103669 v1 Making The Whole Greater Than The Sum Of The Parts: A Review of Fiscal Decentralization in Vietnam Summary Report 2015 Making The Whole Greater Than The Sum Of The Parts: A Review of Fiscal Decentralization in Vietnam Summary Report 2015 Abbreviation BVT Business Value Tax CBFA Committee on Budgetary and Finance Affairs CBRC China Banking Regulatory Commission CBIS Community Boards for Investment Supervision CIT Corporate Income Taxes DOF Department of Finance DP Development Partner DSA Debt Sustainability Analysis GDP Gross Domestic Product GOV Government of Vietnam HIFU HCMC Investment Fund for Urban Development LDIF Local Development Investment Funds LG Local Government LGFV Local-Government Financing Vehicles MOET Ministry of Education and Training MOF Ministry of Finance MOH Ministry of Health MOHA Ministry of Home Affairs MPI Ministry of Planning and Investment NA National Assembly NTP National Target Program ODA Official Development Assistance OECD Organization for Economic Cooperation and Development PAPI Provincial Governance and Public Administration Index PEFA Public Expenditure and Financial Accountability PER Public Expenditure Review PFF Public Financial Funds PIBs People’s Inspection Boards PIT Personal Income Tax PM Prime Minister PPC Provincial People’s Committee PSE Public Sector Efficiency PSP Public Sector Performance SBL State Budget Law SEDP Socio-Economic Development Plan SOE State Owned Enterprise UNDP United Nation Development Program VAT Valued Added Tax VBSP Vietnam Bank for Social Policies VDB Vietnam Development Bank VHLSS Vietnam Household Living Standard Survey WB World Bank Foreword This report provides a review of fiscal decentralization policies in Vietnam and their impact on the Government’s development objectives. It aims to inform reform of central-local fiscal relations in Vietnam to further promote growth and poverty reduction. The report was prepared under the overall guidance of Victoria Kwakwa (Country Director, WB Vietnam), Sudhir Shetty (Chief Economist, WB East Asia and Pacific Region), Rob Taliercio (Lead Economist, WB), Sandeep Mahajan (Lead Economist, WB), James A. Brumby (Practice Manager, WB), and Deepak Mishra (Lead Economist, WB) by a core team including: Quang Hong Doan (Senior Economist, WB); Indira Iyer (Senior Fellow, National Council of Applied Economic Research, India); Jorge Martinez-Vazquez (Regents Professor and Director, Department of Economics, Georgia State University); Nara Monkam (Research Director, Africa Tax Administration Forum, South Africa); Minh Van Nguyen (Senior Economist, WB); Cuong Dinh Pham (Former General Director at the Vietnam Ministry of Finance); Abha Prasad (Senior Debt Specialist, WB); Habib Rab (Senior Economist – Task Team Leader, WB); Anwar Shah (Director of the Centre for Public Economics, Chengdu/Wenjiang, China); and Quyen Hoang Vu (Economist, WB). Research assistance and technical inputs were provided by: Huong Dang (Consultant, WB); Hai Anh La (Economist, Vietnam Academy of Social Sciences); Phuong Anh Nguyen (Research Analyst, WB); Lucy Pan (Economist, WB); Hien Thu (Economist, Mekong Economics); Le Quang Thuan (National Institute of Finance). Linh Anh Thi Vu (Program Assistant, WB) provided administrative support. The State Budget Department and the State Treasury Department at the Vietnam Ministry of Finance; and the Committee for Budgetary and Financial Affairs (CFBA) at the Vietnam National Assembly provided very helpful guidance and advice throughout the study. The team is very grateful to the UNDP Vietnam and CFBA project team on “Strengthening the Capacities of Budget Oversight for People’s Elected Bodies in Vietnam” for support in facilitating dialogue with local authorities in Vietnam. The team is very thankful to local authorities from the following provinces for extensive discussions and feedback throughout the study: Lam Dong; Da Nang; Cao Bang; Bac Kan; Khanh Hoa; Quang Nam; Tay Ninh; Hau Giang; Can Tho; Dong Nai; Son La; Nghe An; Yen Bai; Lao Cai; Ca Mau; Kien Giang; Dien Bien; Quang Ngai; Ha Noi; Ho Chi Minh City; Hai Phong. The team thanks the following colleagues for advice and inputs across different parts of the report: Jairo Acuna-Alfaro (Policy Advisor, UNDP); Cut Dian R.D. Agustina (Economist, WB); James Anderson (Senior Governance Specialist, WB); Cuong Duc Dang (Senior Urban Specialist, WB); Gabriel Demombynes (Senior Poverty Economist, WB); Kari Hurt (Health Cluster Leader, WB); Chris Jackson (Lead Economist, WB); Sang Minh Le (Consultant, WB); Iain Menzies (Senior Water and Sanitation Specialist, WB); Lan Thi Thu Nguyen (Natural Resources Economist, WB); Quang Vinh Nguyen (Senior Water and Sanitation Specialist); Ngoc Thi Nguyen (Consultant, WB); Suhas Parandekar (Senior Economist, WB); Madhu Raghunath (Senior Urban Specialist, WB); Paul Vallely (Senior Transport Specialist, WB); Lorena Vinuela (Public Sector Specialist, WB); Linh Hoang Vu (Poverty Economist, WB); Lan Anh Vu (Human Development Specialist, WB); Ha Thi Tu Vu (Consultant, WB). Review comments at concept and decision stages were gratefully received from: Do Viet Duc (Ministry of Finance, Vietnam); Jonas Frank (Senior Public Sector Specialist, WB); Robert Gilfoyle (Senior Financial Management Specialist, WB); Nguyen Van Hao (Ministry of Finance, Vietnam); Tran Kim Hien (Ministry of Finance, Vietnam); Kai Kaiser (Senior Economist, WB); Lili Liu (Lead Economist, WB); Estelle Liu (Senior Economist, IMF); Vu Nhu Thang (President of National Institute of Finance, Vietnam Ministry of Finance); Theo Thomas (Lead Economist, WB); Vu Thanh Tu Anh (Lecturer of Economics and Director Research, Fulbright Economics Teaching Program); Michel Welmond (Program Leader, WB). The report findings and recommendations were presented at several workshops with the National Assembly; the Ministry of Finance; scholars and researchers at the Vietnam Academy of Social Sciences (VASS), Ho Chi Minh City University of Economics (HEU), and the Harvard University Fulbright Economics Teaching Program; and members of the media. Support for the preparation of this report is gratefully acknowledged from the Australian Department for Foreign Affairs and Trade and the United Kingdom Department for International Development. Table of contents Vietnam Fiscal Decentralization Review: Summary Report 1 Introduction 4 Expenditure decentralization and accountability 8 Expenditure performance of local authorities 16 Local revenue arrangements 27 Intergovernmental fiscal transfers 36 Local borrowing 42 List of Figures Figure 2.1: Per capita spending pre central transfers 10 Figure 2.2: Per capita spending post central transfer 10 Figure 2.3: Expenditure and revenue decentralization 11 Figure 2.4: Local recurrent spending in selected functions (% of total) 11 Figure 3.1: Average divergence between actual and budgeted capital spending 2006-2011 by region (%) 19 Figure 3.2: Lower than average PSP and higher than average PSE 21 Figure 3.3: Higher than average PSP and higher than average PSE 21 Figure 3.4: Lower than average PSP and lower than average PSE 21 Figure 3.5: Higher than average PSP and lower than average PSE 21 Figure 3.6: Central Highlands Provinces against efficiency frontier 23 Figure 3.7: Mekong Delta Provinces against efficiency frontier 23 Figure 3.8: NCCC Provinces against efficiency frontier 23 Figure 3.9: Northern Mountain Provinces against efficiency frontier 24 Figure 3.10: Red River Provinces against efficiency frontier 24 Figure 3.11: South East Provinces against efficiency frontier 24 Figure 4.1: Central Highlands 30 Figure 4.2: Mekong Delta 30 Figure 4.3: NCCC 30 Figure 4.4: Northern Mountains 30 Figure 4.5: Red River Delta 30 Figure 4.6: South East 30 Figure 4.7: Central Highlands 31 Figure 4.8: Mekong Delta 31 Figure 4.9: NCCC 31 Figure 4.10: Northern Mountains 31 Figure 4.11: Red River Delta 31 Figure 4.12: South East 31 Figure 6.1: Local government debt 2006-2012 44 Figure 6.2: HCMC Debt-to GDP ratio 46 Figure 6.3: HCMC Debt service-to-revenue ratio 46 Figure 6.4: HCMC Interest payment/Recurrent Expenditure 46 Figure 6.5: HCMC Overall and Primary Balance (% of GDP) 46 Vietnam Fiscal Decentralization Review: Summary Report 1 BACKGROUND: Local authorities in Vietnam are responsible for over half of total government spending, thanks to fiscal decentralization policies implemented over the past eighteen years. Total government spending in Vietnam is close to 30 percent of Gross Domestic Product; local authorities’ spending is close to 17 percent of GDP. Study of fiscal decentralization therefore is central to understanding government spending in Vietnam and its significant impact on the country’s successful record of economic growth and poverty reduction. MOTIVATION: To date however there has been little analysis of how fiscal decentralization has enabled local authorities to effectively, efficiently and accountably spend money on public services for development. Spending decentralization has grown more quickly than information on the effectiveness of local fiscal policies. This has prompted a number of questions from the National Assembly, the Central Government, and other stakeholders. This report tries to answer some of these questions to help inform future changes to central-local fiscal relations: How are local spending choices aligning with national level objectives? How much spending responsibility do local authorities have? Are current policies and institutions sufficient to ensure that the money is spent well and in line with citizens’ needs? How much do different provinces spend on delivering the same public services and what might explain the differences? What results are being achieved? How effectively are local authorities able to raise revenues to finance local spending? Are fiscal transfers sufficient to effectively cover all local spending needs? Are transfers being equitably distributed? Are there more opportunities for provinces to borrow to meet investment needs? KEY FINDINGS AND RECOMMENDATIONS: There has been significant spending decentralization both in terms of spending assignments and autonomy over resource allocation decisions. Fiscal decentralization policies have helped to channel more spending to the poorest parts of the country where development needs and costs of service delivery are higher. There is also evidence that decentralization has on the whole moved in line with administrative capacity. Institutions of fiscal transparency and accountability however have not kept pace 2 with greater spending responsibility. The revision to the State Budget Law 2002 offers an important opportunity to enhance transparency and public participation in local budgets, and ensure clearer accountabilities in a nested budget system that otherwise leads to overlaps and confusion. The report finds that spending efficiency of local authorities can improve by imposing more discipline on budget implementation. For example, evidence shows that practices such as large carry overs and flexibility on the use of over-realized revenue reduce spending efficiency. The report recommends to significantly reform these practices as part of the SBL 2002 revision process. Despite strong spending decentralization, local authorities have little to no revenue autonomy. Evidence shows that the current arrangements do not have a discernable negative impact on revenue collections. Yet revenue autonomy is an essential part of fiscal decentralization to promote greater accountability and increased investments in high growth potential areas. The report recommends