Report No. 33102-ECA Operationalizing the Health and Education Millennium Development Goals in Central Asia (In Two Volumes) Volume II: Kyrgyz Republic Health and Education Case Studies June 2005 Human Development Sector Unit Europe and Central Asia Region Document of the World Bank CONTENTS Acknowledgements ... Preface........................................................................................................................................................ ................................................................................................................................... 111 iv Key MessagesandPolicyHighlights ......................................................................................................... v 1: The StatusandRelevanceofthe MDGsinCentralAsia .................................................................. 1 Introduction.............................................................................................................................................. 1 Trends .................................................................................................................................................. 2 Education MDGs...................................................................................................................................... 3 MDG: Achieve UniversalPrimary Education.................................................................................... 3 MDG: Promote Gender Equityand Empower Women ...................................................................... 4 Health MDGs........................................................................................................................................... 4 MDG: Combat HIV/AIDS. Malaria and Other Diseases.................................................................... 4 M D G : Reduce Child Mortality........................................................................................................... 6 MDG+: Addressing Adult Mortality by Targeting Non-communicable Diseases and Trauma .............88 M D G : Improve Maternal Health ........................................................................................................ MDG: Eradicate ExtremePoverty and Hunger .................................................................................. 9 M D G : Ensure Environmental Sustainability .................................................................................... 10 Prospects ................................................................................................................................................ 10 Regional Comparisons....................................................................................................................... 11 2: Operationalizingthe HealthandEducationMDGsinCentralAsia: A Frameworkfor Policyand ProgrammaticAction Application o f the Framework to the Kyrgyz Republic Health and Education Sectors: Key Findings. 15 .............................................................................................................................. 17 Sub-optimal Allocation Across Groups. Service Delivery Levels and Inputs...................................... 23 Policy and Programmatic Directions ..................................................................................................... 29 References .................................................................................................................................................. 39 BOXES Box 1: Conceptual Underpinnings of the Framework for Operationalizingthe Health and Education MDGsinCentralAsia............................................................................................................................. 16 Box 2: The Kyrgyz Health Reforms and Impact on Achieving the Health MDGs................................... 27 Box 3: Strengthening Provider and Consumer Incentives for Achieving the Health and EducationMDGs..................................................................................................................................... 29 FIGURES Figure 1: Trends inPrimary CompletionRates. 1998-2003 ....................................................................... 3 Figure2: Trends inIncidence o f TB, 1990-2003 ....................................................................................... 5 Figure3: Trends inPrevalence o f HIV (infection rates per millionpeople), 19962003........................... 5 Figure4: USMR, 2000 ................................................................................................................................ 6 Figure 5: Trends inUSMR, WDI & TMD, 1990-2003 .............................................................................. 7 Figure 6: Trend inMMR, 1980-2002 ......................................................................................................... 8 Figure 7: Primary Completion Rate, 2002................................................................................................. 12 Figure 8: Ratio o f Girlsto Boys inPrimary and Secondary Education (%), 2002.................................... 12 Figure9: Maternal MortalityRatio (modeled estimate, per 100,000 live births) ................................... 13 Figure 10: PovertyMDGTop priority and HighPriority Countries......................................................... 14 i Figure 11: There i s a Close CorrelationBetween Poverty Rates and Education Outcomes ..................... 17 Figure 12: Health Outcomes and Extreme Poverty Rates inthe Kyrgyz Republic are Broadly Correlated................................................................................................................................................ 18 Figure 13: The Health Sector Has Been Getting a Smaller and Smaller Share o f the Government Budget Since 1996 (health expenditures as % of total public expenditures) ....................................................... 21 Figure 14: Public Expenditures on Health are not Pro-poor (per capita health expenditures at 1995 prices, Kyrgyz Soms) ..................................................................................................................... 23 Figure 15: Shares o f Public Health and Primary Health Care are Small inthe Overall State 25 Figure 16: Distribution of Visits (primary care + outpatient specialist).................................................... Budget ..................................................................................................................................................... 27 Figure 17: Access to Hospital Care ........................................................................................................... 28 TABLES Table 1: Prospects o f ECA Countries Achieving the MDGs ...................................................................... 2 Table 2: U 5 M R Percentage o f Improvement. 1990-2003 ........................................................................... 7 Table 3: Overall Impact on Life Expectancy According to Different Scenarios (Central Asia) .................9 Table 4: Increase inIncidence of TB, 1990-2003 ..................................................................................... 13 Table 5: Selected Countries: GNIPer Capita and Public Spending on Health ......................................... 20 Annex Table 1: InternationalMDGs and their Modificationfor the Kyrgyz Republic ............................ 32 Annex Table 2: Cost-Effective Health Interventions for Reaching HealthMDGs and Coverage intheKyrgyz Republic............................................................................................................................ 33 Annex Table 3: List o f Organizations Whose RepresentativesWere Interviewedto Identify Health MDGBottlenecks and Interventions....................................................................................................... 38 11 ACKNOWLEDGEMENTS This report was prepared by a team led by Sarbani Chakraborty (ECSHD), Team Leader and principal author (Kyrgyz Health MDG Case Study and Chapter 2). The team included: Michael Mertaugh (Kyrgyz Republic Education Case Study), Elina Manjieva, Alan Wright, Shweta Jain and Julie Wagshal (all ECSHD) and Adilet Meimanaliev (Consultant). The peer reviewers for the study are: Sebnem Akkaya (Country Economist, ECSPE), Eduard Bos (Lead Population Specialist, HDNHE) and Rosemary Bellew (Lead Education Specialist, HDNED). The team benefited greatly from comments provided by the Quality Enhancement Review (QER) Panel consisting o fDaniel Ritchie, Akiko Maeda, Agnes Soucat, Robert Prouty, Rama Laksminarayan and Mercy Miyan Tambon. Comments were also providedby Chris Lovelace, Arup Banerji, Armin Fidler, Maureen McLaughlin and Peyvand Khaleghian. This volume is a product of the staff o fthe International Bank for Reconstructionand Development / The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views o f the ExecutiveDirectors o f The World Bank or the governments they represent. The World Bank does not guarantee the accuracy o f the data included in this work. The boundaries, colors, denominations, and other information shown on any map inthis work do not imply any judgment on the part o f The World Bank concerning the legal status o f any territory or the endorsement or acceptance o f such boundaries. iii PREFACE 1. The objective o f this report i s to present a generic framework for analysis and policy action for fuller attainment of the health and education MDGs in Central Asia. The framework focuses on the service delivery aspects o f achieving the MDGs and emphasizes: a) monitoring outcomes to identify groups which are not achieving MDGs, b) identifying interventions which are effective in raising MDG performance, c) targeting interventions to groups which are not meeting MDGs, and d) strengthening incentives to service providers and users to raise MDG outcomes for these groups. Where budget financing i s a binding constraint - as it usually i s - it also involves selectivity in the use o f budget resources to focus on improving outcomes for groups which are not meeting MDGs. ii. TheframeworkisillustratedwithcasestudiesonhealthandeducationintheKyrgyzRepublic. As the cases studies show, application o f the approach requires good information on the status o f MDG attainment, and the constraints to fuller attainment - including, crucially, qualitative dimensions o f MDG attainment, and resource constraints. The Kyrgyz Republic was chosen to illustrate the framework because prior analytical and project work provide a relatively rich base o f knowledge and experience and because there i s significant government ownership o f the key principles involved in the approach. The case studies do not aim to present new analysis o f the technical issues inthe health and education sectors in the Kyrgyz Republic. Rather, they aim to bring together relevant existing sector knowledge and lessons o f experience in order to derive the options for moving ahead more aggressively on MDG attainment inthe health and education sectors. iii. Theoptionsthatarepresentedinthecasestudiesarenotrecommendationsperse. Theywere derived as logical implications o f the identified capacities and constraints for MDG attainment, and are offered as options for Government consideration in the event that the political context endorses more aggressive pursuit o f MDG goals. Since capacities and constraints differ by country, the specific options for fuller MDG attainment will also differ. But the recommended fiamework for analysis and policy action i s intended to apply to all the Central Asian countries. Exceptions are Kazakhstan, where the more comfortable revenue prospects may not require the budget selectivity that would be needed to implement the options elsewhere in CentralAsia, and Turkmenistan, where we lack the requisite knowledge o f MDG status and the capacities and constraints for fuller attainment, and where we are not engaged in a dialogue on these issues. iv. The report acknowledges that fuller attainment o f the health and education MDGs will require key intersectoral actions (e.g. improving water and sanitation systems, reducingindoor air pollution). The report does not explicitly focus on these intersectoral linkages, except, in so far as these are related to scaling-up intersectoral actions in the context o f community-based health initiatives. The report also recognizes that health and education generate important spill-over benefits (i.e., education improves the uptake o f health behaviors and has positive impact on improving maternal and child health, healthy children are able to learn better). However, it does not focus on the analysis and policy implications of these synergies. v. The primary audience for the report i s policy makers in the Central Asian countries. The secondary audience i s donors and other stakeholders working on poverty reduction, health and education inthe CentralAsian countries. iv KEY MESSAGESAND POLICYHIGHLIGHTS vi. The Millennium Development Goals (MDGs) are internationally recognized as a key yardstick for measuring development progress. Inthe Europe and Central Asia (ECA) Region, the interpretation o f the MDGs differs slightly from the other regions. For the E C A Region, the global MDG poverty threshold o f U S $ l per day i s adapted to US$2.15 per day reflecting higher costs associated with heating, winter clothing and food. The education MDG focuses on completion o f the compulsory cycle - rather than the international target o f primary school completion; and for health, the focus i s on improving life expectancy through reduced adult mortality as well as measures to improve maternal and child mortality. This implies emphasizingnon-communicable diseases and trauma inaddition to communicable diseases. vii. Progress on MDGs in ECA will require accelerated action on the MDGs in Central Asia. According to the 2005 World Bank Global Monitoring Report, the ECA countries are at risk o f not achieving the health MDGs. Among C A countries, Central Asian countries' are at high risk o f not attainting the health MDGs. While all o f the Central Asian countries have attained full coverage at the primary-school level, there are significant shortfalls from the goal o f universal coverage through grade 11, which i s becoming the norm for the compulsory cycle. Of even greater concern i s the need to restore education quality in all grades following the deterioration o f teaching and learning conditions. Uzbelustan i s not likely to achieve the poverty MDGwhile the other Central Asian countries will achieve the poverty MDGonly with continuedprogress and Tajikistan i s one out o f three E C A countries unlikely to achieve the MDG on access to water. viii. The objective o f this report i s to present a generic framework for analysis and policy action for fuller attainment o f the health and education MDGsin Central Asia. The report recognizes that economic growth, alone, i s not likely to be sufficient for attainment o f the health and education MDGs. Attainment of these goals will also require scaling-up service delivery. A second key point to remember i s that although many sectoral reforms could improve welfare, only those reforms that particularly focus on those at the low end o f the distribution would affect MDGs. ix. The framework focuses on the service delivery aspects o f achieving the MDGs and emphasizes four key areas o f intervention that apply across Central Asian countries: (a) monitoring outcomes to identify groups which are not achievingMDGs, (b) identifying interventions which are effective inraising MDGperformance, (c) targeting interventions to groups which are not meeting MDGs, and (d) strengthening incentives