2 4.3 6.2 8.1 Aid 2.1 2.4 3.0 2.3 2.5 2.6 3.6 6.3 Fiscal Deficit -3.7 -3.0 -3.2 -3.7 -4.0 -4.6 -5.9 -8.4 Primary Balance, incl. grants -0.1 0.3 0.3 -0.7 -1.0 -1.3 -1.5 -2.2 2.14 Capital expenditures are Figure 2.3: Transfers vs. capital formation as % of GDP low and insufficient for attaining growth targets for the infrastructure sectors. The average capital spending during the Tenth Plan was 4.4 percent of GDP. After 2007, it increased to 7 percent of GDP, in 2008/09. However, actual fixed capital formation was extremely low (3.2 percent of GDP in 2009) relative to the country’s needs. This could undermine infrastructure growth targets. 2.15 Capital expenditure includes spending that does not result in capital formation. Prior to FY09, fixed capital formation was less than three percent of GDP. Total capital spending appears to be high in nominal figures, but almost 20 percent reflects the increasing transfers to two loss-making public corporations. Much of the remainder is explained by the ―transfer payments‖ in the capital budget, which are often recurrent in nature (see page 4 and Box 2). 2.16 Though still within the fiscal framework, domestic borrowing has increased, from an average of 1.1 percent of GDP in the Tenth Plan to 2.1 percent of GDP in 14 FY08 and FY09. Although the domestic borrowing level is within sustainable levels, it is on the high side. When juxtaposed with the budget structure, current expenditure amounts to 65 percent, with a growth rate of 17 percent, and this remains an area that needs close attention in case revenue should decline, or associated risks emerge. Box 1: Transfers Transfers (current and capital) to local governments, non-profit organizations, ―financial aid‖, etc. increased in FY08 and FY09. The transfers doubled from 4.4 percent of GDP in FY04 to 8.2 percent of GDP in FY09, and are estimated to reach 10 percent in FY10. The FY09 transfers (8.2 percent of GDP) are equivalent to current spending (8 percent of GDP) and much higher than capital formation expenditure (3.2 percent of GDP). Transfers could be the largest spending category in the medium term. In FY09, transfers went toward funding education (3.3 percent of GDP, mainly salaries), safety nets (0.5 percent of GDP), local governments (1.2 percent of GDP) and others, including PAF (3.2 percent of GDP). Compared to FY08, spending under this category has increased by 46 percent and is expected to increase further raising questions about transparency. The non-transparent nature of the transfers masks actual spending under appropriate budget heads, raises targeting and monitoring issues, and affects operational efficiency. For example, the salaries in education are under-estimated by two percent of GDP and overall consumption expenses by 0.5 percent of GDP. Budget Spending - Major Transfers : Major Activities Economic Heads 100% Current spending net of transfer 90% Capital Formation Grant transfers ( current and capital) 80% Total expenditure 2.5 2.6 2.6 3.0 3.3 26.47 70% Education Sector ( Percent of GDP 22.25 60% total) 19.72 50% Local 18.35 Governments 17.40 16.95 40% 0.5 0.5 0.7 1.2 Safety nets 0.7 0.1 0.1 9.84 30% 0.1 0.1 9.00 0.5 Education salary 7.68 8.02 8.20 7.29 7.02 7.29 20% 5.26 5.62 1.5 1.6 4.37 4.60 4.34 1.6 1.7 2.54 3.18 10% 1.9 2.19 1.96 2.42 0% 2005 2006 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2007 2008 2009 Budget 15 2.17 Revenue growth (2.4 percentage points of GDP from 2007 to 2009) Figure 2.4: Recurrent Expenditure and Revenue as remains lower than that of current Percentage of GDP spending (2.6 percentage points of GDP), resulting in a deficit of resources for financing capital expenditures. A World Bank staff estimate suggests that if remittance growth drops below 10 percent per year, it could result in lower revenues. (See diagram). 2.18 More funds have become available for government-funded projects because of higher aid disbursements. In the three years leading to FY10, there was a 94 percent increase in funds for projects wholly financed by the government (from NPRs 93 billion in 2007 to NPRs 180 billion in 2010). This was made possible by the increase in aid. However, this has raised questions about spending quality because a majority of government funded projects are poorly designed and do not have realistic completion dates. Further, with more aid for financing the priority sectors and activities, the foreign resources may have prompted the government to expand spending on low- priority activities. 2.19 The primary balance (deficit), including grants, widened from 0.2 percent of GDP in the Tenth Plan to 1.4 percent of GDP, during the first two years of TYIP (2008-2009). The gap resulted from the increased public spending, generous but still insufficient levels of aid grants, and lower growth in revenue compared to increased current expenditures. The primary balance deficit – though sustainable with generally low public debt in 2010 (see Box 4) – needs to be monitored in case remittances taper off and lower revenue, which could widen the gap further in the medium term. Continued growth of current expenditures and the possibility of a slowdown in revenue growth remain potential sources of fiscal risk. 2.20 Fiscal outcomes have changed dramatically during the two plan periods. In the Tenth Plan, the rise in current expenditure was offset by increased external assistance and revenue, while domestic borrowing remained low. A different fiscal dynamic emerged in the first two years of the TYIP (2008 and 2009), due largely to the relaxation of the government’s previously cautious fiscal stance: the increase in current expenditures was largely financed by revenue and foreign aid, and domestic borrowing was the source of financing the increase in the capital expenditures. 16 Figure 2.5: Change in Expenditure and Financing in 2007-2003 and 2009-2007 Changes :2007 to 2003 Changes :2009 to 2007 Capital Expenditure Capital Expenditure Current Expenditure Current Expenditure External financing External financing Domestic borrowing Domestic borrowing Domestic revenue Domestic revenue -0.50 0.00 0.50 1.00 1.50 0.00 1.00 2.00 3.00 2.21 Nepal is still a ―moderate‖ debt-distress country. There has been progressive improvement in performance on debt sustainability and therefore Nepal’s risk assessment has improved – from a high to a moderate level of debt distress20. The improvement in sustainability indicators was largely due to an appreciation of the Nepali rupee and lower-than projected disbursements by donors. The most recent joint IMF/World Bank DSA (2010) still places Nepal under a moderate level of debt distress. However, a financial risk scenario, in the context of a delayed impact of the global crisis, weakened macroeconomic variables, structural impediments to growth, and vulnerabilities in the financial sector (and associated risks), could result in breaches of the debt sustainability threshold. Even after accounting for these risks, Nepal continues to be rated as moderately debt-distressed. Debt-related risks are closely tied with fiscal stability, particularly as domestic borrowing requires a tight fiscal stance in the short run, which could be more accommodative in the medium term, provided the vulnerabilities are addressed with better policies aimed at strengthening the financial sector and providing a better environment for stronger, more stable growth. Other Fiscal Risks 2.22 Nepal’s fiscal framework faces risks despite its largely prudent stance. These are related to the outstanding debts of state-owned enterprises, the restructuring of two state-owned banks, the rising pension obligations and the quasi-fiscal activities of semi-state institutions. 2.23 Outstanding debt and the cost for restructuring state-owned enterprises remain high at 4-6 percent of GDP. Two state monopolies, Nepal Electricity Authority and Nepal Oil Corporation, reported cumulative losses equivalent to 2 percent of GDP in 2008. A 2007 government-commissioned study21 rated both NEA and NOC as ―very bad‖, the lowest in a five-category rating. Recapitalization of two state banks would cost an estimated 2-3 percent of GDP. 20 Joint IMF-World Bank Debt Sustainability Analysis of 2007. 21 High-Level Study for Improvement of Public Corporations. 17 2.24 Significant off-budget activities, often driven by donors, also remain as risks. Typically, the off-budget expenditures are recurrent in nature and can have serious implications on treasury management. Many development boards, commissions22, and the welfare fund23 of the army and police forces remain outside the budget, even though they have both revenue and expenditure implications. These activities have budgeting and accountability implications and could harm overall governance. Another, possibly single-largest, source of risk is the off-budget external assistance estimated to be equivalent to aid flowing through the budget. This has high recurrent cost implications especially after programs financed in this manner are handed over to the government.□ 22 Extraordinary funds spent on these entities could be as high as 3-4 percent of GDP (PFM 2007). 23 Army revenue not accounted in the consolidated fund could be 2 percent of GDP (PFM 2007). 18 Box 2: Budget Implementation There are major concerns about the budget, including: (i) size – it is expansionary; (ii) composition – recurrent expenditures are increasing while capital formation is very low; and (ii) implementation capacity – especially due to deteriorating law and order and the late approval of annual plans. On average, 15 percent of capital and 10 percent of recurrent expenditures remained unspent between FY03 and FY09. If one discounts the last trimester reallocation from slow- to fast-spending budget lines, the estimated capital expenditure in FY09 amounts to only 70 percent of what was budgeted. This level of capital spending can jeopardize future growth. The low spending also points to unrealistic implementation planning. Also the opportunity-cost of not being able to allocate more funds to better- performing activities remains significant. The adverse impacts of the late approval of annual plans are high, while the efficacy of mid-term budget evaluation is low. Last-trimester capital spending remained high in FY09 (74 percent of total capital expenditure), leading to questions about the sustainability of outputs and their quality, poor implementation plans and high virement of funds, and virement of funds resulting from re-budgeting during implementation, which affects the predictability of funds. Aware of the cost of bunching up last-trimester expenditures, the government in FY10 announced many first-time, innovative measures to improve implementation. These included: (i) Multi-year contracting for priority projects to enhance budget predictability; (ii) Third-quarter fund virement from low-spending activities to performing projects for strengthening performance-based allocations; (iii) Multi-location tender submissions, including e-tenders to prevent local disruption of tendering; and (iv) Announcing 36 core monitoring indicators in 11 sectors, drawn from annual plans, to enhance accountability and link strategy to outputs and outputs to budget. Recurrent Expenditure Budget Target & Actual Capital Expenditure Budget Target & Actual (As % of GDP) (As % of GDP) 20.0 14.9 12.0 9.5 9.8 13.4 12.4 11.8 12.0 15.0 11.5 11.6 11.5 10.0 6.8 8.0 6.2 5.4 5.5 5.4 5.7 10.0 12.7 11.2 6.0 7.6 10.6 10.3 10.5 10.2 10.6 5.0 6.5 4.0 5.5 4.5 4.3 4.6 4.5 2.0 0.0 0.0 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09* 2009/10 Recurrent Expenditure (w/o repayments) Budget Target 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09* 2009/10 Recurrent Expenditure (w/o repayments) Actual *Actual figures are revised estimate of 2008/09 Capital Expenditure Budget Target Capital Expenditure Actual *Actual figures are revised estimate of 2008/09 . Third Trimester Expenditure as percent of Total Expenditure 71.24 73.75 65.10 65.53 59.02 54.34 43.80 40.96 41.91 41.49 2005 2006 2007 2008 2009 Capital expenditure Recurrent expenditure 19 Box 3: Trends in Non- Budgetary Aid and its Impact Accurate, comprehensive and transparent reporting of foreign aid in national budgets helps improve public accountability; resource allocation overtime; donor alignment, and more robust budget processes. In 2010, foreign aid financed 28 percent of the Nepal budget and 74 percent of capital expenditure. A significant portion of the foreign aid is outside the budget; for example, in 2008 for every one US dollar of aid flowing through the budget there was US$ 0.60 in aid outside the budget. The difference between ODA disbursements recorded in the OECD’s Development Assistance Committee (DAC) database and the aid disbursements recorded in government budget data provides a crude proxy for measuring off-budget aid—which includes, among others, certain types of technical assistance and aid to NGOs. As shown in graph below, ODA disbursements appear to be consistently higher than the aid disbursements accounted for in the government’s budgetary data. In 2008 (the latest year available in the DAC database), aid reported in Nepal’s budget was only 63 percent of total aid to the country, as reported to the DAC by donors (the average during 1975-2007 has been around 65 percent). In some years, the discrepancy has been even more significant; for example, in 2003 the budget captured only 44 percent of total aid recorded in the DAC database. Trends in non-budgetary aid to Nepal, 1975-2008 800 Total aid recorded, in million US$ 700 600 500 400 300 200 100 0 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 total aid recorded in budget total net disbursements reported by donors (recorded in the DAC database) Source: OECD DAC; Budget data (actual) 1. Donor data reported in the OECD DAC database follow calendar years, while aid reported in the GoN budget follows fiscal years (July-June). An exact comparison based on this data therefore may not be fully appropriate; this chart aims to show the different trends. The DAC database reports disbursements by DAC and non-DAC donors and multilaterals. The DAC donors’ assistance has been close to 70 percent of total net disbursements between 2000 and 2008. While the coverage of the DAC donors’ and the multilaterals’ assistance is comprehensive, non-DAC donors’ contributions are not as well-recorded. For example, in the case of Nepal, the contributions of significant donors, such as India and China, may not be fully captured in either the budget or the DAC database. The high off-budget aid may indicate donors’ concerns with state’s implementation and oversight capacity, and therefore a desire to implement programs outside the formal government channels. The government of Nepal has a policy of allowing direct implementation by donors, but requires them to report their spending comprehensively at the end the fiscal year. This, however, is seldom followed up, making it difficult to align investments to outcomes. Availability of foreign aid has enlarged the fiscal space and improved the government’s ability to finance social expenditures and reach the MDGs. However, the presence of a large external budget makes it difficult to align public expenditures with the government’s anti-poverty strategy and undermines the primacy of the budget as the principal instrument of national policy. As budget absorption capacity increases, the donors should: (i) In timely fashion, share information on disbursements and commitments 20 thewith government; (ii) Align their objectives with national objectives and priorities; and (iii) Progressively channel more of the external aid through the budget, giving national authorities control over a larger share of resources. 3. SECTOR EXPENDITURES AND OUTCOMES Key messages  Inclusive development and donor support of the peace process has translated into more aid. This, together with impressive revenue collection, has increased funding levels to sectors that expand services to underserved populations.  Funds have begun flowing to new sectors. But it is important that decisions are based on institutional capacity, transparency and structured governance, and the trade-offs are rationalized on the basis of priorities. Summary This chapter reviews the spending by sectors, its recent trends and challenges to meet the planned outputs. Section A reviews the allocations to and within major sectors and subsectors, and the status of aid-funding and pro-poor expenditure. Section B examines the choices that are made. A. Allocation of Budget by Sector 3.1 Sectoral allocations have been poverty-focused and inclusive. The Poverty Reduction Strategy (PRSP) or the Tenth Plan (2002 to 2007) aimed to shift more Figure 3.1: Public Expenditure Major Sectors resources towards social and general % of GDP administration sectors while lowering 14.00 Public Expenditure - Major Sectors expenditure in the economic sectors. 12.00 Percent of GDP 10.00 3.2 Spending in the social sector 8.00 averaged around 6 percent of GDP, and 6.00 was marginally higher than in the Ninth 4.00 Plan. Spending in general administration 2.00 increased from an average of 1.65 percent 0.00 of GDP in the Ninth Plan to 1.95 percent of Ninth Plan Tenth Plan Interim Plan Budget 2010 GDP in the Tenth Plan. This increase Social Services Economic Services ( two years) General Administration reflected higher spending on security- related activities. On average, there was a 2 percentage point decline in economic-sector spending between the Ninth and Tenth plans, as more capital-intensive projects matured or were completed. The social and political conflict that severely affected program implementation during the Tenth Plan, also lowered the absorption capacity of the economic sector. 3.3 The TYIP sharpned focus on inclusion and strengthened spending for inclusive service delivery. It marked a deepening of the resource allocation shift, 21 enabling the state to deliver services more inclusively compared to previous plans (see below). Implementation of budgetary reforms recommended by Public Expenditure Review Commissions 2001 and 2005; alignment of aid with sector strategy through the MTEF; and the translation of policies into resources for addressing public demands for more state services to underserved regions and populations collectively contributed to higher resource flows to services. There was also increased focus on inclusion. Major sectoral resource shifts under the TYIP are as follows:  Spending on education increased from an average 2.8 percent of GDP in the Tenth Plan to 3.5 percent of GDP in the first two years of the TYIP, and was projected to reach 4.2 percent of GDP by the end of FY10. The Education-For-All (EFA) strategy is anchored by the SWAp, and more than one-half of the education sector’s resources go toward primary education, with more teacher deployment and scholarships to underserved regions (see Chapter 4).  Health sector spending averaged at around 1.4 percent of GDP in 2009, after averaging less than 1 percentage point during the Tenth Plan (the annual averages in Ninth and Tenth plans were 0.91 and 0.88 percent respectively). The 2010 budget estimated health spending would be 1.7 percent of GDP. The health SWAp seeks to deepen the reach of essential health care services (in the rural areas). One half of the spending in health goes into services that are preventive in nature and assist in reducing maternal and infant mortality rates and increasing immunization coverage, contraceptive use, and improving nutrition (see Chapter 5). Figure 3.2: Social Services - Sectoral Trends (% of GDP) 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 Education Health Drinking Water & Local Other Social Sanitation Development Services Ninth Plan Tenth Plan Interim Plan ( two years) Budget 2010  Spending on drinking water and sanitation is increasing but still remains less than 1 percent of GDP. Average spending in this sector in the Tenth Plan was 0.45 percent of GDP and was budgeted to increase to 0.8 percent in 2010. The yearly outbreaks of waterborne diseases – especially in districts of the far-Western regions and core city areas with little or no access to adequate piped water – can be attributed to this low investment. In line with the policy of expanding access to drinking water, an average 67 percent of the sector’s spending goes to rural and community-managed 22 drinking-water activities. This mode of implementation was adopted to ensure sustainability through collective decision-making. However, the project-completion rate remains low, as it is demand-driven and requires full community participation.  