some gradual steps that Vietnam can take to phase in some independence for local authorities on revenue policy decisions. Evidence shows that intergovernmental fiscal transfers are equalizing both from the center down to provinces, but also from provinces down to districts. This has helped to close large imbalances across provinces and across districts. This in turn has contributed significantly to development particularly in the poorer provinces. There is scope to further strengthen the allocation norms used to determine spending needs, which in turn feed into estimates of balancing transfers, for which the report makes a number of recommendations. The report also points to ideas for strengthening the system of targeted transfers from an input driven approach to a more performance-based one. Current limits on local borrowing have helped to maintain local debt at prudent levels, but these limits are not based on economic rationale. Evidence shows that provinces such as HCMC could potentially borrow more and maintain prudent levels of debt. However, higher borrowing also needs to be matched by increased revenue autonomy to make sure that local authorities do not run into liquidity problems. It will also require local debt to be brought onto the budget. This is critical for transparency. Based on the above, the report recommends a revised approach to debt ceilings. CONCLUSION: A combination of these measures should help to maintain the government’s successful redistributive policies to meet social objectives, and 3 enable high economic potential provinces to invest in growth enhancing initiatives, whilst enhancing transparency to ensure accountability. Lack of transparency and coordination in intergovernmental fiscal relations can lead to fragmentation and unequal development across the country. Intergovernmental fiscal relations should ensure that development efforts of all provinces exceed the sum of their individual parts so that they can help eliminate extreme poverty and promote shared prosperity in Vietnam. INTRODUCTION 4 01 The State Budget Law 2002 (SBL 2002) has enabled decentralization of important fiscal responsibilities to local authorities over the past ten years. This report responds to demands for more analysis of fiscal decentralization policies in Vietnam and the extent to which these have delivered on their stability, equity and efficiency objectives. It aims to inform future changes to the system of intergovernmental fiscal relations through revisions to the SBL 2002 and adoption of 2016-2020 Stability Period regulations. The SBL 2002 establishes the key principles for spending assignments, revenue arrangements, and intergovernmental fiscal transfers. It also grants provincial authorities a fair degree of autonomy to determine fiscal relationships with districts and communes within their jurisdiction. The current system allows a reasonable level of differential treatment across provinces to take account of their specific circumstances. There is general acknowledgement that the SBL 2002 has provided a solid framework for public finance management including intergovernmental fiscal relations. At the same time, after 10 years of implementation a number of issues arise such as: clarity of spending responsibilities and local level accountability including for national priorities and objectives; spending performance of local authorities; the effectiveness of local revenue arrangements in meeting spending needs and the potential for increased revenue autonomy for provinces; the extent to which provincial authorities are promoting or impeding central government’s redistribution efforts; and the potential for increased debt financing for local authorities. This report builds on existing research and aims to provide new analysis and perspectives on the above areas through extensive study of available data, consultations with central and local authorities, and reviews of laws, regulations and policies at central and local levels. It covers the following five pillars of fiscal decentralization: (i) expenditure decentralization and accountability; (ii) expenditure performance of local authorities; (iii) local revenue arrangements; (iv) intergovernmental fiscal transfers; and (v) local borrowing. INTRODUCTION 5 Local fiscal policies have Local authorities have been granted increasing a significant impact on levels of fiscal responsibilities since the adoption of Vietnam’s development. This report looks at the the State Budget Law (2002). Local fiscal policies in extent to which fiscal Vietnam have a significant impact on the country’s decentralization policies record of economic growth and poverty reduction. have achieved their They determine decisions on public service delivery development objectives and the overall development trajectory of the country. Poor spending or revenue decisions at local level, coupled with lack of transparency can lead to inefficient spending, unequal development across the country and reversal of the country’s successful track record in delivering inclusive growth. Despite the importance of fiscal decentralization in Vietnam’s development, there has been little study on the successes and failure of policies and institutions in this area. This report responds to demands for more analysis of fiscal decentralization policies in Vietnam and the extent to which these have delivered on their intended objectives of equity and efficiency to promote economic growth and poverty reduction. It aims to inform future changes to the system of intergovernmental fiscal relations through revisions to the SBL 2002 and adoption of 2016-2020 Stability Period regulations. The report aims to inform The SBL 2002 establishes the key principles for changes to the SBL 2002 spending assignments, revenue arrangements, and and 2016-2020 Stability intergovernmental fiscal transfers. It also grants Period regulations, which provide the frameworks for provincial authorities a fair degree of autonomy intergovernmental fiscal to determine fiscal relationships with districts and relations in Vietnam. communes within their jurisdiction. There is general acknowledgement that the SBL 2002 has provided a solid framework for public finance management including intergovernmental fiscal relationships. After 10 years of implementation of the SBL 2002, 6 a number of issues arise in relation to the system of intergovernmental fiscal relations, such as: clarity of spending responsibilities and local level accountability including for national priorities and objectives; spending performance of local authorities; the effectiveness of local revenue arrangements in meeting spending needs and the potential for increased revenue autonomy for selected provinces; the extent to which provincial authorities are promoting or impeding central government’s redistribution efforts; and the potential for increased debt financing for local authorities within prudent limits. This report builds on existing research and aims to provide new analysis and perspectives on the above areas through extensive study of available data, consultations with central and local authorities, and reviews of laws, regulations and policies at central and local level. It analyzes five pillars of fiscal decentralization to assess the extent to which these are promoting or hindering the government’s development objectives. These five pillars are: (i) expenditure decentralization and accountability; (ii) expenditure performance of local authorities; (iii) local revenue arrangements; (iv) intergovernmental fiscal transfers; and (v) local borrowing. 7 EXPENDITURE DECENTRALIZATION AND 8 ACCOUNTABILITY 02 Key issues: local authorities in Vietnam have historically accounted for a significant share of total public sector spending, which has increased further since the adoption of the SBL 2002. Feedback from implementation of the SBL 2002 and earlier studies have highlighted that transparency and accountability of local budgets have not kept pace with the rising delegation of spending responsibility. Objectives: the objectives of this chapter are to assess the extent to which public spending decisions have been decentralized in Vietnam, and options for strengthening the institutional arrangements around spending assignments and accountability to promote allocative and productive efficiency. Key findings: local authorities in Vietnam represent a large share of the local economy, particularly in poorer provinces. The latter have also seen higher levels of per capita spending compared to richer provinces, thanks to the redistributive nature of fiscal transfers in Vietnam. Local authorities in Vietnam are responsible for around 55 percent of general government spending. They account for over 75 percent of total capital spending, and in key social service delivery areas such as education (90 percent), economic services (80 percent) and a growing share of health (from 72 percent of recurrent spending in 2006 to 88 percent in 2011). Decentralization within provinces, down to district authorities has also increased in the 2006-2011 period. In half of all observations, district spending represents more than 45 percent of total local spending. District authorities account for the majority of recurrent spending in both health and education within provinces, though decentralization of capital spending has been less pronounced. An institutional review of expenditure assignments across different tiers of government suggests that the current 9 arrangements would benefit from some specificity, including on: exclusive spending mandates of central government, exclusive responsibilities of different tiers of government within one jurisdiction; and exclusive responsibilities of local authorities. On budgeting and accountability, the chapter finds that the compressed budget calendar and current appropriation structure warrant review to strengthen legislative oversight. At the same time, there is considerable scope for improving budget transparency and participation through clearer communication of budget policies, and publication of the draft budget proposals. There is currently limited participation in the budget preparation process in Vietnam, when evidence from elsewhere, and partly also from Vietnam shows better outcomes and satisfaction with service delivery when participation is promoted. Recommendations: (i) more explicit assignments on service delivery for different tiers of government within a jurisdiction; (ii) decompression of the budget preparation calendar; (iii) greater autonomy over budget approval authority of people’s councils and clearer provisions on budget appropriations; (iv) elimination of minimum allocation requirements for sectors; and (v) increased transparency and participation in the budget process. 