for providers and users to raise MDGoutcomes. x. Where budget financing i s a bindingconstraint - as i s usually the case - it also involves selectivity inthe use o f budget resources to focus on improving outcomes for groups that are not meeting the MDGs. xi. No such generic framework for addressing the health and education MDGs currently exists in Central Asia. Prior work on the MDGs in Central Asia has focused on individual elements o f the framework (e.g. mentioning geographic disparities or the need for financing) and on monitoring the literal MDGs. Other work has focused on estimating the costs o f attaining the health and education MDGs. Central Asia includes the following countries: Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan and Uzbekistan. Given the difficulties o f data availability from Turkmenistan, it i s difficult to estimate the MDGs for this country. V While the role o f sound policies i s implicitly acknowledged inall the work there has been less attention to clearly spelling out the direction o f these policy actions from a sectoral perspective. The proposed framework brings together key operational requirements for attaining the health and education MDGs, and provides a direction for enhanced financing for the MDGsinCentral Asia. FINDINGS POLICYRECOMMENDATIONS AND xii. Monitoring Outcomes to Identify Groups that are not achieving the Health and Education MDGs: Disaggregatedmonitoring o foutcomes is instrumental to operationalizing an MDGstrategy that focuses on groups at the low end o f the distribution. Since health and education outcomes change slowly, in addition to monitoring outcomes, it is important to identify a set of intermediate variables that can serve as proxies for outcomes and monitor these from a disaggregated angle. Intermediate variables for monitoring the MDGs have been defined as a part o f the MDGagenda. Analysis that combines these two levels o f indicators will contribute to strengthening the analysis o f non-income dimensions o f poverty in poverty reduction strategies (PRS) and help in the formulation o f monitorable actions for attainment o f the MDGs in the context o f PRSPs, medium-term expenditure frameworks (MTEF) and sectoral programs. There are few and fragmented examples o f monitoring disparities in MDG outcomes and intermediate variables in Central Asia (e.g. the United Nations Development Assistance Program for Uzbekistan, the Kyrgyz Republic MDGProgress Report). Inthe few cases where this data i s available, it i s not integrated into a results-oriented framework that promotes policy actions and monitoring. xiii. The bottlenecks to regular disaggregated monitoring o f health and education MDGs and integration into policy are: (i) historical orientation, since there was limited use o f disaggregated data under the pre-transition social and political system, (ii) lack o f incentives for line ministries to focus on disaggregated analysis and monitoring, since these policy objectives are not promoted through macro policy instruments such as the PRSP and the medium-term budget framework, (iii) limited technical capacity in line ministries and other institutions (such as the National Statistical Office that i s responsible for poverty monitoring) and (iv) gaps inthe quality and quantity o f available data. For example, there are reliability and validity problems with measuring many o f the health outcomes and disincentives for providers and patients to report communicable diseases. Monitoring o f learning outcomes, especially among disadvantaged groups, i s still at the pilot stage. While there are good examples o f monitoring and evaluation initiatives inthe health and education sector, these are largely donor financed and has not been institutionalizedwithin the relevant ministries. xiv. Identifying Interventions that are Effective for Raising MDG performance: Globally, there i s good evidence on interventions which have proven effective for raising MDG performance. For example, in the health sector, interventions such as Directly Observed Therapy (DOTS)for addressing tuberculosis, and promoting lifestyle changes to address cardiovascular diseases can target the health MDGs. Ninety percent o f health MDG interventions can be provided through public health and primary health care services. Inthe education sector, it i s recognized that the availability o f textbooks and trained teachers and adequate amounts o f instructional time on the core curriculum are essential inputs for effective learning. The report finds that these international lessons o f experience are often not reflected in policy priorities in the health and education sectors. Beyond the international lessons o f experience, the report also recognizes the country-specific nature o f many o f the constraints to fuller MDG attainment. It emphasizes the importance o f correctly diagnosing the country-specific constraints to better MDG performance and evaluating pilots to identify the most feasible and cost-effective approaches for improving MDGperformance. Inthe health sector, an important lesson o f experience i s that community- based health interventions are an important complement to conventional professional services for addressing problems such as micronutrient deficiencies, brucellosis and cardiovascular diseases. But these methods need to be refined to reach some target groups. In the Kyrgyz Republic and Tajikistan, v1 while aggregate levels of coverage o f safe motherhood programs in high, there are disparities in utilization between urban and rural areas and between regions. In Tajikistan, geographical access was identified as a key barrier to continued use o f antenatal care. Inthe education sector, measures to exempt low-income children from paying textbook rentals charges have been effective in improving school attendance, but the number o f children covered by these interventions needs to be linked better inrelation to the extent of need in each locality. xv. Targeting interventionsto groups which are not meetingMDGs: The report emphasizes the importance o f targeting interventions to groups which are not meeting the MDGs. The most important interventions which needs to be better targeted in order to achieve better MDG outcomes i s budget financing for health and education. Budget financing on health in the Kyrgyz Republic and Tajikistan i s only US$7 and U S 2 per capita. The Tajikistan health sector i s almost entirely sustained through out-of- pocket payments, generating severe access barriers for the population. The Kyrgyz Republic spends only 4 percent o f GDP on education - which i s much lower than most countries in ECA. Despite some progress inchanging allocationpatterns from hospital-based health care to primary health care and public health (which are more closely related to MDGoutcomes), the hospital sector inthe Kyrgyz Republic still receives 70 percent o f budget financing, while public health barely receives 5 percent. Inthe education sector, secondary vocational and higher education - which are less relevant than basic education to attainment of the education MDGs - consume a disproportionate share o f budget financing. Within the budget allocations for basic education, too little i s provided for educational materials which are critical for effective teaching and learning. Budget allocations and health and education needs o f regions and communities are not well matched and do not take into account regions' and communities' capacities to contribute. In the education sector, this leads to perennial gaps in learning achievement in the poorest schools. Budget execution for non-salary items i s nominal. xvi. The key bottlenecks to the problems with targeting budget financing are: weak linkages between public resources and strategic expenditure priorities of the Government, wide divergence between initially approved budgets and the allocations for non-protected items with large cuts in the provision of complementary inputs in social and economic development, poor internal and external controls and weaknesses inmonitoring and evaluating the impact of public spending. xvii. Inaddition to financing, another bottleneck is inadequate sectoral attention to evaluating and scaling-up service delivery innovations that can help reach disadvantaged groups. For example, in both the health and education sectors, this means ensuring the recruitment and retention o f skilled health and education personnel in rural areas, ensuring regular supply o f textbooks and medicines, scaling-up interventions that promote community-based and population-based health services to address problems such as micronutrient deficiencies, brucellosis and cardiovascular diseases. A key constraint to more targeting i s limited information on the groups which are not reaching MDG health and education goals, on the constraints to MDG achievement, and on cost-effectiveness o f pilot interventions to raise MDG performance. xviii. Strengthening Incentives to service providers and users to improve MDG outcomes: The 2004 World Development Report emphasizes improving incentives for reaching the poor with better services. The report finds that the Kyrgyz Republic has already made important progress on this front. In the health sector, providers are paid according to outputs and have legally enforceable contracts for achieving outputs and management flexibility in determining budget allocation and inputs. Primary care i s completely free for patients enrolled with a primary care doctor. These incentives have improved the utilization o f primary and outpatient services. For example, between 2001 and 2003, there has been a drop in the use o f outpatient services among the 4" highest income quintile and the richest. In contrast, utilization rates among the poorest quintile and third quintile have remained stable, while they have increased slightly for the second poorest quintile. In the education sector, the Ministry o f Education vii (MOE) i s piloting performance incentives for teachers. It will be important to evaluate these efforts, and to, scale these up if they are found to improve teaching performance and learning achievement. Further performance incentives may also help achieve the health MDGs. Examples include performancebonuses for primary care doctors for reaching specific, monitorable outcomes such as the continuity o f antenatal care inrural areas, improving the coverage o f DOTs and numbers o f patients completing DOTs, as well as adoption o f healthy lifestyles by communities. WHAT D O THE FINDINGS FROMTHIS REPORT MEAN THE KYRGYZ FOR REPUBLIC? xix. The first implication of the findings of this report is the need for adopting a results-based framework for monitoring the MDGs: In the context o f the MDG Progress Reports, a fledgling monitoring framework i s already in place. There i s a need to build upon and institutionalize this framework to involve line ministriesand other relevant groups. Existing data could be usedto buildupon this framework - linking to the extent possible, outcomes to intermediate indicatorsiprocesses, outputs and inputs. The framework could also explicitly acknowledge gaps in data and identify actions for addressing these gaps. Inthe absence o f such as framework, it will not be possible to make the linkages between the National Poverty Reduction Strategy (NPRS) and the MDGs, to integrate the MDGs into the medium-term budget framework (MTBF), to hold line ministries and local governments responsible for results and streamline and harmonize donors support for the health and education MDGs. A results-based framework would also help inmonitoring progress on intermediate results, inbuilding accountability and inkeepingkey stakeholders, including civil society, informed about the progress towards the MDGs. A results-based monitoring framework for the MDGs needs to include disaggregated analysis of outcomes. Butit also needsto go beyond outcomes to identify some o fthe underlying factors (determinants) that are responsible for disparities in outcomes. While the health and education sectors need to lead the formulation o f such as framework, other actors also need to be involved. These include donors, non- governmental organizations, service providers and uses, and the National Statistical Office (which i s responsible for poverty monitoring). The overarching impetus for such a framework needs to be provided by the Comprehensive Development Framework and National Poverty Reduction Secretariats. The results on monitorable actions under the results framework could be integrated into the MDG Progress Report which i s already producedregularly by the Government in collaborationwith the United Nations. Giventhe importanceof a results-basedmonitoringframework for tracking the MDGs,it would be importantto have a comprehensiveframework inplacewithinthe next 12-18 months. The key steps inthe process are: 0 Creating a monitoring framework that includes currently available data on MDG health and education outcomes and intermediate indicators disaggregated by oblast. 0 Integrating disaggregated data from the framework into the National Poverty Reduction Strategy (NPRS) progress report. 0 Identifying targets for improving intermediate variables (such as coverage o f key programs) and linkingthe achievement o fthese goals to monitorable policy andprogrammatic action. xx. The second area that needs attention is strengthening the medium-term expenditure framework (MTEF) process and linking budget execution more closely with the stated policy priorities of the Government (poverty reduction and MDGs). A medium-term expenditure framework, if properly implemented, provides a very powerful tool for operationalizing the health and education MDG. A well-implemented MTEF can improve the effectiveness and governance o f budget management. It can help the Kyrgyz Republic link outputs and outcomes to ensure consistency with sectoral expenditure frameworks and can generate incentives for line ministries to improve allocation and technical efficiency. The Kyrgyz Republic has implemented the first level o f the MTEF (i.e., the medium-term fiscal framework) and i s currently strengthening the next level, i.e., the medium-term ... Vlll budget framework (MTBF). The next stage i s to move to an MTEF inwhich line ministries' expenditure programs are defined on the basis o f outputs and outcomes and linked to a results agreement. This will take time, but it is important to advance the process by building the capacity o f line ministries and the Ministry of Finance to move in this direction. The implementation of the MTEF to date in the Kyrgyz Republic has involved a fragmented process without strong ownership and leadership. The limited technical capacity o f line ministries to formulate defensible and well-documented sector expenditure programs has also been a constraint. Nonetheless, there i s strong donor support for the MTEF process and the current Government i s also committed to continuing efforts on the MTEF. Donor support for the poverty reduction strategy process and the development o f a health Sector Wide Approach (SWAP) also provide opportunities for strengthening the linkages between the MDGs and expenditure allocations inthe context o f the MTEF. The key next steps to operationalize the MTBF in the context o f the MDG agenda include: 0 Developing the first Sector Expenditure Plans (SEP) for the health and education sectors that includes programs for achieving the MDGs in the context o f sector strateges. These should be identified as priority programs with full financing; 0 Ministry o f Finance makes changes inthe current budget classification system allowing tracking of budget execution on priority programs; 0 Indicators for tracking progress on sector expenditure plans include MDG intermediate indicators (e.g. coverage and utilization o fpriority MDGprograms). xxi. The third implication is the need for effective targeted sectoral interventions: A stronger focus on disaggregated data analysis o fMDGoutcomes and underlying factors needs to be complemented by the identification of targeted interventions that will alleviate the constraints to fuller attainment o f the MDGs. Many potential interventions have already been identified. The challenge i s to carefully evaluate pilot interventions including pre-conditions for scaling-up, the associated costs and implementation requirements, and integration into sector programs and sector expenditure plans. The same i s true for strengthening provider and user incentives to achieve the MDGs. This i s not a small task and will require Ministries o f Health and Education to prioritize the MDGs in their day-to-day work and create institutional structures that facilitate coordination between the various vertical programs (as inthe case o f health). Ownership of the MDG agenda i s instrumental for this to happen. Inboth sectors, elements o f ownership are already present. For example, MDGs feature prominently inthe Government's new health reforms strategy and the education sector i s already implementing key interventions to achieve fuller attainment o f the education MDGs.Donor commitment to supportingthe health and education MDGs and ownership o f the MDG agenda beyond line ministries i s critical for generating incentives for further sectoral focus on the MDGs. From the sectoral perspective, it i s also important to pay attention to the management flexibility needed for maximal effectiveness o f provider inputs. In the education sector, conflicting responsibilities for school management and deficiencies o f the school financing formula create a situation in which local governments, which are intended to manage their schools, have neither the incentives nor the means to reconfigure schools or deploy teachers more efficiently. Some priority actions inthis area include: 0 Strengthening rural service delivery by training and equipping feldsher (community health workers) and rehabilitating Feldsher Ambulatory Points (FAPs). 