GEORGIA PUBLIC EXPENDITURE REVIEW Building a Sustainable Future June 2017 Report No: 114062-GE GEORGIA PUBLIC EXPENDITURE REVIEW Building a Sustainable Future June 2017 Macroeconomics and Fiscal Management Europe and Central Asia Region GEORGIA – GOVERNMENT FISCAL YEAR January 1 – December 31 CURRENCY EQUIVALENTS (Exchange Rate Effective as June 28, 2017) Currency Unit = Georgian Lari (GEL) US$1.00 = 2.4172 GEL ABBREVIATIONS AND ACRONYMS ADB Asian Development Bank MOLHSA Ministry of Labor, Health and Social Affairs CIT Corporate Income Tax MRDI Ministry of Regional Development and CVD Cardiovascular Disease Infrastructure DALYs Disability Adjusted Life Years NBG National Bank of Georgia DCFTA Deep and Comprehensive Free Trade Area NCDs Non-Communicable Diseases EBRD European Bank for Reconstruction and NHA National Healthcare Association Development NPL Non-Performing Loans ECA Europe and Central Asia OECD Organization for Economic Cooperation ECOPA European Consensus-Platform for Alternatives and Development EIB European Investment Bank OOP Out of Pocket ERP External Reference Pricing PER Public Expenditure Review ESCO Electricity System Commercial Operator PHI Private Health Insurance EU European Union PIM Public Investment Management FDI Foreign Direct Investment PIT Personal Income Tax FIZs Free Industrial Zones PPP Public Private Partnership GDP Gross Domestic Product PPT Percentage Point GEL Georgian Lari SOEs State Owned Enterprises Geostat State Department of Statistics of Georgia SSA Social Services Agency GIZ Gesellschaft für Internationale Zusammenarbeit TIMSS Trends in International Mathematics and HIV-TB Human Immunodeficiency Virus Tuberculosis Science Study HMR High Mountainous Regions UHC Universal Healthcare IFC International Finance Corporation USAID United States Agency for International IFI International Financial Institution Development IMF International Monetary Fund VAT Value Added Tax IRP Internal Reference Pricing WHO World Health Organization MDR-TB Multi-Drug-Resistant Tuberculosis HNP Health, Nutrition and Population MIP Medical Insurance Program PPA Sector Power Purchase Agreements MOE Minister of Economy WDI World Development Report MOED Ministry of Education NCDC MOF Ministry of Finance TSA Targeted Social Assistance Regional Vice President Cyril Muller EFI Vice President: Jan Walliser Country Director: Mercy Tembon Senior Director: Carlos Felipe Jaramillo Director: John Panzer Practice Manager: Maria Gonzalez Miranda Task Team Leader: Congyan Tan, Mariam Dolidze Table of Contents Acknowledgements i Executive Summary ii Chapter 1. Macroeconomic and Fiscal Challenges 1 A. Overview 1 B. Recent Macroeconomic Developments and Vulnerabilities 2 C. Fiscal Developments 11 D. Key Challenges 19 E. The Way Forward 24 Chapter 2. Strengthening Health Sector Spending 26 A. Overview 26 B. Policy and Institutional Context 27 C. Overview of Health Spending 30 D. Health Spending and Key Health System Goals 35 E. Key Challenges 46 F. The Way Forward 49 References 53 Annexes 55 List of Figures Figure 1.1. GDP Growth and Per Capita GDP 3 Figure 1.2. Drivers of Growth, 2011–2016 3 Figure 1.3. 2015-2016 Growth, ECA Countries 3 Figure 1.4. Trade and Current Account Deficit 3 Figure 1.5. General Government Revenues, Expenditures and Fiscal Deficits 6 Figure 1.6. Public, Private and Total Debt 6 Figure 1.7. Drivers of Fiscal Deficit, Change in Fiscal Balance 6 Figure 1.8. Fiscal Balance in Georgia 11 Figure 1.9. Gross Public Debt Composition 12 Figure 1.10. General Government Deficit Financing by Source 12 Figure 1.11. Pre-reform Tax Collections in Estonia and Georgia 14 Figure 1.12. Structure of Tax Collections 15 Figure 1.13. Tax Revenue for Georgia and Other Countries by Income Groups 15 Figure 1.14. General Government Expenditure 16 Figure 1.15. Composition of General Government Spending by Economic Classification 16 Figure 1.16. General Budget Decomposition by Functions 18 Figure 1.17. Rising Social Spending and its Components 18 Figure 1.18. Tax Expenditures in Select OECD Countries and Georgia 23 Figure 2.1. Total Health Expenditures by Sources of Financing, 2001-2014 31 Figure 2.2. Share of Government Expenditures on Health during Periods of Coverage Expansion 31 Figure 2.3. Functional Breakdown of Public Health Expenditures, 2001-2014 33 Figure 2.4. Composition of UHC Spending by Beneficiary Group 33 Figure 2.5. Out-of-Pocket Spending vs. Government Health Spending, Georgia and Comparator Countries, 2014 35 Figure 2.6. Residents by Gender and Age, 2015 36 Figure 2.7. DALYs (2012) by Sex 37 Figure 2.8. Infant and Under-Five Mortality Rates, Georgia vs. Comparator Countries, 2014 37 Figure 2.9. Leading Causes of Death, All Ages, Georgia, 2014 39 Figure 2.10. Outpatient Visits and Hospitalization Rates, 2009-2015 39 Figure 2.11. Number of Outpatient Visits Per Capita in Georgia and Comparator Countries, 2014-2015 40 Figure 2.12. Inpatient Procedures Per 100,000 Population Per Year in Georgia and Comparator Countries, 2014 40 Figure 2.13. Caesarian Section Rates Per 1000 Live Births, 2014 41 Figure 2.14. Utilization rates, Georgia and Comparator Countries 41 Figure 2.15. Among People Who Reported Being Ill in the Last Six Months, Share Who Consulted a Health Care Provider, 2010 and 2014 42 Figure 2.16. Share of People Who Reported Needing Hospitalization in the Last Year but Were Not Hospitalized Due to Cost, 2010 and 2014 43 Figure 2.17. Share of Consultations Where Medicine Was Prescribed But Not Purchased Due to Cost, 2010 and 2014 43 Figure 2.18.The Impact of Household’s Health Spending on Poverty, 2015 45 Figure 2.19. Control of Hypertension – International Comparison 48 List of Tables Table E. 1. Policy Options for Consideration v Table 1.1. Diverse Response to Slowdown 8 Table 1.2. Macroeconomic Trends and Forecasts, 2013-20 9 Table 1.3. Medium Term Consolidated Fiscal Framework, 2008-2020 10 Table 1.4. Deep Dive into State Spending and the Key Drivers of Expansion 17 Table 1.5. Tax Expenditure Estimates by Tax Type in 2015 22 Table 2.1. Summary of UHC Benefits and User Charges, 2015 28 Table 2.2. Health Financing Indicators, Georgia and Comparator Countries, 2014 30 Table 2.3. Public Spending on Health and Social Protection, millions GEL, 2012 – 2016 32 Table 2.4. CVDs and Other Non-Communicable Diseases Dominate Disease Burden (2000, 2012) 38 Table 2.5. UHC Tracer Indicators for Prevention and Promotion 42 Table 2.6. Distribution of Public Health Expenditures in 2010 and 2014 44 Table 2.7: Proportion of households with catastrophic health expenditures above 5, 10 and 15 percent thresholds (%) 45 Table 2.8. Incentives Facing Patients and Providers by Type of Care 47 List of Boxes Box 1.1. Policy Responses to Economic Slowdown in the Region 8 Box 1.2. Corporate Income Tax Reform: Estonia’s Experience 14 Box 1.3. Tax Expenditures and Best Practices 20 Box 2.1. Defining Universal Health Coverage 29 Box 2.2. Behind the High Out-of-Pocket Health Spending in Georgia 34 Box 2.3. Gender Differences in Health in Georgia 36 List of Annexes Annex 1. Policy Recommendations of Previous PERs (2012, 2014, 2015) 55 Annex 2. Policy and Institutional Context 61 Annex 3. Breakdown of Public Health Expenditures 66 Annex 4. Health Service Delivery System 68 Annex 5. The Role of Private Health Insurance in Georgia 70 Annex Tables Annex Table 1. Share (%) of the Population Covered by any Health Insurance, 2007, 2010 and 2014 63 Annex Table 2. Summary of UHC Benefits and User Charges, 2015 64 Annex Table 3. UHC Program Coverage by Region, 2016 (First Six Months) 66 Annex Table 4. Breakdown of UHC Program by Type of Spending, 2013 - 2015 66 Annex Table 5. Breakdown of Spending on Vertical Programs 67 Annex Table 6. Primary Care Visits, 2009 – 2015, millions 68 Annex Table 7. Number of Hospitals and Hospital Beds by Regions, 2015 69 Annex Table 8. Number of People Covered by Private Insurance Companies, 2013-2015 70 Annex Figures Annex Figure 1. Georgia’s Path to UHC 63 Annex Figure 2. VHI Share of Private Health Spending in the European Region, 2014 81 Annex Figure 3. Relationship between PVHI and OOP Share of Total Health Spending, 2012 72 Acknowledgements This Public Expenditure Review was prepared by a Bank team led by Congyan Tan (Co-Task Team Leader, Senior Economist) and Mariam Dolidze (Co-Task Team Leader, Senior Economist), and comprising Aparnaa Somanathan (Senior Economist), Hui Sin Teo (Health Specialist), Nino Moroshkina (Consultant), Rouselle Lavado (Economist), Eko Pambudi (Research Analyst), Elena Bondarenko (Economist), Cesar A. Cancho (Economist), Musharraf Cyan (Consultant), and Mariela Sanchez (Consultant), with a much appreciated support from Zakia Nekaien-Nowrouz (Program Assistant), Sarah Nankya Babirye (Program Assistant). The report was edited by Christopher Stewart. This report is a joint work by the Macroeconomic and Fiscal Management (MFM) Global Practice and the Health, Nutrition and Population (HNP) Global Practice. The primary authors of the report are Congyan Tan and Mariam Dolidze from MFM for Chapter 1, and Aparnaa Somanathan from HNP for Chapter 2. The team is grateful for comments and advice from Mona Prasad (Senior Country Economist), Nistha Sinha (Senior Economist), Rajul Awasthi (Senior Public Sector Specialist), and Carolina Sanchez (Senior Director). The team is grateful for overall guidance provided by Mercy Tembon (Country Director), John Panzer (Director), Maria Gonzalez Miranda (Practice Manager, MFM), Miria Pigato (Practice Manager, MFM), Enis Baris (Practice Manager, HNP), and Genevieve Boyreau (Program Leader). The team acknowledges the excellent collaboration of the Georgian authorities, in particular the Ministry of Finance and the Ministry of Labor, Health and Social Affairs, as well as development partners. And the support and guidance of peer reviewers Vinaya Swaroop (Lead Economist), Dino Merotto (Lead Economist), Ajay Tandon (Lead Economist), Elene Imnadze (Representative), and Owen Smith (Senior Economist) are highly appreciated. i EXECUTIVE SUMMARY Introduction 1. This Public Expenditure Review (PER) was prepared at the request of the Ministry of Finance of Georgia; its analysis of public spending is designed to inform the Georgian authorities on the fiscal policies that support growth and equity. The Georgian government’s sound macroeconomic and fiscal management has provided a solid foundation for the country’s strong economic performance in the past: after a decade of fast growth (averaging over 5 percent), and despite a downturn at the wake of the global financial crisis in 2008-09, Georgia has become an upper-middle- income country with a per capita GDP of $3,767 (2015). Buoyant levels of economic activity also helped reduce the poverty rate at $2.5/day from 47 percent in 2010 to 31 percent in 2015. As Georgia strives to meet the challenges of a fluid global economic environment, this PER is intended to support the government’s efforts to secure macroeconomic and fiscal sustainability to promote growth and equity. 2. This is a testing time for Georgia’s public finances, as the country faces slowing growth and a difficult regional context. Georgia is an open economy vulnerable to external shocks; as the geopolitical events and the plunge in oil prices substantially weakened the major economies in the region, Georgia’s growth has slowed down. With the regional economy projected to remain weak, Georgia may face diminished revenues. The government has increased excises and taken one-off measures to boost revenues, but future revenue options remain limited by legislation. Moreover, public spending faces pressure going forward—stemming from rising current expenditures (including for social programs, such as healthcare), the government’s plan to expand public investment projects, as well as increase net lending to state-owned enterprises (SOEs). Balancing these pressures to ensure that fiscal policies remain supportive to macroeconomic stability while ensuring that priority spending is adequate to support growth and social objectives will be a key challenge for the authorities going forward. 3. This Public Expenditure Review (PER) is organized in two chapters; the first focusing on the overarching macro-fiscal challenges facing Georgia, and the second “zooming in” to the challenges for public spending in the health sector. In particular, the PER provides an overview of the recent macroeconomic and fiscal developments highlighting major drivers of rising spending, and analyzes potential revenue gains that could be derived from eliminating tax expenditures (Chapter 1). The second part of the report carries out a detailed review of the health spending in Georgia, providing both the context and rationale for implementing the much needed Universal Healthcare (UHC) program in 2013, while highlighting the need for properly managing the existing short-term cost pressures, as well further improving the system’s long-term efficiency and sustainability (Chapter 2). 4. This PER is an integral part of the programmatic series of PERs (2012, 2014, and 2015) providing new analyses and recommendations that are complementary to the existing ones. Many of the analyses and recommendations from previous PERs stay valid to-date (see annex 1 for the recommendations and the progress of government adopting them). The PER (2012) analyzed the rising expenditure pressures on social and capital spending, and presented options for fiscal consolidation, including measures to improve selectivity in capital expenditures, enhance the sustainability of the pension program, and the coverage of targeted social assistance. The PER (2014) examined the spending on social protection, health, and education, and provided recommendations including strengthening ii the Social Service Agency, increasing the UHC drug coverage, strengthening preschool education, and improving general and vocational education. It also analyzed quasi-fiscal spending and intergovernmental fiscal relations. A special volume of the PER (2014) studied Georgia’s Public Investment Management and how it could be further enhanced to improve the efficiency of public investment. The PER (2015) highlighted the presence of spending pressures from social programs and analyzed spending efficiency, which led to policy options to direct the redistribution policies towards greater equity, improve agriculture subsidy programs and local government spending. This PER, as a continuation of the series, while focusing its analyses and recommendations strategically on new and specific challenges, it also draws a selected number of recommendations from previous PERs, notwithstanding that most issues covered in previous PERs remain crucial and need to be addressed in parallel. Key Findings 5. Georgia faces growing fiscal risks, linked to rising spending and limited upside to revenues. Fiscal pressures have intensified, as spending increases led to the broadening of the fiscal deficit (from 2.9 percent in 2014 to 3.8 percent of GDP in 2015, and further to 4.1 percent of GDP in 2016). In particular, social spending has been on the rise; among its components the spending on the UHC Program has been notably larger than programmed in recent years. Going forward, the government envisions higher social spending and capital investment. Meanwhile, growth in 2015-2016 slowed to five-year lows of 2.9 percent and 2.7 percent, respectively, while downside risks linked to the weak regional economic outlook continued to weigh on Georgia’s growth prospects, generating pressures on fiscal revenues. The replacement of the profit tax with a dividend tax —which came into effect in January 2017—is expected to result in tax revenue losses of 1.5 percent of GDP in 2017. Although these losses are projected to be offset by a one-off hike in exercise taxes, over the short- and medium-term, the government will have to balance its needs to expand public spending in some critical areas to support growth and poverty reduction with a limited resource envelope. 6. The macro-fiscal analysis indicates that a consolidation on spending is needed to secure a sustainable fiscal path. The revenue options are limited: the so-called Liberty Act narrows the options for tax revenue generation to raising excise taxes, property taxes and reducing tax exemptions1. The excise taxes have been raised multiple times, including the most recent increase approved in December 20162 while gains from removing tax expenditures are no longer large, as the country’s tax system is generally considered efficient. Against this background, regaining fiscal space will need to rest largely on spending cuts, especially given that the recent budget overrun pushed total public spending-to-GDP ratio3 beyond a budget cap envisaged in the Liberty Act4. To adjust spending within means, the government has already undertaken a number of measures, including the containment of the wages and government purchases. In addition, it is advisable to avoid ad-hoc increases in social spending (including through the introduction of new social programs or raising social benefits on existing programs), deepening 1 The Liberty Act (2014) forbids an introduction of new tax or an increase in tax rate (with the exemption of excise) without a referendum. 2 This includes a sharp increase in excise rates on tobacco, fuel and vehicles. 3 Public spending includes current expenditures and increase in non-financial assets. 4 Under the Liberty Act, budget approved by the Parliament needs to have consolidated expenditures lower than 30 percent of GDP. iii pension system reforms5, and maintaining capital expenditures at 5-6 percent of GDP6. Other areas for achieving efficiencies should also be pursued, including on issues identified in earlier PERs—such as improving public investment management, achieving greater control on the transfers from the central government (including lending and equity transfers), and strengthening the management of off-budget contingencies such as those that may have been building up in state owned enterprises (SOEs) and other government’s contingent liabilities7. 