10 EXPENDITURE DECENTRALIZATION Local government Local government spending constitutes an important spending has played share of provincial economies in Vietnam. It an important role in delivering services to has played a particularly important role in poorer the poorest parts of the provinces that are sparsely populated. This reflects country. efforts to expand service delivery into poorer areas, where costs of service provision are also higher. Central transfers to local authorities have helped in this redistribution process (figures 2.1 and 2.2). Figure 2.1: Per capita spending Figure 2.2: Per capita spending pre central transfers post central transfer Amount (’000 VND) Amount (’000 VND) [1-000} [1000-2000} [2000-3000} [3000-4000] [4000-5000] [5000-6000] [6000-7000] [7000-8000] [>=8000] No data [1-000} [1000-2000} [2000-3000} [3000-4000] [4000-5000] [5000-6000] [6000-7000] [7000-8000] [>=8000] No data Local authorities are Local authorities are now responsible for just over responsible for over half of total government spending, which is high half of total government by international standards (figure 1.3). In Vietnam, spending, which is high by international although higher levels of spending have also standards. been matched by higher levels of local revenue, local authorities have little to no autonomy over revenue policy and administration. Despite this, with decentralized revenue and unconditional balancing transfers accounting for more than 75 percent of 11 core spending on average across all provinces, local authorities have had a fair amount of discretionary resources at their disposal. Over three quarters of Local authorities’ share of total government recurrent recurrent spending in spending in important areas like education, health, social sectors is under economic services, and public administration has the control of district authorities. been high and increased further between 2006 and 2011. For education, local authorities account for 80- 90 percent of total recurrent spending and for health the ratio is around 75-80 percent. Decentralized revenue and unconditional balancing transfers help to cover over three quarters of local spending, suggesting high level of discretionary resources available to local authorities. F  igure 2.3: Expenditure and revenue Figure 2.4: Local recurrent spending in selected decentralization functions (% of total) 35% 100% 95 Sub-national Expenditures (% of total expenditures) Sub-national Expenditures (% of total expenditures) 70 80 70 80 30% 91 90 90 90 80% 88 88 87% 86% 88% 86% 87 86 25% 85 79% 60 60 20% 71% Vietnam 60% Vietnam 80 80 80 50 50 79 78 79 15% 30% 29% 40% 75 40 40 27% 28% 25% 10% 23% 72 72 71 70 70 70 30 30 68 69 20% 5% 66 66 10 20 10 20 65 64 64 0% 0% 62 60 2006 2007 2008 2009 2010 2011 2006 2007 2008 2009 2010 2011 0 0 20 Recurrent Social0security/Central 10 30 (%) 40 50 60 70 80 0 10 20 30 40 50 60 70 80 90 100 Administration Economic Sub-national Revenues (% of total revenues) Transfers to sub-national(% of total sub-nation revenues & grants) Central government share of social security (%) Education Health There is significant Vietnam exhibits a high degree of spending decentralization within decentralization within provinces, which has provinces, bringing increased over the last two Stability Periods. Half of resource allocation decisions closer to the all observations in 2011 showed that districts were people. spending at least 45 percent of total local recurrent spending. District authorities have been responsible for most of the recurrent spending in both education and health within most provinces. Districts’ share of local recurrent spending in education has remained relatively high and constant in the last two Stability 12 Periods, but in health there has been a marked shift in spending decentralization to districts. In contrast, most of the capital spending (around 70 percent on average) is carried out by provincial authorities. Higher capacity local In Vietnam, the evidence shows that on average authorities have higher higher capacity local authorities have higher levels levels of spending responsibilities. of spending responsibilities. Experience in other countries has shown that decentralization too quickly can have serious adverse impacts when local administrative capacities are low. The evidence in Vietnam also shows that provinces that have more people in rural areas have also decentralized more responsibility to the districts. This is a good sign since more rural provinces tend to have lower population density, therefore decentralization can help to better take account of local preferences. There is scope for The SBL 2002 currently assigns the same clarifying spending expenditure responsibilities to both central and responsibilities between center and province. provincial authorities i.e. most spending assignments are shared/concurrent. Sector legislation in health, education and transportation for example provides some clarity in terms of assigning exclusive responsibility for regulation to central government. In general, however, having so many shared functions between center and province causes ambiguity over spending assignments. Some functions should Some functions should rest exclusively with central rest exclusively with authorities because their costs and benefits are central government. national in scope e.g. national defense and security, foreign policies). In Vietnam, however, local authorities are also expected to participate in funding of these services. Several local authorities have indicated that this creates ambiguities and pressures on local budgets, which should be addressed through clearer assignment of exclusive responsibilities to central government. In the case of non-exclusive/ shared functions, the explicit responsibilities of center 13 and provincial authorities should be clearly set out based on agreed criteria. Within a jurisdiction, Provincial level resolutions on revenue and service delivery expenditure assignments have provided significant responsibility should rest clarity on spending and revenue assignments with only one level of government. for provincial, district and commune authorities. However, in many cases, both province and district authorities are also assigned the same service delivery responsibility. Therefore unlike center- province division of responsibilities, there are explicit assignments for authorities within provinces; but in some cases, all tiers within the province are explicitly responsible for the same services. Whilst different orders of government can be responsible for different tasks within a jurisdiction, it is better to avoid assigning different orders of government to perform the same task within a jurisdiction. This leads to overlap and makes it difficult to hold each level of government accountable. The nested budget Although the SBL 2002 has introduced significant system complicates clarity on the roles and responsibilities of different budget preparation and actors, the nested budgeting system complicates monitoring. Some SBL 2002 provisions also budget preparation and monitoring. Vietnam is dilute accountability to one of few countries that still operates such a People’s Councils. nested system, which creates several challenges. Firstly, the multi-layered budget approval process significantly compresses the budget calendar. This gives Provincial Councils little time to review the draft budget. Secondly, the authority of, and therefore accountability to, local councils is diluted by the following provisions in the SBL 2002: “veto rights” of higher levels of government over budgets approved by lower tiers (Article 47) – which albeit is applied only in cases where lower tiers have made incorrect allocations; and the lack of formal requirement to seek legislative approvals before the executive makes changes to budget appropriations. Central fiscal rules and Despite increased responsibility over spending, 14 norms on minimum central fiscal rules and norms affect local autonomy allocations impact over budget decisions in selected areas and distort adversely on local autonomy. resource allocation. There are minimum allocations set for education and science and technology without due consideration of actual needs or the level of service provision by central authorities within a province. A minimum allocation to an area such as science and technology without link to capacity can lead to waste. Rules on the need to spend 50 percent of over-realized revenue on wages 50 percent of over-realized revenue on wages in the next budget year and 50 percent on important capital projects fail to take advantage of recognized budget priorities and leads to the inefficient use of resources. There is much budget Great efforts have been made to ensure publication of information in the public budgetary information, but current practices on budget domain but limited public disclosure do not encourage citizen participation. participation. The general public is not able to contribute to the budget preparation process as the budget is only published after it is approved by the legislature. The current budget classification structure, budget table templates, and coverage of fiscal activities in budget documents do not facilitate active citizen participation in scrutinizing public finances and providing meaningful feedback to local authorities. The highly technical presentation of budget documents prevents the public from understanding let alone analyzing budget policies. Budget policies should be Budget disclosure should be complemented with clearly communicated to efforts to encourage citizen participation as disclosure promote participation by itself will have limited impact on accountability for service delivery. At a minimum this would involve publication of the draft budget submitted to the National Assembly and People’s Councils. But in addition, the government should look at further developing its existing Citizens’ Budgets to communicate the State Budget in a simple way so that it is understood by as many people as possible. Evidence from the PAPI 15 survey show a very clear link between participation in public finance and project design and satisfaction with service delivery at the local level. Recommendations on expenditure decentralization Issues Recommendations Simplify Articles 31 and 33 in the SBL 2002 stating Lack of exclusive spending assignments for that all tiers may spend on any functional area central government in SBL 2002 leads to local but that some functional areas are the exclusive authorities taking on the responsibilities of responsibility of central government. This may also central government. need to be addressed in the upcoming Law on Local Government. The current budget preparation calendar is Review Article 45 of SBL 2002 to decompress the too compressed to enable local councils to budget calendar, including earlier start to budget scrutinize the draft budget. preparation process. State Budget Law 2002 SBL provisions that give upper tier authorities Review Article 47 of SBL 2002 to assign more “veto rights” over budgets approved by lower clearly authority to local councils for budgets at tier governments dilute local councils’ authority. their own level. SBL provisions on changes to budget Review Article 49 of SBL 2002 to indicate that appropriations, undermines accountability to any adjustments to appropriations need to be local councils as the latter are not required to mandates by the legislature. approve re-appropriations. The general public is not able to contribute to the budget preparation process as the budget Require publication of budget proposal when it is is only published after it is approved by the submitted to the local legislature. legislature. The lack of explicit assignments between Each provincial resolution should set out the center and province for service delivery can Stability Period regulations explicit responsibilities of center versus provincial lead to ambiguity and undue pressures for local authorities within that jurisdiction. budgets. For shared functions, enabling multiple orders of government to provide the same service/ Provincial resolutions on revenue and spending perform the same task within a jurisdiction assignments should set explicit responsibility for can lead to overlaps and poor