Report No. 61694-KH Cambodia More Efficient Government Spending for Strong and Inclusive Growth Integrated Fiduciary Assessment and Public Expenditure Review (IFAPER) November 30, 2011 Poverty Reduction and Economic Management Unit East Asia and Pacific Region Document of the World Bank GOVERNMENT FISCAL YEAR January 1 – December 31 CURRENCY EQUIVALENTS (Exchange Rate Effective as of May 25, 2011) Currency Unit: Cambodia Riel (CR) U.S. $1.00 = CR 4078 WEIGHTS AND MEASURES Metric System Vice President: James W. Adams Country Director: Annette Dixon Chief Economist and Sector Director: Vikram Nehru Country Manager: Qimiao Fan Lead Economist: Mathew A. Verghis Task Team Leader: Ivailo Izvorski/Leah April ABBREVIATIONS AND ACRONYMS AOPs Annual Operational Plans NER Net Enrollment Rate ASSDP Agricultural Sector Strategic Development Plan NGOs Non-Government Organizations BCR Benefit-Cost Ratio NIS Newly Industrialized States BSP Budget Strategic Plans NPF Net Present Value CAR Council for Administrative Reform NSDP National Strategic Development Plan CDC Council for the Development of Cambodia O&M Operations and Maintenance CESSP Cambodia Education Sector Support Project ODA Official Development Assistance CMDG Cambodia Millennium Development Goals OECD Organization for Economic Cooperation and Development CMS Central Medical Store OOP Out of Pocket COA Chart of Accounts PAP Priority Action Plan CSF Commune Sangkat Fund PAR Public Administrative Reform D&D Decentralization and Deconcentration PASEC Program for the Analysis of Education Systems of the Confemen (Francophone Africa) DFID Department for International Development PB Program Budgeting ECE Early Childhood Education PEFA Public Expenditure and Financial Accountability EDB Essential Drugs Bureau PETS Public Expenditure Tracking Survey EMIS Educational Management Information System PFM Public Financial Management ESP Education Strategic Plan PFMAP Public Financial Management and Accountability Project EFA FTI Education for All Fast Track Initiative PFMRP Public Financial Management Reform Program FMIS Financial Information Management System PHDO Provincial Health Department Office FWUCs Farmer Water Users Community PISA Program of International Student Assessment GDP Gross Domestic Product PMU Project Implementation Unit GFATM Global Fund for Aids, Malaria and Tuberculosis POC Priority Operating Cost HEFs Health Equity Funds PPP Public Private Partnerships HSP2 Second Health Strategic Plan PREAC Prequalification Evaluation and Award Committee IFAPER Integrated Fiduciary Assessment and Public QIP Qualified Investment Projects Expenditure Review IMF International Monetary Fund SAW Strategy for Agriculture and Water MAFF Ministry of Agriculture, Fisheries and Forestry SDGs Service Delivery Grants MEF Ministry of Economy and Finance SIDA Swedish International Development Cooperation Agency MDG Millennium Development Goals SNFL Sub-National Finance Law MOEYS Ministry of Education, Youth and Sports SOAs Special Operating Agencies MOH Ministry of Health SOEs State-Owned Enterprises MOWRM Ministry of Water Resources and Meteorology TIMSS Trends in International Mathematics and Science Study MRD Ministry of Rural Development UN United Nations MTEF Medium-Term Expenditure Framework VAT Value Added Tax NCD Non-communicable Disease WHO World Health Organization ACKNOWLEDGEMENTS The Integrated Fiduciary Assessment and Public Expenditure Review (IFAPER) was prepared by a team led by Ivailo Izvorski that included Huot Chea, Sodeth Ly, Paavo Eliste, Timothy Johnston, Emre Ozaltin, Suhas Parandekar, and Leah April. Generous financial support from AusAID, DFID, and SIDA is gratefully acknowledged. On the government side, we benefitted greatly from frequent consultations with H.E. Dr. Hang Chuon Naron, Dr. Sok Saravuth, Dr. Youk Bunna, Mr. Bun Roith, Mr. Kim Phalla, and Mr. Pen Thirong (MEF); H.E. Dr. Rith Vuthy (CDC); H.E. Mr. Im Sethy and Mr. Proum Pheap (MoEYS); H.E. Mr. Chan Tong Yves and Mr. Mong Leng (MAFF); H.E. Mr. Hiv Sarim (MRD); H.E. Mr. Veng Sakhon (MoWRM); H.E. Prof. Koeut Meach (MoH) and H.E. Dr. Youk Sambath (MoH); H.E. Mr. Ngo Hongly (CAR); Dr. Kun Nhem, Mr. Pha Eng Veng, and Kongsok Phallakun (Customs). We are grateful to the staff of the Departments of Budget and Finance and of Planning in the MAFF, MRD, MoWRM, MoH and MoE, to the Heads of the Departments of Extension and Crop Production in MAFF, to the Head of the Department of Rural Roads in MRD, to the Departments of Investment Cooperation, Budget and of Economic and Public Finance Policy, to the PFMRP Secretariat in MEF, and to the staff of CDC. Valuable comments were received at several rounds of presentations of the individual chapters to government officials, development partners, Bank staff, and for the agriculture chapter also students at the Royal University of Agriculture. We received valuable suggestions, input and guidance from Qimiao Fan, Stephane Guimbert, Robert Taliercio, Mathew Verghis, Luc Christiaensen, Gary Reid, Patrick Labaste, Peter Murphy, and Jos Verbeek, Simeth Beng, Tsuyoshi Fukao, Frederico Gil Sander, Patrick Grady, Ronald Points, Ahsan Ali, Christopher Robert Fabling, Giovanni Ganelli, Joel Turkewitz, and Sudhir Chitale. Ray Purcell and his colleagues from Mokoro provided a background paper on agriculture. Timothy Purcell (consultant, unrelated to Ray Purcell) also contributed to the agriculture chapter. Diane Northway and Daniel Whitaker prepared the initial draft of the health chapter. Linna Ky, Sophorn Kith, and Narya Ou provided excellent logistical and administrative support. Shane Perkinson, Lynn Gross, and Narya Ou helped format parts of the document. Galina Andronova-Vincelette, Sajitha Bashir, Ron Hood, and Sergiy Zorya were the peer reviewers. Table of Contents Table of Contents .......................................................................................................................................... 4 Overview ........................................................................................................................................................ i Agriculture ................................................................................................................................................ ii Health ....................................................................................................................................................... iii Education ................................................................................................................................................. iii Conclusions and pricing of recommendations ......................................................................................... iv Chapter 1. The Strategic Setting: Restoring and Expanding Fiscal Space ............................................. 1 A. Introduction ....................................................................................................................................... 1 B. Recent Economic Developments and Challenges ............................................................................. 2 C. The Fiscal Setting ............................................................................................................................. 6 D. Government Debt and Debt Sustainability ....................................................................................... 9 E. Fiscal Space .................................................................................................................................... 10 F. Conclusion and Policy Recommendations ...................................................................................... 13 Chapter 2. Composition of Government Spending, Priorities, and Opportunities ............................... 15 A. Introduction ..................................................................................................................................... 15 The Level of Public Expenditures........................................................................................................... 16 The Composition of Government Expenditures ..................................................................................... 18 Private Spending on Public Goods.......................................................................................................... 24 Conclusion and Policy Recommendations .............................................................................................. 26 Chapter 3. Government Spending on Agriculture ................................................................................ 27 A. Introduction ..................................................................................................................................... 27 B. Overview of Agriculture, Irrigation, and Rural Roads ................................................................... 27 C. The Level and Composition of Government Spending on Agriculture .......................................... 28 D. Budget Processes and Performance ................................................................................................ 31 E. Efficiency and Effectiveness of Government Spending in Agriculture .......................................... 32 F. Conclusion and Policy Recommendations ...................................................................................... 35 Chapter 4. More Health for the Money ................................................................................................ 38 A. Introduction ..................................................................................................................................... 38 The Health Sector in Cambodia .............................................................................................................. 38 Trends and Composition of Health Expenditures ................................................................................... 41 The Efficiency and Equity of Health Expenditures in Cambodia ........................................................... 45 Conclusion and Policy Recommendations .............................................................................................. 50 Chapter 5. Structure and Quality of Education Expenditures ............................................................... 52 A. Introduction ..................................................................................................................................... 52 Achievements and Challenges of the Education Strategic Plan.............................................................. 53 Education Expenditures: Patterns and Trends......................................................................................... 56 Teachers and Teacher Wages .................................................................................................................. 58 Provision of Textbooks ........................................................................................................................... 59 Household Spending on Basic Education ............................................................................................... 60 Public-Private Partnership in Higher Education ..................................................................................... 61 Conclusion and Policy Recommendations .............................................................................................. 61 Chapter 6. Sustaining Budget Credibility and Moving Toward Greater Financial Accountability ...... 63 A. Introduction ..................................................................................................................................... 63 B. The State of Public Financial Management .................................................................................... 64 C. Deepening Platform One Reforms, Advancing Platform Two Activities ....................................... 69 D. FMIS Implementation: Challenges and Risks................................................................................. 73 E. Public Administration Reforms and Decentralization and Deconcentration .................................. 74 F. Conclusion and Policy Recommendations ...................................................................................... 76 Annexes ...................................................................................................................................................... 78 References ................................................................................................................................................... 91 FIGURES Figure 1. Output is growing anew after a crisis-induced contraction in 2009 .............................................. 3 Figure 2. The fiscal deficit remains large after widening in 2009 ................................................................ 3 Figure 3. Growth and inflation have been high since the mid-1990s ........................................................... 5 Figure 4. Real exchange rate is depreciating ................................................................................................ 5 Figure 5. Revenues excluding grants are lower in Cambodia than in other low-income countries .............. 8 Figure 6. The value of exemptions from paying taxes remains large ........................................................... 8 Figure 7. General government spending is in line with that in countries with similar per-capita income .. 17 Figure 8. Wages are rising but outlays on maintenance and operations are trending down ....................... 19 Figure 9. Wage outlays rose sharply in 2009, largely due to increases for defense and security staff ....... 19 Figure 10. The number of public employees other than in defense and security is small … ..................... 19 Figure 11. … as is the civilian wage bill..................................................................................................... 19 Figure 12. The share of wages for education has declined since 2007 ....................................................... 20 Figure 13. Capital spending is larger than in countries with similar levels of income per capita............... 20 Figure 14. Government education outlays are low compared with other countries .................................... 23 Figure 15. Schools, and rural roads account for half of private spending on public goods ........................ 24 Figure 16. Political parties provide the bulk of contributions ..................................................................... 25 Figure 17. Overall government spending on agriculture has increased … ................................................. 30 Figure 18. … while donor support has declined ......................................................................................... 30 Figure 19. Recurrent spending has grown steadily in all three ministries .................................................. 30 Figure 20. Government capital spending has increased substantially......................................................... 30 Figure 21. The key outputs achieved have been broadly on an upward path ............................................. 32 Figure 22. But the record on costs has been mixed..................................................................................... 32 Figure 23. BCRs for agriculture and rural roads are rising … .................................................................... 33 Figure 24. … but they are still below potential........................................................................................... 33 Figure 25. Infant and child mortality are down sharply, neonatal only modestly....................................... 40 Figure 26. High rates of malnutrition persist .............................................................................................. 40 Figure 27. Facilities deliveries have increased significantly, with slower progress on contraception ........ 40 Figure 28. Private providers are most frequently consulted for curative care ............................................ 40 Figure 29. Government spending has increased sharply in nominal terms … ............................................ 42 Figure 30. … and as share of GDP and overall government spending ....................................................... 42 Figure 31. Spending on drugs and medical supplies is more than double wage outlays ............................ 42 Figure 32. Most budget is non-programmatic ............................................................................................. 42 Figure 33. Better-off households spend more on healthcare, 2007............................................................. 43 Figure 34. User fees are an important source of funding, 2009 .................................................................. 43 Figure 35. User fees generate substantial incomes ..................................................................................... 44 Figure 36. The source of funding for national hospitals varies, 2008-09 ................................................... 44 Figure 37. Spending on primary care is more pro-poor than hospital spending ......................................... 46 Figure 38. Poor people have closer access to health centers than hospitals................................................ 46 Figure 39. Budget allocations are linked to beds but not to hospital outputs ............................................. 47 Figure 40. Provincial allocations vary ........................................................................................................ 47 Figure 41. The distribution of doctors and primary midwives varies considerably .................................... 48 Figure 42. Stockouts persist despite rapid increases in spending for pharmaceuticals ............................... 48 Figure 43. The cognitive ability of very young Cambodian children has declined .................................... 54 Figure 44. Predicted probability of school attendance and enrollment as a function relative-ranking ....... 54 Figure 45. Government education spending fell relative to GDP, 2003-2009 ............................................ 56 Figure 46. Payment arrears quickly dropped and fully eliminated in 2007 … .......................................... 65 Figure 47. …while end-quarter disbursements improved quickly ............................................................. 65 Figure 48. Priority spending steadily improved together with high growth of investment spending … ... 66 Figure 49. …and overall annual disbursements by expenditure profiles rose markedly. ........................... 66 Figure 50. Economic downturn resulted in a contraction of revenues in 2009 … ..................................... 69 Figure 51. …while direct tax growth rate remained volatile ...................................................................... 69 Figure 52. Priority sector spending share of total domestic budget is declining. ...................................... 70 Figure 53. TSA’s cash balances have been decreasing since mid-2009 ..................................................... 70 TABLES Table 1. Cambodia Key Macroeconomic Indicators..................................................................................... 4 Table 2. Cambodia: General Government Fiscal Accounts, 2000-2010....................................................... 7 Table 3. Structure of Government Revenues, 2000-2009 ............................................................................. 9 Table 4. General government spending surged in 2009 after rising modestly during 2005-08 .................. 16 Table 5. Overall public spending in Cambodia is large .............................................................................. 17 Table 6. The average wage for education has risen but more slowly than the average civil service wage 21 Table 7. Spending across most functional categories rose .......................................................................... 22 Table 8. Foreign aid finances more than half of government spending in transportation and health ......... 23 Table 9. Transfers from the central government to sub-national levels are small ...................................... 24 Table 10. Communes’ priorities for development are yet to be aligned with domestic donor support ...... 25 Table 11. The quality of public outcomes is similar to that in countries with similar per-capita income .. 26 Table 12. Growth in agriculture has been strong ........................................................................................ 28 Table 13. Strong growth in agriculture, modest government spending ...................................................... 29 Table 14. Cambodia is on track to meet many health MDGs ..................................................................... 39 Table 15. Access to primary and higher education services ....................................................................... 53 Table 16. Improving the Quality and Efficiency of Education Services .................................................... 55 Table 17. Institutional and Capacity Development for Educational Staff for Decentralization ................. 56 Table 18. Cambodia’s education expenditures are lower than in relevant comparators ............................. 57 Table 19. Education Expenditures, 2009-13 ............................................................................................... 58 Table 20. Teacher profiles and teacher wage bill, 2001/02 to 2009/10 ...................................................... 59 Table 21. Household education spending varies by quintiles, 2009 ........................................................... 61 Table 22. Total higher education expenditure per source of funding ......................................................... 61 Table 23. The 3 + 2 proposal: key catalysts for financial accountability.................................................... 74 BOXES Box 1. Scope, Definitions, and Calculation of the Expenditure Data used in this Report .......................... 15 Box 2. Recording Expenditures of Donor-Funded Projects and /Programs ............................................... 22 Box 3. What does an Integrated Financial Information System (IFMIS) do? ............................................ 67 Box 4. The new classification encourages the use of government public financial systems ...................... 68 Box 5. The World Bank Experience with Treasury and Financial Management Information Systems ..... 71 Box 6. Incentive Measures for Implementing the Reform Program .......................................................... 72 ANNEX DISPLAYS Annex Display 1. NSDP 2006-2010 targets and the Cambodia Millennium Development Goals ............. 78 Annex Display 2. General government financing by functional classification........................................... 80 Annex Display 3. General government by functional classification: treasury executed component .......... 80 Annex Display 4. General government by functional classification: externally financed component ....... 81 Annex Display 5. Externally-financed technical assistance and NGOs’ own resources ............................ 81 Annex Display 6. NSDP resource requirements and actual allocations ..................................................... 82 Annex Display 7. MOH program budget allocation, 2009 ......................................................................... 82 Annex Display 8. Government budget allocations for health, 2000-2009 .................................................. 83 Annex Display 9. Employment at the Ministry of Health ......................................................................... 83 Annex Display 10. CMS invoice prices for ciprofloxacine are higher than international prices................ 84 Annex Display 11. Prices for ciprofloxacine are lower at private pharmacies than abroad ....................... 84 Annex Display 12. Performance indicators, 2004-2010 ............................................................................. 85 Annex Display 13. High level traffic light performance Indicator for the Platform Two .......................... 89 Annex Display 14. Regional comparison of PEFA scores ......................................................................... 90 Overview Cambodia has made substantial progress in economic reconstruction since the 1993 elections helped return peace to the war ravaged country. Economic growth averaged 7 percent a year subsequently for two decades, bringing per-capita incomes to about $760 in 2010, the global economic crisis notwithstanding. There has been progress in developing the institutions of effective state. The level and effectiveness of government spending has steadily increased, and donor support has remained large. Attention now has to turn to rapidly reducing the constraints to the government’s ability to deliver fair, effective and cost-efficient public services that support sustainable and inclusive growth. Stronger institutions, enhanced transparency, and better governance and accountability will be the critical components to ensure the goals of higher living standards without social exclusion are met. Fiscal prudence and an increase in revenues because of strong growth and better collection helped narrow the fiscal deficit sharply in the decade up to 2008. At the same time, the government increased its deposits with the central bank by more than twenty-fold to $732 million between 2003 and 2008. This large cushion – especially given the lack of government access to capital markets – enabled the authorities to boost spending in 2009 to an all-time high, helping mitigate the impact of the global economic crisis. Cambodia’s level of government spending is similar to that in other countries with comparable levels of income per capita and so are its development outcomes. There is also little difference in the composition of Cambodia’s spending relative to comparable countries. Nonetheless, there are two important issues. First, capital expenditures account for a substantial share of budget outlays, an appropriate share for a country with substantial development needs. Ensuring that these outlays are matched by realistic allocations for maintenance is crucial for ensuring that those capital investments are fully realized and support growth. Second, a hike in wages for the military in 2009 boosted wage outlays substantially, much more than increases in the salaries for non-military public employees since 2007. There may be scope for further wage raises in some (non-military) areas, provided that the fiscal space becomes available and that pay scales are decompressed under the framework of an effective pay reform. The composition of government spending is broadly similar with countries with comparable per- capita GDP. Still, Cambodia is lagging on three of the nine Millennium Development Goals. Within the country’s limited resource envelope, there should be room to reallocate expenditures to meet the MDGs and support stronger and inclusive growth. The recommendations in this report suggest specific steps to achieve substantial efficiency gains in agriculture, health and education that will enable directing more funds to priority areas without increasing the overall level of spending to GDP. Donor support remains substantial at about one-third of general government outlays or about 40 percent of overall public spending when off-budget donor support for technical cooperation and NGOs working on public goods is counted. This reliance on donors implies that the government and donors need to work harder to ensure better aligning strategic planning, budget preparation, and aid coordination. Progress in improving public financial management and financial accountability has been solid, but has yet to match the government ambitions and targets. Budget credibility is established at a robust pace. Reforms that devolve greater budget responsibility to program managers supported by enhanced internal controls and accountability mechanisms are currently in progress. The management structure of the Ministry of Economy and Finance (MEF) needs to be strengthened to implement successfully the Financial Management Information System (FMIS). And focus needs to shift to delivering on all the i public financial reforms targeted. This has to be supplemented by determined efforts to align government strategic goals, budget priorities, and donor support. At present, donor support is insufficiently coordinated and, as a result, is of more limited efficiency than intended. Firstly, collaboration among donors must increase to help ensure budget credibility. At a minimum, the discussion on producing expenditure reports for donor-financed projects using the government chart of accounts (COA) should be resolved. As a matter of priority, all donor support to the government, whether for current spending, capital expenditures, or for technical assistance, needs to be brought on budget and properly classified. Classifying all donor projects as capital spending as done at present does not allow the government to have a clear picture of all public outlays and, more importantly, to plan properly the adequate maintenance needed for the large capital spending carried out today. Presenting a comprehensive and properly classified budget will require government leadership and full donor cooperation. This will help strengthen accountability, facilitate planning, align budgetary spending with strategic goals, and contribute to assuring citizens that their money is well spent. Better integrating cross–cutting public administration, decentralization and public financial management reforms remains an important challenge. The allocation of responsibility for service delivery and resource management is linked to staffing and how the whole civil service is organized. Delays in comprehensive civil service and institutional reform directly influence progress in public financial management and put at risk the sustainability of many hard-earned gains. Assigning and devolving functions will align better responsibilities and sources of funding and will help improve the efficiency and effectiveness of service delivery. The functioning of Cambodia’s processes for managing and spending public money needs to be set in the broader context of the manner in which the overall public sector operates. Cambodia has made major progress in developing the foundations for an effective state after decades of conflict. At the same time, a host of issues constrain the government’s ability to deliver core services and all available studies point to persistent high levels of corruption. Establishing more effective, transparent and rule- based governance remains at the heart of Cambodia’s development challenge (and at the center of the World Bank’s support for the country’s future). It is important to appreciate that the analysis and the recommendations contained in this report should be read as a compliment to the on-going work to enhance transparency, accountability, and participation and that enhancing the value derived from public finances depends on progress in both areas. The issues of improving the efficiency and equity of government spending while strengthening public financial management and improving accountability and governance are reflected in the report’s analysis of three sectors. Agriculture, health and education were selected in extensive consultations with the authorities. A brief summary of the key issues in each of these sectors follows. Agriculture Agriculture is a government priority as one of the core sources of economic growth and export earnings. Although from a low base, government spending on agriculture has risen from 1.3 percent of GDP in 2007 to 1.5 percent in 2009, mainly through significant increases in outlays on irrigation and rural roads.1 This period of rising public spending on agriculture overlapped with relatively high annual growth rates of agricultural output. Although insufficient time has elapsed to assess with confidence whether there is a causal relationship between the level of public spending and agricultural growth in Cambodia, available evidence suggests that public expenditures – along with improvements in the policy 1 In this report, spending on agriculture includes outlays on agriculture, rural roads, and water resources. ii environment – may indeed have had a significant impact in boosting and sustaining agricultural growth. The analysis confirms that government spending has rightly focused on the provision of public goods and services, which have high rates of returns, rather than on subsidies. There is, however, a good potential for further efficiency gains on resources spent. Although Cambodia has no policy-induced distortions in agriculture, a complex set of structural constraints have hampered the effectiveness and efficiency of public expenditures. These include a lack of continuity in public funding, especially for O&M costs, for extension, and agricultural research. Deeper institutional issues that could limit the long-term effectiveness of these expenditures also need to be addressed. Other challenges include the excessive focus on rehabilitating primary irrigation infrastructure and a neglect of secondary and tertiary systems, a neglect of maintenance of irrigation and rural roads, and the slow pace of development of new technologies for rural roads. There is a need to better prioritize expenditures to increase the effectiveness and efficiency of public spending for the agriculture, health and education three sectors. The government could get better value for money by reallocating spending within the existing envelope on agriculture. In the short term, more resources for agricultural research, extension and secondary irrigation canals and smaller allocations for strategy and planning are one of the recommendations for the authorities to consider. Researching mitigation and adaptation measures will help prepare Cambodia for climate change. An extension of funds within the existing budget envelope – or larger allocations from donor grant resources – to secondary canals and associated infrastructure will also bring substantial benefits. Over the medium term, the authorities are invited to consider the economic viability of new technologies for rural road rehabilitation and maintenance. Health Cambodia has achieved remarkable health gains over the last decade. Overall spending on health increased, but out-of-pocket spending outlays still predominate, putting a burden on the poor and making catastrophic care too expensive for people of all income levels. While spending more on certain categories of health care could help improve health outcomes, it is crucial that any such allocations be accompanied by measures to increase the efficiency of existing spending and be funded, at least through the medium term, through savings within the sector. First, the efficiency of expenditures on health care could be enhanced by more efficient purchasing, particularly for pharmaceuticals, medical equipment, and supplies. Savings could exceed $50 million a year, or one-third of government health spending, the equivalent of 0.4 percent of GDP. Second, budget allocations for central departments, hospitals, and provinces should be based on transparent formulas and clearly agreed outputs. Third, mechanisms for budgeting and expenditure tracking should be strengthened at operational district and facility levels. Fourth, the shift toward semi-autonomous service delivery arrangements for hospitals and some health districts should be accompanied by measures to limit increases in user fees and to establish appropriate financing arrangements to ensure quality and continued access by the poor and near-poor. Education Cambodia has achieved substantial progress in improving access to primary and tertiary education. The emphasis now needs to be on improving access to early childhood education and secondary education and raising quality of outcomes. As in other fields, education in Cambodia remains highly dependent on external donor support. Even with extensive external aid, however, overall government spending is still only as large as outlays by households, placing a large burden on segments of the population. iii Government spending on education is in line with government targets but lower relative to GDP than in countries with comparable per-capita GDP. Before considering increasing in overall spending on education, the authorities are invited to consider the potential savings from the large decline in enrollment in recent years. Such savings would be reaped only if the authorities can reallocate resources from schools and districts with declining enrollment. A small part of these savings – about 0.1 percent of GDP – could be used to allocate sufficient funds to increase the availability of textbooks. Some of these savings or additional donor support can provide additional resources for student scholarships. The framework of planning and budgeting works well in education in general. There is scope for improving the program budget sub-program for providing school operating budget. While devolving more responsibility, the authorities should work to ensure that governance issues are adequately addressed so that outcomes are improved. The capacity of the Ministry of Education, Youth and Sports also needs to be strengthened to process data for decisions related to policy planning and implementation. In particular, monitoring of the Educational Strategic Plan needs to provide for the opportunity to revise policy actions on the basis of the level of achievements. Educational quality as measured by student assessment needs to be mainstreamed and scaled up from the current project-based approach to provide a solid metric on the basis of which to take decisions and to evaluate whether public resources are well spent. Conclusions and pricing of recommendations In summary, the report presents the following key messages: x Increasing fiscal space through reducing tax exemptions, tightening collection, and restructuring existing spending should help offer room for larger allocations to areas crucial for sustaining strong and inclusive growth. This includes infrastructure, adequate funding for rehabilitation and maintenance of existing capital spending, research and extension in agriculture, textbooks and guidebooks in education. x Improving the quality, coverage and classification of items in the government budget is crucial for designing, executing and monitoring government policy and strengthening accountability. The authorities should strive to bring all government spending on budget, including disbursements on donor-financed projects, donor-financed technical cooperation and donor top-ups of civil service wages. Agreement is needed on the expenditure reports on donor-supported projects prepared using the government’s chart of accounts – which will enable their better integration into the budget. x There needs to be better alignment between the budget, strategic planning and donor support. Sector strategies need to be designed with full cooperation between donors and the government. Achieving these goals will require improved collaboration among donors to overcome existing fragmentation and heightened government leadership. x Advancing reforms of public financial management is crucial for ensuring public money is spent effectively and budgets, responsibilities, and rewards are better aligned. x Public financial management reforms need to be better integrated with reforms of the public administration and decentralization and deconcentration. The allocation of responsibility for service delivery and resource management is linked to wage levels, staffing and the internal organization of the government. x The functioning of Cambodia’s processes for managing and spending public money needs to be set in the broader context of the manner in which the overall public sector operates. Substantial iv progress has been achieved in developing the foundations for an effective state. But several issues constrain the government’s ability to deliver core services and all available studies point to persistent high levels of corruption. x Establishing more effective, transparent and rule-based governance remains at the heart of Cambodia’s development challenge (and at the center of the World Bank’s support for the country’s future). Progress in improving public finances is critically dependent on progress in enhancing transparency, accountability, and participation. The analysis in the report is the basis for suggestions for the authorities to consider related to the level and composition of government spending, public financial management, and more specifically for the sectors of agriculture, health, and education. All of these recommendations have fiscal implications. The most important ones are summarized in the table below. IFAPER recommendations: summary of revenue and cost implications Recommendations Fiscal impact per year Note Reduce tax exemptions saves 0.6%GDP Potential size of savings 6-7 %GDP Agriculture Overall: broadly neutral Recommendations suggest reallocating spending within the existing envelope Health Overall: cuts spending Improve the efficiency of pharmaceutical, saves 0.4%GDP Equivalent to a third of equipment, and medical supply health outlays expenditures Increase outlays on health equity funds spend 0.01-0.05%GDP Education Overall: broadly neutral Free space from the 12.5% drop in saves 0.02%GDP primary enrollment Spend more on scholarships and student spend 0.05%GDP textbooks Note: For details, consult the list of recommendations at the end of Chapters 1-6. v Chapter 1. The Strategic Setting: Restoring and Expanding Fiscal Space A. Introduction Cambodia has made remarkable progress since the UN-supported elections in 1993. Full peace was achieved in 1998, underpinning macroeconomic stability and enabling the authorities to advance structural reforms and donors to accelerate financial support. These developments have allowed Cambodia to take full advantage of its location in the most dynamic region in the world. Real GDP rose by 7.4 percent a year on average during the 1990s before quickening to 9.7 percent a year during the 2000s, the fastest pace in developing East Asia after China. Rapid growth has helped to more than triple per capita income since the early 1990s to $760 in 2010 and to reduce the poverty rate from 47 percent in 1993-1994 to 21 percent in 2008.2 The government’s balance sheet improved substantially, foreign investment rose markedly, and the rapid rise of textile production put Cambodia firmly on track to become integrated into regional and global production networks. The 2008-09 global economic and financial crises interrupted two decades of strong growth, but the economy handled the stress well. Real GDP is estimated to have declined to 0.1 percent in 2009 as exports of garments and tourist arrivals declined and construction output contracted. Drawing down on the large deposits it had built earlier, the government increased its spending to limit the extent to which the crises depressed domestic demand. As a result, the fiscal deficit more than doubled to 8.1 percent of GDP in 2009. With growth recovering robustly in 2010 and over the medium term, the authorities are aiming to narrow the fiscal deficit and rebuild government deposits as insurance against future shocks. At the same time, the authorities are exploring options to increase the available fiscal space and its possible uses. The government is supplementing these discussions with efforts to increase the efficiency and effectiveness of public spending. This Integrated Fiduciary Assessment and Public Expenditure Review (IFAPER) analyzes the structure and efficiency of government spending in Cambodia, with an emphasis on agriculture, education, and health. This analysis is supplemented by a review of public financial management. The overriding objective of the recommendations presented, and the key motivation of the IFAPER, is to support sustained strong growth in output and employment, to help reduce poverty, and to ensure fiscal sustainability in Cambodia. The report is organized as follows. The remainder of Chapter 1 presents the strategic background, including macroeconomic and fiscal developments. The chapter concludes with options for restoring and expanding the government’s fiscal space. Chapter 2 discusses the level and composition of public spending. Government spending relative to GDP is broadly in line with levels in countries with a similar per capita income as are the outcomes of this spending. For the economy to continue to grow strongly, however, the authorities have set themselves ambitious goals for rapidly increasing the efficiency of spending. The recommendations presented in this chapter are prepared with this goal in mind. 2 The mean estimate of poverty in 2008 is accompanied by a larger than previously observed standard deviation because of the small sample size (3,548 households) and the surge in food and fuel prices that year. A large--scale household survey was carried out in 2009, and when these data become available, they will give a clearer picture of the poverty profile in Cambodia. 