0 Introducing performance incentives for feldshers and family doctors to achieve better coverage and quality o f health MDGs. 0 Designing and implementing on a pilot basis a performance bonus system for primary care doctors and teachers. i x REPORT STRUCTURE xxii. The report consists o f a Main Report and Annexes. The Main Report includes two Chapters. Chapter 1 discusses the relevance o f the MDGs for Central Asian countries and also highlights data problems in trachng MDG progress in Central Asia. It compares the MDGs in Central Asia with other countries o f similar income level and evaluates the implications for policy action. Chapter 2 discusses a fi-amework for operationalizing the health and education MDGs in Central Asia drawing upon the key findings o fthe Kyrgyz Republic health and education case studies. The Annex Report consists o f the two case studies. X 1:THE STATUSAND RELEVANCEOFTHE MDGsINCENTRALASIA INTRODUCTION 1. World leaders from 189nations agreed and signed the UNMillennium Declaration, in September 2000, binding them to a global project to decisively reduce extreme poverty in all its key dimensions. The eight MDGs that derive from this Declaration provide an agenda for global action. The first seven goals are directed at reducingpoverty inall its forms: hunger, a lack o f income, education and health care, gender inequality, and environmental degradation. While each goal i s important, collectively they form a comprehensive and mutually reinforcing approach to alleviating poverty. 2. In the past decade, the social, political and economic conditions in Europe and Central Asia (ECA) have been continuously changing. The massive social infrastructure inheritedby these countries in the early 1990s made them appear better than countries in other regions with similar incomes. The economic downturns of the 1990s however, reversed many o f the social indicators and some are still catching up with the 1990 levels. There has recently been a robust turn around inECA with all countries beginning to experience positive economic growth. While growth i s starting to lead to poverty reduction inmost countries, the pace of change is slow and it isyet to be seen ifthis will translate into comparable progress towards meeting the MDG goals, which will also require significant resource reallocations and major public sector and institutional reforms. 3. Even within the E C A region, as we move from west to east, there i s a significant change in the likelihood o f countries meeting the MDG goals. Kazakhstan, the Kyrgyz Republic, Tajikistan and Uzbekistan are countries of particular concern. Poverty in the CIS countries rose faster than any other region duringthe 1990s. GDP fell, and employment and wages plummeted. However, the resurgence o f growth inthe recent years and the better financing and targeting o f social protection has reduced poverty inmost countries. 1 Table 1: ProspectsofECA CountriesAchievingthe MDGs WQS m 7 msslnal HIUAIRS t wotar 11w1t&y T3 wtrnteirea Access KW Source: World Bank, 2005 Trends 4. The recent development trajectory o f Central Asian countries can be tracedback to their pre-1990 communist governments. On the one hand, their economies were primarily dependent on Russia for supply of raw materials, which made them vulnerable to external shocks. On the other hand, widespread availability o f education, health care and social benefits had created artificially high levels o f social indicators in the early 1990s. The impact o f the breakdown o f the Soviet Union, then followed by the Russian financial crisis was particularly acute in the Central Asian economies (Kazakhstan, the Kyrgyz Republic, Tajikistan and Uzbekistan) compared to the other ECA countries. It led to the accumulation of considerable debt and cuts in health and education spending by these countries. The rapid contraction o f state funding was reflectedinworsening healthindicators and stagnant education indicators. 2 EDUCATIONMDGS MDG: Achieve UniversalPrimaryEducation Indicators: N e t primary enrollment ratios and primary completionrates. Target: Ensure that by 2015, children everywhere, boys and girls alike, will be able to complete a fullcourse ofprimary schooling. 5. When discussingthe education MDG goal o f achieving universal primary education, there are a couple o f different indicators used to measure the progress o f a country, including net primary enrollment ratios (NER), primary completion rates (PCR), and net secondary enrollment ratios. At the current pace o f improvement, Kazakhstan, the Kyrgyz Republic, Tajikistan and Uzbekistan might be able to meet the education MDGtarget. Figure1: TrendsinPrimary CompletionRates, 1998-2003 1120 - ' O B O1 1040 - CI - c ;lWO I I : u 5-20 I 880 MO -Kazakhstan -KyrgyzRepublic Tajikistan -, -, Uzbekistan Source: Data from World Development Indicators 2005 6. In the case of ECNCIS it is important to note that due to their pre-transition communist governments, these states entered the 1990s with extremely high enrollment rates - numbers that were at the levels o f present day OECD countries. The net enrollment rate for U S at present i s 94, and thus inthis report, for any country with NER greater than 90, the likelihood o f achieving this MDG target i s very high. As Figure 1 illustrates, despite the volatility o f the past decade these state have impressive enrollment and completion rates. Nonetheless, in the Kyrgyz Republic both primary NERs and PCRs have dropped over time. And there i s a need to control this decline. Inthe case o f Uzbekistan, PCRs are at 103, but looking at the different sources o f NERs, it i s clear that even though there has been improvement, the pace i s hard to detect. In the past decade, the NERs and the PCRs have either deteriorated (Kyrgyz Republic) or recovered (Tajikistan) to pre-transition levels, but have not improved beyond that. And thus we believe that all the Central Asian countries might be able to meet the MDG education targets ifthey can control any further deterioration o f the education system and can continue to improve even if at a slow pace. Kazakhstan has improved its NER, but has barely maintained the PCR and looks likely to meet the MDG goal. 3 MDG: Promote GenderEquity andEmpowerWomen Indicators: Ratio o f girls to boys enrolled in primary school and percentage o f women in parliament. Target: Eliminate gender disparity in primary and secondary education preferably by 2005 and inall levels o f education no later than 2015. 7. Measuring health, education and labor market differences in men and women illustrates that both have paid the price o f economic transition. Kazakhstan, the Kyrgyz Republic and Uzbekistan are likely to meet this MDG, but Tajikistan poses a problem. There has been a steady decline in the ratio of girls to boys inbasic education, with the current number stating that 88 girls are found for every 100boys in grades 5-9. Further analysis o f the region, however, shows that women are increasingly at a disadvantage. While women enjoy a considerable presence in schools and in the labor market, their minority position i s apparent when barriers to their achievement are considered in the most prestigious sectors o f education and economy. Another symptom o f the erosion o f status o f women in Central Asia i s their low representation in national parliaments: in Kazakhstan women's share o f seats declined from 13 percent in 1995 to 10 percent in 1999. Inthe Kyrgyz Republic, the percentage o f women inmanagerial positions i s only 29 percent and the salary ratio o f women to men has declined from 1996 to 2000 (from 73 percent to 68 percent). At the same time, in the Kyrgyz Republic, the poverty levels among female- headed households inlower as compared to male-headed households. Therefore, even though the gender MDGis currently not a problem, it should be followed for the next couple of years because the countries have not attained complete gender equality yet. HEALTHMDGS 8. One of the problems with tracking the health MDGs in E C A i s the quality o f the data. Among E C A countries, the problems o f data quality are the greatest inthe eight countries o f Central Asia and the Caucasus. Inthese countries, official infant and child mortality rates, as reportedto WHO, are considered to be substantial underestimates. Surveys in these countries have shown much higher rates. There are other problems indata quality such as inaccurate population data which affects the denominator for these estimates. MDG: CombatHIV/AIDS, Malaria andOther Diseases Indicators: HIVInfectionratesandincidence ofTuberculosis. Target: Halt and begin to reverse the spread o f HIV/AIDS by 2015; halt and begin to reverse the incidence o f malaria and other major diseases by 2015. 9. Inall the four countries the prevalence ofHIVrates and the IncidenceofTB numbers have risen drastically over the past decade. 10. As Figure2 shows, InKazakhstan, TB has spread the fastest rising from 57 to 145 infected cases (per 100,000 people) between 1990 and 2003. InTajikistan the number was close to 107 in 1990 and though the number fell for three consecutive years, it has picked up again and now the level i s much higher at 168. 4 Figure 2: Trends inIncidenceof TB, 1990-2003 . . 180 , Tr- of TR, 1990-2003 150 g 90 9 0 60 30 0 1990 1993 1996 1999 2002 Source: Data from World Development Indicators 2005 11. This goal o f halting HIV/AIDS and beginning to reverse the spread by 2015 was largely developed with Sub-Saharan Africa in mind. In the Central Asian countries the epidemic i s still in its early stages, and one o f the main modes of transmission i s through injecting drug users. The risk of spread o f disease is very high in these states, even though it gets undermined due to the low absolute numbers compared to the rest o f the world, but if not kept in check it could increase rapidly. Observing HIV/AIDS data from ECEMAIDS, plotted in Figure 3, we see that HIV/AIDS has risen the fastest in Uzbekistan going from zero newly-diagnosed HIV cases per million people in 1996 to 70.4 cases in2003. That is, the number o freportedcases has been doubling almost ever year. Figure 3: TrendsinPrevalenceof HIV (infectionratesper millionpeople), 1996-2003 Trends in Prevalence of HIV 80 (hi%*- 'IlieRpeepl@#%4333 -10 1996 1999 --Tajikistan 2002 ~ -Kazakhstan -Kyrgyz Republic tUzbekistan Source: from the European Center for the Epidemiological Monitoring o f AIDS. WHO & U N A I D S collaboration center on AIDS. HIV/AIDS Surveillance inEurope Endyear report 2003, N70. 5 12. As the table below shows, the worst affected country in the region i s Kazakhstan where the cumulative rate o f HIV infection in 2004 was 31.4 per 100,000 population. The rate of new HIV infections (new cases) indicates that Uzbekistan experiencing the most rapid growth o f the epidemic in the region. Unlike the other countries, in Tajikistan, HIV/AIDS does not seem to be a problem of any concern. There are currently only 6.7 infected cases per million, as reported by ECEMAIDS. Nonetheless, there i s no scope for complacency since the risk factors for spread o f the disease are present in Tajikistan. For example, the number o f registered drug users (RDU) in Tajikistan has more than doubled between 2001 and 2003 and the number o f RDUs using heroin increased from 63 to 75 percent duringthe same period(Godinho et a12005). 13. Since the likelihood o f reaching this MDG goal i s based on whether either the HIV/AIDS or tuberculosis targets are likely to be met, all the CIS states appear unlikely to achieve it. High migration rates, limited capacity o f governments and civil society to implement effective preventive responses, and low levels o f awareness o f HIV and STIs. Drug transit and growth o f local consumer markets i s also contributing to the problems. MDG: ReduceChildMortality Indicator: Under five mortality rate Target: Reduce the under-five mortality rate by two-thirds between 1990 and 2015 14. Analyzing trends in IMR and U 5 M R i s a tricky issue as it brings to fore the data issues o f the region. As figure 4 shows, there are significant differences in the reported USMRs across data sources. For example, in Kazakhstanthe U 5 M R reported can vary from 25.15 to 73 per 1,000 live births. It i s the same for Tajikistan where the 2002 Demographic and Health Survey (DHS) reported an IMR o f 87 per 1000 live births- a four-fold difference with national data. Figure4: USMR,2000 120 - rn WDI w TMD Country specific source Kazakhastan tQgyRe(xlMic Tajikistan m i s t a n Source: Data fromWorld DevelopmentIndicators 2005, TransMONEE database & Kazakhstan DHS 1999. 15. Interms of the trends inUSMRs, inTajikistan, the Kyrgyz Republic and Uzbekistan, it has been falling consistently but at a very slow pace. The following table shows the progress counties have made over the past 15 years inlowering there U5MRs. 6 Table 2: USMRPercentage of Improvement, 1990-2003 Kazakhstan -15.9 Kyrgyz Republic Tajikistan 20.2 Uzbekistan 12.7 49.3 Source: Data from World Development Indicators 2005 & TransMONEE database. 16. As the above numbers illustrate, in the case o f Kazakhstan WDIAJNICEF considers child mortality to have been on a rise moving from 63 deaths per 1,000 to 73 per 1000. TMD, on the other hand, reports that the U 5 M R has been falling consistently going from 34 to 21.7 deaths per 1,000, and since it hit a highof 36.5 in 1995 it has fallen by over forty percent in a decade. At this rate Kazakhstan will be able to meet the child mortality MDGgoal according to TMD data but it has almost no chance of meeting the goal according to WDI data. Thus, it i s important to look at definitional differences across data sources. Figure 5: Trends inUSMR,WDI & TMD, 1990-2003 90 - T 140 , -Kazakhstan _._. - 80 ------Kyrgyz Republic --.I----- - _I_* ------Tajikistan 60 -Uzbekistan _._. - 40 10 - 20 - I 0 ` 0 1990 1994 1998 2002 Source: Data from World Development Indicators 2005 & TransMONEE database. 