planning and specific tasks within a jurisdiction. budgeting. Central fiscal rules and norms affect local Reconsider minimum allocation norms for sectors autonomy over budget decisions and distort or types of expenditure in the next Stability Period. resource allocation. Budget Law regulations The current budget classification structure, Further develop the existing Citizens’ Budget to State budget table templates, and coverage of fiscal communicate in a more accessible manner budget activities in budget documents. policies of local authorities. EXPENDITURE PERFORMANCE OF LOCAL 16 AUTHORITIES 03 Key issues: Earlier studies have indicated a lack of credibility in public spending plans, due to current policies on use of over realized revenue, extra budgetary financing and carry over practices. This leads to a loss of budget transparency, which can impact negatively on productive efficiency. This is a matter of prime concern to policy makers now as Vietnam faces tightening fiscal conditions with slowing revenue mobilization. Objectives: The objectives of this chapter are to determine if the above spending management practices are impacting on spending efficiency, and the implications for reform to promote greater budget credibility and transparency. Key findings: Local spending plans lack credibility when compared to actual outturns. This is particularly the case for capital spending, which is often more than 50 percent of what was budgeted, well above good practice guidelines of maintaining spending within 5 percent of budget. The lack of credibility arises partly due to the flexibility accorded to the executive to change budget appropriations approved by the legislature, which was discussed in chapter 2. However there are also several institutions and policies embedded in the SBL 2002 that enable the executive to take advantage of the flexibility over budget appropriations, including: use of over- realized revenue; access to extra budgetary financing; and carry over practices. These can impact on productive efficiency. Analysis of productive efficiency finds that there are some clear differences across provinces. On average, provinces in the Northern Mountains spend 42 percent more and those in the Central Highlands 35 percent more to achieve the same outputs/ outcomes as the most efficient provinces. In the South East and the Mekong River Delta on the hand, the average is closer to 7 percent and 9 percent respectively. Some of these disparities are associated with differences in population density, and therefore conditions that local 17 authorities cannot control. But poorer performers within a region also tend to have less credible spending plans and higher levels of carry overs compared to their peers in the same region. The share of wages and salaries in recurrent spending also seems to play a role, including in terms of crowding out other critical spending. Recommendations: (i) spending of over-realized revenue should not be tied to specific categories and should be approved by legislature; (ii) extra budgetary activities should be integrated into the overall budget; (iii) limit the level of carry over spending as a share of budget; (iv) introduce population density criteria in allocation norms to estimate spending needs of local authorities; (v) review wage bill and application of salary reform policies. 18 EXPENDITURE PERFORMANCE With increased spending With increased spending responsibility delegated to decentralization there local authorities, it is important to look at expenditure is more demand for spending performance performance across provinces. This has been a topic information. of high interest, particularly for the National Assembly and central authorities because reporting on service delivery performance has not kept pace with the increased spending decentralization. In addition to this, tightening fiscal conditions in Vietnam calls for a better understanding of how much different provinces are spending to deliver the same services, which should help identify areas of potential efficiency gains. The study finds that local spending plans lack Local spending plans lack credibility, which reduces transparency and impacts credibility, which reduces on efficiency. This is particularly the case for capital transparency and impacts spending, which is often more than 50 percent on efficiency. of what was budgeted, well above good practice guidelines of maintaining spending within 5 percent of budget (figure 3.1). The lack of budget credibility is partly due to the flexibility accorded to the executive to change budget appropriations approved by the legislature. As recommended in chapter 2, this warrants review of Article 49 in the SBL 2002 to clarify rules on changes to appropriations. The higher than budgeted spending is also due to spending from contingency reserves, and extra budgetary funding sources for capital investment projects. Figure 3.1: Average divergence between actual and budgeted capital spending 2006-2011 by region (%) 19 190 172 159 2006 2007 2008 2009 2010 2011 140 116 117 104 84 90 50 52 47 40 32 22 19 -10 Central Highlands Northern Mountain Northern Central and Mekong River Delta Red River Delta South East Central Coast Source: Staff estimates based on published state budget data Policies on over realized There are several institutions and policies embedded revenue, off-budget in the SBL 2002 that enable the executive to take financing, and carry overs reduce budget credibility. advantage of flexibility over budget appropriations, and which this study recommends be reviewed. The first is around use of over-realized revenue, for which, according to regulations, half has to be channeled to salaries and the other half to capital regardless of actual spending needs. The second policy area is on the use of off-budget financing mostly for capital investments, which ranges from 5 to 20 percent of the budget. These extra-budgetary sources lack transparency and create a distorted picture of budget allocations. The third policy area contributing to lack of credibility in spending plans relates to the practice of carry overs. Unlike other countries, there are currently no limits on the level of carry overs, which can go up to 50 percent for some provinces. Lack of spending plan The above institutions and policies on budget credibility dilutes link management can impact adversely on the efficiency between budgets and of spending. Maintaining a credible spending plan is a plans. necessary, albeit not sufficient, condition to promoting more effective and efficient government spending. In addition, poor credibility of spending plans dilutes the link between local development plans and local budgets. Despite the strong role of local spending in provincial economies in Vietnam, the role of public 20 finances is insufficiently covered in local development plans reviewed during this study. Differences in spending The study looks at various measures of spending efficiency across efficiency across provinces and the extent to which provinces are due above budget management practices might affect to factors outside of those measures. Using available data on three province’s control but also due to lack of sectors (education, health and transportation), the credibility in spending study computes Public Sector Performance indices, plans. reflecting the outputs and outcomes achieved in the three sectors. It uses this to derive Public Sector Efficiency scores (figures 3.2-3.5). The analysis finds that within regions there are important disparities in efficiency between provinces. Some of these disparities are associated with differences in population density. But poorer performers also tend to have less credible spending plans and higher levels of carry overs compared to their peers in the same region. There are also cases where provinces are very efficient but are performing poorly on critical service delivery outcomes, meaning that these provinces could consider increasing local spending in those particular sectors and spending less in some others. Public sector performance and efficiency indicators across provinces 21 Figure 3.2: Lower than average PSP and Figure 3.3: Higher than average PSP and higher than average PSE higher than average PSE 1.2 Public Sector Performance Public Sector E ciency 2.5 Public Sector Performance Public Sector E ciency 2.5 3.0 1.0 2.0 2.0 2.5 0.8 1.5 2.0 1.5 0.6 1.5 1.0 1.0 0.4 1.0 0.2 0.5 0.5 0.5 0.0 0.0 0.0 0.0 Lam Dong Binh Duong Hai Duong Hai Phong Dong Nai Tien Giang Binh Dinh Nam Dinh Khanh Hoa Can Tho Ho Chi Minh Tay Ninh Thai Binh Ha Noi Ben Tre BRVT Hau Long Kien Vinh An Dong Bac Phu Binh Ninh Bac Ha Hung Binh Giang An Giang Long Giang Thap Lieu Yen ThuanThuan Ninh Nam Yen Phuoc Mekong River Delta Northern Central Red River Delta South and Central Coast East CH Mekong River NCCC Red River Delta South East Delta Figure 3.4: Lower than average PSP and lower Figure 3.5: Higher than average PSP and lower than average PSE than average PSE 1.2 Public Sector Performance Public Sector E ciency 1.6 Public Sector Performance Public Sector E ciency 1.0 1.4 1.4 1.4 1.2 0.8 1.0 1.2 1.2 0.6 0.8 1.0 1.0 0.4 0.6 0.4 0.8 0.8 0.2 0.2 0.0 0.0 0.6 0.6 Lang Son Tuyen Quang Thai Nguyen Quang Binh Dak Nong Quang Nam Soc Trang Bac Giang Cao Bang Ha Giang Lai Chau Son La Vinh Phuc Quang Tri Ninh Binh Kon Tum Dak Lak Phu Tho Lao Cai Hoa Binh Dien Bien Ha Tinh Yen Bai Gia Lai Bac Kan Tra Vinh TTH Ca Mau 0.4 0.4 0.2 0.2 0.0 0.0 Central Mekong Northern Northern Mountain Red Da Nang Nghe An Thanh Hoa Quang Ngai Quang Ninh Highlands River Central and River Delta Central Coast Delta Northern Central and Central Coast Red River Delta 74 Performance Public Sector On average provinces Based on the above, a productive efficiency frontier in poorer regions spend is estimated based on the maximum level of output 35-40 percent to deliver attained by local authorities in Vietnam for a given set the same level of services as the most efficient of inputs. The purpose is to review the efficiency of provinces. provinces relative to such an efficiency frontier. The frontier is derived based on PSP scores against per capita spending of the most efficient provinces. This analysis provides an approximate idea of the difference in spending across provinces for achievement of a given level of outputs and outcomes. It finds that on average, provinces in the Northern Mountains region spend 42 percent more and those in the Central Highlands 35 22 percent more to achieve the same outputs/outcomes as the most efficient provinces. In the South East and the Mekong River Delta on the hand, the average is closer to 7 percent and 9 percent respectively. As noted above, some of these differences are associated with geographical characteristics and population density, but also partly due to budget management practices. Within the regions, the analysis finds a number of Within regions, provinces outliers for which productive efficiency is considerably that are less efficient lower than for other provinces in the same region also have less credible (figures 3.6-3.11). For the poorer regions (Northern spending plans. Mountains and Central Highlands), the outliers also happen to have the least credible spending plans and highest levels of carry over spending. In the Northern Central and Central Coast region, the province of Danang performs very well in terms of service outputs and outcomes, but it also spends much more despite high population density and urbanization. Econometric analysis of possible determinants of the Determinants