1 Chapters 3, 4 and 5 discuss the level, composition and effectiveness of government spending on agriculture, health, and education, respectively. The analysis suggests that government spending in these areas is lower relative to GDP than in countries with similar levels of income per capita. The results of spending, however, are consistent with income rankings. The analysis concludes that while more funds to specific areas could help improve some outcomes, larger budgetary allocation will be possible only if more fiscal space is found and if the efficiency of government spending and the quality of service delivery are improved. Chapter 6 discusses Cambodia’s progress and challenges in reforming public expenditure policy and public finances. Achievements improving budget credibility under Platform One of the Public Financial Management Reform Program have been remarkable, but sustaining progress remains a key challenge. Platform Two broadens the reforms and aims to devolve greater budget responsibility to program managers through enhanced internal controls and accountability mechanisms. The chapter underscores the key areas where additional attention is needed to help advance the reform agenda. These include improved donor collaboration, strengthened MEF management structures, alignment of all public- sector reforms, and the need for public administration and compensation reform to improve government efficiency and strengthen the quality of public services. B. Recent Economic Developments and Challenges Strong growth has returned after the crisis-induced contraction in economic activity in 2009. The authorities’ emphasis has shifted from crisis response to policies needed to ensure strong and inclusive economic expansion over the long term and a transition to middle-income status. Output, Employment, and Inflation Cambodia’s concentrated export structure, large dependence on foreign investment and foreign aid, and small domestic market contributed to the contraction of output in 2009. Real GDP is estimated to have declined to 0.1 percent in 2009 after expanding by an average of 9.7 percent between 1999 and 2008. The drop in output reflected the contraction in garments production and exports and in the construction and tourism sectors, despite a robust increase in agricultural production. Cambodia became a clothing exporter in the mid-1990s, and garments grew rapidly from a $27 million export industry in 1995 to $3.0 billion in 2010. The industry is now the largest foreign exchange earner in Cambodia, accounting for about 60 percent of the country’s export receipts and nearly a quarter of industrial employment. The tourism industry has also expanded rapidly, with tourism receipts rising tenfold over the last decade to $1.8 billion in 2010. The economy has rebounded strongly from the crisis. Resurgence in tourist arrivals and garment exports appears to have helped real GDP to expand by 6.0 percent in 2010 probably by as much in 2011. The 2009 fiscal stimulus also helped (Figure 2). Garment exports rose by 24 percent in 2010, reflecting the recovery of retail sales in the US, the most important market for Cambodia. Tourist arrivals are on an upward trend and appear to have increased by about 16 percent so far in 2011, the same pace as during 2005-2010 on average. Agriculture prospects remain strong, with paddy production rising by 13 percent a year on average since 2005. Thanks to an abundant harvest, paddy prices were broadly unchanged during 2010, a year in which many countries in the region suffered from a surge in weather-related increases in food prices. About 70 percent of the labor force is at least partially employed in agriculture, mostly for subsistence, and the sector has the highest percentage of the poor in the country at about 79 percent.3 3 World Bank 2005, “Cambodia Rural Sector Strategy Note: Towards a Strategy for Rural Growth and Poverty Reduction”, June 2 Figure 1. Output is growing anew after a crisis- Figure 2. The fiscal deficit remains large after induced contraction in 2009 widening in 2009 (in percent) (In percent of GDP) 17 24 20.1 20 19.6 12 16 12 7 8 Overall balance (excl. grant) 4 Total expenditure and net lending 2 0 -4 -3 -8 2010e 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 1996 1995 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010e Industry GDP Growth Sources: NIS and World Bank Sources: MEF, IMF, and Bank staff estimates Headline inflation declined to 3.1 percent in 2010 from 5.3 percent in 2009. Given the dollarization of the economy and the tight link between the riel and the US dollar, there is a risk that further increases in international food and fuel prices may have substantial negative effects on domestic prices. However, the strong harvest and the ample stocks have so far limited the impact. Balance of Payments The recovery in domestic investment and consumption is likely to have widened the current account deficit again in 2010. The external shortfall appears to have increased to 7.1 percent of GDP in 2010 from 5.4 percent in 2009 and 2.8 percent on average during the 2000 to 2008 period. The deficit was largely driven by the trade deficit as both imports and exports recovered strongly. Net capital inflows increased during 2010, financing the external shortfall and boosting foreign exchange reserves. During 2010, capital inflows rose by about 25 percent to US$1.3 billion (nearly $0.3 billion of net inflows over 2009), reflecting a pickup in foreign direct investment (FDI) and larger loans from foreign banks. Foreign exchange reserves rose to $2.7 billion at the end of 2010, equivalent to approximately four months of imports, the highest they had been for two decades. And Cambodia’s good economic prospects caused inflows of foreign direct investment (FDI) to recover in 2010-2011 after declining in 2009. It is expected that inflows of FDI will increase from 6.7 percent of GDP in 2010 to 7.1 percent in 2011. Monetary Policies and Banking Developments The economy is effectively dollarized, limiting policymakers’ monetary options. The role played by the local currency has declined with time, with riel-denominated deposits accounting for only 3.1 percent of total bank deposits at the end of 2010. Pervasive dollarization has been a key factor in leading the central bank to keep the riel tightly linked to the US dollar, with fluctuations limited to a band of ±1 percent. Even within this straitjacket, the central bank eased monetary conditions in 2009 by cutting the minimum reserve requirement ratio from 16 percent of gross deposits to 12 percent and by eliminating restrictions on real estate lending. 3 Table 1. Cambodia Key Macroeconomic Indicators 2001 … 2006 2007 2008 2009 2010e 2011f Output, Employment and Prices (Percent change year-on-year unless indicated otherwise) Real GDP 10.7 10.8 10.2 6.7 0.1 6.0 6.0 Industrial production 11.2 18.3 8.4 4.0 -9.5 13.5 10.5 Consumer prices 0.7 4.2 14.0 12.5 5.3 3.1 7.5 Foreign Trade, BOP and External Debt (In millions of U.S. dollars unless indicated otherwise) Current account balance (excl. official transfers) -347 -522 -705 -1280 -1,203 -1,570 -1,631 (% GDP) -8.7 -7.1 -8.1 -12.4 -11.6 -13.5 -13.0 Current account balance (incl. official transfers) -45 -47 -296 -700 -562 -828 -1,179 (% GDP) -1.1 -0.6 -3.4 -6.8 -5.4 -7.1 -9.4 Trade balance -523 -1,078 -1,343 -1,801 -1,634 -1,698 -1,859 Exports of goods 1,571 3,693 4,089 4,708 4,196 4,687 5,483 Imports of goods 2,094 4,771 5,432 6,509 5,831 6,384 7,342 Foreign direct investment 150 475 866 795 520 774 890 External debt 1,321 2,245 2,555 2,808 3,054 3,514 3,813 (% GDP) 33 31 29 27 30 30 30 Short-term debt 224 209 218 218 218 218 218 Debt service ratio (% exports of g&s) 2.9 1.4 1.0 1.0 1.2 1.0 1.0 Foreign Exchange Reserves 548 1,097 1,616 2,164 2,367 2,653 3,237 (months of imports of g&s) 2.7 2.4 3.1 3.4 4.1 4.3 4.4 Financial Markets Domestic credit (% change y-y) -4.1 35.7 70.7 51.1 19.9 35.3 35.0 Exchange rate (Riel/US$, eop) 3,900 4,061 4,003 4,081 4,169 4,053 4,100 Real effective exchange rate (2000=100) 99.7 92.4 95.1 111.7 111.3 108.0 .. (% change y-y) -0.3 2.1 3.0 17.4 -0.4 -2.9 .. Memo: Nominal GDP (million US$) 4,004 7,350 8,754 10,284 10,328 11,608 12,529 Sources: IMF, Ministry of Economy and Finance, and World Bank staff estimates. Growth of credit to private sector gained momentum again in 2010 after a significant slowdown in 2009. Bank liquidity surged, as deposits rose to a high of $4.3 billion by the year ending 2010. The ratio of loans to deposits remained broadly unchanged at 74 percent. Most lending over the last year was directed to domestic trade, tourism-related activities, and some manufacturing. Strengthened banking regulation, mainly regarding the standardized charts of accounting practices and secured transactions, have also helped to increase trust in the banking system. 4 Figure 3. Growth and inflation have been high since Figure 4. Real exchange rate is depreciating the mid-1990s (Cambodian riel and US dollar) (in percent) Average Real GDP Growth Average Inflation 140 3,400 135 3,500 10 9.2 130 125 3,600 7.8 8.2 120 8 7.5 3,700 6.7 115 6.2 110 3,800 105 3,900 6 4.9 100 95 4,000 4 3.1 90 4,100 85 4,200 80 2 75 4,300 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 0 1996-00 2001-05 2006-10 2000-10 Real effective Against dollar (RHS) Sources: NIS and World Bank staff estimates. Sources: National Bank of Cambodia and IMF. Poverty and the Millennium Development Goals Reducing poverty and achieving the other Cambodia Millennium Development Goals (CMDG) are the key priorities under the government’s National Strategic Development Plan.4 Other socioeconomic priorities identified in the National Strategic Development Plan (NSDP) are agriculture, physical infrastructure, private sector development, health, and education. To achieve these goals and to promote development according to these priorities, the authorities are also counting on prudent spending of well-targeted public resources. Usually, health, education, and transport are sufficiently funded under the NSDP, while funding for rural development and agriculture are somewhat short of the targets (Annex Display 6). The recent United Nations and Asian Development Bank report on the millennium development goals (MDG) commended Cambodia for making steady progress toward meeting the MDGs, especially on the goal of reducing poverty to 19.5 percent by 2015.5 Against these overall solid advances, the report noted Cambodia’s slower progress in reducing the under-5 mortality rate and the maternal mortality rate and in achieving universal primary education (Annex Display 1). 4 Economic policies have been guided by five-year plans since 1996. The plans have centered on maintaining political and macroeconomic stability, strong growth and poverty reduction. During the first Social Economic Development Plan (1996- 2000), real GDP growth averaged 7.5 percent. Growth quickened substantially under the second plan (2001-2005), later known as National Strategic Development Plan (NSDP), with output rising 9.2 percent a year. The global economic crisis notwithstanding, growth during the NSDP (2006-2010) remained strong at averaged 6.7 percent a year, broadly in line with the plan’s target. The updated NSDP targets economic growth of 6 percent a year and per capita income of nearly $1,000 by 2013. 5 Joint UNDP, ADB and UNESCAP Report “Paths to 2015: MDG Priorities in Asia and the Pacific” 2010/11 5 C. The Fiscal Setting Fiscal prudence and an increase in revenues because of strong growth and better collection helped narrow the fiscal deficit sharply in the decade up to 2008. Government spending began to rise in 2006 from relatively low levels, however, increasing to nearly 15 percent of GDP by 2008. At the same time, the government increased its deposits with the central bank by more than twentyfold to $732 million between 2003 and 2008. This large cushion enabled the government to boost spending in 2009 to an all- time high, helping mitigate the impact of the global economic crisis. The Fiscal Balance During and After the Crisis The sharp decline in economic activity during the global economic crisis led the government to widen the fiscal deficit, in part to cushion the impact of the economic downturn on domestic demand. This was decisive and appropriate, even though concerns have been raised about the resulting large increase in the wage bill. The fiscal balance shifted from a surplus of 0.3 percent of GDP including grants in 2008 into a deficit of 4 percent deficit in 2009 (Table 2). Almost all of the deterioration reflected larger spending, split almost equally between an increase in investment outlays and higher public wages. Tax revenues eased little despite the slowdown because the sectors most affected by the crisis – including garments – had been granted generous tax exemptions earlier, which meant that their tax payments were low anyway. Fiscal policy is the main instrument for macroeconomic management given the pervasive dollarization and the de facto peg of the riel to the dollar. As the economy recovers from the crisis, the government is working to reverse the 2009 fiscal relaxation. This should ensure fiscal sustainability and align the short-term considerations that guided fiscal policy during the crisis with long-term policy objectives as elaborated in the NSDP. Furthermore, the limits of fiscal policy for short-term economic management are well known. These include long implementation lags and administrative uncertainties. The fiscal deficit appears to have narrowed to 2.2 percent of GDP in 2010 in large part due to a rise in revenues. Revenues rose because of the stronger economy, improved enforcement of import taxes and duties, and the doubling of the road tax on vehicles.6 On the expenditure side, the government curbed civil service promotions and recruitment and curtailed other discretionary spending. 6 From 2010, the government introduced tax on fixed assets (land and buildings, both completed and under construction) at an annual rate of 0.1 percent for assets with value equivalent or above CR100 million (about $25,000). 6 Table 2. Cambodia: General Government Fiscal Accounts, 2000-2010 (in percent of GDP) 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010e Total revenues and grants 13.0 12.5 13.2 11.8 12.0 12.4 14.0 14.1 15.1 15.6 17.9 Revenue 10.2 10.0 10.5 9.6 10.3 10.3 11.5 11.9 12.0 11.5 13.1 Tax 7.5 7.2 7.4 6.6 7.7 7.6 7.9 9.5 9.9 9.4 10.6 Nontax (incl. Capital Rev.) 2.7 2.8 3.1 3.0 2.6 2.7 3.5 2.3 2.1 2.1 2.4 Grants 2.7 2.5 2.7 2.2 1.7 2.1 2.6 2.2 3.1 4.1 4.8 Total expenditures and net lending 15.0 15.2 16.8 15.5 14.8 12.8 14.2 14.7 14.8 19.6 20.1 Current expenditures 8.7 8.9 9.4 9.9 8.9 7.9 8.5 8.7 9.0 11.2 11.0 Capital expenditures 6.4 6.2 7.4 6.5 6.0 4.9 5.8 6.1 5.8 8.4 9.4 Cash adjustment .. .. .. 0.9 .. .. .. .. .. .. 0.3 Overall balance (incl. grants) -2.1 -2.6 -3.6 -3.7 -2.9 -0.4 -0.2 -0.7 0.3 -4.0 -2.2 Overall balance (excl. grants) -4.8 -5.2 -6.3 -5.9 -4.5 -2.5 -2.7 -2.9 -2.8 -8.1 -7.0 Financing 2.1 2.6 3.5 3.7 2.9 0.4 0.2 0.7 -0.3 4.0 2.2 External 2.3 2.4 3.9 3.5 3.1 3.0 2.4 3.2 2.5 2.5 2.4 Official creditors 2.3 2.4 3.9 3.5 3.1 3.0 2.4 3.2 2.5 2.5 2.4 Domestic -0.2 0.3 -0.4 0.5 -0.2 -2.4 -2.0 -2.2 -2.5 1.8 0.2 Debt Amortization 0.0 0.0 0.0 -0.3 -0.1 -0.1 -0.2 -0.3 -0.2 -0.3 -0.3 Pending Memorandum Items: Grants 2.7 2.5 2.7 2.2 1.7 2.1 2.6 2.2 3.1 4.1 4.8 Loans 2.3 2.4 3.9 3.5 3.1 3.0 2.4 3.2 2.5 2.5 2.4 Technical assistance 1/ 7.9 6.9 5.9 5.8 5.5 4.7 4.8 3.7 3.8 3.3 3.3 Total 13.0 11.8 12.4 11.5 10.3 9.7 9.7 9.0 9.3 9.9 10.5 Sources: IMF, TOFE by the Ministry of Economy and finance; and World Bank staff estimates. 1/Technical assistance and other direct supports to CSOs (not channeled through the Ministry of Economy and Finance.) Government Revenues Revenues excluding foreign grants increased by 1.3 percent of GDP between 2000 and 2009, an increase that was smaller than in 2010 alone. The rise through 2010 reflected exclusively the broadening of the tax base, simplification of the tax system, and improved compliance. Even at the increased level in 2010, however, revenues as a share of GDP are lower in Cambodia than in the low- income countries on average (Figure 5).7 The value of exemptions has been on the decline but still remains large at about 4.5 percent of GDP (Figure 6). The bulk of tax revenue derives from indirect, consumption-based taxes. The emphasis on indirect taxes, especially on VAT, is appropriate given the large informal economy, the declining share of international taxes, and the potential for tax evasion (Table 3). However, one-third of all revenues are from VAT – and 60 percent of these are on imported goods – leaving the budget directly exposed to external shocks. Receipts from corporate income taxes are small, largely because of exemptions, and are much lower than in Lao PDR and Vietnam where they account for about one-third of tax collection. 7 The average includes the Kyrgyz Republic with its legacy from the period of the USSR of high taxes. Excluding the Kyrgyz republic, the average drops to 13.5 percent. 7 Figure 5. Revenues excluding grants are lower in Figure 6. The value of exemptions from paying taxes Cambodia than in other low-income countries remains large (In percent of GDP in 2008) (In percent of GDP) 40 7 35 6 30 5 25 4 20 3 15 Garment Total 2 Investment 10 1 5 0 0 Cambodia Low-income ASEAN East Asia OECD Source: World Development Indicators, the World Bank. Source: Cambodia General Department of Customs and Excise Budget Financing The government drew down its deposits with the central bank to finance the wider deficit in 2009. Up until 2008, by contrast, foreign grants and concessional loans financed almost all of the fiscal deficits as other options are limited. Borrowing on non-concessional terms is prohibited by the budget law, the government has not accessed domestic or international capital markets, and privatization receipts have been negligible. Government deposits with the central bank were reduced from $732 million, or 7.1 percent of GDP, at the end of 2008 to $540 million (about 5 percent of GDP) at the end of 2009. As the economy recovers, it is imperative for the government to rebuild these reserves to restore this important cushion. Foreign aid rose sharply during the 2000s but has appeared to have tapered off. Cambodia’s aid per capita increased from $31 in 2000 to $51 in 2008 and has remained little changed since then. In addition to the foreign aid channeled through the budget, foreign donors provide substantial technical assistance to the government and to NGOs. (For more details, see Chapter 2.) While aid alignment and coordination among donors has improved, there is more to be done to align donor support with the government’s strategic planning and the annual budget. Some of the key issues that arise include: x Donors have largely supported primary infrastructure and capital projects (such as the construction of main roads, bridges, and irrigation canals) while leaving to the national budget to finance secondary and tertiary infrastructure and subsequent maintenance. Given the limited room that is available within the national budget, adequate funding for subsequent maintenance often cannot be secured. x A substantial amount of foreign aid is channeled through technical cooperation and NGOs. As a first priority, there is a need for the government when preparing the annual budget to take into account the foreign aid given to NGOs for carrying out government-like functions. x As a medium-term priority, the donors and government would be well-advised to consider how to fold most donor support to Cambodia into the budget. This will require a shift away from project management units or equivalent entities to established government systems, much better coordination 8 of support from the different donors, and a much greater leadership role from the government in preparing for and requesting such support. Table 3. Structure of Government Revenues, 2000-2009 (in percent of total) 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009e Revenues 100 100 100 100 100 100 100 100 100 100 Tax revenues Taxes on income and profits 8.9 8.4 6.9 7.4 6.6 7.8 9.1 10.9 11.3 13.4 Taxes on property (rental) 0.5 0.5 0.5 1.0 0.5 0.5 0.6 0.6 0.7 0.8 Taxes on consumption 33.6 35.7 36.3 34.3 40.0 39.8 36.6 41.1 46.7 43.8 o/w VAT 25.7 25.8 24.3 23.1 26.2 25.5 24.4 26.2 28.3 27.7 o/w Excise taxes 7.8 9.9 11.9 11.2 13.8 14.3 12.2 14.8 18.4 16.1 Turnover Tax 0.9 0.6 0.4 0.6 0.3 0.4 0.4 0.4 0.3 0.3 Taxes on international trade 27.0 24.0 24.0 22.1 23.1 21.5 18.7 21.7 19.8 20.1 o/w Import duties 25.9 23.3 23.1 21.1 22.2 20.8 18.0 20.9 19.4 18.7 Other taxes 2.4 2.9 2.5 3.2 3.9 3.3 3.8 5.6 3.6 5.2 Nontax revenues Tourism (inc. visa fees) 1.8 2.7 3.4 3.5 4.1 4.9 4.5 4.8 3.9 3.8 Other 22.9 24.5 25.0 26.1 20.5 16.9 15.3 12.1 10.1 10.4 Capital Revenues 2.1 0.6 0.9 1.7 0.9 4.8 11.0 2.8 3.7 2.2 Sources: Ministry of Economy and Finance, TOFE. e= estimate. D. Government Debt and Debt Sustainability At the end of 2009, Cambodia’s external debt (almost all of it composed of public and publicly- guaranteed obligations) amounted to about 28 percent of GDP in nominal terms and 26 percent in net present value (NPV) terms. The debt stock rose to nearly 30 percent of GDP by the end of 2010, however, reflecting increased assistance from donors in the face of the global recession. The size of domestic public debt is currently small at about 1 percent of GDP. On a net basis, subtracting the stock of government deposits, domestic debt is negative relative to GDP. The government has also issued guarantees to state owned companies, wholly owned by the government. The guarantees cover several contracts entered into by the state owned companies with private companies. For example, a state-owned company entered into a power purchase agreement (PPA), in which the company agreed to purchase electricity whether there is demand for the electricity or not. The government guarantees the power purchase in case the state owned company cannot pay for it. A recent debt sustainability analysis indicates that Cambodia is at a moderate risk of debt distress.8 Potential risk may emerge as the government has doubled its legal ceiling of borrowing, adopting a more relaxed rule for 2011 and onwards to finance more capital projects. Borrowing on less concessional terms could also add to the risks, as will liabilities from contingent obligations related to hydro-power plants, other major public infrastructure investments. Therefore any increased loan ceiling should be carefully contemplated, as it could significantly limit the government’s ability to respond to any future crisis. 8 Cambodia’s Joint IMF/World Bank Debt Sustainability Analysis (in collaboration with the Asian Development Bank), October 2010, Washington DC. 9 Stronger capacity in debt management is an essential prerequisite for prudent use of foreign borrowing, including on less concessional terms. Such capacity will become even more important when Cambodia begins considering a domestic government bond market. With an equity market now in operation, the authorities need to consider carefully the costs and benefits of government domestic borrowing (see below). E. Fiscal Space Fiscal space refers to the availability of budgetary resources without jeopardizing the sustainability of the government’s financial position or the stability of the economy.9 Governments can create fiscal space by raising taxes or non-tax revenues, by securing additional external grants, by reducing lower priority expenditures, or by borrowing at home or abroad. Reducing the Fiscal Deficit and Rebuilding Deposits The first priority for the government in relation to fiscal space is to reduce the fiscal deficit and rebuild the stock of government deposits. The fiscal deficit needs to be reduced further over the medium term and the stock of government deposits needs to be rebuilt up to at least its pre-crisis level over the next several years, for example, by 2015. In the absence of access to capital markets, Cambodia will continue to rely on these government deposits (cash balance with the central bank) as a cushion against adverse external shocks. In due time, the authorities will begin to consider options for future access to capital markets. Such an option may help limit the need to hold a large unremunerated (by law) cash cushion with the central bank. There are strong arguments in favor of creating fiscal space in addition to the cushion needed against future shocks. Cambodia’s fixed investment – both by the government and the private sector– amounted to 19 percent of GDP in 2007-2009, below the 25 percent average among the 13 countries that managed to grow by at least 7 percent a year for 30 years according to the Growth Commission Report.10 Given the limited capacity of the private sector in Cambodia and the country’s already substantial dependence on external savings, creating more sustainable fiscal room for larger public investments and for enhanced maintenance will be crucial. Fiscal space will also ultimately be needed to bring off-budget spending and some private outlays on public goods on budget. The current donor funding for such activities will likely decline over time or be redirected to the budget if the transfer is done transparently. Increasing Fiscal Space: the Role of Revenues Measures to increase revenues need to be weighed against concerns about whether these will hamper economic growth more than help government expenditure on public goods. Our finding that Cambodia has a lower ratio of revenues to GDP than countries with similar levels of per capita income suggests that there is room to increase revenues without negatively affecting growth. There is scope for further reforms to level the playing field for companies and improve revenue collection. The government might wish to consider standardizing the accounting records of local businesses, strengthening the capacity of the tax administration, and institutionalizing the tax assessment system to limit scope for discretion. It is crucial to expand the use of the banking system more widely among the population (as announced in 2005) for tax collection purposes. Efforts to improve tax 9 Heller, Peter (2005) and World Bank (2009). 10 Strategies for Sustained Growth and Inclusive Development, 2008, Growth Commission, www.growthcommission.org. 10 collection and the recent crackdown on log and timber smuggling are likely to boost revenues, and similar actions should be considered in other sectors of the economy. Reducing tax evasion will bring substantial revenues and will result in companies, both domestic and foreign, being treated more equitably. In 2006, the Ministry of National Assembly and Senate Relations and Inspection revealed that nearly 500 companies owed millions of unpaid taxes.11 Another study found that only about one-fourth of taxes due were collected in 2005.12 Besides weak enforcement, the other main reason for this low revenue collection is the large informal sector in the economy. Many businesses are operating outside the financial system and without having registered with the government. The absence of standardized accounting records for businesses is also a factor. Facilitating registration, introducing requirements for business records, and strengthening the tax administration should all be pursued. Dealing with tax evasion and smuggling is at the top of the government agenda, but emphasis should also be put on reviewing the public inventory and registration of state assets (mainly land) to discover any unregistered and untaxed rentals to the private sector. Stricter enforcement of both the law on taxation and the recently passed anti-corruption law will also reduce evasion. It is also important to rethink the extent of the tax incentives currently granted to companies. Such reconsideration should include changing the way tax incentives are provided by moving away from tax holidays towards investment allowances, tax credits, and accelerated depreciation. It is estimated that exemptions from paying customs duties and excises result in a loss of about 4 percent of GDP a year. Other incentives that are harder to quantify are likely to amount to about 2 to 3 percent of GDP and include a holiday from paying corporate income taxes for up to 9 years for qualified investment projects (QIP) approved by the Council for the Development of Cambodia (CDC). The overall cost of these exemptions, as a result, is an outsized 6-7 percent of GDP a year. Some of these exemptions could be revisited and scaled down, helping increase fiscal space. While tax exemptions help investors to overcome the disadvantage they have relative to “first movers” into a market, maintaining them for an extended period of time serves little useful economic purpose. A lower tax rate – applicable to all investors – would be preferable. In addition, tax incentives are rarely among the most important factors that investors consider when selecting a location, which tend to be political stability, the availability of decent infrastructure, electricity at reasonable prices, and wage costs not out of line with productivity. Improved public management of extractive industry resources should help increase further fiscal space and improve living standards. Cambodia offers one of the few remaining untapped fields in the world. And it is up to the government and the people of Cambodia to convert wealth underground into resources to support economic growth and prosperity. Fiscal policy plays a critical role in ensuring that natural resources are not a curse and are used prudently. A transparent fiscal framework and heightened government accountability are also crucial elements of successful natural resource management. A similar issue arises with allocations of land for economic concessions. Ensuring a fair price for such concessions will not only contribute to bolstering fiscal revenues and ensure that land is efficiently used, but is a cornerstone of efforts to improve government accountability (also see chapter 6). 11 The Cambodia Daily, July 15-16, 2006. 12 Economics Institute of Cambodia, 2006, “Assessment of Corruption in Cambodia’s Private Sector.” 11 Increasing Fiscal Space: the Role of Foreign Aid Cambodia is still dependent on aid. While the government provides strong leadership in driving the aid effectiveness, the aid environment is crowded with a large number of donors and of uncoordinated, fragmented projects. While donor aid has overall been crucial for the country’s recovery, it is becoming increasingly difficult to ensure that an aid level of about $0.9 billion a year delivers the needed development results in practice – not least because the “low hanging fruits” have already been picked. The government’s current policy choice, supported by donors, is to ensure complementary division of labor among donors and the alignment and harmonization of aid. There has to be full alignment between donor support, strategic priority-setting under the NDSP, and the annual government budgets. Donor support to the government – including for off-budget spending financed by technical cooperation – and to NGOs needs to be firmly under the government’s overview to ensure there is no duplication of effort. As a priority, donor-financed off-budget spending also needs to be brought on budget. There is also a need to harmonize and align systems and procedures. As in other countries, dealing with the different donor requirements and formats (for example, for project preparation and appraisal, reporting, M&E, or audit) imposes significant transaction costs on both the government and the donors. The government’s response has been to push for harmonization of systems at various levels. These have included delegated cooperation arrangements, multi-donor trust funds, and other forms of pooled financing and, in some cases, common manuals for financial management and procurement. Where government systems are weak, this should help improve them to acceptable donor levels. An important issue still unresolved is the existence of many project implementation units established by donors. The unpredictability of aid makes it hard to integrate external assistance into the government’s fiscal processes. This unpredictability encompasses both discrepancies between commitments and disbursements and the tendency for many donors to commit only one year ahead, which makes government programming of aid within its overall fiscal and sector strategies very difficult. The limited available data on aid makes it difficult to monitor and increase the effectiveness of aid. To date, donors have not supplied accurate, comparable, and timely data to the CDC, making it hard to align development assistance with stated government priorities. Increasing Fiscal space by Rationalizing Existing Spending Even with a relatively low level of government spending, there is room to rationalize public expenditures. Chapter 2 focuses on specific options, supplemented by the analysis in Chapters 3-5. Increasing Fiscal Space by Securing Government Access to Capital Markets Cambodia’s fiscal law rightly does not allow the government to borrow at non-concessional terms nor does it allow the issuance of government bonds. Nonetheless, the 2011 Budget Law allowed the issuance of $25 million of Treasury bills in case of urgent cash need but subject to approval by the National Bank of Cambodia. In addition, while the law prohibits the government from borrowing from the central bank, it allows the bank to temporarily extend credit to the National Treasury at a maturity of up to three months.13 The aggregate amount of outstanding credit to the government cannot exceed 13 Article 24 of the Law on the Organization and Conduct of the National Bank of Cambodia 12 10 percent of the ordinary domestic budgetary revenues (excluding grants and proceeds from the sale of assets) for the previous financial year. The government has not exercised this option for more than a decade now. Given the very high financial dollarization of the economy and lack of demand for securities denominated in local currency, the government has not accessed the domestic market for either deficit financing or for cash management purposes. The government is considering accessing domestic capital markets in the future, however. The nascent Cambodia Stock Exchange is likely to be operational later this year and this could open a new opportunity for considering the future development of a domestic bond market. Achieving such a goal will require substantial effort in improving government capacity, nonetheless. A recent debt management performance assessment (DeMPA) revealed that Cambodia meets the minimum requirements for five of the 15 debt management performance indicators.14 These include the indicators on the legal framework, managerial structure, coordination with monetary policy, and debt records, and debt management. The indicator on domestic market borrowing was not rated, as Cambodia does not issue any government securities in its domestic market. The assessment identified the following areas that require improvement and could be considered as priorities for capacity building and reform: (i) debt management strategy development and evaluation of debt management operations; (ii) audit of debt management activities, policies and operations; (iii) improving capacity to prepare a debt sustainability analysis; (iv) more systematic information sharing with the NBC and within and among relevant government departments; (v) taking a more proactive role in comparing and negotiating loan terms offered; (vi) preparing procedures for on-lending; (vii) improving cash balance management; (viii) enhancing operational risk management; and (ix) preparing and publishing regular and comprehensive debt statistical bulletins. The External Debt Division of the authority has been operational since 1997 and adequately staffed, but not formal debt management strategy has been adopted15, nor medium-term debt management strategy covering currency composition, the refinancing, and interest-rate risks has been developed. F. Conclusion and Policy Recommendations Restoring and increasing fiscal space needs to be a high priority for the government in the short term. This chapter makes the following recommendations: x As the economy recovers, it is important to rebuild stocks of government deposits at least their pre-crisis level to ameliorate the negative impact of future shocks. Over time, the government needs to consider adopting a framework for the transparent use of these deposits. x The other priority in relation to fiscal space is to narrow the fiscal deficit over the medium term. x There is scope for further reforms to improve revenue collection. Measures to increase tax net include lowering threshold for annual profit tax, raising tobacco and alcohol taxes, implementation of the property tax, capital gain tax, standardizing the accounting records of local businesses, strengthening the capacity of the tax administration, and institutionalizing the tax assessment system to limit scope for discretion. Better management of extractive resources and 14 DeMPA, November 2009, joint World Bank and UNCTAD assessment. 