7. The sustainability of Georgia’s public finance also rests on the successful implementation and cost management of the much needed UHC System. Since its introduction in 2013, the UHC has become the fastest growing social program, and also a regularly underestimated component of social spending in the state budget. While early gains in health care coverage and equity and financial protection derived from the UHC are remarkable, fiscal space constraints and inefficiencies in its implementation pose threats to financial sustainability of the program and its achievements. This PER identifies three major areas of inefficiency, in particular: i. The Social Service Agency (SSA), responsible for the administration of UHC, remains a relatively passive purchaser of services with little administrative capacity to manage claims expenditures effectively. This difficulty is compounded by the complex payment system for hospitals with various tariff-setting and copayment rules for different types of hospital care, which enables providers to game the system, further undermining SSA’s ability to control costs. ii. The service delivery structure is skewed towards costly hospital and emergency services due to the fragmentation of the primary care system and misplaced incentives for the providers. The primary care system is fragmented and offers little value for money for patients relative to specialist or hospital care. Moreover, primary care providers are paid a fixed rate per patient, which creates incentives for the primary care doctors to push higher-risk patients towards hospital care. Moreover, hospitals are paid based on activity, which creates incentives to pull patients towards ambulance and inpatient care, and towards emergency care in particular (WHO, 2016). iii. There is considerable scope for restructuring spending on pharmaceuticals. Out-of-pocket (OOP) expenditures in Georgia are very large,8 and about two-thirds of OOP spending by the population is on medicines (Ministry of Health Labor and Social Affairs, 2016). Limited coverage of outpatient medicines by the public sector means that patients have little choice 5 The measures on wages and salaries and pensions are drawn from PER (2014). 6 A spending cut on current expenditures is envisioned in the approved 2017 budget. 7 While the latter analysis is out of the scope of this particular PER, it remains a key element in Georgia’s fiscal strategy, and a crucial component of ongoing and future work by the Bank and other International Financial Institutions. PIM guidelines and SOEs risks are part of our dialogue through the Development Policy Operation Lending (DPL) operation, which was initiated based on the findings of PER (2014). The IMF is providing a technical assistance on fiscal risk of SOEs, and the Bank also plans assistance in this area. 8 Reducing the OOP burden of health care expenditures, particularly on outpatient drugs, remains one of the biggest challenges for the health sector in Georgia. Analysis of Household Utilization and Expenditures Surveys (HUES) from 2010 to 2014 showed that OOP spending on drug expenditures represent a significant burden for those with chronic illnesses. Georgians spend in excess of 4 per cent of GDP on medicines annually and this is twice the Organization for Economic Cooperation and Development (OECD) average. Moreover, the vulnerability to fall into poverty is high in Georgia because of the large share of OOP spending by households on healthcare. iv but to pay OOP for drugs at pharmacies. This increases the risk of impoverishment from OOP costs, creates barriers for accessing timely preventive care services, and is associated with an overreliance on costly hospital and emergency care services since inpatient medicines are fully covered under UHC. Current reimbursement and purchasing mechanisms further contribute to inefficient spending. 8. Addressing these constraints to improve financial sustainability and performance in the health sector would require a range of reforms in the short, medium and long term. Going forward, the Ministry of Health Labor and Social Affairs (MOHLSA) would need to move towards more active purchasing of health care, while incrementally strengthening the SSA’s capacity to effectively monitor and control costs. These efforts should be complemented by measures geared to simplifying the healthcare payments system and reduce distortions associated with the current payment methods. In the short run, given fiscal space constraints, there is little scope for expanding coverage for outpatient drugs. However, steps can be taken now to standardize and reduce prices of medicines which the UHC program pays for, in particular, through the adoption of a reference pricing system and consolidated procurement. Once fiscal space improves, an immediate priority would be to increase public coverage of outpatient drugs. Expanding outpatient drug coverage would help improve use of primary care services and reduce the reliance on hospital services. However, significantly re-orienting healthcare use away from hospital services and towards primary care would require significantly more investment in the health sector than what can be generated through efficiency savings alone in the short-run. 9. The attached table summarizes the key recommendations of the report. Table E. 1. Policy Options for Consideration Policy Area Issues Options for Consideration Sequencing Macro-Fiscal Reigning in government Focus efforts on consolidating spending. Short-to spending is challenging, medium-term  Continue to contain government spending on given the need to public wages (including those in legal entities of accommodate social public law) and purchases of goods and services (in spending needs in critical the short-term); align hiring/firing/remuneration areas, while keeping the frameworks with the new civil services law9 (long- composition of spending term). pro-growth.  Refrain from introducing new social programs, especially programs that overlaps with TSA.  Avoid ad hoc increase of pension benefits; introduce indexation of basic pension benefits to inflation; gradually transition towards a contributory pension system.10  Maintain capital spending at 5-6 percent of GDP.  Improve efficiency through implementing Public Investment Management guidelines developed in 2016. 9 New Civil Service Law was adopted in October 2015, enactment has been postponed to July 1, 2017 by 6 months. 10 Draft Pension Law was endorsed by the government for public consultation. v Policy Area Issues Options for Consideration Sequencing Macro-Fiscal The scope for raising Remove tax expenditures, including those on profit tax Medium-term revenues by tax policy and VAT; and tax long-term capital income. is limited by law, and the efficiency gains from further streaming tax system are modest. Health Low cost generic Adopt a reference pricing system to improve the Short-to medicines are generally efficiency of spending on pharmaceuticals in the medium-term less available in retail public sector, using external reference pricing (ERP) pharmacies, compared to and/or internal reference pricing (IRP). more expensive originator Increase the availability, affordability and prescription products, skewing of generic medicines. consumption towards higher-priced medicines, and increase the spending on pharmaceuticals. Health Complex payment systems Enhancing quality and efficiency of hospital care: Short-to for hospitals creates the SSA to continue standardizing the payment rates medium-term opportunities for these and using its leverage as single purchaser further institutions to game the in this area; the SSA to introduce quality and access system and charge higher standards as part of the agreements with hospitals cost. participating in the UHC and vertical programs. Health11 SSA remains a relatively Empower SSA to move towards active purchasing: Short-to passive purchaser of medium-term  Establish clear goals for the SSA, and set up services, which prevents mechanisms for monitoring its performance and it from managing claims regularly report to the public. expenditures more effectively.  Develop the SSA’s analytical capacity to support day-to-day operations, including improvement of IT systems and strengthening of staff analytical skills; introduce a regular management reporting system to foster a culture of regular use of data and evidence.  Re-design the SSA’s internal organizational structure to enable different program teams to work together to tackle overlaps and inefficiencies, and to encourage the development of new and innovative ways of improving purchasing.  Optimize administrative procedures and reduce bureaucracy (paperwork) by minimizing activities that do not add value.  Consider introducing one contract/agreement per provider for the UHC Program and all of the vertical programs. This would enhance the negotiating power (leverage) of the SSA in relation to providers. The analysis of the purchasing system and recommendations are drawn entirely from WHO (2016), a situation analysis of the 11 purchasing system that was carried out by the WHO at the same time as this PER. vi Policy Area Issues Options for Consideration Sequencing Health Over-reliance on costly Strengthen primary care as a way to reduce the reliance Medium-term hospital and emergency on hospital and emergency care services. services as primary care  Expand public sector coverage of essential drugs is weak, and limited or in primary care by expanding the outpatient drug no coverage of drugs in benefit. primary care  Integrate the rural doctor program into the UHC Program; adjust the capitation rate for risk or patient needs; Introduce a performance-based component to SSA reimbursement for primary care. Health Weak primary care, Re-orient the health service delivery system away Medium-to inadequate coverage of from reliance on hospitals towards greater emphasis long-term outpatient drugs and a on primary care, and better coordination between generally fragmented primary and hospital care by: service delivery system  Improve access to quality primary care services. severely compromised the effectiveness and  Strengthen coordination of care: MOLHSA to efficiency in managing incrementally strengthen care coordination by non-communicable aligning incentives of patients and providers with diseases (NCDs). the right objectives and putting in place the right infrastructure and tools, especially electronic data management systems. vii Chapter 1. Macroeconomic and Fiscal Challenges A. OVERVIEW 1.1. Following several years of robust growth performance in the aftermath of the financial crisis of 2008-2009, Georgia’s macroeconomic outlook and fiscal position have deteriorated again, amid a weak external environment and recent policy changes: i. Prior to the geopolitical and oil-price shocks that shattered the region of the South Caucasus in late 2014, Georgia benefitted from a benign external context, marked by strong inflows (including through FDI and remittances), that fueled a buoyant domestic demand, which became the country’s main growth driver. The external accounts largely mirrored these developments, and the foreign inflows supported broad current account deficits (9-11 percent of GDP) up to 2012. The trend was temporarily halted in 2013, as elevated uncertainty due to government transition caused a slowdown in domestic demand and halved economic growth. Net exports led growth during 2013, resulting in a dramatic narrowing of the current account balance. ii. The domestic uncertainties faded relatively quickly, but by the time they did, the external environment was no longer benign, and Georgia could not return to the earlier growth model. The country’s key economic partners (and external income sources) saw their own growth falter as the oil price collapsed in the second half of 2014, and other geopolitical events materialized. Thus, following a very short-lived recovery in 2014, Georgia has struggled to revive its main engines for economic activity. Meanwhile, the external position has weakened—the current account deficit widened back to its historical levels— but this time within a much less favorable context (and outlook) for the external inflows that may help sustain it. The fiscal accounts have also deteriorated, partly driven by large budget overruns, however the effect of this unintended boost in domestic demand has been limited. 1.2. Against this background, the fiscal challenges have intensified, and press both over the short and medium term. Budget overruns have widened the fiscal deficit from 2.9 percent of GDP in 2014 to 3.8 percent of GDP in 2015, and reached 4.1 percent of GDP by the end of 2016. In 2015, larger-than-envisaged spending was driven by an overshoot of the UHC program, as well as a hike in unanticipated loans and transfers to state-owned enterprises (SOEs). In 2016, teacher salaries, social benefits, including the UHC, as well as unbudgeted transfers12 contributed to the increase of financial assets have explained the fiscal loosening. 1.3. At the broader macroeconomic level, it will be important to recognize that the space for counter-cyclical macroeconomic and fiscal policy is limited. Given the risks that external adverse conditions may become long-lived, it will be important to ensure a sustainable fiscal and external position that can withstand further shocks, were they to materialize in the future. In recent years, Georgia’s fiscal space has been gradually consumed by policies that seek to promote growth but lack thorough analysis 12 These were directly paid to the state entities by the donors, which includes transfers to the Tbilisi and Batumi municipalities under EBRD-funded city-bus projects, payments for Defense Ministry contracts and other activities. 1 of effectiveness and also consistency with the economic cycle. Among such policies the most prevailing ones are: the replacement of the corporate income tax with a dividend tax, the planned implementation of an ambitious infrastructure investment program13, or the rising social spending. With fiscal space narrowing, it will be critical for Georgia to rebuild the fiscal buffers by: (i) mobilizing domestic revenue when possible; (ii) consolidating public expenditure, (iii) increasing the efficiency and effectiveness of spending; and (iv) ensuring that the remaining composition of public spending remains pro-growth and protects essential (and high-quality) social and investment programs. More concretely, with the Liberty Act restricting the scope for domestic revenue increases, the government needs to focus on ensuring that the overall expenditure envelope is consistent with maintaining fiscal sustainability14. 1.4. In parallel, Georgia must continue building up macroeconomic and financial resilience from its existing strengths. Throughout this period, the authorities (through the National Bank of Georgia) have appropriately maintained exchange rate flexibility, while focusing monetary policy to preventing strong pass-through effects. Prudent banking supervision has supported the health of the financial system at large, despite the strong vulnerabilities derived from the large financial dollarization prevailing in the country. Going forward, maintaining the integrity of these macroeconomic and financial fundamentals will be critical to jump-starting the economy sustainably, in particular, by continuing to promote the development of robust and healthy private sector. 1.5. This chapter is structured as follows: the next section sets out the macroeconomic challenges and vulnerabilities. Section C discusses fiscal developments and the rising cost of new programs and policies, and section D highlight the key challenges. Finally, Section E outlines policy recommendations. B. RECENT MACROECONOMIC DEVELOPMENTS, OUTLOOK AND VULNERABILITIES Recent Developments 1.6. Georgia enjoyed robust growth in 2010-2014. During this period, GDP grew at 5.6 percent per year on average, supported by external inflows and increased government spending, which boosted investment and consumption (Figure 1.1, 1.2). Economic activity dipped in 2013, reflecting a one-off effect linked to the uncertainties of the government transition, but activity rebounded in 2014 to 4.6 percent, lifted by the last strands of a benign external environment. 13 The tax revenues are expected to stay lower over the medium term under the new policy to only levy tax on dividend. A large infrastructure program is planned for 2017. However, investment projects in Georgia are prone to efficiency loss due to the lack of a sound public investment management (PIM) system. The PER (2014) reviewed Georgia’s PIM system, and identified the lack of a functional system for projection screening, appraisal, review, selection and implementation as the key challenge towards efficient public investment. Currently there is no established PPP framework, but the ADB is working with the authorities on this. 14 Under an unsustainable fiscal policy, where the government cannot sustain its spending with its revenues, government will see its arrears accumulating from required payment on public debt obligation and government contracts. In this case the government may be forced to cut expenditures, sometimes on vital government programs, such as healthcare or pension which could be very unpopular measures. 2 1.7. Hit by external shocks at the end of 2014, Georgia’s growth in 2016 was the slowest since the 2008-2009 crisis. The regional environment sharply deteriorated, with elevated uncertainty that arose from geopolitical tensions, and with the plunge of oil prices and the slowdown of China hitting many economies in the region (Figure 1.3). Georgia, a small open economy, quickly felt the chill with a significant weakening in external demand and a loss of competitiveness due to larger depreciation in the currencies of its trading partners, which more than offset the benefits stemming from cheaper oil imports.15 Georgia’s growth decelerated to 2.7 percent, while total exports fell by 4 percent in 2016 after falling by 23 percent in 2015. A fall in household consumption also contributed to the slowdown in growth. Figure 1.1. GDP Growth and Per Capita GDP Figure 1.2. Drivers of Growth, 2011–2016 (%, $US) (Percent) 20.0% 8.0% 5000 15 15.0% 6.0% 4000 10.0% 4.0% 10 3000 5.0% 2.0% 5 2000 0.0% 0.0% 0 -5.0% -2.0% 1000 -10.0% -4.0% 0 -5 -15.0% -6.0% -20.0% -8.0% 2008 2009 2010 2011 2012 2013 2014 2015 2016e GDP per capita (in U.S. dollars) Private Consumption Public Consumption Real GDP (percent change) Investments Net Exports GDP growth Source: Estimates based on Geostat statistics. Source: Bank staff calculations. Figure 1.3. 2015-2016 Growth, ECA Countries Figure 1.4. Trade and Current Account Deficit (Percent) (% of GDP) 10 8 6 Real GDP Growth, % 2.90 3.04 2.68 4 2.71 2 0 -0.60 -2 -0.46 -4 - 3.70 -2.50 -6 -3.90 -8 -10 2015 2016e 9.90 - -12 Source: Geostat; 2016 growth based on Staff estimates. Source: National Bank of Georgia. 15 Imports of oil products, which accounted for 10 percent of total imports in Georgia, registered a 28 percent fall in 2015. 