of relative relative efficiency of provinces in Vietnam confirms efficiency of provinces in some of the hypotheses in the fiscal decentralization Vietnam are in line with literature. Findings show that dependence on transfers the fiscal decentralization can impact negatively on efficiency. There may be other literature. factors (not covered in the analysis) such as capacity to develop sound budgets, timely budget allocations and capacity to implement projects. Findings also show that population density, urbanization rates, and budget transparency have a positive impact. Surprisingly, higher per capita GDP was found to reduce efficiency, though this may point to residents in poorer localities being more active in monitoring the quality of service delivery. The credibility of spending plans has a slight negative impact on efficiency, and the share of recurrent spending in overall spending seems to have a strong negative impact on efficiency. Provincial performance against efficiency frontier across regions 23 Figure 3.6: Central Highlands Provinces against efficiency frontier 2.5 Hanoi 2.0 PSP scores 1.5 Hau Giang Lam Dong Gia Lai Dak Lak Kon Tum 1.0 Dak Nong 0.5 0.0 13.0 13.2 13.4 13.6 13.8 14.0 14.2 Log of per capita spending Figure 3.7: Mekong Delta Provinces against efficiency frontier 2.5 Hanoi 2.0 1.5 PSP scores Tien Giang Long An Can Tho Hau Giang 1.0 Vinh Long Ben Tre Kien Giang Dong Thap Tra Vinh Ca Mau Bac Lieu Soc Trang 0.5 0.0 13.0 13.2 13.4 13.6 13.8 14.0 14.2 Log of per capita spending Figure 3.8: NCCC Provinces against efficiency frontier 2.5 Hanoi 2.0 PSP scores 1.5 Da Nang Binh Dinh Nghe An Khanh Hoa Quang Ngai Ha Tinh Binh Thuan 1.0 Hau Giang Phu Yen Thanh Hoa Quang Tri Quang Nam Ninh Thuan Thua Thien Hue Quang Binh 0.5 0.0 13.0 13.2 13.4 13.6 13.8 14.0 14.2 Log of per capita spending Figure 3.9: Northern Mountain Provinces against efficiency frontier 24 2.5 Hanoi 2.0 PSP scores 1.5 Hau Giang Thai Nguyen Lang Son 1.0 Bac Giang Phu Tho Ha Giang Hoa Binh Son LaCao Bang Lai Chau Tuyen Quang Bac Kan Yen Bai Lao Cai Dien Bien 0.5 0.0 13.0 13.2 13.4 13.6 13.8 14.0 14.2 Log of per capita spending Figure 3.10: Red River Provinces against efficiency frontier 2.5 Hanoi 2.0 1.5 PSP scores Thai BinhNam Dinh Hai Duong Hai Phong Hau Giang 1.0 Ha Nam Vinh Phuc Quang Ninh Bac Ninh Hung Yen Ninh Binh 0.5 0.0 13.0 13.2 13.4 13.6 13.8 14.0 14.2 Log of per capita spending Figure 3.11: South East Provinces against efficiency frontier 2.5 Ho Chi Minh 2.0 Hanoi 1.5 PSP scores Dong Nai Hau Giang BRVT 1.0 Binh Phuoc An Giang Tay Ninh 0.5 0.0 13.0 13.2 13.4 13.6 13.8 14.0 14.2 Log of per capita spending Richer provinces tend On the latter, the study looked specifically at how to have a lower share of the composition of recurrent spending, in particular 25 recurrent spending going the share of salaries and wages in overall recurrent to wages and salaries. spending, interacts with productive efficiency across regions and provinces. Overall, provinces in wealthier regions not only have lower local spending as a share of the local economy, but also a lower share of recurrent spending going on wages and salaries – this may be because poorer provinces have to offer higher wages and salaries to attract staff, and not necessarily because the civil service is leaner in richer provinces. Within poorer regions, there are outliers in terms of high share of salaries and wages and low productive efficiency. It is important to review those cases in more detail to ensure that the wage bill is not crowding out critical operations and maintenance or goods and services spending, which would further exacerbate efficiency concerns. Issues Recommendations Policies on over-realized revenues in the SBL Ensure in the SBL and associated regulations, that 2002 and associated regulations lead to lack of the approval of the use of excess revenues (by the credibility of spending plans and loss of fiscal NA or provincial Councils) for spending takes place State Budget Law 2002 discipline. in the context of a Supplementary Budget. Off-budget debt financing is likely to increase over time, which would exacerbate problems Integrate all debt and other off-budget financed of low credibility spending plans without activities into local budgets to ensure transparency stricter controls in the context of overall budget and comprehensiveness. decisions. Carry over practices lead to loss of credibility Limit carry over spending in the SBL to only in spending plans, and lack of transparency as investment spending. Delete Article 63 allowing carry over spending is not broken down in the carry over of unspent revenue. budget. Introduce population density criteria in allocation Population density has a large and significant norms to estimate spending needs of local impact on the relative cost of service delivery authorities (i.e. inversely proportional). regulations Stability The current budget classification structure, Further develop the existing Citizens’ Budget to Period budget table templates, and coverage of fiscal communicate in a more accessible manner budget activities in budget documents. policies of local authorities. Productive efficiency differentials across regions Reaffirms earlier recommendations on addressing are often related to level of development and policies that lead to loss of spending plan credibility. Follow up analysis population density, but within regions there are It is also recommended to study some of the outliers, many of whom suffer from low credibility outlying provinces in more detail constraints to of spending plans and high carry overs. productive efficiency. Several provinces in poorer regions exhibit high Review wage bill and application of salary reform share of salary in recurrent spending and low policies in outlying provinces. productive efficiency. LOCAL REVENUE ARRANGEMENTS 26 04 Key issues: local authorities in Vietnam have seen increased levels of local revenue in the 2006-2011 period, but continue to have little to no autonomy over revenue policy and administration. Provinces where sharing rates declined in subsequent stability periods may have disincentives to maximize revenue effort, which points to a review of the revenue sharing arrangements. These provinces also highlight a deficit in infrastructure financing, which points to a review of potential for increased revenue autonomy. Objectives: the objectives of this chapter are to assess whether the current revenue arrangements are impacting on the revenue performance of local authorities, and how these could be addressed through reforms of the revenue sharing arrangement and the potential for more revenue autonomy in selected provinces. Key findings: compared to other countries, decentralized revenues in Vietnam constitute relatively large shares of national GDP (9-10 percent) and general government revenue (33 percent excluding extra budgetary sources). The share of decentralized revenue over total local revenue however has declined over time. This is partly because richer provinces have been making higher contributions of shared revenue to central government for the purposes of redistribution through transfers. Analysis in this chapter does not suggest that this has had a significant negative impact on revenue effort by these net contributing provinces. On average provinces in Vietnam collect roughly a little over 60 percent of their potential estimated revenues. But performance across provinces varies a lot due to factors such as local capacity and the level of revenue 27 decentralization as measured by share of 100 percent locally retained revenue. This, together with the needs of richer, high growth potential provinces, highlights the need to consider opportunities for more revenue autonomy. The revenue sharing arrangement also warrants review to improve its transparency and equity. In particular, major taxes end up credited to provinces where major firms are headquartered rather than where the output is produced or consumed. More equitable sharing arrangement should promote local economic performance and revenue performance across the provinces. Recommendations: (i) consider moving to sharing of VAT and CIT on a formula basis rather than on a derivation basis; (ii) centralize natural resource taxes; (iii) enable provinces to impose surtaxes on PIT, local businesses and excises; (iv) increase autonomy over determination of user fees at local level; (v) consider property tax in the long run. 28 LOCAL REVENUE ARRANGEMENTS Levels of local revenue Local authorities in Vietnam have seen increased levels have increased but of local revenue in the 2006-2011 period to match revenue autonomy has not. spending decentralization, but continue to have little to no autonomy over revenue policy and administration. Local authorities in Vietnam have two main sources of revenue outside of transfers from upper tiers of government: (i) revenue retained 100 percent by local authorities, which is the closest that local authorities come to “own source revenue;” and (ii) revenue that is shared with upper tiers of government, which are pooled and redistributed across the country. Compared to other countries, decentralized revenues Decentralized revenues in Vietnam constitute relatively large shares of national as a share of GDP are GDP (9-10 percent. General government revenue relatively large in Vietnam. to GDP is around 20 percent of GDP) and general government revenue (33 percent excluding extra budgetary sources). Decentralized revenue as a share of local GDP on the whole is relatively small – less than 7 percent for half of all observations from 2006 to 2011. The reason is that central revenues constitute the bulk of general government receipts. It therefore does not necessarily reflect lower revenue effort on the part of local authorities. The share of decentralized revenue over total local The share of decentralized revenue however has declined over time. This is partly revenue however has because richer provinces have been making higher fallen. contributions of shared revenue to central government for the purposes of redistribution through transfers. Transfers in turn have constituted a growing share of local financing. There is however good One concern about this trend is that provinces where evidence of strong 29 sharing rates declined in subsequent Stability Periods revenue effort by provinces. may have disincentives to develop their economies and to the extent they affect it, to maximize revenue effort. The marginal benefit of raising more revenue for these provinces gets diminished with the decreasing tax sharing rates, which get adjusted at the beginning of each stability period with the objective of channeling those funds to the central government for redistribution and other budget objectives. An analysis of revenue effort by local authorities does On average, provinces in Vietnam have relatively not suggest that this has had a significant negative strong revenue effort. impact on revenue effort by the net contributing provinces (figures 4.1-4.6). Performance across provinces however varies a lot due to factors such as local capacity and the level of revenue decentralization as measured by share of 100 percent locally retained revenue (figures 4.7-4.12). This, together with the needs of richer, high growth potential provinces, highlights the need to consider opportunities for more revenue autonomy. Local nominal GDP growth and nominal decentralized revenue growth by region 30 Figure 4.1: Central Highlands Figure 4.2: Mekong Delta 60% 35% Kon Tum Can Tho 