15 Debt Management Strategy is being finalized 13 land will also provide scope, as of now difficult to estimate, for increased revenues and better government accountability. x There is also a need to rethink the scope of tax incentives for particular industries. Gradually eliminating some of these exemptions – that in total are estimated to amount to about 6-7 percent of GDP a year –would allow the authorities to introduce a lower tax rate applicable to all investors. x Existing rules and regulations need to be better enforced. The recent strict enforcement on illegal log and timber activities should be replicable to other sectors so that smuggling, a prime source of revenue loss, can be reduced. Similarly, stringent rule-based procurement activities should be enhanced as over 60 percent of contracts above the competitive bidding threshold escaping competitive bidding through regulatory loopholes.16 16 IMF, 2010, Cambodia Article IV Consultations Report, Washington DC. 14 Chapter 2. Composition of Government Spending, Priorities, and Opportunities A. Introduction Cambodia’s level of government spending is similar to that in other countries with a comparable level of income per capita and so are its development outcomes. There is also little difference in the composition of Cambodia’s spending relative to most relevant comparators. There are two important issues related to the composition of spending that need immediate government attention. First, ensuring that the substantial capital expenditures are matched by appropriate allocations for maintenance is crucial for stronger and sustainable growth. Donor support will be essential for this to become reality. The government also needs to work with donors to ensure the proper classification of donor-supported investment projects in the government fiscal accounts. Second, a hike in wages for the military in 2009 boosted wage outlays substantially, much more than increases in the salaries for non-military public employees since 2007. There may be scope for further wage raises in some non-military areas, provided there is fiscal space and that pay scales are decompressed under the framework of an effective pay reform. Donor support remains substantial at about one-third of general government outlays or Box 1. Scope, Definitions, and Calculation of the Expenditure Data used in this Report about 40 percent of overall public spending when off-budget donor support for technical Scope: The term “general government spending” used cooperation and NGOs working on public in this report includes outlays by the central and local goods is counted. This substantial reliance on (provincial) governments, both domestically and donors implies that the government and donors externally financed, with the exception of the need to coordinate better to ensure that following: technical cooperation funded by donors; strategic planning, budget preparation, and aid expenditures by NGOs even if funded by its own coordination are fully aligned. resources; and private spending on what are public goods. Stronger economic growth and a robust Data compilation: Domestically financed spending increase in nominal revenues has made it data including provincial revenues and expenditures is possible to allocate more resources than obtained from the annual TOFE reports prepared by the planned under the National Strategic Ministry of Economy and Finance. The externally Development Plan and its 2008 Mid-term financed spending data is compiled using official Review. However, spending more money does development assistance (ODA) disbursement data not always mean that progress has been (reported by the CDC). The two data sets are combined achieved as projected. Cambodia is lagging on to obtain General Government Spending. three of the nine Millennium Development Goals. Within the country’s limited resource While these data are available, lack of consolidation envelope, there should be room to reallocate makes difficult monitoring and planning. expenditures to meet the MDGs and support growth. The recommendations in this chapter suggest some steps to take in this direction, including increasing the share of resources devoted to maintenance, increasing the wage bill only if efficiency gains are found and if reforms of the existing pay system alter incentives, and increasing resource allocations to the key sectors identified in the NSDP. The analysis presented in this chapter is based on data for general government spending that the team compiled from information provided by different government departments and donors. In addition to general government spending, there are expenditures on public goods by non-government entities that could, under different circumstances or in different countries, be carried out by the 15 government. Such spending includes outlays by NGOs on health and education predominantly externally funded and expenditure by political parties and others. A brief overview is provided in Box 1. The rest of the chapter is organized as follows. Section B analyzes the level of government spending and section C discusses the composition of government outlays. Section D presents and analyzes the results of a recent survey of private spending on public goods at the commune level. Section E concludes with recommendations to help to improve government policies. The Level of Public Expenditures After modest increases during 2006 to 2008, government spending surged in 2009 as a result of the large government fiscal stimulus package and a sharp rise in some public wages. Following the fiscal adjustment that began in 2004, government spending was nearly 15 percent of GDP by 2008, but this level had barely changed since the start of the decade (Table 4). However, in 2009, spending increased sharply by 5 percent of GDP, of which 0.7 percentage points reflected the decline in nominal GDP as a result of the economic crisis and 1.5 percentage points was due to the hike in wages, primarily for the military. The rest reflected the broader fiscal stimulus. Table 4. General government spending surged in 2009 after rising modestly during 2005-08 (in percent of GDP) 2001 2002 2003 2004 2005 2006 2007 2008 2009e Total expenditures and net lending 15.2 16.7 15.5 14.8 12.8 14.2 14.7 14.8 19.6 Current expenditures 8.9 9.4 9.9 8.9 7.9 8.5 8.7 9.0 11.2 Wages 3.1 3.5 3.3 3.5 3.2 3.3 3.0 3.2 4.6 o/w Civil administration 1.4 1.8 1.8 1.6 1.6 1.7 2.2 2.1 2.5 Other current 5.5 5.8 6.2 4.8 4.1 4.6 5.0 5.1 5.9 o/w: Interest 0.1 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 Provincial expenditures 0.2 0.1 0.5 0.5 0.5 0.6 0.6 0.8 0.7 Capital expenditure 6.2 7.4 6.5 6.0 4.9 5.8 6.1 5.8 8.4 Domestically financed 1.7 2.0 1.8 1.4 1.2 1.3 1.2 1.5 2.3 Externally financed 4.6 5.4 4.7 4.6 3.7 4.5 4.8 4.3 6.1 Memo items: Adjusted expenditures Current spending adjusted 10.1 10.7 11.3 9.9 8.8 9.5 9.7 9.9 12.4 Current from the MEF 8.9 9.4 9.9 8.9 7.9 8.5 8.7 9.0 11.2 Adjustment for externally financed 1.2 1.3 1.4 1.0 0.9 1.0 1.0 0.9 1.2 Capital spending adjusted 5.0 6.1 5.1 5.0 4.1 4.7 5.0 4.9 7.3 Locally financed capital spending 1.7 2.0 1.8 1.4 1.2 1.3 1.2 1.5 2.3 Adjusted externally financed 3.3 4.1 3.3 3.6 2.9 3.4 3.8 3.3 5.0 Sources: IMF, Ministry of Economy and Finance, and World Bank staff estimates. There are some outlays on public goods that not included in the general government spending figures and these need to be taken into account. These outlays are detailed in Table 5 and include: x Technical cooperation: These are outlays fully financed by donors focused on strengthening the technical skills of government officials and on institutional building. These outlays amount to about 3 percent of GDP, or one-third of overall foreign assistance. In most other countries, this item is classified as spending by the general government. In Cambodia, it is not presented as such to reflect the limited control that the government has over how it is used and the lack of any co- financing or future commitments by the government. As part of our recommendations, we 16 suggest that the government and donors consider working closely together to agree the scope of these activities and to consolidate all such spending under the general government budget. x Outlays by NGOs: These are substantial outlays on public goods, including on education and health, financed by NGOs (own resources). These same NGOs also often help to implement government programs that are in the same areas and are presented on-budget. When NGOs carry out spending on public goods that, in other countries or under other circumstances, would be carried out by the government, it is important that the authorities consider the size of and areas for such spending in preparing the budget. Donor support to NGOs is no more predictable over a longer period of time than its support to the government and can pose similar dilemmas and create distortions. x Private spending on public goods: Private donors – typically political parties – finance part of the spending by local governments on schools, hospitals, and roads (see Section D below). General government spending and outlays on public goods by non-government entities amounted to about 20 percent of GDP in 2007-08 and as much as 24 percent in 2009. These numbers are not excessively high, but are closer to the upper limit for countries with similar levels of per-capita GDP. Reflecting them properly in the budget needs to be a medium-term priority, but taking them into account when preparing the National Development Strategic Plan and the budget needs to be of immediate concern. This should be accompanied by efforts by the government and donors in partnership to closely align donor support with the strategic plan and the budget. Table 5. Overall public spending in Cambodia is Figure 7. General government spending is in line with large that in countries with similar per-capita income (in percent of GDP, 2009) (in percent of GDP, 2008 and 2009) Total 24.0 30.0 Kenya General Government Spending 19.6 Cambodia/1 Burkina Faso Ghana Benin 25.0 Domestically Financed 14.0 20.0 Mali Cambodia (2009) Lao PDR Externally financed (grants and loans) 5.6 15.0 Bangladesh Cambodia (2008) 10.0 Technical Cooperation1/ 2.9 5.0 NGOs financed from domestic donors 1.0 0.0 2/ 450 500 550 600 650 700 750 800 Private spending on public goods 0.5 Sources: MEF and World Bank staff estimates. Source: World Development Indicators 1/ 1/ Notes: Financed by external donors and executed by Note: Overall public spending for 2009 as in Table 5. government. 2/ On the basis of a commune level survey (see below). The above aggregation of government spending and private spending on public goods provides a useful insight into the share of redistribution of resources in the economy. Data that allow such calculations for other countries at similar income levels are often not available. Therefore, the rest of the analysis – especially when we compare Cambodia with its neighbors or countries with similar levels of income – will focus on the general government budget. 17 Until 2008, Cambodia’s general government spending relative to GDP was similar to that in countries with similar levels of per capita income (Figure 7). However, few countries with Cambodia’s per-capita income introduced the kind of fiscal stimulus package that Cambodia did in 2009, by drawing down on government deposits. As a result, spending relative to GDP in Cambodia in 2009 was probably at the higher end of the spectrum of comparable countries. That was a timely and decisive move. As the stimulus is withdrawn, the government is likely to maintain some of the higher spending to help improve much needed public service delivery, in line with increased revenues and ongoing efforts to reduce tax exemptions and tax evasion. While an increased revenue envelope will also help alleviate important constraints, including infrastructure – especially in the power sector – the government needs to ensure that additional funds are spent wisely, efficiently and effectively. Foreign donors finance about one-third of general government spending through grants and loans. Donor support is expanding rapidly in dollar terms but was declining – until the start of the crisis – relative to GDP. Together with their technical cooperation grants for capacity and institutional building and their financing of NGOs, foreign donors finance about 40 percent of overall public spending. How donor support will evolve over time will depend on a variety of factors, including the ability and the intention of the government to raise substantial additional revenues of its own, the need for higher levels of government spending given the level of per capita income, and its implementing capacity. At the same time, it is imperative for the government and donors to make more progress on aligning donor support with the strategic directions for the economy and the budget. Formally bringing donor support on budget should also help the government to plan and monitor the execution of fiscal policy. The Composition of Government Expenditures The composition of general government expenditures was little changed up until 2008. However, the fiscal stimulus in 2009 and the hike in wages – mainly for defense and security personnel – resulted in a substantial increase. While most elements of the fiscal stimulus are set to be withdrawn, the wage increases are here to stay. Economic Classification Our classification of budget allocations in this section is broadly consistent with the government’s goals presented in the NSDP. The surge in military and defense wages in 2009, however, squeezed outlays for maintenance, which are critical for keeping the capital stock in decent shape.17 Wage Bill After changing very little earlier, the wage bill rose sharply in 2008 and 2009 as a share of GDP. In addition to the 2009 hike for defense and security staff, there has been a more gradual increase in wage outlays for other public employees since 2007 (Figure 8-Figure 9). Several other features of the wage bill and public employment stand out: x The overall size of the wage bill is not higher than in countries with comparable per-capita GDP. x Public employment outside the defense and security sectors -- about 1.5 percent of the population – is smaller than in other countries with populations of a similar size (Figure 10). 17 The authorities in Cambodia classify the overall wage bill into two components: (i) defense and security and (ii) civil administration. The latter includes wages for teachers and medical doctors. 18 However, the difference is not that significant when Cambodia is compared with countries with similar income levels and, importantly, with Singapore. Figure 8. Wages are rising but outlays on maintenance Figure 9. Wage outlays rose sharply in 2009, largely and operations are trending down due to increases for defense and security staff (in percent of current spending) (in percent of GDP) 50 5.0 45 4.0 2.1 40 3.0 1.7 1.6 1.9 0.8 1.0 1.7 1.6 35 1.8 2.0 30 2.5 1.0 2.2 2.1 1.8 1.8 1.6 1.6 1.7 1.4 25 0.0 2001 2002 2003 2004 2005 2006 2007 2008 2009e 2001 2002 2003 2004 2005 2006 2007 2008 2009e Operating cost (excl. wage)/Current Spending (%) Civil adm wage Defense/security wage Wage/current spending (%) Sources: MEF and Bank staff estimates Sources: MEF and Bank staff estimates x Given the size of employment in Cambodia and its budgetary space, wage outlays for staff in areas other than defense and security are also smaller than in other countries with similar levels of per-capita GDP (Figure 11). x The share of the public wage bill accounted for by defense and security is oversized. The recent rise – although coming on the heels of reductions as a share of total outlays since the early 2000s – has limited room for further wage increases for civilian employees. Figure 10. The number of public employees other than Figure 11. … as is the civilian wage bill in defense and security is small … (in percent of GDP, 2008 or latest available) (civil employees, in percent of the population, 2008) Mongolia 8.0 Malaysia 7.0 Philippines China 6.0 Timor-Leste 5.0 Vietnam 4.0 South Korea Thailand 3.0 Lao PDR 2.0 Indonesia 1.0 Singapore PNG 0.0 Cambodia 0 1 2 3 4 5 Source: World Bank analysis Source: World Bank analysis x The share of wages for education in the total wage bill has declined since 2007 ( x Figure 12). Education wages have decreased to 45 percent of total wage outlays, but employees in the education sector have increased to 59 percent of all total civilian civil servants (2009). This means the average wage educational employees is less than the national average civil service 19 civilian wage. This raises the question whether teachers’ compensation has been properly prioritized (Table 6). x The wage structure is very compressed. The average ratio of the highest to the lowest civil service wage (compression ratio) is about 3.5 in Cambodia, whereas it is about 5 or 6 in the lower-middle-income countries of Eastern Europe and about 10 in the advanced economies.18 There may be some room in the pay structure to strengthen incentives for better performance, a topic to which we will return below. As in all countries at Cambodia’s level of development, however, meaningful decompression is hard to achieve. And the recent increases in military wage rates further expanded their share of the overall wage bill, limiting any scope for adjustments in the non-military portion. x Wage outlays are dominated by allowances, that is components that have little or nothing to do with the skill and responsibility requirements of the position being held by the person receiving that remuneration, and which are assigned through non-transparent procedures that generate resentment among many staff. The base wage amounts to only 44 percent of the total remuneration of civil servants, compared with as much as 90 percent in advanced countries. Figure 12. The share of wages for education has Figure 13. Capital spending is larger than in declined since 2007 countries with similar levels of income per capita (in percent of total) (in percent of GDP) 100 12.0 Mongolia 10.0 Vietnam Capital Expenditures (% of GDP) 80 8.0 Cambodia 60 Urganda 6.0 Kenya 40 Nepal India 4.0 20 2.0 Togo Bangladesh Indonesia 0 0.0 2001 2004 2007 2009e 0 500 1,000 1,500 2,000 US$ GNI per capita Defense Security Health Education Source: MEF Source: World Development Indicators, latest available year. Note: Capital spending adjusted as discussed in text. x Increases in total remuneration over one's career are not reliably enough linked to performance relative to one's formal job. x There is significant reliance on donor financing to cover wage bill costs for a subset of public administration positions which needs to be phased out. The large salary supplements funded by donors are mostly not reflected in the official data. At the end of 2009, the government terminated the performance-based salary supplement (the Merit-based Pay 18 The compression ratio for a category is defined as the average wage in that category relative to the average wage of the lowest category. See Nunberg, Barbara (1995) and De Tommaso (1997) for detailed discussions. 20 Initiative) due to concerns that it promoted unequal pay for equal work.19 The off-budget scheme was jointly financed by donors and the government and covered only 2 percent of permanent civil servants at the time of its termination. That scheme was replaced by the so-called Priority Operating Cost (POC), financed solely by donors. Although the levels of top-ups under the new scheme are lower than under the previous one and the government is not involved in its funding, issues of inequality remain. As a result, this is just a short-term fix, whereas the issue still needs a permanent solution to ensure that equal jobs result in equal pay, as supported by the public administration reform agenda. Table 6. The average wage for education has risen but more slowly than the average civil service wage (in thousands of riels per month unless otherwise indicated) 2003 2004 2005 2006 2007 2008 2009 Education wage 172 170 177 204 300 320 323 Overall wage 145 147 167 189 246 325 433 Civilian wage 161 162 179 209 331 394 421 Share of education in the total number of civilian 50 51 52 53 56 56 59 civil servants (in percent) Share of education in the total civilian wage bill 53 54 51 52 50 45 45 (in percent) Sources: Ministry of Economy and Finance and World Bank staff estimates. Implementation of pay and employment reforms is necessary to enhance accountability and improving public service delivery. One viable option involves looking at a possibility of piloting a decompressing wage structure to reduce the gap between public and private sector wages for similar qualifications, responsibilities, efficiency, and results. This can be done by integrating donor-financed salary supplement expenditures (at present about $20 million a year) and government-financed wage outlays, and then gradually utilize saving from other areas to phase out donor financing and absorb these costs into the budget. Reform needs to include a substantial increase in the base wage as a share of the total wage (i.e., an increase in the component of salary that is linked to skills and responsibility requirements of the position being held by the person receiving that remuneration) and linking salary progressions over time with performance. With sufficiently well managed human resources, narrowing the gap between the public and private sector wages have already proven successful. A number of public, autonomous, and semi- autonomous institutions including those under the special operating agency arrangements, namely the Phnom Penh Water Supply Authority, Electricity of Cambodia, the National Bank, public higher education institutions, and many public health centers and hospitals, are among those that have markedly improved their performance. See Chapters 4 and 5 for specific initiative compensation reforms toward improving the basic services. Capital Spending Thanks to the fiscal stimulus, capital spending by the general government surged to 7.3 percent of GDP in 2009 up from 4.9 percent in 2008.20 Increased disbursements of donor-supported investment 19 The MBPI scheme provided an average top-up salary supplement of $300 a month for regular staff but as high as a few thousand dollars a month for high skilled recipients (doctorate level). 20 These data on capital spending by the government are adjusted. The adjustment deducts from the “externally financed projects” the portion estimated to be accounted for recurrent spending. Consult Box 2 for details. 21 projects accounted for two-thirds of the increase. Even before the surge in capital spending in 2009 due to the fiscal stimulus, government capital outlays in Cambodia – including spending financed by generous donor support – were larger than in countries with similar levels of income per capita ( Figure 13). The figure is slightly higher – by about 0.3 percent of GDP – if private outlays on capital goods are included (see subsequent sections). Proper classification of donor-supported investment projects into current and capital expenditures is essential. At present, donor-financed investment projects are classified fully as capital outlays although they have a sizable current component. Fixing this will ensure that the government has a clear picture of the size of capital outlays and the amount of the needed maintenance. Box 2. Recording Expenditures of Donor-Funded Projects and Programs The general government spending tables capture the public sector’s outlays on donor-financed projects but record all outlays as “externally financed capital expenditures.” An analysis of these donor-funded projects reveals that genuine capital expenditures range from about 95 percent in infrastructure projects to about one- third in education and health projects. Based on this, it is estimated that only about 70 to 80 percent of the expenditures of the donor-financed projects should be classified as capital spending. As a result, government capital expenditures have been adjusted down from an average of 6.3 percent of GDP in the government presentation to 5.3 percent (Table 4). Functional Classification of Spending Reflecting government priorities, spending increased unevenly across functional categories during the 2000s. The most salient features of the functional classification of government spending are: Table 7. Spending across most functional categories rose (in percent of GDP) Summary Functions 2001 2002 2005 2007 2008 2009e Core government 5.4 5.4 3.2 4.1 4.0 7.4 Economic services 2.4 3.3 2.9 3.1 3.8 4.7 o/w Agriculture 0.5 0.5 0.2 0.4 0.4 0.6 Transport 1.6 1.8 1.4 1.5 1.8 2.4 Rural development 1.1 0.7 0.5 0.7 0.6 0.6 Social services 5.7 5.7 4.5 4.8 4.1 4.7 o/w Health 1.5 1.5 2.2 2.1 2.0 2.5 Education 1.6 2.4 1.7 1.9 1.6 2.1 Other 3.0 3.2 2.3 2.3 3.9 2.1 Total 17.7 18.3 13.4 14.9 16.3 19.6 Source: Ministry of Economy and Finance. x Allocations to the “core government” (the general administration and the judiciary, defense, and security functions) rose by more than those to other functional categories between 2001 and 2009 (Table 7 and Table 8).21 The increase was especially sharp in 2009, reflecting the hike in wages 21 For Cambodia, core government functions include the Royal Palace, the National Assembly, the Senate, the Council of Ministers, the Constitutional Council, the State Secretariat of Public Functions, the Council for the Development of Cambodia, the Ministry of National Assembly-Senate Relations and Inspection, the Ministry of Foreign Affairs and International Cooperation, the Ministry of Economy and Finance, the National Election Committee, the National Audit Authority, the Ministry of Justice, the Ministry of Defense, and the Ministry of Interior. 22 for defense and security staff. As a result, spending on the core government functions accounted for 38 percent of general government outlays in 2009, up from 25 percent in previous years. x In contrast with most other functional sectors, the core government is almost fully financed from domestic resources. Donor support has financed 6 percent or less of the total allocation in this component every year of the past decade. x Spending on transportation increased by 0.8 percent of GDP from 2001 to 2009, driven by a threefold increase in the dollar value of external support. As a result, the density of the transportation network expanded during the 2000s and its quality improved substantially. x General government spending on health increased by 1 percent of GDP at the start of the decade to 2.5 percent by 2009. About half of government health outlays are funded by donors. Nonetheless, government spending on health amounts to just one-third of overall health spending in the country, leaving households to cope with the remaining costs (Chapter 4). x Public education outlays rose by 0.5 percent of GDP since the start of the decade, tripling in dollar terms because of larger domestic allocations, which now cover more than 80 percent of the total. Education is second only to the core government in terms of the share of spending financed from the country’s own revenues. Nonetheless, education outlays relative to GDP are lower than in other countries, including those with similar levels of per capita income (Figure 14). Table 8. Foreign aid finances more than half of Figure 14. Government education outlays are low government spending in transportation, agriculture, compared with other countries and health (in percent of GDP) (in percent and percent of GDP) Financing, average 5.0 Change, 2001-2009, 2008-09, in percent in percent of of total 4.0 gov't GDP spending External Domestic 3.0 Core government 2.0 7.2 6.1 93.9 2.0 Transport 0.8 3.3 69.5 30.5 Health 1.0 4.4 51.6 48.4 Education 0.5 1.7 20.1 79.9 1.0 Agriculture 0.1 0.2 67.2 32.8 Rural development -0.6 -3.2 43.1 56.9 0.0 Environment 26.7 73.3 Cambodia Nepal Thailand Vietnam Singapore Sources: MEF, CDC, and World Bank staff estimates. Sources: 2010 WDI, ADB Indicators, and World Bank staff estimates. x The allocations for agriculture and rural development rose by a modest amount between 2007 and 2009. A large but declining share of these allocations is financed by foreign donors, especially for capital projects in irrigation. The government spends less on agriculture than other countries with similar income levels (see Chapter 3). There has been progress on decentralization and de-concentration. The process of decentralization and de-concentration (D&D) began in 2002 when the central government and donors began transferring 23 funds to communes and sangkats.22 An organic law on D&D was adopted in 2008, paving the way for establishing additional sub-national administrations at the district and provincial levels as well. While the central government has not yet transferred any funds to these other sub-national levels, donors, by contrast, do provide direct support to them. The total allocations to the communes, sangkats, and district and province levels amount to only about 3 percent of general government spending. Transfers to communes and sangkats from the central government’s own resources via its Commune and Sangkat Fund (CSF) amount to about 2 percent of domestic revenues, and due to the limited absorption capacity of these sub-national administrations, the government has proposed capping this allocation at 4 percent (Table 9). Table 9. Transfers from the central government to Figure 15. Schools, and rural roads account for half sub-national levels are small of private spending on public goods (In percent of general government spending) (In millions of U.S. dollars, average annual 2005-2009) Irrigation/ Pagodas Others, $11 $17 agriculture Fund Transfer 2007 2008 2009 $3 , 20% 32% 5% CSF via the Treasury 1.7 1.4 1.3 Commune Donors 0.5 0.8 0.7 Government - - - District Donors 0.3 0.3 0.3 Government - - - Province School/edu, Donors 0.7 0.9 0.6 $12 , 22% Rural roads $11 Total 3.2 3.4 3.0 21% Sources: NCDD and World Bank staff estimates. Source: Survey of private spending on public goods carried out for this report in July and August 2010. Private Spending on Public Goods Private spending on public goods plays a major role at the commune level in Cambodia. A recent survey of commune-level spending commissioned for this report suggests that political parties, domestic companies, and private individuals provide grants of about $30 million (0.3 percent of GDP) for communal programs, mostly roads, schools, and irrigation (Figure 15). These grants are supplemented by support from the same donors to the communes of as much as $25 million a year for restoring or building pagodas. Although the amount of private spending on public goods is relatively modest in aggregate, most communes consider it to be vital as it often amounts to at least half of the total transfer from the central government via the CSF. However, a majority of commune officials interviewed for the survey reported that the execution of this private spending is not as efficient as that of the expenditures financed by the government. Political parties provide almost two-thirds of the private spending on public goods at the commune level (Figure 16). This support still appears to reflect actions by individuals and members of the party working groups assigned to each constituency. 22 Municipalities and provincial capitals are divided into sangkats. 24 Private spending on public goods varies greatly among communes and does not necessarily correspond with the priorities of the commune’s development plans or investment programs (Table 10). Finding a way for communes to coordinate better with private donors mirrors the challenge that the central government faces with international donors – more work is needed to ensure that strategic plans, budgets, and donor support are aligned at all levels. Figure 16. Political parties provide the bulk of Table 10. Communes’ priorities for development are contributions yet to be aligned with domestic donor support (in millions of U.S. dollars, average 2005-2009) (rank based on survey) Company, Others, $6 , Political Priority Ranking $2 , 3% 11% parties Areas of developments (Mean score of 1-4 Community, $32 scale) 59% $3 , 5% Road improvement 3.8 Irrigation system 3.5 Agricultural extension service 3.4 Hygiene and clean water 3.2 Health centre development 3.1 Charity/ School development 3.1 individual $12 Electricity infrastructure 2.9 22% Pagoda development 2.2 Source: Survey of private spending on public goods carried Source: Survey of private spending on public goods carried out for this report in July and August 2010. out for this report in July and August 2010. Quality of Outcomes of Government Spending The quality of the outcomes affected by public spending in Cambodia is in line with that in other countries with similar levels of per capita income. We reached this assessment by comparing a number of indicators, including life expectancy, child mortality, primary school completion rate, and the ratio of girls to boys in primary and secondary education in Cambodia with other countries with similar levels of per capita income (Table 11). However, Cambodia aspires to achieve middle-income status, which means that it needs to review its existing pay and compensation policies to ensure appropriate performance incentives. 25 Table 11. The quality of public outcomes is similar to that in countries with similar per-capita income Life expectancy at Mortality rate, under- Primary completion Ratio of girls to boys in birth, total (years) 5 (per 1,000) rate, total (% of primary and secondary relevant age group) education (%) 2000 2008 Change 2000 2008 Change 2000 2008 Change 2000 2008 Change Cambodia 56.9 61.0 4.1 106.4 89.5 16.9 47.2 85.0 37.8 81.7 89.7 7.9 Nepal 61.7 66.7 5.0 85.4 51.4 34.0 65.8 -- -- 77.1 -- -- Uganda 46.2 52.7 6.4 157.8 134.9 22.9 -- 56.1 -- 92.8 98.9 6.2 Bangladesh 61.3 66.1 4.8 91.3 54.4 36.9 59.5 57.5 -1.9 -- 106.1 -- Togo 59.7 62.5 2.8 121.7 97.5 24.2 63.2 61.3 -1.9 -- Lao PDR 61.0 65.0 4.0 86.3 61.3 25.1 69.3 74.7 5.4 81.3 87.2 5.9 Kenya 52.8 54.2 1.4 127.7 126.6 1.1 -- 79.5 -- 97.6 95.9 -1.7 Vietnam 71.9 74.4 2.4 30.0 13.6 16.4 96.4 -- -- 93.1 -- -- India 61.3 63.7 2.4 93.5 68.8 24.8 72.4 93.6 21.2 78.9 92.2 13.3 Ethiopia 51.4 55.2 3.8 148.0 108.5 39.5 23.0 52.1 29.1 65.1 85.2 20.1 Source: World Development Indicators Conclusion and Policy Recommendations This chapter has analyzed the level and composition of government spending and its alignment with donor support and the government’s strategic directions. In the short-term, the government is encouraged to consider the following recommendations: x Bring all government spending on budget, including externally financed investment projects, donor-financed technical cooperation, and donor top-ups of civil service wages. x Improve the quality and widen the scope of government finance statistics, including by recording all government expenditures, both domestically and externally financed, in government databases and budget documents. At present, budget documents include only an aggregate estimate of externally financed spending derived from data from the Public Investment Program. x Properly classify current and capital spending and financing, beginning with reclassifying externally financed investment projects, which at present include a sizable current component. For an integrated and comprehensive general government budget, externally-financed spending needs to be fully and properly classified using the government chart of accounts. (Also see Chapter 6.) The authorities are also invited to begin work on measures that will bring medium-term results: x Ensure that adequate operations and maintenance outlays will be included in future budgets for every capital project now implemented with no adverse effects on fiscal sustainability. The Medium-Term Expenditure Framework (MTEF) and Strategic Budget Plans (BSP) are the logical place to begin this exercise. The NDSP and the annual budget laws should also reflect this requirement. x Rationalize technical cooperation activities by moving toward sector-wide approaches and joint programming between the government and donors. Initial steps should involve the consolidation of the numbers of independent project management units and project implementation units to increase effectiveness and efficiency. 26 Chapter 3. Government Spending on Agriculture  A. Introduction Agriculture is a high priority for the Cambodian government as it is one of the core sources of the country’s economic growth and export earnings. Starting from a low base, government spending on agriculture rose to 1.3 percent of GDP in 2007 and to 1.5 percent in 2009, mainly through significant increases in outlays on irrigation and rural roads. This period of rising public spending on agriculture overlapped with robust annual growth in agricultural output. Although insufficient time has elapsed to assess with confidence whether there is a causal relationship between the level of public spending and agricultural growth in Cambodia, the evidence provided in this chapter shows that public expenditures may indeed have had a significant impact in increasing and sustaining agricultural growth, along with conducive policy environment. The analysis confirms that Government budget spending has been generally going for the provision of public goods and services, which have high returns on investments. However, the analysis also shows that there is a good potential for further efficiency gains on resources spent. This chapter focuses on areas with highest potential efficiency gains to increase the value for money from investments in core public goods and services such as extension, irrigation and rural roads. This is a first attempt to carry out such an analysis in Cambodia, and even in the Greater Mekong sub- region. Based on extensive data gathering and surveys, this chapter analyzes the efficiency and effectiveness of agricultural sector expenditures in Cambodia and assesses various options for increasing the impact of government expenditures on agricultural growth. Other challenges include an excessive focus on rehabilitating primary irrigation infrastructure and a neglect of secondary and tertiary systems, a lack of maintenance of irrigation and rural roads, and the slow pace of developing or adopting new technologies to reduce future maintenance costs. There is also a need to better prioritize agricultural and related infrastructure expenditures, both by type and by geographic location, to maximize their impact on growth. The rest of this chapter is organized as follows. Section B discusses recent developments in the agriculture sector in Cambodia. Section C presents an overview of government spending during the 2000s. The budget process and its relationship to sectoral development strategies are discussed in Section D. Section E presents our analysis of the efficiency and effectiveness of government spending using cost- benefit analysis of selected public investments. Section F presents our conclusions and policy recommendations. B. Overview of Agriculture, Irrigation, and Rural Roads Agriculture continues to be the mainstay of Cambodia’s economy and has recently emerged as an important source of growth. The sector expanded by 4.4 percent per year on average in the decade between 1998 and 2008 and by 5.4 percent in 2009 (Table 12). Although slower than the overall GDP growth rate, this pace of expansion has been higher than the 3.8 percent average for developing East Asia. Rice is the dominant crop, whose production accounts for about one-fourth of agriculture GDP and half of crop value added. Rice is grown on 84 percent of the country’s cultivated land and by about 85 percent of the rural residents. Despite the predominance of rice, there has been substantial diversification during the last decade. The production of non-traditional food and industrial crops such as maize, cassava, and soy beans has risen rapidly, partly in response to increasing demand from livestock 27 growers. Other crops that have significantly increased in output include tobacco, peanuts, sesame, and rubber. Table 12. Growth in agriculture has been strong (percent change, constant prices) 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Agriculture 4.4 1.7 -2.5 10.5 -0.9 15.7 5.5 5.0 5.7 5.4 Crops 8.6 12.1 3.0 -4.8 21.9 -2.3 27.6 8.2 6.6 5.8 Livestock & Poultry -1.1 1.0 -1.1 5.7 3.9 5.6 8.2 3.7 3.8 5.0 Fisheries 5.0 5.9 0.6 1.7 -1.7 5.6 3.8 0.8 6.5 6.0 Forestry & Logging -13.0 -13.8 -4.3 -3.0 0.8 5.1 7.0 1.1 0.9 1.1 Industry 31.2 11.2 17.1 12.0 16.6 12.7 18.3 8.4 4.0 -9.5 Services 8.9 11.1 7.7 5.9 13.2 13.1 10.1 10.1 9.0 2.3 GDP 10.7 7.4 6.6 8.5 10.3 13.3 10.8 10.2 6.7 0.1 Sources: National Institute of Cambodia (July 2011) The agriculture sector has a considerable potential to contribute to economic growth and exports given Cambodia’s good climate, substantial land area, and surface water . With appropriate technical and institutional support from the public sector, there is significant potential to produce sustainable increases in crop yields by increased use of agricultural inputs, improved farming techniques, and improvements in irrigation infrastructure. The challenge will be for Cambodia to seize this opportunity in the agriculture sector. Constraints to agricultural development need to be overcome if the sector is to drive stronger growth and help to reduce poverty. Investment in rural infrastructure has been low, which has not conducive to the development of internationally competitive agriculture. Cambodia has the least developed road network in the region with the smallest percentage of paved roads. More than 70 percent of the country’s unpaved rural roads are barely accessible or impassable during the rainy season, which isolates some parts of the country and impedes domestic and external agricultural trade. Despite demonstrated increases in agricultural productivity when access to water is assured, irrigation is still limited in its effective coverage. Currently there are around 2,400 irrigation schemes in Cambodia, covering a total of 1.046 million season hectares (582,085 hectares of wet season irrigation and 245,288 hectares of dry season irrigation). According to the Ministry of Water Resources and Meteorology (MOWRAM), that is at most 40 percent of the total arable cropping land. The untapped agricultural capacity can be attributed to volatile rainfall patterns, with the attendant higher risks to the use of purchased inputs for small farmers, and the reduced capacity to undertake crop activities during the dry season. C. The Level and Composition of Government Spending on Agriculture Three government ministries focus on agriculture and rural infrastructure in Cambodia. These are the Ministry of Agriculture, Fisheries, and Forestry (MAFF), the Ministry of Water Resources and Meteorology (MOWRAM), and the Ministry of Rural Development (MRD). MAFF is responsible for supporting dry land agriculture, upland and lowland crops, rice production and marketing and for agricultural support services relating to research, extension, and farmer education. It is also responsible for water resource-related functions for irrigated agriculture, fisheries development, and forest and catchment programs. It is organized into four thematic areas: general agriculture, rubber development, forestry, and fisheries. MOWRAM is responsible for the development and management of all water resources. It has two technical departments. The Engineering Department is responsible for the rehabilitation or construction of irrigation/water resources infrastructure, as well as its operation and maintenance. The Irrigated Agriculture Department deals with institutional issues related to the development and maintenance of irrigation infrastructure, including the establishment of Farmer Water Users Committees (FWUCs). The mandate of the MRD covers other rural infrastructure, including rural 28 roads, water supply and sanitation, ethnic development, community development, and rural economic development. Overall government spending on agriculture – both current and capital and financed both from own resources and by donors – amounted to 1.5 percent of GDP in 2009, somewhat higher than in 2007. Recurrent spending accounts for the bulk of the total, although the share of capital spending has been increasing (Table 13). External donor support for agriculture has declined from about 42 percent of total government agriculture spending in 2007 to about one-third in 2009 (Figure 18). Cambodia’s public spending on agriculture is relatively low when compared to countries with similar levels of per capita GDP. Out of ten countries, only four have equal or lower ratios of public spending on agriculture to GDP, and only Bangladesh has a lower ratio to agriculture value added. Public expenditure on R&D in Cambodia is relatively low at 0.1 percent of agricultural GDP compared with 0.2 to 0.6 percent elsewhere in Asia. This points that quality of public spending is at least as important as relative levels of spending. While public spending has in no doubt contributed to high rates of agricultural growth in Cambodia and Bangladesh, both countries also share similarities in terms of conducive and relatively distortions free policy environment, which have contributed to the efficiency of public spending and encouraged private sector contribution to sector performance. Table 13. Strong growth in agriculture, modest government spending Government spending on agriculture, 2004, in percent of: Growth, agriculture Public R&D in percent GDP, in percent, Agriculture value added, of GDP, 2000 Relative to … agriculture 1990-2005 in percent of GDP, 2003-05 GDP GDP Bangladesh 1 1.7 0.44 3.2 20.0 China 6 11.3 0.43 3.7 12.8 India 1 11.7 0.34 2.5 17.6 Indonesia 1 3.1 0.21 2.3 15 Philippines 3 5.1 0.41 2.4 14.9 Sri Lanka 3 5.3 0.64 1.4 15.9 Thailand 3 11.7 n.a. 1.8 10.1 Kenya 2 4.1 2.68 2.6 25.9 Uganda 5 4.1 0.5 3.9 25.6 Bolivia 2 6.8 n.a. 2.9 12.9 Cambodia 1.5 2.9 0.1 5.3 (2008) 32.4 Sources: World Bank staff estimates for Cambodia, and World Development Indicators for the other countries. Recurrent expenditures have been growing faster than GDP. The MAFF as a service provider has the largest recurrent budget (CR78 billion in 2009), followed by the MRD (54.6 billion) and then by MOWRAM (CR31.4 billion). Annual recurrent expenditure growth rates between 2000 and 2008 for the MRD (21.4 percent) and MOWRAM (15.1 percent) exceeded the growth rates of the total recurrent budget rates in real terms, while MAFF recurrent spending grew by 9.2 percent (Figure 19). 