17. WHO and Soviet countries have defined "live births" differently leading to varying rates. Most CIS countries, when recording official data in hospitals, still use the definitions established in the Former Soviet Union (FSU). Contrary to the World Health Organization (WHO), the Soviet practice was not to count as live births the premature and low-birth-babies who die within seven days. That can lead to under estimations o f the true infant mortality rate by some 20 percent. Yet, observing the current trend as seen in figure 5, it is clear that none of states except may be Uzbekistan, will be able to meet the child mortality MDGtarget o f a two-thirds reduction o f the 1990 baseline child mortality rate. Tajikistan TMD data i s only available from 1990-1994. 7 MDG: ImproveMaternalHealth Indicator: Maternal Mortality Rate. Target: Reduce the mortality ratio by three-quarters between 1990 and 2015. 18. In the case o f maternal mortality rates (MMR) again, the numbers are significantly different across data sources. However, the trends remain the same. Current data indicate that none o f the countries will be able to meet the MDGtarget o f a 75% reduction inmaternal mortality rates. Figure6: Trend inMMR,1980-2002 70 / Kazakhstan Kyrgyz Republic Tajikistan Urbekistan Source: Data from the TransMONEE database. 19. TMD reports that MMR in Kazakhstan declined from 20.3 (death/100,000 live births) to 8.3 while WHO reports a fall of over 43% starting at 54.8. InTajikistan we have a problembecause MMR has risen since the 1990 benchmark according to TMD data. At this pace it definitely will not meet the maternal health MDG goal and due to fluctuations in number overtime one cannot pick a trend either. UNDP national MDG progress reports say that the Tajikistan government recognizes this problem and has started implementing programs on reproductive health and family planning, and malnutrition issues amongst women to improve the situation. MDG+: ADDRESSING ADULT MORTALITYTARGETING BY NON-COMMUNICABLE DISEASES AND TRAUMA 20. In comparison to other regions (e.g. Sub-Saharan Africa and South Asia), infant and child mortality i s less o f a problem in ECA. In contrast, adult mortality i s a concern, especially from non- communicable diseases (NCD) and trauma. If the health MDGs are widely interpreted as improving health outcomes inthe population usinglife expectancy as a measure, then it i s critical for ECA countries, including Central Asia, to address adult mortality. A World Bank study has calculated the impact o f addressing NCDs and trauma in Central (World Bank, 2004). The results shows that in comparison to other sub-regions within ECA, reducing maternal and child mortality in Central Asia i s still likely to reap considerable health benefits especially in Tajikistan (Table 3). Nonetheless, simultaneously addressing cardiovascular diseases and other external causes o f death i s likely to have an even greater impact on life expectancy, exceeding the impacts under reducing child mortality. The impact i s particularly large for Kazakhstan, and exceeds the E C A average. This indicates that to achieve better health outcomes - measured in terms o f life expectancy -Central Asian countries will have to focus on the dual burden of child and adult mortality. While the core MDGs do not specifically focus on NCDs and trauma, Target 8 8 (have halted and begun to reverse the incidence o f malaria and other diseases) does provide an opportunity for Central Asian countries to define an agenda for action on the NCDs. Table 3: OverallImpact on Life ExpectancyAccordingto Different Scenarios (CentralAsia) Source: World Bank, 2004 MDG: EradicateExtremePovertyandHunger Indicator: Proportion o f populationbelow $2.15 per day. Target: Halve between 1990 and 2015, the proportion o f people whose income i s less than $1 a day and halve between 1990 and 2015, the proportion o fpeople who suffer from hunger. 21. While the MDG indicator and target include $1 a day, a higher poverty line o f $2.15 a day i s considered more appropriate in Central Asia given the extra expenditure on heat, winter clothing, and food. The collapse o f the Soviet planned economy, introduction o f market prices on basic goods and political instability (e.g. the civil war inTajikistan) contributedto a growth inpoverty throughout the sub- region in the first half of the 1990s. The return o f growth since 2000 has made the CIS economies some o f the fastest growing in the world. On one hand, though the share o f poor declined in the Kyrgyz Republic by one-eight between 2000 and 2003, prospects for further poverty reduction in the region are less promising. Even though rising unemployment rates were expected given the end o f the socialist system, the creation of high-value addedjobs has been slow, and unable to replace the jobs lost. A large share o f employment, in the CIS i s in low productivity occupations such as subsistence agriculture and a failure to expand the scope o f more productive employment will fundamentally limit the poverty reduction impact o f growth. 22. According to ECAPOV database that uses 1996 PPP-corrected U S D and an aggregate consumption basket consisting o f the same components for all four Central Asian countries, poverty has decreased in all o f these countries with the sharpest decline inKazakhstan. The poverty levels identified inECAPOV study are different fromthe official poverty statistics because of different methodologyused in assessing individual welfare. The poverty measures are influenced by i)consumption basket chosen, (ii) poverty threshold, and (iii) conversion factor. Hence, when doing international comparison the full extent o f the dataset in one country (e.g. Kyrgyz Republic) cannot be exploited because the other countries may not collect as rich the dataset (e.g. Uzbelustan). 23. Tajikistan has the highest percent o f people living on less than $2.15 per day, although it has decreased significantly in the period from 1999 to 2003 from 69.9% to 45.5%. In the Kyrgyz Republic 9 Uzbekistan it has decreased from 7.3% to 5.4% duringthe same period. Kazakhstan- the only low-middle the poverty rate according to ECAPOV has gone down from 40.1% to 27.2% during 2000-2003 and in income country o f the four - has decreased most rapidly its poverty rate, from 14% in 2000 to 7.1% in 2003. Still, the poverty rate in Kazakhstan seems to be higher than in Uzbekistan. This outcome i s somewhat counter-intuitive as Kazakhstan has a much higher GDP per capita and general standard o f living than Uzbekistan. These figures differ significantly from the poverty estimates based on country- specific consumption baskets provided by the relevant national statistical agencies or World Bank calculations based on them. Thus, at $2.15 usingnational consumptionbasket the absolute poverty rate in Uzbekistan in 2000 was 71.7% and in Kazakhstan in 2001, it was estimated to be 4.9%. InTajikistan when using national consumption basket the poverty rate at $2.15 poverty line at 1996 PPP-corrected U S D the poverty rate decreased from 75.9% in 1999 to 55% in 2003 (a 20% decrease as opposed to 24% shown when usingECAPOV methodology). For the Kyrgyz Republic, according to nationalpoverty line, the absolute poverty rate fell from 52% in 2000 to 40% in 2003. Most importantly for this Study, all o f the available data for all Central Asian countries, except Uzbekistan, show a general decline in absolute poverty during the past four years. However, it i s questionable whether the rate o f such decline and its steadiness will allow the four countries to achieve the poverty MDGtarget by 2015. MDG: EnsureEnvironmental Sustainability Indicator: Proportiono f people with access to improved water source. Target: Integrate the principles o f sustainable development into country policies and programs and reverse the loss o f environmental resources; halve the proportion o f people without sustainable access to safe drinking water by 2015; achieve a significant improvement in lives o f at least 100 million slum dwellers by 2020. 24. Although access to most utilities remains high, the quality o f electricity, water, gas, heat and other infrastructure and energy services has deteriorated due to under-funding and poor maintenance. While data availability i s a challenge, available survey data suggests that neither water supply nor electricity i s fully reliable. Water represents a key and scarce resource in Central Asia and numerous concerns remain in areas o f both the quantity and quality o f available water. Much progress has been achieved in making water available to over 90 percent o f urban homes in all the states, and to at least three-quarters o f rural households (with the exception o f Tajikistan, where around a quarter o f families only use improved water sources). However, management o f water resources i s made difficult by their very uneven distribution both on the regional and national scales. Current data trends show that Kazakhstan and Uzbekistan might be able to meet the sustainable environment MDG. Tajikistan looks unlikelyto meet it, while inthe case of the Kyrgyz Republic it is hard to detect the state o f development due to a lack o f data. Once, the issue o f water access has been dealt with it i s then important to concentrate on the quality o f water being supplied because that seems to be an upcoming problem. PROSPECTS Following is a summary of the previous discussion o f each o f the MDGs. 10 Chart 1:MDGSummary 1 MDGI MDG2 MDG3 MDG4 MDG5 MDG6 MDG? poverty school equality child maternal HlVlAlDS 8 water MDG target likely to be achieved Too hard to tell whether MDG target will be met or not Source: World Bank, 2005 25. As Chart 1clearly indicates, while the healthMDGspresent the greatest challenge for the Central Asian countries, the education, environment and poverty indicators are also o f much concern. Inthe case o f universal primary education, although primary enrollment and completion rates have improved in the past decade, the countries will needto accelerate a little over the trend observed for the MDGtargets to be met. Gender equity in education does not pose a problem in any o f the countries except Tajikistan. For Kazakhstan, Kyrgyz Republic and Uzbekistan, the gender MDG targets are close to or have already been met. Child and maternalmortality are moving inthe right direction inmost countries. However, with the slow progress in achieving reduction in mortality, as well as concerns on the delivery and quality o f critical medical services, imply that the Central Asian countries, except Uzbekistan, look unlikely to meet the child and maternal mortality MDGs. 26. The HIV/AIDS and TB incidence also pose a significant issue for the region. While the absolute numbers may be lower than other regions, the HIV epidemic has been growing extremely fast. The epidemic i s still inits early stages inCentral Asia, but with an increase in "risky" behaviors, such as drug use and protected sex, HIV/AIDS i s now spreading rapidly from the high-risk groups, to the "bridge" populations and into the general population. Tuberculosis i s also a large and growing problem in these countries. The disease i s fueled by ineffective approaches to diagnosis and treatment, poor coverage o f effective treatment protocols and weak, deteriorating health systems. Thus, none o f the Central Asian states are likely to meet the HIV/AIDS & TB incidence MDG. 27. As for sustainable environment MDG, the sub-region definitely has taken significant strides in increasing access to improved water sources. There has however, been little investment in water infrastructure so that water quality i s becoming a serious issue with drinkingwater frequently not meeting basic biological and chemical standards. Kazakhstan and Uzbekistan might be able to reach the MDG target, while Tajikistan probably will not be able to do and due to lack o f data in Kyrgyz Republic we cannot say anything. Regional Comparisons 28. The discussion in the previous sections has focused on the trends in MDG goals within the Central Asian countries. It i s however, also important to compare them to countries from around the world. The Kyrgyz Republic, Tajikistan and Uzbekistan are lower income countries while Kazakhstan is a low-middle income country. Based on population and income levels the following part o f the report compares the progress inMDGsto countries from other regions. 11 29. Inthe period since 1998 many countries have maintainedor improved their highlevels of school enrollment. The CIS countries entered the 1990s with a widespread coverage o f education services that enabled them to achieve almost universal coverage incompulsory education. Even though, some of these achievements eroded during the 1990s, primary completion rates are typically higher than comparator countries. As figure 7 and 8 illustrate, in comparing Central Asian states with countries o f different regions with similar incomes and populations levels, we find that the Central Asian nations fare much better inabsolute terms incase of the education and gender equality MDGs. Figure 7: Primary CompletionRate, 2002 "1 Figure 8: Ratio of Girlsto BoysinPrimary and SecondaryEducation (YO),2002 1 81 Source: World Development Indicators, 2005 30. When we compare the health MDGs however, the story changes considerably. Even though maternal mortality rates (figure 9) across Central Asian countries are lower than the comparator countries, the rise of diseases such as Tuberculosis overtime has been more drastic inthe region. 12 Figure 9: Maternal Mortality Ratio (modeled estimate, per 100,000 live births) Maternal mortality ratio (modeled estimate, per 100,000 live births) 1600 - 1200 800 400 0 31. Given our set o f comparator countries, TB has risen the fastest in the Central Asia region. Uzbekistan, Kazakhstan, Tajikistan and the Kyrgyz Republic all have positive growth rates o f incidence of TB. Table 4: Increase inIncidence of TB, 1990-2003 Country I %increasein incidenceof TB, 1990-2003 ~ Uzbekistan 69 Kenya 445 Yemen, Rep. -33 Kazakhstan 154 Ecuador -32 Belarus 39 Syrian Arab Republic -38 Guatemala -26 Country % increaseinincidenceof TB, 1990-2003 Tajikistan 57 Rwanda 188 Eritrea 22 KyrgyzRepublic 138 Togo -1 Lao PDR -11 Source: Data from World Development Indicators 2005. 