50% R² = 0.0086 30% R² = 0.042 Ca Mau Decentralized Revenue growth Decentralized revenue growth Hau Giang 25% Vinh Long 40% Long An Dak Nong 20% Ben Tre Kien Giang 30% Gia Lai Dong Thap 15% Dak Lak Soc Trang 20% Lam Dong 10% Bac Lieu 10% 5% Tra Vinh Tien Giang 0% 0% 0% 5% 10% 15% 20% 25% 30% 35% 0% 10% 20% 30% 40% Local GDP growth Local GDP growth Figure 4.3: NCCC Figure 4.4: Northern Mountains 50% 70% 45% Quang Nam 60% Ha Giang R² = 0.2615 R² = 0.0304 40% Decentralized revenue growth Quang Ngai 35% 50% Decentralized revenue growth Hoa Binh 30% Khanh Hoa 40% Son La 25% Thai Nguyen Quang Tri Quang Binh 30% Bac Giang 20% Thanh Hoa Cao Bang Bac Kan Da Nang Binh Thuan Tuyen Quang 15% Lao Cai 20% Phu Tho 10% Dien Bien Lang Son Binh Dinh 10% Lai Chau 5% 0% 0% 0% 10% 20% 30% 40% 0% 10% 20% 30% 40% Local GDP growth Local GDP growth Figure 4.5: Red River Delta Figure 4.6: South East 45% 35% 40% R² = 0.0903 Binh Duong Ha Nam Ninh Binh R² = 0.2248 30% Decentralized revenue growth 35% Hai Duong Hung Yen Decentralized revenue growth 25% 30% Ho Chi Minh Bac Ninh 20% 25% Nam Dinh Thai Binh Tay Ninh 20% Binh Phuoc 15% Vinh Phuc Ha Noi Dong Nai 15% - Vung Ba Ria 10% Tau 10% 5% 5% 0% 0% 0% 10% 20% 30% 40% 50% -40% -20% 0% 20% 40% Local GDP growth Local GDP growth Relative efficiency of local authorities in mobilization of decentralized revenue 31 Figure 4.7: Central Highlands Figure 4.8: Mekong Delta 20% 12% Decentralized revenue/local GDP 18% Decentralized revenue/local GDP Kon Tum 10% 16% Dong Thap Long An Ca Mau 14% 8% Can Tho 12% Gia Lai Tien Giang Vinh Long 10% Lam Dong 6% Soc Trang Dak Lak Hau Giang Dak Nong Bac Lieu 8% Kien Giang Ben Tre 6% 4% Tra Vinh 4% 2% 2% 0% 0% 0.00 0.20 0.40 0.60 0.80 1.00 0.00 0.20 0.40 0.60 0.80 1.00 Efficiency score Efficiency score Figure 4.9: NCCC Figure 4.10: Northern Mountains 35% 18% Lai Chau 16% 30% Decentralized revenue/GDP Decentralized revenue/GDP Da Nang 14% 25% 12% Thai Nguyen Phu Tho 20% Bac Giang 10% Lao Cai Cao Bang 15% 8% Yen Bai Son La Bac Kan Ninh Thuan Ha Tinh Quang Ngai Hoa Binh Ha Giang Khanh Hoa 6% Tuyen Quang 10% Quang Tri Binh Thuan Dien Bien Quang Nam Binh Dinh Lang Son 4% Thanh Hoa 5% Quang Binh Phu Yen 2% 0% 0% 0.00 0.20 0.40 0.60 0.80 1.00 0.00 0.20 0.40 0.60 0.80 1.00 Efficiency score Efficiency score Figure 4.11: Red River Delta Figure 4.112: South East 25% 16% Binh Duong 14% Vinh Phuc 20% Binh Phuoc Decentralized revenue/GDP 12% Decentralized revenue/GDP Ha Noi 10% 15% Ninh Binh Dong Nai Hai Duong 8% Ho Chi Minh Hai Phong Tay Ninh 10% Bac Ninh Hung Yen 6% Ha Nam - Vung Ba Ria Thai Binh 4% 5% Nam Dinh Ta 2% 0% 0% 0.00 0.20 0.40 0.60 0.80 1.00 0.00 0.20 0.40 0.60 0.80 1.00 Efficiency score Efficiency score The revenue sharing Whilst the revenue sharing arrangement does not 32 arrangement warrants necessarily point to lower effort in collecting revenue on reform including the part of net contributing provinces, there are a number revamping the system of assigning revenue to of elements that warrant review. Shared revenues in where they are collected Vietnam are split based on where revenues are actually rather than where the tax collected rather than where the tax is incurred (the so- is incurred. called “derivation principle”). This raises questions concerning the fairness of the system, especially for the Value-Added Tax (VAT) and the Corporate Income Tax (CIT). For example if an enterprise is operating in Son La province, but is headquartered in Hanoi, the enterprise’s VAT and CIT liabilities will be owed to the province of Hanoi (unless it is a firm with unified accounting, in which case CIT will be paid to the central government). The major taxes therefore end up credited primarily to the few jurisdictions where the headquarters of the enterprise are located or the place of business registration, and not necessarily where the outputs are produced or sold. More equitable sharing arrangement should promote local economic performance and revenue performance across the provinces. There are several options to address this including sharing VAT and CIT on a formula basis. Formula for the former could be based on criteria such as provincial population, GDP, or level of consumption per capita, while that for the latter could be by payroll (labor), assets (capital), and sales (output). In addition, there should be consideration to centralize natural resource taxes to address disparities and avoid or correct negative economic externalities, while introducing production-related excises, which might be fully or partly assigned to local governments as part of compensation of environmental damage of the extraction process. Aside from this, however, increasing revenue autonomy is one of the most significant issues with the current local revenue arrangements in Vietnam. As noted in chapter 3, this is important for efficiency of spending, but also to enable richer provinces to fill their infrastructure financing deficit, as discussed in chapters 5 and 6. Providing local Increasing revenue authorities better access to some revenue handles at the autonomy is one of margin will strengthen fiscal responsibility, as it promotes 33 the most significant linkages between the financing of local spending policy issues with the current local revenue decisions and those that benefit from these expenditure. arrangements. Several There are several options that Vietnam can consider to options are proposed enhance local revenue autonomy. in the study including surcharges on existing This could entail for example providing local governments bases or more autonomy in determining fees and a closed list of selected taxes with discretion to set rates charges. within a band. Surcharges, or piggybacking, on central taxes may provide additional subnational revenue. This involves imposing a surcharge on tax bases defined by the central government. Surcharge tax rates are frequently subject to both upper and lower limits set by the central government. Surtaxes could be imposed on Personal Income Tax (which is small now but likely to grow) or on excises for example. More autonomy over revenue policy might also involve increasing independence over determination of user fees at subnational level. Local authorities would need to balance the revenue from fees against the cost of collecting those fees. It would also need to take into account the extent to which such fees and charges are regressive. Requiring local authorities to set the fee levels below the actual cost of service provision imposes an unfunded mandate, which can lead to poor service provision. Over the longer-term the government can consider the full introduction of modern property taxation. Local authorities have a comparative advantage in identifying and valuing properties because they are familiar with the housing and land available there. Given the complexities of a modern property tax system however, this may take a while to introduce. Issues Recommendations 34 Current flexibility over spending of over realized Eliminate provisions in the SBL 2002, which include State Budget Law 2002 revenue leads to negative incentives to issues such as revenue carry over and use of over- underestimate revenue in local budgets, which realized revenue. The latter needs to be mandated impacts negatively on overall credibility of the plan. by the legislature through appropriations. Consider two options: (i) moving to sharing VAT Revenue sharing on a “derivation basis” leads on a formula basis; and (ii) centralizing natural to concentration of major revenues in selected resources tax and CIT for redistribution through the provinces. transfer system. Increased revenue autonomy is needed for Consider introducing a system of surtaxes (on regulations better revenue effort and to help net contributing personal income, business profits, and excises) and Stability Period provinces meet their infrastructure financing ne more autonomy over fee setting. Over long-term eds. consider property tax. 35 INTERGOVERNMENTAL FISCAL TRANSFERS 36 05 Key issues: revenue sharing arrangements reviewed in chapter 4 are designed to address vertical imbalances arising out of the gap between local revenues and spending needs. Balancing transfers are meant to address horizontal imbalances across provinces. One issue that arises is the extent to which the transfer system is delivery more equitable resource distribution across and within provinces. Aside from revenue sharing and balancing transfers, there are also targeted transfers that face challenges on how to link national objectives with results on the ground. Objectives: the objectives of this chapter are to assess the extent to which balancing transfers are achieving their aim to equalize resource availability not just across provinces but also across districts, and recommend opportunities for reform. Another objective of this chapter is to identify ways to reform the system of targeted transfers to more effectively deliver on their intended targets. Key findings: revenue sharing arrangements and the system of balancing transfers in Vietnam have played a significant role in narrowing vertical and horizontal fiscal imbalances respectively. There are however growing vertical imbalances due to local authorities’ increased spending responsibilities. Trading off richer provinces’ right to retain a higher share of shared revenue for growth enhancing spending to allow more redistribution is welcome though also points to potentially greater need for revenue autonomy (chapter 4) and access to debt financing (chapter 6), particularly for some of the richer provinces. Increased spending responsibility of both provincial and district authorities are covered by increased balancing transfers. The latter at the center to provincial level are underpinned by a transparent, norms-based system. At province to district level, the publication of stability period rules and regulations 37 have also helped to improve transparency. The system of fiscal transfers has helped to promote greater equity in resource distribution both across and within provinces. There is evidence of fairly wide variation in per capita spending across districts. This may reflect the diversity in conditions of different districts, though it is difficult to say for sure what explains this remaining disparity. Target transfers continue to play an important role in local spending even though its relative share has fallen over time. The lack of predictability in targeted transfers poses serious challenges. Resourcing is not closely aligned with the targets and objectives, which are quite ambitious. Complex, input-based guidance reduces flexibility and increases burden of reporting. There is a real opportunity to review the institutional arrangements for NTPs for the coming stability period aiming to simplify the system and increasing transparency, monitoring and accountability Recommendations: (i) link stability period balancing transfers to inflation so do they do not decline in real terms; (ii) selected adjustments to allocation norms to determine spending needs; (iii) ensure greater predictability of targeted transfers to provinces and districts; (iv) simplify institutional structure for NTPS and link resourcing to performance. 