There has been a shift in local development spending after the Tenth Plan, at both aggregate and subsector levels. Spending rose to 1.5 percent of GDP in 2009 from less than 1 percent of GDP in the Tenth Plan. (Tenth Plan average spending in local development was 0.9 percent of GDP). Rural infrastructures, targeted programs, and grants to local governments benefitted from this policy. There has been a doubling of spending on targeted programs from an average of 18 percent in the social sector as a whole during the Tenth Plan to 35 percent in 2010 (see Chapter 6). However, the absorption of the grants declined in 2009, due largely to the absence of elected local governments, the reported capture of resources by interest groups and/or stalling of implementation of infrastructure development activities24. Besides the ―other‖ social sector activities, government spending on cantonment management and peace and reconstruction also contributed to the increased expenditure in this social sector sub-category.  Investment in agriculture has been low, despite the high dependence25 of the rural population on farming for food security. Investment in the agriculture sector, inclusive of irrigation, is under 2 percent of GDP. The availability of subsidized cereals across the border in India, migration of people in search employment within Nepal and across national borders, unavailability of extension services when required, slow expansion of irrigation systems and failure to implement the “Agriculture Perspective Plan� have collectively contributed to declining share of the sector’s contribution to GDP26. Figure 3.3: Economic Services - Sectoral Trends percent of GDP 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 Ninth Plan Tenth Plan Interim Plan ( two years) Budget 2010 24 In 2009, local government spending, under grants transfer, declined to 30 percent of total sector spending from an average of 50 percent of total. 25 About 70 percent 26 PFM, 2007 23  The ratio of spending on the economic sector vs. the social sector declined from 1.1 to 0.6 between the Ninth and Tenth plans and hovered at around 0.5 during the TYIP, in favour of the social sector. Low investment in infrastructure27 and increased resource-flows to the social sector carry significant trade-offs and may therefore require a policy realignment for securing the right balance. This is true especially if infrastructure deficit is an obstacle to growth. The ratio of infrastructure spending to social-sector expenditures declined from an average of 0.6 to 0.4 between the Ninth and the Tenth plans and to 0.2 in the TYIP.  Road connectivity between and within districts has been a TYIP focus, and the investment on feeder roads28 doubled that of the Tenth Plan period. This investment is still underestimated because the Ministry of Local Development (MoLD) spends almost the same amount on this road category. The quality and strategic alignment of local-road expansion under the MoLD is often questioned, because of the low technical capacity of local governments; roads built by local populations are often washed away by the Monsoon rains. The under-spending on road rehabilitation and maintenance also raises questions about the quality of the infrastructure.  Rural electrification from the central grid and off-grid has remained a government focus since 2006. Nearly half of the investment in hydropower has a rural focus, and there has been an increase of 10 percent in total spending in this subsector, compared to that of the Tenth Plan. Large- and medium-scale hydro projects, as a subsector, began to make a comeback in the power sector from FY10. Nearly 50 percent of the budget that year went to this subsector, to address the nation’s chronic power shortages. The policy makers need to address the potential crowding-out effects of large investments in big and medium-sized hydro projects. Investment on rural electrification in FY10 was half that of 2009.  Defense spending declined after 2006 but that on maintaining public security (police) increased. The post-conflict period has witnessed a large number of protests and strikes, which has raised the costs of maintaining public security. 3.4 Government financing of the budget has increased thanks to impressive revenue growth and aid availability. Across sectors, the government contribution to sectoral funding increased in 2010 compared to the 2007 budget. There has been a marginal rise in aid availability as a percentage of the budget (27-28 percent). The composition of aid has also shifted towards cash. Growth in remittances has translated into higher revenues on taxable consumption. Higher resource availability has allowed the government to expand services. While the lower dependence on aid is positive, it can also raise the risk of non-implementation of activities in the eventuality of a decline in revenue. 3.5 Both aid and government funding availability raised fund flows to government-funded projects which, however, tend to be poorly designed. The 27 Infrastructure – power, roads and communication. 28 Investments made through the Department of Roads. 24 government has reduced the number of fully-funded projects29 by 15 percent between 2007 and 2010, but their share of the total funds has increased. This raises questions about efficient use, particularly in the absence of feasibility studies and because activities are open-ended without timeframes for completion. Improved planning for the use of public funds could lessen the delays in replenishment by government of counterpart funding for aid-supported activities. Table 3.1: Sectoral Spending Trends in Percentage of GDP P Sectoral Spending Trends in Percentage of GDP Tenth Plan Interim Plan Budget 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 Social Services 5.37 5.36 5.22 5.15 5.49 5.57 6.36 6.96 9.64 11.71 Education 2.50 2.84 2.68 2.67 2.91 2.94 2.95 3.29 3.74 4.18 Health 0.83 0.88 0.78 0.78 0.84 0.93 1.07 1.25 1.45 1.68 Drinking Water & Sanitation 0.55 0.42 0.44 0.48 0.34 0.42 0.56 0.57 0.68 0.81 Local Development 1.05 0.81 0.98 0.81 0.85 0.82 1.20 1.11 1.94 2.27 Other Social Services 0.44 0.41 0.34 0.41 0.56 0.47 0.58 0.74 1.83 2.77 Economic Services 5.35 4.48 3.82 3.72 3.89 3.69 3.68 4.74 4.75 5.96 Agriculture 0.55 0.60 0.40 0.37 0.37 0.41 0.43 0.43 1.11 0.72 Irrigation 0.93 0.69 0.48 0.46 0.40 0.44 0.48 0.50 0.61 0.71 Forestry 0.30 0.36 0.33 0.33 0.34 0.28 0.26 0.26 0.26 0.31 Industry 0.13 0.19 0.17 0.12 0.10 0.08 0.08 0.08 0.10 0.13 Other Economic Services 0.43 0.37 0.30 0.48 0.40 0.60 0.53 1.58 0.58 0.78 Infrastructure 3.02 2.27 2.14 1.96 2.28 1.89 1.90 1.88 2.09 3.31 Transportation 1.25 1.08 0.83 0.84 0.78 0.71 0.95 0.95 1.19 1.77 o/w Air Transportation 0.12 0.04 0.00 0.00 0.00 0.01 0.01 0.02 0.05 0.11 Communication 0.22 0.23 0.52 0.23 0.26 0.20 0.19 0.20 0.23 0.21 Power 1.55 0.96 0.79 0.89 1.24 0.97 0.76 0.73 0.66 1.32 General Administration 1.71 2.00 1.85 1.71 1.80 1.91 2.50 2.20 1.88 2.02 o/w Police 1.16 1.30 1.25 1.16 1.24 1.23 1.30 1.53 1.22 1.31 Defence 0.84 1.25 1.48 1.56 1.84 1.71 1.51 1.36 1.49 1.38 Others* 4.80 4.34 4.69 4.52 4.37 4.08 4.33 4.45 4.50 4.63 Total 18.08 17.43 17.07 16.66 17.40 16.95 18.38 19.72 22.25 25.69 * Royal Palace, Constitutional Body, Loan Payment and Miscellaneous 3.6 Pro-poor funding – defined as activities benefitting rural populations – is now close to 50 percent of total expenditure, having increased as a share of the budget since 2008. The increase has gone to the social sector, and both Table 3.2: Pro-Poor Expenditure (as %of Expenditure) education and health spending Pro - Poor Expenditure ( as percent of expenditure) stand out for their rural focus. The TYIP ( two Budget MoLD is in the process of Tenth Plan years) 2009/10 deepening targeted programs Social Sector 24.87 31.92 35.84 which, with increased block Education 13.57 14.35 14.07 grants, could directly benefit large Health 3.34 4.60 4.95 rural populations. Pro-poor Other social sector 1.56 4.12 6.38 Local development 5.00 6.84 8.61 expenditure under the MoLD has Economic Sector 8.29 10.29 10.42 increased from an average of 5 Agriculture( incl. irrigation) 1.44 2.08 1.87 percent of total expenditure in the Transport 2.25 2.85 4.39 Tenth Plan to 10 percent in 2010. Power 1.78 1.45 1.20 29 Budget lines 25 Figure 3.4: Financing 2007 and 2008 Budget Financing 2007 Budget Financing 2009 Budget Defense Defense Genreal Administration Genreal Administration Power Power Roads Roads Agriculture ( incl.irrigation) Agriculture ( incl.irrigation) Local Development Local Development Drinking water and Sanitation Drinking water and Sanitation Health Health Education Education 0.00 20.00 40.00 60.00 80.00 100.00 120.00 0.00 20.00 40.00 60.00 80.00 100.00 120.00 Cash aid Non cash aid GoN Cash aid Non cash aid GoN The coverage of the Poverty Alleviation Fund and the deepening of its activities contributed to a three-fold increase in ―other social activities‖ between the two plan periods. In the economic sectors, the investment in agriculture roads has steadily increased the spending in transport, helping to improve rural connectivity. B. Moving Ahead – Making Choices 3.7 Nepal’s peace process and economic development need high investment. The government needs NPRs 1,019 billion30 for implementing the forthcoming Three- Year Plan (2011-2013), of which foreign aid is estimated to be around 24 percent. The rehabilitation cost of former combatants31 alone could cost about 1.0-1.5 percent of GDP, and subsequent steps in the peace process (reintegration and skills development) could add another 1 percentage point of GDP, in addition to the proposed severance packages. Reform and strengthening of the security sector will also entail significant additional costs. An investment outlay of NPRs 55 billion per annum through 2015 to meet the MDG targets is a good reference for the potential resource demand under the Three Year Plan. Funding these activities, and staying within the finite envelope, will require prioritization of programs and increased resource availability, sequencing and scaling-up of activities – as well as a MTEF to ensure adequate funds for priority programs. The financial planning and budgeting tools are in place but need to be strengthened and implemented beyond central planning and budgeting institutions – at the line agency levels. 3.8 Nepal does not need to worry about trade-offs in investment levels between sectors or subsectors but about efficiency gains, which can be achieved with effective implementation. The choice is not about whether to invest more in the economic or social sectors, or whether more should be spent on primary or secondary education, or if the EHCS is more important than curative care. Nor is it an either/or choice between current and capital spending. Nepal has increased investment on safety nets and community-managed development (see Boxes 4 & 5), but these investments also need to put to the efficiency test. Raising the efficiency of investments is possible by having sound sector strategies to start with, and through better planning and budgeting, implementation and monitoring. These steps are suggested:  Planning and budgeting effectiveness can be attained by requiring ministries to submit procurement plans with the Annual Work Program 30 Three-Year Plan – NPC, 26 April 2010. 31 Strength of former combatants was 19602 in 2009. 26 Budget for approval. This may require advancing the budgeting preparation process from six months after the start of the fiscal year to three months;  Freeze allocations for non-implemented activities and end virement of funds after nine months from the start of the fiscal year; and  Strengthen monitoring by linking spending to outputs and making it the basis for sectoral allocation for the coming fiscal year by using the feedback to inform the planning and budgeting process. 3.9 Proper prioritization can ensure fund predictability for national projects. Budget prioritization has ensured adequate funds for priority projects to some extent, but since 2008 over 80 percent of the budget has undermined by priorities defeating the intent of the exercise. The high number of priority projects and ―rationalization‖ of fund flows during implementation – which has always been far below allocation – has led to disenchantment in the line ministries. This remains an obstacle to bringing about noticeable behavioural changes from need-based budget planning to availability-based budgeting. The uncertain resource availability continues to encourage line ministries to propose budgets above their implementation capacity – in anticipation of cutbacks. Other pending issues are the tendency to categorize government-financed projects as Priority One activities while pushing aid-funded projects as Priority Two activities in the expectation that donors will continue to fund them. This approach to prioritization can be mitigated if the resource ceiling of Priority One projects is limited to the cash budget and funds are released against reported outputs. 3.10 The current MTEF exercise has set new criteria for tightly aligning resources to the Three-Year Plan. The new provisions follow the nine pillars of the Three-Year Plan with emphasis on inclusive development and employment. The criteria represent an improvement in capturing national priorities but fall short in addressing the following issues: (i) supporting sectoral strategies, other than outcomes of the Plan; (ii) implementation capacity of the line ministries; (iii) cross-sectoral leveraging of outputs that link to outcomes; and (iv) as noted above, efficient use of limited cash. 3.11 Nepal has yet another choice to make: deciding on a federal model that is not only acceptable politically but that also makes economic sense in decentralising authority to local units. There has been some discussion about the political structure of the federal arrangement, but the economic viability of the units and their inter- linkages remain to be debated and formalized. (see Box 6). 27 Box 4: Safety Nets In 2008, under the TYIP, the government expanded public service coverage to reach underserved populations. The expenditure on safety nets increased from an average 0.6 percent of GDP in 2008 to 3.0 percent of GDP in 2010, or 11 percent of budget. This policy shift has benefitted widows, the disabled, people over the age of 70 years, and new and young mothers. The support has also funded nutrition programs for children. The spending has had positive impact – cash transfer – and increased state presence to the underserved regions. But, with reported poor targeting and monitoring of these activities, a fresh look in targeting and strengthening of monitoring systems may be necessary before further expansion and/or increase in the levels of cash transfers. Social Protection Expenditure ( as pertcent of GDP and Total Expenditure) 14.00 12.00 0.01 0.03 10.00 0.01 4.02 0.03 1.34 8.00 6.00 0.01 3.23 0.00 0.02 3.86 1.35 1.62 4.00 0.85 1.34 0.00 0.00 0.00 0.00 2.9 4.30 2.00 1.60 1.80 1.67 1.83 3.53 2.0 3.33 2.63 1.1 0.5 1.0 1.26 0.5 1.07 0.5 1.02 1.08 0.5 0.00 0.01 0.01 0.03 0.03 0.01 0.03 0.08 0.19 0.15 0.01 Labor Market Interventions Pension Social Safety Nets 1 2 3 4 5 6 7 8 Education and Health Social Funds Social Care Service Subsidies % of GDP 28 Box 5: Community-Managed Development Since 1998, the government has been trying to channelize self-help, a characteristic intrinsic to Nepali society, for development. The Local Governance Act was the first step towards taking resources closer to the people. It became a major policy focus after the conflict intensified at the turn of the century, and has continued after 2006. Resource flows to communities for implementing local, need-based programs, rose from 1 percent of total expenditure in FY00 to 7 percent in FY09, and is budgeted to increase to 8 percent in FY10. The spending on local community-managed projects is expected to increase further under the Three-Year Plan, which has employment, inclusive service delivery and broad-based development as major outcomes. The impact of the community-managed programs funded so far is visible. About 500 to 750 kilometers of rural roads (incl. track opening) have been added annually; one-third of all schools in Nepal are now community managed; and the community-based integrated management of childhood illness and nutrition covers all of the country’s 75 districts. Further, small-hydropower plants in isolated communities already generate over 4.5 Mw of electricity, and there are hundreds of small units that supply power to specific communities. Nepal has plans to have at least one community-managed micro- hydro unit in each village by FY2010/11. Resources to Community E 7.7% x 6.8% p T e 5.9% o n t d 4.1% a i 3.9% 3.6% l t u 2.0% r 1.5% 1.4% e 1.0% 0.0% FY 1999 FY 2000 FY 2001 FY 2002 FY 2003 FY 2004 FY 2005 FY2006 FY 2009 FY2010 29 Box 6: Federalism and Social Service Provision Nepal announced itself a federal republic after peace was achieved and the monarchy abolished. But the transition has not been smooth; the new form of governance has not been agreed upon and the new constitution has not been prepared. However, concept papers produced by the committees for constitution drafting indicate a three-tier (federal, provincial, and local) government system and propose expenditure allocations to these layers. The shift to federalism could impact how the public services discussed in this report are delivered. Federalism can be efficient in producing benefits at local levels in accordance with residents’ preferences. But when externalities (benefits accruing to more than one local jurisdiction) or redistribution concerns are involved, higher level government, even central, should join the decision-making. Education: Primary education tends to thrive under local management, and higher education is more suited to centralized supervision, because the latter is more closely linked to externalities, such as the labor market and other economic factors. But, because primary education also has externalities (as mobility of population increases) and because the redistribution impact is high, it can benefit from central government’s involvement —- especially in financing, to ensure minimum standards and the access by the poor. Parental involvement is also desirable, which is why Nepal has placed School Management Committees at the centre of education policy. The concept paper on allocation of ―economic powers‖ envisages basic education being administered by local governments, technical education by provincial governments, and higher education (including the policy agenda for Education For All) by the federal government. Health: It is normal for preventive and diagnostic basic health care to be provided by local governments. Curative facilities, especially hospitals and institutions for disease control, have externalities and scale economies that require more centralized government control, and the concept paper takes this position. Preventive care, while tailored to serving local communities, also has enormous externalities, and so may need a measure of central government involvement, e.g., in setting minimum standards. Interrelationships with the private sector and administrative capacities of different levels of the government are also important to consider when assigning health system responsibilities to different levels of government. Recommendations: International practice suggests it is best to leave primary education and preventive health care to local governments, and the GoN seems to be leaning this way. But there are transitional issues to consider: many local governments, especially new ones, are unlikely to have the capacity to handle new administrative tasks, and need to be empowered to do so. The priority should be to: (i) quickly enhance the administrative capacity of the new local governments; (ii) engage with central government to plan and implement minimum standards where externalities and redistribution impact are significant; (iii) allocate responsibilities beyond primary education and basic health care; and (v) carefully design the financing structure – including revenue assignments, conditional transfers, and non-conditional grants – to ensure service levels while protecting macroeconomic stability. 