29 Figure 17. Overall government spending on Figure 18. … while donor support has declined agriculture has increased … (in percent of government outlays on agriculture) (in percent of GDP, 2007-09) 250,000 500,000 36% 1.5% 32% 450,000 42% 200,000 400,000 2007 1.4% 2007 350,000 2008 2008 150,000 300,000 1.3% 2009 2009 250,000 100,000 200,000 150,000 50,000 100,000 50,000 0 0 Agriculture Irrigation Rural roads TOTAL MAFF MOWRAM MRD TOTAL Sources: Ministry of Economy and Finance TOFE Tables and Source: CDC database and DIC. MAFF Department of Finance Within the recurrent budget, the share of salaries and wages is low in all three ministries. Wages account for one-third of the MAFF recurrent budget, 17 percent of MOWRAM’s, and 15 percent of the MRD’s. Operating costs are significantly higher in the MAFF, while in the MOWRAM and the MRD operating budgets are primarily used for operations and maintenance. Two key issues emerge. First, maintenance allocations have been inadequate given the extent of capital spending on both irrigation and rural roads. Second, the low share of wages is a concern. Staff recruitment, retention, and motivation have become a major problem, making it impossible to deliver planned work programs and effective and efficient public services. Figure 19. Recurrent spending has grown steadily in Figure 20. Government capital spending has increased all three ministries substantially (in billions of riels) (in millions of riels and in percent of government spending) 35 90 300,000 Billion Riels Billion Riels MRD, Real recurrent expenditure 80 30 MOWRAM, Real recurrent expenditure 250,000 64% 70 68% 2007 25 MAFF, Real reccurent expenditure 60 200,000 2008 20 2009 50 150,000 59% 15 40 30 100,000 10 20 5 50,000 10 0 0 0 MAFF MOWRAM MRD TOTAL Source: MEF TOFE Tables Source: MEF TOFE Tables 30 Irrigation and rural roads have been given priority in the small but expanding capital budget financed by own revenues. Capital expenditure in the MOWRAM accounts for more than half of the total capital outlays on agriculture, and the rest is spent in the MRD. The capital budget of MAFF has been negligible since 2004. Only 3 percent of MAFF’s budget is allocated to extension and only 5 percent is dedicated to agricultural research. The shortfalls in domestic operating budgets for extension and research have been covered by donors, who allocate about 31 percent of their funding to these functional areas. These are also the areas where donor dependency is the highest within the agriculture sector. During 2007-2009, the share of government funding in total funding was only 3.5 percent for extension and 41 percent for agricultural research. Donor funding has been significantly lower in the areas of rural roads and irrigation investments where it has concentrated mainly on new or rehabilitated infrastructure. In total, external donors financed about 57 percent of public spending on agriculture during 2007- 2009. This heavy dependence on donor fund in agriculture in general has created a number of issues, one of which is whether this funding is fiscally sustainable and another is the proliferation of project implementation units, which have attracted key skills and personnel away from the government to donor projects. Meanwhile, ministries are focused on projects rather than on policy and budget strategy, while the priorities of the donors may not be always aligned with national development goals. D. Budget Processes and Performance The budget processes within the three ministries are governed by systems put in place by the Ministry of Economy and Finance (MEF). The three-year rolling Medium-term Expenditure Framework (MTEF) forecasts and allocates resource flows to priority sectors within the government based on priorities set out in high-level planning documents such as the NSDP and in sectoral strategies. Budget Strategic Plans (BSPs), introduced under the Public Financial Management Reform Program (PFMRP) in 2007, are critical policy and budget planning tools for ministries. The BSP is a performance-oriented instrument used by line ministries to prepare their medium-term and annual budgets and their program budgets. Despite some progress, the links between the NSDP, the MTEF/BSP, and the annual budget remain weak. The recurrent, capital and donor budgets are not yet fully integrated (externally financed spending is only partially captured in the government budget and is not classified using the government chart of accounts). The BSPs should become the central planning tool for the implementation of the Strategy for Agriculture and Water (SAW) and should fully cover both domestically financed and donor- supported spending. At present the effectiveness of the BSPs is constrained by limited capacity and by overlaps with the project implementation units (PMUs). The desirability of moving away from projects towards a more integrated and programmatic approach is well understood by international donors. The constraints to improving budgeting are systemic and are to some extent beyond the control of the line ministries. The MEF is responsible for public finance management, and the PFMRP is working on improving the system, though the multi-year PFMRP timetable recognizes that these reforms will take some time to implement. Agricultural productivity, irrigation, and rural roads are among the top priorities for national and sector strategies. The NSDP identifies two key actions, namely increasing public investment in agriculture and increasing yields through extension services. The Agricultural Sector Strategic Development Plan (ASSDP) stresses the importance of agricultural commercialization. The SAW, which replaced the ASSDP, provides a strategic framework for the implementation of the NSDP. 31 Government spending on agriculture and irrigation has deviated from the projected costing in the government’s strategy documents. The government has spent less on research and extension than was projected in the ASSDP and the SAW and more on forestry, ensuring market access for agricultural products, and irrigation. The actual amounts that it spent on the institutional framework and forestry reform were in line with the ASSDP projections. The government’s investment on irrigation was substantially higher than had been projected in 2005. E. Efficiency and Effectiveness of Government Spending in Agriculture In this section, the performance of the government in achieving its objectives is measured in terms of output, increased incomes for the immediate beneficiaries of public expenditure, and reduced expenditures by the same beneficiaries. Most of the SAW objectives for research, extension, and irrigation are aimed at increasing the incomes of farmer beneficiaries. In the case of rural roads, the immediate objectives of SAW are to increase the incomes of farmers and to reduce prices and costs of transport. Also included are the wider objectives of promoting growth and reducing poverty. The levels of output from public expenditure have been sustained since 2002, whereas the costs per unit of output have been variable. According to government data, outputs of both irrigation and rural roads have been on the rise, with the share of rehabilitated irrigation schemes and rural roads rose to more than 5 percent of the total. There are large variations in the annual unit cost figures, but broadly unit costs in constant prices have remained largely unchanged for extension, have increased for irrigation, and have declined for rural roads. The factors that influence unit costs include poor reporting, possible errors in the recording of rural road maintenance, and increased prices and crop margins. The simple aggregation of data for a wide range of irrigation works may also have added to the problem. The limited available international evidence suggests that unit costs in Cambodia are high for extension and rural roads and low for irrigation (Figure 21 and Figure 22). Figure 21. The key outputs achieved have been Figure 22. But the record on costs has been mixed broadly on an upward path (actual and trend lines, constant 2009 prices) (in riels and in percent of GDP, 2002-2009) Extension contacts New irrigation land 180 Extension ($/farmer) Road rehabilitation 120,000 Road maintenance 1,200 160 Irrigation Rehab ($/hectare) Index of Cost per Standard Output Rural Road Rehab ($/km) 140 100,000 1,000 120 Farmers / area irrigated 80,000 800 100 km roads 60,000 600 80 60 40,000 400 40 20,000 200 20 0 0 0 2002 2003 2004 2005 2006 2007 2008 2009 2004 2005 2006 2007 2008 2009 Source: World Bank staff calculation. Source: World Bank staff calculation. The analysis of economic returns to public expenditures is based on benefit-cost ratios (BCRs). The BCR is the ratio of the net present value (NPV) of benefits to the NPV of costs that would have been calculated if economic appraisals had been done over the past decade. The BCRs were low (close to 1) in 2007 but rose to more than 2 after the commodity price increases since 2007 and despite irregular donor funding and poor coordination with research (Figure 23 and Figure 24). 32 Figure 23. BCRs for agriculture and rural roads are Figure 24. … but they are still below potential rising … 6 5 Actual Potential 4 Benefit Cost Ratio (2009) 3 2 1 0 Extension Irrigation Rural Roads Note. BCRs were calculated as discounted net benefits generated for every unit of expenditure incurred, which includes both capital and recurrent expenditure. This makes it possible to compare infrastructure sectors, such as rural roads, with those delivering public services, such as extension. The BCRs were calculated for each year, based on actual trend of unit costs of a year and expected benefit stream valued at market prices. The bar chart compares the BCRs for each year to potential BCRs that would have been theoretically achieved if the performance of public expenditure had met its stated targets or if there would have been a balance between rehabilitation and maintenance expenditures. For example, BCR for extension is calculated by assuming the average actual diffusion rate of 3 farmers per extension worker per visit, while potential BCR was calculated by assuming average diffusion rate of 5 farmers per extension worker per visit, which is a target set by Department of Extension of MAFF. Since 2007, public expenditures on agricultural extension have yielded high returns, while returns to irrigation investments have remained low despite the rise in prices since 2007. This is largely due to the disproportionate expenditure allocated to rehabilitating irrigation schemes that have limited or no maintenance operations. With better targeting of public funds, it should be possible to double the returns from public expenditures on irrigation. Public expenditures on rural roads have yielded acceptable and steadily growing returns throughout the last decade. If output targets remain stable and are achieved, spending on extension and rural roads would yield some of the highest returns of all government expenditures on agriculture. Agriculture BCRs have increased since 2007 after remaining around unity until 2006 in part reflecting rising benefits along with higher agriculture prices. This has made returns to extension investments as attractive as investments in roads, peaking at a BCR of 3 in 2008. While the average costs per farmer contacted by public extension officers in Cambodia are comparable to international norms, there are large differences in unit costs between government and NGO extension services, with the latter being significantly higher when including all local and international overhead costs. The results of beneficiary survey, which was carried out as part of PER analysis, show that farmers generally perceive the quality and technical knowledge of Department of Agricultural Extension (DAE) extension staff good, and expressed relatively high satisfaction rate with its services. Yet, the effectiveness of public extension spending has been relatively low when comparing reported diffusion rates with potential diffusion rates, with latter resulting a significant gap between actual and potential BCRs (3 vs 5). Public spending on agricultural extension in Cambodia has potential to continue to generate high rates of return. The current DAE extension model where extension officers train lead farmers, who in turn then train their peers, seem to be working fairly well. There is still a considerable scope to increase farm productivity levels, especially for wet season rice, which indicates that extension activities could continue to provide significant benefits over medium to longer term. While cost savings would certainly 33 have a positive effect on BCRs, the key variable that affects the performance of extension expenditures is the adoption rate per worker and the resulting increase in diffusion rates, as well as more intensive adoption of new technology packages by farmers. For example, PER analysis shows that in order to reach the effectiveness benchmark target of $1,500 per year in incremental increase in farm value added per extension worker would require the incremental increase of 100 new adopters per extension worker per year up from current average level of 50 farmers, and associated increase of farm value added of about 10 percent above baseline levels (ie extensive messages). Similar results could be achieved with only additional 20 new adopters per year per extension worker if farmers adopt a full package of farming technologies (ie a combination of improved seeds and inputs, and better farming techniques), which was estimated to increase incremental farm value added of about 50 percent above current levels (ie intensive messages). Increasing long-term performance of public extension activities requires improvements in its institutional functionalities. The main issue related to current performance of DAE extension services is insufficient operating budgets which results shorter and less frequent than optimal visits to farmers to help increase knowledge and spur the adoption of more new technologies, but also broader institutional issues. Achieving better balance between operating costs and recurrent costs, coupled with broader institutional changes which devolve more MAFF budget resources and staff to provincial and district levels would go long way to deliver extension messages along extensive margins by reaching larger number of farmers. Harmonization of extension messages and delivery approaches between various service providers (government, NGOs, private sector), and elimination of conflicting messages, would also improve value of money for extension spending by all. Finally, better integration of agricultural research and extension delivery functions will ensure that public spending on development of new crop varieties, and information on research trials on fertilizer and soil management techniques for specific agro-ecological conditions will reach farmers quickly and in easily accessible format, providing a principal basis for intensification of extension messages. Irrigation The BCRs for irrigation remained close to 1 throughout the 2000s, suggesting that irrigation is performing well below its potential. Since 2007 when rice prices have been rising, the BCRs for irrigation have stayed constant due to increasing unit costs (although they still compare favorably with international norms). This is a major concern, and MOWRAM needs to devote more attention to prioritizing its investment activities. The wide variability of BCRs among schemes shows that irrigation investments can be competitive even with the currently high unit costs if they are well selected and designed, fully implemented, and properly maintained. The actual BCRs from irrigation investments amount to less than half of their potential returns. The main reasons for this are incomplete rehabilitation and inadequate maintenance. It has been estimated that only about 70 percent of the command area could be actually irrigated in Cambodia, largely because of lack of rehabilitation. This in turn is the result of public funds being spent on primary canals at the expense of secondary and tertiary canals, although there is increasing recognition in MOWRAM of the need include latter as integral part public irrigation works investments. Meager funding for maintenance is another key reason for the limited effectiveness of Cambodia’s irrigation sector. It is likely that the lack of proper maintenance is reducing benefits by up to 10 percent each year. Therefore, we recommend that about 5 percent of current rehabilitation costs should be allocated to maintenance in each subsequent year. This means that about one-third of total expenditures on irrigation would be devoted to maintenance. There is also the possibility of using new technologies in irrigation investments, which would reduce the need for costly maintenance works in the future. The government has been encouraging the creation of Farmer Water User Communities (FWUCs) to 34 undertake local maintenance works. While this is valid in principle, it may take time before FWUCs are in the position to carry such financial cost. It is therefore unrealistic to place full responsibility for maintenance on the FWUCs, meaning that the optimum share of public funding to be devoted to maintenance is about one-third. Rural Roads Expenditures on rural roads have yielded very large benefits in recent years, with BCRs of higher than 2 since 2005 and reaching 3 in 2008 and 2009. Between 100 and 600km of rural roads have been rehabilitated (an average of about 1 percent each year) and 300 to 1,000km have benefited from periodic or routine maintenance (an average of about 2.3 percent each year). The BCRs for rural roads depend heavily on the level of maintenance efforts. The recent increase in expenditures on maintenance will further increase the effectiveness of these investments, although the share of the budget assigned to maintenance is still far below the optimum level of about 45 percent of total roads spending. The costs of building and maintain rural roads are increasing with the growing shortage of laterite. New materials and techniques will need to be found and piloted, such as bamboo concrete and engineered earth methods if the progress in increasing rural access is to be sustained, but this will require higher upfront capital outlays and more capacity building. Improving rural road infrastructure can significantly reduce poverty. An inadequate road system is a major bottleneck to economic development. Investment in rural roads yields high returns in terms of poverty reduction in developing countries, and this has been demonstrated to be the case in Cambodia as well. It is estimated that road rehabilitation can result in a 30 percent reduction in the costs to farmers of transporting their surplus production to market and may increase the farm gate price for those farmers who do not transport their surplus product to market themselves. Improving rural roads helps rural populations to gain access to key services and increases their opportunities to generate non-farm income. F. Conclusion and Policy Recommendations The effectiveness of public spending on agriculture could be substantially increased in Cambodia. The challenges identified in this chapter include lack of continuity in public funding, especially for O&M costs, for extension and agricultural research and the need to address deeper institutional issues which could limit long-term effectiveness of these expenditures. Other challenges include the excessive focus on rehabilitating primary irrigation infrastructure and a neglect of secondary and tertiary systems, a neglect of maintenance of irrigation and rural roads, and the slow pace of the development of new technologies for rural roads in response to the declining supplies of laterite. There is a need to better prioritize expenditures to increase the effectiveness and efficiency of public spending for the three sectors. The recent increases in agriculture prices have boosted BCRs to levels. While increasing agricultural spending on public goods and services is generally desirable, we believe that there is little room to do this relative to GDP within the existing budget envelope in the short-term, unless offsetting reductions are made in other areas. Should such room be found, the prime candidates for increased allocations should be extension, irrigation, and rural roads. The recommendations in this chapter would therefore focus primarily on improving the overall efficiency and effectiveness of existing agriculture spending levels. 35 On the basis of this chapter’s analysis, we offer the following recommendations for improving the performance of public expenditures on agriculture, irrigation, and rural roads: x Reallocate more budget resources on agricultural extension. Government funding for extension is low as a share of its agriculture budget and, as a result, these services are dependent on donor support. Public extension has potential to deliver high returns in Cambodia. There is an opportunity to significantly increase government by reallocating it from functional areas of lesser value for money. However, any spending increases, both from government and donors, should be accompanied by improvements of institutional functionality of MAFF extension systems and establishment of monitoring systems to measure the effectiveness of expenditures on these functions. Better harmonization of service delivery standards between various extension providers (government, NGOs and private sector) and elimination of conflicting messages would also improve the efficiency of extension spending for all. x Do not ignore funding for agricultural research. Analysis of efficiency of agricultural research spending in Cambodia was constrained by data limitations, but international evidence shows that agricultural research can have very high rates of return. While increased public spending for agricultural research is justified, it should come with institutional changes which clarify the specific functions of the country’s various research institutions and by consolidating laboratory capacities in various government units in order to reduce overlap and waste of resources. This can be done by establishing joint planning and evaluation systems by the government, donors and private sector to identify research needs and resource requirements. Technical expertise of donors could play an important role in building more effective research capacity in Cambodia. Better integration of national agricultural research institutions with extension services could further increase public spending efficiency on research, as well as extension. x Prepare for climate change. Related to two above recommendations, long-term efficiency of public spending on extension and research could be further improved by focussing more research efforts on development of new crop varietals and dissemination of knowledge on improved water storage and soil moisture preservation at farm-level – i.e. “low regret” investments that combine increased agricultural productivity with climate adaptation and mitigation against negative effects from reduced and more variable growing seasons. x Increase the effectiveness of irrigation investments. Public funding for new and rehabilitated irrigation schemes needs to be extended to secondary canals and associated infrastructure. The rehabilitation of tertiary canals using public funds is also justified provided that it includes arrangements the costs of operation and maintenance to be recovered from the end users. There also a need for more selectivity in funding irrigation schemes. x Increase the share of the budget designated for maintenance of irrigation and rural roads. In the case of irrigation, the optimal level of maintenance expenditure should be about 33 percent of total investment, and in the rural roads sector, an average of 45 percent is required for periodic and routine maintenance combined. This level of maintenance should ensure that the next major rehabilitation is not required for at least 10 years. x A longer-term measure to improve the efficiency of spending on rural roads would require investigation of economic viability of new road rehabilitation and maintenance technologies. New technologies to consider include the use of bamboo reinforced concrete (for busy roads) and engineered earth. The government could usefully invest in or create incentives for others to invest in promoting skills in these technologies. 36 A key weakness of public spending in the three line ministries is the absence of coordination between policies, programs, and the budget process. This weakness is compounded by duplication of programs and unrealistic budget estimations with no identified sources of funding. We make the following recommendations for improving government expenditure planning and management: x Use the BSPs within each ministry as the key policy and budget planning tool to allocate and manage all resources to the sectors. The government should consider using the BSPs as the central planning tool for the implementation of the Strategy for Agriculture and Water (SAW) and applying them to both domestically financed budget allocations and donor allocations. x Reinforce the annual budget process by requiring MAFF, MOWRAM, and the MRD to assign “hard” and enforceable budgets to each department through the BSPs. The current system of partial program budgets should be absorbed into this comprehensive approach. x Use the Treasury system to record expenditure on a departmental basis. This would require a more comprehensive financial reporting (currently called an expanded version of the TOFE) and greater devolution of responsibility for expenditures to budget entities. x Introduce an annual review and planning process for the annual and medium-term budgets that includes both the government and its development partners that puts BSPs at the heart of the process and that integrates the SAW into government planning. The latter will require the coordination and integration of all the budget planning and management instruments that have been put in place by the government under the PFMRP. The donors, in consultation with the government, are preparing to map the way forward, and this should be done as a joint government/donor activity. The SAW initiatives are an opportunity to establish systems for regular and informative appraisal, monitoring, and evaluation of expenditures and for these to be fed back into future budgeting decisions. 37 Chapter 4. More Health for the Money A. Introduction Cambodia has made remarkable gains in improving the health of its population in the last decade. Reaching the government goal of universal health coverage however will require increased allocations to some components of health outlays. Given overall constraints to the fiscal envelope, such increases need to be accommodated predominantly by reallocation within the existing health budget by bolstering the efficiency and improving the quality of health spending. Since 2000, overall spending on health has increased, but out-of-pocket spending outlays continue to predominate. This puts a burden on the poor and risks making catastrophic care too expensive for people at all income levels. Recurrent government health spending has also risen as a share of total public outlays, but allocating any further resources – including for wage increases – will depend on available fiscal space. An increase – an even the introduction – of the payroll healthcare tax would discourage formal sector activity and thus would not be a recommended option for further resource mobilization. Investing in health and nutrition is essential for strengthening human capital. Targeted increases in certain categories of health allocations are possible, as discussed in the report, but need to be accompanied by improved efficiency of existing spending and financed at least in the near to medium term from savings achieved within the sector. First, the efficiency of expenditures could be enhanced by better purchasing, particularly for pharmaceuticals, medical equipment, and supplies. Second, budget allocations for central departments, hospitals, and provinces should be based on transparent formulas and clearly agreed outputs, and mechanisms for budgeting and expenditure tracking should be strengthened at operational district and facility levels. Third, existing health financing initiatives – including health equity funds for the poor, performance-based contracting for operational districts and hospitals, and midwifery incentives – should be increasingly integrated into national institutions and budgets. Fourth, the shift toward semi-autonomous service delivery arrangements for hospitals and some health districts should be accompanied by measures to limit increases in user fees, and to establish appropriate financing arrangements to ensure quality and continued access by the poor and near-poor. Fifth, performance management and monitoring need to be strengthened. Finally, given that the licensed and unlicensed private sector continues to be the predominant source of health care, the government should strengthen the regulation of private medical practices and pharmacies and introduce appropriate regulations for dual practice. The rest of the chapter is organized as follows. Section B discusses in brief the health sector in Cambodia. Section C reviews trends and composition in health expenditures. Section D discusses efficiency and equity of health expenditures in Cambodia. Section E offers recommendations. The Health Sector in Cambodia Health is a priority sector for the government as articulated in the government Rectangular Strategy. Interventions in support of the strategy include the construction of referral hospitals and health centers and the provision of free health care for the poor, and continued prevention and treatment programs for communicable diseases. Other interventions include the promotion of maternal and child health, and the adoption and enforcement of health laws and regulations to ensure the quality and safety of health services, medicines, and food. The Strategy’s policy statements are further elaborated in the Second Health Strategic Plan (HSP2) for 2008-2015. The Plan focuses on the development of health systems to improve health outcomes, and has established five sector-wide strategies to achieve these goals: (1) health service delivery; (2) health care financing; (3) human resources for health; (4) health information systems; and (5) health system governance. National health financing strategies and policies 38 are further specified in the National Charter on Health Financing (1996) and the Strategic Framework for Health Financing (2008-2015). Cambodia is on track to meet many of its health MDGs, but challenges remain in the areas of child nutrition and neonatal mortality. Overall life expectancy rose from 52 years in 1998 to 63 years in 2008 for men and from 56 years to 67 years for women over the same period (Table 14). Both infant and under-5 mortality have declined since 2005 largely due to improved socioeconomic conditions, higher vaccination coverage, and greater access to curative care (Figure 25). Nonetheless, this general progress masks large disparities within the population. For example, child mortality in 2005 was three times higher for Cambodians from the lowest wealth quintile than for those in the highest (CDHS 2005). Table 14. Cambodia is on track to meet many health MDGs Goal Measure 2000 2005 2015 2010 Recent data Assessment DHS DHS target target Reduce Children under-5 124 83 65 85 54 (2010) On track child mortality rate per mortality 1,000 live births Children under-1 41 70 90 85 91 (2008) On track immunized against measles Improve Maternal 437 472 250 350 206 On track maternal mortality ratio per (revised (revised (2010 health 100,000 live NSDP) NSDP) CDHS) births Combat Prevalence of - - 135 214 693 (2009) Not on track HIV/AIDs, smear-positive malaria and TB per 100,000 other population diseases People living with - 0.6% 1.80% 2.00% 0.7% (2008) On track HIV, 15-49 years (achieved) old Malaria severe - - 0.10% 0.25% 0.35% Not on track case fatality rate (2009) Sources: CDHS, CDC, UNDP, MOH Health Information System, Note: CDHS 2010. High rates of child malnutrition pose a risk to Cambodia’s long-term economic development. Malnutrition persists because of child feeding practices, limited access to water and sanitation, food insecurity, and lack of access to health care and to treatment for acute malnutrition. Stunting in children under the age of 5 declined from 49 percent 2000 to 40 percent in 2010, but the percentage of underweight children was little changed during 2005-2010 at 28 percent, while wasting rose from 8 percent to 11 percent (Figure 26). Malnutrition is high across income quintiles, while wasting is the highest among the urban poor. Child malnutrition diminishes cognitive development and reduces adult productivity. The causes and consequences of malnutrition are multisectoral, and thus require a coordinated government response (MOH 2008). Maternal mortality has declined and is on track to meet the MDG, due in part to significant increases in the coverage and use of maternal health services, but is still high. The maternal mortality ratio was estimated to have been 472 deaths per 100,000 live births in 2005, and has declined to an estimated 206 deaths per 100,000 live births in 2010 (CDHS 2010). Fertility rates declined from 4.9 in 1995 to 2.9 in 2009, while modern contraceptive prevalence increased relatively slowly to 35 percent in 2010, short of the NSDP target of 40 percent for 2010. Births at public health facilities increased 39 dramatically from 22 percent of the total in 2005 to 54 percent in 2010 (Figure 27). The quality of maternal health care remains a concern in both the public and private sectors, however. Access to emergency obstetric care is also limited, particularly in rural areas, due to shortages of trained personnel and equipment to deal with delivery complications or to perform cesarean sections. Figure 25. Infant and child mortality are down Figure 26. High rates of malnutrition persist sharply, neonatal only modestly (in percent) (deaths per thousand births) Sources: CDHS 200, 2005, 2010 Sources: CDHS 2005 Figure 27. Facilities deliveries have increased Figure 28. Private providers are most frequently significantly, with slower progress on contraception consulted for curative care (in percent) (in percent) 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Urban Rural Sources: CDHS 2000, 2005, 2010 Sources: CSES 2007 Despite steady progress in reducing the incidence of disease and death rates, Cambodia faces a dual burden of both communicable and non communicable disease. Cambodia is among the countries hardest hit by the HIV/AIDS epidemic in the region, with 0.7 percent of adults affected. The burden of non-communicable diseases already exceeds that of communicable diseases. According to the World Health Organization (WHO), 832 deaths per 100,000 in 2004 in Cambodia were due to non- communicable diseases and 660 from communicable diseases. Twenty percent of adults in the poorest three quintiles say that they smoke cigarettes daily, compared to only 11 percent of adults in the richest quintile (CSES 2007). A significant increase in tobacco taxes would likely help discourage smoking – particularly among the poor where smoking is most widely spread – and raise additional revenues. 40 The use of curative care has increased in both the public and private sectors, but Cambodians prefer the private sector. About two-thirds of care episodes (excluding self-care) take place in the private sector. While outpatient consultation rates in the public sector have increased, the use of the licensed and unlicensed private sector has grown even faster. In urban areas, the major sources of care are private pharmacies, private clinics, unlicensed drug shops, and national hospitals. Home visits to providers and the use of unlicensed drug shops have declined over the past five years. In rural areas, unlicensed drug shops, visits to homes of trained health workers, and private pharmacies are the most commonly used sources of care (CSES 2004, 2009). Private health care provision has expanded rapidly since it was legalized in 1993. The private provision of health services includes private hospitals and clinics, individual practitioners (licensed and unlicensed), and various drug sellers and pharmacies (licensed and unlicensed). Official data probably substantially underestimate the actual number of unlicensed facilities and practitioners. The Ministry of Health (MOH) counted 169 private hospitals, clinics, and policlinics in 2009 with 758 inpatient beds. There were reportedly 2,300 private “consulting rooms,” of which 80 percent were registered. There are two major for-profit hospitals in Phnom Penh and one in Siem Reap. Private sector health training has grown since the establishment of the International University in Phnom Penh in 2002. The MOH is establishing licensing procedures for individual health practitioners by working with the various professional councils in Cambodia. The MOH is taking steps to strengthen the regulation of the private sector, but conflicts of interest due to dual practice remain a challenge and regulation of private pharmacies needs to be further strengthened. The ministry focuses on increasing the proportion of private facilities that are licensed, and is seeking to improve the reporting of health statistics by the private sector. There is still no regulation of prices or quality of private health care. Dual practice is tolerated but is not regulated or subject to clear codes of conduct. The pharmaceutical market has expanded rapidly since imports were legalized in 1993. According to the MOH, there were 797 licensed pharmacies and 1,111 unlicensed drugstores in Cambodia in 2009. The number of unlicensed pharmaceutical retailers has been reduced by improved regulation but remains high. The Essential Drugs Bureau (EDB) is charged with private sector regulation, including tackling medicine counterfeiting. However, it does not have enough staff to test medicines and has no power to confiscate fraudulent goods or to punish traffickers. Trends and Composition of Health Expenditures Government health spending amounts to only one-fifth of overall health outlays in Cambodia but as a share of GDP is broadly similar to that in countries with comparable per-capita incomes. Government spending on health rose from about 0.8 percent of GDP in 2000 to about 1.1 percent in 2009, and increased further to about 1.4 percent of GDP in 2010. External support to health also increased rapidly in the past decade, and remains higher than government outlays (Figure 29 and Figure 30).23 As a share of overall government spending, health expenditures rose to more than 11 percent in 2009, above the regional average but below the NSDP target of 13 percent. Government health spending remains centralized. The MOH controls about 70 percent of expenditures, despite the government’s intention to decentralize spending. Most of the remaining health spending is delegated to provincial governors, who manage purchases and cash advances on behalf of health facilities – with about 70 percent of the provincial health budget spent on utilities, fuel, and other 23 Estimates of total health spending and the percentage of out of pocket spending vary, due to differing estimates of OOP spending (most based on the 2005 CDHS data) and differing data sources for external financing. The 2010 CDHS data, once available, will enable an updating of these estimates. 