38 INTERGOVERNMENTAL FISCAL TRANSFERS Vietnam has a relatively Vietnam has developed a relatively transparent, rules- transparent, rules-based based system of intergovernmental fiscal transfers. system of fiscal transfers. Separate formulae are used to determine recurrent and capital spending needs of provinces (and of districts). The formulae use population, geographic location (e.g. mountainous, rural/urban) and other criteria to take account of the specific circumstances, and therefore funding needs, of different provinces. This has helped develop a highly predictable system of balancing (or unconditional) transfers. Revenue sharing arrangements and the system of balancing transfers in Vietnam have played a significant role in narrowing vertical and horizontal fiscal imbalances respectively. A general trend both across and within provinces, however, is growing vertical imbalances due to local authorities’ increased spending responsibilities. In other words, 100 percent locally retained revenues are not keeping pace with increased spending decentralization. Revenue sharing arrangements are helping to cover the fiscal gap, but overall spending responsibilities are growing faster. This means that horizontal imbalances (i.e. decentralized revenue/local spending) have also grown. To cover horizontal imbalances, fiscal transfers have increased over subsequent Stability Periods. This has been facilitated by a selected number of richer provinces that are contributing a growing share of their shared revenues to the central government for redistribution through balancing transfers. Richer provinces are The study argues that trading off richer provinces’ right transferring higher to retain a higher share of shared revenue for growth 39 revenue to allow more enhancing spending to allow more redistribution for redistribution. greater equity seems like the correct policy choice in Vietnam. The evidence shows that increased spending responsibility of both provincial and district authorities are being covered by increased levels of balancing transfers. The latter at the center to provincial level are underpinned by a transparent, norms-based system. At province to district level, the publication of Stability Period rules and regulations by the provincial governments have also helped to improve transparency. The current system The current institutional arrangements for fiscal has promoted more transfers have helped to promote greater equity equity across and within in resource distribution both across but also within provinces, but there is wide variation in per provinces. The effectiveness of central government’s capita spending across redistribution policies is a matter of concern in districts. countries where intermediate provincial authorities are granted responsibility for determining transfers to lower orders of government. However, the evidence shows that balancing transfers are helping to equalize levels of resourcing across provinces and across districts. However, there is evidence a fairly wide variation in per capita spending across districts. This may be reflecting the diversity in conditions of different districts but there is no conclusive evidence on this. The principles used in determining spending needs through use of allocation norms based on population, school going population, geographic location and other, have contributed to the above results. The study suggests that these norms could be further refined. For example, as noted in chapter 3, the government could use population density as a means to estimate spending needs. Less densely populated provinces will have higher spending needs than more densely populated provinces with equal population. Targeted transfers should It may also be important to adjust allocation norms to take 40 become more predictable account of: (i) the level of service provision by different and shift towards a orders of government within the same jurisdiction (e.g. performance-based framework. in where provincial authorities are a major supplier of services within a district, allocation norms should be adjusted down compared to other districts where provincial engagement is less); (ii) levels of private sector service provision within a jurisdiction. It is also important to take account of possible negative externalities (e.g. congestion, pressure on local capacity) created by non- residents or temporary migrants using local services. In those cases, it makes more sense for the upper tier to provide compensating transfers than adjusting the norm. Targeted transfers continue to play an important role in local spending even though its relative share has fallen over time. The lack of predictability in targeted transfers poses serious challenges for both provincial and district authorities. It impedes on planning, budgeting, and delivery of NTP targets. The level of resourcing however is not closely aligned with the targets and objectives, which are quite ambitious. Complex, input- based guidance on NTP implementation reduces flexibility and increases burden of reporting. There is a real opportunity to review the institutional arrangements for NTPs for the coming Stability Period. However, moving to more output or performance- based transfers has to be a gradual process. It is not advisable to move the whole system to performance- based one. Performance-based systems have their own challenges in terms of incentives and the quality of reporting, which impact on their effectiveness. The government may therefore consider a sequenced approach in rolling out a performance-based transfer system. The new system could for example target first those provinces that have higher capacity, with poorer provinces still receiving transfers on the basis of socio- economic characteristics or inputs. As the new system gets tested and piloted, it could be gradually expanded and tailored (e.g. in terms of performance measures) to poorer provinces. In the meantime, for the latter, it is important to generate as much information as possible on what existing targeted transfers are delivering in 41 terms of outputs and development outcomes. This will provide important inputs in the design of grants that are tailored to the specific challenges different categories of provinces. Recommendations on fiscal transfers Issues Recommendations Law 2002 Allocation norms to estimate spending needs Amend Articles 31 and 33 in the SBL 2002 to Budget State and determine balancing transfers include clarify the exclusive responsibilities for central areas that should be exclusively the central government, thereby eliminating the need for norms government’s responsibility. in these areas. Maintaining balancing transfers for provinces Index balancing transfers to the rate of inflation in and districts constant over the Stability Period the formula for the next Stability Period. does not take account of rising costs. Allocation norms for central and provincial authorities do not adjust for the level of service Services for a particular function within a jurisdiction provision by different orders of government should be provided exclusively by one order of within the same jurisdiction, which could penalize government, even if higher orders of government areas where upper tiers are not providing any play a role in regulation or provision of subsidies Stability Period regulations 2002 services. Adjust norms down for provinces with significant Allocation norms do not adjust for different private sector provision. In case this leads to lower levels of private sector service provision within access, upper orders of government should address a jurisdiction through targeted transfers or social protection schemes. Allocation norms do not compensate for negative Compensated affected provinces and districts externalities (e.g. congestion, pressure on local through upper order transfers or direct provision of capacity) created by non-residents or temporary services, not through higher allocation norms. migrants using local services Use of allocation norms based on physical Ministry of Finance can issue guidance on the assets or the share of recurrent spending leads appropriate use of allocation norms to determine to distorted incentives spending estimates. Lack of predictability in targeted transfers puts Institute more discipline and predictability in target pressure on local budgets and creates unfunded transfers from center to province and province to mandates. districts. Different funding sources for NTPs add Explore greater use of country systems by transaction costs due to differing financial Development Partners supporting NTPs. management and procurement procedures. NTP targets are not adequately linked to level of resourcing, the system does not link Move to a system of output-based conditional National Assembly resources to performance, and high level targets grants, complemented by the existing socio- regulations make it difficult to meaningfully monitor NTP economic criteria used to determine eligibility. performance. Coordination across NTPs is difficult due to Review options to consolidate NTPs and other fragmented management and complex, input- target programs and focus on fewer output and based guidelines. outcome indicators. LOCAL BORROWING 42 06 Key issues: the system of intergovernmental fiscal relations in Vietnam, as illustrated in the previous two chapters, is geared to redistributing locally collected revenues, the bulk of which are contributed by a handful of provinces that point to a growing infrastructure financing deficit. Local borrowing has accordingly emerged as an important topic in Vietnam particularly for those provinces that are not able to satisfy their capital spending needs through the existing local revenue and transfer arrangements. Objectives: to assess whether selected provinces could potentially borrow more to address this deficit, and highlight reforms to the existing institutional framework to enable this. Key findings: borrowing by local authorities over the 2006-2011 period has remained very low (less than 3 percent of GDP). The five largest cities account for just over 40 percent of total local debt. Local authorities have several sources of debt financing including development banks (38 percent of total in 2011), state treasury (29 percent), local bonds (22 percent), central government on-lending (8 percent). The SBL 2002 requires local authorities to maintain total outstanding debt from these sources at below 30 percent of annual capital budgets, except for HCMC and Hanoi for whom the ceiling is 100 percent of annual capital budget (likely to be raised to 150 percent from 2016). At times several provinces have breached this ceiling, which in the first place is not a good indicator of local authorities’ actual borrowing capacity. An illustrative debt sustainability analysis of HCMC shows that under specific assumptions, the province could borrow more than the current limits prescribed in the SBL 2002 and maintain sustainable levels of debt. However, without greater access to decentralized revenue, the province would face liquidity pressures, thereby reiterating recommendations on revenue autonomy in chapter 5. 43 Aside from direct debt liabilities, local authorities also need to look more closely at indirect liabilities and contingent liabilities, potential sources of which include: public financial funds; local SOEs; and banking sector stress from payment arrears to local contractors. Recommendations: (i) restate fiscal rules related to local debt in the SBL 2002 and adopt more standard qualitative fiscal rules in the revised law; (ii) adopt specific thresholds on local debt stock and debt servicing in line with debt sustainability targets; (iii) improve transparency and reporting on public debt, including through accounting for subnational debt in the local budget balance and through producing statistical debt bulletins in more advanced provinces; (iv) transition to a two-tier local debt system with central government facilitating more market access for advanced provinces and greater ODA on-lending for less advanced provinces. 