30 4. EDUCATION Key Messages  Nepal will meet the MDG target for the net enrolment (NER) in primary education, but is likely to fall short in producing sufficient high school graduates;  Gender-parity has been reached but not in the inclusion agenda of caste coverage;  Rationalization of deployment of teachers through Per Child Funding (PCF) and elimination of the relief-teacher quota system can ensure cost efficiency in teacher appointments;  Community management has increased schooling efficiency but there are issues about the politicization of the SMCs;  Increases in education-funding levels have expanded current expenditure but raised questions about sustainability. Summary This chapter reviews education-sector spending trends, progress and alignment of resources to education strategy, focusing on primary and secondary education. Section A assesses progress and challenges in the education sector and outcomes so far. Section B reviews spending patterns. Section C reviews planning, budgeting and implementation issues, and Section D summarizes the chapter and the recommendations. A. Education in Nepal: Strategy, Outcomes and Challenges 4.1 The literacy level and gender-parity index of Nepal is comparable to those of regional countries. Nepal’s gender-parity index is 86 and literacy level (15–24-year age group) in 2007 is 79 percent below that of Sri Lanka, Maldives and India, economies much larger than Nepal’s. However, with strong government commitment to education since the early 1990s, Nepal’s government has implemented a series of national programs with impressive outcomes. Figure 4.1: Comparative Literacy in the 15–24-year Age Group, by Percent and GPI Source: EFA Global Monitoring Report 2007 Literacy 15-24 year age (GPI) Literacy 15-24 year age (Percent) 1.20 1.03 1.01 1.01 120.0 1.00 0.89 0.86 98.0 97.0 0.85 100.0 82.0 79.0 0.74 74.0 0.80 72.0 69.0 80.0 60.0 0.60 34.0 40.0 Literacy 15-24 0.36 Literacy 15-24 year 0.40 age, GPI 20.0 year age, % 0.20 - - 31 Primary Education 4.2 Nepal is set to attain the MDG goal in primary education. Progress in primary education has been remarkable and the MDG target is likely to be attained by 2015. The net enrolment rate (NER)32 increased from 66 percent in 2001 to 92 percent in 200933 and is on track to attain Figure 4.2: Primary Education Dropout rates the MDG target of 99 percent in 2015. Nepal’s progress 100 towards ensuring the survival rate to grade 5 in 2006 is 80 Dropout comparable with other South Asian countries (Figure 4.2) 60 (Grade 1), and is estimated to have reached 81 percent in 2007. 40 (%),2006 National targeted programs to encourage students from 20 marginalized groups to complete primary education 0 Survival (scholarships), ―reach out‖ programs of schools under rate to Bangladesh Bhutan Pakistan Maldives Nepal Afganistan India Sri Lanka Grade 5, school management committees, more teachers and early (%),2006 childhood studies in native languages are reasons given Source: EFA Global Monitoring Report for the improvement of the survival rate. 2010 4.3 Decentralized management of schools through community management forms the cornerstone of the government’s strategy to increase equity in access to education across gender, caste and ethnicity. The second SWAp (School-Sector Reform Program for 2010–2014, with an investment of US$ 2.6 billion)) anchors the national plan of action for Education For All (EFA) and aims to: Improve primary education outputs by raising learning achievements to 71 percent from the 53 percent measured in 2009; Increase the survival rate to 79 percent from 58 percent (2009); and, reduce the dropout rate to 2 percent from 18 percent (2009) by 2014. 4.4 Nepal is progressing well towards closing the gap in the Net Enrolment Ratio (NER) in primary education in the region. In 2007, Nepal’s primary education NER was 80 percent, just above that of Pakistan, and is estimated to have reached 92 percent in 200934. Figure 4.3: Primary NER and GER in South Asia 2007 140.0 124.0 103.0 111.0 112.0 111.0 108.0 120.0 91.0 87.0 87.0 89.0 96.0 92.0 100.0 80.0 80.0 66.0 60.0 40.0 Primary NER 20.0 - - - Primary GER 4.5 The disparities across regions, poverty quintiles and ethnic groups have also been narrowed. The progress in the NER for girls is striking – with an increase of 27 percent in seven years (2001–2008) and is now comparable with that for boys. 32 Expressed as a percentage of children enrolled at an appropriate level of schooling. 33 Flash Report – II. 34 Flash report 2009. 32 Similarly, the NER achievements across regions, poverty quintiles and ethnic groups are also noteworthy – they have increased by 20-30 percent across regions, poverty quintiles and ethnic groups in seven years. In terms of religious minorities, primary education enrolment for Muslims increased by 50 percent during the review period, as madrasas (religious schools) curriculums cover primary education subjects. The impressive enrolment is also observed for Terai/Madhesh Janajatis (indigenous people), Dalits and Terai middle castes. The increase in enrolment for these groups ranged from 30-40 percent in seven years. Primary enrolment rates have also increased across regions, and are most pronounced in mid-Western and Eastern regions. The enrolment gaps between income groups have narrowed and the enrolment status of 80 percent of the population is almost at par across different quintiles. The gains in enrolment of the poorer 60 percent are higher than those of the upper two quintiles. However, narrowing the geographical disparity (NER for Terai is 85 percent; Mountains and Hills 90+ percent) will require more resources, better school infrastructures and incentives for both students and teachers to be in school in the Terai districts (Table 4.1). Table 4.1: Primary Level (Grades 1-5) Net Enrolment Rates s Table: Primary Level (grades 1 - 5) Net Enrollment Rates NLSS-I NDHS*† NLFS-I* NDHS† NLSS II NDHS† NLFS - II* 1995/96 1996 1999 2001 2003/04 2006 2008 All 56.6 64.8 71.2 66.2 72.5 78.7 88.2 Boys 66.6 73.3 78.8 71.9 77.9 80.0 89.2 Girls 46.4 56.2 63.3 60.3 66.9 77.4 87.0 Gender Parity Index 69.7 76.6 80.3 84.0 85.9 96.7 97.6 Poorest 38.3 48.8 . 60.9 55.8 76.9 . 2 48.4 59.5 . 61.9 68.4 78.0 . 3 61.3 61.7 . 62.1 80.9 79.9 . 4 69.5 71.6 . 70.8 81.9 77.1 . Richest 76.4 89.5 . 78.8 84.8 82.7 . Urban 71.0 84.1 87.5 75.3 83.0 78.0 92.2 Rural 55.7 63.3 69.3 65.4 71.1 78.8 87.6 Mountains 46.6 58.4 69.8 69.9 76.9 79.6 90.8 Hills 65.3 72.3 79.7 76.7 79.5 86.0 92.2 Terai 51.1 59.1 64.3 57.1 66.4 72.9 84.8 Eastern 59.2 68.3 71.2 65.6 74.0 78.5 87.1 Central 50.8 59.8 65.3 60.5 62.2 69.9 83.5 Western 70.3 73.2 82.4 71.1 84.7 84.5 92.5 Mid Western 52.0 55.8 67.8 68.4 79.8 85.1 92.8 Far Western 46.6 70.4 75.5 72.9 74.2 85.3 91.3 Brahman/Chhetri 70.3 88.4 85.1 83.6 85.1 84.9 93.9 Terai Middle Castes 43.2 58.5 62.9 51.9 61.9 67.0 83.0 Dalits 54.8 50.4 67.1 57.2 68.3 76.7 83.0 Newar 72.1 80.4 91.8 77.2 89.1 89.9 92.1 Hill Janajati 55.5 63.7 74.0 74.3 70.7 81.4 90.7 Terai Janajati 47.5 55.1 69.9 58.2 71.7 82.6 91.8 Muslim 36.3 27.2 45.3 23.6 48.3 46.0 71.6 Other Minorities 48.0 74.4 56.6 74.9 80.8 72.2 82.8 Sources: Nepal Living Standards Survey 1995/96, 2003/04; Nepal Demographic and Health Survey 1996, 2001, 2006; Nepal Labor Force Survey 1999, 2008; Staff Estimates 4.6 Gender parity in primary education has been achieved. The gender-parity index is now 0.98 in 2008 compared to 0.84 in 2001 and is comparable to regional 33 countries (Figure 4.1). Factors that may have contributed to this improvement include, inter alia, the success of awareness and back-to-school campaigns, the rise in income levels, incentive measures such as scholarships, uniforms and textbooks, changes in school management, and overall improvements in infrastructure that facilitate access and retention, particularly of girls (Table 4.1)35. Figure 4.4: Gender-Parity Index at Primary Level l 1.20 1.081.08 1.021.00 1.01 1.01 1.00 0.960.96 0.97 0.96 1.00 0.82 0.78 0.80 0.63 0.60 0.40 Primary NER GPI Primary GER GPI 0.20 - - - Source: EFA Global Monitoring Report 2010 Secondary Education 4.7 The NER for secondary schools has reached 50 percent and is on the rise. The rate for secondary schools increased by 20 percentage points in seven years, from 31 percent in 2001 to 51 percent in 2008. The growth in the net enrolment of girls is higher at the secondary level (girls: 21 percent; boys: 18 percent) and the gap between overall enrolment of girls vs. boys is narrowing –less than 5 percent in 2009, compared to 10 percent in 2001. Overall growth in enrolment across geographical, regional and income quintiles averaged around 20 percent, with notable increases in the far-Western region and the poorer quintiles. However, the NER of richer households is double that of the poorest quintile, and there are further disparities between social groups. For example, the NER for the Newar (an urban ethnic group) is almost three times that of Muslims (a religious minority). The gaps in enrolment between social groups will need special policy attention to reduce the disparities. 35 Past household survey data in Nepal closely tracked administrative records data. These more recent figures will be validated by the ongoing Nepal Living Standards Survey III. 34 4.8 Nepal has made progresses in the secondary school NER compared to neighboring countries. In 2007, the rate was 42 percent in Nepal – ahead of those for Bangladesh and Pakistan. In 2008, it increased to 51 percent. Figure 4.5: Secondary NER and GER in South Asia a 90.0 83.0 80.0 69.0 70.0 56.0 55.0 60.0 48.0 50.0 43.0 45.0 41.0 42.0 40.0 33.0 32.0 28.0 26.0 Secondary, NER 30.0 20.0 Secondary, GER 10.0 - - - - 4.9 The gain in the secondary-level NER is more distinct among the poorer 60 percent than among richer quintiles. During the review period (2001-08), the NER of the lowest quintile grew by 86 percent compared to 18 percent for the richest quintile, though the level is still half that of the richest quintile. This increase in NER of the poorest quintile is largely due to base effect, access to affordable schooling, rising incomes, targeted scholarship programs, awareness campaigns, and participation of CBOs/SMCs, which reach out to parents and students in their respective catchment areas. The NER among the Newar and Brahmins/Chhetri is double that of Muslim, Dalits and Terai middle castes (Table 4.1). The SSRP puts in place measures to further narrow the NER gap between these caste/ethnicity groups. It is a step in the right direction, and implementation of programs under this component needs to be closely monitored. 4.10 Gender parity in secondary education was 0.91 in 2008. Nepal’s gender-parity index is comparable to neighbouring countries (Figure 4.4), and the progress made is impressive – a 25-points increase from 2001 to 2008. Figure 4.6: Gender-Parity Index at Secondary Level l 1.20 1.071.06 1.061.07 1.00 0.93 0.920.93 1.00 0.83 0.760.76 0.80 0.60 0.380.38 0.40 Secondary NER, GPI 0.20 Secondary GER GPI - - - - 35 4.11 However, a large number of primary-level graduates do not join secondary schooling. More than half of the primary-level students do not join secondary schools, and only one-half of them complete secondary schooling. Plausible reasons for low secondary-school completion rates are, among others: (i) economic – the poor are 8 times less likely to complete secondary schooling than the rich, as low income compels them to join the labor force earlier than the richer households; (ii) tradition – associated with low income, it forces many children to join their elders for seasonal odd-jobs in India, especially those from far-Western and Western regions; (iii) lack of access to technical institutions – this makes job-market skills a more desirable alternative to formal schooling; and (iv) lack of awareness that education can bring more assured and higher returns. This low secondary-school completion rate tends to reduce the growth potential of the country and slows improvements in quality of life. Fewer girls join secondary schools and, among those who do join, fewer complete the 10th grade – 34 percent compared to 38 percent for boys. Continuation of schooling, especially for the underserved groups, continues to be a pressing issue (as indicated on the previous page). Table 4.2: Secondary Level (Grades 6-10) Net Enrolment Rates s Table: Secondary Level (grades 6 -10) Net Enrollment Rates NLSS-I NDHS*† NLFS-I* NDHS† NLSS II NDHS† NLFS - II* 1995/96 1996 1999 2001 2003/04 2006 2008 All 27.7 31.6 36.1 30.8 40.0 46.8 50.5 Boys 34.1 37.7 42.0 35.0 43.5 50.6 52.7 Girls 20.8 25.3 29.8 26.5 36.2 43.1 48.1 Gender Parity Index 61.1 67.0 70.9 75.7 83.2 85.3 91.2 Poorest 9.9 14.0 . 16.8 13.4 31.3 . 2 15.9 19.5 . 19.3 24.6 39.1 . 3 24.2 26.1 . 25.9 42.3 45.4 . 4 34.7 36.0 . 36.1 55.4 56.0 . Richest 52.1 60.4 . 55.5 65.7 65.3 . Urban 48.8 56.3 50.2 49.8 61.1 56.6 65.8 Rural 26.1 29.4 34.3 28.7 36.8 45.2 48.1 Mountains 17.3 22.5 27.8 27.0 42.4 41.7 47.8 Hills 31.1 35.9 37.8 34.7 41.5 54.9 55.2 Terai 25.8 28.8 35.9 28.0 38.3 40.8 46.5 Eastern 33.4 39.2 37.2 31.6 43.5 47.7 51.6 Central 26.5 29.6 32.8 28.8 36.6 41.1 49.0 Western 31.9 36.6 45.0 37.8 48.0 51.5 55.3 Mid Western 21.7 22.6 35.4 26.9 30.3 52.6 46.3 Far Western 17.5 23.1 26.9 25.9 40.6 46.8 49.0 Brahman/Chhetri 38.0 56.9 47.7 55.9 57.5 61.0 64.5 Terai Middle Castes 22.7 23.1 33.3 24.8 31.7 25.4 36.9 Dalits 8.1 9.5 19.2 14.6 21.2 25.4 32.4 Newar 42.4 54.9 50.5 51.4 58.6 61.2 72.8 Hill Janajati 22.1 28.6 32.3 28.5 34.6 41.4 49.5 Terai Janajati 19.7 19.8 36.9 20.4 36.0 36.6 50.3 Muslim 5.1 10.7 17.3 7.7 6.3 23.8 23.6 Other Minorities 24.7 33.6 27.7 32.3 74.7 58.0 47.4 Sources: Nepal Living Standards Survey 1995/96, 2003/04; Nepal Demographic and Health Survey 1996, 2001, 2006; Nepal Labor Force Survey 1999, 2008; Staff Estimates 36 Reform measures implemented under EFA – SWAp I to improve education status 4.12 The key reforms to improve education quality under SWAp I have been: (i) Most importantly, decentralizing the sector by measures such as voluntary transfer of schools to local management committees; (ii) Initiating licensing of teachers; (iii) Opening textbook publishing and distribution to the private sector; and (iv) Providing resources (since 2009) for hiring teachers based on the student enrolment (per-child-fund), as opposed to the previous system of centralized teacher recruitment (see Section B, Education Expenditures). These actions were taken with the understanding that overall education quality can be improved by enhancing school governance, raising quality of teachers, increasing availability of textbooks at the beginning of school years and providing adequate teaching materials. About 85 percent of schools are funded through public resources. 4.13 School management committees have played a major role in reducing the number of out-of-school children. Nepal began transferring school management to local committees in 2002 as a strategy to improve primary education. A preliminary review of a sample of schools under SMCs has shown that they have made major contributions towards a 50 percent reduction in out-of-school children in three years after the school handover. This was possible through increased participation of beneficiaries in parent-teacher associations; involvement of community-based organizations and civil society groups in supporting schools; improvements in school infrastructure, and parents working closely with teachers. Despite these gains, there are concerns about the increasing interest of political parties in school management committee elections. There is a need to understand the political and economic ―incentives‖ that have interested national parties in SMC elections – something that should perhaps be of concern to parents, teachers and students, rather than central political parties. Box 7: Community-Managed Schools By 2010 one third of all schools had been transferred to elected School Management Committees (SMC), which started with the Seventh Amendment to the Education Act in 2001. These give the elected parents authority to appoint teachers, execute the budget, mobilize resources and monitor progress. The SMCs have brought an increase in enrolment and attendance at schools by both teachers and students. The number of out-of-school primary-age children has dropped by 27 percent (Dalits 33 percent, Janajati 29 percent); teacher attendance has risen by 15 percent and that of all students by 38 percent. Girl-student numbers have increased by 7 percent annually in the SMC-managed schools. These trends will be vetted by an on-going evaluation. The successes, however, are not unthreatened. The SMC elections have begun to attract the interest of political parties, and that may not be in the interest of the parents, students or teachers. It is possible the politicians want to use the authority of SMCs, and student energies, for partisan gain. In Gorkha, in late June 2010, for instance, Maoist supporters clashed with Nepali Congress members after the latter's candidate 37 won a SMC election. Study of the political-economy of schools and the political interest in SMCs is urgently required to find out how to prevent political parties from capturing the educational resource. 4.14 Schools in the private sector have contributed to the growth in enrolment. While public schools and government-supported community schools are responsible for most of the education (85 percent of the all schools; 87 percent of total enrolment), the private sector’s role in raising enrolment is vital for Nepal to attain its MDG targets. The private sector is currently responsible for 15 percent of total school numbers and 13 percent of enrolment. It is more active at higher secondary level (Grade 11 and 12) than at primary level; and its involvement is likely to rise further in rapidly-urbanizing towns and cities where disposable income is highest. The approach paper of the Three-Year Plan envisages a 40 percent private-sector investment in education, implicitly raising the sector’s role. Challenges 4.15 The role of public schools in providing education to the underserved regions remains critical for attaining the planned education outcomes. There were increases in the establishment of schools and enrolment in Nepal’s southern plains between 2004 and 2009. Enrolment rose by 3 percent and school establishment by 1 percentage point during the review period. Because of its sheer size and reach, the public sector will remain the backbone of education in far-flung and underserved regions, and in urban centers for those who cannot afford to send their children to private schools. 4.16 Teacher management continues to be an issue in Nepal, even though the country has made tremendous progress in delineating the respective roles of central government agencies, regional, district and sub-district offices, and school management committees, in the recruitment and management of teachers. The key remaining challenge is to ensure that this reform is carried through fully, and not left to languish. Only then will the benefits be fully sustainable. Completing these reforms would require: (i) All centrally appointed teachers to be grandfathered, and new hiring to be decentralized, at school level; (ii) Teacher quality to be ensured through strict licensing and certification procedures; (iii) Teacher service rules and regulations to be amended to provide clear career paths for teachers, with permanent appointment of the best teachers at their schools, ensuring that the central government covers pensions and other benefits; and (iv) Ensuring sufficient legal protections for teacher appointments to forestall intimidation or interference by school management committees. 