3 1.8. The deterioration in the external position has led to heightened macroeconomic vulnerabilities. The current account deficit widened to 12.0 percent of GDP in 2015 and further to an estimated 13.3 percent in 2016 (Figure 1.4). Although FDI and official loans remain resilient sources of financing, the large external imbalance generated significant depreciation pressures on the lari. In particular, throughout 2015 and 2016, the lari depreciated by 42 percent against the US dollar, making the Georgian economy—which is characterized by a high level of total external debt and dollarization— more vulnerable. With approximately 80 percent being denominated in foreign currency, Georgia’s public external debt rose to 34 percent of GDP by the end of 2016, up from 27 percent of GDP two years earlier, largely driven by depreciation.16 Total gross external (both public and private) debt rose to 108 percent of GDP by end-201617 (Figure 1.6). The banking sector in Georgia also has high exposure to the exchange rate risk, as over 60 percent of loans are denominated in US dollars. So far, despite a small uptick in the non-performing loans18 (NPLs) to total gross loans ratio from 3.5 percent in the end of 2014 to 3.7 percent by the end of 2016, the deterioration in asset quality has been contained, supported by sound supervision.19 1.9. In response to these shocks, the NBG largely maintained the flexibility of the lari, and raised interest rates to curb inflation. The depreciation of the lari helped safeguard a modest level of reserves of US$2.8 billion in 2016, equivalent to approximately three months of imports20. Meanwhile, inflation pressures emerged from the pass-through of the depreciation, and annual inflation accelerated from 1-2 percent in 2014 to 5-6 percent during 2015. In response, the NBG raised its policy rates in increments about every month, from 4 to 8 percent in 2015. As the annual inflation subsided to almost zero toward the end of 2016, the NBG cut the policy rate to 6.50 percent in September. In response to the emerging inflation expectations caused by the recent depreciation and a significant increase in excise taxes on tobacco and fuel for 2017, the NBG raised the policy rate again from 6.5 to 6.75 percent in January 2017. The monetary policy is geared towards maintaining price stability, with inflation targets at 4 percent for 2017 and 3 percent for the medium-term. 1.10. A budget overrun in 2015 and 2016 raised concerns about the government’s fiscal management, but helped dampen the adverse impact on growth. Although the government planned a budget deficit of 3.0 percent of GDP both for 2015 and 2016—very much in line with the deficit of 16 This includes the government’s Energy Sector Power Purchase Agreements (PPAs), which the Bank staff identifies as a foreign exchange denominated contingency. In particular, these are power purchase and risk sharing agreements issued by the government to the energy SOE Electricity System Commercial Operator (ESCO) to guarantee profitability upon completion of energy projects. It is, however, not officially reflected as a government’s liability. 17 Out of total external debt, 30.5 percent is state debt (mostly from international financial institutions on concessional terms), 9.5 percent comes from SOEs, 20 percent from the banking sector, 20 percent from the corporate sector, and 20 percent from intercompany loans. 18 NPLs are defined by the loans that are more than 90 days overdue. 19 This is calculated following the IMF definition to identify NPL to be loans over 90 days overdue. Under an alternative more conservative definition used by the NBG which accounts loans over 30 days overdue as NPLs, the non-performing loans (NPL) to total gross loans ratio rose from 7.6 percent in the end of 2014 to 8.1 percent by the third quarter of 2016. 20 The increase in reserves from $2.5 billion in 2015 is also partly attributable to the higher reserve requirements for dollar deposits, introduced by the NBG in June 2016. 4 the previous year, larger-than-envisaged spending led to a widening of the deficit to 3.8 percent of GDP (Figure 1.5) in 2015 and further to 4.1 percent of GDP in 2016. Half of the overrun (0.3-0.4 percent of GDP) was attributable to the mounting spending pressure on health expenditures, as the beneficiaries started to take up services from the Universal Health Care (UHC) and inefficiencies emerged in its third year of implementation. The other half of the extra spending arose from the transfers and net lending to state-owned enterprises, which more than doubled in 2015.21 After a budget overrun of 0.8 percent of GDP in 2015, the excess spending led to a fiscal deficit of 4.1 percent of GDP—wider than the budget target of 3.0 percent of GDP for 2016. The economic policy of the administration that came into power after October 2016 elections introduced several fiscal measures on expenditure and revenue side, which are expected to improve fiscal trajectory. However, the pressures on spending are also becoming higher, putting forward greater risks to sustainability. 1.11. In order to return to the sustainable path, the government developed a fiscal consolidation program at the end of 2016. The commitment to implement proposed program has been reflected it in the State Budget Law 2017 and government’s medium-term framework. Moreover, this commitment served as a basis for an agreement with the IMF, with an Extended Fund Facility approved by the IMF board in April22, 2017. The IMF-supported program intends to safeguard growth against external vulnerabilities and thus envisages the deficit to be reduced to 3.1 percent of GDP by 2020. This rests on the successful implementation of a set of ambitious medium-term fiscal plans including efficiency gains in current spending. Namely, restrictions have been imposed for the management of the wage bill on central and local level of the government and additional spending controls on local governments for purchase of goods and services. Efforts to increase efficiency in public healthcare aim to increase emphasis on the most vulnerable and generally improve management of claims. On the revenue side, the agreement acknowledges the government’s decision to increase taxes to offset revenue losses from the corporate income tax reform and finance additional capital spending. According to the program, the authorities are committed to take additional measures, if needed, to meet the fiscal targets. 1.12. The most recent developments in early 2017 were positive. Growth was estimated to recover significantly, at 5 percent in the first quarter of 2017, driven by strong credit growth and a sharp recovery in remittances. Export also recovered by 30 percent year-on-year, but net exports s contributed negatively to growth due to import and remittances dynamics. The fiscal accounts were consistent with the announced consolidation policy, which envisages a flat spending envelope (in nominal terms) and sharply increased tax collections, mostly from excises and VAT. As a result, the overall fiscal balance reached a surplus in the first quarter of 2017. However, the effect of the profit tax reform introduced from January 1, 2017 has yet to be felt, and some of the spending pressures may emerge later in the year. Outlook and Vulnerabilities 1.13. The policy mix has helped support economic growth in the short term, and the IMF- supported program is designed to help Georgia reduce economic vulnerabilities going forward, but key sustainability challenges remain. The regional slowdown is weighing down on Georgia’s own 21 These are primarily IFI (EIB, EBRD, World Bank) financed loans recorded as the central government’s lending to the state energy company. 22 The program entails a three-year Extended Fund Facility (EFF) with access of SDR 210.4 million in funding, which supports future fiscal consolidation of the government. 5 prospects (Figure 1.3). In particular, Georgia’s growth is likely to remain below its long-term potential (estimated at about 4-5 percent) over 2017-2018. Against this background, we present an illustrative “baseline” scenario, which reflects the announced policy plans for the approved 2017 Budget with no additional measures, would likely derive into a fiscal consolidation insufficient to bring deficits back to prudent levels (Table 1.1). The scenario shows that important downside risks remain, including in the external environment. Were these to materialize, they could harm growth further in the short and medium term, worsening the fiscal deficits. This suggests that there is an acute need for the Georgian authorities to shift its priorities into rebuilding fiscal space. To this end, the new IMF program focuses on structural reforms to generate higher and more inclusive growth, focusing on: improving education; investing in infrastructure; making the public administration more efficient; and improving further the business environment to boost the private sector as a growth engine. Figure 1.5. General Government Revenues, Figure 1.6. Public, Private and Total Debt Expenditures and Fiscal Deficits (% of GDP) (% of GDP) -2.8 -2.6 120.0 107.9 108.2 40.0 -2.9 -2.0 -3.6 -3.8 -4.1 100.0 80.3 83.8 82.3 83.9 30.0 -4.0 80.0 -6.4 -6.5 75.0 74.2 20.0 -6.0 60.0 54.7 57.0 54.1 56.6 40.0 10.0 -8.0 -9.2 20.0 34.0 26.8 28.2 27.3 32.9 25.6 0.0 -10.0 2008 2009 2010 2011 2012 2013 2014 2015 2016e 0.0 2011 2012 2013 2014 2015 2016e Revenues and Grants Total Expenditure Public Debt Private Debt Overall �iscal de�icit Source: MOF. Source: NBG. 1.14. Government spending will remain Figure 1.7. Drivers of Fiscal Deficit, Change in high over the short- to medium-term. Social Fiscal Balance spending (including the UHC), is likely to remain (% of GDP) under pressure and containing costs will be Revenues and Grants challenging without structural reforms in this 5 Social bene�its 0 area. With an aging population and upcoming Capital Expenditure Fiscal Balance Composition, % Other expenditures pension reforms, government spending will 3 Fiscal de�icit, % GDP(RHS) Fiscal deficit, % GDP increase further. In addition, the government GDP change plans to significantly raise capital expenditures 1 starting 2017, with a focus on improving road, -5 -1 2009 2010 2011 2012 2013 2014 2015 2016e energy and seaport infrastructures. Over the medium term, the government has announced -3 plans to continue shifting spending towards public investment (which will lead to a rise in -5 -10 capital expenditures of over 8 percent of GDP), while current expenditure will be consolidated Source: MOF and Bank staff calculations. 6 (Table 1.2). In 2017, the approved budget also envisaged increases in the costs of pensions (due to the full year effect of in-year increases in 2016), teacher salaries (similar full year effects), and benefits to the residents in the mountainous region. This planned fiscal expansion in 2017 (a continuation of spending increase from mid-2016) is expected to contribute to a widening of fiscal deficit at over 4 percent of GDP in 2017. These commitments were undertaken before the new IMF program. 1.15. A wider divergence between revenues and expenditures is expected over the medium term if no restraining measures are implemented. In contrast to the expected increases in expenditures, the revenue side will be affected by the upcoming tax policy change to replace the profit tax with a dividend tax, which is expected to reduce tax revenues by 1.5 percent of GDP in 2017. Remedial measures were approved in December 2016, including a sharp increase in excise taxes on tobacco, fuel and vehicles. In particular: (i) a topping-up to the tobacco excise by 60 tetri is expected to generate 0.6 ppt additional tax collections; (ii) the doubling of the petrol excises (150 percent increase in gas excises and 165 percent increase in diesel excises) are expected to generate 0.8 ppt of GDP; and (iii) the introduction of car excises will generate another 0.1 ppt of GDP. In addition to this, the MOF will generate a one-off revenue of 0.5 ppt of GDP by collecting VAT advance payments and abolishing the preferential treatment to postpone VAT taxation on imports from the so called “golden list” of companies. Overall, these measures are expected to cushion the revenue drop in 2017 and help maintain the fiscal deficit at about 4 percent of GDP. However, in 2018-2019, the revenue stream is expected to decline as the profit tax exemptions start to be extended to financial institutions. Other things constant, some recovery could start from 2020, when the positive impact of the dividend tax on business would begin to emerge with the acceleration in private sector growth; a gradual recovery on tax revenues is projected from this point onwards. Box 1.1. Policy Responses to Economic Slowdown in the Region The recent external shocks have a severe impact on Georgia and other economies in the Europe and Central Asia (ECA) region. Russia’s economy—along with other oil producers—have tumbled into recessions while others, mostly small open economies, have also seen their exports and remittances drop sharply, led by the demand contraction of their key trading partners. Many countries’ good progress toward economic development and poverty reduction are at risk, which has prompted governments to rethink their fiscal policy responses. In general, such policy responses to the economic slowdown largely vary across the region. Some countries are pursuing countercyclical policies through fiscal stimuli, while others have cut expenditures to rebuild fiscal buffers (Table B1.1). Bank staff estimates suggest that growth of most economies in the region remains below potential, which helps justify the use of counter-cyclical policy measures. However, not all countries have the resources to enter a period of fiscal expansion to rescue anemic growth—and having some resources does not necessarily mean that the expansion will not hamper fiscal sustainability, especially if not adequately calibrated. For less affluent economies, fiscal loosening can lead to accruing unsustainable levels of debt that may eventually prove difficult to reign in. In addition, in case this economic downturn turns out to be very persistent, the government would find it necessary to adjust sooner to the new realities by aligning spending better with the reduced revenues before fiscal gap widens too much. Belarus, Moldova and Ukraine, with high levels of debt and widening fiscal deficit, have chosen to consolidate. As for Georgia, Bank Staff estimates a one percent negative output gap for 2015. The gap widened from 2.9 percent of GDP in 2014 to 3.8 percent in 2015 close to the ECA average. The level of public debt of Georgia has risen to 41 percent of GDP, above ECA (developing economies) average of 38 percent. The Georgian authorities have increased spending albeit unintentionally, helping stimulate economic growth. However, this strategy led to a greater fiscal deficit and public debt, which the government had to address by generating more revenues from hiking excises, further constraining future revenue options. 7 Table 1.1. Diverse Response to Slowdown Country Output Gap, % Growth Slowdown Fiscal Bal- Debt, % GDP, Policy Response Potential GDP, 2015 2014-2015, ppt ance, % GDP, 2015 Change 2015 Armenia -0.92 -0.6 -4.8 48.8 Stimulus Azerbaijan -6.25 -0.9 -6.2 18.6 Stimulus Belarus -7.47 -5.6 1.7 48.2 Consolidation Georgia -0.99 -1.8 -3.8 41.4 Stimulus Kazakhstan -2.27 -3 -7.8 21.9 Stimulus Kyrgyz Repub- lic 1.08 -0.5 -3 68.3 Stimulus Moldova -1.58 -5.3 -2.3 46.5 Consolidation Russia -5.86 -4.4 -3.5 14.6 Consolidation Tajikistan 0.49 -0.7 -2.1 38.4 Stimulus Turkmenistan 0.58 -3.8 -0.7 21 Consolidation Ukraine -12 -3.3 -1.2 80.3 Consolidation Uzbekistan 3.49 -0.1 0.4 10.5 Stimulus Source: Bank staff calculations. 1.16. Mitigating the risks to fiscal sustainability will require offsetting measures. As noted earlier, the current macroeconomic projections (Tables 1.1 and 1.2) illustrate a “passive” path based on the government’s announcements of remedial measures made by the authorities early in the year. The WB estimates show that, with the government’s first round of these measures, largely based on increases in excises and some cost reduction in public administration, the operational surplus (excluding capital spending) would improve from 2.4 percent of GDP in 2017 to 4.7 percent of GDP by 2020. However, the fiscal deficit would remain above 4 percent of GDP due to a considerable scaled-up in public investments envisioned in the government’s four-point-program. Under the IMF package, the government has committed to undertake further expenditure rationalization to bring the deficit down to about 3 percent of GDP by 2020 (Table 1.2). At this stage, risks remain on how the government’s future plan will be implemented, in particular to make space for the expansion in capital expenditures through the consolidation of current expenditures. In absence of a sharp consolidation of current spending, maintaining a sustainable fiscal path would require public investments to remain between below 7.5 percent of GDP in the medium-run. Going forward, budget revenues and fiscal balances could worsen, given that there are downside risks to growth as the regional prospects remain uncertain. 8 Table 1.2. Macroeconomic Trends and Forecasts, 2013-20 2013 2014 2015 2016e 2017p 2018p 2019p 2020p Actuals Forecasts (Percent change, unless otherwise indicated) National Accounts                 Real GDP (percent change) 3.4 4.6 2.9 2.7 3.5 4.0 4.5 5.0 GDP nominal (in billions of U.S. dollar) 16,140 16,508 13,988 14,333 13,733 14,860 15,994 17,298 GDP per capita (in U.S. dollars) 3,600 3,676 3,759 3,853 3,675 3,973 4,276 4,625 Consumer price index -0.5 3.1 4.0 2.1 5.7 3.3 3.0 3.0 (In percent of GDP, unless otherwise indicated) Investment and saving Gross investment 24.8 29.8 32.1 32.4 33.3 35.0 35.8 36.2 Public 5.9 5.6 7.0 6.5 8.0 8.3 8.8 9.0 Private 18.9 24.2 25.1 25.9 25.3 26.7 27.0 27.2 National Savings 19.1 19.2 20.3 19.1 20.6 20.2 21.2 21.9 General Government Operations         Revenues and grants 27.7 28 28.1 28.4 29.4 28.7 28.3 28.5   Of which: Tax revenues 24.8 25.1 25.1 25.8 26.4 25.9 25.6 25.8 Grants 0.9 1.0 1.0 0.8 0.9 1.0 0.9 0.8 Expenditure and net lending 30.2 31.0 31.9 32.5 33.5 33.4 33.1 32.8 Current expenditure 24.4 25.4 24.9 26.0 25.5 25.0 24.3 23.8 Of which: interest payments 0.9 0.9 1.0 1.2 1.4 1.4 1.3 1.3 Capital expenditure and net lending 5.9 5.6 7.0 6.5 8.0 8.4 8.8 9.0 Primary balance -1.7 -2.0 -2.8 -3.0 -2.7 -3.3 -3.5 -3.0 Overall fiscal balance -2.6 -2.9 -3.8 -4.1 -4.1 -4.7 -4.8 -4.3 (In percent of GDP, unless otherwise indicated) External Sector                 Current account balance -5.8 -10.6 -11.9 -13.3 -12.6 -12.2 -11.4 -9.9   Exports of goods and services 44.5 42.6 44.6 44.0 49.1 48.3 48.1 48.5   Imports of goods and services 57.4 60.6 61.5 58.9 66.1 63.9 62.5 62.8 Remittances 4.7 4.6 4.2 4.3 4.8 4.7 4.8 4.9 FDI (net) 5.1 8.1 9.0 10.0 10.2 9.4 8.8 8.1 Gross International Reserves (in months of total imports) 3.6 3.2 3.3 3.9 3.6 3.7 3.7 3.9 (in millions of dollars) 2,823 2,699 2,521 2,627 2,809 2,904 2,797 2,677 Memo item: IMF program deficit (in percent of GDP) -2.6 -2.9 -3.8 -4.1 -4.1 -3.8 -3.5 -3.1 Source: Georgian authorities; and Bank staff estimates and projections. 