44 LOCAL BORROWING Some provinces are not The system of intergovernmental fiscal relations in able to satisfy capital Vietnam, as illustrated in the previous chapters 4 and spending needs through existing revenue and 5, is geared to redistributing locally collected revenues, transfers. the bulk of which are contributed by a handful of provinces, which at the same time increasingly point to a growing infrastructure financing deficit. Local borrowing has accordingly emerged as an important topic in Vietnam particularly for those provinces that are not able to satisfy their capital spending needs through the existing local revenue and transfer arrangements. Borrowing by local authorities over the 2006-2011 Local borrowing has period has remained very low (less than 3 percent remained low in the 2006- of GDP). The five largest cities account for just 2011 period. over 40 percent of total local debt. Local authorities have several sources of debt financing including development banks (38 percent of total in 2011), State Treasury (29 percent), local bonds (22 percent), and central government on-lending (8 percent). Figure 6.1: Local government debt 2006-2012 40,000 3.0% 35,000 2.77% 2.5% 30,000 Disbursement Repayment 2.0% 25,000 2.17% Outstanding debt Outstanding debt in % of GDP 20,000 1.78% 1.5% 1.62% 15,000 1.47% 1.0% 1.18% 1.22% 10,000 0.5% 5,000 0 0.0% 2006 2007 2008 2009 2010 2011 2012 Source: Ministry of Finance Local borrowing The State Budget Law 2002 (SBL 2002) and the Public should be part of local Debt Management Law 2009 (PDML 2009) stipulate the 45 authorities’ balance golden rule that provincial governments cannot borrow sheets to meet recurrent expenditures. The SBL also requires local authorities to maintain balanced budgets, which means that any borrowing is treated off the balance sheet. Even though aggregate local borrowing levels have been relatively low to date, it is critical to integrate local borrowing on local authorities’ balances sheets to ensure accurate fiscal reporting. The SBL 2002 requires local authorities to maintain total outstanding debt from these sources at below 30 percent of annual capital budgets, except for HCMC and Hanoi for whom the ceiling is 100 percent of annual capital budget. At times several provinces have breached this ceiling, which in the first place is not a good indicator of local authorities’ actual borrowing capacity. An illustrative debt sustainability analysis of HCMC An illustrative debt shows that under specific assumptions, the province sustainability analysis of could borrow more than the current limits prescribed in HCMC shows that under the SBL 2002 and maintain sustainable levels of debt. specific assumption, the As illustrated in figure 6.2 below, the anticipated debt province could potentially stock for HCMC remains affordable when comparing to borrow more than the current limits in the SBL local GDP as the indicator for repayment capacity. 2002, but that it could run into liquidity pressures However, without greater access to decentralized without greater revenue revenue, the province would face liquidity pressures. autonomy. This is illustrated in figure 6.3, which shows that the debt service to revenue ratio could rise higher than 25 percent in some years. Therefore if HCMC wants to borrow more, it also needs more access to local revenues, which reiterates recommendations on revenue autonomy in chapter 5. It is also important to note that higher borrowing could also lead to interest payments crowding out other important spending, as these payments are likely to rise significantly. Figure 6.2: HCMC Debt-to GDP ratio Figure 6.3: HCMC Debt service-to-revenue ratio 46 30.0% 35.0% 25.0% 30.0% 25.0% 20.0% 20.0% 15.0% 15.0% 10.0% 10.0% 5.0% 5.0% 0.0% 0.0% 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 HCMC Debt-to GDP ratio Threshold HCMC Debt service-to-revenue ratio Threshold Figure 6.4: HCMC Interest payment/Recurrent Figure 6.5: HCMC Overall and Primary Balance Expenditure (% of GDP) 8% 0.5% 7% 0.0% 2014 2015 2016 2017 2019 2020 2021 2022 2023 2024 2018 2025 6% -0.5% 5% -1.0% 4% -1.5% 3% -2.0% 2% -2.5% 1% 0% -3.0% 2014 2015 2016 2017 2018 2019 2020 2021 2022 2024 2025 2023 -3.5% HCMC Interest Payment/Recurrent Expenditure Balance (% of GDP) Primary Balance (% of GDP) Source: Staff estimates This analysis points to The above analysis points to the need for reviewing sub- the need for reviewing national fiscal rules in Vietnam. There are generally three Vietnam’s current fiscal rules for local borrowing. types of indicators for controlling local government debt including limits on: (i) debt service, (ii) new borrowing in a fiscal year and (iii) on total outstanding debt. Borrowing limits might be regulated by combined indicators. So local governments should meet multiple criteria, e.g. limits on both debt service and debt stock. The State Budget Law might set out general principles or indicators against which to monitor local debt (e.g. debt service/ local revenues, outstanding debt stock/local budget). But quantitative thresholds should be set separately in secondary regulations. Aside from direct Aside from direct debt liabilities, local authorities liabilities, local authorities also need to look more closely at indirect liabilities 47 also need to watch indirect and contingent and contingent liabilities. Cross country experience liabilities. suggests that explicit contingent liabilities, especially loan guarantees, may be the most frequent type of indirect liability but implicit ones are often the most costly. The latter include bailouts of SOEs, banks and local governments. It also includes natural disasters contingent liabilities, which can significantly add to governments’ balance sheet risk in as much as they imply additional leveraging and tend to be triggered in times of financial stress, with realized costs having a major impact on a country’s fiscal position and debt sustainability Recent developments Recent developments in Vietnam, particularly in the suggest greater banking sector and SOEs underscore the importance monitoring of Public of monitoring contingent liabilities. Rapid credit Financial Funds, local growth in 2008-2009 with insufficient due diligence, SOEs, and banking sector coupled with the effects of the global economic crisis, stress. has led a build-up of Non-Performing Loans and a rapid deterioration in the balance sheets of domestic banks in Vietnam. The crisis also brought to the fore the vulnerability of SOEs and the lack of transparency around their operations and finances. MOF/MPI have estimated total SOE debt as at September 2011 to be closer to US$20 billion, or roughly 15 percent of GDP. More specifically for local authorities in Vietnam, there are potentially three important sources of contingent liabilities: (i) Public Financial Funds; (ii) local SOEs; and (iii) banking sector stress as a result of payment arrears to capital construction projects. Local authorities The successful implementation of any rules and 48 should report more management of fiscal risks requires transparency and systematically on its debt and other contingent reporting arrangements. Although local authorities liability obligations. are required to follow a large number of steps and regulations to contract debt, there is little formal requirement on reporting and disclosure of local debt. Increased transparency is important not only for the government’s internal management purpose but also for private investors to take timely and adequate decisions. For the more advanced provinces, the need for transparency is particularly important. These are provinces like HCMC that plan to go increasingly to debt markets for infrastructure financing. Secondly, SOEs in these provinces are also turning increasingly to the market on account of falling budgetary subsidies. The requirement for greater transparency and disclosure should apply across all provinces, but in terms of sequencing, the more advanced provinces should already start preparing debt statistical bulletin covering its domestic and external (on-lent) debt. Overall, subnational borrowing should be accounted for in the local budget balance. The government may consider following a two-tier The government may wish strategy in relation to local borrowing. The central to pursue to a two-tier government can target the more advanced provinces strategy to strengthening to strengthen the enabling environment for better local debt management capacity. access to debt markets. This will require appropriate capacity building support for fiscal planning and debt management, in addition to greater transparency and disclosure as mentioned above. For less advanced provinces, the central government could start with greater ODA on-lending to begin with. As capacity improves, these provinces can also gradually progress to increased access to debt markets. However, this would need to be done in a careful and sequenced manner to prevent build up of unsustainable debt as experienced in other countries. Recommendations on local debt 49 Issues Recommendations Conflicting rules in SBL 2002, requiring provinces Revising rules in SBL 2002 to require current to maintain balanced budget while allowing them account balance only, and account subnational to borrow for capital spending. budget deficit in the overall national budget deficit. State Budget Law 2002 Revamping the thresholds to better monitor both debt service (e.g. debt service/local revenues) Existing quantitative ceiling of subnational debt and debt stock (e.g. outstanding debt stock/local does not reflect provinces’ debt sustainability, budget). Considering balancing the operational solvency or repayment capacity. and capital budgets in the longer-run. Quantitative thresholds should be set separately in regulations and decrees. Including borrowings for cost-recovery projects and on-lending in the total subnational debt amount. Concerns over “off-budget” borrowings Monitoring closely borrowings of SOEs and Local Development Investment Funds. Including subnational debt data in the national Public Debt Bulletin. Considering interactive way Strategy for strengthening debt management to collect subnational debt information from LGs. Missing subnational debt data in the national Encouraging frontier provinces to develop debt Public Debt Bulletin. No provincial public debt statistical bulletin covering both domestic and bulletins despite interest of potential investors. external (on-lent) debt and other fiscal information that is of interest to credit rating agencies and potential investors. At the central level, developing comprehensive capacity methods for local borrowing capacity assessment, Limited capacity in public debt management at and strengthening MOF capacity. At the local level, both central and local governments. strengthening DOFs capacity to develop medium- term fiscal plan and public debt management program, and closely monitor contingent liabilities. Central government facilitating appropriate “enabling environment” to ensure smooth supply Evolving fiscal landscape and debt profile at both of loanable funds for the frontier provinces, while central and local levels. promoting ODA on-lending in net budget recipient provinces. The World Bank in Vietnam 63 Ly Thai To Street, Hanoi Tel: (84-4) 3934 6600 Fax: (84-4) 3935 0752 Website: www.worldbank.org.vn