41 purchases on behalf of health facilities. Little information exists about the composition of spending below the provincial level, or on the actual value of commodities distributed to health facilities. A health Public Expenditure Tracking Survey therefore raised concerns about the proportion of health expenditures reaching service delivery level (World Bank 2008). The MOH worked to improve expenditure reporting and recently estimated that Provincial Health Department Offices (PHDO), health centers, provincial and district referral hospitals each receive about a third of provincial allocations. Data on investment outlays are not readily available. Figure 29. Government spending has increased Figure 30. … and as share of GDP and overall sharply in nominal terms … government spending (in thousands of U.S. dollars, 2000-2009) (in percent and in percent of GDP) 250,000 12% 11% 10% 200,000 9% 8% 150,000 7% 6% 100,000 5% 4% 3% 50,000 2% 1% 0 0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 % of GDP % of expenditure ODA Own Resources Sources: CDC Aid Effectiveness Report 2010 and MEF. Sources: MEF and World Bank staff estimates. Government expenditures on drugs and medical supplies are high and more than double wage outlays. Public expenditures on drugs are also substantially higher than the international average. Pharmaceutical spending more than tripled in nominal terms from 2004 to 2009, rising from 40 percent of the health budget in 2004 to 46 percent in 2009 (Figure 31). Much of the budget’s operating costs are for purchase of equipment and nonmedical supplies at the central and subnational levels, with the result that nearly 60 percent of the total health budget is spent on procurement. Wages and salaries, including all allowances, rose rapidly from a low of 13 percent of health spending in 2004 to 20 percent in 2009. Figure 31. Spending on drugs and medical supplies is Figure 32. Most budget is non-programmatic more than double wage outlays (in percent of government health spending) (in millions of U.S. dollars) 140 120 NCD HSS 1% 100 15% CDC 80 11% 60 40 20 0 2004 2005 2006 2007 2008 2009 RMNCH 10% Non- Drugs and medical supplies program Programmes of action (ADD & PAP) 63% Operating costs (excluding drugs) Salaries Source: Ministry of Economy and Finance. Source: MOH 2009 Annual Operational Plan. 42 About one-third of overall government health spending is allocated to specific programs and there is a continued mismatch between stated national priorities and program allocations. All three health goals and five sector-wide strategies of HSP2 are incorporated into the annual planning and budgeting process, yet their allocations remain limited (Figure 32). All other budgeted activities are consolidated as “non-program activities” and include cost items that are managed centrally such as wages, the procurement of drug and medical supplies, and other support activities. Only 1 percent of domestically financed spending in 2009 was programmed for non-communicable diseases (NCDs) and just 10 percent for reproductive, maternal, neonatal, and child health. However these figures understate actual budget resource allocation to NCDs and MCH, which are largely incorporated into non-program budget lines for health care in health centers and hospitals. A larger share of government health expenditures are financed from external assistance than own revenues, raising concerns about government health spending and about the harmonization of goals between the government and donors. Donor financing rose from about $67 million a year in 2000-2002 to more than $210 million in 2009.24 Bilateral donors accounted for about 40 percent of the external support for health in 2009, while multilateral agencies, the Global Fund for Malaria, Tuberculosis and Aids, and non-profit organizations each accounted for about 20 percent. Very large MCH expenditures are financed by non-official donations through NGO hospitals for children (which account for the majority of child inpatient and outpatient contacts in Cambodia) and are not recorded in AOPs. Donors have increased their efforts to coordinate their assistance following an OECD survey in 2006 that revealed coordination was not effective. For example, seven donors now provide support to the MOH under the Second Health Sector Support Program, either through pooled funds or discrete funding mechanisms using common procedures in line with the MOH’s Second Health Sector Strategic Plan. Other than government health spending, overall health spending consists of out-of-pocket payments – including official user fees – spending by health equity funds, and outlays financed by voluntary insurance. Consider each of these sources in turn. Figure 33. Better-off households spend more on Figure 34. User fees are an important source of healthcare, 2007 funding, 2009 (in U.S. dollars) (in thousands of U.S. dollars) 12,000 180 160 10,000 140 120 8,000 100 80 6,000 60 40 4,000 20 0 2,000 1 2 3 4 5 Wealth Quintile Health centers Public hospitals 0 Private hospitals Private clinics HCs PRHs Nat'l Hospitals Pharmacies and stores Home care Others other delivery para clinic IPD OPD Source: CSES 2007. Source: MOH Health Financing Report 2009. 24 These outlays are supplemented by direct support from NGOs – externally financed – to hospitals and health providers, which are not included in the public budget and expenditures. See Chapter 2 for details. 43 Out-of-pocket spending accounts for the bulk of overall health outlays. Three-fourths of private health spending is for private sector treatment and drugs. Although half of OOP payments are made by those in the wealthiest quintile, OOP spending represents a considerable burden on the poor (Figure 33). The poor spend a smaller percentage of their income on health than the rich, often because of severe cash constraints they forgo care or seek low-cost alternatives, but catastrophic spending affects all income categories. Nonetheless, catastrophic spending is lower in Cambodia than in some countries in the region, including China and Vietnam – countries that depend even more heavily on user fees (Wagstaff and others 1999; GiZ 2011). User fees – just one component of out-of-pocket-spending – are an important source of funding for public health facilities but are a barrier to access for the poor and near-poor. The 1996 Health Financing Charter introduced user fees at public facilities. Currently, 60 percent of the fees are retained by facilities to provide financial incentives for staff, 39 percent to be retained for the facility’s operational expenses, and 1 percent to be remitted to the government. Reported income from user fees rose from $1.3 million in 2001 to $15.9 million in 2009, with national hospitals generating about two-thirds of the total (Figure 34). User fees amount to 45 percent of the income of national hospitals but comprise only 5.8 percent of overall expenditure at the provincial level. Figure 35. User fees generate substantial incomes Figure 36. The source of funding for national (in percent of total income) hospitals varies, 2008-09 (in thousands of U.S. dollars) 100 $12,000 $10,000 80 $8,000 60 $6,000 40 $4,000 20 $2,000 $0 0 KBH KOH ADH NPH Calmette KSFH health centers referral hospitals central institutions CBHI HEFs User Fees User Fees CMS RGC BUDGET Sources: MOH Health Financing Report 2008 Sources: MOH Health Financing Report 2009, CMS distribution data from 2008. Notes: KBH = Kantha Bopha Hospitals; KOH = Kossamak; ADH = Ang Doung; NPH = National Pediatric Hospital; KSFH = Khmer Soviet Friendship Hospital. The poor are exempt from paying (official) user fees but they make other out of pocket payments. In 2009, 1.5 million user fee exemptions were granted, most of them at the health center level. More national hospitals are becoming Public Administrative Establishments, with the autonomy to increase revenues from user fees. As a result, they are becoming increasingly reluctant to grant fee exemptions unless the government explicitly reimburses them for the lost income. In contrast, the Kantha Bopha pediatric NGO hospitals, which received $3.5 million in budget transfers and $30 million in external funding in 2009, provides free services to the poor (Figure 36). Health Equity Funds have emerged as an important source of income for public facilities to treat those who can least afford it. Health Equity Funds (HEFs) covered about three-quarters of the poor in 44 2009 in 50 of Cambodia’s 77 operational districts. Most HEFs are operated by NGOs with donor funding, with a government financed scheme covering national hospitals and some operational districts. In 2008, provincial hospitals that worked with HEFs received 25 percent of their total income from the HEFs (Figure 35). Evidence suggests HEFs reduce health-related debt and are a more effective means of ensuring free services for the poor than fee exemptions managed by facilities (Bigdeli and Annear 2009). The government is introducing health insurance for private salaried workers and plans to introduce compulsory health insurance for civil servants. Pursuant to the 2002 Social Security Law, a 2007 sub-decree established the framework for the National Social Security Fund. An employment injury scheme financed via a 0.8 percent payroll tax was initiated in 2008. The scheme covered all firms with eight or more workers, and 400,000 workers from 1,200 enterprises had registered by 2009. The government plans to launch the private sector health insurance scheme in 2011 but has postponed the start of the civil service scheme. Many details of the schemes, including the benefit packages and the level of payroll contributions, have yet to be determined. However, the potential – and advisability – for raising further revenue through earmarked payroll contributions is constrained by the limited size of the formal sector, the likelihood of substantial evasion, and evidence from other countries that suggests payroll taxes can severely inhibit formal sector growth, especially in low-income countries. Moreover, such insurance schemes can be funded from general government revenues rather than by payroll taxes (Savedoff 2004). Voluntary Community-Based Health Insurance (CBHI) schemes have been launched in various parts of the country over the past decade, but their coverage remains limited. CBHI schemes target the non-poor or near-poor and have contributed to an increase in the use of health services, although enrollees tend to be those living close to health facilities or those with more frequent illness. About 100,000 people appear to be covered and turnover is high as even modest insurance premiums are often a burden. International experience suggests that CBHI and similar schemes for the informal sector rarely achieve sufficiently wide coverage without substantial government subsidies to stimulate enrollment, or by making the schemes quasi-compulsory (World Bank 2010). Scope for further increases in the level of government spending on health relative to GDP will depend on several factors. The most important factor is the availability of fiscal space. Given the level of per capita incomes and the level of informality in Cambodia, higher and new taxes do not appear to be a convincing option at this time. And while higher tobacco tax rates may be a useful tool to discourage smoking, new revenues from these taxes need not be spent on health. Improved tax collection and reprioritizing other government spending are the most promising options for increasing general revenues for health (see Chapter 2). A careful examination of the composition of non-program spending and increased transparency of sub-provincial spending similarly will create fiscal space through efficiency gains. Working with donors to better alignment their support with government priorities is critical. The Efficiency and Equity of Health Expenditures in Cambodia Health outcomes adjusted for incomes and the level of spending in Cambodia are close to the global average. Nonetheless, outcomes are substantially lower than most countries in developing East Asia – which also tend to have much higher incomes per capita (World Bank 2010). Efficiency and Equity of Health Service Provision Coverage and access to health care have increased. The MOH developed the Health Coverage Plan (HCP) in the early 1990s, leading to the current system of Health Operational Districts (ODs), which typically cover a population of 150,000-250,000 with a single district or provincial hospital and with 8-15 health centers. The number of operating health centers amounted to 984 in 2009. Access to public health facilities has increased significantly. Government spending on primary care tends to benefit the poor and 45 rural populations, whereas spending on provincial and national hospitals tends to benefit those that are better-off (Figure 37). The average distance people in the poorest quintile have to travel to access their nearest health center fell by 36 percent between 2004 and 2007 to 4.5 km, but distance remains a barrier to access to referral hospitals (Figure 38). Efficiency indicators related to hospitals appear to have improved, but allocations are still based primarily on the number of beds, with little regard to the total number of patients (Figure 39). Bed occupancy rates (BOR) for provincial referral hospitals (PRHs) averaged 64 percent in 2008, ranging from 24 percent in Kep PRH to 100 percent in others (such as Kirivong and Siem Reap). At national hospitals, the BOR averaged 84 percent. The average length of stay was about six days at both national and provincial hospitals, with some evidence of longer stays for HEF patients. Shifting to case-based payments rather than fee-for-service could reduce the incentive to overtreat and could help control costs. Payments can also be adjusted to induce incentives for quality improvement. Allocations to provinces appear to be driven by historical expenditures and direct negotiations, while needs-based or output-based budgeting is not yet introduced. And there are large variations in overall per capita allocations (Figure 40). At the same time, the allocation of external financing to provinces primarily reflects donor priorities. Although the Central Medical Store (CMS) has developed a logistics system for forecasting and distributing drugs according to needs, per capita allocations in 2008 ranged widely among provinces, from $1.28 in Phnom Penh to $12.40 for Pailin. Similarly, a comparison of vaccination coverage and out-patient usage rates with provincial per capita budget allocations shows a wide variation among provinces. The per-capita allocation of doctors and nurses across provinces also varies. Cambodia has 0.2 doctors and 1.1 nurses and midwives per 1,000 people with significant variation among provinces (Figure 41). While the MOH has made progress in posting a primary midwife to nearly all health centers, the number of secondary-trained personnel has decreased in the past decade. Cambodia’s primary health system relies on nurses for primary care, and the ratio of nurses and midwives per doctor is 5.4, which is higher than both the ASEAN (4.7) and the global (2.1) averages (Kanchanachitra and others 2011). Figure 37. Spending on primary care is more pro- Figure 38. Poor people have closer access to health poor than hospital spending centers than hospitals (in percent) (in kilometers) Sources: CSES 2007 and World Bank staff estimates. Sources: MEF (2009), CMS (2008), CDC (2009). 46 Improving Procurement of Pharmaceutical and Medical Supplies More efficient purchasing of pharmaceuticals and medical supplies could generate savings equal to a third of the 2010 health budget, or nearly 0.4 percent of GDP. Government spending on pharmaceuticals has increased substantially since 2005, but stock-outs of essential drugs at facilities persist (Figure 42). A review of a sample of CMS invoices suggests the MOH is paying on average six times the international reference prices for essential drugs. In 2010, the government spent about $65 million on pharmaceuticals and $14 million on medical supplies, from a total health budget of $150 million.25 More efficient purchasing could generate up to $50 million a year (0.4 percent of GDP) in savings – which could be used to further improve drug availability and to finance other critical health priorities. About 96 percent of drugs used at the provincial level are purchased by the MOH. Only 3.5 percent of drugs are purchased out of the provincial government budget allocation, and the rest from retained user fee income. Figure 39. Budget allocations are linked to beds but Figure 40. Provincial allocations vary not to hospital outputs (in U.S. dollars per capita) (in U.S. dollars) 25 $800 $700 20 $600 $500 15 $400 $300 10 $200 $100 5 $0 0 Ratanakiri Takeo Kratie Kandal Banteay Meanchey Pailin Kep Pursat Battambang Siem Reap Kampot Kg Cham Svay Rieng Stung Treng Kg Thom Mondulkiri Kg Chhnang Prey Veng Preah Vihear Phnom Penh Odar Meanchey Kg Speu Koh Kong Sihanoukville Gov't Budget/bed per month Gov't Budget/discharge External CMS Provincial Budget Sources: MEF budget data (2009), CMS distribution data Source: CSES 2007. (2008), CDC data (2009). Drugs and supplies are purchased by the MOH and distributed by Central Medical Stores, with the MOH Department of Drugs, Food, Medical Devices, and Cosmetics (DDF) responsible for quality monitoring. The DDF determines the essential medicines list in partnership with national programs and is responsible for quality monitoring. Procurement decisions meanwhile rest with the Prequalification Evaluation and Award Committee (PREAC), subject to approval from the MEF’s Department of Public Procurement. The procurement unit in the MOH currently reports directly to the Minister of Health. PREAC procurement has not yet been harmonized with international best practice. Medicines and equipment supplied via the HSSP2 and the Global Fund for AIDS, Malaria, and Tuberculosis (GFATM) are procured through separate international bidding, using donor-specific procedures. A review of sample CMS invoices suggests that medicines and vaccines purchased through international agency procedures are close to international reference prices. 25 No data were available on costs of medical equipment and non-medical supplies purchased at central and provincial level – representing an additional $20 million expenditure in 2010 – but further savings could likely be achieved. 47 Figure 41. The distribution of doctors and primary Figure 42. Stockouts persist despite rapid increases in midwives varies considerably spending for pharmaceuticals (per 100,000 of population) (in millions of U.S. dollars and in percent) 40 60 14 35 12 50 30 Expenditure USD million 25 10 Stockout percentage 40 20 8 15 30 10 6 5 20 4 0 Kratie Pursat Pailin Ratanakiri Takeo Modulkiri Siem Reap Kep Sihanouk Ville Battambang Svay Rieng Kompong Cham Udor Meanchey Kandal Preah Vihear Kompot Bantheay Meanchey Prey Veng Kompong Speu Kompong Thom Koh Kong Kompong Chhnange Steung Treng 10 2 0 0 2001 2002 2003 2004 2005 2006 2007 2008 Doctors Midwives (primary) Expenditure Reported stock-out at facilities Sources: CSES 2007. Sources: MEF budget data; MOH HIS. Improving efficiency and coordination of budgeting and planning process The budget strategic plan is the MOH’s input into the MEF’s medium-term expenditure framework. The BSP identifies medium-term objectives and indicators for health sector priorities. On the basis of the medium-term framework, the MEF issues a budget circular which guides the preparation of the annual budget request by the line ministries. The MOH BSP currently does not include subnational health spending, however, and has limited information on external financing. ODs, PHDs, and the MOH prepare comprehensive Annual Operational Plans (AOPs) as tools to implement the annual budget. The final MOH-level AOP is compiled from consolidating plans prepared by separate MOH departments, provinces, national programs, and hospitals. The PHDs are responsible for consolidating AOPs prepared by ODs and referral hospitals into provincial AOPS. The AOP are more complete than the government budget, combining capital and recurrent expenditures, own revenues and external partner financing, and using a program/sub-program/activity classification. The MEF budget preparation process, by contrast, relates to government recurrent expenditures, largely excludes external funding, and is based on economic expenditure categories. The BSP, annual budgets and AOPs should converge in terms of coverage, types of spending, and sources of financing through the further implementation of the Public Financial Management Reform Plan. The MOH has prepared a Financial Management Improvement Plan, which includes the key elements of budget reforms from PFMRP and identifies key planned activities and responsibilities. Consolidating demand-side and supply-side health financing initiatives Evidence suggests that Health Equity Funds have helped increase the use of health services, reduced health-related debt, and are more effective in these functions than fee exemptions (Tangchanroensanthien and others 2011; Annear 2010). Scaling up HEFs nationally by 2015 with the current benefit package would require over $16 million a year, including $2.3 million for identification of the poor (USAID and URC 2010). The benefit package could also be expanded to include essential public services for the non-poor: for example, subsidizing deliveries and emergency obstetric care would cost just over $1 million. Substantial synergies could be obtained by coordinating HEFs with the planned insurance schemes for formal sector workers and civil servants. The draft Master Plan for Social Health 48 Protection calls for establishing a Coordinating Committee at the Council of Ministers, and creating a single social health protection agency. A transition strategy could be to establish a semi-autonomous agency under the MOH to manage HEF and CBHI schemes, with policy coordination among HEF/CBHI and social health insurance schemes ensured by the national Coordinating Committee. In 2009, the government designated former NGO contracting districts as Special Operating Agencies (SOAs), giving them limited autonomy and making them eligible to receive performance- based Service Delivery Grants (SDGs). Ten NGO-contracted ODs and five provincial referral hospitals were initiated as SOAs in 2009, followed by a second wave of 15 new SOAs in 2010. The HSSP2 provides Service Delivery Grants on the basis of performance contracts between the MOH and the SOAs. The SOAs in turn sign performance contracts with health facilities. The overall ceiling for SDGs is based on a modified per capita formula, of which 80 percent is provided as a base allocation, 15 percent is allocated on the basis of the agency’s performance, and 5 percent on the basis of contract negotiations. Currently 80 percent or more of each SDG “base allocation” and performance allocation can be spent on staff incentives, in addition to a 60 percent share of revenue from user fees and HEFs. In addition to providing counterpart funding for the SDG grants, the MEF agreed to provide flexible program-based budgets (PBB) directly to the SOAs. From 2011, the program-budget allocations are released to the SOAs through the provincial treasuries, rather than through the MOH as was the practice earlier. The SOAs and SDGs face many challenges and are not yet fully aligned with government financing and civil service policies. Challenges include inadequate arrangements for performance monitoring by the MOH and the PHDs, a lack of clarity regarding rules for the use of service grants, and the late arrival of government funding for operating costs. While performance varies significantly among SOAs, monitoring data suggest that SOAs with few exceptions as a group have maintained and in some cases improved their performance on key indicators. The SDG incentive payments to staff depend on donor financing, and it remains unclear if RGC is willing or able to finance SDGs in their current form through the budget. Several evaluation activities are planned for SOAs and the SDGs in 2011, which should help further improve the approach. A key next step will be to integrate the performance monitoring, provider payments, and oversight mechanisms of SDGs with those of the HEFs. A payment of $15 to midwives introduced in 2007 for each delivery in health centres (and $10 in public hospitals) appears to have contributed to an increase in such deliveries. Payments to staff at central institutions and national hospitals amounted to $0.3 million (2009 data), and midwifery payments at provincial levels amounted to $1.6 million (2008). The latter is equivalent to 10 percent of the base wages for provincial health staff. RGC has committed to maintaining the midwifery incentives through 2015, after which they may be integrated into other demand- or supply-side financing schemes. The initiatives and incentives discussed above provide additional compensation and motivation to health workers but need to be integrated into a medium-term wage reform and an updated health financing strategy. Each initiative uses a different set of rules to distribute revenue to staff without guidelines on how much additional revenue staff can or should receive. This creates a risk that facilities increasingly will raise fees or seek to maximize revenue from costly services, making services less efficient and less accessible for the near-poor and informal sector. Health finance reforms and wage reforms should be linked to establish payment mechanisms that create appropriate incentives for quality, efficiency and equity for health facilities, while ensuring appropriate compensation and incentives for health workers. The Priority Operating Cost incentive scheme for service delivery staff could also be subsumed under the more integrated HEF incentive schemes. To be effective, higher and more decompressed salaries will require effective monitoring and appraisal of staff performance. 49 Conclusion and Policy Recommendations The government’s long-term objective is to achieve universal coverage with quality health care within the available fiscal space. Achieving such a goal will require substantial further improvements in the efficiency of health spending, health coverage, and the quality of service delivery. The main message of this chapter is that government could make significant progress toward its key long-term health objective through a combination of more efficient purchasing of health commodities; delegating more resources to service delivery level; consolidating various health financing initiatives to create appropriate incentives for service quality and equity; strengthening monitoring and evaluation of public services; and improved regulation of the private sector. The path to universal coverage will need to be guided by some key policy decisions. First, the government will need to clarify its long-term vision for the desired “mix” between the public and private sector for both primary care and the hospital sector. Second, decisions will need to be made regarding the balance between supply-side and demand-side financing (including whether to keep user fees relatively low with continued supply-side subsidies or to make a significant shift of resources toward demand-side financing). Third, the authorities will need to clarify the extent to which hospitals and other health services will be granted autonomy, and how to establish necessary financing mechanisms and regulations to ensure autonomous entities fulfill their social mandates. Fourth, the government needs to address challenges of health worker compensation and motivation by establishing clear guidelines for dual practice, and aligning and integrating various initiatives that seek to incentivize health workers with the evolving salary reform agenda. Achieving the long-term objective will require measures in the short- and medium-term to improve the efficiency of health spending and the budget processes. Substantial savings could be reaped from the following specific recommendations. In the short-term, increase the efficiency of purchasing for pharmaceuticals and medical supplies: x Improve the efficiency of pharmaceutical, equipment, and medical supply expenditure through more competitive procurement and international price benchmarking. Savings could reach $50 million a year, or one-third of government health spending, the equivalent to 0.4 percent of GDP. Use savings to further increase drug availability and to finance other priority programs, including steps toward universal health coverage. x Convert the Central Medical Stores to an autonomous Public Administrative Establishment, responsible for both procurement and distribution of pharmaceuticals and medical supplies. The reformed entity would operate on commercial principles with a social mandate, overseen by a board of directors, including representatives of MOH and MEF. The pharmaceutical budget would be delegated to service units, who would be required to purchase from the new agency except in exceptional circumstances (e.g., stockouts, or specialty items for national hospitals). The government is also invited to consider the following short-term measures to increase the efficiency and quality health service delivery: x Integrate and institutionalize performance-based grants into demand-side financing to improve performance of health facilities, consolidate existing incentive schemes for motivation of health workers, and link reform of these to medium term wage reforms. Update the Health Financing Charter, including updating rules for revenue sharing with health staff. 50 x Strengthen the monitoring and evaluation of Special Operating Agencies (SOAs) and Service Delivery Grants, and integrate with other payment systems for HEFs and CBHI. Consider whether SDGs could be: (i) integrated into the program-based budget transfer mechanism within the national system; and/or (ii) combined with HEF and other demand-side payments as an integrated performance-based transfer to health service delivery units. x A significant increase in tobacco taxes could help lower smoking rates and reduce the future burden of chronic disease, particularly for poor. As part of the medium-term PFM reform agenda, increase the equity and efficiency of the MOH resource allocation: x Delegate a larger share of the operational budgets to ODs and to facilities below the provincial level. Establish pilot budget entities at PHD and possibly operational district and facility levels, and/or make ODs and PRHs “budget centers,” to enable budget allocations and expenditure monitoring. x Deepen and extend program budgeting to provinces and consider ways to allocate more of the budget from non-program to program funding. Consider adapting the administrative level approach used by the Ministry of Education, Youth and Sport (e.g., primary care, secondary care), while retaining program codes linked to NSDP for central expenditures. x Further improve Budget Strategic Plans supplemented by the AOP process. Strengthen medium- term planning within the MOH through the three-year-rolling plan process linked with the Budget Strategic Plans, and expand BSPs to include subnational and external financing. x Establish transparent criteria to allocate budgets to provinces. Allocation formulas should include at a minimum population size, number of facilities, and possibly the poverty or health status of population served. Over the medium to long term, the authorities are invited to consider measures to ensure adequate social protection for the poor and reduce out of pocket spending: x Scale-up and integrate the health equity funds and various forms of voluntary and compulsory health insurance into a comprehensive demand-side financing regime. The introduction of social health insurance for private sector workers and civil servants should be done in a coordinated manner. Establishing appropriate national institutional arrangements for HEFs and CBHI is a critical priority. Insurance for informal sector workers will have to be subsidized to expand coverage. This measure is estimated to cost $16 million a year. x Ensure access of the poor and near-poor to hospital services. The shift toward semi-autonomous service delivery arrangements for hospitals should be accompanied by measures to limit increases in user fees, and to establish appropriate financing arrangements to ensure quality and continued access by the poor and near-poor. x Strengthen regulation of user fees and update Health Financing Charter. x Payroll taxes tend to discourage formalization of economic activity and result in substantial evasion, especially in countries with low levels of GDP per capita and already pervasive informality. The authorities should avoid further increases in payroll tax rates. 51 Chapter 5. Structure and Quality of Education Expenditures A. Introduction Cambodia has achieved substantial progress in improving access to primary and tertiary education. The focus now has to turn to access to early childhood education and secondary education and, much more importantly, to improving the quality of education outcomes and addressing issues about equity across the country’s regions. As in other fields, education in Cambodia remains highly dependent on external donor support, and substantial progress has been achieved aligning this support with government priorities. But at 2.3 percent of GDP, overall government spending on education – both from own resources and donor support – lags the levels in countries with similar levels of income per capita and most countries in East Asia. Private spending on education, as a result, is almost as large as the overall spending by the government. While such household involvement helps strengthen accountability, it also strains households financially. The prevalence of private higher education institutes calls for tighter regulation to ensure consistent standards. As in some other sectors, larger allocations for education may help improve the quality of outcomes. Given constraints on the overall fiscal space, however, the first steps need to focus on finding room within the sector. The large decline in enrollment in recent years offers scope for substantial saving if the authorities reallocate resources from schools and districts with declining enrollment. Additional resources, if the overall fiscal stance allows, could be used to rationalize teacher wages and offer incentives to better-performing teachers, expand early childhood education, and support a scholarship program for lower secondary education that has had a large positive impact in the past. The framework of planning and budgeting works well in general. There is a manageable set of clearly defined targets and a functioning system of programs that align budgetary and donor resources towards meeting the targets. A small set of “core breakthrough indicators” are closely monitored together with other indicators for which the Educational Management Information System (EMIS) collects and provides a high quality of timely data. There is scope for improving the program budget sub-program that provides school operating budget by streamlining procedures and providing school with greater autonomy for approved expenditure items and post-hoc fiduciary reviews on a sample basis. The capacity of the Ministry of Education, Youth and Sports (MOEYS) needs to be strengthened to process data for decisions related to policy planning and implementation. In particular, the Educational Strategic Plan (ESP) needs to provide a mechanism to revise policy actions on the basis of the level of achievements. This is particularly important from a regional perspective because there is substantial variation on some indicators and resources and policy attention needs to respond to those regional variations. The rest of the chapter is organized as follows. Section B describes the setting of education within the government’s Rectangular Strategy and presents the achievements during 2006-2010 on the basis of access, quality, and institutional development. Section C analyzes government education spending with a focus on program budgeting. Section D examines the issue of teachers and teacher wages and Section E focuses on textbooks. Household financing of education in Cambodia is explored in Section F. Section G discusses government spending on higher education. Section H presents the policy recommendations. 52 Achievements and Challenges of the Education Strategic Plan The Education Strategic Plan (ESP) for 2006-2010 covers three strategic areas with a small set of clearly defined indicators. The strategic areas of the ESP are access to education, quality of education and capacity. There has been solid progress for the strategic area of access for primary education and higher education; pre-school and secondary education now need increased policy attention. Progress has been most impressive in improving access to primary education, with the enrollment rate up to about 95 percent in 2010 from an already high 91.3 percent at the beginning of the ESP 2006-10. The enrollment target was exceeded for upper secondary and nearly met for higher education (Table 15). Progress has been slow for early childhood education (5- year olds) and for lower secondary education. Policy measures are likely to require attention to both the supply of school places and the affordability for parents. Future versions of ESP should provide analysis of the reasons why certain indicators, such as the secondary enrollment rate were not met and identify corrective measures. While the enrollment rate for primary education rose, the number of students enrolled fell by about 12.5 percent from 2003 to 2010. At the same time, upper secondary enrollment increased by about 50 percent, but coverage still remains low at less than 20 percent of the population. Cambodia could reap the benefit of this “demographic dividend” in terms of smaller class sizes for the primary level, but retraining primary teachers and adjusting the school physical plant for pre-school or for the secondary level would be demanding. Table 15. Access to primary and higher education services (in percent unless indicated otherwise) Actual 2005-06 Target 2009-10 Actual 2009-10 Target 2013-14 Total Female Total Female Total Female Total Female Enrollment, 5-year olds 27.3 27.9 50 50 39.8 40.5 60 60 Net Admission Rate 82.6 81.8 95 95 92.4 92.7 100 100 Primary NER* 91.3 89.7 96 96 94.8 94.6 98 98 Lower Second.NER* 31.3 30.4 50 50 32.6 34.6 52 52 Upper Second. NER* 11.3 9.9 18 17 19.4 19.4 24 24 Students in Higher Education 91,000 29,000 162,000 58,000 145,000 52,000 206,000 96,000 Source: Education Strategic Plan 2009-2013, MOEYS. *Net Enrollment Rate Early childhood education requires renewed policy attention as there are multiple economic, equity and efficiency benefits to investing in this level. Organized quality childcare benefits women, especially those with low incomes, enabling them to work outside the home. Children who have benefitted from early childhood education are also likely to do better in higher education levels. The evidence from Cambodia about the decline of cognitive ability of children due to the lack of quality childcare programs is stark and disturbing (Figure 43). The graphs show deterioration with age in TVIP (Peabody Test of Vocabulary in Images) and WJ (Woodcock-Johnson psycho-educational battery) scores, with the deterioration worse for children from lower income groups. 53 Figure 43. The cognitive ability of very young Cambodian children has declined Source: Naudeau, Martinez and Filmer, 2010. Demand constraints are a major barrier Figure 44. Predicted probability of school attendance to participation in secondary education. and enrollment as a function relative-ranking Opportunity costs of time spent in school Attendance from school visits Enrollment tend to increase significantly for children at 1 1 lower secondary level. One policy tool to 0.8 0.8 deal with this - the government’s scholarship program - has been very Probability Probability 0.6 0.6 successful at retaining children in school. A 0.4 0.4 rigorous impact evaluation conducted with 0.2 0.2 the help of World Bank indicates a large impact of the scholarship program that 0 -25 -15 -5 5 15 25 0 -25 -15 -5 5 15 25 provided scholarships ($45-$60 per year) to Relative ranking Relative ranking lower secondary students. Scholarship Recipients Non-recipients Recipients Non-recipients Source: Filmer and Schady, World Bank, 2009. recipients were 20 percent more likely than those who did not receive any scholarship to be enrolled in school (Figure 44). Policy attention is required to maintain the level of funding for the scholarship program and preserve the hard-won gains for children from poor families, particularly girls. Progress raising quality – the second strategic area of the ESP – has lagged improvements in access, with continued very high repetition and dropout rates. Only 38 percent of the children who enroll in grade 1 continue through grade 9. Pupil-teacher ratios have not moved much, and around 50 pupils per teacher makes for crowded classrooms that are not conducive to effective teaching and learning practices. There has been some improvement regarding the repetition rate, but repetition remains very high, especially in grade 1. Student assessment has yet to take firm roots in Cambodia, moreover. Cambodia has not so far taken part in an international assessment such as PISA or TIMSS, although the country might take part in a South-East Asia extension of the PASEC assessment for francophone Africa. 