9 Table 1.3. Medium Term Consolidated Fiscal Framework, 2008-2020 2008 2009 2010 2011 2012 2013 2014 2015 2016e 2017p 2018p 2019p 2020p % of GDP Revenues and Grants 30.7 29.3 28.3 28.2 28.9 27.7 28.0 28.1 28.4 29.4 28.7 28.3 28.5 Tax revenues 24.9 24.4 23.5 25.2 25.5 24.8 25.1 25.1 25.8 26.4 25.9 25.6 25.8 grants 3.2 2.2 2.3 0.9 1.0 0.9 1.0 1.0 0.8 0.9 1.0 0.9 0.8 Other revenues 2.5 2.7 2.5 2.1 2.4 2.0 1.9 2.0 1.7 2.1 1.8 1.8 1.9 Total Expenditure 37.1 38.5 34.8 31.8 31.7 30.2 31.0 31.9 32.5 33.5 33.4 33.1 32.8 Current Expenditure 28.3 30.1 26.4 23.1 23.3 24.4 25.4 24.9 26.0 25.5 25.0 24.3 23.8 Compensation of 5.3 5.9 5.4 4.7 4.6 5.2 5.2 5.0 5.2 4.8 4.7 4.5 4.4 employees Purchases of goods 8.5 6.4 5.5 5.0 5.0 3.8 3.9 3.8 4.1 4.0 3.9 3.7 3.6 and services Interests 0.6 1.1 1.0 1.2 1.0 0.9 0.9 1.0 1.2 1.4 1.4 1.3 1.3 Subsidies 2.7 2.5 1.8 1.8 2.0 2.0 2.1 2.1 2.2 2.0 2.0 2.0 2.0 Grants 0.1 0.2 0.1 0.1 0.1 0.1 0.0 0.3 0.1 0.1 0.1 0.1 0.1 Social benefits 7.2 8.6 7.8 6.8 7.1 8.5 9.6 9.6 10.0 10.1 9.9 9.8 9.6 Other expense 3.9 5.4 4.8 3.6 3.6 3.9 3.6 3.1 3.2 3.1 3.0 2.9 2.8 Operational (current) 2.4 -0.8 1.8 5.2 5.6 3.3 2.6 3.2 2.4 3.9 3.7 4.0 4.7 Balance Capital Expenditure 8.8 8.4 8.4 8.8 8.4 5.9 5.6 7.0 6.5 8.0 8.4 8.8 9.0 Capital 8.0 8.0 7.4 7.7 7.3 5.1 5.0 5.6 5.1 6.2 6.5 7.0 7.3 Net Lending 0.8 0.4 1.0 1.1 1.1 0.8 0.6 1.4 1.4 1.8 1.9 1.8 1.7 Overall fiscal deficit -6.4 -9.2 -6.5 -3.6 -2.8 -2.6 -2.9 -3.8 -4.1 -4.1 -4.7 -4.8 -4.3 real 2008 Lari, million Revenues and Grants 5,854 5,372 5,515 5,904 6,425 6,367 6,737 6,961 7,220 7,771 7,852 8,091 8,555 Tax revenues 4,753 4,479 4,576 5,270 5,669 5,703 6,043 6,218 6,565 6,978 7,086 7,319 7,745 grants 617 397 444 192 230 205 231 249 212 238 274 257 240 Other revenues 484 497 495 443 525 459 463 494 444 555 492 515 570 Total Expenditure 7,081 7,067 6,790 6,660 7,041 6,954 7,446 7,901 8,265 8,855 9,138 9,463 9,846 Current Expenditure 5,396 5,525 5,155 4,824 5,173 5,605 6,106 6,166 6,615 6,740 6,839 6,947 7,144 Compensation of 1,008 1,083 1,053 976 1,022 1,195 1,256 1,249 1,315 1,269 1,286 1,287 1,321 employees Purchases of goods 1,614 1,175 1,070 1,040 1,103 866 944 939 1,046 1,057 1,067 1,058 1,081 and services Interests 121 202 194 247 215 203 205 257 302 370 383 372 390 Subsidies 512 459 357 366 437 469 516 523 560 529 547 572 600 Grants 12 37 12 11 14 13 10 65 28 26 27 29 30 Social benefits 1,379 1,579 1,526 1,422 1,579 1,965 2,303 2,369 2,547 2,670 2,708 2,802 2,882 Other expense 750 991 941 762 803 895 871 772 818 819 821 829 841 Capital Expenditure 1,685 1,542 1,635 1,836 1,868 1,348 1,340 1,735 1,650 2,115 2,298 2,516 2,702 Capital 1,524 1,468 1,448 1,605 1,628 1,168 1,192 1,386 1,289 1,639 1,778 2,001 2,191 Net Lending 161 73 187 230 240 181 149 349 361 476 520 515 510 Overall fiscal deficit -1,227 -1,694 -1,275 -756 -616 -587 -709 -940 -1,045 -1,084 -1,286 -1,372 -1,291 Source: MOF data and staff estimates. 10 1.17. External vulnerabilities may be protracted, given the weak external demand and high fiscal deficit. Growth of Georgia’s exports and remittances is projected to remain subdued in the short- to medium-term, as economic prospects of key trade partners remain gloomy. In addition, the widening fiscal deficit will add to the external vulnerabilities. The external imbalance is likely to persist, with the current account deficit at over 10 percent of GDP over the medium-term, which could result in further weakening in the currency. C. FISCAL DEVELOPMENTS 1.18. Fiscal management was largely prudent in Georgia in the aftermath of the 2008/09 global crisis, but has become loose in the past three years. From 2009 to 2013, the government was committed to rebuilding fiscal buffers after a large stimulus deployed during the financial crisis left the country largely indebted.23 During this period, general government expenditures were successfully consolidated from 38 percent of GDP in 2009 to 30 percent in 2013. Meanwhile, as revenues stayed buoyant at 28-29 percent of GDP aided by a strong tax collection, fiscal deficits narrowed from 9.2 percent of GDP to 2.6 percent (Figure 1.8). However, from 2013 the government re-oriented their spending to address social needs, which resulted in an increase of social spending from 7 percent of GDP in 2012 to 10 percent in 2016, as well as a widening fiscal deficit (Table 1.3). Fiscal Revenue 1.19. On the revenue side, Georgia’s current tax system is characterized by a reliance on flat- rate taxes. Georgia’s tax system is comprised by six flat taxes: the three major taxes, which include the Corporate Income Tax (CIT), the Personal Income Tax (PIT) and the Value Added Tax (VAT) are levied at flat rates of 15 percent24, 20 percent and 18 percent, respectively. Excise taxes are applied to a handful of goods, including tobacco, alcohol, fuel and cars. Imported goods are subject to custom taxes of 0 percent, 5 percent and 12 percent depending on categories. Finally, there is a property tax, which is a local tax with rates set by each local government to up to one percent. In terms of tax administration, electronic filing is currently the norm in Georgia. Figure 1.8. Fiscal Balance in Georgia (% of GDP) 15 10 5 0 -5 -10 Real GDP growth Cyclically adjusted �iscal balance Fiscal balance Source: Estimates based on Geostat statistics. 23 This stimulus package was mainly financed by donors and bilateral partners. 24 Under CIT, there are also a dividend tax of 5 percent and an interest income tax of 5 percent. 11 Figure 1.9. Gross Public Debt Composition Figure 1.10. General Government Deficit Financing (% of GDP) by Source (% of total) 50 150 40 100 30 50 20 0 10 -50 0 2009 2010 2011 2012 2013 2014 2015 2016 b 2009 2010 2011 2012 2013 2014 2015 2016 b External: public Domestic: T-bills Domestic: other Historical privatization Change in deposits External Domestic Source: MOF. Source: MOF. Note: negative change in deposits means accumulation of government savings. 1.20. Georgia’s current tax system is a result of a series of reforms implemented about a decade ago. Many countries in the ECA region24 adopted flat taxes in the hope of boosting economic growth, in particular after the successful experiences of the Baltic countries. Tax reforms that brought in flat taxes to Georgia aimed at the simplification of the tax regime, and strengthening of tax administration. These were carried out between 2004 and 2007, and the outcome has generally been considered successful: through its tax reforms, Georgia increased its tax revenues from 14.5 percent in 2003 to 25.8 percent of GDP in 2007; revenues stabilized afterwards, with the tax share at 25.3 percent of GDP in 2015. Georgia ranks high in VAT, PIT and CIT productivities compared to others in the region (PER 2014). 1.21. To preempt any reversal of these tax reforms, the Liberty Act was enacted in 2014 to ban the introduction of new state taxes or rate increases without a referendum for most taxes. This Act excludes the property tax, which has limited potential for revenue increases since property tax is paid only by the formal employees with monthly income above GEL 40,000. Excise taxes are also an exception from this rule, and have been raised frequently, especially for tobacco and alcohol. The most recent increase, in January 2016, is projected to generate a 16 percent increase in excise tax revenues (about 0.4 percentage points of GDP). 1.22. VAT, CIT and PIT represent the bulk of fiscal revenue—and constituted nearly 80 percent of total tax collections in 2015. VAT, being the largest source of revenue, accounted for 40 percent of total revenues. PIT contributed 25 percent collections and CIT 12 percent of total revenue, while another 10 percent was collected under excises. Collections from CIT, PIT, VAT, and excises are 3.23, 7.01, 11.07, and 2.74 percent of GDP respectively. 25 Belarus, Bosnia and Herzegovina, Bulgaria, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Macedonia, Romania, Russia, Serbia and Turkmenistan. 12 1.23. The transition to the dividend tax will have major implications on the size and composition of fiscal revenues. In 2015, the government legislated to replace the CIT by a dividend tax starting in 2017, anticipating higher investment from the reinvestment of profits. This change was modeled after Estonia’s CIT reform of 2000 (Box 1.2). Under this change, the CIT rate of 15 percent is only applied to profits when they are distributed as dividends, and the profits not distributed are no longer taxed. A USAID study suggests that this could boost long-run investment, potentially adding up to 1.4 percentage points to growth. However, in the short run, it is estimated that the government’s new initiative to change the CIT tax into a dividend tax will reduce public revenues by about 1.5 percentage points of GDP, about half of the existing CIT revenues, in the first full year of enactment (likely in 2017). 1.24. The government expects the raise of the excises on tobacco, fuel and vehicles to fill the revenue hole left by the transition to dividend tax. The 2017 budget envisioned large increases of excises—55 percent for filtered and 100 percent for non-filtered cigarettes. Excises on fuel and gasoline for cars were also raised considerably—gasoline by 100 percent, diesel by 167 percent, natural gas by 150 percent, and lubricants by 100 percent. The magnitude of increases in excises on cars varies across cars of different age: notably new cars and cars purchased within one year are charged by 53 percent more, while those within two years old are charged by 27 percent more; cars over 7 years face significantly higher excises—by 60 percent more for cars within 7-10 years of age and by over 100 percent for those over 10 years old. These measures are expected to generate about GEL 550 million in revenues which is about 1.5 percent of GDP for 2017. However, higher fuel excises could raise the cost of public transportation; and have implications on the budget since the service is subsidized by the government. Other than the tax revenue measures, the government also expects to generate an additional income of GEL 150 million in 2017 via non-tax revenues. 1.25. The proposed tax policy change is likely to affect the poor more. The government’s recent policies to raise indirect taxation (excises) and cut the direct tax burden (profit tax) may have implications from the equity point of view. The PER (2015) showed that there is a stark contrast between Georgia’s direct and indirect taxes in distributional effect: direct taxes concentrate more in the top deciles, indirect taxes are more evenly distributed. In other words, the cut in taxes companies need to pay would benefit the riches in the country; while the raise of excise rates on tobacco, fuel and vehicles would affect the less well-offs of the population more; for instance, higher fuel excises would translate into higher transportation costs and prices for necessities, leading to higher living cost which will weighs more heavily on the poor. 13 Box 1.2. Corporate Income Tax Reform: Estonia’s Experience Estonia introduced its corporate income tax reform in 2000, under which profit is no longer taxed unless it is distributed as a dividend. This reform was billed as a means to attract investment, promote business and increase employment. Many attributed this tax reform to the robust growth performance in the 2000s—which catapulted Estonia to the high income status. Many studies acknowledge that the reform had a positive impact on investment and productivity. Some argue that this strengthened the competitiveness of Estonia’s tax system, and set the country on the path to eventually become the most competitive among OECD members,26 in attracting a large influx of foreign direct investment. Between 2000 and 2004, investment growth in Estonia was 39 percentage points faster than in neighboring Latvia and Lithuania. Micro-evidence show increases in capital stock, the investment rate of return, and total factor productivity over the medium term. The undistributed profits were, in many cases, retained as savings— which helped improve the liquidity situation for firms, especially for SMEs, and also allowed firms to fare better during the financial crisis. The results are, however, less motivating in terms of revenue collection. During the first three years of the reform’s implementation, a collapse of CIT revenues (from 2 percent to 1 percent of GDP) was observed despite robust growth. And the dividend tax payments did not catch up for the next few year but total tax revenues as percent of GDP recovered to the pre-2000 level in four years. Today, Georgia bears some striking similarities with Estonia before its tax reform. Both economies were middle income countries, with growth hit by external shocks—the recession of Russia’s economy, and sought for growth-promoting policies in difficult times. Both countries had tax reforms to simplify their tax systems. However, there are a few notable differences in the tax system of the two countries. First, prior to the reform, Estonia’s collection of total tax revenues were much higher compared to Georgia’s current revenues: Estonia’s total tax revenue was about 34 percent in 1995-1999, while this ratio was at about 25 percent in Georgia in 2012-2016. Second, CIT plays a more important role for Georgia’s revenues: CIT collection accounted for 13.3 percent of total tax revenues in 2015, more than doubling that of Estonia in 1999. These differences suggest that the proposed reform would have a greater revenue impact hence higher fiscal cost for Georgia. Figure 1.11. Pre-reform Tax Collections in Estonia and Georgia 35 25 Total Tax Revenue, % GDP 30 CIT, % Tax Revenue 20 25 20 15 15 10 10 5 5 0 0 EST: 1996 GEO: 2012 EST: 1997 GEO: 2013 EST: 1998 GEO: 2014 EST: 1999 GEO: 2015 EST: 2000 GEO: 2016 26 Estonia was selected for having the most competitive tax system among 34 OECD countries in 2015 by Tax Foundation. 14 Figure 1.12. Structure of Tax Collections Figure 1.13. Tax Revenue for Georgia and Other (million GEL) Countries by Income Groups (median, % of GDP) 9000 26 8000 24 7000 22 6000 20 5000 18 4000 16 3000 14 2000 12 1000 0 10 2013 2014 2015 -1000 CIT PIT VAT domestic Low income Lower middle income VAT imports VAT refund Excise domestic upper middle income high income Georgia Excise imports Other domestic other customs Source: MOF. Source: IMF Revenue Longitudinal Dataset (WoRLD), June 2016. Public Expenditure 1.26. In term of government expenditure, Georgia is considered moderate relative to its regional peers, with a large allocation to wages and social transfers. Between 2006 and 2016, Georgia’s general government expenditure-to-GDP ratio averaged around 30 percent, below the ECA average of 38.2 percent of GDP (Figure 1.14). Over the period 2006-2016, its current expenditure has varied in 70- 80 percent of total expenditures. Around 60 percent of expenditures are allocated to wages and social transfers. Pensions account for about 16 percent of total expenditures where a rising trend is observed. 1.27. The Economic Liberty Act also lays out a set of fiscal rules in view of keeping a small and disciplined government. This Act, adopted in September 2009 and entered into force in 2014, introduced: (i) upper limits for the approved consolidated budget expenditures, at 30 percent of GDP, (ii) limits to the budget deficit at 3 percent of GDP, and (iii) ceilings to public debt, set at 60 percent of GDP. Under normal circumstances, which is not in situations of emergency, martial law, or economic recession, if the rules are violated, the government shall submit an additional two-year budget plan to return to the set limits. These rules and parameters are generally consistent with standard fiscal rules worldwide. 1.28. However, the fiscal rules are perceived to have lacked the ex-post monitoring and disciplining features to become truly effective. The Liberty Act (in particular, the expenditure cap) applies to the budget law approved by the Parliament. However, the recent budget executions have risked its credibility, creating the impression that the Law has no “teeth” to push on the implementation of remedial actions when the actual executed budget does not comply with the budget plan. In 2015, 15 total expenditures already went beyond their legal limit of 30 percent of GDP, but since the original approved budget law was lower than 30 percent of GDP, the fiscal rule did not trigger any additional remedial action, even if the actual budget execution suffered overruns. The approved 2016 budget law was also within the parameters established under the Liberty Act, while the actual spending has again exceeded the spending ceiling of 30 percent of GDP. Fiscal rules in Georgia require deeper analysis going forward. 1.29. Current expenditures remain high given the government’s commitments to social programs. Social sectors constitute, by far, the biggest spending item—accounting for more than half of total current expenditures (Figure 1.15). Social spending rose by 2.6 percent points of GDP in 2015 compared to a year ago, and in 2016 another increase of 1.6 percent points is observed. The UHC program contributed to the largest increase of 0.7 percent of GDP in 2015, and a recently announced 12.5 percent increase in pension benefits contributed to 0.5 percent of GDP in 2016. In addition, rising spending from targeted social assistance, the universal healthcare program (UHC), education, culture and religious affairs activities represented additional pressures. Figure 1.14. General Government Expenditure Figure 1.15. Composition of General Government (% of GDP) Spending by Economic Classification (Percent of GDP) 12.0% 31.8% 32.0% 31.7% 10.0% 8.0% 31.0% 6.0% 30.2% 30.5% 4.0% 29.5% 2.0% 0.0% 28.5% 2011 2012 2013 2014 2015 total wages, goods and Services subsidies and grants Social bene�its other expenses Capital and net lending Source: IMF WEO, Ministry of Finance and World Bank staff Source: MOF. calculations. 16 Table 1.4. Deep Dive into State Spending and the Key Drivers of Expansion (In GEL million and percent of GDP)   2014 2015 2016 Percent Percent Million Million Million   of GDP of GDP GEL GEL GEL Increase Increase Total expenditure, State Budget 8,377 9211 2.6% 10,260 3.0%  Changes due to the programs initiated and scaled up since 2014:  1.3%   1.7% Universal healthcare program (UHC) 338 573 0.7% 681 0.3% Agro projects (MOA total) 265 262 0.0% 321 0.2% Pensions 1335 1399 0.2% 1570 0.5% Social programs other than pensions (incl. assistance) 617 615 0.0% 680 0.2% General education vouchers 374 430 0.2% 507 0.2% Enterprise development agency (MOED) 2 23 0.1% 22 0.0% Social and healthcare benefits (Ministry of Defense) 7 8 0.0% 41 0.0% Social benefits to high mountainous regions 0 0 0.0% 20 0.1% Source: Ministry of Finance. Note: 2016 budget is estimate based on new spending items that are not reflected in the budget law yet. 1.30. Pension spending has been rising significantly, reaching 4.7 percent of GDP in 2016. The existing pension, which is universal and non-contributory, aims at protecting the elderly against poverty at the minimum subsistence level, but does not provide adequate income replacement to the employed at their retirement, manifested by a low replacement rate of 19 percent by 2016. As the largest spending item in the budget, pension costs continue to rise with ad hoc increases in benefits—including increase from GEL 160 to GEL 180 per month, per person, starting from July 2016 and to take full year impact in 2017. The pressure is expected to ease in the long run as the government adds a savings pillar to the currently non-contributory pension system and introduces regular indexation to the basic pension. The indexation of the basic pension, if it is linked to inflation, will reduce the pressure for large, sporadic increases in benefit levels27, given that the raises in pension benefits over the past few years have outpaced inflation. The government plans to provide a matching contribution to the individual’s contribution under the savings pillar to facilitate the transition towards a contributory system, which will incur some fiscal cost in the beginning; however, if managed properly, the new system could mitigate the pressure for ad hoc increases in basic pension in the long run. Indeed, if the government were to choose an exceedingly generous indexation and matching scheme, they may need to consider other parametric changes in the future to make it financially sustainable, such as an increase in retirement ages. 27 The Bank team is supporting this reform via development policy operations (DPO). 