4.17 These measures should help address issues such as problematic teacher deployments36, mismatched teacher data37, and non-standardized teacher service rules 36 Provisions of the existing Act and Regulations make it difficult for the GoN to effect necessary transfers from schools with low PTRs to schools with high PTRs. Further, there is a historical neglect of certain regions of the country, such as the Terai areas. Such problems can be addressed by ensuring that future budgets channel more resources to historically deprived areas. 37 There is a mismatch between the number of identified teacher posts and salary payments. This may be partly due to ghost teachers in some areas, but more likely it is because grants or operational subsidies made to schools are often used to hire additional teachers under the decentralized rahat or PCF schemes. From FY2010, the rahat (relief teacher scheme) is effectively closed and no new hires will be made under this scheme. Furthermore, these hires are not appropriately recorded as teacher salary payments, which produces a mismatch between the 38 and career paths38. They should help to fully implement the decentralization strategy adopted by the GoN in 2001/02. 4.18 Because of these issues, there are disparities in teacher deployments across regions. For instance, the Terai districts have a pupil/teacher ratio of 63 – well above the national average of 42 and policy target of of 45. The policy of redeploying teachers to regions where student/teacher levels are high from schools with lower student/teacher ratios risks political interference, as explained elsewhere with reference to the Nepal Teacher Association. 4.19 Finally, while Education For All (EFA) is the government’s priority, rising numbers in basic schooling are causing an enrolment bulge and a low completion rate of 36.4 percent at secondary school level. This undermines the quality of human resources required for the country to achieve its higher growth targets. The economic factors enumerated above, and lack of easy access to affordable education, have turned many prospective students away from higher education. The contribution of this population to economic growth can be strengthened by market-driven, targeted skills-development programs at secondary school level to meet the demands for skills abroad and within the domestic labor market. Nearly 300,000 people enter the job market each year, and nearly 60 percent have not completed secondary schooling. If Nepal wishes to move to higher growth, there must be a concerted effort to enhance the skills of its population and reform the labor regulations. 4.20 To summarize, the EFA policy has brought about remarkable progress in education but more remains to be done. Education in Nepal is now more inclusive, but there are still disparities between regions, castes and ethnic groups. Various education indicators suggest there is much room for improvement. Adult literacy is 79 percent, primary school completion is running at 80 percent and that of secondary schools at a very low 40 percent, while the quality of education in general is poor. Further, the school completion rates are lower for lower-income quintiles and underserved castes and ethnic groups, suggesting the need for more and better-targeted investments to reduce these gaps. centralized Personal Information System, with some 108,000 registered posts, and the actual salary payments made, which reflects a teacher cohort of about 200,000. 38 There is a need for a minimum set of standards for service rules to ensure equity in teacher career development, while allowing school management committees to fully exercise their roles and responsibilities. At present the numerous teacher categories complicates teacher management. Furthermore, the use of rahat , or relief, teachers should be a temporary phenomenon, with such teachers eventually being absorbed into regular teacher service or their contracts terminated as they become no longer needed . 39 B. Education Expenditures 4.21 The education sector budget has increased to 4 percent of GDP (2010) from 2.5 percent in 2001. With the continuation of the EFA policy and expansion under the School Sector Reform Program (SSRP), the investment is expected to reach 7 percent of GDP by 2015. Table 4.3: Educational Sector Expenditure: Percent of Education Sector Budget Consolidation of ( 2003 to 2010) primary Tenth Plan Interim Plan education and Budget gradual Sub sectors 2003 2004 2005 2006 2007 2008 2009 2010 Monitoring and expansion of Administration 5 5 4 5 6 4 3 4 secondary Primary Education 59 58 55 56 57 56 60 37 schooling is SSRP strategy. Secondary Education 23 22 24 24 25 26 24 19 Educational In 2009 more Development 27.9* 1 2 5 4 2 1 1 than half of the Higher and Technical Education 12 13 12 11 12 13 12 13 education sector Source: MoF and Staff spending went to Estimates * Large portion of education development is spent on Primary primary Education- teachers' salary education. Secondary education received 25 percent, while less than 15 percent went to higher education. The low investment in higher education is a reflection of the government’s intent to consolidate the gains made through EFA, and supports its policy to encourage greater private-sector participation in higher education. This is reflected in the government’s statements in the approach paper of the Three-Year Plan (2011-2014). 4.22 Ninety percent of the education expenditure is recurrent, with teacher salaries absorbing 78 percent, leaving less than 10 percent for capital expenditures for expansions and/or additions to the existing physical assets39. Expenditure on scholarships and school supplies are maintained at the previous years’ levels. This questions the efficacy of targeted programs to attract and bring more children in to schools, especially those from the underserved population. The low spending on teaching-learning activities (supplies make up just 6 percent) will lower the quality of education being imparted to students. Table 4.4: Education spending - Economic Head s 2009 2010 2005 2006 2007 2008 Budget Budget Current 93 92 93 88 91 91 o/w salary 3 2 2 2 2 2 o/w subsidy transfer (mostly salaries) 79 81 80 79 76 76 o/w Scholarship 0 4 4 3 2 4 o/w supplies 7 4 5 4 7 6 Capital 7 8 7 12 9 9 o/w capital grants 6.6 7.7 6.0 9.8 8.0 8.9 39 Expenditure on teachers’ salaries in Bangladesh, India, and Pakistan are between 95-98 percent of recurrent expenditures, which tend to account for more than 95 percent of total education expenditure. 40 4.23 Salaries account for more than 78 percent of the spending in the education sector. This is an underestimate of true salary cost in education. Much of teacher salary is under current transfer (76 percent) while only 2 percent of the education budget is recorded under the salary economic code. This non-transparency of recording of the actual level of teacher salary under appropriate economic heads Box 8: Teacher Relief Quota (Rahat): has to do with the way the education sector is funded. As a policy, In 2000, the government put a freeze on the creation government provides operational of new teacher posts. With the conflict at that time, subsidies to non-profit institutions internal migration increased and people moved from (schools) and, staying within norms, their birth places in search of better economic opportunities. With the simultaneous increase in the schools can invest this funding as they number of students entering schools, these migrations see fit. In practice, however, teachers are collectively contributed to unmanageably high hired and deployed by central and district student/teacher ratios. The government responded by level institutions (ministry, department, hiring teachers through the ‘teacher relief quota’, or district education offices) under the rahat, as a means of circumventing the hiring freeze. teacher relief quota, and salaries are paid This practice still continues, but with a difference: through operational subsidy. This non- While the rahat hiring practice used to be based on transparent means of hiring teachers student/teacher ratios, nowadays it has been extended to cover all community schools and those under (Box Seven) increases ambiguity in community management. Current records show that financing of salaries and future cost to 38,787 teachers were hired under the rahat quota. the treasury. It also challenge governance in the hiring of teachers, although the In principle, the overall management and distribution World Bank’s project performance audits of the rahat quota resides with the Ministry of have so far not shown significant Education and the department. But since there are no governance problems. Besides the two comprehensive guidelines for the distribution of teacher quotas, ambiguity has crept in about systems of financing teacher salaries ( appointment authority and terms of reference for salary and subsidy codes), little is known teachers. Student/teacher data are not readily of the use of VDC grants and funds available. And, unallocated subsidy fund is carried over allocated from consultancies and other to next fiscal year instead of being returned to the service code (line item 2.07) that also treasury. An investigation committee recently found fund teacher salaries. Currently, it seems, that the distribution of teacher quotas is subject to the government has limited knowledge of influence from lobby groups. the actual cost to the treasury of teacher Source: Investigation Committee Report on Rahat Quota submitted to salaries40. Prime Minister’s Office, March 2010. 4.24 The Personnel Information System (PIS) records 109,098 teacher posts. With the addition of a further 38,787 listed under the rahat quota, the total teaching complement is officially 147,885, at a cost of NPR24.3 billion annually (Table 4.8). This is judged to be a considerable underestimation, as discussed in earlier paragraphs. The 2010 budget has made an appropriation of an additional NPR1 billion to fund teachers through the Per Child Funding modality and, to date, there has been no recruitment through this modality. Over 66 percent of total teacher recruitment is in primary schools and 17 percent in both lower-secondary and secondary schools. 40 Investigation Committee Report on Rahat Quota submitted to Prime Minister’s Office, March 2010,pg-19. 41 Table 4.5: Number of Teachers and Salaries in 2010 0 Teacher strength and salary 2010: Total Teacher pos t Teacher under Rhat quota teachers Total Salary Salary per Salary Salary per Salary (in teacher per (in teacher Salary (in School Level Teacher Million) annum Teacher Million) per annum Number Million) Primary School 80176 12747 158988 22211 3205 144298 102387 15952 Lower Secondary 16204 2781 171624 9636 1473 152864 25840 4254 Secondary 12718 2759 216937 6940 1370 197406 19658 4129 Total 109098 18287 167620 38787 6048 155929 147885 24335 Source: MoF 4.25 Three channels of teacher recruitment have challenged not only the estimation of actual cost of teacher salaries to the treasury, but distribution of teachers according to schooling needs. Firstly, the Personnel Information System does not have a complete name–to-post list, as over 30,000 temporary teachers are hired under the vacant retired teachers’ post discretion of lobby groups. Secondly, due to patronage and influence over rahat quotas by strong lobby groups (Box Seven), recruitment through this channel continued until the government’s policy to recruit teachers under the Per Capita Financing (PCF) modality brought an end to the practice this year. The recent controversy over rahat allocations also brought the issue to the forefront. Thirdly, even after making necessary budget appropriations in the 2010 budget, the PCF recruitment process has yet to take firm root. To ensure transparency of teacher recruitment, there is an urgent need to implement the PCF policy fully. The recruitment of 16,290 teachers under the School Sector Reform Program (SSRP – see section below) could be the best place to implement this policy first. For the PCF modality to be effective, no new teachers should be hired under the rahat quota, and with the retirement of teachers all vacant posts should be terminated, so that the PCF modality is mainstreamed as the only channel for teacher recruitment. 4.26 There is room for improving teacher deployment. For instance, the high concentration of teachers in Hill and Terai regions, and high student/teacher ratios in the Terai, are partly due to poor deployment and historical neglect. Future budgets should correct this by allocating increased resources to the areas that need it the most. In this regard, the government should develop an educational development index to measure educational outcomes in order to channel finances to resource-starved areas. Teacher re-deployment is difficult to achieve, since no other incentives are provided to encourage teachers to move; the direct costs of such redeployment are also borne by the teachers themselves. The strategies of decentralization and per-capita financing are expected to help improve deployment issues in schools in the coming year. 4.27 Scholarship programs have doubled, but targeting issues persist. Allowing communities to select scholarship recipients may even prolong the targeting problem. To ensure equality of access for underserved populations, scholarship programs have tripled in the last five years – especially in 2010, when these rose to NPR1.8 billion from NPR900 million the previous year. It must be noted that the 2005 PER review of 42 the government raised targeting issues for scholarships which could only be vetted on completion of the ongoing review of efficacy of the program. Nevertheless, the scholarships amount to 4 percent of education budget, an allocation that is expected to increase under the SSRP (US$ 44 million). Table 4.6: Scholarship Budget (in thousand NPR) ) Scholarship Budget Rs. 000 Actual Actual Actual 2007/0 Budget Budget Economic Heads/Line Item 2005/06 2006/07 8 2008/09 2009/10 Scholarship under education ministry 744160 822078 812525 953625 1772504 Scholarship under other Ministries 500 1360 16844 20396 94300 Total Scholarship 744660 823438 829369 974021 1866804 4.28 The education sector budget is largely dependent on government funding. In 2010, foreign aid amounted to 31 percent of the education budget – marginally higher than in 2007 (29 percent). Aid has leveraged government funding in the education Figure 4.7: Education Sector--Source of Funding Education: Sources of Funding 2007 Education: Sources of Funding 2010 Non cash 4 Non cash 1 o/w cash grant 14 o/w cash grant 23 Cash 96 Cash 99 Education: Sources of Education: Sources of Loan Funding 2010 Loan Funding 2007 7 11 Grant 24 Grant 18 Government 69 Government 71 0 50 100 150 0 50 100 150 sector from NPR16 billion in 2007 to NPR32 billion in 2010 (or 18 percent of the total 2010 revenue target). Without a dedicated revenue code to finance the rising investment in the education portfolio, there could be financing problems if the aid should taper off. The high fixed costs of the education sector could tax the treasury, not just in terms of salaries but also pensions, unless the government dedicates a revenue code to fund teacher salaries to protect the sector’s funding requisites, as the government plans to expand investment in infrastructure and agriculture during the next interim plan. 4.29 The aid availability has provided a cushion to government to finance project activities that are solely financed by the state. Government funds for such project activities – that have no aid funding – have tripled from NPR7 billion in 2007 to 43 NPR29 billion in 2010. This increase in education funding has benefitted the education administration and, more specifically, teacher salaries (by an additional NPR7 billion), non-formal education and the national literacy program (an additional NPR1 billion) and teacher pension payments. 4.30 National programs that are implemented throughout the country amounted to 22 percent of education spending in 2001 and are declining – now making up 17 percent of Table 4.7:Education Spending by Development total education expenditure. There appears to Regions and Ecological Belts be no attempt at redistribution of this savings Interim by targeting regions with lower investment. Ninth Plan ( Two western regions and the mountain belt Plan Tenth 2007 to still experience lower educational investment 2002 Plan 2008) compared to other regions. The comparatively Development low investment in the regions mentioned Regions above partially explains their low secondary Eastern 20.9 21.1 22.1 Central 35.1 33.8 30.7 education outcomes. However, a causality Western 22.1 22.1 22.3 relationship between investment and Mid Western 12.7 13.1 14.1 education outcomes is hard to establish across Far Western 9.3 10.0 10.7 regions and ecological belts with investment as the only variable. A more detailed analysis Ecological Belts is required to establish the reason behind the Mountain 11.8 12.5 12.3 poor education outcomes in those regions. Hill 57.8 56.5 54.3 Terai 30.4 31.0 33.4 Table 4.8: Primary and Secondary Education Spending by development Regions and Ecological Belts Secondary Education Spending by Development Regions Primary Education Spending by Development Regions and and Ecological Belts Ecological Belts Interim Interim Plan Plan ( Ninth Plan ( 2007 to Ninth Plan 2007 to 2002 Tenth Plan 2008) 2002 Tenth Plan 2008) Development Regions Development Regions Eastern 21.4 21.3 22.3 Eastern 23.3 22.1 22.1 Central 32.5 33.4 30.5 Central 30.0 29.8 30.4 Western 22.5 21.6 21.4 Western 25.1 24.7 24.7 Mid Western 13.8 13.6 14.9 Mid Western 12.1 12.6 12.4 Far Western 9.9 10.0 10.9 Far Western 9.4 10.8 10.4 Ecological Belts Ecological Belts Mountain 12.6 12.1 12.8 Mountain 11.8 11.9 11.5 Hill 53.5 54.4 55.9 Hill 56.7 57.2 53.4 Terai 34.7 33.7 32.6 Terai 30.7 30.7 33.8 4.31 The increase in education spending (1.3 percentage point of GDP) has made little difference in lowering the disparity in primary and secondary education spending across regions and ecological belts. While the education budget increased from 2.8 percent of GDP in 2002 to 4.2 percent in 2010, the net of national level programs – programs that are implemented across the country and regions – the levels of education investment in primary and secondary schooling, across five development regions and three ecological belts have not changed. This raises doubts about whether the investment is aimed at lowering the disparities in these areas. 44 Going Forward: 4.32 The School Sector Reform Program (SSRP) – SWAp II (2009-2014) – finances both the recurrent and the development expenditures of all school education, but with an emphasis on primary education. The program focuses on the three pillars of Access, Inclusion, and Quality in: (i) Basic Education (Grades 1-8) including Early Childhood Education and Development (ECED) and Literacy and Lifelong Learning; (ii) Secondary Education (Grades 9-12); and (iii) Institutional Capacity Strengthening. 4.33 Access and Inclusion in Basic Education focuses on expanding access to all children in Nepal through four main mechanisms: (i) physical expansion of the schooling system; (ii) identifying and reaching the hardest-to-reach students from disadvantaged, marginalized and poor backgrounds; (iii) providing access to safety nets/educational guarantee schemes; and (iv) traditional schooling. In addition to these primary tasks, the SSRP will also help expand the number of centers offering courses on Literacy and Lifelong Learning and the number of ECED centers, which are both school- and community-based. 