54 Table 16. Improving the Quality and Efficiency of Education Services Actual 2005-06 Target 2009-10 Actual 2009-10 Target 2013-14 Total Female Total Female Total Female Total Female Primary Pupil Teacher Ratio 50.8 45 49.2 45 Repetition Rate 11 9.8 2 2 8.9 7.8 5 5 Dropout Rate 11.6 11.9 6 5 8.3 7.9 5 5 Survival grade 1 to 6 49.3 48.9 75 76 61.7 63.8 NA NA Lower Secondary Pupil Teacher Ratio 31.7 28.5 32.2 NA Repetition Rate 2.5 1.7 1 1 2.3 1.5 1 1 Dropout Rate 22.8 24.9 13 14 18.8 19.4 12 14 Survival grade 1 to 9 26.3 24.3 54 54 37.2 37.9 NA NA Source: Education Strategic Plan 2009-2013, MOEYS. Student assessment carried out as part of the CESSP project shows encouraging results, but additional efforts are needed to bring student assessment into the mainstream. Two sets of results are available from a sample-based national assessment that forms part of the CESSP project (sample size varies between 6000 and 7000 and is nationally representative). The results indicate still a low absolute level of performance in grade 3 for the two tested subjects of Khmer language and mathematics, but performance improved strongly from 2005-06 to 2008-09. (The Khmer score rose from 40.4 percent to 54.1 percent on average). For grade 9, scores have remained stable at 68 percent for Khmer language and improved from 41.2 percent to 43.8 percent for mathematics. Detailed analysis of the changes in scores would help reveal whether the improvements are due to measurement effects or due to improvement in student abilities. If student abilities have improved, it would be very useful to attempt an analysis of the data to understand the causal factors. Student assessment is a better way to gauge the quality of an education system rather than reliance on the measurement of inputs and indicators such as repetition rates. Test scores are better presented in terms of proficiency scores, as in the case of PISA, where questions are weighted according to the level of difficulty so that information can be obtained from the entire distribution of test scores. Tests can also be used to enhance accountability and provide feedback to the teachers and school administrators. For this to happen, the government needs to upgrade the quality of testing and provide for regular universal testing rather than one based on a sample basis. A preliminary examination of quantitative indicators for institutional and capacity development – the third strategic area of the ESP – indicates a lack of progress from 2006 until 2010. Education expenditures declined from 18.5 percent of overall recurrent government spending in 2006 to 17 percent by 2009 and amounted to 2.5 percent of GDP in 2009 (Table 17). The level of spending on education leaves Cambodia behind most countries with similar levels of income per capita and most economies in East Asia. The earlier Priority Action Program (PAP) was replaced during the ESP2006-10 with program budgets (PB) as part of the government-wide reform of public financial management (see Chapter 6 for more details). The PB execution rate improved from 62.1 percent in 2007 to 90 percent in 2009. Dependence on development partners for technical assistance has declined by much less than targeted under the ESP 2006-10 as measured by total person-months. It is not, however, clear that this is the proper metric to capture MOEYS’s dependence on technical assistance. 55 Table 17. Institutional and Capacity Development for Educational Staff for Decentralization (in percent and in person months) Actual Target Actual Target 2006 2009 2009 2013 Education (in percent of government recurrent budget) 18.5 20.0 17.0 21.0 Program budget spending (in percent of total program budget, 2007) 62.1 100 89.9 95 Program budget implementing units audited internally (in percent) 42.4 45.0 16.7 NA Schools subject to financial annual audit (in percent) 9 5 1.1 NA Technical assistance (person months) 576 382.5 516 339 Source: Education Strategic Plan 2009-2013, MOEYS. The authorities have laid the groundwork for conducting impact evaluations of the program of scholarships for lower secondary students, and preliminary results indicate a very strong positive impact from scholarships to enrollment. But the results appear not to have been used to support an increase in outlays for scholarships. This program is an example where financing shifted from external partners to government’s own resources. The authorities should reconsider extending the program and increasing its reach, even with continued assistance from development partners. Education Expenditures: Patterns and Trends Education expenditures have increased in absolute amounts but declined relative to GDP in line with reduced donor support. General government spending on education – including both own- resources and donor support – fell from 3 percent of GDP in 2003 to 2.4 percent in 2009 (Figure 45, Table 18, Table 19). Education spending financed from own resources rose by 0.2 percent of GDP over the same period. Government outlays are below the 3.8 percent of GDP spent in developing East Asia, the 4.6 percent average for IDA countries, and the 4.8 percent world average. Together with large household spending on education, overall education outlays in Cambodia declined from about 6 percent of GDP in 2004 to about 5 percent in 2009 (see below for a discussion of household education spending). The government introduced a program- Figure 45. Government education spending fell relative to based budget in 2007 following the GDP, 2003-2009 earlier Priority Action Programs (PAP) (In percent of GDP) under the government-wide PFM reform. Education is one of the priority Donors as % of GDP sectors for the government, where the 3.50% budgeting process is being successfully Govt. as % modernized, with PB slated for an 3.00% of GDP increasing proportion of expenditures during the current ESP. 2.50% 1.41% The largest PB allocation in the past 1.26% 1.05% 0.60% 2.00% three years has been for quality 1.05% 1.01% 0.65% primary education and the trend is programmed to continue in the ESP 1.50% 2009-13. About half of the budget in sub- program 1.2 is for school operational 1.00% 1.74% grants that are used by schools to acquire 1.58% 1.47% 1.35% 1.50% 1.45% 1.57% didactic materials. Many PB programs are 0.50% within the capital budget that is financed mostly by development partners. For sub- 0.00% program 1.2, there are resources for 2003 2004 2005 2006 2007 2008 2009 construction of schools and sanitation Source: Education expenditures from ECORYS, 2011, and WDI facilities. 56 Table 18. Cambodia’s education expenditures are lower than in relevant comparators (in percent of GDP unless indicated otherwise) EAP ECA LAC MNA SAS SSA IDA FTI World Average 3.8 5.0 5.0 5.2 4.6 4.8 4.6 4.8 4.8 Median 3.5 5.1 4.9 5.5 3.8 4.6 4.2 4.6 4.8 Maximum 6.3 8.3 13.3 8.6 7.9 13.3 13.3 13.3 13.3 Minimum 1.6 2.6 2.4 2.7 2.6 1.4 1.4 1.4 1.2 Countries with data/All countries 12/35 33/58 19/36 9/21 5/8 24/48 38/78 24/37 104/209 Source: Education Strategic Plan 2009-2013, MOEYS. Note: EAP = East Asia and Pacific: ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MNA = Middle East and North Africa; SAS = Southeast Asia; SSA = Sub-Saharan Africa Several other PB allocations and issues stand out. For example, the program of scholarships for poor students (sub-program 1.4) was financed earlier via external support but has now become part of the government’s recurrent budget. Only half of the PB budget for textbook development (sub -program 5.3) was spent because of problems with the procurement of textbooks for the secondary level in 2008. Human resource development (sub-program 5.4) is largely accounted for by a program that awards teachers with bonus payments for teaching in remote or disadvantaged areas. The inclusion of a special program for strengthening governance (sub-program 5.5) is noteworthy, although implementation has been weak. The increased priority to expanding early childhood education and higher education is in line with government efforts to improve coverage. The expansion of the capital budget is largely due to the accelerated implementation of the EFA FTI-CF program after past procurement-related problems were resolved. Program 3 for youth and sports is slated to increase substantively as it starts from a very low base. Finally, the budget for sub-program 5.5 (construction of safety boxes) is slated to rise from CR1.2 billion in 2009 to CR18 billion in 2011. It will be important for the government to examine the actual implementation level so far under the last program to ensure alignment with the ESP priorities. Other than this item, the remaining PB allocations, and the accompanying documentation and justification are of a very high quality and reflect well on the planning and budgeting capacity of the MOEYS. The process for planning and budgeting for educational expenses is based on a solid footing with a close integration between the targets, policy measures and expenditures. The PB instrument appears to be working well in education, though there are individual PB elements that need to be made more efficient. One of the problems with regard to implementation in PB has been the lack of flexibility and autonomy provided to individual spending units that has led to low budget execution and subsequent reductions in the budget. Particularly with regard to school operating budgets, increased autonomy to schools with incentives to encourage timely execution would lead to greater efficiency. Equity would improve as the dependence on parental contributions would be reduced. 57 Table 19. Education Expenditures, 2009-13 (in millions of riels unless indicated otherwise) 2007 2008 2009 2010 2011 2012 2013 (A) Government budget 506,644 659,971 754,666 824,879 950,185 1,046,419 1,165,415 Wages 372,088 435,384 505,749 609,960 686,815 756,449 842,388 Non-wage 134,556 224,587 248,917 214,919 214,919 263,370 323,026 (B) Donor Support 355,193 274,384 260,535 314,462 408,585 412,672 417,252 Government total (A+B) 861,837 934,355 1,015,201 1,139,341 1,344,322 1,476,711 1,625,241 (In percent of GDP) 2.5 2.2 2.4 2.4 2.6 2.6 Program budget 78,272 92,587 111,745 162,272 226,550 265,601 304,132 In percent of recurrent 15.5 14.0 14.81 19.7 24.2 25.0 25.2 Selected items from PB: General Education 48,526 64,011 74,394 111,538 135,896 154,138 170,184 Early childhood education 434 641 686 6,553 7,130 7,700 8,267 Quality primary education 30,046 38,427 43,794 43,775 44,331 54,270 60,456 Secondary education 12,195 16,983 18,949 36,360 50,131 55,672 62,193 Scholarship poor students 3,409 4,001 6,438 8,230 9,280 10,342 11,497 Expansion non-formal 2,441 3,959 4,527 6,860 7,724 8,505 9,471 education Higher education 1,677 2,501 3,062 3,493 15,439 27,908 41,335 Youth and sport 531 1,237 1,496 2,417 2,996 3,390 3,884 development Strengthening governance 309 441 439 707 914 1327 1492 Source: ESP 2009-2013 for 2010-13; ECORYS, 2011 for 2007-2009 aggregates; MOEYS and MEF for PB budget 2007-09. Teachers and Teacher Wages The wage bill for education rose four-fold in nominal terms from the start of the decade and now accounts for three-fourths of education spending. As a share of GDP, the wage bill increased from 0.8 percent of GDP at the start of the decade to 1.1 percent in 2009, a pace higher than that of overall education outlays. Four-fifths of the increase is due to higher wage rates, with the rise concentrated in 2006-2008. Nonetheless, the average monthly teacher salary of about $115 is broadly equal to what a recent study termed “the living wage for garment workers.”26 About 68 percent of primary teachers and 50 percent of secondary teachers hold a second job after regular working hours to supplement their incomes.27 The second occupation is private tutoring for 42 percent of teachers at the primary level in urban areas (87 percent of lower secondary teachers across urban and rural locations). Private tutoring by public school teachers has serious equity implications and should be discouraged. A fifth of the increase in the wage bill was due to a rise in the number of teachers. The number of teachers grew from 71,000 in 2001-02 to 83,000 by 2009-10, but increases in student enrollment has meant that the student-teacher ratio has remained high (Table 20). There is a marked improvement in the qualification of teachers, however. The share of teachers with only a lower secondary qualification level fell from 66 percent in 2001-02 to 46 percent in 2009-10. Still, the share of teachers with tertiary degree was little changed at about 7 percent. 26 Chandararot, Kang and Liv Dannet. 2009. “Living Wage Survey for Cambodia’s Garment Industry,” Cambodia Institute of Development Studies, Cambodia. 27 Results from the PETS survey as described in World Bank, 2005. 58 Table 20. Teacher profiles and teacher wage bill, 2001/02 to 2009/10 (in percent and in U.S. dollars) 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 Non-teaching staff 11,141 11,884 13,266 14,591 15,643 16,667 17,180 17,340 17,135 Teaching staff 70,608 73,642 76,350 77,897 78,606 77,974 79,823 81,350 82,820 Age Distribution of Teachers (in % of total) 30 or younger 5.2 5.3 5.1 4.8 4.7 4.4 4.6 4.1 3.9 30 – 39 years 65.6 63.1 60.4 59.2 57.9 56.6 53.7 50.1 45.6 40 – 49 years 23.3 25.8 28.6 30.2 32.0 33.1 35.7 38.9 43.1 50 and older 5.9 5.8 5.9 5.8 5.4 5.9 6.1 6.9 7.3 Distribution of Teachers by Educational Qualification (in % of total) Primary 41 38 37 37 37 36 38 39 39 Lower Secondary 31 36 39 41 40 40 39 37 34 Upper Secondary 17 14 12 10 10 9 9 11 14 Tertiary 10 11 11 11 12 12 12 12 11 Average Annual Teacher Wage (in dollars) Nominal wage 455 551 564 577 629 786 1121 1285 1378 PPP adjusted 1443 1759 1837 1863 2014 2490 3412 3556 3895 Source: HRMIS Database as reported by ECORYS, 2011. Incentives for teachers to work in remote areas appear to have had an effect. The number of teachers in remote areas increased from 986 in 2001-02 to 1,750 in 2008-09, an increase that is four times as large as the rise in the overall number of teachers in the country. PB sub-program 5.4 directed about $21 million over the 5 years of the ESP to incentive payments for teachers to work in remote areas. The ESP sets a target of deploying 95 percent of new teacher graduates to under-staffed schools, disadvantaged and remote areas. It is difficult to expect high quality outcomes given the teachers’ low level of qualifications and relatively low pay. Higher wages in other fields, notably in the private sector, reduce the pool of people willing to become teachers. Low wages have been anecdotally linked to teacher absenteeism. For example, during surprise visits in 2006, about 15.6 percent of lower secondary teachers and 7 percent of primary school teachers were absent.28 The availability of teacher substitutes is rare and lack of teachers translates into little or no learning at school. Teacher pay needs further improvement in a framework of clear rules on qualification and teacher performance. Ultimately, improving the quality of teacher instruction holds the key to the country’s future prosperity. Allowing a further erosion of the qual ity of teachers will have detrimental long-term impact on Cambodia. Provision of Textbooks Textbooks are a crucial input for better education. This is especially true for students in classes of 50 or more. Textbook provision, however, is beset with three problems: an inadequate budget, issues related to the procurement and distribution of textbooks, and the content development of textbooks. With the average pupil-teacher ratio about 50, the individual interaction between teacher and students is minimal and textbooks are even more important to help students follow the pace of a lesson and study after class hours. Ideally, each student should have a set of textbooks for each subject that the student should be able to take home, to be replaced approximately every four years. These textbooks need to be in addition to 28 World Bank, 2008, Teaching in Cambodia. 59 sets of workbooks with exercises. There is a lot that needs to be done for this to become reality in Cambodia. Budgetary resources for textbooks are inadequate. During 1979-1995, the quantity of textbooks distributed failed to keep pace with rising student numbers and the situation was marked with serious deficiencies. The Basic Education Textbook Project (BETP) with support from the ADB and the UNICEF provided $28 million during 1996-2002 (about CR16 billion a year). Since 2002, the provision of textbooks was included in the PAP budget. During 2002-2006, the PAP budget allocated about CR13 billion a year. Only part was spent, from CR7 billion in 2003 to CR12 billion in 2006. Since 2007, textbooks are part of the PB program, and allocations rose to CR19 billion for 2011. There is considerable scope for improvement of the procurement and development of textbooks . A new curriculum was introduced in 2006, but only half of the students now have new textbooks and very few teachers have received the accompanying teacher manuals. Until recently, the textbook market was monopolized by the semi-public entity Publishing House (PDH). To ensure all children have textbooks and teachers have the appropriate manuals, the budget will need to be increased by about 25 percent from current levels.29 To improve the quality of provision and the budget execution levels, the recent move towards opening the textbook market for competition by moving away from a government-owned publishing house is correct and needs to be sustained. However, there are problems of textbook development and uncertainties that have so far restricted development of a robust set of suppliers – the government should explore policy options to deepen the market – possibly separating tasks of textbook development, printing and distribution. One option that has worked in several countries is the provision of a range of titles for each subject and grade, with teachers and school communities determining the appropriate selection of titles. Household Spending on Basic Education Households’ direct spending on education is substantial. Households in Cambodia spent approximately CR1.1 billion a year on education, or about 2.5 percent of GDP, the same amount as spent by the government both from own resources and from donor support (Table 21).30 The extent of household contributions suggests that parents have a greater stake in the quality and efficiency of the provision of educational services. At the same time, these large outlays place a heavy burden on households. The large inequality of spending by the different income quintiles implies substantial inequality in access and ultimately outcomes. Analysis based on the CSES 2008 suggests that the lowest quintile accounts for about 6 percent of expenditures, while the top quintile accounts for more than half.31 Therefore, within the same public school system some students have a qualitatively different experience because their parents supplement the public provision with their own resources – for instance, by buying textbooks not provided by the government or by getting private tutoring. This introduces a wedge in terms of educational outcomes which tends to exacerbate inter-generational social inequity: the children of the well-off are educated better and so will tend to find better labor market outcomes. Public education, as a result, amplifies rather than mitigates social inequity. 29 Fukao (2011a) provides details of the analysis underlying the estimated budget needs. 30 The Cambodia Socio-Economic Survey (CSES), 2009. 31 Fukao (2011b) provides the Lorenz analysis on the basis of which this claim is derived. 60 Table 21. Household education spending varies by quintiles, 2009 (in billions of riels) Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 All School Fees 19.22 15.84 27.99 75.34 250.86 389.26 Tuition 11.90 13.29 25.04 35.47 74.78 160.49 Textbooks 11.63 11.18 13.95 22.34 40.51 99.61 Other Supplies 12.72 11.09 15.08 23.33 37.77 99.99 Allowances 26.86 25.46 26.97 44.64 101.36 225.29 Transportation 11.70 11.18 11.88 31.02 74.61 140.40 Gifts to teachers 0.80 0.76 1.37 2.48 4.59 10.00 Total 94.70 88.83 121.99 234.14 583.29 1,122.96 Source: CSES 2009. Public-Private Partnership in Higher Education Enrollment in both public and private institutes of higher learning rose sharply over the last decade. Enrollment increased from 8,902 students in 1998 to 124,947 students in 2008, with some 34 public (mostly autonomous) higher education institutions (HEIs) and 55 private ones spread over more than 111 campuses. This extended higher education opportunities to larger parts of the population. A unique characteristic of higher education in Cambodia is its large private financing. Private spending pays for tuition at private universities and fees at public HEIs. By contrast, public spending on higher education has been small, mostly in the form of salaries and scholarships, amounting to 3.3 percent of the total public education spending in 2007 leaving Cambodia behind its neighbours (Table 22). Table 22. Total higher education expenditure per source of The real cost of higher education is funding declining relative to GDP per (in thousands of US dollars unless indicated otherwise) capita, in large part due to Public Private Total In % of GDP competition among the HEIs to HEIs HEIs attract students. The cost fell to Public Funding 8,150 -- 8,150 0.1 69 percent of GDP per capita by 2007 Private Funding 14,993 27,364 42,357 0.5 from 130-160 percent in 2000. Total 23,143 27,364 50,507 0.6 Source: Orivel (2009). The rapid increase in access to higher education due to private provision bodes well for Cambodia as it indicates the very high demand for education. To make the most of these private resources, the government should consider supporting more research and development in priority areas, where there is ample economic evidence of the need for public intervention because of positive externalities. The private sector will be an important partner in this area. The government should also focus on improving the regulation of course offerings by private providers to ensure a consistent quality of education. Greater civic involvement will also help support transparency in this case. Conclusion and Policy Recommendations There has been considerable progress in Cambodia regarding providing access to primary education, but more remains to be done to improve quality in primary education and to provide access with quality for other levels of education. Cambodia currently spends 2.4 percent of GDP on education, in line with the goals in the ESP. This level of spending is much lower than in other countries with broadly similar per-capita incomes, but 61 raising at this stage it can be accomplished to the extent the authorities can reallocate expenditures within the budget or by securing more grant donor support, given the overall level of government spending and private outlays on public goods. The authorities are also invited to consider the space that would be freed by the 12.5 percent decline in the number of primary students in the past five years. This space could be augmented by possible increases in donor support, provided strategic planning, budgets, and donor aid are aligned as discussed in Chapters 1 and 2. The authorities need to consider reallocating the possible savings from streamlining schools and staff as a result of this decline and a potential increase in donor support to the following areas: x Early childhood education: Resources allocated to increase access to early childhood education need to be spent timely and appropriately. Use public pre-schools, community pre-schools and home- based programs to scale up the level of provision. x Textbook budget: Textbooks need an improved budgetary allocation. Reaching the target of one textbook per subject for each child will need at least a 25 percent increase in the budget allocation even with efficiency gains. The estimated cost is about 0.02 percent of GDP a year. x Scholarship program: With assistance from development partners, provide additional resources for student scholarships. The estimated cost is about 0.03 percent of GDP a year. Over the medium term, the following recommendations will help improve performance of the education sector: x Strategic Planning: The ESP or equivalent strategic plan needs to be accompanied by a process of feedback where the achievement or non-achievement of targets is analyzed and corrective actions undertaken together with a reasoned calibration of the targets. x School quality: Educational quality as measured by student assessment needs to be mainstreamed and scaled-up from the current project-based approach. 62 Chapter 6. Sustaining Budget Credibility and Moving Toward Greater Financial Accountability A. Introduction Cambodia has made significant progress in reforming public expenditure policy and public finances after embarking in 2004 on a Public Financial Management Reform Program (PFMRP). Achievements improving budget credibility under Platform One – the first stage of the program – have been remarkable, but sustaining progress remains a key challenge. Budget credibility has largely been attained despite slippages and deficiencies. Platform Two broadens the reforms and aims to devolve greater budget responsibility to program managers with enhanced internal controls and accountability mechanisms. A carefully balanced and sequenced approach is necessary to improve the internal control environment while simultaneously devolving responsibility to line ministry departments and sub–national governments and entities. This chapter focuses on progress made since 2004 under the PFMRP and provides a general overview of the implementation of public expenditure and financial management. The chapter reviews the key accomplishments of Platform One (2004–2008) and outlines the remaining challenges. The chapter analyzes reforms identified under Platform Two with a special emphasis on the implementation of the Financial Management Information System (FMIS). The analysis highlights the key issues and risks faced by the government associated with the dual challenge of preserving the gains made over the past several years and sustaining the reform momentum. The functioning of Cambodia’s processes for managing and spending public money needs to be set in the broader context of the manner in which the overall public sector operates. Cambodia has made major progress in developing the foundations for an effective state after decades of conflict. At the same time, a host of issues constrain the government’s ability to deliver core services and all available studies point to persistent high-levels of corruption. Establishing more effective, transparent and rule- based governance remains at the heart of Cambodia’s development challenge (and at the center of the World Bank’s support for the country’s future). It is important to appreciate that the analysis and the recommendations contained in this report should be read as a compliment to the on-going work to enhance transparency, accountability, and participation and that enhancing the value derived from public finances depends on progress in both areas. This chapter underscores four broad areas where additional attention and action is needed. First, collaboration among donors must increase to further support and ensure that budget credibility is maintained and the reforms envisioned under Platform Two are advanced. At a minimum, the discussion on producing expenditure reports for donor-financed projects using the government chart of accounts (COA) should be resolved. Second, the management structures of the Ministry of Economy and Finance (MEF) needs to be strengthened to ensure successful implementation of the FMIS. Third, the 3-year implementation plan (IP3) and the sub-national financial reform agenda need to be aligned better with the PFMRP to ensure a feasible and sequenced approach to the Deconcentration and Decentralization (D&D) reforms. Fourth, it is important to embark on public administration and compensation reform to support the broader government reforms and strengthen the quality of public services. The rest of the chapter is organized as follows. Section B takes stock of the state of public financial management in Cambodia. Section C discusses the need for deepening Platform One reforms and advancing activities under Platform Two. Section D discusses the FMIS implementation. Section E analyzes public administration reforms and decentralization and deconcentration. Section F presents the policy recommendations. 63 B. The State of Public Financial Management The last IFAPER (2003) concluded that a weak public expenditure and financial management system had high costs in terms of allocative and operational efficiency and created unacceptably high levels of fiduciary risk to public funds. Budget execution suffered from delays and an unpredictable release of funds due to cash shortages as a result of poor cash management practices. This resulted in a build–up of arrears. Accounting and reporting systems were deficient, with a weak control environment presenting opportunities for corruption. Deficiencies in budget formulation and execution undermine the allocative efficiency and effectiveness resulting in the misallocation of resources across sectors, regions over time. Previous attempts to correct these problems were made through pilot initiatives, including the Accelerated District Development and Priority Action Programs in concert with a “sector wide” approach (Swap). These approaches improved the alignment of resources with policy objectives, but were hampered by ongoing concerns around the control environment and weak institutional capacity pointing to an overall inability to provide robust treasury support operations. To address this overarching and systemic reform agenda, the government developed a long–term phased platform approach–based program in December 2004. “The Public Financial Management Reform Program (PFMRP): Strengthening Governance through Enhanced Public Financial Management,” presented a detailed and sequenced action plan for reform. Because the envisioned transformation of the PFM system was a long–term undertaking, the government identified stages in the process of the reform at which a determination would be made before progressing to the next stage. Each stage was designed to change the performance of the PFM system and provide the platform on which further stages could be undertaken. The PFMRP is led by the Ministry of Economy and Finance (MEF). The PFMRP’s Platform One aims to make the budget a credible instrument for strategic and operational management of public resources. Activities under Platform One focus on improving resource mobilization (government revenue generation and collection), enhancing budget comprehensiveness and accuracy, and increasing budget implementation effectiveness. Budget comprehensiveness improved with the adoption of the 2008 Organic Budget Law (OBL) and the inclusion of donor funding and partial inclusion of Road Fund expenditures. The introduction of Budget Strategic Plans for both domestically- and externally-funded spending with costed priority sector strategies helped align the budget to sector priorities. Budget credibility has been enhanced as a result of several measures. These include strengthened budget formulation and execution processes, improved revenue performance and revenue forecasting, harmonized budget classification and chart of accounts, better cash management, and the introduction of a mid-year budget review process. A realistic medium–term macro–fiscal framework (MTMFF) was developed. Progress was achieved in consolidating government bank accounts, strengthening the Treasury Single Account (TSA), and expanding the use of commercial banks for government transactions. This, along with the recent relocation of the Cash Management Unit to the General Department of the National Treasury (GDNT) has helped improve cash management thereby ensuring greater predictability, reliability and availability of financial resources.32 Payment arrears were successfully eliminated in 2007 and have not resurfaced since then (Figure 46).33 Budget execution, monitored quarterly, steadily improved from a disbursement ratio of 64-68 percent during 2004-2005 to 93–96 percent on average 32 “Cambodia PFMR – Review by the External Advisory Panel,” February 2010. 33 Payment arrears are defined as appropriate payment orders outstanding for 90 days or more. 64 during 2006-2010 indicating improved procurement planning and streamlined payment processes (Figure 47) (see Annex Display14 for detailed performance indicators during the period 2004-2010). Positive steps to strengthen the public procurement framework and improve the overall procurement environment in the country have been taken. These include the enactment of the national procurement Sub-Decree No. 105 in October 2006 and the official publication and issuance of the supporting Implementing Rules and Regulations for Public Procurement (IRRPP) including standard bidding documents (SBDs) in July 2010. These are mandatory for all government units to procure with funds from the budget. Figure 46. Payment arrears quickly dropped and fully Figure 47. …while end-quarter disbursements eliminated in 2007 … improved quickly (In billion of riels) (In percent of annual budget) 600 100% 500 80% Actual 400 Target 60% 300 40% 200 20% 100 0% 0 2004 2005 2006 2007 2008 2009 2010e 2004 2005 2006 2007 2008 2009 2010e 1st Q 2nd Q 3rd Q 4th Q Source: MEF. Source: MEF. A draft procurement law that integrates the domestic and donor–financed procurement systems is scheduled to be submitted to parliament later this year. In parallel, a review and update of the Procurement Manual for externally-funded projects is near completion. The manual incorporates the feedback received during the last five years of use and will provide further guidance and procedures to enable effective implementation by project executing agencies. Several proposals to strengthen the government’s procurement capacity are also under consideration. An assessment of the government’s public financial management systems and overall performance using the Public Expenditure and Financial Accountability (PEFA) framework was conducted in 2009.34 The assessment indicated the strengthening of budget credibility in that the budget now provides a fair indication of the actual financial resources available to spending entities for providing services through greater predictability of revenues and reasonably good expenditure control. A regional comparison indicates that Cambodia’s PEFA scores are higher than or similar to those of Laos PDR but below those of Thailand.35 Cambodia scores the highest in the policy-based budgeting dimension among the three countries (see Annex Display 16). The Merit-Based Pay Initiative (MBPI) allowance introduced during Stage 1 of the PFMRP was significant in motivating staff engage in the reform program. Despite ongoing difficulties in inter- 34 “Public Finance Management Assessment Cambodia: Based on the Public Expenditure and Financial Accountability Framework (PEFA) Final Report,” February 2010. 35 Regional comparison must be treated with cautions as one country’s scoring may not be fully comparable to scoring of other countries. 65 department coordination and cooperation, significant change was achieved. Some managers introduced change management ideas and amended business processes with good results.36 Platform Two broadens the reforms and aims to devolve greater budget responsibility to program managers through enhancement of internal controls and accountability mechanisms. This dual strategy requires a carefully balanced and sequenced approach. Officially launched in December 2008, Platform Two reform activities were initiated across government and in all line ministries. The decision to cover all of government resulted in uneven execution and delays to an already slow implementation. Additional institutional arrangements, such as the introduction of the Independent Procurement Agent (IPA) hampered progress. In response, the MEF refocused its efforts on a series of strategic activities – particularly around the implementation of the FMIS – in key line ministries. Designed in the first quarter of 2010 and launched in April of that year, the catalyst measures to accelerate the implementation of PFMRP strategy identified several outputs to be achieved by the end of 2010. Figure 48. Priority spending steadily improved together Figure 49. …and overall annual disbursements with high growth of investment spending … by expenditure profiles rose markedly. (In percent of annual budget) 40 Capital exp growth y/y 5 125% Priority spending (% of GDP, r. scale) 30 4.5 100% 20 4 10 3.5 75% 0 3 50% -10 2.5 25% -20 2 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 0% 2004 2005 2006 2007 2008 2009 2010 Salary Goods/Services Capital Source: MEF. Source: MEF. Good progress has been made on catalyst Platform Two activities and remaining Platform One activities. Revenue policy and administration continue to deepen and improve. The use of the banking system has broadened, and consolidation of the government’s procurement systems will be advanced with the planned adoption of the public procurement law in 2011. The budget now provides a greater indication of the actual resources available to spending entities thereby further enhancing its credibility as a policy document and implementation tool. Priority and investment spending continues to increase reflecting sound expenditure policy (Figure 48). The significantly improved disbursement profile – in particular in capital spending – points to more developed framework for budget execution (Figure 49).37 36 “International Merit Base Pay Initiatives (MBPI) Evaluation, Final Independent Evaluation Report,” October 2009. 37 Priority ministries include Health, Education, Agriculture, Rural Development, Women’s Affairs, Justice, Land, and Labor. 66 Efforts made toward increasing financial accountability under Platform Two have been Box 3. What does an Integrated Financial considerable. Much of the work centers around Information System (IFMIS) do? five interlinked catalysts focused on FMIS The FMIS automates (computerizes) the financial implementation. The policy for developing budget operations of budget and treasury units and is usually entities below ministerial level has progressed. built around a core treasury system supporting key budget Internal audit has been strengthened and internal execution functions. The system: x Improves the accuracy of budget transactions, audit departments in several line ministries are in execution and reporting across organizational units place. A new integrated uniform account code and budget entities; structure is being developed. It will unify budget x Increases the efficiency and transparency through and accounting functions into a single structure. direct payments to vendors reducing uncertainty surrounding payments; Financial Management Information Systems x Supports decentralized operations through (FMIS) enable governments to control aggregate centralized web–based solutions providing access to spending and deficits and prioritize expenditure authorized budget users at all levels (provinces, across policies, programs and projects for districts). allocative efficiency and equity. The MEF has set out its definition and vision for the implementation of a fully Integrated FMIS across the government. Along with a series of implementation and strategy plans around key areas (including a functional review, change management, communications, business process reengineering, and FMIS project planning) the MEF is introducing the leadership and management structures required to lead and implement this complex, multi-year reform. The FMIS is planned to be implemented in two phases. The first phase’s revised sequencing proposal will include the core treasury and budget modules to be implemented within MEF: MEF departments, the General Department of the National Treasury, and all provincial treasuries while having electronic interface (e-forms) with all line ministries. The remaining functional modules – purchasing, asset management, and interfaces (payroll, tax revenue, and debt management) – will then be put in place in all line ministries and provinces as part of the second phase. The first phase is a Treasury–centric approach that will help ensure the implementation of the core functionality at the national treasury and the provincial treasuries. Spending ministries at the center and spending departments in the provinces will submit – using a web-based portal where possible – their transactions to the central treasury or the corresponding provincial treasuries for processing. The web- based portal will provide easier access to financial information to the spending ministries and departments. Under this approach, implementation of the central system becomes easier and less costly allowing the government to speedily capture all payment transactions enabling greater fiscal control and more timely financial information. The budget entity is a key concept for introducing greater financial and programmatic accountability at all levels of the government. A budget entity is an organization that uses or receives public resources to accomplishing a function for which the government is responsible.38 Autonomous public organizations and sub–national governments considered initially as budget entities or cost centers. The MEF plans to ask pilot ministries to prepare their 2013 budget submissions under the new structure. Budget entities will be piloted in key line ministries as part of the broader phased approach for putting into operation the FMIS which also requires the reengineering of business processes, a new 38 Ministry of Economy and Finance, 2010, “Framework for Indentifying Budget Entities.” 