17 Figure 1.16. General Budget Decomposition Figure 1.17. Rising Social Spending and its by Functions Components (% of GDP) (% of GDP) 16.0% 30.7% 30.6% 30.5% 30.0 24.4 25.4 24.9 23.1 23.3 30.3% 30.5% 20.0 10.0 6.0% 29.5% 29.5% 0.0 2011 2012 2013 2014 2015 -4.0% 28.5% total Public Administration Defence and public order Economic affairs and environment Utilities Social Affairs Source: Ministry of Finance. Source: Ministry of Finance. 1.31. The UHC program accounts for about one fourth of all social transfers, and spending pressures from this program have built up. UHC spending has been steadily increasing since its introduction in 2013, and so far it has constituted approximately 75 percent of total health spending in 2016. Budget overruns in the UHC in 2016 added to the concern that health expenditures are not managed efficiently. Nevertheless, total public health spending in Georgia is 2.9 percent of GDP according to the 2016 budget outcomes —low compared to upper-middle-income countries, which spend an average of 4.3 percent of GDP on public health (see Chapter 2 for a detailed discussion of the health sector, and proposals to further strengthen the effectiveness of the UHC while promoting its fiscal sustainability). 1.32. Georgia’s public spending on education has also been growing recently, primarily driven by increases in teacher salaries. Public spending on education has risen over the past four years by 1 percentage point of GDP—to a budgeted 3.8 percent in 2016—out of which some 3.1 percent of GDP relates to the central government, and the rest is managed by local administrations for preschool education. This expenditure level is somewhat lower than education spending in other countries in the region (PER 2014), and while educational outcomes—measured by the Trends in International Mathematics and Science Study (TIMSS) (2009)—are weak, and teacher quality is perceived to be one of the pressing issues in Georgia’s education system. The amount of the transfers to primary and secondary schools for expenditures is determined by a capitation-based formula. The central government is responsible for capital investment in the primary and secondary schools, while maintenance and teacher salaries are covered by the school grants. Tertiary education financing (10 percent of the total education expenditure), has been identified by the government as a priority area requiring systematic approach to strengthen linkages with the labor market demand and quality of service. 1.33. The Law on High Mountainous Regions (HMR), adopted in July 2015, will add to social spending in 2017 onwards. The objective of the HMR law is to “ensure equal social and economic development of Georgia’s regions in order to solve the social economic problems of the residents of high mountainous regions”. The law seeks to achieve this objective by providing supplementary benefits to teachers, pensioners, and socially vulnerable households in these regions. The estimated annual cost is 0.1 percent of GDP. 18 1.34. The government views capital investment as another key policy priority. Given the large infrastructure gap in Georgia (PER 2012), capital expenditures have averaged 24 percent of total expenditures (about 8 percent of GDP) since 2006. A sustained level of public investment was a key growth engine in earlier years and stimulated the economy during the crisis periods in 2008- 2009. During 2013-2014, capital expenditures dropped to below 6 percent of GDP, as the government reoriented resources towards social spending. However, in 2015 and 2016 there was a marked return to the public investment policy—consolidated capital expenditures accounted for 7 and 6.6 percent of GDP in respective years, out of which in 2015 the central government spent 3.6 percent of GDP, the local governments spent 2.1 percent, and the rest went to lending to SOEs. Various donors covered 40 percent of these capital expenditures (2.8 percent of GDP), while the rest was financed by domestic sources. The government plans to expand investment in capital from 2017 onwards. 1.35. The 2017 budget envisions consolidation on the administrative part of current spending. The allocation for the wages of public employees is reduced by 5 percent, leading to savings of 0.6 percentage point of GDP. Spending on goods and services will be kept the same as in 2016, which is expected to generate a saving of 0.4 percentage point of GDP. 1.36. In terms of ministerial budget allocation, three ministries absorbed 50 percent of the central government’s spending. The Ministry of Labor, Health and Social Affairs (MOLHSA) accounted for 31 percent of state budget expenditure in 2015, the Ministry of Education and Science spent another 9 percent, while the Ministry of Regional Development and Infrastructure (MRDI) consumed 10 percent in 2015 out of the budget of GEL 9.3 billion. Transfers to local administrations amounted to GEL 1.2 billion (or 13.5 percent of total budget). The remaining 58 state budget spending entities—including 10 line ministries—accounted for an additional 36 percent. This structure has been broadly maintained for the period of 2014-2016. 1.37. As for fiscal deficit financing, the major source relates to external loans from donors, although domestic financing has been playing an increasing role in recent years. External public debt averaged 30 percent of GDP since 2009 and remained the largest source of financing for Georgia’s fiscal deficit, the majority of which was financed by development partners including the World Bank, IMF, ADB, EU, EIB, GIZ, and EBRD, and bilateral partners. There is only one commercial bond, a Eurobond of US$500 million, equivalent to nearly 5 percent of total external public debt. However, in 2015 the government increased financing from domestic sources by tapping into more T-bill issuances, and in 2016 the government drew down reserves and scaled up asset sales.28 KEY CHALLENGES D.  1.38. The deterioration in the fiscal position, coupled with weak growth, heightened external vulnerabilities, spending pressures and an anticipated revenue reduction in sight, highlight the significant fiscal risk Georgia is facing. Putting growth back on a fiscal prudent footing would call for measures to boost revenues and/or curtail (and better prioritizing) spending. The major fiscal challenges Georgia is facing is as follows: 28 Sale of SOEs, buildings, land and other state properties. 19 (i) The scope for raising revenues by tax policy is limited by law and the efficiency gains expected from further streaming the tax system are positive, but limited. (ii) The exercise of reigning in government spending must be accompanied by a careful balancing act—given the need to protect key, high-quality social spending while keeping the composition of spending pro-growth. (iii) As a key driver of fiscal pressures—and a priority area for poverty reduction and increased welfare in Georgia—the health sector must be subject to such a delicate “balancing act”. This will require implementing proposals to further strengthen the effectiveness of the UHC, while promoting its fiscal sustainability (Chapter 2). Revenue Mobilization 1.39. Options to raise revenues in Georgia are limited to very few activities by law. As noted earlier, the Liberty Act was enacted in 2014 to ban the introduction of new state taxes or rate increases without a referendum for most taxes (not including excise tax), with the intention to preserve the gains in the design of the current tax system design. Since the enactment of the Act, the government has been raising excises, but there is a limit to generate revenues this way. With limited scope to raise other taxes, closing tax exemptions presents a viable option permitted under the legislation. Box 1.3. Tax Expenditures and Best Practices  What are tax expenditures? Tax expenditures are provisions in the tax code that generate preferential treatment of certain taxpayers. They usually favor an industry, activity or class of persons with social or other policy objectives.  Modalities. Tax expenditures can take a number of forms including tax exemptions, allowances, preferential tax rates, deductions, credits, and other special tax treatments. These measures may be viewed as alternatives to other policy instruments, such as spending or regulatory programs. While some tax expenditures are justified, inevitable, or derived from the difficulties of administering taxes, they can also imply foregone revenue and potentially unfavorable economic and social impacts.  Adverse Effects. The unfavorable impacts of tax expenditures can be related to their potential ineffectiveness, inefficiency and inequality. Some tax expenditures are offset by underlying economic forces and other tax provisions. Many tax expenditure schemes are a response to various interest groups—a context that leads to loss of efficiency in resource allocation by favoring particular sectors. Tax expenditures also tend to be less equalizing, for instance, most non-refundable tax expenditures would exclude non-taxpayers who are among the poorest group from receiving benefits. Other pernicious effects include difficulties in estimating tax revenue, added complexity into tax laws, increasing costs of law enforcement, rent seeking, and elusiveness of the government size. In addition, the costs of tax expenditures are also not evident in the budget.  Measuring the cost. Countries apply various methodologies to evaluate the economy-wide benefits and costs from tax expenditures including revenue impact and the effects of tax measures on resource allocation and income distribution. They also estimate the impact of removing a few key tax exemptions, both on revenue and on the economy as a whole.  Mitigating measures. To address these potential impacts to make informed decisions going forward, governments evaluate the economic and social impact of tax expenditures with a focus on relevance, effectiveness and efficiency. To support this effort, countries implement systems and reporting requirements to create transparency with regard to the revenue, inventory of tax expenditures, their nature, legal basis, past and likely impacts of exemptions. Even in absence of immediate action on reducing or eliminating tax expenditures, it is important to present tax expenditures and their fiscal costs in the budget to enhance transparency, public awareness, and allow for a robust policy discussion. 20 1.40. Tax expenditures identified in Georgia for income taxes arises from a list of exemptions covering profit tax, long-term capital gains, foreign income, and others. Income tax in Georgia, as described earlier, consist of taxes on personal income and on corporate income (profit). Article 99 under the Tax Code provides exemption from the standard 15 percent profit tax, to activities primarily in the sectors of agriculture, secondary education, health, and tourism, in addition to religious organizations, charities, international organization and government. Another tax expenditure relates to the assets held for more than two years—regarded in Georgia to be long-term capital, and gains on long-term capital are exempt from income tax. Income earned abroad by residents of Georgia is not subject to the income tax.29 In addition, there is a list of exemptions and deductions under Article 82 of the tax code.30 The cost of these three tax expenditures in terms of foregone tax revenues in 2014 and 2015 are reported in Table 1.4. A notable observation is that despite the large number of activities involved, revenues foregone under Article 82 constitutes the smallest amount out of the three. Their total estimated cost is GEL 72.6 million in 2015 or approximately 0.2 percent of GDP. Another source of corporate income tax expenditures is the Free Industrial Zones (FIZs) which are designed as tax free export-processing zones31. Due to lack of information the estimates cannot be obtained, but existing studies suggest such tax expenditures are limited.32 1.41. Value-added tax expenditures exist for agricultural goods, financial services, education and healthcare spending. For most business activities Georgia’s VAT applies a standard rate of 18 percent. Expenditures on education, healthcare, and financial services, and certain agricultural goods, are subject to zero rates, meaning that these sectors are exempt from output tax and receive a credit for any input tax paid.33 The estimates for VAT tax expenditures amount to GEL 695 million in 2015, larger than those from income taxes. 29 Although our methodology views the exemption in this case as a tax expenditure, it is worth pointing out that all OECD countries except for the United States exempt income earned abroad by residents. Long-term capital gains are exempt from income tax. 30 There are approximately 50 exemptions from income tax enumerated in Article 82, such as income received as alimony; awards received by sportsmen and their coaches at Olympic games for winning and/or getting medal place at world and/or European championships; lottery winnings less than GEL 1,000; capital gains from the sale of an apartment (house) held for more than two years and from the sale of a motor vehicle under ownership for more than six months; income from specified gifts of inheritance (not to exceed certain amounts); and so on. 31 Georgia has two FIZs, one in Kutaisi and the other in Poti. 32 A study on FIZ (IFC, 2012) shows that the operations in these FIZs are quite limited. 33 We estimate that 50 percent of expenditures on agriculture, fishing, and forestry are exempt from VAT. 21 Table 1.5. Tax Expenditure Estimates by Tax Type in 2015 Tax expenditures (million GEL, unless otherwise specified) Income tax 1. Article 82 exemptions 0.5 2. Article 99 exemptions 255.4 2. Foreign source income 9.1 3. Exemption for realized long-term capital gains 63.0 Sub-total 328.0 VAT: 5. Exemption for food 226.6 6. Exemption for financial services 36.2 7. Exemption for education 241.4 8. Exemption for healthcare 190.4 Sub-total 694.6 Grand-total 1022.6 As percent of GDP (%) 3.22 Note: Bank staff estimates 1.42. The scope of further revenue mobilization through eliminating tax expenditures is positive, but limited. Bank staff estimates that tax expenditures account for 3.2 percent of GDP in 2015, which is close to other estimates, for instance, the estimates of 3.1 percent of GDP in 2013 (IFC&ECOPA, 2014). This is very much in the mid-range compared to OECD countries (Figure 1.6), consistent with the fact that the Georgia tax code is perceived to be broadly efficient with limited deductions, exemptions, credits, and special treatments, and the application of flat tax rates with a very narrow zero-rate bracket34,35. Among the identified tax expenditures, a number of them are not uncommon: Georgia’s VAT treatment of expenditures on education, health, and even for food, is also the typical practice in many countries, primary out of social equity objectives. The exemption made to financial sector activities are 34 According to Article 81(1) of Georgia’s tax code, a natural person, whose taxable employment income during the calendar year does not exceed GEL 6,000, shall be entitled to deduct from this income personal allowance – GEL 1,800. This is a very modest and limited income tax exemption. In fact, the administrative costs to the Government of Georgia of collecting personal income tax on such small amounts would likely exceed the revenue yield from the tax. Although technically speaking, such income should be included in the comprehensive income tax base. Since we lack data on the number of individuals would qualify for this exemption and the foregone tax revenue from the exemption is likely to be so small, we do not attempt to estimate this tax expenditure. 35 This, however, does not rule out the possibilities of downward bias in estimated tax expenditures which may arise due to a number of factors, including the conservative methodology employed and the underreporting of income exempt from tax. 22 often available in the tax codes of many EU countries, given the difficulty in measuring the value-added in the sector. Most countries do not tax foreign income. Measures to tax the profits of all companies in the agriculture, education, health and tourism industries is likely to yield an income less than 0.8 percent of GDP. Taxing realized long-term capital gains and eliminating VAT treatment for selected food items can yield at most 0.2 percent of GDP. Figure 1.18. Tax Expenditures in Select OECD Countries and Georgia36 (% of GDP) 9 8 7 6 5 4 3 2 1 0 Source: OECD (2010) and Bank staff calculations. Managing Public Spending 1.43. With a major reduction in government revenues from 2017 onward, the planned fiscal expansion is likely to protract the current high level of deficits. The envisioned fiscal expansion is expected to keep total government spending over 32 percent of GDP over the medium term37. Bank staff projections show that with measures enforced from 2017 fiscal deficits could still remain over 4 percent over the medium term unless the scale-up program for public investments is reconsidered or additional consolidation measures are in place. 1.44. Georgia is under intense pressure on social spending. Social spending as percent of GDP has grown from about 7 percent in 2011-2012 to nearly 10 percent in 2015; and with the government’s commitment, it will be sustained at a level close to 10 percent of GDP over the medium term. The UHC and the pension are major drivers behind the increase. In particular, the overruns of the UHC budget in 2015-2016 highlighted the rapidly growing demand for health services and the inefficiency of the healthcare system (see Chapter 2 for a more detailed discussion). Pension, the largest social program in Georgia, has experienced multiple benefit increases in an ad-hoc manner over 2015-2016. In addition, 36 Based on OECD (2010), Tax Expenditures in OECD Countries, Paris: OECD Publishing and United States, National Commission on Fiscal Responsibility and Reform (except Georgia, which is based on our calculations). 