4.34 The emphasis of SSRP under this pillar – secondary education – will provide access to ―soft technical and vocational education and training (TVET)‖ programs that improve the students’ ability to transit from school to work environments if they choose to drop out after Grade 10. This will at least equip them with a basic understanding of the world of work and the opportunities available to them. 4.35 The final focus in the SSRP is to strengthen and improve the capacity of the schooling infrastructure to deliver on the above. The emphasis will be on improving capacities at the school level with a focus on the School Management Committees and other community-level organizations, and at the district level for improved planning and execution. A major emphasis of the SSRP will be to improve monitoring and evaluation of both the program, and in particular the ability to carry out international standard learning assessments. To focus on improving governance in the education sector, the following components will be implemented against the backdrop of the major reforms that have already taken place in Nepal including: (i) decentralization of authority to schools and school management committees; (ii) grant-based financing; (iii) decentralized recruitment of teachers by the SMC and financed by teacher grants through the provision of grants to schools on the basis of Per Capita Financing (PCF), which makes up 57 percent of all SSRP funding; and (iv) provision of scholarships for children from disadvantaged or marginalized backgrounds who historically have been excluded from the schooling system. 4.36 The SSRP aims to: improve the NER for basic education to 85 percent by 2014, from 73 percent in 2009; increase the basic education completion rate from 41 percent in 2009 to 66 percent in 2014; and, aim to reach gender parity index to 98 percent in 2014 from 96 percent in 2009. To achieve these outcomes, nearly 70 percent of SWAP-II41 is channelled towards basic education, and the remaining 30 percent to secondary education and institutional development. 41 SWAp –II total investment is US $ 2.6 billion. 45 4.37 To improve the educational outcomes across the lagging regions and ecological belts, the new SWAp (SSRP) should focus on redirecting education spending to these two lagging regions (the mid- and far-western regions and mountain and Tirai districts). C. Planning, Implementation and Monitoring Issues 4.38 Teacher management issues need to be prioritized and the reforms around decentralization have to be fully implemented for their impacts to become fully evident. There needs to be consolidation of teacher service rules and reduction in the number of categories of teachers. This would help address some of the issues raised above about the mismatch between teacher numbers in the Personal Information System and the actual payments being made for teacher salaries. The problem of teacher deployment exists for two reasons: past historical misallocations and irregularities and anomalies in allocations based on political patronage, rather than strict quantifiable need. With the move to decentralized hiring, it is anticipated that both these issues can be corrected in the long run. In the short-term, re-deployment within VDCs might be the most politically viable solution for correcting deployment issues. 4.39 The targeting of scholarships also needs more attention, in addition to improving the delivery mechanism. The government spends 4 percent of the education budget on scholarships aimed at increasing school attendance of children from marginalized groups. The targeting of scholarships can be improved to ensure that the poorest and the most-disadvantaged receive the scholarships intended for them42. A rigorous analysis of inclusion- and exclusion-errors in scholarship distribution is scheduled for the first time later this year. 4.40 The voluntary transfer of schools to local SMCs has slowed. The handover of schools to SMCs was a turning point in reforming education in Nepal but only one- third of the schools have been transferred in early 2010. The aim for the SSRP is to complete transfer of 88 percent of schools to SMCs by 2014. A detailed diagnostic study of school management is being planned for this school year to understand better the factors that encourage communities to take over the management of schools and determine what factors of school management help achieve better outcomes. A growing concern is that national-level political parties have shown increasing interest in running SMCs, and there is sufficient anecdotal evidence that such political capture is becoming more common place. This is another issue that needs further investigation. 4.41 The flash reports and the EMIS need to be further strengthened to provide detailed and disaggregated information on the coverage of targeted populations. The flash reports have assisted in informing policymakers of progress during the school year. However, this data is seldom used for planning and budgeting of the education sector, since the outputs are not aligned to expenditure. 4.42 The non-availability of textbooks at the start of the school year also remains a concern. The government’s textbook printing corporation has resisted the opening up 42 Flash Report II. 46 of textbook printing and distribution of school textbooks to the private sector. Introducing a reverse subsidy through grants to SMCs for purchasing textbooks based on need – even at higher prices than those set by the government – can assist the SMCs to ensure textbook availability before the beginning of each school year. 4.43 The disbursement of funds for building schools and/or classrooms is slow. The use of multi-year contracting for construction of schools and classrooms can be expedited only if procurement plans are approved with budget approvals. This is not the case in education sector. Approvals of procurement plans with budget should help raise efficacy of infrastructure investments.□ D: In Conclusion 4.44 The achievements in the education sector need to be protected and expanded focusing implementation in the following areas: 4.45 Strengthening the teacher management to enhance teacher accountability, teacher deployment, and overall schooling quality in Nepal. This requires that a standardization of rules pertaining to teacher recruitment, service rules, teacher professional development be developed in close coordination with representatives of teacher unions. This will require the complete abolition of relief quotas (or raahat teachers) and ensure that all teacher hiring is decentralized to school management committees. 4.46 Expedite the transfer of school management to SMCs. There is a need to protect the gains made in education by SMCs by providing more block grants to schools doing well for improving education quality. An independent, comprehensive social audit of the functioning of the SMCs can further strengthen the accountability and governance of SMCs and assist in mainstreaming SMCs to higher levels of schooling. 4.47 Revisit the scholarship program for better targeting and monitoring. The distribution of scholarship through SMCs and in schools through CBOs can assist in the identification of right beneficiaries and also facilitate faster fund release to the student. Scholarships are a priority of the education and therefore they can be made more effective by the release of one-third of the funds during the first trimester of the budget cycle. 47 5. HEALTH Key Messages  With significant donor dependency, and an expanding package and coverage of services that are free to the client, the sustainability of health sector spending is of concern and requires an in-depth examination of fiscal space questions. This should focus on inefficiencies, including, but not limited to, the role of the state in its provision, purchasing and stewardship roles, the management of drugs and equipment and the management of non-communicable diseases.  To further improve equality in health outcomes, the recommendable allocation to Essential Health Care Services is insufficient by itself. Budget allocations should be aligned with the Gender Equality and Social Inclusion Strategy and should be based more on beneficiary size, accessibility, and service delivery cost rather than, as is practiced now, on the needs of existing facilities.  In order to reduce the Out-Of-Pocket Expenditures by the poor a detailed examination of those expenditures is required to establish for which services, drugs and diseases they are incurred.  Devolution of basic health services to health posts and sub-health posts, managed by the communities and private institutions, can be the way to manage basic health services that are transparent and accountable to the beneficiaries.  Initiation of Annual Work Program and Budget process, with procurement plans, by the month of October and its approval by budget announcement date will improve absorption capacity of the sector. Summary This chapter reviews public expenditure in the health sector. Section A assesses progress in the sector, expenditure and challenges. Section B reviews public spending, and Section C examines planning and budgeting issues. A. Health: Outcomes and Challenges Table 5.1: MDG Achievement Achievement Target MDG/Impact Indicator 1991 1996 2001 2006 2009 2010-11 2015 Maternal Mortality Ratio 539 539 415 281 229 250 134 Total Fertility Rate 5.3 4.6 4.1 3.1 2.9 3.0 2.5 Adolescent Fertility Rate (15-19 years) NA 127 110 98 NA 98 70 CPR (modern methods) 24 26.0 35 44 45.1 48 55 Under-five Mortality Rate 158 118.3 91 61 50 55 38 Infant Mortality Rate 106 78.5 64 48 41 44 32 Neonatal Mortality Rate 49.9 43 33 20 30 16 Attended delivery 18.7 28.8 % of underweight children 49.2 48.3 38.6 39.7 34 29 HIV prevalence among pregnant women aged 15-24 yrs NA NA NA NA NA Halt and reverse trend 48 70 65 71 75 85 TB case detection and success rates (%) NA 79 89 89 88 89 90 Full Immunization 60.1 80.1 88.8 48 90+ Malaria annual parasite incidence per 1,000 NA 0.54 0.40 0.28 NA Halt and reverse trend Source: Nepal Family Health and Demographic and Health Surveys 1991, 1996, 2001, and 2006. 2009 estimates from Maternal Mortality and Morbidity Study in 8 districts and Mid-Term Survey for NFHP II of family planning, maternal, newborn and child health. 5.1 Nepal has made considerable Figure 5.1: Health Indicators Regional and Inernational progress in reducing maternal and Comparison under-five mortality rates. The Bangladesh 2004 88 Maternal Mortality Ratio (MMR) Cambodia 2005 83 declined from 415 to 229 deaths per 100,000 from 2001 to 2009. Nepal 2006 61 Staus of Under-five Mortality Significant gains have also been India 2005/06 57 made in terms of other heath Indonesia 2002/03 46 indicators: the infant mortality rate Philippines 2003 40 2006 DHS, declined to 41 per 1000 in 2009 Tanzania 2004/05 macro 112 Interantiona from 64 in 2001; the neonatal Ethiopia 2005 IInc. 123 mortality rate from 43 in 2006 to 20 in 2009; total fertility rate from 4.1 India 2005/06 44 in 2001 to 2.9 in 2009; and under- Indonesia 2002/03 Children Fully 52 Immunized five mortality declined from 91 in Cambodia 2005 67 2001 to 50 in 2009. The reasons for Philippines 2003 70 Bangladesh 2004 73 the achievement include the joint Nepal 2006 83 efforts, through the health SWAp, by Tanzania 2004/05 71 government, donors and NGOs to Ethiopia 2005 20 expand and improve access of underserved regions and population, Bangladesh 2004 13 availability of aid, and higher per- Assisted capita disposable income43 -- and Nepal 2006 19 Delivery Cambodia 2005 44 higher household expenditure in health care compared to education44. 48 India 2005/06 Philippines 2003 60 The introduction of targeted Indonesia 2003 65 incentives in 200545 also helped to Tanzania 2004/05 43 raise the number of childbirths at Ethiopia 2005 6 health institutions. Childbirth attended by medical practitioners increased from 17 percent in 2006 to Tanzania 2004/05 3 Nutritional Status of Children: 27 percent in 2009. Another major Wasting Cambodia 2005 7 achievement is the immunization coverage, which was estimated at 90 Ethiopia 2005 11 percent in 2009. The immunization coverage in 2001 was 60 percent. Nepal 2006 12 Bangladesh 13 5.2 Nepal has reduced under-five 2004 child mortality and expanded Source: 2006 DHS, Macro International Inc. immunization coverage much faster than India, even though the latter has the advantages associated with a stronger economy, lower fertility rate and better female education. In 2006, Nepal’s under-five child mortality rate was 61 per 1000 and it improved to 50 in 2009. Immunization coverage has also improved, from 83 percent in 2006 to 89 percent in 2009. All major health outcome indicators – attended birth delivery, maternal mortality ratio, infant mortality rate, contraceptive use, malaria and annual parasite incidence – have 43 Per Capita GNDI in 2001 was US$ 298 and it was US$ 588 in 2009. 44 Household expenditure in health was 5.7 percent compared to 2.8 percent in education. (National Health Accounts – 2009) 45 Entitlements of Rs. 500, and Rs. 1500 in the Terai (plains), Hill and Mountain regions, respectively. 49 improved compared to 2006. But, Nepal’s performance in child nutrition – even when compared to low performing countries – remains poor. Although nutrition has improved compared to 2001, the range of stunting, wasting and underweight extends from moderate to severe. Further, over 24 of percent of women – aged 15 to 49 years – are underweight, lower than the situation in Bangladesh and Ethiopia. Likewise, anemia among women is 35 percent, lower than that in Cambodia. Table 5.2: Total Health Expenditure as Percentage 5.3 There is wide disparity in health of GDP in SAARC Countries outcome indicators. The infant mortality in rural areas is 73 percent, higher than Table: Total Health Expenditure as percentage of GDP in SAARC Countries that in urban areas. Infant mortality in Per Capita Health Expenditure in terms of USD the Mountain region is almost twice that Country 2001 2002 2003 2004 2005 of the Hills. Similarly, infant, child, Afghanistan 3 11 13 16 20 under-five and maternal mortality rates Bangladesh 11 11 12 13 12 among the poor are significantly higher Bhutan 48 41 43 48 52 than the non-poor (DHS 2006). China 48 54 62 71 81 India 21 22 26 30 36 5.4 The private sector invests more in Maldives 153 151 174 208 316 health than both donors and the Nepal 12 13 13 14 16 government. The state’s invests only Pakistan 10 11 12 13 15 Sri Lanka 32 34 38 45 51 23.7 percent in health while donors Source: WHO, 2008 contributed 20.8 percent and the private 46 sector 55.5 per cent . Of the private- sector expenditure in total health-sector spending, household out-of-pocket (OOP) expenditure is 90 percent, and the balance is spent by non-governmental organizations (NGOs). 5.5 Per-capita health expenditure has increased by more than 50 percent in nominal US dollar terms in five years (2001 to 2006). Per capita health expenditure was US$ 12.5 in 2001 and US$ 19 in 2006,I – low by international or regional standards. Table 5.3: Health Care Expenditure by Funding Source and Major Funcatios (% of Totoal Health Expenditure) 5.6 High remittance growth will further increase OOP health expenditure. Health Care Expenditure by Funding Source and Major Functions ( % of Total The doubling of per capita GDNI (from Health Expenditure) US$ 298 in 2001 to US$ 588 in 2009), and Governm Out-of- Res t of the ent Pocket worl d because households spend more on health than education47, and availability of more Source Al l opa thi c hos pi tal a nd out pa tient ca re 23.9 3.4 49.7 7.7 26.4 0.0 health care institutions, will contribute to Al l opa thi c medi ci nes 0.2 31.4 0.2 the increase. Ba s i c medi ca l a nd di a gnos tic s ervi ces 7.2 1.3 0.0 MCH a nd FP a nd reproductive hea l th s ervi ces 0.7 2.1 3.6 Immuni za tion 2.0 0.0 1.6 5.7 The national health care market is STDs 0.5 0.4 1.6 specialized – the private sector is in Vector-borne di s ea s e 0.2 0.0 0.3 curative care while public institutions Al l other publ i c hea l th s ervi ces 3.4 0.1 0.3 Government a dmi ni s tra tion of hea l th a nd concentrate on providing basic health care hea l th-rel a ted s oci a l s ecuri ty 1.7 0.0 5.7 services. The three main health Pri va te hea l th a dmi ni s tra tion a nd hea l th i ns ura nce 0.0 1.9 7.9 expenditure channels are: the demand for Source: Nepal National Health Accounts, July 2009 curative care and services by the private 46 Nepal National Health Accounts, July 2009. 47 NLSS – II 50 sector, basic health care services provided by government institutions, and health insurance/health-related social-security activities funded by donors. The OOP expenditure is largely spent on private goods – allopathic medicine and care (42% of total national health expenditure) – while government and donors have concentrated on providing public goods, basic health services through state institutions and increased access to these services through health-related social security and insurance schemes. While the private sector complements and, to some extent, competes with the state sector, it is the public sector, supported by donors, that provides the basic services (27 percent of total health expenditure)48. This includes all micronutrient supplements, safe motherhood and immunization services and large portions of the national program on family and Table 5.4: Health Related Expenditure by Funding reproductive services. Source and Major Functions ( % of Total ) Rest of Health care capital formation and Private the investment in health personnel are led by Government sector world the private sector – nearly two-thirds of Source 23.0 74.1 2.9 expenditure in both fields is borne by the Capital formation of health care provider private sector, and this trend continues. institutions 16.6 50.0 0.5 Education and 5.8 Non-state health service providers training of health personnel 3.2 23.5 2.4 have led the way in the expansion of Source: Nepal hospitals, beds and personnel. The number National Health of hospital beds available to the population Accounts, July 2009 increased by 25 percent during the review period (2004 to 2010). In 2004, there was one bed for 1,756 persons; this number rose to one bed for 1,321 persons in 2010. Non-state health institutions added 4.4 hospitals each year compared to two hospitals added by the state. The corresponding increase in the number of Table 5.5: Status of Health Infrastructure and hospital beds was Personnel -2010 727 in the private S Description Year 2004 Year 2009 Growth NO rate sector compared 1 Hospitals (public sector) 88 98 11 to 639 in the 2 Hospitals (non-government and private) 118 160 36 public sector. The 4 Hospital beds (public sector) 5,767 8,961 55 number of beds 5 Hospital beds (others) 8,315 11,951 44 per hospital is 6 Sub health post 3,142 3,132 0 marginally higher 7 Health Post 705 674 -4 at public hospitals 8 Primary health centre 178 215 21 9 Traditional medicine dispensary 182 190 4 (92 beds per 10 Health workers 7,491 31,212 317 hospital) 11 Nurses 10,099 29,579 193 compared to the 12 Medical Doctors 4,012 9,495 137 private hospitals 13 Female community health volunteers 47,000 50,000 6 (75 beds per Source: MoH, 2010. hospital). There has also been a noticeable increase in the number of health personnel. The number of health workers has tripled (4,744 per year); and that of nurses and doctors grew by more than double (3,896 nurses and 1,097 doctors per year). The number of female community health volunteers rose to 50,000 in 2010. 48 All basic medical care, all other public health services, health related social security and health insurance. 51 5.9 Retention and availability of skilled health personnel are major issues at state health institutions. Almost 25 percent of the sanctioned posts at public institutions remain vacant and Table 5.6:Status of Human Resources -State Institutions similarly, on an average, Position Sanctioned Filled Vacant % of filled Share % 12 percent of nursing and positions paramedic posts remaine Medical doctor 1,062 816 246 76.84 4.34 unfilled. Absenteeism is Nursing staffs another issue at public 5,935 5,307 628 89.42 24.25 including ANMs health institutions. Paramedics 10,642 9212 1,430 86.56 43.48 According to a recent Other 6,838 6,394 444 93.51 27.94 survey, only 64-84 Total 24,477 21,729 2,748 88.77 100 percent of posted doctors, 68-81 percent of nurses Source: Annual report, DoHS, 2007/08 and 81-92 percent of paramedics were available at their posts.49 Absenteeism is higher in remote districts. The reasons health personnel prefer to work at private urban facilities are attractive salaries, better facilities, and better housing and education access for dependents50. 