32. Figure 10 highlights high and top priority countries in relation to the poverty MDG, i.e. those where levels o f poverty are high and from 1990 to (for most data) 2000 progress was slow or non- existent. As i s evident from this figure, most o f the low-income and low-middle income countries from the East Asia and LAC regions have controlled their poverty numbers, and the Central Asian states need to work towards this goal. Even though, Sub-Saharan Africa needs to do the most work, all four of the Central Asian countries are also marked on the map. While Tajikistan i s a top priority country due to extreme poverty numbers, Kazakhstani s a highpriority country because even though the absolute level o f poverty i s not extreme, it i s unlikely to meet the poverty MDG goal. Given the additional data provided by ECAPOV I1we can now say that both Kyrgyz Republic and Uzbekistan are also highhop priority countries . Figure10: PovertyMDGTop PriorityandHighPriority Countries 14 2: OPERATIONALIZINGTHEHEALTHAND EDUCATIONMDGSIN CENTRAL ASIA: A FRAMEWORKFORPOLICYAND PROGRAMMATICACTION 33. As Chapter 1has shown, the health and education MDGs and MDG+ are relevant for Central Asian countries. Accelerating progress on health and education MDGs inCentral Asia will be critical for MDGprogressinECA. The objective o fthis Chapter is to present a generic framework for analyzing and addressing the health and education MDGs in Central Asia. The framework is applied to the health and education sectors in the Kyrgyz Republic. The Kyrgyz Republic was chosen to illustrate the framework because prior analytical and project work provide a relatively rich knowledge base and experience and because there i s significant government ownership o f the key principles involved in the approach. The case studies do not aim to present new analysis o f the technical issues inthe health and education sectors inthe Kyrgyz Republic. Rather, they aim to bringtogether relevant existing sector knowledge and lessons o f experience in order to derive the options for moving ahead more aggressively on MDG attainment in the health and education sectors. The Chapter begins by presenting the framework highlighting the underlying conceptual bases for the framework. Next, the key findings o f the Kyrgyz health and education case studies usingthis framework are presented. 34. The report identifies a generic framework for analysis and policy action for fuller attainment o f the health and education MDGs in Central Asia (see Box 1 for analytical underpinnings o f the framework). The report recognizes that economic growth, alone, i s not likely to be sufficient for attainment o f the health and education MDGs. Attaining these goals will also require scaling-up service delivery. The framework focuses on the service delivery aspects o f achieving the MDGs and emphasizes: (a) monitoring outcomes to identifygroups which are not achieving MDGs, (b) identifyinginterventions which are effectiveinraising MDGperformance, (c) targeting interventions to groups which are not meetingMDGs, and (d) strengthening incentivesfor providers and users to raise MDG outcomes. 35. Where budget financing i s a binding constraint - as i s usually the case - it also involves selectivity inthe use o f budget resources to focus on improving outcomes for groups that are not meeting the MDGs. 36. No such generic framework for addressing the health and education MDGs currently exists in Central Asia. Prior work on the MDGs in Central Asia has focused on individual elements o f the framework (e.g. mentioning geographic disparities or the need for financing) and on monitoring the literal MDGs. Other work has focused on estimating the costs o f attaining the health and education MDGs. While the role o f sound policies i s implicitly acknowledged in all the work there has been less attention to clearly spelling out the direction o f these policy actions from a sectoral perspective. The proposed framework brings together key operational requirements for attaining the health and education MDGs, and provides a direction for enhanced financing for the MDGs in Central Asia. 15 Box 1:ConceptualUnderpinningsof Framework for Operationalizingthe Health andEducation MDGsinCentralAsia rhere i s a substantial literature on the linkages between investing in health and education services and outcomes. Findings from key studies are summarized below. Inessence the studies point to the following key elements: targeting o f interventions and groups and strengthening service delivery through incentives and accountability. Based on a review o f recent empirical and theoretical literature, Filmer, Hammer and Pritchett (1998) (Health Policy inPoor Countries: Weak L i n k s in the Chain), highlight the disappointing experience with implementation o f primary health care. They emphasize the evidence o f two weak links between government spending o n health and improvements inhealth status. First, the capability of governments to provide effective services varies widely - so health spending, even on the "right" services, may lead to little actualprovision of services. Second, the net impact of government provision of health services depends on the severity of market failures. Evidence suggests that these are the least severe for relatively inexpensive curative services, which often absorb the bulk o f primary health care budgets. Government policy in health care more usefully focus on mitigating market failures intraditional public health activities, and inmore developed settings, failures in the market for riskmitigation. Inanother paper, Filmer (2003) (Determinants of Health and Education 0utcomes)argues that the supply o f services that affect health and education outcomes starts with global technological knowledge and goes all the way to whether teachers report to work and communities maintain water pumps. Theperformance of public expenditure in producing outcomes varies considerable across countries. There are large differences in achievements at the same levels of expenditures and similar achievements with large differences in public expenditure. Spending more through thepublic sector is not always linked to better outcomes. This is not to say that spending cannot be useful - the way resources are used i s critical. In"Rising to the Challenge: the Millennium Development Goals for Health". Wagstaff et a1argue that "Extra Government Spending is Necessary But Not Sufficient - Sector Strengthening is Also Required, and Spending Needs to be Targeted." The case against extra government health spending i s built o n empirical evidence that shows that additional government health spending has no perceptible impact o n infant and under-five mortality, once other influences are held constant. But this analysis has limitations. It refers only to child mortality, not health outcomes. Child mortality differs from other outcomes (say maternal mortality and newborn health) that are highly dependent on the quality o f health services. It i s unlikely that maternal mortality i s inelastic to government health spending. The results are also general. They indicate what happens to child mortality as a whole, in an average country. As a result, they hide significant spending effects among specific subpopulations and specific types o f countries. Finally, the results indicate what would happen to child mortality if additional government spending were to take theform of proportional scaling-up of all government health programs. They do not show what would happen if extra spendingfocused on specific subpopulations or specific programs. In "Public SDending and Outcomes: Does Governance Matter" Rajkumar and Swaroop (2002) examine the role o f governance - measured by level o f corruption and quality o f bureaucracy - and ask how it affects the relationship between public spending and outcomes. They examine whether the efficacy o f public spending can be explained by the quality o f governance. The authorsfind that public spending on health lowers child and infant mortality rates in countries with good governance. The results also indicate that as countries improve their governance, public spending on primary education becomes effective. The results o f this study are particularly relevant for countries, where public spending on education and health is relatively low, and the state o f governance i s poor. The 2004 WDR: Making Services Work for Poor People, focuses o n accountability as a key determinant o f effective service delivery. The WDR2004 identifies three types o f accountability relationships. These relationships share inter-linkages. The three comers o f the accountability triangle are represented by citizens (or clients), politicians (policymakers) and service providers. The inter-relationships represented by the sides o f the triangle represent the roles o f "voice" (between citizens and politicians), client power (between citizens and providers) and "compact" (between politicians and service providers). The WDR2004 framework identifies "voice" and "compact" as embodying the long routes to accountability and "client power" as the short route to accountability, and recognizes that the importance o f these accountability relationships will vary with sector and country-specific political characteri&ics. 16 APPLICATION OF THE FRAMEWORKTHE KYRGYZ TO REPUBLIC HEALTH EDUCATION AND SECTORS: KEY FINDINGS 37. The Kyrgyz Republic, similar to other countries o f Central Asia, has committed to achieving the health and education MDGs. The MDGs are consistent with the policy objectives o f the Comprehensive Development Framework (CDF) and the National Poverty Reduction Strategy (NPRS). In the health sector, the MDG targets adopted by the Kyrgyz Republic are consistent with the international targets, except under communicable diseases, brucellosis i s also included. The education MDG for the Kyrgyz Republic i s interpretedto include universal completion o f an eleven-year cycle o f primary and secondary education with acceptable learning outcomes. Because school attendance i s a requirement for effective learning, it also includes hllschool attendance as a subsidiary goal3 The Key Findings from the Kyrgyz Republic Health and Education Case Studies are: 1. MonitoringOutcomes to Identify Groupswhich are not Achievingthe MDGs 38. The education and health case studies reveal that aggregate health and education MDG data inthe Kyrgyz Republic mask disparities in outcomes by urban and rural areas, oblasts and the poor versus the non-poor. Figures 11and 12 demonstrate the relationshipbetween poverty rates and education and health outcomes. As expected, this shows a broad correlation. Urban-rural differentials are also large. For example, the single variable for which there are substantial enrollment deficits at the compulsory level i s residential location. Although children in rural areas tend to start school earlier than children in urban areas, they also tend to leave school at an earlier age: there i s a sharp fall-off inrural enrollment at age 17 whereas inurban areas enrollments taper offmuch more gradually. Figure 11: There i s a Close Correlation betweenPoverty Rates and Education Outcomes 100.0 90.0 +YO of Grade 8 80.0 Students with Lowest 70.0 Score in Natural 60.0 Sciences 50.0 -%ofGrade8 40.0 Students with Lowest 30.0 Score in Math 20.0 10.0 0.0 2003 Povsrty Incidence Headcount Index ( O h ) , Source:Center for Public Opinion Studies and Forecast,Monitoring of Learning Achievement (Grade S), Bishkek, 2002-2003, andWorld Bank,Kyrgyz Republic: Enhancing Pro-Poor Growth, Report No. 24638-KG, August 23, 2002. 3 UnitedNations 2003. 17 Figure 12: Health Outcomes and Extreme PovertyRates inthe Kyrgyz Republicare Broadly Correlated e E x t r e m e poverty rate e I M R f r o m A R I p e r 10000live births ~ -8mX- uBrucellosisper 100,000 ~ , - Source: Health data from Ministry o f Health, Poverty data from National Statistical Office as reported inthe 2004 World Bank Poverty Assessment 39. Despite the importance o f disaggregated analysis, the report finds that this type o f analysis i s not presented in a systematic way in key policy documents o f the Government. For example, the Kyrgyz National Poverty Reduction Strategy (NPRS) only presents aggregate health and education indicators and aggregate trends. This presents only half the picture about the MDGs and linkages with the non-income dimensions of p ~ v e r t y . ~ 40. In addition to data on health outcomes, to operationalize the MDGs in the context of medium- term policies, it i s important to focus on the disaggregated monitoring o f intermediate variables that are good proxies for MDG outcomes. Most o f these indicators have already been defined and focus on monitoring coverage of key services such as immunizations and antenatal care, and the net enrollment rate for education. These proxy indicators are key for linkingthe MDGs to poverty reduction strategies, medium-term expenditure frameworks and sector programs. The case studies found that data on intermediate indicators i s fragmented and disaggregated data i s hard to come-by (see Annex Table 2). In situations where data i s available there i s incomplete integration o f this data into a results-based framework that can promote policy actions and monitoring o f the MDGs. The bottlenecks to systematic analysis o f disaggregated data and integration into a results-based framework include: (i) historical orientation, since there was limited use o f disaggregated data under the pre-transition social and political system, (ii) lack o f incentives for line ministries to focus on disaggregated analysis and monitoring, since these policy objectives are not promoted through macro policy instruments such as the NPRS and the medium-termbudget framework, (iii)limited technical capacity in line ministries and other institutions (such as the National Statistical Office that i s responsible for poverty monitoring) and (iv) gaps in the quality and quantity o f available data. For example, there are reliability and validity problems with Government o f Kyrgyztan. 2004. PRSP Progress Report. The Government o f the Kyrgyz Republic, NPRS Secretariat, Bishkek,The Kyrgyz Republic. Government o f Kyrgyztan. 2003. Poverty Reduction Strategy Paper. The Government o fthe Kyrgyz Republic, NPRS Secretariat, Bishkek, The Kyrgyz Republic 18 measuring many o f the health outcomes and disincentives for providers and patients to report communicable diseases. Monitoring o f learning outcomes, especially among disadvantaged groups, i s still at the pilot stage. While there are good examples o f monitoring and evaluation initiatives in the health and education sector, these are largely donor financed and has not been institutionalizedwithin the relevant ministries. 2. IdentifyingInterventionswhich are Effective inRaisingMDGPerformance 41, Globally, there i s good evidence on interventions which have proven effective for raising MDG performance. For example, inthe health sector, interventions such as Directly Observed Therapy (DOTS) for addressing tuberculosis, and promoting lifestyle changes to address cardiovascular diseases can target the health MDGs (Annex Table 2) Ninety percent o f health MDGinterventions can be provided through public health and primary health care services. In the education sector, it is recognized that the availability o f textbooks and trained teachers and adequate amounts o f instructional time on the core curriculum are essential inputs for effective learning. The report finds that these international lessons o f experience are often not reflected inpolicy priorities inthe health and education sectors. As Annex Table 2 shows, the coverage and quality o f health interventions for raising health MDG performance in the Kyrgyz Republic shows a mixed picture. Coverage o f programs such as the neonatal integrated package, safe motherhood and nutrition i s limited. These programs are largely donor-led initiatives generating concerns about institutional and fiscal sustainability. 42. Beyond the international lessons o f experience, the report also recognizes the country-specific nature o f many o f the constraints to fuller MDG attainment. It emphasizes the importance o f correctly diagnosing the country-specific constraints to better MDG performance and evaluating pilots to identify the most feasible and cost-effective approaches for improving MDGperformance. Inthe health sector, an important lesson o f experience i s that community-based health interventions are an important complement to conventional professional services for addressing problems such as micronutrient deficiencies, brucellosis and cardiovascular diseases. But these methods need to be refined to reach some target groups. In the education sector, measures to exempt low-income children fi-om paying textbook rentals charges have been effective in improving school attendance, but the number o f children covered by these interventionsneeded to be linkedbetter inrelation to the extent o f need ineach locality. 