37 Meanwhile quasi-fiscal spending including spending on SOEs remains an upside risk. 23 the increase in teachers’ salaries and the new program to provide additional benefits to the population in the high mountainous region also added pressures on the spending side. 1.45. The recent surge in social spending also presents a key challenge in maintaining a pro- growth budget composition. The increase in social spending from 2012 was accompanied by a notable reallocation away from the focus on productive spending. However, as infrastructure need in the country remains substantial, public investment rebounded in 2015-2016, which is likely to be followed by an ambitious infrastructure investment plan in 2017. Given limited upside on revenues and rising social spending, an expansion in public investment will put fiscal sustainability at risk. 1.46. In addition, government investment projects in Georgia are prone to efficiency losses due to the lack of a sound public investment management (PIM) system. The PER (2014) identified the lack of a functional system for projection screening, appraisal, review, selection and implementation as major obstacles. E. THE WAY FORWARD 1.47. The government can continue to create fiscal space from the revenue side, but the gains could be modest. Since the Georgian tax system is broadly efficient, staff estimates show that total tax expenditures (mainly attributable to VAT zero rates and profit tax exemptions) would amount to 3.2 percent of GDP in 2015—which is very much compared to Georgia’s peers. Since exemptions on agriculture, education, healthcare and financial services are the usually candidates for tax exemption worldwide, and would be politically difficult to remove, the government could eliminate profit tax exemptions on some sectors, the exemptions on long-term capital, and VAT on some food items which will likely yield a revenue of less than one percent of GDP. 1.48. In this light, the government needs to focus its efforts on streamlining and consolidating and increasing the effectiveness of public expenditures going forward. As the government has prioritized social spending and public investment, bringing back government outlays into a prudent path will require the government to consider taking measures on the other expenditures, such as containing the purchases of goods and services and public employee’s wages and salaries38. The government also needs to refrain from promising new benefit programs. To manage social spending more efficiently, it is advisable for the government to avoid any more ad hoc increases in pension benefits while deepening reforms on the pension system to ensure its long-term sustainability, including properly indexation of the basic pension and transition towards a contributory pension system in the long run39. The health sector has also been a major cost driver; recommendations to ensure that it is effectively managed so that it achieves fiscal sustainability while delivering on its promise are presented in Chapter 2. 38 A cut on current expenditures is envisioned in the approved 2017 budget. As examined in PER (2014), public sector compensation policy in Georgia has been inadequate; and developing rule-based compensation policies are necessary to contain increases in government spending on wages and salaries. Facing the challenge of fiscal sustainability, it is recommended that the government puts on hold increasing wages and salaries until a sound public sector employee remuneration law is in place. 39 This draws from analyses and recommendations from PER (2014). 24 1.49. To facilitate fiscal consolidation and ensure credibility in fiscal policy making, it is advisable for the government to make spending decisions within the parameters formulated by the Liberty Act. The executed budget in 2015 was higher than the 30-percent-of-GDP cap, and this is likely to happen again in 2016. To signal the commitment to prudent fiscal management and avoid the perception of poor management of budget, the government needs to maintain budget execution within the spending limits under the rules laid out in the Liberty Act over the medium term. If rising spending pressures of 2017 push spending beyond the limits, the government should prepare a consolidation plan to return within the limits over the following two years (as an action triggered by the escape clause of the Liberty Act), which will send a clear signal of commitment to fiscal sustainability to the general population and the donors. 1.50. A reasonable level of productive spending needs to be sustained, in particular for capital investment. Infrastructure needs remain high for Georgia, including after signing the Associated Agreements including the agreement on Deep and Comprehensive Free Trade Area (DCFTA). Due to increasingly tight fiscal space consumed by rising social spending, it is advisable for the government to keep capital expenditures at 5-6 percent of GDP. 1.51. The government also needs to continue strengthen the effectiveness of public investment. Investment spending has been an area where gaps in effectiveness has been identified; to maximize the use of public investment spending, the establishment of a sound public investment management (PIM) system remains crucial, especially at project identification and appraisal stage (PER 2014). 25 Chapter 2. Strengthening Health Sector Spending A. OVERVIEW 2.1. Georgia has made significant progress in improving access to health services under the UHC Program. The introduction of the UHC Program in February 2013, aimed at improving the general population’s access to good quality health care, has benefited more Georgians, particularly those relatively less well-off, by improving access to health services when ill and reducing the likelihood of impoverishment or catastrophic out-of-pocket (OOP) spending on health care. 2.2. The recent overruns in health spending have, however, highlighted a key challenge that the government faces in maintaining the sustainability of the UHC Program. Since the implementation of the UHC Program, health spending has risen sharply (from 4.0 percent to 8.4 percent of total government spending between 2012 and 2015). Notwithstanding these increases, it is still almost the lowest share of government spending among European states and below the European Region middle- income country average of 10.5 percent in 2014 (WHO 2016). 2.3. The UHC Program has consistently overspent its budgeted amount in recent years, which raises concerns about the sustainability of the program. In addition, considerable challenges remain in the health sector in Georgia that could undermine efforts to sustain and improve on the progress achieved to date. Primary care utilization rates are increasing but remain low, exacerbated by a shallow outpatient drug benefit under the UHC Program. Both patients and providers face strong incentives to use costly hospital and emergency care services. All of these contribute to an inefficient service delivery system that is poorly suited to addressing the health care needs of an aging population with a large burden of chronic non-communicable diseases (NCDs). 2.4. Going forward, the government needs to ensure the financial sustainability of the UHC Program in order to maintain the progress made to date, and further deepen coverage and financial protection. To achieve these higher level health system goals, it is critical that efforts are made to increase efficiency and obtain better value for money from current spending. In the short term, reducing duplicate coverage among those who have both Private Health Insurance (PHI) and UHC coverage, as well as further strengthening the Social Services Agency’s (SSA) management and purchasing capacity would generate efficiency savings that can be used more productively within the sector. Using these savings to improve access to lower-cost, generic medicines at the outpatient level, would help reduce the reliance on expensive hospital and emergency services. Reorienting service delivery away from hospital services and strengthening primary health care will certainly improve outcomes and efficiency in the long term, but this reorientation of service delivery would require additional investments in the health sector that cannot be achieved through efficiency savings alone. 2.5. This chapter reviews health expenditure patterns in Georgia through the lens of the major sector objectives, followed by a discussion of key policy issues. Section B provides the policy and institutional context of the recent reforms. Section C provides an overview of health spending in Georgia, particularly in the context of the UHC Program, and examines the costs and sustainability of the system. Section D examines the performance of public health expenditures relative to long-term trends in health outcomes, and other health system objectives such as equity, financial protection and efficiency in order to contextualize the outcomes of the ongoing health reform. Although the health impacts of the reform will only become apparent in the medium term, a number of preliminary outcomes are already beginning to emerge, specifically with respect to the financing scheme and greater equity in the health 26 system. Section E presents the key challenges facing the health sector. Section F addresses key policy priorities and a set of both short-term and medium-term recommendations. B. POLICY AND INSTITUTIONAL CONTEXT 2.6. Georgia’s Social-Economic Development Strategy (“Georgia 2020”) identifies the need to improve the general population’s access to good quality health care as a key policy priority aimed at strengthening human capital. 2.7. Critical reform areas identified include: a) Improving public health care spending by increasing the public share of health financing, as well as increasing efficiency, equity, and inclusiveness of public health spending. b) Improving the quality of care through better regulation, professional development, introducing clinical guidelines, and strengthening mechanisms for patient protection. c) Increasing the affordability of pharmaceuticals through the use of evidence-based clinical practice, improved regulations, and inclusion of medicines for priority diseases into the publicly-funded package, and centralized procurement. d) Strengthening primary care for early diagnosis of both communicable diseases and NCDs, and better outpatient management. 2.8. In February 2013, the Government of Georgia launched the UHC Program. Universal health coverage is, by definition, about addressing many of the goals referred to above. The UHC Program marked a significant shift in how health care is financed and health services are purchased in Georgia, as well as the culmination of nearly two decades of health system reforms in Georgia.40 The UHC Program extended publicly financed entitlement to health care coverage to the entire population.41 The nature of the program is noncontributory, in the sense that Georgians do not have to contribute for enrollment. Enrollment involves registering with the primary care provider of choice. The benefits package covers a range of primary and secondary care services and limited essential drugs (Table 2.1). Administratively, the reform transferred responsibility for purchasing health care services from private insurance companies to the SSA under MoLHSA, thus putting in place a platform to shift from passive to active purchasing. 2.9. The health financing reforms introduced since 2013, and backed up by significant increases in public health spending, have moved Georgia closer to European norms. These include: (i) near universal population entitlement to publicly financed health care; (ii) free visits to family doctors; (iii) referral and prescribing systems; (iv) a single purchasing agency; and (v) higher public spending on health (WHO 2016). Sustaining the coverage achieved to date and deepening coverage through better financial protection against OOP costs are the policy priorities for the Government of Georgia. 40 Beginning in 1995, Georgia shifted from social health insurance (SHI) (1995-2004) and competitive private health insurance (PHI) (2004-12) to general revenue financing for health with complementary PHI (2013 onwards). The health financing reforms were accompanied by service delivery reforms such as the rationalization of hospitals and the introduction of family doctors and gatekeeping. Other reforms were aimed at increasing pharmaceutical market competitiveness by reducing regulatory and capital requirements for market entry (2009) and introducing prescription requirements (2014). Efforts to strengthen primary care, improve quality of service delivery, strengthen provider payment and purchasing arrangements are ongoing. 41 Those who had private insurance on July 1, 2013 are not eligible for UHC Program coverage, but can access a minimum package of services if they can demonstrate to the SSA that they no longer have Voluntary Health Insurance (VHI)(about 1 percent of the population). In practice, however, eligibility is hard to verify as private insurers are not required to report their beneficiaries to MoLHSA or the SSA. 27 Table. 2.1: Summary of UHC Benefits and User Charges (2015) Type of Benefit User Charges Former Medical Former Beneficiaries of the Veterans All Others (Previously Insurance Program Program for Pensioners etc. Uninsured) (MIP) Beneficiaries Planned outpatient Free Free Free Free care Outpatient special- Free Free Free 30% copayment ist visits Essential drugs Covered up to GEL 50 Covered up to GEL 100 per Covered up Not covered (around 50) per year (GEL 200 for year for pensioners (GEL 50 for to GEL 50 per pensioners). children 0-5 years). year 50% copayment 50% copayment. Diagnostic tests Free Free Free 30% copayment (basic lab tests) Diagnostic tests Free Free for most. Free 30% copayment (ultrasound, ECG, x-ray) 10-20% copayment for CT scans. Normal delivery Covered up to GEL 500 Covered up to GEL 500 Covered up to Covered up to GEL 500 GEL 500 C-section Covered up to GEL 800 Covered up to GEL 800 Covered up to Covered up to GEL 800 GEL 800 Elective surgery Covered up to GEL Covered up to GEL 15,000 Covered up to Covered up to GEL 15,000 per year per year. GEL 15,000 15,000 per year. per year 10% copayment (pensioners). 30% copayment. 20% copayment (children 0-5 years, people with disability, students, teachers). Chemotherapy, Covered up to GEL Covered up to GEL 15,000 Covered up to Covered up to GEL hormone and 12,000 per year per year. GEL 12,000 12,000 per year. radio therapy per year 10% copayment (pensioners). 20% copayment. 20% copayment (children, students, teachers, people with disability). Emergency Free Free Free Free outpatient care Emergency Free 10% copayment (pensioners). Free Covered up to GEL inpatient care 15,000 per case. 20% copayment (children 0-5, people with disability, students, 30% copayment. teachers). Source: WHO, World Bank and USAID (2016). 28 Box 2.1: Defining Universal Health Coverage UHC means that all people can use the promotive, preventive, curative, rehabilitative and palliative health services they need, of sufficient quality to be effective, while also ensuring that the use of these services does not expose the user to financial hardship. This definition of UHC embodies three related objectives: 1. Equity in access to health services - everyone who needs services should get them, not only those who can pay for them; 2. The quality of health services should be good enough to improve the health of those receiving services; and 3. People should be protected against financial risk, ensuring that the cost of using services does not put people at risk of financial harm. UHC is, therefore, about progressively reducing both the ill-health burden and the economic burden of disease. Increasing effective coverage of health interventions and reducing dependence on OOP payments, as well as increasing financing from prepaid and pooled sources, are key to making progress towards UHC. Three dimensions–(effective) health services, finance, and population–are typically represented in what has come to be known as the coverage cube: Notably, this representation in the UHC coverage cube excludes several critical dimensions of health financing and service delivery. These include care that is foregone due to lack of access and/or high OOP spending, and indirect costs incurred in seeking care such as transport costs and foregone income. It also does not account for quality of health care and efficiency of health expenditures. Many of these issues are relevant in Georgia–in particular high OOP spending and efficiency–and are discussed below. Source: WHO 2010, WHO 2015. 29 C. OVERVIEW OF HEALTH SPENDING 2.10. Total health spending in Georgia–at 8.5 percent of GDP in 2014 (Table 2.2)–is much higher than the average for upper-middle-income countries (7.0 percent) and approaching the EU average (10 percent) (World Bank 2016). Cross-country data show that total health spending tends to rise with income, as populations expect and demand better coverage and quality of services, and as new and more expensive medical technologies are introduced. Richer countries also tend to have older populations with more NCDs and a greater need for chronic care. 2.11. Nevertheless, while the total level of health spending in Georgia is high compared with countries of a similar income level, public health spending as a share of GDP is below average and relatively very low by European and international standards. Georgia’s public health spending started from an extremely low base and, despite recent increases, the share of public health spending remains lower than in many other comparable countries. In 2014, public spending in Georgia comprised just 28 percent of total health spending, the lowest in the European region (Table 2.2). In addition, the split between government and OOP financing is comparable to that of a low-income country.42 Table 2.2: Health Financing Indicators, Georgia and Comparator Countries (2014) Total Health Total Health Government Share OOP Spending as a Health Share Expenditure Per Expenditure as of Total Health % of Total Health of Total Country/Region Capita (current % of GDP Expenditure (%) Spending Government US$) Spending (%) Georgia 374 8.5 28.2 66.0 8.9* Armenia 162 4.5 43.0 53.5 7.0 Croatia 1,050 7.8 81.9 11.2 14.0 Estonia 1,248 6.4 78.8 20.7 13.5 Latvia 921 5.9 63.2 35.1 9.8 Lithuania 1,063 6.6 67.9 31.3 13.4 Russian Federation 893 7.1 52.2 45.8 9.5 Slovenia 2,161 9.2 71.7 12.1 12.8 Upper-middle- 494 7.0 66.6 24.9 -- income countries ECA 481 6.1 48.4 43.9 -- Source: World Bank 2016. Note: * 2015 for Georgia health share of total government spending. 42 As national incomes rise, health spending not only rises but its structure also typically changes from a heavy reliance on OOP payments to more prepayment and pooling (Fan and Savedoff 2014). 