67 comprehensive budget classification and unified account code structure, strengthened internal audit and controls, and comprehensive recording and Box 4. The new classification encourages the use of reporting.39 government public financial systems The new budget classification (and chart of accounts) The FMIS system will facilitate the introduction referred to as the Integrated Uniform Account Code of an Integrated Uniform Account Code Structure will enable budget integration and encourage the Structure. A new six-segment classification use of the government’s public financial system via budget system proposed by the MEF integrates the budget support as donors will be able to have a better understanding how funds are being accounted and where and accounting functions into a uniform single expenditures are incurred. structure. The new coding structure reflects function, administrative, economic, program, This is made possible by “source of funds” and source of fund, economic, and geographic and a “geographic” classification coding. The “source of funds” framework that will cover the needs of all the users classification supports the integration of externally financed projects into the budget and fiscal reporting and captures of financial data of the government, ranging from government own sources and external (donor-financed) the parliament, the National Audit Authority, line sources. The “geographic” classification identifies locations ministries and sub–national governments to of incurred expenditures distinguishing among central, citizens and donor agencies. provincial, municipal, district, and commune levels of administration. As the single, integrated central repository of information and source for budgetary control, the FMIS will provide managers consolidated and more reliable and timely information. Reporting will become an output of the system and not a process. Once transactions are entered into the system, reports are generated. The FMIS provides opportunities for improving the existing system through extensive automation, integration and streamlined operations. These can only be realized if appropriate business processes are fed into the FMIS. Therefore, moving from manual and fragmented systems to a highly automated and integrated system requires a complete business process review supported by robust change management measures. The new business processes will root out rent seeking behavior. Once computerized, the new business processes will automate workflows, eliminate duplication, superfluous layers and steps, and reduce the need for the proliferation of different letters and forms. Electronic notification to managers will replace management involvement in transactions and reduce unnecessary approval processes thereby eliminating opportunities for collusion and corrupt practices. Steps have been taken to create a unified chart of accounts (COA) for use by all levels of government. Within the MEF, the General Department of the National Treasury (GDNT), the Budget Department, Local Finance Department, and the Department of Investment and Cooperation (DIC) are working to finalize development of the uniform account code structure by integrating the separate charts of accounts used by the central, provincial, district, and commune levels of government. Developing a coherent view of the accounting model and standards has been slow, due in part to conflicting policy advice from donors. Recently, consensus within the government has developed around the implementation of International Public Sector Accounting Standards (IPSAS) and work on understanding these standards is at an early stage. As an important first step, the GDNT has undertaken a gap analysis between current systems and IPSAS standards to develop a roadmap and strategy to move toward IPSAS and International Financial Reporting Standards (IFRS). A program is under consideration to support the implementation of the road map. Such a program will further strengthen treasury operations and functions critical for the successful implementation of the FMIS. 39 Planned FMIS pilot ministries are: Health; Education, Youth and Sports; Agriculture, Forestry and Fisheries Rural Development; and Public Works and Transport. 68 Internal auditing has been introduced. Internal audit departments were established in the line ministries and agencies although many are not fully operational. An internal audit framework that includes the internal audit standards and the charter for the internal audit function is an important component of a robust internal control environment and requires further strengthening. With the framework in place, audit manuals, plans, and procedures can be developed by the MEF Internal Audit Department for the internal audit departments in the line ministries. A uniform and consistent set of standards which is currently at the final stage of preparation for all ministries to follow is critical in ensuring the proper controls and procedures are in place across government. C. Deepening Platform One Reforms, Advancing Platform Two Activities Significant progress has been accomplished under Platform One and Platform Two, but challenges remain. The PEFA assessment compared progress since 2002 and concluded that budget credibility requires additional strengthening in the areas of revenue policy and administration, cash management, debt management, and procurement. Expanding the fiscal space remains a dual challenge due to a low ratio of revenues to GDP and an overdependency on trade taxes under the PFMRP. Revenue collection dipped in 2009 due to the negative impacts of the global financial crisis (Figure 50 and Chapter 1 for more details). Under the ASEAN free trade agreement (AFTA) the government is obligated to reduce trade taxes, but the direct taxation system is hampered by questionable investment incentives and weak compliance (Figure 51). Direct tax collection remains well below levels in developing East Asia and the overall tax administration remains inadequate. In 2010, the government approved the Revenue Mobilization Strategy, identifying potential measures for addressing these deficiencies. The introduction of the national receipt for non–tax revenue aims to improve revenue administration. The clarification of the petroleum tax and the introduction of property and capital gains tax schemes should improve collection and expand fiscal space. Improving revenue administration and broadening the tax base is of particular importance. Such efforts need to be accompanied by measures to strengthen the accountability for the use of public funds. Figure 50. Economic downturn resulted in a Figure 51. …while direct tax growth rate remained contraction of revenues in 2009 … volatile (percent change relative to budget) (percent change year-on-year) 50% 33.0% 40% 23.0% 30% 16.0% 14.0% 20% 3.0% 10% 0% -5.0% -8.0% -10% 2004 2005 2006 2007 2008 2009 2010 2004 2005 2006 2007 2008 2009 2010e Source: MEF. Source: MEF. The share of domestically financed priority spending has steadily declined. This indicates that the additional fiscal space created has been used for non–priority sectors (Figure 52). While linking program priorities and budget is to be addressed more fully during the third stage of the PFMRP, initial steps can be taken now, including the integration of spending financed from own resources and from external support, the adoption of the MTEF, and the alignment of budget strategic plans with sector priorities. 69 The PEFA noted that budget predictability for spending agencies is low and the variance in expenditure composition exceeded the aggregate deviation by 10 percent in two of the last three years ending in 2008. This is due, in part, to irregular allocations from a large reserve (contingency) budget. The reserve amounts to about 10 percent of budgeted expenditure and the unpredictable way it is allocated throughout the year contributes to non–transparent budget management practices. Predictability 40 is somewhat better in education, health and agriculture. Despite significant improvements to cash management, in particular in strengthening the TSA, cash flow management practices have not evolved sufficiently. A number of issues related to cash flow forecasting data collection and sharing continue to have serious implications for ensuring significant resources on hand to meet the government’s cash flow requirements. Imminent cash shortages have reemerged since 2010, threatening hard earned achievements on payment arrears (Figure 53). The Cash Management Unit (CMU) has identified as a priority the need for expenditure planning information and full cooperation with line ministries and MEF departments, but the CMU does not have the authority to require departments to submit cash flow forecasts. Figure 52. Priority sector spending share of total Figure 53. TSA’s cash balances have been domestic budget is declining. decreasing since mid-2009 (in percent of domestically-financed spending) (in billion of riels) 39.0 1,200 37.0 1,000 TSA in Riel (NBC) 35.0 Total Available Cas 800 33.0 600 31.0 400 29.0 200 27.0 0 Sep-08 Sep-09 Sep-10 Mar-08 Mar-09 Mar-10 Mar-11 Nov-08 Nov-09 Nov-10 Jan-08 Jan-09 Jan-10 Jan-11 May-08 May-09 May-10 Jul-08 Jul-09 Jul-10 25.0 2004 2005 2006 2007 2008 2009 2010e 2008 2009 2010 2011 Source: MEF. Source: MEF. There has been solid progress implementing the Debt Management and Financial Analysis System (DMFAS) and preparing the debt management legal framework. By contrast, debt data integration has not been successful. The GDNT is unable to capture debt data resulting in incomplete government financial reporting. The institutional fragmentation of the debt and cash management functions remains. A well functioning and fully integrated debt and cash management operation will help ensure improved cash management and planning. The public procurement system currently remains fragmented because of two separate legal and regulatory frameworks applicable for procurement financed from the national budget and by external donors. Each is governed by a separate sub-decree, separate implementing rules, regulations and processes under the oversight of two different MEF departments. Significant benefits of economy and efficiency can be obtained by consolidating the two systems into a single unified national public procurement system. No explicit provisions and procedures exist in the legal framework on the remedial 40 While the TSA strengthening has helped improve budget comprehensiveness by consolidating government bank accounts, the recent PEFA assessment indicated that there is a significant extent of unreported extra-budgetary operations using possibly non-declared non-tax revenues that may be inconsistent with public policy objectives. 70 actions to be taken against those involved in fraud and corruption in the procurement process or during contract execution (including collusive, coercive and obstructive practices). The capacity of line ministries and agencies to effectively carry out procurement remains weak and the local market of suppliers and contractors is underdeveloped and constrained by governance issues. There is no system for ministries and agencies to monitor performance and the outcomes of procurement. The main weakness of the policy–based budgeting dimension is the continued poor integration of capital and recurrent expenditures. This is exacerbated by having a separate unit (the DIC) in MEF and project implementation units (PIUs) within line ministries that manage all aspects of donor–funded projects. While stronger links have been made between the annual budgeting and medium–term planning through the BSP, more can be done to capture external financing projects to ensure availability of funding of all project costs and strengthen the link of donor–financed programs with the policy priorities of the government. Robust costing information on domestic and external resource requirements along with sufficient identification of objectives and targets linked to sector priorities will enhance the BSP credibility as policy tool. Donor practices in several areas continue to challenge the government’s ability to advance public financial management reform. The PEFA rated donor practices consistently low in: (i) the predictability of direct budget support; (ii) the use of country systems; (iii) the completeness and timeliness of budget estimates for project support; and (iv) the frequency and coverage of reporting on actual donor flows for project support.41 Achieving budget comprehensiveness and transparency remains a key challenge to the government, primarily because of the exclusion of a significant portion of public expenditures financed 42 by donors. This is especially problematic when external financing of expenditure remains outside the government’s stated policy goals. Moving towards use of government systems is also important. As an initial step, donor projects can Box 5. The World Bank Experience with Treasury use the government COA to report data, enabling the and Financial Management Information Systems MEF to capture this data for broader planning, Since 1988, the World Bank has financed 87 treasury budgetary, and reporting purposes. The relatively and financial management information system projects higher marks for budget credibility under PEFA are in 51 countries. The four key factors contributing to the somewhat misleading so long as PIUs continue to success of the treasury and FMIS projects were: manage donor–sponsored projects within ministries x an emphasis on developing the human resource capacity necessary to implement and fully operate outside the government budgeting, accounting and the systems reporting mechanisms. The lack of transparency in x the presence of strong political commitment and many grant-funded donor projects implemented by ownership by the borrowers NGOs erodes budget credibility. In addition, putting x close World Bank supervision more donor funding on budget promotes a greater x flexible project management sense of urgency in getting formal government Source: The World Bank’s Experience with Treasury systems running, facilitates greater government and Financial Management Information Systems (1984 responsibility and oversight, and serves as an – 2010), 2010, draft World Bank working paper. important catalyst for implementing change. These 41 With the exception of the first indicator related to the predictability of direct budget support which received a ‘C’ grade, all indicators related to Donor practices received a grade of ‘D’. 42 Most of the recurrent cost implications of donor financed projects on the national budget are not adequately identified or considered and represent significant future funding and program service delivery liabilities. The PEFA notes that stronger linkages between the current and capital budgets will strengthen the credibility of the budget over the medium term by increasing the probability of sufficient financial resources being budgeted for the provision of projected required service levels. 71 issues are critical within the broader context of aligning donor support with government strategic planning and budget. Recent initiatives allow for incorporation of a requirement to classify donor-financed resources and expenditures using the government Box 6. Incentive Measures for Implementing the Reform chart of accounts, as stipulated in a revised Program draft Financial Management Manual for donor-financed projects. The provision maintains In August 2005, the government and development partners the flexibility to meet donors’ budgeting and piloted and jointly financed a Merit-Based Pay Initiative (MBPI) to provide better incentives for the implementation of financial reporting requirements until improved the PFMRP. Largely because of equity considerations, the harmonization and alignment can be agreed and government cancelled the MBPI on January 1, 2010 and implemented. Agreement among donors should replaced it seven months later by a new incentive plan, the be a priority and accelerated. Priority Operating Costs (POC). Progress on the PFMRP remains encumbered The POC scheme is part of a twin-track approach for broader compensation reform encompassing: (i) a longer-term by deficient technical skills and weak capacity strategy as defined by the National Program for despite ongoing capacity building efforts. A Administrative Reform (NPAR) and (ii) the temporary POC proposed twining program will provide training in arrangement to address service delivery and reform accounting, budgeting, auditing, and basic implementation issues. Both tracks are expected to progress management, using in part MEF’s training in parallel with regular reviews to: (i) assess progress toward compensation reform; (ii) feed lessons from POC to entity, the Economic and Finance Institute (EFI). compensation reform; and (iii) decide on the merits of Within the MEF, this training will need to be revising, continuing, or stopping the POC arrangement. Whatever the outcome, addressing overall compensation supplemented by capacity development efforts led from within line structures and departments. This issues is a critical factor for sustaining and further progressing reforms across government. should help sustain progress on the PFMRP and other reform initiatives such as D&D, public administration reform, and good governance (see Box 5 on the key role human resource development played in the implementation of FMIS in other developing countries). Improving accountability through robust internal and external audit functions supported by strengthened parliamentary oversight remains problematic. Although work on internal audit has been ongoing since 2006, the recent PEFA assessment noted that internal controls remain weak. While templates for audit reports have been developed by the MEF Internal Audit Department (IAD), they are not uniformly applied or used by line ministries. With donor support, internal audit has improved in key line ministries, including the Ministry of Health (MOH), the Ministry of Agriculture, Forestry, and Fisheries (MAFF), and the Ministry of Rural Development (MRD), although these are conducted largely to meet donor reporting requirements rather than those of the government. As the government moves more financial management and accountability to the line ministries and sub-national governments, audit plans will need to incorporate a risk–based approach currently under support and guidance provided by the staff of MEF’s Internal Audit Department and concentrates on the line ministries. A stronger National Audit Authority (NAA) will strengthen fiscal transparency and encourage greater use of country systems by donors. Two issues are of particular importance: the conduct by the NAA of annual compliance and financial audits and the timely and comprehensive publication of results that meet international audit standards. A recent peer review of the NAA by the New Zealand Audit Office concluded with three recommendations. Those are: (i) building NAA existing systems and methodologies for its auditing activities; (ii) preparing the NAA for a wider contribution under the PFM reforms; and (iii) continuing the implementation of its strategies for developing its infrastructure including its workforce and a comprehensive and sustainable ICT platform.43 43 The review was conducted in 2010. Most recommendations are either omitted or given low priority in the NAA’s work plan. The PEFA assessment also concluded that external audit is not playing a strong role in holding the 72 Oversight by the legislative branch remains weak. Strengthening the capacity of the parliament, and in particular the commission in charge of finance and banking, will further enhance oversight capacity. Development partners continue to engage with the parliament, especially during the annual budget appropriations through workshops and seminars and supporting the publication of “budget in brief” for citizens to improve understanding of the government budgets by all. The recent regional promotion of National Assemblies and National Audit institutions sharing experiences has been particularly encouraging. D. FMIS Implementation: Challenges and Risks The FMIS project represents both a technical challenge and a fundamental transformation of the way financial decisions are made and how resources are controlled and managed. The choices to be made around authority, accountability, responsibility, and technology will shape and determine the government’s overall financial management and service provision structures for many years to come. Clarity in policy on a number of technical and non–technical issues will facilitate a successful FMIS implementation and help ensure that the system eventually put in place throughout the government is useful and relevant. The ongoing work in preparing the FMIS contract suggests that completion of the system will take time, and while reform is possible, many difficult and transformational changes will need to be committed and agreed losing hard-won policy or procedural gains as the system is rolled out. The FMIS catalyst activities have been advanced, but these should be refocused and realigned to accelerate progress. There are three key areas to help ensure the FMIS begins on the right track, facilitate project implementation, and avoid delays thereby minimizing potential costly contract amendments in the future. First, a functional review and putting into place a new organizational structure to support the reform agenda is necessary. Having the new structure and management in place by the time of contract award will facilitate a smoother implementation of the FMIS. Second, a new chart of accounts that integrates budget and accounting into a uniform account code structure should be agreed and implemented. And third, a revised set of business processes will need to be fairly advanced to allow for testing, adjustments and subsequent training of staff before the core system starts operation in early 2013. These set of activities should underpin a well–defined and integrated FMIS action plan that incorporates all the elements of a successful FMIS implementation. Three groups (line accountability, budget execution, and accounting and reporting standards) led by two Secretaries of State and a MEF Secretary General have been established to develop and implement the FMIS action plan. To accelerate the implementation of the FMIS, the MEF proposes a further prioritization of five core catalyst areas that revolve around FMIS implementation. The “3 + 2 proposal” presented in March 2010 attempts to further prioritize and accelerate work of the groups working on first set of catalysts (Table 23). Further efforts are needed to integrate the work of the three groups into a single plan endorsed by development partners. Under the proposed Treasury–centric model, the General Department of the National Treasury will process all transactions and the centralized accounting and payment systems at the Ministry of Economy and Finance will remain. Such a proposal does not bring about the accountability envisioned under Platform 2. Under this scenario, internal audit in the line ministries becomes less important. executive branch of government to account. The NAA does not publish its reports on time (a summary 2006 audit report was published in late 2009), giving the public no way of knowing whether public funds are spent according to their intended purpose. 73 Although envisioned as an interim solution, work on developing proposals for the powers and responsibilities to be assigned at different levels of budget entities should be accelerated to help determine how quickly to empower budget entities in the pilots and eventually throughout government. The MEF has developed a comprehensive strategy for managing staff to support the FMIS implementation. The strategy aims to manage employee resistance and build the capacity for change within the MEF, line ministries and provincial Treasuries.44 Although a Table 23. The 3 + 2 proposal: key catalysts for financial accountability business process impact assessment has started and a certain level of involvement 3 Key Catalysts (Elements) Key Milestones and awareness by departments achieved, Line Accountability · Budget Entity acceptance of these changes across · Internal Control System government remains a significant challenge, Accounting Standards · Budget Classification a situation that decreases the likelihood of · Chart of Accounts initial acceptance. Because the advent of · Recording/Report automated business processes will result in Budget Execution System · Commitment process · Procurement Process fewer opportunities to engage in rent · Payment Process seeking behavior, strong and sustained leadership from MEF, the line ministries, 2 Key Catalysts Key Milestones and the project implementation teams will Financial Management · Strategic implementation plan be critical for the success of the FMIS. Information System (FMIS) · Contract award and ·Project Management Experience with the PFMRP suggests that the FMIS implementation will be Capacity Development · Training · HRM slower than previously envisaged. The · Institutional Structure FMIS as a concept and in practice is new to Source: MEF. Cambodia. Given the lack of experience, the complexity, and scope of this significant task, further consideration should be given to the time required for piloting and rolling out Phase One of the project. Under current plans, the MEF anticipates awarding a single contract for Phase One that includes a piloting phase and rollout of core treasury and budget functions to all provincial treasuries over a four year period. The World Bank’s experience implementing FMIS projects throughout the world suggests that comprehensive FMIS projects take a minimum of 6–7 years to complete (including the project design, procurement, development of information systems, and capacity building).45 Given that the procurement process for the FMIS system commenced in 2006, lengthening the time for the project and awarding the contract in multiple phases appears to be the way forward. E. Public Administration Reforms and Decentralization and Deconcentration Integrating cross-cutting public administration and finance reforms with reforms in health, education and other sectors remains an important challenge. Determining responsibility for service delivery and resource management is linked to staffing and how the whole civil service is organized. Delays in comprehensive civil service and institutional reform directly influence progress made in public financial management and put at risk the sustainability of many hard earned gains. Assigning and devolving functions is a key aspect for achieving greater alignment between resources, responsibilities 44 See MEF’s “Change Management Strategy,” unpublished report prepared for the PFMRP, 30 th April 2009. 45 The World Bank’s experience with Treasury and Financial Management Information Systems (1984 – 2010) Working Paper Draft, November 2010. 74 and awards. Future progress on Platform Two reforms, including FMIS implementation, could be at risk unless these broader public administration issues are addressed. As the FMIS becomes operational, many current functions, assigned authorities and responsibilities, and changes brought about by the revised business processes will need to be addressed through organizational restructuring and enhancing the skills and capacity of existing staff. Launching a concerted effort to strengthen the process for working with senior officers in ministries in the implementation of the three public sector management reforms – Public Administration Reform (PAR), PFM and Decentralization and Deconcentration (D&D) could heighten a sense of personal responsibility for their success. Such mechanisms could include consultations with the leadership of line ministries and agencies on how the three cross-cutting reform efforts can most effectively help them to meet their mandates, as well as on how those inter-locking reform efforts could be adjusted to better support line ministries in their efforts to deliver services and implement their programs. Having small sets of senior officers of line ministries report periodically to the Supreme Council for State Reform on how the PFM- PAR-D&D reform initiatives are being implemented in their ministries could foster greater linkages, understanding, and acceptance for the reforms across government. Additionally, greater communication and consultations among the technical leadership working on these reforms especially in harmonizing the development of regulatory and implementing guidelines would mitigate the risks of conflicting and incompatible processes thereby minimizing unnecessary delays to reform implementation.46 Public administration reform should be viewed as part of a wider reform process to ensure that policy-making, budgeting, service delivery and accountability are aligned to achieve the strategic national development goals. The government aims at transforming the public administration to become an “effective public service provider” by making the public service more transparent, responsive and efficient through motivation, loyalty, and professionalism. Toward this end, the Council for Administrative Reform (CAR) is developing key policies including human resources management, human resource development, and public service policies. The government has in place a policy to increase basic salary by 20 percent per annual; however, civil servants’ salaries remain significantly low compared to those of the private sector. Low salary levels limit the ability of the public sector to deliver services in every area, and in particular, in PFMRP, and present salary distortions. The MBPI scheme and the subsequent POC schemes attempted to address these issues. The National Program for Administrative Reform (NPAR) sets the overall framework for addressing the structures of government, improving human resource management instruments and compensation reform. The government, led by CAR and the development partners within the framework of the PAR-TWG agreed to identify viable proposals for sequencing compensation and broader public administration reforms by the end of 2011 as envisaged as overall compensation issues remains a critical factor for sustaining and further progressing reforms across government, in particular, the PFMRP. The sub-national reform agenda needs to be aligned more closely with the PFMRP. This will require sustained and expanded dialogue on these reforms, clarification of roles and responsibilities, and efforts to avoid duplications and non–alignment of reforms. The Organic Budget Law (OBL) adopted in 2008 sets the legal and regulatory framework for all aspects of national and sub–national government budget 47 management processes. The recent draft sub-national finance law (SNFL) sets forth the financial regime and property management of the sub–national administration. These laws require careful review to ensure 46 These initial recommendations and ideas were developed by a World Bank team working on the three reforms for future discussions with government. 47 See Chapter 1, Article 3of the Organic Budget Law (Law on Public Finance System) 2008. 75 they are complimentary, do not contradict one another, and create sufficient space for the establishment of mechanisms and other laws that support sub–national financing.48 The proposed management of administrative accounts under the SNFL is of concern. The OBL calls for the use of the TSA and significant progress over the years under the PFMRP has been made closing accounts. Allowing governors to manage, open, and close ‘administrative accounts’ under the SNFL will erode the government’s efforts to improve the already strained cash management systems and contribute to cash shortages further jeopardizing the sustainability of Platform One reforms. Careful consideration should be given as to whether or not governor’s administrative accounts are needed at this point in time and the fiduciary risks associated with such accounts given current weak management structures. The three-year implementation plan for the D&D reforms outlines a set of complex compliance and inspection arrangements. Compliance could largely be achieved through external and internal audits. While the ministries will need to perform some regulatory oversight functions to ensure that national strategies, policies and procedures are followed, how this is to be done will require careful consideration. The implementation plan (IP3) envisions a fiduciary safeguard structure with an internal control policy and regular independent audit of accounts. The responsibility for conducting regular independent audits at the sub–national level is unclear. If these are to be conducted by the NAA, that institution will require significant capacity development to fulfill this function. Clarity around the financial reporting arrangements of sub-national administrations (SNAs) is needed. The production of annual financial statements by sub–national entities and an annual audit by the NAA would be a positive step in terms of governance, accountability and transparency. The SNA will need to move toward IPSAS to be in alignment with the national government’s accounting practices and standards. The IP3 states that a single integrated management information system for planning, budgeting, monitoring, evaluation, financial management (and accounting), contracting and reporting will be implemented. Careful consideration and planning is needed to ensure the appropriate linkages are made with the FMIS that will be rolled out to the provincial level and to determine whether or not an FMIS for provinces, districts and municipalities will be able to deliver the benefits envisaged. A key aspect of the decentralization agenda concerns the discussion and agreement on functions and what level of government is responsible for their execution. Devolving responsibilities to the provinces and districts not only has financing and resource implications but reaches across a number of institutional, organization and human resource considerations that require agreement within the government. As consensus evolves on these sensitive issues, determining appropriate resource management and allocation issues becomes clearer. F. Conclusion and Policy Recommendations The government remains committed to its detailed and sequenced action plan for “ Strengthening Governance through Enhanced Public Financial Management.” Work should continue on strengthening Platform One accomplishments. Platform Two broadens the reforms and aims to devolve greater budget responsibility to program managers through enhanced internal controls and accountability mechanisms. A carefully balanced and sequenced approach is necessary to improve the internal control environment while simultaneously de–concentrating control and responsibility out to line ministry departments and sub–national governments and entities. Sufficient time for piloting and testing processes 48 For instances, the OBL may need to be amended to be consistent with the SNFL by allowing governors to order the collection of revenue (currently a central government function). 76 and procedures along with a greater focus on building capacity and engagement with line ministries will help to secure acceptance and success at all levels of government. The “3 + 2 proposal” is a step in the right direction, and the MEF must continue to provide the leadership necessary to ensure the overall success of the FMIS project. Moving to Platform Three activities should occur only after budget credibility has been further enhanced and sustained and the successful rollout of the FMIS achieved. The government is invited to consider the following recommendations to help achieve the stated PFM reform objectives. In the short-term measures to sustaining budget credibility involve further strengthening of Platform One activities: x Develop and implement action plan for the newly approved revenue mobilization strategy to expand fiscal space. Introduce and implement the national receipt regime scheduled to be launched in early 2011. x Further strengthen government cash and bank accounts management including cash flow forecasting and cash planning. Expand the use of banks for government transactions to all provinces. Reconsider the devolution of authority around ‘administrative accounts’ in the SNFL. x Enact and implement the new procurement law. Move toward the integration of the national public procurement systems under a single normative/regulatory body to ensure efficiency, transparency, and accountability. At the same time, moving toward financial accountability will include advancing key Platform Two Activities: x Design and implement a well–defined and integrated FMIS implementation action plan underpinned by the newly established line accountability, budget execution, and financial standards groups. Such a plan that brings together the three working groups (line accountability, accounting standards, and budget execution) of the FMIS Steering Committee should be in place and substantial progress made by the time of the FMIS contract award. The plan also should incorporate a roadmap that illustrates the transition from the Treasury–centric approach to a wider rollout of an FMIS that provides greater financial management and accountability for sub–national administrations and line ministries. The timeline for implementing the entire program should be recalibrated. x Improve engagement and capacity of the NAA to conduct external audits and enhance legislative oversight and engagement. Because of the number of donors currently supporting or planning to support the NAA, assistance must be coordinated and agreement on the nature of that assistance (based on the recommendations of the peer review) should be agreed. x Design and implement a comprehensive treasury support program to facilitate the implementation of the FMIS, sustain and improve cash management, and the movement towards the adaptation of IPSAS standards. x Continue to build capacity for both short and long–term programs including through the twinning program, bilateral and multilateral cooperation. Introduce professional training and certification programs for accounting, internal auditing and procurement. 