3. Targeting Interventions which are Effective inImprovingMDGPerformance 43. The report emphasizes the importance o f targeting interventions to groups which are not meeting the MDGs. There are two key elements to improving targeting: (i) ensuring that budget funds are allocated to the right interventions and the right groups, (ii) mitigating service delivery bottlenecks in reaching these groups with the effective interventions. The following section discusses the situation in the Kyrgyz Republic with regards to each o f these issues. 3a. Levelsand OptimalAllocation of Budget Financing for Health and Education 44. As the 2004 WDR argues, "the budget i s a critical link on the long route to accountability connecting citizens to providers through politicians and policymakers. The biggest payoffs to service delivery are likely to come for a few key actions o f governments, Le., spending wisely and predictably in line with priorities and coordinated across sectors5 World Bank, World DevelopmentReport:Making Services Work for PoorPeople. InternationalBank for ReconstructionandDevelopment.The World Bank, WashingtonDC. 19 45. One o f the key bottlenecks for achieving the health and education MDGs inthe Kyrgyz Republic i s the budget. In the last few years, the Kyrgyz Republic has achieved progress on improving aggregate fiscal disciple through fiscal consolidation.6 However, the goal o f using the budget for allocative efficiency and equity and strengthening operational impact has yet to be achieved. K e y weaknesses in PEM include: weak linkages between public resources and strategic expenditure priorities o f the Government, wide divergence between initially approved budgets and the allocations for non-protected items with large cuts inthe provision of complementary inputsinsocial and economic development, poor internal and external controls and weaknesses inmonitoring and evaluating the impact o f public spending. The present M&E systemi s very simple and i s limitedto monitoring inputs.' 46. Health Sector': The Kyrgyz Republic continues to spend very low amounts on health - in fact way below the average levels of public spending for the E C A region (4.4 percent o f GDP). This i s partly explained by macroeconomic constraints. As Table 5 shows, countries with similar or slightly higher Gross National Income (GNI) per capita spend the same amounts as a percentage o f GDP or even less. Public financing as a percentage o f total health expenditures i s the highest in the Kyrgyz Republic and Ghana. Under the 2006-08 MTBF, public financing for health (with PIP) i s estimated to increase from 2.8 percent in2005 to 3.5 percent in2008. Table 5: SelectedCountries: GNIPer CapitaandPublic Spendingon Health Country IGNI per capital Public spendingon health IPublic financing as YOof total I as YOof GDP HE Cambodia 290 2.1 17 Kyrgyz Republic 290 2.2 51 Ghana 270 2.3 41 380 0.8 25 230 1.2 5 47. What i s less well explained i s why health spending has been steadily declining as a percentage o f total government spending and how this problem can be addressed in the future (Figure 13). One o f the key factors in this decline i s the inability o f the Government to meet expenditure targets due to over commitment as well as difficulties inrevenue collection. This has lead to ad-hoc cuts inthe healthbudget both at the Republican and local government levels. Wages, pensions and a few other items are prioritized while all non-salary inputs are vulnerable to cuts. At the Republican level this has meant that budget execution for the health insurance program for vulnerable groups which i s defined in terms o f an allocation per beneficiary rather than inputs such as salaries has been vulnerable to greater cuts. For example, in 2003, while overall budget execution o f the Ministry o f Health portion o f the Republican budget was almost 90 percent, budget execution for the MHIF program was only 50 percent. In2004, in the first and second quarters, budget execution was only 29 percent and improved only when donors raised concerns. Local government allocations for health typically cover non-salary expenditures and 'World Generalgovernmentfiscal deficit has declinedfrom 6 percentofGDP in2001to 4 percent in2004. Bank.2005.KyrgyzRepublic:ProgrammaticPublicExpenditureReview: ConceptNote,The World Bank, WashingtonDC. The sectiononhealthsector financing andbudgeting drawsheavilyfromthe following documents: Chakraborty S andKutzinJ. 2004.KyrgyzRepublic:PublicExpenditureandInstitutionalReview (PEIR): Health Chapter, The World Bank, WashingtonDC. WorldBank. 2004: Kyrgyz Republic: PovertyAssessment:HealthChapter, The World Bank, WashingtonDC. Various publicationsby Kutiz JandJakab Met a1publishedby the WHO-Dfid HealthPolicyAnalysis Unit.Healthfinancingnotesproducedbythe World Bankteam. 20 although these allocations are planned and included in the health budget at the beginning o f the year, budget execution i s poor. Since the health sector has co-payments, it i s argued that they can cover their own expenditures on non-salary items. There i s little recognition o f the fact that co-payments are formalization o f a previously informal source, and the public portion o f the budget is calculated taking into account co-payments. Ifbudget allocations are less than expected, patients have to pay formal co- payments and well as other payments to cover the gap inpublic funds. Figure13: The HealthSector HasBeenGettinga Smaller and Smaller Share of the Government Budget Since 1996 (healthexpenditures as YOof totalpublic expenditures) I 15% 13.5% ~92.3% I 11.7% A 10.1% 9.9% ^^, 1995 1996 1997 1998 1999 2000 2001 2002 2003 +State Budget -c-Republican Budget Local Budget Source:WHO HPAUAnalysis basedonTreasuryandMHIF data 48. Education Sector: Enrollments in primary and secondary education increased by 18% and in higher education more than doubled between 1990 and 2000.' Real education expenditures fell by 69% over the same period, implying an implosion o f education resources with few historical parallels throughout the world. Although there has been some slight progress toward recovery since then and further increases planned, central budget resources per student remain well below halfo f what they were in 1990. The sharp decline in budget financing for education occurred in spite o f a roughly constant budget share for education - averaging about 20% o f total government expenditures, and reflected both the fall inGDP and the much smaller share o f the public sector under a market economy. This i s a lower share o f GDP spending on education than in other countries in the region with the exception o f Russia and Tajikistan, and considerably below the average for the OECD. Expressed in real terms, public expenditures on education have fallen even more sharply - to just one-third o f their 1990 level, reflecting the combined effect o f the smaller public sector and the sharp decline in GDP. Under the 2006-08 MTBF,public financing for education (with PIP) is estimated to increase from 4.4% o f GDP in 2005 to 6.0% o f GDP in2008. Michael Mertaugh,KyrgyzRepublic Public ExpenditureReview: BudgetingandExpenditures inthe Education Sector, February24,2003. 21 49. The impact o f reduced resources on health and education services in the Kyrgyz Republic . include: Since public financing for health covers the Program o f State Guarantees (PSG) and the MHIFprogram for vulnerable groups, ifthe allocated levels ofpublic financing are less than what i s required to provide the PSG, the gaps have to be met through other means, i.e., greater dependence on out-of-pocket payments andor cutting services through implicit rationing. Analysis o f household data shows that, in all income groups, . households spent a greater share on health care in2003 than in2000 (Jakab, 2005). Widespread deprivation o f basic educational materials needed for effective teaching and learning and for modernization o f education programs. The recent National Survey o f Primary Education Quality found, for example, that 80 % of primary schools lacked a complete supply o f textbooks for students, 70 % lacked teachers' guides, 20 % lacked . desks and chairs for students, 70 % needed repairs to school furniture, 23 % o f schools lacked water supply, and 39 % lacked telephones." L o w salaries for health and education personnel, which has eroded teachers' and health workers' morale and motivation, induced many teachers and health personnel to take . additionaljobs, and exacerbated problems o f corruption." A cessation o f new school construction which has led to excessively intensive use of existing school facilities. Only 10 percent o f primary schools operate on a single shift. . Fully 81 % of rural schools and 71 % o f rural schools operate on double shifts. Nine percent o frural schools and 19 % o furban schools operate on triple shifts.12 The development o f parental contributions as an important source o f financing for school maintenance, fuel, and other necessities inurban schools. Parental contributions to urban schools often exceed $100 per year per student. Parents also contribute to schools in rural areas, but widespread poverty means that income from this source i s very limited. . The reliance on parental contributions i s a major source o f inequity in quality o f education. The re-emergence o f arrears in teachers' salaries and arrears to health care providers. While the republican budget as o f M a y 2005 didnot have any arrears on teacher salaries, local budget arrears currently equal to 9.8 million Soms. MHIF as o f January 1, 2005 covered its arrears to providers but in 2004, total MHIF arrears to providers were 260 loMonitoring Learning Achievement: National Survey o f Primary Education Quality, Center for Public Opinion Studies and Forecast, Bishkek, 2001. l1Teacher salaries - which averaged 857 s o m per month in2001 - are low inboth absolute and relative terms. In absolute terms, they are only halfthe minimumconsumption level for individuals, not to mention households. As a result, teachers lack motivation, and are compelledto work at other jobs inorder to support themselves and their families. Inhigher education, low faculty salaries have contributed to a serious problemo f corruption, with students oftenpaying for admission and grades. This has ledto a serious deterioration inquality o f education. More significantly, it has ledto cynicism on the part o f employers and the public about the significance o f higher education diplomas, except for the few institutions which have been able to prevent or control corruption. Teacher salaries are also low ina relative sense. Average monthly earnings inthe education sector are just 857 som, versus the considerably higher average earnings inall other sectors except forestry and health. It i s notable, inparticular, that average teachers' salaries are less than 40 % o f average earnings inpublic administration. l2ibid 22 million soms (US$6 million at current exchange rates). Moreover, there are arrears by local and republicanbudgets for utilities. SUB-OPTIMAL ALLOCATIONACROSS GROUPS, SERVICE DELIVERYLEVELS INPUTSAND 50. In resource constrained environments such as the Kyrgyz Republic, ensuring that funds are allocatedto the right interventions, groups and inputsi s key. Accelerating action on health and education MDGs will require focusing on public health and primary health care services inhealth and primary and basic secondary care for education, especially among disadvantaged groups. Public financing o f these goods can be justified according to public and merit goods criteria. Ensuring that there i s financing available for non-salary inputs i s also key. Historically, health spending inthe Soviet Union was skewed towards hospital-based, specialized care. 51. Historically, in the health sector, public health and primary health care services, especially in rural areas, were very poorly funded. The input-based budget allocation process favored regions which had large hospitals and urban polyclinics, which typically meant that resource allocation favored tertiary health and urban areas over rural areas. Figure 14 describes the current allocation pattern for local, republican and MHIF financing. This shows that poor oblasts such as Naryn and Batken do not receive any republican funding and are largely financed from local government contributions supplemented by categorical grants. Republican financing i s largely concentrated in Bishkek city, which also has a prosperous city government that i s able to increase financing for health. Figure 14 also shows that MHIF financing has an equalizing effect on health financing since it i s risk adjusted for vulnerable groups (number o f children, women and pensioners in the population). One o f the key challenges of the health reforms has been changing the budget formulation pattern in the health sector to match the output-based payment system (i.e., a per capita allocation for each person covering the Program o f State Guarantees and the MHIF Program for Vulnerable Groups). Although the MHIF program for vulnerable groups has now been in operation for several years, per capita financing for the PSG has yet to be implemented. Local governments also continue to define allocations by inputs. Figure 14: Public Expenditureson Health are not Pro-poor (per capita health expenditures at 1995 prices, Kyrgyz Soms) 350.0 300.0 250 0 200.0 150.0 100 0 50.0 0 0 E]local republican MHIF 1 23 52. The current approach to education budget allocation treats all schools equally - as through they had equal needs, equal performance, and equal access to other resources. This approach essentially promotes a level playing field in terms o f publicly provided inputs. But since some communities supplement central budget resources with very significant contributions from other sources, this equal- input approach leads to average levels o f learning achievement that are distinctly unequal, as the Monitoring Learning Achievement surveys document. It also results in equal allocations to schools that are, by themselves, only marginally keeping schools alive. For schools that rely exclusively on central budget financing, these allocations lead to substandard learning conditions and substandard learning outcomes. A strategy that gives first priority to attainment o f the education MDGwould need to allocate education budget resources very differently, giving priority to schools with sub-standard learning achievement. This would mean introducing a deliberate slope to the playing field in terms o f inputs in order to ensure that the lowest-performing schools meet national standards for learning achievement. In the current situation o f severely constrained budgets, it would also mean reducing - in some cases, possibly even eliminating - central budget allocations for higher-performing schools with access to other sources o f financing. 