30 Figure 2.1: Total Health Expenditures by Source 2.12. There has been a substantial increase of Financing (2001-14) in the government’s budget allocation to health in recent years due to the 100% implementation of the UHC. Between 2012 and 2014, the public share of total health spending 75% increased substantially from 19 percent to 28.2 50% percent (Figure 2.1), with much of the increase 25% associated with the introduction of the UHC Program. From 2012 to 2015, the health budget 0% more than doubled, increasing from 4.0 percent 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 to 8.4 percent of total government spending, and International Aid as a percentage of GDP from 1.3 percent to 2.8 Voluntary Health Insurance Taxes Mandatory Health Taxes Paid by employer percent. In this respect, Georgia is experiencing State budget sources a steep increase in its health sector spending, OOP Mandatory Health Taxes Paid by employee which is consistent with other middle-income Regional budget sources countries’ experience at the time of UHC introduction (Figure 2.2). Sources: National Health Accounts 2001-14; MoLHSA 2016. Figure 2.2: Share of Government Expenditures on Health During Periods of Coverage Expansion Brazil and India 12 10 Philippines, Thailand and Vietnam 8 18 16 6 14 4 12 10 2 8 0 6 4 2 0 Brazil India Philippines Thailand Vietnam Source: Authors’ estimates. Note: Arrows indicate when policy commitments were made to significantly expand coverage. 31 Table 2.3: Public Spending on Health and Social Protection (millions of GEL)(2012-16) 2016 Function 2012 2013 2014 2015 (estimated) Total General Government 8,285 8,120 9,023 10,140 11,209 Spending MoLHSA Budget 1,794 2,126 2,718 2,978 3,312 Health programs 329 435 656 854 951 UHC 0 70 420 647 720 MIP 153 240 68 0 0 Preventive programs 8 31 44 68 84 Vertical programs 168 94 124 139 5 Other 0 0 0 0 142 Social Programs 1,398 1,639 1,971 2,034 2,273 Administration 47 44 51 54 51 Other incl. capital 19 8 41 36 37 Source: MoF. 2.13. Public spending on health in Georgia is mainly drawn from general tax revenues and allocated to the UHC Program and vertical programs, all of which are administered by the SSA. The UHC Program accounts for the largest share of public spending on health (75 percent in 2016) and has been increasing over time due to the expansion of UHC benefits and the inclusion of state health insurance programs, including some vertical programs (Table 2.3). 2.14. Public spending on health is allocated largely to curative care services provided at hospitals. Over three-quarters of public spending on health is on curative care and, in particular, inpatient curative care accounted for 55 percent in 2014 (Figure 2.3). Moreover, 67 percent of total public spending was on hospitals and 25 percent on ambulatory care providers (MoLHSA 2016). Similarly, the bulk of UHC Program spending was on emergency inpatient care. These shares have risen over time– reflecting the priority given to adequately covering curative care, particularly hospital services under the UHC Program. The share of public spending devoted to outpatient drugs is exceptionally low (less than 0.5 percent), leaving much of this to be purchased out-of-pocket (Annex 2 provides a more detailed breakdown of spending on the UHC and vertical programs). Compared to most OECD countries, the share of public spending on curative care is higher and that on medicines considerably lower in Georgia, reflecting wide coverage of outpatient drugs in most OECD health systems. 2.15. Notably, health spending on administration has declined since 2013 (Figure 2.3). This reflects a shift from a multipayer, PHI system with high administrative costs to a system with a single payer. Prior to 2013, Georgia’s public spending on health administration was considerably higher than most OECD countries, including Switzerland that also has a PHI-based system. Experience across a range of OECD countries has shown quite consistently that multipayer, PHI-based systems tend to have higher administration expenditures than single-payer, publicly insured systems. Georgia’s experience since 2013 is consistent with this trend. 32 Figure 2.3: Functional Breakdown of Public Figure 2.4: Composition of UHC Spending by Health Expenditures (2001-14)(% of GDP) Beneficiary Group 2013 100% 2011 2009 80% 2007 2005 60% 2003 2001 40% 0% 20% 40% 60% 80% 100% 20% 0% 2014 2015 2016 Former MIP (18% of bene�iciaries) Pensioners, children, students (25% of bene�iciaries) Former uninsured (56% of bene�iciaries) Source: National Health Accounts 2001-14 (MoLHSA 2016). Source: Social Services Agency and MoLHSA 2016. 2.16. Budget overruns of the UHC Program since 2014 have led to concerns about the financial sustainability of the program. In 2013, the UHC Program budget execution was much lower than planned (69 percent). However, by 2014, the actual UHC Program spending was one-third higher than planned. In 2015, the program’s planned budget was 39 percent higher than in 2014, and actual spending was again higher than planned. More generally, the UHC Program budget was expanded by GEL 96 million in late 2015, causing an unexpected widening of the fiscal deficit. The 2016 budget for UHC was initially set at the same level as in 2015 (GEL 570 million), but nearly 90 percent of the budget had already been executed by September 2016. 2.17. The budget overruns are largely due to unanticipated increases in the demand for health care. The SSA analysis of demand indicates that much of the increase was due to “pent-up” demand among those who were previously uninsured or lacked coverage for specific interventions (for example, glaucoma surgery). As Figure 2.4 shows, spending on those who were previously uninsured accounted for nearly 45 percent of the UHC Program spending in 2014–the first full year of implementation, and this ratio remains 35 percent of the total cost of UHC in 2015 and 2016. 2.18. Compared to the UHC, costs of the vertical programs are easier to control (WHO 2016). One reason for this is that, as opposed to the UHC where costs were incurred in many hospitals, the vertical program has centralized procurement through public international tenders that lowers the price of drugs by an average of about 10-12 percent. A second reason is that volume control is achieved through targeted program management. Tariffs and service packages are also fixed across all providers, which helps cap expenditures. 2.19. Private health insurance remains modest. In 2015, half a million people (14 percent of the population) had PHI coverage, mostly through corporate policies provided by employers. This includes around 214,000 state employees (about 6 percent of the population) who have PHI paid for by the Ministry of Defense and the Ministry of Internal Affairs in addition to being covered by the UHC 33 Program. The MoLHSA estimates that demand for PHI has risen since 2015 as people look for ways to cover expenses not covered by the UHC Program (WHO 2016). Since private insurance companies are not obliged to report to the SSA, there have been concerns about the duplication of services by private insurers and the UHC Program. Appendix 4 reviews the role of PHI in Georgia in light of the international evidence on PHI. A key point to note is that no country in the world has achieved UHC by relying on voluntary PHI alone. 2.20. Despite rising public spending on health, OOP remains the dominant source of financing for health in Georgia, filling the void of health spending that is not covered through public sources (Box 2.2). OOP spending in Georgia is estimated to be 66 percent of all health spending, far higher than other countries in the region (Figure 2.5). High OOP payments are inimical to the goals of UHC, which can create barriers to accessing needed health services and absorb household resources that could otherwise be used for more productive purposes and potentially impoverish households. 2.21. High OOP is likely to distort people’s incentive towards overuse of costly services that are free under the UHC Program, such as inpatient emergency services. For all of these reasons, most countries support the principle that payment for health care should be based on ability to pay and access to services based on need, and put in place pre-pooling or risk-pooling mechanisms that reduce reliance on OOP for health care. International norms suggest that OOP spending as a share of total health spending should be less than 20 percent, in order to reduce the risk of impoverishment and catastrophic health spending. Box 2.2: Behind High OOP Health Spending in Georgia Although the UHC Program has been associated with a reduction in the OOP share of total health spending (73 percent in 2010 to 66 percent 2015), Georgia still has one of the highest shares of OOP in the region. This is attributable to the following factors: Limited coverage of outpatient drugs in the UHC Program: While medicines are provided free of charge to patients through MoLHSA’s vertical programs and for inpatient use, the UHC Program has a very limited outpatient drug benefit. Selected groups (the poor, veterans, and pensioners) are eligible for 50 percent reimbursement, while other groups are not eligible for the drug benefit at all. The annual claim limit per person is low,43 and prescriptions from rural doctors are not accepted–requiring patients to visit family doctors instead. Spending on outpatient medicines has consistently comprised less than 0.01 percent of total UHC Program costs. Complex copayments policy: The copayments policy does not provide anyone with adequate depth of coverage due to the presence of the annual cap on benefits. In addition, for those beneficiaries who are not eligible for free care, the patient copayment at the hospital is calculated as 30 percent of the hospital price or the maximum SSA tariff, whichever is lower. Patients also have to pay hospitals any difference between the SSA tariff and the hospital’s price. To add to this complexity, hospitals’ prices vary widely and different entitlements and rules that apply for copayment waivers for different groups cause confusion and undermine transparency (WHO 2016). 43 Having annual claim limits is unheard of in European health systems. 34 Figure 2.5: OOP Spending vs. Government Health Spending, Georgia and Comparator Countries (2014) 80 OOP share of total health spending (%) Georgia 60 Armenia Indonesia Russia 40 Vietnam Latvia China Lithuania Estonia 20 Turkey Croatia Slovenia 0 0 3 6 9 12 15 Government health spending share of GDP (%) Sources: National Health Accounts 2001-14 (MoLHSA 2016); World Bank 2016. D. HEALTH SPENDING AND KEY HEALTH SYSTEM GOALS 2.22. Georgia has made steady but modest progress in health outcomes in the two decades to 2017. Infant mortality has declined significantly from 41.6 per 1,000 live births in 1995 to 8.6 in 2015, and under-five mortality declined from 44.3 per 1,000 live births to 10.2 in this same period (Figure 2.6). Average life expectancy in Georgia at 74 years is about comparable to other countries at a similar level of income, but remains below the EU average of 81 years. 2.23. Improvement in life expectancy over the years has been slow. Georgia was far closer to European averages in the 1970s than it is today. Moreover, in 2012, Georgia’s healthy life expectancy was just 65, which is nine years lower than overall life expectancy at birth. This lost healthy life expectancy represents nine equivalent years of full health lost through years lived with morbidity and disability. There also appears to be a distinct gender gap in life expectancies in Georgia (Box 2.3). Despite good progress in infant and under-five mortality rates, Georgia continues to fare poorly in these health outcomes relative to other countries in the region and at a similar level of income. 35 Figure 2.6: Infant and Under-Five Mortality Rates, Georgia vs. Comparator Countries (2014) Infant mortality Under-�ive mortality 125 200 125 200 Ratio per 100,000 live births 50 50 Rate per 1,000 live births Indonesia 25 Indonesia Vietnam Vietnam Armenia Turkey Armenia Turkey Georgia Georgia China 10 10 China Russia Russia Latvia Latvia Lithuania 5 5 Croatia Croatia Lithuania Estonia Slovenia Estonia Slovenia LOWER UPPER MIDDLE MIDDLE LOWER UPPER LOW INCOME INCOME INCOME HIGH INCOME MIDDLE LOW INCOME INCOME MIDDLE HIGH INCOME INCOME 1 1 250 500 1000 2500 10000 35000 100000 250 500 1000 2500 10000 35000 100000 GNI per capita, US$ GNI per capita, US$ Box 2.3: Gender Differences in Health in Georgia For analysis of gender differences in health, it is important to assess both sex-based and gender-based differences in health and related outcomes. While “sex” captures the biological differences between women and men, “gender” is widely used to refer to how social norms and culture shape women and men’s expected roles, responsibilities and opportunities. Biological factors predispose men or women to certain types of conditions and illnesses. Social and cultural norms can also shape gender-based health risks of women and men by influencing health-seeking behaviors. In order to meet the needs of all in the population, public spending on health needs to address any sex- and gender-based differences in health risks. As noted by WHO, “Gender equality in health means that women and men, across the life-course and in all their diversity, have the same conditions and opportunities to realize their full rights and potential to be healthy, contribute to health development and benefit from the results.” As a rapidly aging country, Georgia has more women than men in the population due to the relatively longer life expectancy of women (Figure 2.7). From the 3.7 million estimated total population in Georgia, about 52 percent are women (WDI 2015), but the male-to-female ratio is not uniform across age cohorts. Older age groups have more than twice as many women as men–an outcome of gender differential in life expectancies. While women over 65 represent 16.5 percent of the total female population, men over 65 make up only 11 percent of the male population. Georgian women’s life expectancy is estimated at 78 years while that of men’s is 71 years (estimates reported in WDI). 36 Figure 2.7: Residents by Gender and Age (2015) Figure 2.8: DALYs lost (2012) by Sex and Cause (% of population) (% of DALYs) 80+ 75 - 79 70 - 74 65 - 69 Women 60 - 64 55 - 59 50 - 54 45 - 49 40 - 44 35 - 39 30 - 34 Men 25 - 29 20 - 24 15 - 19 10 - 14 5-9 0% 20% 40% 60% 80% 100% 0-4 Communicable, maternal, perinatal and nutritional conditions Noncommunicable diseases Men Women Injuries Source: Bank staff calculations. Source: Estimated Disability-adjusted Life Years (DALYs) by cause and WHO member state (WHO 2014). There are biological and behavioral differences that lead to greater ill health among men but there appear to be no gendered barriers to accessing health services. The number of male deaths exceeds number of female deaths until age 70-74 years. At least part of this outcome is rooted in behaviors that put men at risk of higher morbidity and mortality. Men are more likely to die from external causes, such as violence, injuries, and the effects of prolonged use of alcohol and tobacco. Men are also more likely to use tobacco. According to the Household Utilization and Expenditure Survey (HUES 2014), 65 percent of men aged 15 or older, compared to only 4.5 percent of women, have ever consumed tobacco products. An analysis of DALYs shows that men experience higher lost DALYs than women do. Cardiovascular disease (CVD) accounts for the dominant share of the burden of disease (BoD) for both women and men. The BoD due to injuries is, however, disproportionately higher for men than women, a gender pattern that is found in most countries (Figure 2.8). Nevertheless, in the HUES survey data, women are more likely than men to report illness in the previous 30 days or six months. As is commonly found across the globe, it is possible that gender norms could prevent men from reporting illnesses in the household survey. The utilization of health services in case of illness, which is also self-reported in the HUES, shows no significant gender differences. Taken together, the evidence suggests that the main source of gender difference in health in Georgia may lie in differences in risky behaviors that put men at a higher risk than women for ill health. The 2010 Law on Gender Equality recognizes that equal access to medical assistance should be ensured, and that mothers and children might need special consideration, but measures addressing men’s health issues are conspicuous by their absence. 37 2.24. NCDs comprise the majority of the burden of disease in Georgia today, similar to that in most European countries. NCDs account for more than 80 percent of the BoD in Georgia, and 94 percent of all deaths. CVDs alone account for 38 percent of the BoD (Table 2.4), and 69 percent of all deaths (Figure 2.9)(WHO 2014). The top three health risk factors in Georgia are dietary risks, high blood pressure, and tobacco smoke–all of which are highly correlated with NCDs. There is substantial progress to be made on NCD-related outputs and outcomes. In 2010, only 33.4 percent of adults with elevated blood pressure were on medication for the condition (a proxy indicator for hypertension treatment coverage), while diabetes treatment coverage was estimated to be just 16.3 percent (NCDC&PH 2010).44 Georgia’s population is also aging rapidly. By 2030, the population aged 65 years and over is expected to account for one-fifth of the population. Table 2.4: CVDs and Other NCDs Dominate Disease Burden (2000, 2012) Type of Medical Condition 2000 2012 Cardiovascular diseases Non-communicable 37.7 38.1 Malignant neoplasms Non-communicable 10.5 10.9 Mental and behavioral Non-communicable 8.8 9.3 Injuries Injuries 7.6 7.1 Musculoskeletal diseases Non-communicable 4.5 4.8 Respiratory diseases Non-communicable 2.9 4.1 Neonatal conditions Communicable 5.0 4.0 Neurological conditions Non-communicable 3.0 3.1 Infectious and parasitic Communicable 3.0 2.4 Respiratory infections Communicable 2.4 1.4 Source: Estimated DALYs by cause and WHO member state (WHO 2014). 2.25. At the same time, while communicable diseases comprise a very small proportion of the overall disease burden, they continue to be a public health concern in Georgia. For example, HIV incidence has increased from 5.4 per 100,000 persons in 2005, to 19.3 in 2015. Multidrug-resistant tuberculosis (MDR-TB) also continues to be a problem in Georgia, with MDR-TB and HIV/TB co-infection being a particular challenge. Effective diagnosis and treatment of these diseases, therefore, continues to remain important. 44 Diagnosis rates for hypertension and diabetes not available. Proxy for diabetes treatment coverage is measured by the percentage of adults aged 18-64 with raised fasting blood glucose or currently on medication for raised blood glucose. 