5.10 There are wide variations in cost and per-capita expenditure across institutions and income quintiles. Rich households spend 24 times more than the lowest-income quintile on health care. Costs of Table 5.7: Household Expenditure by Facility and Per diagnostic services are lower in Capita government institutions while Government Private medicine costs are higher in Household expenditure Institutions Institutions comparison to private Diagnostic and other services NRs. 99 136 institutions. It is noteworthy Medicine cost NRs. 538 465 that the rich spend 1.5 times Richest Poorest Per Capita Household Expenditure Quintile Quintile more in government Mean expenditure per person NRs 255.54 10.75 institutions than at private Share of household spent on health (%) 7.26 2.63 institutions, because of the Source: Nepal national Health Accounts, lower cost. Also, the utilization 2009 7.26 2.63 rates of various health services consistently increase with income. One example of the use of skilled birth attendance is shown in Table 5.8 below: 49 RTI International – December 2009. 50 RTI International – December 2009. 52 Table 5.8: Percentage of women who had skilled birth attendance e Features 1996 2001 2006 2009 National Average 9.0 10.8 18.7 28.8 Income quintile First (poorest) 2.6 2.5 4.3 8.5 Second 4.9 4.7 9.3 26.8 Third 6.1 6.7 11.9 22.6 Fourth 7 12.6 21.7 38.9 Fifth (wealthiest) 35 43.9 55 57.7 B. Public Health Expenditure 5.11 Health sector spending is growing. As a share of GDP, spending has nearly doubled in five years from 0.9 percent in 2006 to 1.7 percent in 2010 and the focus of public spending is on the provision of inclusive, basic health services. Donor funding plays a large role in financing the health sector – under the SWAp-I it provided 48 percent of 2010 health sector budget. But donor funding in the health sector is still an underestimate because a substantial amount of donor funds are reportedly spent outside the government system and through NGOs and CBOs51. The out-of-budget funds remain a challenge in improving sector-wide governance, particularly in terms of accountability and ownership. To a certain extent, the out-of-budget funds have distorted the incentive structure within the ministry with salary differences, better office environments and training opportunities. 5.12 Government-funded activities have increased by 1.7 times from 2007 to 2010, with more aid availability and revenue growth. Spending in fully-government-funded projects has increased from NPR3.8 billion in 2007 to NPR6.8 billion and, much of this increase has gone into state hospitals, which have high recurrent cost. 5.13 Government funds a large proportion of the health budget – up to 52 percent now, from 49 percent in 2007. The government’s commitment to new initiatives also helped raise this share. The structure of aid has also changed: cash grants increased by 11 percent between 2007 and 2010, while overall aid, as a share in health budget, decreased by three percent. The gap was filled by increases in government funding. Non-cash grants declined by 17 percent during the review period. The reduction in non-cash grants, as a share of total aid, may have to do with poor governance in public health institutions and management and the resulting donor preference to deliver services through non-government channels. 51 National Health Accounts, 2009. 53 Figure 5.2: Sources of funding 2007and 2010 Health: Sources of Funding 2007 Health: Sources of Funding 2010 Non cash Non cash o/w cash grant o/w cash grant Cash Cash Health: Sources of Health: Sources of Loan Funding 2007 Funding 2010 Loan Grant Grant Government Government 0 20 40 60 80 0 20 40 60 80 100 5.14 Health public expenditure is geared towards providing essential health care services (EHCS) and accounts for half of the total health spending. This structural shift in spending emphasis on basic services is the result of the PRSP strategy that was anchored by the first health SWAp and was continued by the TYIP. Within the total EHCS sub-sector spending, the spending emphasis has been on providing services through health posts and sub-health posts (38 percent, mostly current expenditure); and Table 5.9: Health Sector expenditure - Percent of Health Sector Budget ( 2003- through programs such 2010) as immunization (21 Three Year percent), family Tenth Interim Budget planning (10 percent), Plan Plan 2010 and tuberculosis control Essential Health Care Services 62.2 47.7 48.9 Beyond Essential Health Care Services ( mostly (7 percent). hospitals) 13.1 26.4 31.2 Private Sector/NGO Sector Development 6.7 1.8 1.0 5.15 A moderate Sector Program Management 2.5 2.9 2.3 structural shift has Health Financing Resource Management 3.0 3.9 4.7 Logistic Management 9.2 15.0 9.5 taken place in tertiary Human Resources Development 1.9 1.3 1.6 care. There has been an Integrated MIS 1.4 1.0 0.8 addition of two new Source: MoF and Staff estimates. state hospitals/facilities each year from 2004, causing spending in ―beyond essential health care services‖ to increase by 10 times in nominal terms – the increase mostly went to hospitals. Spending on hospitals increased from an average of 13 percent of total health spending in Tenth Plan to 31 percent in 2010 and is expected to rise further as more public sector health facilities are pressed into service in urban areas. The SWAp-I that anchors the health sector strategy has not only helped to protect basic health care services but has also increased funding for EHCS by three times from 2004 to 2010. This increase took place from a high base: EHCS spending was already 62 percent of the health sector spending in the Tenth Plan. The rapid expansion of spending in basic services and hospitals has led to reduction in expenditure on health service delivery through non-government institutions. Spending through non-public 54 institutions, as a percentage of public health expenditure, has declined by less than one percentage point from seven percent during the Tenth Plan. However, there has been a corresponding increase of nine percent growth in expenditures by NGOs in health services52. These expenditures are outside the budget. 5.16 Recurrent expenditures comprise about 80 percent of the health sector budget. The spending on salaries, as a share of total health expenditure, has decreased between two plans – Tenth Plan (27 percent) and TYIP (16 percent). However, in nominal terms, the salary cost increased by 38 percent during the review period. Supply of medicines and equipment and transfer of funds for targeted programs/health related social security have increased by 2.5 times respectively. As a result of the renewed emphasis on urban expenditure – hospitals – capital formation has tripled, from five percent of health spending in 2005 to 16 percent in 2010. Table 5.10: Health spending - Economic Heads s 2009 2010 2005 2006 2007 2008 Budget Budget Current 91 83 84 75 80 79 o/w salary 40 35 27 23 19 16 o/w subsidy transfer 18 14 14 14 18 19 o/w supplies (medicine and equip.) 26 27 37 32 39 38 Capital 9 17 16 25 20 21 o/w capital formation 5 11 9 14 15 16 5.17 Salaries in the health sector are expected to increase. Currently only 30 percent of the total national health personnel are hired within the public sector 53. The demand for health services is also expected to rise due to the planned expansion of basic health services under second health sector plan, abolishment of user fees for basic care, increase in population and migration from rural to urban areas. The human resource information system at the Ministry of Health and Population has records of 35,000 health personnel, and this is also an underestimate54. 5.18 Provision of the EHCS is focused more is the Central Development Region and the central hill districts. Disparities in EHCS spending across development regions and belts are visible55. More than 90 percent of medicines and three-quarters of capital funds are spent in the Central Development Region where about 40 percent of Nepalis reside. The focus of targeted programs under the EHCS is in Terai, where 52 National Health Accounts, 2009. 53 Total national health personnel 120,286 (MoH, 2010); approved positions in MoH, 35,000. 54 Human Resource Information System is believed to contain only 75 percent of total personnel, MoH April 2010, NHSP – Implementation Plan. 55 Population distribution by region as of 2001 (%): Eastern- 23, Central-34, Western – 20, Mid-western- 13 and Far western- 10 55 over 60 percent of total targeted programs under the EHCS are executed. These disparities across regions and ecological belts have not been addressed under the SWAp. In addition, health expenditures cannot be compared without first disaggregating them into different flows, such as salaries, spending on medicines, and equipment. Quite often, salaries crowd out spending on medicines and equipment – what really matters for beneficiaries. Table 5.11: EHCS-Regional Expenditure by Major Economic Heads-2007 and 2009 EHCS -Regional Expenditure by Major Economic Heads - 2007 and 2009 2007 2009 Trasfers ( targeted Trasfers ( Development Medicin program Medici targeted Regions Salary e s) Capital Salary ne programs) Capital Eastern 23 3 16 8.3 22 4 16 2.4 Central 29 92 43 76.8 30 89 48 74.4 Western 23 3 4 6.1 22 3 16 15.1 Mid-Western 15 2 21 3.9 15 3 10 4.8 Far Western 11 1 16 4.9 11 1 10 3.2 Ecological Belts Mountain 12 2 1 0.6 13 2 2 2.5 Hill 50 94 32 79.7 52 92 38 80.0 Terai 35 4 67 19.7 33 6 60 17.5 Source: MoF and Staff Estimates 5.19 Inequality is less in ―Beyond-EHC‖ expenditures than in EHCS spending. Beyond-EHC spending in hospitals and more on curative care – 31 percent of total health spending – is comparatively low in Mountain districts (a quarter of spending in Hill districts) and in the far-Western Region (a third of that of the central region). Unlike the EHCS programs, Beyond-EHC spending in the Terai districts is favorable compared to the Hill districts but high compared to mountains. These differences in levels of spending across regions are influenced by health infrastructures that are centered in and around district headquarters and expansion of health posts in the Terai during the review period. The deficit of health infrastructures in far-Western Region and Mountain districts should be addressed through more capital investment. In 2010 it was less than 10 percent. Table 5.12: Beyond EHCS-Regional Expenditure by Major Economic Heads-2007 and 20 Beyond EHCS -Regional Expenditure by Major Economic Heads - 2007 and 2009 Trasfers ( Trasfers ( Developme targeted targeted nt Regions Medicine programs) Capital Medicine programs) Capital Eastern 23 22 23 24 22 25 Central 29 34 34 30 36 22 Western 21 20 22 21 19 21 Mid-Western 17 15 12 16 14 21 Far Western 10 9 9 9 9 10 Ecological Belts Mountain 14 9 7 15 9 6 Hill 50 40 57 49 47 60 56 Terai 36 51 36 37 44 34 Source: MoF and Staff Estimates 5.20 There are marked differences in health spending across regions and ecological belts. The net of national health spending (programs that are implemented in all districts – 34 percent Table 5.13: Public Expenditure in Health Sector 2001-2009 by Regions and Geographical Belts as of total health spending) is concentrated percent of health spending. in central Hill districts. Health spending in the central region is seven times Ninth Interim Development Plan Tenth Plan ( 2008 more than that spent in the Far-western Regions 2001 Plan to 2009) Region. The Hill districts receive more Eastern 16 17 16 than eight times the level of expenditure Central 50 48 50 than the Mountain districts. In per- Western 15 15 15 capita terms, the central region Mid Western 11 11 12 residents receive twice what those in Far Western 8 8 7 the far-West get, and the people in the Hills get 30-40 percent more than what Ecological Belts Mountain residents receive. Mountain 10 9 8 Hill 64 61 65 5.21 The expenditure on basic health Terai 27 29 27 care services across regions and ecological belts follows the national pattern. The EHCS spending is higher in central region and mountain districts compared to other regions and ecological belts. However, the health spending patterns in hospitals (classified as ―beyond basic Table 5.14: Public Expenditure in Essential Health Care and Beyond Essential Health Care 2004-2009 by Regions and Geographical Belts Public Expenditure in Beyond Essential Public Expenditure in Essential Health Care 2004-2009 Health Care 2004-2009 by Regions and by Regions and Geographical Belts Geographical Belts Interim Plan Interim ( 2008 to Plan( Tenth Plan 2009) Tenth 2008 to Development Regions Plan 2009) Eastern 17 16 Development Regions Central 47 51 Eastern 21 24 Western 16 15 Central 34 28 Mid Western 12 11 Western 20 21 Far Western 8 7 Mid Western 16 17 Ecological Belts Far Western 11 10 Mountain 10 9 Ecological Belts Hill 60 63 Mountain 13 10 Tarai 29 28 Hill 47 53 Terai 40 37 health‖ expenditures, except for spending on children and women) are concentrated in the district headquarters, and the level of investment is in tune with the number of district headquarters in development regions and ecological belts. 57 5.22 The variation in public expenditure in the health sector alone cannot explain the disparities in health outcomes across development regions and ecological belts. Many inter-dependent inputs – literacy, access to safe drinking water and sanitation, Table 5.15: Health Outcomes by Development Regions and Ecological Belts – 2006 6 Development Regions Ecological belts Mid- Far- Eastern Central Western Western western Mountain Hill Terai Moderately and Severely Thin ( among women age group) percent 8 8.8 5.3 6.3 11.6 4.1 4.2 11.9 Weight-for-age - Percentage of children under five years as malnourished 7.9 11.4 10.9 11.3 12.2 11.5 8.1 12.6 Percentage of women age 15-49 with any anemia 31.1 35.8 31.2 36.7 51.4 21.5 20.7 51.4 Percentage of Children with any diarrhea 11.7 12.2 12.6 10.2 11.9 15.5 12.6 10.7 All basic vaccines 86.2 78.3 88.9 80.8 80.5 71.3 81.6 86.4 Assisted delivery by SBA 17.2 24.7 20.1 14.2 9.6 7.1 22.7 17.5 Place of Delivery home 82.6 74.2 81.8 85.8 89 91.6 78.1 81.6 Place of Delivery government sector 12.8 18.4 11.9 9.5 5.9 4.6 17 11.3 Percentage receiving antenatal care from SBA 45.2 46.5 50.9 43.6 26 32.3 46.7 43.1 Childhood Mortality per 1000 Infant 45 52 56 97 74 99 47 65 Child 15 17 18 28 28 32 16 21 Under-five mortality 60 68 73 122 100 128 62 85 Source: 2006 DHS Mid- Far- Eastern Central Western Western western Mountain Hill Terai Health Public Expenditure : Plan Periods Ninth Plan 16 50 15 11 8 10 64 27 Tenth Plan 17 48 15 11 8 9 61 29 Interim Plan 16 50 15 12 7 8 65 27 Source: MOF and Staff Estimates roads and telecommunications (connectivity), availability of skilled staffs and medical supplies, socio-cultural behaviors and beliefs – collectively influence health outcomes. Some patterns in health expenditure and outcomes: (i) Spending disparities in public health across regions and ecological belts continue and there has been no attempt to redress them; (ii) Poor health outcomes are pronounced in the far-Western Region and Mountain districts, where there has been less than 10 percent of the health spending; (iii) Except for immunization coverage across regions and ecological belts, there is need for more focus on all 10 health outcomes in the far-Western and Mountain districts; and (iv) As health sector outputs are not ―costed‖, it is very difficult to link inputs (investment) to outputs. Nevertheless, poor health outcomes in lagging regions do signal the need for more investments. 5.23 Health sector spending is now 90 percent of the allocation. The spending performance has improved from a low base of 78 percent in 2004 to 90 percent in 2009. The dedicated planning and budgeting unit supported by a health sector reform unit, donor harmonization for reducing transaction costs and tight alignment of 58 resources on priority areas that are recurrent in nature – immunization, cash incentives for safe birth delivery and free health care services up to district hospitals – are reasons for this spending level. Table 5.16: Health sector - Spending Performance (2004- 2009) )) 2004 2005 2006 2007 2008 2009 Budget - NRs. Billion 5 7 8 10 13 16 Expenditure - NRs Billion 4 5 6 8 10 14 Absorption Capacity Health Sector 78 71 76 81 82 89 Essential Health Care Services 82 98 117 126 147 217 Beyond Essential Health Care Services 62 79 78 97 86 86 Private Sector/NGO Sector Development 93 100 99 49 100 159 Sector Program Management 94 52 56 104 79 37 Health Financing Resource Management 92 82 83 89 91 83 Logistic Management 63 47 45 76 75 84 Human Resources Development 73 46 62 65 72 93 Integrated MIS 98 71 83 88 90 93 Source: MoF and Staff Estimate 5.24 More importantly, the government reallocated health funds from non- performing activities to those that worked: (i) Cash incentives for safe birth delivery in 2006, 2007, and 2008; and, (ii) targeted, free basic health care services in 2009. This data is reflected in higher-than-budgeted spending from 2006 under EHCS subsector performance. 5.25 The challenges facing the health sector are: enhancing access, social inclusion and equality in health service utilization. The inequality in access, health outcomes and service utilization remains high. The poor have the largest unmet demand for family planning, make the lowest use of maternal cares services, have the lowest vaccination coverage, and are least likely to seek care when ill. The poor, especially in the Mountain districts, have the lowest physical access to health care. In regional terms, access to health services in the mid- and far-Western regions is limited compared to other regions,especially in terms of nutrition and HIV/AIDS indicators. Going Forward 5.26 Implementation of activities under the second National Health Sector Plan (NHSP II) can improve health outcomes. The current health sector achievements are built upon sufficient resources for priority activities funded under SWAp-I. Improved road connectivity, essential staffs at sub-health posts around the year – minimum, presence of one ANM and/or CMA – improvements in drinking water and sanitation coverage assisted in attaining the outcomes. 5.27 The NHSP II deepens focus on areas that have had little or no progress. NHSP I (2004-10) focused on basic health service delivery and strengthening of health systems. NHSP II had seven components: Essential health care services, beyond EHC services, service delivery through NGOs and private sector, sector program management, health resource management, logistics and human resources development and integrated Management Information System (MIS). The strategy 59 was anchored by the first SWAp in 2004. The NHSP II, while continuing emphasis on the seven pillars, deepens interventions to improve outcomes in the following areas: population; nutrition and reproductive health; child health care; health system performance; and HIV/AIDS and other communicable diseases. The second SWAp has an outlay of US$ 1.5 billion for five years (2011-2015). 5.28 The child health care strategy focuses on reducing the disparity in health outcomes across regions and income quintiles. The NHSP II has set aside 25 percent of total funding for child health care and for reducing the current disparities across regions. Rural infant mortality is 73 percent higher than in urban areas. The infant mortality in the Mountain region is twice that of the Hill region. In general, the health outcomes of poor households are lower and needs urgent attention (DHS 2006). 5.29 Malnutrition in Nepal is among the highest in the world. During 1995-2002, Nepal had the highest number of underweight children in the world. Although there has been some improvement, the progress is far from the ideal and more recent studies suggest that a third of all children weigh less than what they should in the corresponding age group. The NSHSS II has allocated 25 percent of the health sector funding to address this concern. 