53. Full attainment o f the education MDG would also require that central budgets for education finance a broader range o f interventions. If the Republican budget for education were to continue to finance only teacher salaries and school utilities, the increase inbudget allocations which the Government has proposed would serve largely to raise the level o f teacher salaries. Higher teacher salaries are certainly needed, since average official teacher salaries are currently less than half the average salaries in public administration, and below the $1.OO per day poverty threshold. But budget resources are also need to cover other essential inputsto the learningprocess. For schools and communities which are too poor to be able to finance significant education inputs from their own resources, central budget resources would need to cover all essential requirements for effective teaching and learning. This means central budget financing not only o f teachers' salaries and school utilities, but also o f textbooks, teachers' manuals, supplementary reading materials, and teaching and learning materials for classrooms. In addition to getting these classroom learning resources to all schools with sub-standard performance, full attainment o f the educational MDG would require that the central budget support whatever other interventions (such as extracurricular tutoring, student transport, or revision o f the school calendar) may be necessary to ensure that all students attend school and meet national learning standards. These actions would need to be tailored to the particular needs o f each school. 54. Improving regional equity in resource allocation requires adopting an output-based budget formulation process, whereby the cost o f providing a service at an adequate level i s calculated (e.g. a per student or per patient cost). This cost typically includes the cost o f all inputs(staff, medicines, textbooks, other material costs). Given that certain groups are harder to reachthan others, this per capita rate i s then adjusted. Geographic, poverty, age and sex adjusters could be used based on an understanding o f the determinants o f poor performance. 55. The development and application o f a per capita formula for health and education in the Kyrgyz Republic has been delayed for the following reasons: (i)low capacity within relevant government institutions (MOF and line ministries), (ii) lack o f strategic vision for public expenditure management and reforms consistent with the policy priorities o f the Government, (iii)slow progress on identifying alternative institutional arrangements that could support accountability in the flow o f funds under a per capita system. 56. Improving intra-sectoral resource allocation i s another mechanismfor effective targeting. One o f the objectives o f the health reforms inthe Kyrgyz Republic was to improve the allocative efficiency o f health spending by allocating away from specialized, hospital-based care to public health and primary health care services. In the last few years, as a result o f right sizing o f staff and infrastructure in the 24 hospitals sector, the allocations have begun to shift in favor o f primary and other outpatient care. For example, between 2000 and 2004, allocations for primary and outpatient care have increased from 11 percent to 18 percent. Nonetheless, the level o f public funds allocated to inpatient care i s still high at 70 percent (for e.g. EU15 countries allocate only 40 percent o fpublic funds to inpatient care) (OECD, 2004). Inthe education sector, there are opportunities for improving allocative efficiency by shifting financing from vocational secondary education and higher education to compulsory primary and secondary education. 57. Further improvements in intra-sectoral allocations in the health sector are linked to rightsizing in the capital city - which has the largest health infrastructure in the country. This has been difficult due to the following reasons: (i)the powerful lobbying by the medical community in Bishkek and the considerably strong "voice" o f this group as compared with primary care doctors, (ii) the reluctance o f the Bishkekcity government to implement measures to reduce duplication in services inthe city since this could imply closure o f city hospitals. The difficulties o f rightsizing inBishkekhighlight a general lesson inthe Central Asia countries as well as other ECA countries - it is politically very difficult to right size hospitals in capital cities and yet without this, large improvements in allocative efficiency are not possible. 58. The story o f intra-sectoral allocations in the education sector i s similar with a strong political lobby o f urban-based students and parents in favor large public subsidies for higher education. The "voice" o f this group i s considerably stronger as comparedto their rural counterparts. Figure 15: Shares of Public Health and Primary Health Care are Small in the Overall State Budget 0PublicHealth 1 ' ArnbulatoIycare I 0Hospitals ,I 1995 1996 1997 1998 1999 2000 2001 I - Source: WHO HPAU 59. Optimal allocations across inputs are also key to improved service delivery. Under an input- based budget system, non-salary inputs have received insufficient attention. In the health sector, which has moved to output-based payment systems, the budget at the facility level i s not allocated by inputs and providers have flexibility to allocate resources across various inputs. Allocations for salaries and utilities - typically the largest consumers of facility budgets have been slowly declining complemented by increasing allocations for medical supplies, food, drugs and other inputs. This allocation pattern i s much 25 more conducive to quality service delivery. However, in the education sector, where school budgets are still based on input-based allocations, textbooks and other key items are under funded. 3b. Evaluating and Scaling-up ServiceDelivery Innovationsto ReachDisadvantaged Groups 60. A one size fits all approach i s not likely to help intargeting groups that are performing worse on the health and education MDGs. These groups face specific demand and supply-side challenges in accessing and utilizing health and education services. In addition, some o f the health MDGs such as micronutrient deficiencies in women and children, hypertension in adults or brucellosis requires intersectoral action. In the education sector, the Ministry o f Education and Culture (MOEC) i s currently piloting an incentive program for rural teachers. Inthe health sector, one o f the key supply side barriers in providing quality health care in rural areas in the difficulties of attracting and retaining health personnel. The majority o f staff inrural areas are above fifty and when they retire, it is difficult to replace them. There are monetary disincentives since staff i s rural areas are paid very poorly. However, there are also non-monetary factors (intellectual and social isolation, the lack o f amenities such as electricity and transport, absence o f professional support). These factors are harder to address and the fiscal implications of addressing these factors are large. 61. Some countries have addressed this problem by attracting young medical graduates for a fixed term appointment in rural areas with substantial financial incentives (allowances, waiver o f student loans). Another option i s training local recruits in medicine and using community health workers (feld~hers).'~Feldshers in the Kyrgyz Republic are already frontline providers for 25 percent o f the population. There i s a need to determine how to train the feldshers and linking them to the service delivery chain to provide quality services and continuity o f care. Supply o f pharmaceuticals i s also a problem and the Ministry of Health i s currently implementing a rural pharmacy pilot. In one o f the poorest and hard-to-reach oblasts o f the Kyrgyz Republic, Naryn, a pilot intervention i s focusing on community health promotion activities linked to the formal public health, health promotion and clinical service delivery networks. After only one year o f implementation, village health committees are operational, iodine deficiency has been reduced and households are addressing nutritional deficiencies among women and ~hi1dren.l~ This pilot has been evaluated and if successfully scaled-up could have a large impact on achievingthe MDGs. l3World Bank. 2005. Global Monitoring Report. The World Bank, Washington DC. l4Schuth T. 2004. Community Action for Health inKyrgyzstan: Approach and First results of the Pilot Project in N a r y oblast, Kyrgyztan. Kyrgyz-Swiss Health Reform Support Project, Bishkek, The Kyrgyz Republic. 26 Box 2: The Kyrgyz HealthReforms and Impact on Achieving the Health MDGs The Kyrgyz health sector, similar to that o f other countries o f Central Asia, was highly dependent o n public financing and 87 percent o f the health sector was funded from public sources. Accessibility to health services and financial protection was high. This system achieved significant gains in health outcomes and morbidity and mortality from infectious diseases was low. With the collapse o f the Soviet Union, the Kyrgyz Republic, like other CA countries, faced a severe financing problem inthe health sector. Health services almost collapsed and informal payments emerged as a major problem. Vaccine-preventable diseases such as meningitis began to re-emerge, and the prevalence o f TB beganto grow. Designinga ComprehensiveHealthReformProgram: Leaders inthe health sector became acutely aware o f the multiple problems facing the health sector and realized that they were facing a dual challenge: (i) addressing new problems such as financial inaccessibility and the deterioration o f existing health services, (ii) a resource-constrained environment where public financing for health could not feasible be increased rapidly to reach pre-transition levels. A comprehensive health sector reform program was designed to address these multiple challenges. The key elements o f this program were: (i) designing a health financing system that would improve financial accessibility and generate adequate provider incentives for good performance, (ii) redress regional inequities in financing, (iii) strengthen public health and primary health care services so as to improve the cost-effectiveness o fhealth services. Implementation o f the reforms began in 1996 and in1997, the Mandatory HealthInsurance Fund(MHIF)was formed. Financing for the MHIF is largely frombudget funds and social insurance contributions only constitute 13 percent o f total MHIF financing. A s the Singe-Payer in the health system, the MHIF pools funds from various sources, thereby enhancing risk pooling and reducing fragmentation in financing. As a strategic health purchaser, the MHIF contracts with public providers to deliver health services. The contracts with providers are o n an output-basis. For example, hospitals are paid according to the volume of cases and primary care doctors o n the basis o f number o f enrolled population. The MHIF finances a State Guaranteed Benefits Package under which: (i) primary care i s free for all citizens o f the country provided they enroll with a primary care doctor (also called family doctor), (ii) secondary care requires a co-payment but for vulnerable groups (women, children, elderly and persons with social benefits), the co-payments are reduced. Recognizing that payments for drugs was a major financial access barrier, the MHIF has developed a Drug package. Vulnerable groups pay reduced co-payments for drugs too. The Benefits Package has been widely advertised among the population and the MHIF has a hotline where patients can cross-check information given by providers to complain about services. Although ownership o f health facilities i s still public (largely owned by local governments), health facilities have the autonomy and flexibility to allocate their budgets to achieve desired outcomes. Hospitals have developed business plan. Co-payments are retained by hospitals. Health providers have the autonomy to determine staffing levels and hire and fire staff. The health reforms were implemented on a staggered basis. The early reform oblasts were Issyk-kul, Chui, N a r y and Talas. The new reformers are Batken and Djalal-abad and the there are those which have not yet joined the healthreforms (Bishkek city). KeyResults of the HealthReforms: Health insurance coverage inthe country i s 95 percent and the entire population o fthe country has access to free primary care; Inearly reform oblasts, the increase inper-hospitalizationpaymentrate is 20 percentagepoints lower than inlate reformoblasts. The distribution o futilization o fhealthcare services across socio-economic groups i s more equal in2003 than it was in2000. This i s the case for both outpatient and inpatient care. This shows that the health reforms appear to play an equalizing role on utilization across the country. 27 Figure 16: Distribution of Visits (primary care +outpatient specialist) 18 16 14 12 10 +Visit rate 2001 a -%--Visit rate 2003 6 Poorest 2nd 3rd 4th Richest Source: Jakab et al. 2005. (Jakab et a1.2005 "Who benefits from the Kyrgyz Single-Payer System: Analysis o f the incidence o f Public Expenditures" Forthcoming) Figure 17: Access to Hospital Care 10',-: +Hospitalization rate 6 2001 -a+--Hospitaliz#ation rate 4 2003 0 ' 3 Poorest 2nd 3rd 4th Richest Source: Jakab et al. 2005. (Jakab et a1.2005 "Who benefits from the Kyrgyz Single-Payer System: Analysis o f the incidence o f Public Expenditures" Forthcoming) The prevalence o f acute respiratory infections - a key factor for infant and U 5 M R has declined significantly. This i s due to improved quality o f care and access to key drugs for treating AN. Immunization coverage i s almost 99 percent. This i s a huge improvement over the post-transition period when immunization services collapsed leading to outbreaks o f meningitis and other diseases among children. Although the number o f TB cases has not decreased significantly, mortality rate from TB i s showing continuous decline. This i s attributable to the successful implementation o f DOTS. The hospitalization rate for key health problems inthe population - ulcers, asthma and hypertension - has been reduced, which points to the success to access and quality of primarycare. Source: Jakab et al: Who Benefits from the Kyrgyz Single-Payer System: Analysis o f the Incidence o f Public Expenditures, Chakraborty S et al: The Kyrgyz Health Reforms: Lessons for other CIS-7 countries, Various papers byKutzinJ, WHO-DfiD Health Policy Analysis Unit,2000-2003. 