77 Annexes Annex Display 1. NSDP 2006-2010 targets and the Cambodia Millennium Development Goals CMDG- 2005 2008 2010 2015 Eradicate - Poverty & Hunger (*) 1* Poverty levels -- percent of population 34.7 30.1 25 19.5 2* Poverty levels -- percent rural population 39.2 34.7 People below food poverty line -- percent of 3* 19.7 18 13 10 population Enhance Agricultural Production and Productivity 4 Paddy yield per hectare (tons) 1.97 2.74 2.8 3.0 Irrigated area – including supplemental irrigation 5 20 - 33.5 ( percent of rice area) Land Reforms: Land Titles to farmers -- percent of 6 12 20 24 Total agri. Land Improvements in Health (*) 7* Infant Mortality Rate per 1,000 live births 66 66 60 50 8* > 5 Mortality Rate per 1,000 live births 82 82 75 65 9* Maternal Mortality per 100,000 live births 473** 461*** 350 250 Births attended by skilled health personnel -- 10* 44** 58 70 87 percent 11* HIV/AIDS prevalence, percent of adults 15-49 1.0 0.7 0.7 <0.6 12* Malaria Cases – fatality per 100,000 2.70 2.10 1.20 0.78 13* TB smear positive cases, per 100,000 676 617 564 464 Married women using modern birth spacing methods 14* 20.1 26 40 60 ( percent) Improvement in Nutrition percent of children aged 0 – 59 months who are 15 moderately or severely stunted (height-for-age more 37 39.5 28 22 than 2 standard deviations below normal) percent of children aged 0 – 59 months who are 16 moderately or severely wasted (weight-for-height 7 8.9 10 9 more than 2 standard deviations below normal) 17* percent of women aged 15-49 years with anaemia 47 NA 32 19 18 percent of children aged 0-59 months with anaemia 62 NA 52 42 Improvements in Education (*) Net Enrolment: Primary Schools -- Total; Boys; 91.9; 93.0; 94.4; 94.8; 96.0; 95.5; 100 for 19* Girls -- percent 90.7 94.0 96.0 all Net Enrolment: Lower Sec. Schools --Total; Boys; 26.1; 27.1; 33.9; 32.0; 43.0; 41.0; 20* 100 Girls -- percent 24.8 35.0 45.0 21* Gross Completion Rate Grade 6 -- percent 53.1 85.6 88 100 22* Gross Completion Rate Grade 9 -- percent 30.18 49.1 53 100 23* 6-14 years out of school ( percent) 18.7 17 15 10 Rural Development Rural Roads rehabilitated – Kames (out of total 24 22,700 24,140 26,658 32,000 28,000) Safe Drinking water access -- percent rural 25* 41.6 40.49 43.49 50.09 population 26* Sanitation access -- percent rural population 16.4 23.24 25 33 Environmental Sustainability (*) 27* Forest Cover -- percent of total area 60 59.09 57.99 60 78 28* Fuel Wood dependency: Households -- percent 83.9 73 61 52 Access to safe water source -- percent of urban 1 1 29* 35 51 75 80 population Access to improved sanitation -- percent or urban 30* 55 - 67 74 population Gender Equity (*) 31* Mainstream gender in all spheres Number of ministries/institutions that have a 2 15 20 27 formulated a Gender Mainstreaming Action Plan Number of ministries/institutions that have b implemented their Gender Mainstreaming Action 2 10 15 27 Plan Female share of wage employment -- agriculture, 52.5; 53.5; 52.5; 53.5; 50; 50; 50; 50; 32* industry, services ( percent) 27.0 27.0 37 50 Level of awareness that violence against women is a 33* 4.5 30 50 70 crime ( percent) Number of protection orders issued by courts, based a on the Law on the Prevention of Domestic Violence 0 NA 40 120 and the Protection of the Victims Number of victims of domestic violence who b 0 600 900 2000 received counselling by qualified personnel. Reforms 34 Accelerate Governance Reforms To be developed Sustain high Macro-Economic Growth (*) 35 Annual GDP Growth at constant prices - percent 13.3 6.7 3.0 36 Per Capita GDP at constant prices (000 Riels) 1,400 2,998 3,284 37 Rate of Inflation percent 6.2 19.7 6.0 Improve Budget Performance Total government budget revenues - percent of 38 11.80 13.3 12.6 GDP Total government budget expenditure -- percent of 39 14.9 15.7 15.4 GDP Targets/Indicators Accelerate Industrial Growth & Employment Annual Growth in manufacturing - constant prices 40 10.2 3.1 14.2 ( percent) 41* Working children aged 5-17 years -- percent 22.3 - 10.6 8 Tourism 42 Annual Tourist arrivals (in hundreds of thousands) 1,421 2,125 2,276 4,450 De-mining, Victim Assistance (*) 43* Casualties (deaths and injuries) 797 271 212 125 44* Area affected cleared of mines and ERW -- percent 30 48 58 87 Infrastructure Length of paved roads (primary & secondary) out of 45 2,100 2,186 2,580 4,100 11,618-kms Energy 46 Per capita use of electricity - Kwh 54 139 153 350 Sources: CDHS 2005, Population Census 2008 and MTR 2008. Note: Cambodia MDGs are marked with * 79 Annex Display 2. General government financing by functional classification (in millions of U.S. dollars) Functions 2001 2002 2003 2004 2005 2006 2007 2008 2009e 1 Core Government 215 231 186 171 201 280 355 412 768 1.1 General Admin & Judiciary 109 127 83 63 90 153 202 216 423 1.2 Defense 71 68 68 68 71 80 92 123 234 1.3 Security 36 36 36 41 40 48 61 72 110 2 Economic Services 96 142 148 159 182 290 264 394 492 2.1 Agriculture 21 22 18 24 12 108 34 42 67 2.2 Transport 63 79 66 91 89 73 127 187 248 2.3 Other Economic Services 13 29 49 26 71 99 105 167 170 2.4 Environment 2 17 20 22 15 17 7 6 7 3 Rural Development 43 30 11 49 32 39 60 58 63 4 Social Services 228 246 276 261 282 360 410 426 490 4.1 Health 58 65 76 91 137 157 178 204 257 4.2 Education 64 105 107 110 108 145 165 164 214 4.3 Other Social Services 106 76 93 60 37 58 67 58 19 5 Other 121 136 207 131 147 116 196 405 217 6 Emergency & Food Aid - - - - 3 0 0 0 1 Total 703 785 828 771 846 1,085 1,285 1,695 2,032 o/w Treasury-Executed 465 512 590 530 580 671 870 1,167 1,464 o/w Domestically financed 442 490 578 530 569 671 834 1,158 1,445 o/w Externally financed 262 295 250 241 277 414 451 537 587 Sources: MEF, CDC, and Bank staff estimates Annex Display 3. General government by functional classification: treasury executed component (millions of U.S. dollars) Functions 2001 2002 2003 2004 2005 2006 2007 2008 2009e 1 Core Government 175 180 188 187 201 242 303 370 728 1.1 General Admin. & Judiciary 69 76 85 78 91 115 151 174 383 1.2 Defense 71 68 68 68 71 80 92 123 234 1.3 Security 36 36 36 41 40 48 61 72 110 2 Economic Services 36 38 41 48 61 77 93 118 185 2.1 Agriculture 8 10 10 10 12 14 14 16 20 2.2 Transport 7 5 5 14 21 23 33 47 85 2.3 Other Economic Services 24 26 28 26 31 44 51 59 75 2.4 Environment 1 2 2 2 2 3 3 4 5 3 Rural Development 3 5 4 19 15 18 21 33 36 4 Social Services 113 147 153 167 176 212 256 268 317 4.1 Health 33 42 44 49 56 64 85 92 131 4.2 Education 53 74 76 82 86 109 123 125 176 4.3 Other Social Services 27 31 34 37 34 40 47 51 9 5 Other 137 143 205 109 128 123 197 378 199 Total 465 513 590 530 580 671 870 1167 1465 percent of GDP 12 12 13 10 9 9 10 11 14 Sources: MEF and Bank staff estimates 80 Annex Display 4. General government by functional classification: externally financed component (in millions of U.S. dollars) Functions 2001 2002 2003 2004 2005 2006 2007 2008 2009e 1 Core Government 40.0 50.7 0.0 0.0 0.0 38.2 51.6 41.7 40.4 1.1 General Admin. & Judiciary 40.0 50.7 0.0 0.0 0.0 38.2 51.6 41.7 40.4 1.2 Defense - - - - - - - - - 1.3 Security - - - - - - - - - 2 Economic Services 59.7 104.4 106.0 110.8 120.3 213.4 170.6 276.3 306.4 2.1 Agriculture 13.2 12.1 8.1 14.4 0.3 94.5 19.5 26.1 47.0 2.2 Transport 56.0 73.9 60.6 76.8 68.2 50.0 93.3 139.5 162.6 2.3 Other Economic Services - 3.1 19.1 - 39.4 54.4 53.7 108.8 95.2 2.4 Environment 1.0 15.3 18.2 19.6 12.3 14.6 4.0 1.9 1.6 3 Rural Development 40.0 25.5 6.6 30.1 17.0 21.2 38.9 24.9 27.3 4 Social Services 114.7 98.7 123.2 94.1 106.5 147.9 154.4 157.9 173.6 4.1 Health 25.1 22.8 32.2 42.9 81.3 93.2 92.9 111.9 126.1 4.2 Education 10.3 30.8 31.4 28.0 22.1 36.0 41.3 38.9 37.3 4.3 Other Social Services 79.3 45.2 59.6 23.2 3.1 18.7 20.1 7.0 10.1 5 Other - - 1.7 6.8 18.8 - - 27.5 18.5 6 Emergency & Food Aid - - - - 3.0 0.4 - 0.3 1.0 Total excluding budget & BOP support 238.8 272.1 237.6 241.8 265.6 414.1 414.8 528.6 567.1 Budget & BOP Support 23.0 22.7 12.4 - 11.1 - 36.0 22.4 19.9 Total including budget & BOP support 261.7 294.8 250.1 241.8 276.7 414.1 450.8 551.0 587.0 In percent of GDP 6.6 6.9 5.4 4.5 4.4 5.7 5.2 5.3 5.7 Sources: CDC, MEF, and Bank staff estimates Annex Display 5. Externally-financed technical assistance and NGOs’ own resources (millions of U.S. dollars) NGO's own Technical assistance resources Functions 2007 2008 2009e 2008 2009e 1 Core Government 56.3 78.2 69.2 0.2 0.1 1.1 General Admin. & Judiciary 56.3 78.2 69.2 0.2 0.1 1.2 Defense 1.3 Security 2 Economic Services 73.8 68.0 75.3 3.0 6.1 2.1 Agriculture 25.9 30.0 42.8 1.2 1.3 2.2 Transport 4.1 3.5 9.4 - - 2.3 Other Economic Services 39.6 30.3 16.3 0.2 0.2 2.4 Environment 4.3 4.2 6.7 1.5 4.6 3 Rural Development 28.1 36.2 33.0 1.3 1.4 4 Social Services 81.4 103.4 112.4 100.4 95.6 4.1 Health 27.2 37.4 48.4 41.9 35.7 4.2 Education 28.3 39.7 38.5 26.8 27.5 4.3 Other Social Services 25.9 26.3 25.4 31.7 32.3 5 Other 7.5 11.3 6.2 0.1 0.2 6 Emergency & Food Aid 1.9 2.3 3.2 - - Total excluding Budget & BOP Support 248.9 299.4 299.2 104.9 103.3 Budget & BOP Support - 13.9 0.5 - - Total including Budget & BOP Support 248.9 313.3 299.7 104.9 103.3 As a percent of GDP 2.9 3.0 2.9 1.0 1.0 Sources: CDC and Bank staff estimates 81 Annex Display 6. NSDP resource requirements and actual allocations (in millions of U.S. dollars) Actual NSDP Summary of Functions MTR 2006-2009 (2006-2010) 1 Core Government 994 220 270 1.1 General Administration 994 220 270 1.2 Defense - - - 1.3 Security - - - 2 Economic Services 1,440 1,410 1,775 2.1 Agriculture 251 350 570 2.2 Transport 634 550 690 2.3 Other Economic Services 542 410 515 2.4 Environment 37 100 120 3 Rural Development 220 350 420 4 Social Services 1,686 1,150 1,390 4.1 Health 796 600 720 4.2 Education 687 550 670 4.3 Other Social Services 202 - - 5 Other 936 370 225 Total (excluding defense and security) 5,276 3,500 4,080 Financing (excluding defense and security) 5,276 3,500 Domestically financed 3,288 1,099 Externally financed 1,989 2,401 Sources: NSDP 2006-2010, Mid-Term Review, and Bank staff estimates Annex Display 7. MOH program budget allocation, 2009 (in millions of riels unless indicated otherwise) Program Central In % National Central National SOA Program Budget total Centers Departments Hospitals Funds Maternal and child health 7,763 2.2 6,543 754 466 0 Communicable diseases 17,665 5.2 6,145 2,227 9,294 0 Non-communicable diseases 888 0.3 733 56 100 0 Health systems strengthening 13,309 3.9 1,705 2,001 2,886 6,720 Non-program 303,195 88.4 26,690 246,320 30,184 0 Total 342,819 100.0 41,816 251,3578 42,926 6,720.0 Source MOH Budget Book 2009 82 Annex Display 8. Government budget allocations for health, 2000-2009 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 National nominal health expenditure 101.8 129.7 164.4 173.0 192.1 224.5 260.8 343.3 426.8 505.2 GDP 14,083.0 15,617.0 16,781.0 18,535.0 21,438.0 25,754.0 29,849.0 35,039.0 45,583.0 45,026.0 National health expenditure at constant prices 101.8 126.4 159.0 164.4 174.1 191.9 213.0 263.3 291.5 315.8 National Health Expenditure as a % of GDP 0.72% 0.83% 0.98% 0.93% 0.90% 0.87% 0.87% 0.98% 0.94% 1.12% Per capita expenditures on health in US$ nominal rates 2.17 2.85 3.57 3.38 3.61 4.09 4.40 5.83 7.79 9.20 Per capita expenditures on health in US$ (constant rates 2000=100) 2.17 2.78 3.45 3.21 3.27 3.49 3.59 4.47 5.32 5.75 Public health expenditures as % of GDP 0.72% 0.83% 0.98% 0.93% 0.90% 0.87% 0.87% 0.98% 0.94% 1.12% Total national public expenditures at current prices 1215.48 1415.63 1491.17 1758.14 1745.65 1967.46 2354.69 3460.41 4587.71 4439.69 Total national public expenditures at constant prices (2000=100) 1215.48 1379.18 1442.44 1670.65 1582.45 1681.35 1923.22 2653.36 3133.08 2774.80 % of total 8.38% 9.17% 11.02% 9.84% 11.00% 11.41% 11.08% 9.92% 9.30% 11.38% Annex Display 9. Employment at the Ministry of Health 2009 2008 2007 2006 2001 1996 Medical Doctors 2,159 2,196 2,165 2,108 2,055 1,247 Medical Assistants 1,210 1,258 1,277 1,294 1,425 1,458 Pharmacists 421 434 434 386 382 n/a Primary Nurses 5,052 5,186 5,014 4,524 3,808 n/a Secondary Nurses 3,385 3,534 3,492 3,346 4,275 n/a Primary Midwives 1,797 1,844 1,850 1,796 1,243 n/a Secondary Midwives 1,437 1,478 1,351 1,047 1,786 n/a Total health staff 16,504 17,002 16,630 15,434 15,927 15,594 Total MOH staff 17,988 18,592 18,272 17,047 17,810 18,233 Doctors per 100,000 population 22.76 23.50 23.58 23.63 29.02 25.58 Nurses and midwives per 100,000 78.86 81.92 80.18 74.40 92.60 108.69 Nurses per doctor 3.46 3.49 3.40 3.15 3.19 4.30 Health professionals to other staff 92 91 91t 91 89 86 Sources: Personnel Dept, MOH; staff calculations 83 Annex Display 10. CMS invoice prices for ciprofloxacine are higher than international prices (in U.S. dollars, 500mg tab/capsule) Source: Sample invoices from provincial hospitals; World Health Organization and Health Alliance International, Medicine Price Surveys website, 2010 (http://www.haiweb.org/medicineprices/). Annex Display 11. Prices for ciprofloxacine are lower at private pharmacies than abroad (in U.S. dollars, 500mg tab/capsules, 14 tablets) 30 25 20 15 10 5 0.56 0 Dominican Republic… Yemen (Sana) Argentina (La Plata) Cambodia (Phnom Penh) Ghana (Accra) Philippines (Manila) Bangladesh (Dhaka) UAE (Abu Dhabi) Pakistan (Lahore) Chile (Santiago) Suriname (Paramaribo) Armenia (Yerevan) Nigeria (Abuja) Colombia (Bogota) Kyrgyzstan (Bishek) Russia (Moscow) Thailand (Bangkok) Ecuador (Quito) Turkey (Bursa) Indonesia (Depok) Egypt (Cairo) Jordan (Amman) Source: World Health Organization and Health Alliance International, Medicine Price Surveys website, 2010 (http://www.haiweb.org/medicineprices/). 84 Annex Display 12. Performance indicators, 2004-2010 PLATFORM 2 (2008) INDICATOR 2004 BASE LINE PLATFORM 1 (2005) TARGET Oct-05 JAN 06 OCT 06 DEC 07 DEC 08 DEC 09 TARGET 1) Revenue out-turn Outturn: 92% ƒ Outturn within 95% of domestic Outturn 67.6% December 2005: 114%. Outturn as of October 06: Revenue outturn Dec +/-5% of forecast used for Revenue outturn Dec Revenue outturn Dec 09: increasingly close to revenue for 2005 92 % (pro rata 83%) 07: 117.6% budget 08: 132.4% 95.1% targeted level in approved budget 2) No accumulation of new Outstanding arrears: riel ƒ No new arrears arising from in-  arrears and steadily 554 billion year transactions.  declining stock reduction of 23% ƒ Reduced by 41,8% from ƒ Reduced by 17% Reduced to 0 % Arrears as defined to be kept Reduced to 0 % Reduced to 0 % achieved in 10 months 372,833MR to at no more than 2% of total 216,811MR. expenditure ƒ [20%] reduction in stock of old arrears at end of 2004. 3) Budget holders ƒ Improvement in profile  increasingly able to as below: commit expenditure in line 1st Quarter [7 ]% st 1 Quarter [9 ]% ƒ 1st Quarter: 15.48% ƒ 1st Quarter: 12.0% ƒ 1st Quarter: 15.0% ƒ 1st Quarter: 14.8% ƒ 1st Quarter: 17.9% with budgets and cash flow forecasts 2nd Quarter[18 ]% 2nd Quarter[25 ]% ƒ 2nd Quarter: 47.5% ƒ 2nd Quarter: 41.0% ƒ 2nd Quarter: 45.0% ƒ 2nd Quarter: 43.8% ƒ 2nd Quarter: 41.8% 3rd Quarter[36 ]% 3rd Quarter[45 ]% 3rd Quarter[67.4 ]% ƒ 3rd Quarter: 72.2% ƒ 3rd Quarter: 62.1% ƒ 3rd Quarter: 67.0% ƒ 3rd Quarter: 71.7% ƒ 3rd Quarter: 71.3% 4th Quarter [64]% 4th Quarter [70]% ƒ 4th Quarter profile ƒ 4th Quarter: 94.3.0% ƒ 4th Quarter: 96.0% ƒ 4th Quarter: 107.0% ƒ 4th Quarter: 105% 96.3%. No transparency on ƒ Plan established and piloted to ƒ Plan was established and releases of mandates or improve transparency on release is implementing along plan to decentralize of mandates and to decentralize with de-concentration commitment and commitment and payment financial control and payment authority authority. streamlines commitment/ payment process implementation. ƒ Internal Audit sections ƒ Internal Audit sections in [5] No plan established yet ƒ 5 Ministries have ƒ 5 Ministries have ƒ 25 Ministries have  ƒ 26 Ministries and 3 ƒ 26 Ministries have in [2] Ministries.  Ministries. established internal audit established internal audit established internal SOEs have established established internal audit department and two are department and two are audit department and 6 internal audit department and 18 operational.  operational are operational department and 6 are operational operational ƒ Percentage of ƒ Percentage of payments to § Percentage achieved: ƒ Percentage achieved:  ƒ Percentage achieved not ƒ Percentage achieved: ƒ Percentage achieved: payments to creditors creditors and staff made through 67% 67% less than 80%. 100% 80.97% and staff made through banking system in 2005 [20] %. banking system in 2005 [5] %. ƒ Percentage of tax ƒ Percentage of tax revenue § Percentage achieved: ƒ Percentage achieved: ƒ Percentage achieved not ƒ Percentage achieved: ƒ Percentage achieved: revenue collected though collected though Banks in 2005 25% 29.9% less than 60% 100% 95.5% Banks in 2005 [2]% [20]%  4) Service delivery units The proportion of ƒ Identify these elements in 2006 (schools, health centers) budgets to be allocated budget. receive an increasing to service units not proportion of funds identified in budget targeted at their levels (and ƒ Develop an improvement profile of goods and services measures meant to be procured for them) increasingly timely. 85 5) Public procurement No systematic bid ƒ Implementation of Decentralization ƒ Sub-decree No. 60 on based on clear rules, announcing and improvements in each of these implemented. procurement is being consistently enforced. No dissemination of bid areas as planned in activities 13.3, reviewed. Decentralization major delay in processing documents. 13.6 and 13.7 of the program. (activity 13.9) is being and payment implemented. Product delivery does ƒ Reduction in price differential to Sub decree is reviewed. ƒ Not possible to evaluate. not meet requirements, 15% Technical assistance Post review process is unavailable of good, recruited to develop post being developed after inability to exchange review process decentralization and about goods, inadequate 70% of procurements now evaluation, payments do conducted by spending not comply with agencies.  contracts 6) Composition of ƒ Target fit of outturn to budget ƒ Profile of achievements: ƒ Profile of achievements: ƒ Profile of Outturn compared to budget ƒ Profile of ƒ Profile of expenditure by type (staff achievements: to be not less than: achievements: achievements: costs, non-staff costs etc) Salaries [92 Salaries [94 ]% Salaries Salaries Salaries 83.0 Salaries Salaries 98.0 % Salaries 115.8 Salaries 107.8 close to approved budget. ]% 73.4% 97.8% % 92.2 % % % Goods/Services Goods/Services [85]% Goods/Services Goods/Services Goods/Services 75.0 Goods/Services Goods/Services 95.0 % Goods/Services 124 Goods/Services106.7 % [81]% 62.6% 102.6% % 99.9 % % Capital [35]% Capital [55]% Capital Capital Capital Capital Capital 85.0 % Capital 96.7 Capital 93.5 % 68.9% 83.3% 63.8% 83.1 % % 7) Better yield achieved ƒ Annual increase in non-tax § Increase in non-tax ƒ Achievement in increase ƒ Achievement in increase ƒ Achievement in Progressive improvement in ƒ Achievement in ƒ Achievement in from tax base through revenue in real terms of [15]%. revenue over the same in non-tax revenue: 6.2% in tax revenue: 24 % over increase in tax revenue yield compared to increase in Non-tax increase in Non-tax improved collection period: 19.15% the same period in 2005. revenue: 21.3% GDP by 0.5% per annum revenue: 33.9% revenue: -13.4% efficiency and planned use of non-tax sources. ƒ Annual growth in tax revenue:  § Achievement in ƒ Achievement in increase ƒ Achievement in increase ƒ Achievement in  ƒ Achievement in ƒ Achievement in increase in tax revenue: in tax revenue: in tax revenue: increase in tax increase in tax increase in tax revenue: revenue: revenue: Direct tax [ 3 ]% Direct tax [ 5 ]% Direct tax: 45.93% Direct tax: 40.6% Direct tax: 42% Direct tax: 40.7% Direct tax: 34.5% Direct tax: -0.5% Indirect tax [9 ]% Indirect tax [9 ]% Indirect taxes: 47.42% Indirect taxes: 35.6% Indirect taxes: 17% Indirect taxes: Indirect taxes: Indirect taxes: -3.3% 16.3% 33.9% 8) Single and orderly Fragmented budget ƒ Develop plan for integration.  Preparing strategic plan ƒ Strategic plan is being ƒ With the introduction of ƒ With the introduction  budget process producing process. No forum for is considering within the developed. new budget classification of new budget good (and integrated) substantive discussion amendment of budget both recurrent and capital classification and BSP budget plans about budget priorities law budget will be presented in both recurrent and and justification. Budget the budget 2007 document capital budget was cycle starts in June ends however the integration of presented in the budget in Dec the preparation processes 2007 document. The is still in the process of BSP preparation has developing mechanism.  been improved and a technical workshop was held with LMs. ƒ Bring forward start of cycle to ƒ New budget framework ƒ The introduction of ƒ Budget preparation  [April] to allow more time for has been developed and Strategic Budget Plan for 2008 started in Ministry preparation and analysis.  discussed including the (SBP), 2007 budget March 2007.  proposal for starting the preparation started in May cycle early (March/April) however, the amendment and new formats and of the budget law proposed processes for budget to start from March.  preparation called Ministry Budget Strategy Framework (MBSF) which requires the budget proposal to include both recurrent and capital budget for three year basis with a statement link to National Strategy development Plan (NSDP). ƒ Budget hearing take place.  ƒ A booklet format of Macro-Fiscal data is being developed. ƒ Form of summary publication for information of general public piloted for 2006 budget 86 9) All significant areas of Major areas of exclusion ƒ Develop and implement a Strategy is being ƒ Strategic plan has been ƒ Strategic plan has been ƒ The establishment of No significant areas of ƒ Strengthening TSA, ƒ Continue strengthening both public revenue and are: investment strategy for inclusion of these developed. developed and is developed and is TSA, the consolidation omission the consolidation of TSA, the consolidation expenditure captured in expenditure funding missing areas. implementing in several implementing in several of Gov' t bank Gov' t bank accounts, of Gov' t bank accounts, both the budget and directly from loans and fronts such as introduction fronts such as introduction accounts, the the implementation of the implementation of accounts of the external financing of new COA, new budget of new COA, new budget introduction of new new COA, new budget new COA, new budget Government. agencies; and classification, new TOFE classification, new TOFE COA, new budget classification, and the classification, and the expenditure funded from with GFS classification, with GFS classification, classification, and the introduction of BSP introduction of BSP are revenue not declared to and MBSF.  and MBSF.  introduction of BSP are all contributed to all contributed to an treasury are all contributed to an improved capture of improved capture of both an improved capture of both revenue and revenue and expenditure both revenue and expenditure in the Gov' in the Gov' t account. expenditure in the Gov' t account. t account. ƒ MEF continues to  ƒ MEF continues to ƒ MEF continues to reduce number of reduce number of reduce number of government banks government banks government banks account, currently at account, currently is account, currently is 90.  626.  187 and 13 are frozen.  Database system is DMFAS DMFAS implementation being developed to implementation with improved improved capturing/ management and monitoring of external monitoring of external financing. financing. 10) Budget has direct Personnel information ƒ Implementation of an agreement Working party formed ƒ Information sharing and ƒ Information sharing and ƒ Information sharing  influence over nominal not used by MEF in between CAR and MEF that CAR and its working closely working relationship have working relationship have and working staff roll (establishment). setting staff related will provide information on with CAR been improved and a been improved and a relationship have been budget establishment and planned working group is working group is improved and a changes in the number of posts as established and working established and working working group is a basis for budget formulation and closely with CAR to closely with CAR to established and that CAR will take into account establish a mechanism for establish a mechanism for working closely with the available budget before a systematic flow of a systematic flow of CAR to establish a approving changes in posts. information and clearer information and clearer mechanism for a roles and process of MEF roles and process of MEF systematic flow of and LMs in improving and LMs in improving information and clearer human resource human resource roles and process of management. management. MEF and LMs in improving human resource management. Budget information is   ƒ Civil Service Secretariat Civil Service not considered by CAR is now consolidated a Secretariat in deciding whether to requested for new consolidated a approve changes in recruitment from LMs and requested for new establishment and posts sent to CAR for final recruitment from LMs decision and copy to MEF and sent to CAR for for budget 2007 final decision and copy preparation. to MEF for budget 2008 preparation. 11) Clearer/more accurate No clear legal ƒ Legal framework established. Legal framework is ƒ Legal Framework is ƒ Legal Framework is ƒ TOFE submission is  overview of public framework for supply of being established and being established and being established and not always within 2 finances regularly data to MEF. some improvements some improvements have some improvements have weeks. available (TOFE?) based have been made in been made in capturing been made in capturing on improvements of capturing missing areas. missing areas. missing areas. existing system pending introduction of IFMIS. Accounting data suffers ƒ Progress in capturing some No quantitative measure ƒ No quantitative measure ƒ No quantitative measure No review was made from lack of missing areas. made made made by the internal audit. completeness. Substantial ƒ Inconsistencies reduced to no inconsistencies between more than [5%] by value.  final TOFE for the year (mid-following year) and interim report at end of financial year. 87 12) System in place to Significant budget ƒ Excesses reduced to no more Review completed. ƒ Post budget ƒ No supplementary Supplementary ensure that proposals for excesses. No system in than [2]% of total budget. Problem identified and supplementary does not without source and it does expenditure credit is post-budget supplementary place for ensuring that proposal for reducing excess the total budget. not exceed the total based on priority and expenditure credits are in-year proposals for post-budget budget. urgency and the always accompanied by an new expenditure have supplementary is being provision has always MEF report on realistic identified funding prepared. been caped within options for financing the sources contingency fund. expenditure involved. ƒ System established for ensuring ƒ Review completed. funding sources identified in every Problem identified and case. proposal for reducing post- budget supplementary is being prepared. 13) Institutionalized mid- Mid-year review started ƒ Plan established in advance for Plan established. ƒ Plan established. ƒ Mid-year review was Mid-year review has year budget review feeding late in 2004, results not conduct of review for trialing in conducted in July and the been conducted in July into 2nd half year budget implemented and the 2005. result has been used for 2007 and review result implementation/budget authority and process not improving implementation was used for the preparation for next year. clear of the 2nd half through implementation of the quarterly revenue and second half of the year. expenditure plan and as a basis for budget 2007 preparation. ƒ Review commenced in May and Review commenced in ƒ Quarterly revenue  completed in June. July. and expenditure plan has been prepared and implemented. ƒ Procedures established in Procedures established The cash management advance for implementing unit has been findings. functioning. ƒ Plan established and mandates Further improvement is ƒ Review commended in clear authority and process. needed with regard to July.  mandatory, authority and processes. ƒ Procedures established.  ƒ Further improvement is needed with regard to mandatory, authority and process 14) Annual forecasts of all No cash flow plans in ƒ Quarterly average variance from Cash flow forecasts have ƒ A cash flow forecast was ƒ Quarterly revenue and in-flows and out-flows place forecast cash flow requirement: been prepared but further not prepared properly. expenditure plan has been prepared and regularly 2005 [+ or -15]% improvement is needed. Quarterly revenue and prepared and implemented. updated (based on accurate expenditure plan has been revenue forecasts and good prepared and implemented. budget implementation ƒ The cash management plans) unit which is under reorganized by the budget execution coordination group is working on developing a mechanism for developing cash plan. 88 Annex Display 13. High level traffic light performance Indicator for the Platform Two HIGH LEVEL PERFORMANCE MEASURES FOR PLATFORM AS DECEMBER 2009 PLATFORM OBJECTIVE: To improve accountability for achieving effective financial management so that public servants at all levels feel responsible for how they manage and control public finances, that they are empowered to fulfil those responsibilities effectively and that instruments for holding them accountable are in place and effective. Target Description Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 1) Establishment and application Review of existing rules Design of new rules on a Issue of new rules under Issue of guidance and of clear rules for both responsibility consolidated basis an appropriate authority provision of appropriate and empowerment of all budget training managers. Traffic lights - Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 2) Core FMIS system in place and Structure and System procurement Core system System producing providing timely and reliable capacities in place carried out and contracts implemented satisfactory required accounting data for both within MEF for in place reports in a consistent management and formal managing FMIS and reconciled manner reporting requirements. Hierarchy implementation of accounting reports established producing consistent and reconciled data to senior management in budget entities, within MEF, to Cabinet and to the legislature. Traffic lights Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 3) Successful implementation of Transactions being Required forms of Core FMIS Functional and program revised chart of accounts and full processed in accordance functional and program implementation takes classification introduced alignment between budget and with new economic classification researched account of required accounting classification at the classification and decided forms of functional and level of the formal budget program classification presentation. Traffic lights - Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 4) Full transparency achieved withSpecific measures to be Taxpayers have easy Active taxpayer A tax appeals system with regards to tax taxpayer obligation taken identified and access to comprehensive education campaigns in transparent administrative and liabilities consulted upon. user friendly and up-todate place. procedures and appropriate information about checks and balances tax liabilities and implemented administrative procedures for all major taxes. Traffic lights - Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 5) Achievement of an improved Reporting formats designed Consultation undertaken Arrangements in place for Reporting formats level of transparency in reporting with potential users of suchgathering data to service theimplemented and to Parliament and to the public. information new reporting formats distribution and access arrangements in place. Traffic lights - Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 6) Clear and transparent fiscal relationships between central and locally based public bodies established and adhered to. Traffic lights - Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 7 Clear sanctions in place and Structure of sanctions Changes in law Promulgation and training Effective checks of applied for inappropriate resource reviewed and new implemented to reflect provided compliance and management activity or results. integrated framework new structure. application of sanctions designed in place Traffic lights - Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 8) Clear rewards in terms of trust, Potential freedoms and Changes in law Promulgation and training Objective assessments freedoms and flexibility establish flexibilities identified and implemented to reflect provided being made as a basis for and applied for those budget a framework designed new structure. allowing appropriate managers who meet criteria for freedoms and flexibilities adequate standards of financial on a selective basis. management. Traffic lights - Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 9) Strong post audit arrangementsInternal Audit Strategic Internal Audit functional Reports issued regularly Reports responded to in place through internal audit Plan/Charter in most important for most Entities and actions implemented sections providing system based Developed Government Entities and Implementation of by most managers and risk assessment based audit Manual and Training undertakes systems review Training Programmes coverage. Programmes Developed Traffic lights- Milestone Achievement Milestone 1 Milestone 2 Milestone 3 Milestone 4 10) Minimum capacities Support to EFI and Review of Capacity Review of Capacity Review of Capacity established in key line ministries Personnel Department in Development Plan Development Plan Development Plan with regards to skill set required f place exercising of effective Measures in skill accountability. development set out in the CDP being implemented Traffic lights- 89 Annex Display 14. Regional comparison of PEFA scores PEFA Assessment Comparison Cambodia Thailand Lao PDR Note: Shaded areas represent M2 scoring methodology Overall Tentative Score score A. Credibility of the Budget B B Aggregate ex penditure out-turn compared to original approv ed PI-1 budget M1 B A NR Composition of ex penditure out-turn compared to original PI-2 approv ed budget M1 D Aggregate rev enue out-turn compared to original approv ed A A PI-3 budget M1 A A C+ PI-4 Stock and monitoring of ex penditure pay ment arrears M1 C+ B. Comprehensiveness and Transparency PI-5 Classification of the budget M1 C B C Comprehensiv eness of information included in budget PI-6 documentation M1 B B B B+ D+ PI-7 Ex tent of unreported gov ernment operations M1 C D+ D PI-8 Transparency of inter-gov ernmental fiscal relations: M2 C+ Ov ersight of aggregate fiscal risk from other public sector A D+ PI-9 entities M1 C+ PI-10 Public Access to key fiscal information M1 C B C C ( i) Policy-Based Budgeting A C+ Orderliness and participation in the annual budget process PI-11 M2 A C D+ Multi-y ear perspectiv e in fiscal planning, ex penditure policy PI-12 and budgeting M2 B C ( ii) Predictability and Control in Budget Execution A D+ PI-13 Transparency of tax pay er obligations and liabilities M2 B B C Effectiv eness of measures for tax pay er registration and tax PI-14 assessment M2 C A NR PI-15 Effectiv eness in collection of tax pay ment M1 D+ Predictability in the av ailability of funds for commitment of PI-16 ex penditures M1 C+ A B+ Recording and management of cash balances, debt and B+ D+ PI-17 guarantees M2 C+ B C+ PI-18 Effectiv eness of pay roll controls M1 D+ B D+ Competition, v alue for money and controls in procurement PI-19 M2 C Effectiv eness of internal controls for non-salary ex penditure C+ D+ PI-20 M1 C PI-21 Effectiv eness of internal audit M1 D+ C+ D C ( iii) Accounting, Recording and Reporting C+ C PI-22 Timeliness and regularity of accounts reconciliation M2 C Av ailability of information on resources receiv ed by serv ice B D PI-23 deliv ery units M1 C B+ C+ PI-24 Quality and timeliness of in-y ear budget reports M1 C+ C+ D+ PI-25 Quality and timeliness of annual financial statements M1 D+ C ( iv) External Scrutiny and Audit PI-26 Scope, nature and follow-up of ex ternal audit M1 D+ B D+ B+ C+ PI-27 Legislativ e scrutiny of the annual budget law M1 NU 1/ D C+ PI-28 Legislativ e scrutiny of ex ternal audit reports M1 NU D. Donor Practices D+ D-1 Predictability of Direct Budget Support M1 C Financial info prov ided by donors for budget, reporting on C+ D-2 project, programme aid M1 D Proportion of aid that is managed by use of national D D-3 procedures M1 D 90 References Akroyd, S. and L. Smith, 2007, “Review of Public Spending to Agriculture,” Joint DFID/World Bank study. Annear, Peter Leslie. 2010. “A comprehensive review of the literature on Health Equity Funds in Cambodia,” Manuscript. Phnom Penh. Bigdeli, Maryam & Peter Leslie Annear. “ Barriers to access and the purchasing function of health equity funds: lessons from Cambodia.” Bull World Health Organ 2009;87:000–000. Bloom, E, Bhushan, I, Clingingsmith, D “Contracting for health: evidence from Cambodia: Improving service delivery by contracting out to NGOs.” Brookings Institution, 2006. Cambodian Rehabilitation and Development Board Council for the Development of Cambodia. Cambodia Aid Effectiveness Report. . May 2007 Cambodia Development Cooperation Forum. 2009. Cambodia Aid Effectiveness Report. -----. 2010. Cambodia Aid Effectiveness Report. CRDB/CDC, 2010, “Cambodia Aid Effectiveness Report,” Royal Government of Cambodia. Dans, A. and others. “The rise of chronic non-communicable diseases in southeast Asia: time for action.” The Lancet: Health in Southeast Asia. January 2011. Ecorys, “Public Expenditure Review Cambodia: Education Sector”, Draft Final Report, Rotterdam, March 2011. EIC, 2006, “Assessment of Corruption in Cambodia’s Private Sector”, Economics Institute of Cambodia, July 2006, Phnom Penh, Cambodia Filmer, Deon and Norbert Schady, “Impact Evaluation Brief: Promoting Schooling through Scholarships in Cambodia: Evidence from the Cambodia Education Sector Support Project”, World Bank, 2009 Fukao, Tsuyoshi, “Working Paper on Textbooks Analysis for IFAPER Cambodia”, EASHD, The World Bank, February 2011. Fukao, Tsuyoshi, “Working Paper on Household Education Expenditures for IFAPER Cambodia”, EASHD, The World Bank, February 2011. GiZ and MOH. 2011. “The Cambodian Demographic and Health Survey and the Cambodian Soci- Economic Surveys: Health Expenditure and Utilization Analysis Report.” Phnom Penh. Heller, Peter. 2005. “Back to Basics – Fiscal Space: What It Is and How to Get It”, IMF Finance and Development Quarterly Magazine, June 2005, Vol. 42, No. 2. http://www.imf.org/external/pubs/ft/fandd/2005/06/basics.htm Kanchanachitra, Churnurntai, Magnus Lindelow, Timothy Johnston, and others. “Human resources for health in southeast Asia: shortages, distributional challenges, and international trade in health services,” The Lancet: Health in Southeast Asia. January 2011. King H. et al. Diabetes and Associated Disorders in Cambodia: 2 epidemiological surveys, The Lancet, 2005; 350; 1633-1639 Lall, Ashish, “Returns to Education in Cambodia: Results from the 2007 Socio-Economic Survey”, LKY School of Public Policy, National University of Singapore, November 2008. Law on Administrative Management of the Capital, Provinces, Municipalities, Districts and Khans (unofficial translation), Royal Decree No. NS/RKM/0508/017, Royal Government of Cambodia 2008 Level in Kampong Cham Province.” A Brief Review Paper February 2009 Ministry of Economy and Finance. 2008. Comprehensive Action Plan for PFMRP. -----. 2009 Fiscal Performance Monitor 91 -----. 2009. Draft Medium Term Expenditure Framework 2010-2012. -----. TOFE, 2000-2009. Ministry of Health. 2010. Department of Planning and Health Information. Bureau of Health Economics and Financing. Annual Health Financing Report 2009. -----. 2009. Department of Planning and Health Information (DPHI), Bureau of Health Economics and Financing. Annual Health Financing Report 2008. -----. 2009. Health Information System Annual Report, 2008. -----. 2009. Master Plan for Social Protection. -----. 2009. Contracting of health services in Operational Districts, Final evaluation of contractor performance. -----. 2008. “Health Strategic Plan 2008-2015: Accountability, Efficiency, Quality, Equity,” April 2008. -----. 2006. Comprehensive Midwifery Review. -----. 2006. Health Sector Support Project, Mid Term Review. -----. Annual Operational Plan 2005-2010. -----. Department of Budget and Finance. Budget Book (annual from 2005 – 2010). -----. Department of Budget and Finance. “Expenditure Book” (annual from 2005-2008). -----. Human Resource Development Department . Second National Health Workforce Development Plan 2006-15. -----. Joint Annual Performance Review (annual from 2005-2009). -----. National Health Statistics Reports (2002-2008) -----. National Nutrition Programme. 2008. Nutrition in Cambodia 2008, An Analysis of Nutritional Status, Trends and Responses. MOH and Conseil Sante. 2007. Cambodia health Services contracting Review. MOH and OPM. 2005. Cambodia Health Workers Incentive Survey. MOH HSSP2. 2009 Annual Performance Monitoring Report MOH (USAID/BASICS). 2009. Cost and Funding Projections for the Minimum Package of Activities for Health Centres. -----. (USAID/BASICS). 2009. Cost Projections for the Complementary Package of Activities for Referral Hospitals. MOH, World Health Organization, Belgian Technical Cooperation. 2006. Health Equity Fund Forum Report MOP, 2005, “National Strategic Development Plan (2006-2010)” and MOP 2008, “Mid-Term Review of NSDP (2006-2010)”, Royal Government of Cambodia. MOP (Ministry of Planning) . 2006. A Poverty Profile of Cambodia 2004. -----. National Strategic Development Plan 2006-2010. -----. Public Investment Programme 20010-12. -----. Public Investment Programme 2007-09. -----. Public Investment Programme 2008-10. -----. ROYAL GOVERNMENT OF CAMBODIA Public Investment Programme 2009-11 Murakami, Hitoshi, Mari Nagai, Sadatoshi Matsuoka and Hiromi Obara. 2009. “ Performance-based Financing of Maternal and Child Health Services: Financial and Behavioral Impacts at the Field National Institute of Public Health and National Institute of Statistics. 2006. Cambodia Demographic and Health Survey 2005. 92 National Institute of Statistics (NIS) and MOH. 2011. Cambodia Demographic and Health Survey 2010: Preliminary Report. National Public Health Institute. 2009. National Emergency Obstetric and Newborn Care Assessment in Cambodia. Phnom Penh. National Reproductive Health Program, MOH. 2007. Reproductive Health Costing 2006-2015. Naudeau, Sophie, Sebastian Martinez and Deon Filmer, “Early Childhood Development: Building Evidence in Low-Income Settings: Examples from Mozambique and Canada”, APEIE Workshop Presentation, Ghana, May 11, 2010, World Bank. NBC, Annual Supervision Report of the National Bank of Cambodia, 2007-09, Phnom Penh, Royal Government of Cambodia. Orivel, Francois, “Financing of Higher Education in Cambodia”, Department of Higher Education, Ministry of Education, Youth and Sport, Cambodia, January 2009. PR Office GFATM . Drug Quality Assurance System in Cambodia, Towards a More Responsive System. Savedoff, William. 2004. “Tax based financing for health systems: Options and experiences.” Department "Health System Financing, Expenditure and Resource Allocation" (FER) Cluster "Evidence and Information for Policy" (EIP) WHO Discussion Paper #4. Geneva. Tangchanroensathien, V., and others. “Health-finance reforms in southeast Asia: challenges in achieving universal coverage,” The Lancet: Health in Southeast Asia. January 2011. UN Millennium Campaign, 2010, “Millennium Development Goals Report Card”, End Poverty 2015 Millennium Campaign Overseas Development Institute. USAID and URC (University Research Corporation). 2009. Evaluation Report: Health Equity Funds Implemented by URC and Supported by USAID. Phnom Penh. Vickery C, et al. 2001, Private Practitioners in Phnom Penh: A Mystery Client Study, Draft Report, Phnom Penh Urban Health Project, Options Wagstaff, A., Van Doorslaer E., Van Der Burg, H. et al. “Equity in the Finance of Health care: Some Further International Comparisons”, Journal of Health Economics, 1999. 18: 263-290. WHO. 2010. World Health Report 2010. Geneva -----. 2010. Global Tuberculosis Report 2010. Geneva. -----. 2009. National Health Accounts Database. Geneva. -----. 2008. Commission on Macroeconomics and Health. -----. 2008. Global Burden of Disease. Geneva. WHO and MOH. 2007. Scaling up for better Health in Cambodia. Phnom Penh, Cambodia. WHO-UNFPA-UNICEF-World Bank Joint Country Support. 2009. “Plan for Accelerated Implementation of Maternal and Newborn Continuum of Care as part of Improving Reproductive Health, Mapping of in-country activities” Summary Report, August 2009. World Bank, “Cambodia Public Expenditure Tracking Survey (PETS) in Primary Education”, Report No. 34911-KH, December 2005. World Bank, “Teaching in Cambodia”, Human Development Sector Unit, June 2008. World Bank 2005, “Cambodia Rural Sector Strategy Note: Towards a Strategy for Rural Growth and Poverty Reduction”, June World Bank. 2010. “Health Financing Note: East Asia and the Pacific Region.” Human Development Department, East Asia and Pacific Region. Washington, D.C. -----. 2009. “Public Financial Management Assessment: Cambodia.” Phnom Penh. -----. 2008. Public Expenditure Tracking Survey in Health, Budget Years 2003-4. 93 -----. East Asia and Pacific Region. PREM Sector Unit. 2007. Poverty Profile and Trends in Cambodia: findings from the 2007 Cambodia Socio-Economic Survey (CSES). Phnom Penh. 94