5.30 Nepal has highest HIV prevalence rate in South Asia. Nepal’s HIV prevalence rate in the 15-49 years age group is 0.49 percent – the highest in South Asia. Past public interventions for managing and preventing HIV/AIDS were far from satisfactory. The NSHP II has increased focus on HIV/AIDS and has allocated 15 percent of the health sector funds for activities to reduce spread and care for people with HIV/AIDS. 5.31 Nepal has started from a low baseline, and therefore improving health outcomes will require concerted efforts from service providers to expand coverage, increase investment, create an enabling environment for the involvement of non- government actors and improve cooperation and coordination amongst health service providers and other line ministries, whose inputs (roads, drinking water, sanitation, education, etc.) are also preconditions for improving the outcomes. C. Planning, Implementation and Monitoring Issues 5.32 Planning and budget approval remains slow. The approval of the annual work program budget (AWPB) has generally stretched to the second trimester of the fiscal year, weakening the efficiency gains that could have resulted from the investment. Late approval of the work program has resulted in bunching up of activities in second and third trimesters and this may have compromised the absorption capacity, especially in procurement. Early consultations, sometime in October/November, between donors and the MoH for discussing the next fiscal year’s AWPB can have a meaningful impact by ensuring timely approval of the budget, and for securing ample time for tightly aligning resources with priority activities. 5.33 Procurement plans must also be part of the AWPB. The approval of AWPB, without a procurement plan for major activities, is meaningless. Approval of the budget in itself is not a sufficient condition for timely implementation. The MoH should take advantage of multi-year contracting to ensure that medicines, supplies and 60 equipment are procured and distributed on time. For this to happen, it will have to change the current practice of preparing procurement tenders, after budget approval, and ensure that the budget is based on the AWPB and a procurement plan. 5.34 Poor logistics management has compromised the quality of health services. Persistent problems in procuring medicines, supplies and equipment and their timely distribution to health centers have resulted in low quality of supplies and delays in providing services. Mobile health teams and special camps are pushed to service throughout the year for delivering services in areas with frequent outbreaks of preventable diseases. Early procurement of medicine and its distribution can assist in ameliorating the supply deficits but there are additional issues of storage that need to be addressed for effective service delivery. 5.35 The limited supply of trained personnel has stretched health service units. Shortages of doctors and unfilled doctor vacancies remain major issues in health service delivery. The government has taken measures to reduce this shortfall by appointing new doctors, contracting more doctors and pressing medical students (on government scholarship) to work for two years in rural areas. These measures have helped to ease the deficit. However, doctors and nursing staff prefer not to work in remote regions owing to low salaries and lack of training opportunities for advancing their skills, which need to be addressed in order to ensure presence of trained human resources at the service stations. An effective partnership with private medical colleges in addition to efficient regulatory and accreditation mechanisms could provide a better solution to this chronic problem. 5.36 The Health Management Information System (HMIS) needs to be strengthened. The HMIS is essentially an early warning system of disease outbreaks that has been extended to monitor health outputs and information on services being provided to people from different social groups. The HMIS may not be an appropriate instrument to monitor health outputs, inequality and health spending data. Pressing the HMIS to monitor budget progress outputs, in addition to disease monitoring, may harm the system’s ability to deliver on both functions. A separate health-spending monitoring system that can be linked with outputs is needed to improve efficiency of health expenditures. The AWPB process can serve as an instrument to continuously monitor the government’s efforts, in particular, to reduce inequality. Expansion plans, introduction of new programs and all interventions will need to be assessed based on their contribution to, among others, reduction in inequality. 5.37 The decentralization of health services to local management committees has been ineffective in the absence of elected local bodies. Unlike the education sector, where SMCs are elected, the handing over of health facilities to local management committees is the responsibility of local bodies. This model of handing over facilities without a law and transparent rules has made little difference in either service delivery or facility management. D. In Conclusion 5.38 Nepal has made remarkable progress in health service delivery, and the interventions planned for reaching underserved populations are expected to improve 61 the sector’s performance. The following recommendations would help to widen coverage of health services for meeting the MDG goals: Realign health resources. Essential Health Care Services (EHCS) is the major policy thrust of the government, but its share in total health expenditure has been declining. This needs correction at the ex-ante planning and budgeting stage. There is also a need to continue to focus and recalibrate the EHCS to underserved regions and population groups. The government can also consider contracting out to private and NGO institutions specific non-EHCS services that require specialized, focused care and diagnostic support, rather than increasing block grants to tertiary hospitals and other institutions. Outsourcing of services to medical schools. All development regions have more than one medical school that can be pressed for services through acceptable outsourcing modalities under corporate social responsibility agreements with the GoN when these institutions were set up. The outsourcing should not only expand quality coverage of basic health care services, but also provide an opportunity for new doctors to experience rural health care. This might encourage them to work in those areas even after expiry of the mandatory two-year service for those on government scholarships. Preparation of the AWPB for the coming fiscal year should start in October and be accompanied by procurement plans. A complete budget package – activities backed by a procurement plan, anchored by funds against verifiable output targets – could strengthen the planning, budgeting and implementation processes. Aligning the budget with the results framework of the NHSP II could protect spending from interest and advocacy groups. Public expenditure on health needs to address inequality in service provision and outcomes. Budget allocation has been on the basis of the needs of existing facilities rather than the communities they serve, which tends to maintain or even exacerbate inequalities. A move away from this practice towards allocating budget on the basis of beneficiary size (so that allocation on per- capita basis is considered), accessibility (how accessible services are to users), and service delivery cost would improve equity in access to, and use of, health services. The Gender and Social Inclusion Strategy provides a basis for planning and budgeting basis for addressing inequity. The monitoring of health-sector activities needs to be strengthened. There is a need to assess and evaluate the cost-effectiveness and feasibility of the HMIS for both early-warning and monitoring of budget implementation. The reporting schedule should be aligned with the budget’s fund release for facilitating mid-term fund release and/or correcting variations. There is a need for in-depth analysis of the large OOP expenditures – to understand what the funds are used for – for planning financial protection measures. Establishment of an inter-ministry coordination committee could facilitate the delivery of health services. Close coordination of inputs provided by other line ministries, especially, the expansion of road and telecommunications, improvement of trails and bridges, expansion of female literacy and access to 62 information could facilitate the utilization of existing health services more effectively. Preparation and implementation of a private-public partnership plan could complement the public efforts to improve health service delivery. The increase in remittances has increased the demand for curative services in urban areas and emerging towns. A private-public health policy could be the first step towards making health services financially sustainable. Such a partnership could also free public resources from urban investments for use in expanding basic health services to underserved regions and social groups. An umbrella regulatory mechanism, such as a private-sector regulation Act – with clear roles and responsibilities for state and non-state actors in the health sector – would facilitate implementation of this kind of partnership arrangement.□ 63 6. SOCIAL PROTECTION Key messages  Social protection spending in recent years has increased substantially from 0.5 percent of GDP in 2004/05 to 2.0 percent in 2008/09, and is expected to increase to 2.9 percent in 2009/10.  Increased government expenditures have helped to diversify the social protection system in Nepal to include safety nets and social security or pensions programs, and to a lesser extent labor market interventions, social care services, social funds and fertilizer-subsidy programs.  Cash transfer and school feeding programs constitute the majority of social protection spending. Spending on the three largest safety-net programs (old age, widow, and disability allowances), which are universal, is expected to reach 0.8 percent of GDP in 2009/10, thereby raising safety-net spending to the same range as in most developing countries ( 1-2 percent of GDP).  Coverage of these programs remains limited, and suffers from some overlap of beneficiaries. Pro-poor targeting and/or consolidation of these programs are likely to generate substantial savings, which could be used to better align programs to the needs of beneficiaries within existing budgetary allocations. Summary This chapter reviews the role of public expenditure in providing safety nets to the vulnerable sections of the society to help mitigate potential risks and vulnerabilities associated with various shocks, trends in public expenditure, efficacy and targeting of safety net programs, and makes recommendations for improving public social protection interventions. Section A discusses global shocks that increase vulnerability of the poor; Section B reviews the spending patterns and Section C raises issues related to program coverage, targeting and expenditure adequacy. Section D summarizes the chapter with policy recommendations. A. Poverty, Risk and Vulnerability 6.1. Poverty in Nepal remains high despite a decade of strong growth and remarkable economic and social progress. The percentage of Nepalis living in poverty fell from 42 percent in 1995-96 to 31 percent in 2003-04. This was a remarkable achievement considering that Nepal emerged as recently as 2006 from a 10-year conflict. No recent estimates of poverty rates are currently available. However, given the widespread adverse impact of the global increase in food prices on poverty in other countries in the region such as Bangladesh56, it is plausible that average real 56 Simulations of the impact of the rising prices on household welfare in Bangladesh, using 2005 household-level data, show an estimated increase in the poverty rate by about 3 percentage points, with a 5 percent nominal wage increase, using the 2005 poverty rates as the baseline (World Bank. 2008. ―Poverty Assessment for Bangladesh: Creating Opportunities and Bridging the East-West Divide.‖ World Bank, Washington D.C.). 64 income, particularly of the poor, has declined. There are also signs of stagnating economic growth. GDP grew by 5.3 percent in FY08 from 3.3 percent in FY2007, but decelerated in FY09 to 4.7 percent and is expected to be even lower in FY10. 6.2. The global food price crisis is likely to adversely affect the poor. At the height of the food crisis the Food and Agriculture Organization and the World Food Program estimated that 42 out of 75 districts in Nepal had food deficits. Since the food price crisis, the price of staple commodities has increased by 30-70 percent. Given that the poor spend most of their income on food (the poorest population spends more than 75 percent of their income on food57), they are likely to shoulder a disproportionate burden of rising prices by either consuming less or by replacing consumption with cheaper alternatives. Households also cope by postponing children’s enrolment to school, selling assets, and reducing expenditures on health and medical care. Some of these coping mechanisms can have long-term adverse consequences on human capital accumulation. While it is not clear what future food prices will be, food security remains a major concern in Nepal. 6.3. Increased incidence of climatic shocks and a changing climate add to the vulnerability faced by Nepal’s poor, mainly in rural areas. Nepal is witnessing an increase in temperatures, more intense rainfall and increased unpredictability in weather patterns, including drier winters and delays in the Monsoon rains. Poor crop yields, water shortages and extreme temperatures are driving agricultural output down, leaving rural populations unable to properly feed themselves and/or leading to indebtedness. More than 3.7 million people in Nepal are estimated to require food assistance due to a combination of natural disasters. Some of the heaviest burdens have fallen on women who are on the frontline of climate change: They have to travel further to fetch water and take on the responsibility for feeding their families as men in many poor households migrate seasonally to seek work. 6.4. In addition to women, certain castes and ethnicities are poorer and more vulnerable than others. Past analysis shows that the incidence of poverty in Nepal varies by gender, caste, ethnicity and location. As in much of South Asia, Nepali women and girls fare comparatively worse than men and boys in most human development outcomes (World Bank, 2006). Women’s lack of access to income, education and decision-making perpetuates inferior outcomes on child nutrition and school participation and thereby exacerbates the inter-generational transmission of poverty. Although the poverty headcount declined for 11 castes and ethnic groups between 1995-96 and 2003-04, the rate of decline for upper-caste households was considerably higher than the country as a whole. Large differences in the rate of poverty reduction across caste and ethnic groups point to structural differences in assets, skills and opportunities. Households, particularly in the Mountain and Hill districts of the far- and mid-Western regions are also vulnerable to several types of natural disasters. 6.5. Mechanisms to protect the poor and the vulnerable have received greater prominence in public policy in recent years as part of the government’s efforts to achieve inclusive growth. Evidence from OECD countries suggests that substantial redistributive expenditures can lower inequality and facilitate growth. Social 57 Project paper: ―Additional Financing for the Social Safety Nets Project‖ 65 protection programs are important in reducing deep deprivations associated with extreme poverty and in mitigating risks of the poor from reverting to poverty in the event of a shock. The TYIP and the Poverty Reduction Support Program (PBPRSP) of the government highlight the need to achieve growth with equity. Accordingly, the FY08/09 budget included ―expansion of social security and inclusion‖ as one of its three priority-policy themes. 6.6. Allocations for social protection spending have increased substantially. Increased allocations have funded both the expansion of existing programs and added new ones. More specifically the FY08/09 budget introduced programs such as the provision of immediate relief to conflict-affected households; increased public works (e.g., The Productive Employment and Fundamental Rights for All program – which, on demand, guarantees 100 days of work per annum); substantially expanded spending on safety nets to the elderly, widows and the disabled, and added new eligible groups (e.g., Dalits, endangered ethnicities and residents of the Karnali region); and initiated active labor-market programs (e.g., technical vocational education and training). 6.7. It is important to ensure that social protection allocations are adequate, progressive, and efficient. The social protection system consists of a scattered, poorly funded set of instruments that do not necessarily focus on the poor. The government is cognizant of the need for scaling up, and improving the targeting and administration of its social protection programs. A review of social protection expenditures over the period 2005 to 2010 is being conducted to identify possible short-run fiscal consolidation as well as efficiency-enhancing reforms that would eventually generate fiscal savings. The focus of the analysis is on government expenditures that finance safety nets, labor-market activation programs, pensions, social care services, social funds, as well as related health and education programs that are designed with social protection goals and objectives (such as programs for school feeding and nutritional supplements for children and pregnant women). These programs are generally categorized as social protection spending and a detailed description on them are given in Box A1 in the Annex. A list of the specific programs considered for analysis in this chapter is also provided in the Annex in Table A2. The list is by no means exhaustive but captures most of the expenditures that can be categorized as social-protection spending. Data on some smaller programs was not readily available58. It should be noted that the analysis presented in the following sections are based on estimates of data received from various ministries. A major limitation of this chapter is its inability to assess the benefit incidence of the key social-protection programs due to the lack of household-level data for the period under review. B: Trends in Social Protection Expenditures and Composition 2005-2010 6.8. Social protection expenditures (as a percentage of GDP) in Nepal were until recently lower than most countries in South Asia. In 2004/05 Nepal spent an estimated 0.5 percent of GDP on social protection (including pensions), whereas the average for South Asia was about 2.4 percent of GDP. 58 For example, the analysis does not include smaller programs such as the Public Distribution System (PDS) that provides goods at concessional prices. 66 Figure 6.1: Spending on Social Protection in South Asia (% of GDP, FY04/05) 2.5 2.22.2 2 1.8 1.5 1.11 1.1 1 0.5 0.6 Safety net 0.5 0.31 0.3 Public sector pensions 0 Source: World Bank staff estimates based on country safety net reports as of 2004/0559 6.9. Social protection spending jumped to 2 percent of GDP in FY08/09 and is expected to go up to 2.9 percent in FY09/10. The rest of South Asia has also increased total expenditures towards the social sectors, a likely fallout of the recent food price crisis. Spending on the three largest safety-net programs (old age, widow and disability allowances) is estimated to increase to 0.8 percent of GDP in FY09/10 which is more or less in line with the average spending on safety nets in most developing countries (1-2 percent of GDP.)60 6.10. The social-protection system has diversified over the years but expenditures remain skewed towards public sector pensions. In FY04/05 social-protection spending was mostly for safety nets, health and education programs. Since FY07/08, increased government expenditures have helped to diversify the social protection system to also include social security or pensions and to a lesser extent labor-market interventions, social care services, social funds and fertilizer subsidy programs (see Table A2, Appendix). The rise in spending has been particularly steep for expenditures on pensions: in FY04/05 spending on pensions was at 0.01 percent of total government expenditures, but went up to 4.3 percent in FY08/09. This bias has important implications, since public-sector pensions tend to benefit the non-poor and those with access to the formal sector. 