4. Addressing Incentives for Frontline Providers and Consumers will be a Critical Factor for Achieving the Health and Education MDGs. 62. Under the health reforms (See Box 2), significant progress has been made in changing health financing and delivery mechanisms to generate incentives for performance among providers. In most countries with a successful capitation system for primary care, the capitation formula i s risk-adjusted so 28 that doctors that have a higher percentage o f at-risk population have the incentives to treat these groups effectively. Otherwise, under a capitation system, there i s a danger that high-risk groups will be under- treated. Inaddition to higher payments for high-riskgroups, bonus payments to primary care doctors for achieving preventive and public health goals o f the country are also common. So far, the Kyrgyz Republic has only instituted the first level o f changes in capitation payments. Further refinements, included linked to the MDGhealth goals will be needed (See Box 3). Box 3: StrengtheningProviderand Consumer Incentivesfor Achievingthe Health andEducation MDGs School-based Performance Awards in Chile: Since 1996, Chile has been awarding "top-performing'' schools in each region. Ninety percent o f the award goes directly to teachers (inproportion to their hours o f employment) and 10 percent is allocated to school. Schools are divided into comparison groups within each region o f the country based on location (rural, urban), education level (primary, secondary) and socioeconomic status o f parents. In this way, the performance o f poor rural schools i s not compared head-to-head with that o f richer urban schools. Next, an index o f school performance i s calculated based o n standardized test ingrades 4,8 and 10. This index i s weighted for average test level (37 percent) and improvements in test scores (28 percent) and includes other criteria such as equality o f opportunity- based on student retention and "no discriminatory practice" (Mizala A and Romaguera P 2002. "Evaluacion del Desempeno e Incentivos en la Educacion Chilena." Cuandernos de Economia, 35(2):392-4 17. Promoting Preventive Services: Paying for Performance in Haiti: In Haiti, nongovernmental organizations provide basic health services such as immunizations and prenatal and maternal care. In an effort to improve the effectiveness o f these organizations, a performance-based contracting system was initiated. The providers are operating under a payment system that reimbursed their expenses up to a ceiling. The new system sets performance targets and withholds a portion o f the historical budget, allowing the providers to earn back the withheld amount plus a bonus iftargets are met. After just one year o f implementation, there was a marked increase inimmunization coverage. The percentage o f mothers reporting use o f Oral Rehydration increased as did the number reporting correct use. To motivate staff, the nongovernmental organizations implemented a bonus scheme for staff including community health agents. (Eichler R, Auxila P and Pollock J. 2002. Promoting Preventive Health Care: Paying for Performance inHaiti. Incentives and Enablers for Addressing TB: Incentives and enablers for providers and TB patients is considered crucial for improving TB detection and treatment. Various schemes have been piloted in countries and targeted all those diagnosed with TB or specific groups such as ex-prisoners, those with multi drug-resistant TB or those who have a history o f defaulting on treatment. Incentives and enablers included money, food and transport. Actions requiredby patients included clinic attendance and adherence to treatment. For providers it included actions such as home visits to deliver Directly Observed Therapy (DOTS),patient referral. Positive results o f such schemes include: improved case finding, higher cure rates and improved referral rates to appropriate testingheatment centers (Management Sciences for Health, 2002. D o incentives and enablers improve TB Outcomes) POLICY AND PROGRAMMATICDIRECTIONS 63. The first implicationof the findings of this report is the need for adopting a results-based framework for monitoring the MDGs. In the context o f the MDG Progress Reports, a fledgling monitoring framework i s already in place. There i s a need to build upon and institutionalize this framework to involve line ministries and other relevant groups. Existingdata could be used to build upon this framework - linking to the extent possible, outcomes to intermediate indicators/processes, outputs and inputs. The framework could also explicitly acknowledge gaps in data and identify actions for addressing these gaps. Inthe absence o f such as framework, it will not be possible to make the linkages between the National Poverty Reduction Strategy (NPRS) and the MDGs, to integrate the MDGs into the medium-term budget framework (MTBF), to hold line ministries and local governments responsible for results and streamline and harmonize donors support for the health and education MDGs. A results-based 29 framework would also help inmonitoring progress on intermediate results, inbuilding accountability and inkeeping key stakeholders, including civil society, informed about the progress towards the MDGs. A results-based monitoring framework for the MDGs needs to include disaggregated analysis o f outcomes. Butit also needs to go beyondoutcomes to identifysome o fthe underlying factors (determinants) that are responsible for disparities in outcomes. While the health and education sectors need to lead the formulation o f such as framework, other actors also need to be involved. These include donors, non- governmental organizations, service providers and uses, and the National Statistical Office (which i s responsible for poverty monitoring). The overarching impetus for such a fi-amework needs to be provided by the Comprehensive Development Framework and National Poverty Reduction Secretariats. The results on monitorable actions under the results framework could be integrated into the MDG Progress Report which i s already producedregularly by the Government incollaboration with the UnitedNations. 64. The second area that needs attention i s strengthening the medium-term expenditure framework (MTEF) process and linking budget execution more closely with the stated policy priorities of the Government (poverty reduction and MDGs). A medium-term expenditure framework, if properly implemented, provides a very powerful tool for operationalizing the health and education MDG. A well-implemented MTEF can improve the effectiveness and governance o f budget management. It can help the Kyrgyz Republic link outputs and outcomes to ensure consistency with sectoral expenditure frameworks and can generate incentives for line ministries to improve allocation and technical efficiency. The Kyrgyz Republic has implemented the first level o f the MTEF &e., the medium-term fiscal framework) and i s currently strengthening the next level, i.e., the medium-term budget framework (MTBF). The next stage i s to move to an MTEF inwhich line ministries' expenditure programs are defined on the basis o f outputs and outcomes and linked to a results agreement. This will take time, but it i s important to advance the process by building the capacity o f line ministries and the Ministry of Finance to move inthis direction. The implementation of the MTEF to date inthe Kyrgyz Republic has involved a fragmented process without strong ownership and leadership. The limited technical capacity o f line ministries to formulate defensible and well-documents sector expenditure programs has also been a constraint. Nonetheless, there i s strong donor support for the MTEF process and the current (interim) Government i s also committed to continuing efforts on the MTEF. Donor support for the poverty reduction strategy process and the development o f a health Sector Wide Approach (SWAP) also provide opportunities for strengthening the linkages between the MDGs and expenditure allocations inthe context o f the MTEF. 65. The third implication is the need for effective targeted sectoral interventions. A stronger focus on disaggregated data analysis o f MDGoutcomes and underlyingfactors needs to be complemented by the identification o f targeted interventions that will alleviate the constraints to fuller attainment o f the MDGs. Many potential interventions have already been identified. The challenge i s to carefully evaluate pilot interventions including pre-conditions for scaling-up, the associated costs and implementation requirements, and integrating into sector programs and sector expenditure plans. The same i s true for strengthening provider and user incentives to achieve the MDGs. This i s not a small task and will require Ministries o f Health and Education to prioritize the MDGs in their day-to-day work and create institutional structures that facilitate coordination between the various vertical program (as in the case o f health). Ownership o f the MDG agenda i s key for this. In both sectors, elements o f ownership are already present. For example, MDGs feature prominently in the Government's new health reforms strategy and the education sector i s already implementing key interventionsto achieve fuller attainment o f the education MDGs. Donor commitment to supportingthe health and education MDGs and ownership o f the MDG agenda beyond line ministries i s critical for generating incentives for further sectoral focus on the MDGs. From the sectoral perspective, it i s also important to pay attention to the management flexibility needed for maximal effectiveness o f provider inputs. In the education sector, conflicting responsibilities for school management and deficiencies o f the school financing formula create a situation 30 in which local governments, which are intended to manage their schools, have neither the incentives nor the means to reconfigure schools or deploy teachers more efficiently. 31 Annex Table 1: InternationalMDGsand their Modification for the Kyrgyz Republic [nternational Formulationfor the [nternational 1 Formulationfor the Formulation Kyrgyz Republic Formulation Republic $chieve universal Achieve universal basic Ensure that, by 2015, IIKyrgyz Ensure that, by 2015, ximary education secondary education Zhildren everywhere, children everywhere, boys and girls alike, will boys and girls alike, will be able to complete a full be able to complete a full course o fprimary course of basic secondary schooling schooling Reduce child mortality Reduce child mortality Reduce by two-thirds, Reduce by two-thirds, between 1990 and 2015, between 1990 and 2015, the under-five mortality the under-five mortality rate rate L p r o v e maternal health Improve maternal health Reduce by three-quarters, Reduce by three-quarters, between 1990 and 2015, between 1990 and 2015, the maternal mortality the maternal mortality rate rate Combat HIVIAIDS, Combat HIVIAIDS, Have haltedby 2015 and Have halted by 2015 and malaria and other diseases malaria and other diseases begun to reverse the begunto reverse the incidence o f malaria and spread o f HIVIAIDS, other major diseases have halted and begun to reverse the incidence o f malaria and other major disease (e.g. Brucellosis). 32 m m d m m m m 3 s Y E 1 ~~ Annex Table 3: Listof Organizations whose Representativeswere Interviewed to Identify Health MDGBottlenecks andInterventions 1. Bishkek Territorial Department o f Health Insurance Fund 2. Mandatory Health Insurance Fund 3. Medical Information Center, Ministryof Health, 4. Primary Health Care Department, Ministryof Health 5. Migration Services Department, Ministryof Foreign Affairs 6. RepublicanAIDS Center 7. Project HOPE 8. ICRC 9. Swiss Red Cross, Kyrgyz-Swiss Health Reform Support Project 10. UNAIDS 11. UNICEF, Health Project 12. UNICEF, NutritionProject 13, USAID Regional Mission for Central Asia, Health Specialist 14. Department o f Obstetrics and Gynecology, Kyrgyz Russian Slavic University. 38 REFERENCES Chakraborty, Sarbani, and J. Kutzin. 2002. KyrgyzRepublic Public Expenditure Review: Budgeting and Expendituresinthe Health Sector. InWorld Bank Public Expenditureand InstitutionalReview.The InternationalBank for Reconstruction and Development/The World Bank, Washington DC. Charkraborty, Sarbani et al. 2002. The Kyrgyz Health Reforms: Lessons for other CIS-7 Countries. BackgroundPaper for CIS-7 Paper. The World Bank, Washintgon DC. Eichler,Rena, P. Auxilia, and J. Pollock. 2001. "Promoting Preventive Health Care: Paying for Performance inHaiti." InContractingfor Public Services: Output-Based Aid and Its Applications, eds. P.J. Brook and S. Smith.Washington, DC: World Bank InternationalFinance Corporation. Filmer,Deon, Jeffrey Hammer and Lant Pritchett.2000. Weak Linksinthe Chain: A Diagnosis ofHealth Policy inPoor Countries.World Bank ResearchObserver 15 (2) 199-224. Filmer,Deon. 2003. "Determinants ofHealthand EducationOutcomes." BackgroundNote for the WDR 2004. Godinho, Joana et al. 2003. HIV/AIDS and Tuberculosis in CentralAsia: Country Profiles. World Bank, WashingtonDC. Godinho, Joana et al. 2005. Reversing the Tide: Prioritiesfor HIV/AIDSPrevention in Central Asia. The World Bank, Washington DC. Godinho, Joana et al. 2005. Stopping Tuberculosis in Central Asia: Prioritiesfor Action The World Bank, WashingtonDC. Government o f the Kyrgyz Republic.2004. PRSP Progress Report. The Government of the Kyrgyz Republic,NPRS Secretariat, Bishkek,Kyrgyz Republic. Government of the Kyrgyz Republic.2003. Poverty Reduction Strategy Paper. The Government of the Kyrgyz Republic, NPRS Secretariat, Bishkek, Kyrgyz Republic. Government of Uzbekistan. 2005. Welfare Improvement Strategy Paper of the Republic o f Uzbekistan for 2005 -2010. The Government of Uzbekistan, Tashkent, Uzbekistan. Jakab, MeIitta et al. Forthcoming. "Who benefits from the Kyrgyz Single-Payer System: Analysis of the Incidence of Public Expenditures". WHO-DfID Health Policy Analysis Unit,Bishkek,The Kyrgyz Republic. Kyrgyz News, September 2,2002. Kyrgyz Ministryof Education and Culture, UNICEF, UNESCO, and the Centre for Public Opinion Studies andForecast.2001. MonitoringLearning Achievement: National Survey o f Primary Education Quality, Bishkek. Kyrgyz Ministryof Finance. March 31,2005. Medium-TermBudget Framework 2006-2008. Available at http://www.minfin.kg/publications/feb/spb/prognos.htm 39 Loevinsohn, B.and A. Harding. 2004. Contractingfor the Delivery of Community Health Services: A Review of Global Experience. HNP Discussion Paper. The World Bank. Washington DC. Management Sciences for Health. October 2003. Building An Evidence Base on theRole of Incentives and Enablers in Improving TB Control Program: Summary of Findings to Date. Available at http://www.msh.org/projects/rpmplus/3.5.5.htm Manas Health Policy Analysis Project, Policy ResearchPapers Series, 2000-2003. Available at http://eng.hpap.med.kg;/Articles/ViewSection.aspx?ArticleID=170 Mertaugh, Michael. Kyrgyz RepublicPublic Expenditure Review: Budgeting and Expendituresinthe Education Sector, February 24,2003. MinistryofHealth. 2005: Evaluatingthe ImpactofHealthReforms: The ImpactofEconomic Restructuringon Economic Effectiveness and Accessibility. Joint Evaluation of MOH, Kyrgyz Republic, WHO and the World Bank, Bishkek,The Kyrgyz Republic Mizala, A. and P. Romaguera. 2002. "Evaluacion del Desempeno e Incentivos en la EducacionChilena." Cuandemos de Economia, 35(2):392-417. Rajkumar, Andrew Sunil, and Vinaya Swaroop. 2002. "Public Spending and Outcomes: Does Governance Matter?" The World Bank Policy Working Paper 2840. Washington, DC. Available at http://econ.worldbank.org/view.php?type=5&id=l5176 Schuth,T. 2004. Community Action for HealthinKyrgyzstan: Approach andFirstresults ofthe Pilot Project inNarynoblast, Kyrgyztan. Kyrgyz-Swiss HealthReform Support Project, Bishkek, The Kyrgyz Republic. UNICEF, UNESCO and the Centre for Public Opinion Studies and Forecast. 2002-2003. Monitoringo f Learning Achievement 11: Grade 8; National Survey of Quality of Learning inthe 8Ih Grade. Bishkek. UnitedNations. 2005. UnitedNations Development Assistance Framework: 2005-09. The United Nations, Tashkent, Uzbekistan. UnitedNations. 2004. MillenniumDevelopment Goals NeedsAssessment: Tajikistan, Preliminary Report. The UnitedNations, Dushanbe, Tajilustan. UnitedNations DevelopmentProgram. 2003. The Kyrgyz Republic:MillenniumDevelopment Goals ProgressReport, The UnitedNations, Bishkek. Wagstaff, Adam and Mariam Claeson. 2004. Risingto the Challenges: the MillenniumDevelopment Goals for Health. World Bank, Washington, DC. World Bank. 2005. Republic of Tajikistan-Health Sector Note, Human Development Unit,Europe and Central Asia Region, The World Bank, WashingtonDC. World Bank. 2005. Global MonitoringReport. The World Bank, Washington DC. World Bank. 2004. MillenniumDevelopment Goals for HealthinEurope and Central Asia: Relevance and Policy Implications. The World Bank. Washington DC. 40 World Bank. 2004. Making Services Work for Poor People. The World Bank and OxfordUniversity Press. Washington DC. World Bank. 2003. World Bank estimates producedfor the Joint World Bank-IMF paper on: Supporting Sound Policies with Adequate and Appropriate Financing. International Monetary Fund, Washington DC. Wuhib, Tadesse et al. 2003. "Underestimation of InfantMortality RatesinOne Republicofthe Former Soviet Union." Pediatrics Vol. 111No. 5 May 2003. Accessed on March 21,2005 at ht~://pediatrics.aappuplications.orP/cllll1/5/e596 41