6.11. The composition of social-protection spending is expected to be more pro- poor in FY09/10. Safety-nets spending, which are generally targeted to the poor, is to overtake social-security expenditures, and at an estimated 3.86 percent of total expenditures it constitutes the second-largest social-protection spending (based on the FY09/10 allocation). Expenditures on social protection-related health and education 59 Safety nets include targeted programs, workfare and food subsidies, which can give rise to some variation in what is covered by these estimates. Public-sector pensions refer to spending on civil service and military pensions. Estimates are not fully comparable across countries due to differences in the way that military and private pensions are included. 60 Grosh et al. 2008. ―For Protection and Promotion: The Design and Implementation of Effective Safety Nets.‖ The World Bank: Washington, D.C. 67 programs are to make up the largest component of total spending, at 4.02 percent, as shown in Figure 2. Figure 6.2: Social Protection Expenditure 2004/05–2009/10 (Totals and Percent of GDP) ) 14.00 12.00 10.00 8.00 6.00 3.23 3.86 4.00 0.85 1.34 2.9 2.00 2.0 1.26 0.5 1.07 0.5 1.02 0.5 1.08 0.5 1.0 1.1 0.00 1 2 3 4 5 6 7 8 Labor Market Interventions Pension Social Safety Nets Education and Health Social Funds Social Care Service Subsidies % of GDP Source: Staff estimates based on interviews of relevant ministry officials 6.12. Cash transfers and school feeding programs constitute the core of social protection spending. Nepal is the only country in South East Asia that has adopted safety-net programs as part of its strategy for addressing income insecurity among its old age, widowed and disabled populations. Old age, disability and poor widow allowances are the three largest cash transfer programs and have been scaled up over the years (see Figure 3). Whilst the old age allowance program targeting persons over the age of 75 is universal, the widow allowance is targeted to women above the age of 60 who meet the criteria of not having economic resources, nor care from family members, nor receiving the pensions of their late husbands. The disability allowance is targeted at people aged over 16 years, and with permanent disabilities. Together, the cash-transfer programs reach about 1 million people. The Food-for-Education program is geographically targeted to 11 districts and reaches about 450,000 school children in 4,000 schools. The program provides meals on-site and fortified rations to take home. 68 Figure 6.3: Beneficiaries of Old-Age, Widow and Disability Allowances: 1996-20088 300,000 257,437 250,000 252,025 200,000 150,000 100,000 50,000 6,875 0 1996- 1997- 1998- 1999- 2000- 2001- 2002- 2003- 2004- 2005- 2006- 2007- 97 98 99 00 01 02 03 04 05 06 07 08 Number of Elderly Number of Widows Number of Disabled Source: NLSS 2003/04 in Isik-Dikmelik, A., 2009 6.13. Emerging priorities for the government include using social-protection spending to promote greater social inclusion and harmony. The recent increases social-protection expenditures are only partly explained by the scaling up of existing safety-net and education programs. The increased spending is also due to new programs targeted to specific sections of the population. For example, a scholarship program of NPRs 8.9 billion targeted towards Dalits, girls and disabled children was introduced in FY09/10. Table A2 (Annex) presents the trends in the absolute levels of expenditures across the different programs from FY02/03 to FY09/10. Table A3, which provides trends in the allocations towards the three main cash-transfer programs, shows that some targeting of the old age allowance towards Dalits and Karnali residents, and allowances for Janajatis were introduced in FY08-09. The expansion of a food/cash-for-work program is also planned for the poverty-stricken high- and mid-Mountain areas. The government has also been implementing the Emergency Peace Support Project since 2008 to provide cash transfers to selected groups affected by the conflict, including families of those killed in the conflict, conflict-related widows and orphans, disabled people, and Maoist combatants. C: Social Protection: Program Coverage, Targeting and Efficiency 6.14. Cash benefits, until recently, have been low. The elderly, widow, and disability allowances, until 2008, were low, at monthly stipends of NPRs 200 for the elderly, NPRs 150 for widows and NPRs 125 for the disabled. The average size of the benefits amounted to 24 percent of the most recent 2003/04 poverty line of NPRs 7,896 per person per year, and 38 percent of the food poverty line of NPRs 4,966 per person per year, assuming the beneficiary is able to collect the entire allowance every month. Low levels of benefits that do not meet minimum basic needs are often the norm in many developing countries because of limited financing, trade-offs between coverage and benefit size, and the important concern of not lowering the incentive to work. Recent studies from India and Pakistan show that programs with low levels of benefits are also subject to less leakage than programs with higher benefits, due to increased probability of elite capture of the latter. Nepal is one of the few low-income countries, and the only country in South Asia, that provides a universal old-age social 69 (non-contributory) pension, and it could afford to do so due to modest benefits and a high eligibility age of 75.61 6.15. Changes in program eligibility criteria and increases in benefit levels have fuelled the recent expansion of the SP budget. Under the expansion plans for the Old Age Allowance program (as detailed in the Budget Speech for 2008-09) the eligibility age threshold was decreased to ―70‖ (to ―60‖ in Karnali zone and to ―60‖ for Dalits in all the regions) and the benefit levels were increased to NPRs 500 per month. Calculations using the 2003/04 National Living Standards Survey (NLSS) data62 show that such a projected expansion would have resulted in a 109 percent increase in the number of eligible people (from 360,732 to 752,797) and a 420 percent increase in the budget for the program to NPRs 4.5 billion, given 100 percent uptake of the program. Similarly, changes in the eligibility criteria for the widow allowance program (now called Single Women allowance) were also proposed by the GoN in September 200863. The program is no longer restricted to poor widows but is targeted to all single women above the age of 60 and those whose husbands have died, who were never married, or who separated. An increase in the benefit level to NPRs 500 per month was also proposed. This change in the eligibility criteria and benefit amount also translates into a large increase in the program budget. The numbers from the 2003/04 NLSS indicate that, assuming perfect targeting and 100 percent uptake of the program, the eligible population under the newly-expanded program would increase by 109 percent to 502,256, while the total budget for the program would have to increase by nearly 600 percent to NPRs 3.4 billion. 6.16. The 2008 expansion plan for social assistance also included raising the disability allowance benefits to NPRs 330 to Rs. 1000 per month depending on the level of disability64. However it is not clear how the degree of disability will be determined. The government also introduced a new transfer of NPRs 500 to people of endangered ethnicities (indigenous groups) of all ages as of 2008. 6.17. The actual figures show that the expansion of the budget for all these programs over 2007-08 to 2009-10 to be even larger than some of the above projections, at 718 percent. However, the number of beneficiaries increased by 65 percent (see Table A3). Figure 4 shows the drastic expansion of these programs since 2008: 61 There are only about 12 cases of universal flat programs that pay benefits to citizens above a certain age. Even within this group there are wide ranging differences with respect to eligibility age and benefit levels. There are only few low-income countries in this group (Palacios and Rajan, 2004). 62 NLSS 2003/04 is the most recent detailed living standards survey available and thus is used in calculations in the rest of this note. However, NLSS 2003/04 does not include detailed social protection program information (i.e. receipt, amount of benefit etc.). As part of the Nepal Social Protection Technical assistance the World Bank is undertaking a vulnerability assessment through a household survey which would also include program specific information for SP hence will be able to partially alleviate this strategic omission. 63 Government of Nepal, Budget Speech for Fiscal Year 2008/2009. 64 Government of Nepal, Budget Speech for Fiscal Year 2008/2009. 70 Figure 6.4: Approved Budget of Main Cash Transfer Programs (NPRs Million) 1995/6-2009/10 0 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 Source: MoLD 6.18. Despite this drastic increase in SP expenditures, coverage of programs remains limited. The total budget for the Old Age, Disability, and Widow/Single Women Allowance programs amounted to 0.14 percent of GDP in 2006/07, rose to 0.4 percent in 2008/09, and is estimated to be 0.8 percent of GDP in 2009/10 – representing a 100 percent increase in just one year. Nevertheless, the total coverage of these programs, including a new cash-transfer program for Janajatis, remains limited, at about 3 percent of the total population (see Table A3). Moreover, these programs have a history of budgetary shortfalls, the largest being in the Old Widow Allowance program, which generally results in a loss of entitlement and rationing. In 2008-09, the combined shortfall for all four of these cash transfer programs was NPRs 1.4 billion – almost a third of the total approved budget (see Table A3). 6.19. Programs benefit both the poor and non-poor. The universal nature of these cash transfer programs by definition mean that both poor and non-poor sections of the population benefit from them. However, in most instances65, the objective of universal old-age pension programs is to reach the elderly above a certain age, because they are thought to be among the most vulnerable. In a tight fiscal envelope environment, if the poverty rate among the elderly is not drastically higher, the universal coverage of elderly may not be a good vehicle to provide safety nets to poor. For instance, there is no evidence that the elderly population is poorer than the rest in Nepal. Instead household level data analysis using the 2003/04 National Living Standards Survey (NLSS)66 shows that around 50 percent of the elderly population eligible for old age allowance lives in households that are in the top two expenditure quintiles, indicating a high leakage rate67 for the Old-Age Allowance program.68 Isik-Dikmelik (2009)69 65 Kakwani and Subbarao (2005) ―Ageing and Poverty in Africa and the Role of Social Pensions.‖ UNDP International Poverty Centre Working Paper No. 8. 66 NLSS 2003/04 is the most recent detailed living standards survey available and thus is used in calculations in the rest of this note. However, NLSS 2003/04 does not include detailed social protection program information (i.e., receipt, amount of benefit, etc.). As part of the Nepal Social Protection Technical assistance the World Bank is undertaking a vulnerability assessment through a household survey which would also include program-specific information for SP, hence will be able to partially alleviate this strategic omission. 67 Leakage rate is the proportion of non-poor people among the beneficiaries (inclusion error). 71 also shows a 28 percent poverty rate for single elderly (above age 60) women, using NLSS 2004/05. This implies 72 percent of the Single-Women’s Allowance program benefits go to the non-poor. That said, many countries operate universal social pensions as these programs have a number of advantages, such as being politically and administratively easier to implement than targeted programs. However, being universal, such programs are expensive in addition to having a high level of leakage. In a developing country context, universal social pensions are therefore commonly limited to the very old and provide small benefits, as used to be the case in Nepal prior to the 2008 program expansion. 6.20. There is potential overlap of targeted beneficiaries across programs. The eligibility criteria of the Old-Age Allowance and the new criteria for the Single- Women’s Allowance program point to the overlap in the target group between the two programs. Consolidating the two programs, or changing the eligibility criteria to avoid any potential overlap, will minimize double-uptake of benefits and thus improve the overall the allocative efficiency of SP expenditures within the given budget. 6.21. There is evidence of under-coverage70 in other smaller safety-net programs. A preliminary analysis of participation in public works programs indicates that the poor and the socially excluded groups are not more likely to participate in these programs71. Further analysis is needed to determine whether the selection of sites or the presence of other barriers may be responsible for these participation rates, especially since GoN has proposed to expand the program. 6.22. Under-coverage problems are also pervasive in education programs targeted to marginalized groups. High levels of under-coverage were present for both Dalit and 50 percent Girls’ Scholarship programs at the primary level (81 percent and 88 percent respectively). Despite being targeted to girls from poor families, more than half of the beneficiaries of the 50 percent Girls’ Scholarship program came from families in the top three expenditure quintiles (Isik-Dikmelik, 2009). On the other hand, though there were no provisions requiring the recipients of Dalit scholarships to be from poor families, around 64 percent belonged to households in the bottom two expenditure quintiles. 6.23. Cash-transfer programs appear to have relatively lower overhead costs compared to other social-protection programs (excluding public-sector pensions). Preliminary estimates show that when total expenditures are disaggregated by wage and non-wage costs, social care services are the most cost-intensive, followed by the Poverty Alleviation Fund. While the level of the administrative costs is an important 68 It should be noted that the poverty faced by the elderly population may be different from the poverty of the households they reside in due to differences in intra-household allocation of resources which household level data is unable to discern. 68 Isik-Dikmelik, A. 2009. ―Targeting Systems for Safety Nets in Nepal: A Review.‖ Policy Note, SASSP, World Bank. 69 Isik-Dikmelik, A. 2009. ―Targeting Systems for Safety Nets in Nepal: A Review.‖ Policy Note, SASSP, World Bank. 70 Under-coverage rate is the proportion of poor households that are not included in the program (exclusion error). 71 More than 50 percent of the participants were lower-middle income and upper-middle income households. Similarly more than 50 percent of the participants of public works programs were Brahmin/Chhetris (Isik- Dikmelik, 2009). 72 indicator of efficiency, comparing across these programs is difficult, given the differences in their design and operational features. Figure 6.5: Breakdown of Wage and Non-Wage Costs by Program (2008/09) 120 100 0 9.5 22.94 15 15 80 24.875 31.68 50 60 40 % Avg wage exp. 20 0 % Avg non-wage exp. Source: Staff estimates based on interviews of relevant ministry officials D: In Conclusion 6.24. Although a large increase in social-protection expenditures has taken place in recent years, the amount is in the same range of most countries in the region. A key priority for the GoN would be to ensure that these redistributive expenditures are sustained as a permanent and integral part of the country’s development policy to ensure equitable and inclusive growth. That said, social protection programs are challenging to design and implement, and thus should be carefully managed and regularly monitored. 6.25. Better alignment of programs to needs is likely to generate substantial savings. The recent expansion of SP expenditures appears to have been quick responses to the recent crises and therefore reflect lack a vision and strategy. The majority of the programs benefit both the poor and non-poor, including those targeted to the poor (e.g., 50 percent Girls’ Scholarship program) and suffer from high leakage and under- coverage rates. This implies social-protection expenditures only partially benefit the neediest households. Possible options to increase the overall productivity of government outlays towards redistributive goals include: (i) Reviewing the eligibility criteria and benefit levels of the Old-Age and Widows’ Allowance programs to better align them to needs; (ii) Consolidating the Old-Age and Single Women’s programs to avoid double uptake of program benefits; and (iii) Targeting benefits specifically to the poor. For instance, if the Single Women’s Allowance program was targeted to only poor single women above the age of 60, the savings would amount to NPRs 2.17 billion annually72. To generate further fiscal space, the GoN could also consider targeting the Old-Age Allowance program to a larger section 72 Assuming there are no costs to targeting (Isik-Dikmelik, 2009). 73 of the poor elderly population than is currently covered by the program. This would represent a logical effort towards reducing old- age poverty, as is done in other countries in South Asia, such as India and Bangladesh. This would, however, introduce all the challenges of targeting. 6.26. Adequate funding and strengthening of the administration systems could help improve the overall performance of SP programs. Targeting errors often result from implementation challenges as is the case in many programs in the region73. One of the principal reasons potential beneficiaries fail to apply for programs is the lack of an adequate budget for program administration (Grosh at al, 2009). Adequate funding is required for sufficient staff, enough offices and travel funds to get the staff to the offices, and adequate spending on literature, media campaigns, targeted dissemination of information, and so on. Similarly, sound monitoring-and-evaluation systems allow program managers to implement programs more transparently and effectively. Investing in better targeting and administrative systems may be more costly initially, but would prove more efficient in reaching the poor and most vulnerable and thus generate greater savings in the medium term. 6.27. The decision to expand certain programs over others should be based on rigorous impact evaluations. Further consideration should be given to the possibility of reallocating expenditures within the existing SP budget so that resources are reallocated from less effective to better performing programs. This will require investing in the collection of more information and making use of the existing data. If an existing program is not successful in protecting the poor, other alternatives need to be considered to obtain better impact for the expenditures currently made towards social protection. Evaluation of existing programs will also help to draw lessons on what works and what does not in the Nepal context to help inform the design of new programs customized to the needs of the poor and vulnerable.□ 73 Ferre, C and M. Rashid. 2008. ―Modernizing Safety Nets in South Asia: Improving Regional Responses to Chronic and Transient Poverty.‖ 74