38 Figure 2.9: Leading Causes of Death, All Ages, Georgia (2014) Communicable, maternal, perinatal and nutritional conditions 3% Injuries Other NCDs 3% 6% Diabetes 1% Chronic respiratory diseases, 4% Cancers 14% Cardiovascular diseases, 69% 2.26 A key link between health system inputs and outcomes is the delivery of health system outputs, measured through service use and coverage. While health outcomes can take years to respond to health system reforms such as the introduction of UHC, health service outputs tend to respond more quickly–providing some interim evidence of the impact of the reforms. 2.27. Outpatient visit and hospitalization rates in Georgia have both increased in recent years, reflecting the increase in public spending and coverage for these services. Overall, people in Georgia are more likely to seek care when ill today, than they were five years ago. In 2014, 79 percent of those who were ill in the previous six months consulted a health care provider–a slight increase from 75 percent in 2010 (WHO, World Bank and USAID 2016). On average, there were 4.0 outpatient visits per capita per year in 2015 compared to just 2.3 in 2012, and hospitalization rates have seen a steady increase since 2012 (Figure 2.10). This could largely be explained by the introduction of the UHC Program that offered coverage to a vast number of people in Georgia who were previously uninsured. Figure 2.10: Outpatient Visits and Hospitalization Rates (2009-15) Number of outpatient visits per capita Number of hospital admissions per capita 4.5 0.14 4.0 0.12 3.5 4.00 0.12 3.52 0.10 3.0 0.10 3.22 2.5 0.08 0.09 0.08 0.07 0.08 2.0 2.23 2.30 0.06 0.07 1.98 2.14 1.5 0.04 1.0 0.02 0.5 0.0 0.00 2009 2010 2011 2012 2013 2014 2015 2009 2010 2011 2012 2013 2014 2015 Source: NCDC 2016 39 2.28. Care-seeking behavior in Georgia is, however, skewed towards hospital-based, inpatient care, whereas outpatient services–and, in particular, primary health care–are underutilized. Compared to other countries in the region, outpatient visit rates remain low: there were on average seven outpatient visits per capita in the EU; other countries in the region had outpatient visit rates closer to WHO’s proposed benchmark of five visits per person per year (Figure 2.11:).45 Conversely, the number of inpatient procedures per 100,000 population per year in Georgia exceeds that of other countries in the region–with the exception of Belarus (Figure 2.12). 2.29. The Cesarean section rate is high in Georgia. Georgia has some of the highest Cesarean section rates in the region. The rate even exceeds that of Chile, which has a reputation for having an exceptionally high share of Cesarean section deliveries (Figure 2.13). This could reflect the inefficiency in design of Georgia’s health sector which may have generated incentives for all hospitals to prioritize high-cost services. Figure 2.11: Number of Outpatient Visits Per Figure 2.12: Inpatient Procedures Per 100,000 Capita in Georgia and Comparator Countries Population Per Year in Georgia and Comparator (2014-15) Countries (2014) Belarus 8,749 10 8.9 Georgia 5,488 9 CIS 5,434 8 7 7 6.3 6.3 Moldova 4,898 5.9 6 4.8 Armenia 4,323 5 3.9 4 4 Kazakhstan 4,323 3 Kyrgyzstan 3,114 2 Uzbekistan 2,629 1 0 Turkmenistan 2,623 Azerbaijan 1,865 Tajikistan 1,846 0 2000 4000 6000 8000 10000 Source: MoLHSA 2016; European Health For All database. Source: European Health For All database. 2.30. Overall, the efficiency of public spending on health measured in terms of service outputs is relatively low. Georgia produces fewer outpatient contacts per capita compared to other countries in the region. Low rates of hospitalization (114 per 1,000 population compared to 181 for the European Region in 2013), low average lengths of stay (5.3 days compared to 8.3 in the European Region in 2013) and very low bed occupancy rates (53 percent compared to 80 percent in the European Region in 2013) suggest that the hospital sector does not operate efficiently (Figure 2.14:)(WHO 2016). Data on the number of hospitals and beds per hospital suggest most hospitals operate with fewer than 100 beds. It is questionable whether multiprofile hospitals can function efficiently with so few beds. It is also difficult to ensure quality of care when care volumes are so low (WHO 2016). 45 Cross-country data on hospital admission rates are not available. 40 Figure 2.13: Cesarean Section Rates Per 1,000 Live Births (2014) ECA* 243 EU-15 (2014) 259 Chile (2008) 325 Georgia (2014) 385 0 100 200 300 400 500 Source: European Health For All database; OECD 2015. Notes: *Average for ECA is latest available year (2011-15). Figure 2.14: Utilization Rates, Georgia and Comparator Countries Hospital average length of stay (days) Bed Occupancy Rate (%, acute hospitals) CIS region (2015) 11 CIS region (2015) 108.6 EU-15 (2014) 8.3 ECA* 80.0 EU-15 (2013) 77.3 ECA* 7.9 Georgia (2014) 53.0 Georgia (2014) 5.3 0.0 20.0 40.0 60.0 80.0 100.0 120.0 0 2 4 6 8 10 12 Source: MoLHSA; European Health For All database. Note: *ECA average from latest available year (2011-15) for each country. 2.31. There continue to be gaps in coverage of preventive care, health promotion services, and public health in Georgia. Table 2.5 shows how Georgia fares on a range of tracer indicators for prevention and promotion46 relative to comparator countries. Overall, Georgia’s performance is mixed– doing very well in certain indicators (skilled birth attendance, access to improved drinking water), while other indicators lag behind regional peers (family planning, tobacco nonuse). Improving performance on many of these indicators requires years of consistent investment and steady, incremental progress. 2.32 While performance on some outputs and outcomes are influenced by factors outside the health sector–for example, access to improved drinking water and sanitation depend at least in part on improvements in basic infrastructure, and public utility systems–many can be improved through health sector reforms. Notably, tobacco use can be reduced through a stronger taxation regime, as well as restrictions on advertising, and health promotion initiatives such as smoking 46 Indicators proposed under the Joint WHO – World Bank Universal Health Coverage Measurement Framework (WHO and World Bank 2015). 41 cessation campaigns. There is also room for improvement on immunization coverage in Georgia through a stronger and sustainable immunization program. While it would not be possible to see the impact of Georgia’s UHC Program on these indicators just yet, monitoring these over time would help provide international comparability to the UHC reforms. Table 2.5: UHC Tracer Indicators for Prevention and Promotion Countr/Region Family ANC Skilled Birth DPT3 Tobacco Access To Access To Planning (%) Attendance (%) Nonuse Improved Improved (%) (%) (%) Drinking Water Sanitation (%) Georgia 53 98 100 91 68 100 86 Armenia 55 99 100 93 73 100 89 Croatia NA 94 100 95 64 100 97 Estonia 63 99 99 93 67 100 97 Latvia 69 98 99 92 63 99 88 Lithuania 63 100 100 93 70 97 92 Russian Federation 68 100 100 97 59 97 72 Slovenia 79 100 100 95 80 100 99 Upper-middle- 58 94 95 90 76 93 84 income countries Europe and 56 95 98 93 71 95 90 Central Asia Source: World Development Indicators, latest available year. 2.33. In terms of equity outcomes, intro­ Figure 2.15: Among People Who Reported Being duction of the UHC Program was followed by Ill in the Last Six Months, Share Who Consulted a a significant increase in utilization among Health Care Provider (2010 and 2014) poorer households (WHO, World Bank and 2010 2014 USAID 2016). For instance, the largest increase 100 in consultations among those who reported being % consulting a health care 80 ill occurred among lower- and middle-income provider when ill 60 households (Figure 2.15). These households were less likely to have had insurance coverage before 40 the introduction of the UHC Program. Geographical 20 variation in outpatient contacts suggests, however, 0 that there may be serious inequities in access to Total Rural Third Urban Fourth Second Richest Poorest �ifth �ifth care across the country: there is a huge discrepancy between rural areas and Tbilisi. Ambulance care also favors people living in Tbilisi (WHO 2016). Source: WHO, World Bank and USAID (2016). Note: Darker columns indicate statistically significant results; black: p<0.01; dark grey: p<0.05 or p<0.1. 42 2.34. Financial access has improved for inpatient care, but not necessarily for medicines. The decline in financial barriers to accessing inpatient care was steep among the poorest and the third and fourth income quintiles (Figure 2.16). As a result, the level of unmet need for inpatient care among the bottom 20 percent of the population is now closer to the level of unmet need experienced by richer people. Financial barriers to accessing medicines fell overall between 2010 and 2014, but the decline was only statistically significant for people in urban areas and the top quintile of the population (Figure 2.17). These findings suggest that the UHC Program has not improved access to medicines for 80 percent of households and people in rural areas, although the substantial increase in public spending on inpatient care has helped improve coverage in this area. Figure 2.16: Share of People Who Reported Figure 2.17: Share of Consultations Where Needing Hospitalization in the Last Year but Were Medicine Was Prescribed But Not Purchased Due Not Hospitalized Due to Cost (2010 and 2014) to Cost (2010 and 2014) 2010 2014 2010 2014 30 8 25 % of consultations where medicine was prescribed % of people needing 6 20 hospitalization 15 4 10 5 2 0 Poorest �ifth Third Richest �ifth Total Second Urban Rural Fourth 0 Total Rural Third Urban Fourth Second Richest Poorest �ifth �ifth Source: WHO, World Bank and USAID (2016). Source: WHO, World Bank and USAID (2016). Note: Darker columns indicate statistically significant Note: Darker columns indicate statistically significant results; black: p<0.01; dark grey: p<0.05 or p<0.1. results; black: p<0.01; 2.35. A Benefit Incidence Analysis indicates that the distribution of public spending on health on outpatient care was considerably more pro-poor in 2014 than in 2010 (Table 2.6). The distribution of public spending on inpatient care remained the same although the spending on general outpatient care was considerably more pro-poor in 2014 compared to specialized outpatient care. 43 Table 2.6: Distribution of Public Health Expenditures (2010 and 2014) (%) Income Quintile Inpatient Outpatient-All Outpatient- Outpatient- Outpatient- Total Subsidies Others Ambulance Hospitals Polyclinic 2010 Lowest quintile 20.4 23.8 32.5 17.6 24.8 22.1 2 18.6 21.3 24.2 17.1 22.9 19.6 3 19.3 21.5 16.3 18.7 21.0 19.2 4 20.7 14.9 11.8 22.0 15.7 18.8 Highest quintile 20.9 18.5 15.2 24.6 15.6 20.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 Share in the total subsidy 47.0 14.7 13.2 22.0 3.2 100.0 Concentration Index 0.0113 -0.0826 -0.2222 0.0615 -0.1219 -0.0264 2014 Lowest quintile 21.2 26.2 24.1 23.7 21.6 22.5 2 17.8 26.7 20.4 17.7 22.4 19.2 3 17.6 18.4 18.9 19.5 17.0 18.1 4 21.9 14.8 13.2 16.6 18.2 19.4 Highest quintile 21.5 14.0 23.3 22.5 20.8 20.8 Total 100.0 100.0 100.0 100.0 100.0 100.0 Share in the total subsidy 63.5 13.2 10.0 12.7 0.7 100.0 Concentration Index 0.0096 -0.1675 -0.0529 -0.0144 -0.0505 -0.0234 Sources: Authors’ estimates based on HUES and NHA data for 2010 and 2014. Notes: (i) Benefit Incidence Analysis describes the distribution of public spending across individuals ranked by their household consumption. The methodology and approach used for this analysis is derived from O’Donnell et al. (2008). Public spending data is obtained from Georgia NHA for 2010 and 2014. Utilization of public facilities is derived from HUES 2010 and 2014. Unit costs are obtained by dividing aggregate expenditures by the weighted sum of utilization reported in the HUES, where weights are expansion factors indicating how many individuals in the population are represented by each sample observation. (ii) A positive concentration index indicates a more pro-rich distribution of public spending on health. A negative concentration index indicates a more pro-poor distribution. (iii) Totals may not equal 100 percent in each instance due to the effects of rounding. 2.36. OOP payments for health are “impoverishing” to households. If a household has total consumption expenditures including OOP above the poverty line, while total nonmedical consumption excluding OOP below the poverty line, they could be considered to have suffered impoverishment due to OOP for health. Figure 2.18 shows this graphically based on 2015 data. Households are ranked along the horizontal axis by total consumption. The vertical drip lines represent OOP for health, and the poverty threshold is indicated by the horizontal line. Applying this approach to 2015 household survey data and using an international poverty line of US$2.50/day, it has been estimated that an additional 6.6 percent of Georgian households were poor as a result of OOP for health. In 2010, an additional 6 percent of households were poor due to OOP, implying that risk of impoverishment due to OOP payments has remained unchanged following the introduction of the UHC Program. 44 Figure 2.18: The Impact of Household Health Spending on Poverty (2015) 10 pre-OOP consumption post-OOP consumption 5 0 -5 -10 0 2 4 6 8 1 Cumulative proportion of population, ranked from poorest to richest Source: Authors’ analysis using Household Budget Survey 2015. 2.37. The proportion of households facing “catastrophic” health care costs has changed little from 2010-15. While impoverishing OOP puts the emphasis on crossing the poverty line irrespective of the size of payments, catastrophic health expenditures occur when they exceed some threshold of total expenditure. The choice of threshold is somewhat arbitrary, but 25 percent of total expenditure is commonly used. In both 2010 and 2015, 10 percent of households had health spending over 25 percent of total expenditure (Table 2.7). This ratio remains one of the highest relative to comparator countries. At the lower threshold of 10 percent of total expenditures, the share of households experiencing catastrophic spending has increased from 28 percent to 34 percent over the same period and probably reflects increased spending on pharmaceuticals. Table 2.7: Proportion of Households with Catastrophic Health Expenditures Above 5, 10 and 15 percent Thresholds (%) Household Health Expenditure Year Above 10 percent Above 15 percent Above 25 percent 2010 28 20 10 2011 28 19 9 2012 28 19 9 2013 29 19 9 2015 34 23 10 Sources: Authors’ analysis using Household Budget Surveys. Notes: Health expenditures are considered catastrophic when they exceed a certain percentage of the household budget or catastrophic thresholds. 45 E. KEY CHALLENGES 2.38. Ensuring the financial sustainability of the UHC Program depends crucially on controlling key cost drivers and reducing major inefficiencies in service delivery. The major challenges Georgia is facing on the path toward sustainability and efficiency are as follows: i. Purchasing mechanisms are weak at managing costs effectively. ii. Current design and implementation of the UHC Program generates incentives towards high cost services and undermines efforts to effectively manage chronic conditions. iii. High cost of pharmaceuticals generates pressures on the UHC Program. Purchasing mechanisms are not good at managing costs effectively47 2.39. As the sole purchaser of services for the UHC Program, the SSA potentially has the power to purchase services strategically and manage costs effectively. The purchasing mechanism is the way in which public funds are used to deliver health services. It includes: (i) managing revenues and expenditures; (ii) contracting; (iii) paying providers and setting the right incentives; and (iv) monitoring provider performance, service and quality. A key element of the UHC reform was to transfer responsibility for purchasing publicly financed services from private insurance companies to the SSA. Georgia has in place a single payer system that could potentially manage costs very effectively if the right capacity is in place. 2.40. In practice, the SSA is more of a passive purchaser (WHO 2016). First, in the absence of any selective contracting, the SSA’s main instruments for ensuring services are delivered appropriately are prior authorization and claims management. Both processes are a drain on the UHC Program’s administrative capacity because of the extensive paperwork involved. In practice, all claims from hospitals are reimbursed. Second, provider payment mechanisms are fragmented. The detailed and complex payment system for hospitals with different tariff setting and copayment rules for different types of hospital care enables providers to game the system, which leads to increased administrative costs and constrains SSA’s ability to control costs. Third, the SSA’s organizational set up is itself complex, which results in fragmented work practices. 2.41. The SSA has become more assertive in its role as a strategic purchaser and makes use of the leverage it has over health care providers. For instance, in 2015, the SSA stopped revising its tariffs on a monthly basis to cap increases in tariffs, which is one of the main reasons why UHC Program costs were rising previously. The SSA has also taken steps to standardize tariff-setting rules (for example, for critical and intensive care), which have already led to cost savings. Current design and implementation of the UHC generates incentives towards high-cost services and undermines efforts to effectively manage chronic conditions 47 The analysis in this section is drawn entirely from WHO (2016), a situation analysis of the purchasing system that was carried out by the WHO at the same time as this PER. 46 2.42. The inefficient design of the UHC Program generates incentives for patients to access costly emergency and hospital services. The UHC Program reimburses expenditures on drugs on the Essential Drugs List for up to GEL 50-100 per year for the poor, pensioners and veterans only. The rest of the population pays out-of-pocket for drugs prescribed during primary care visits. Rural doctors, who fall under a parallel vertical program to the UHC Program can prescribe medicines, but these medicines are not reimbursed by the UHC Program. Patients must go to the primary care center at the nearest rayon to obtain a prescription that can be reimbursed. Meanwhile, essential medicines are free when provided as part of inpatient or emergency care under the UHC Program, creating incentives to patients to access emergency and/or hospital services directly. 2.43. A poorly implemented referral system and perverse incentives embedded in the provider payment system have also resulted in underuse of primary care facilities and overuse of hospitals. Although primary care is the designated first point of contact, patients are able to seek specialist care by paying more out-of-pocket. The scope of primary care is poorly defined and fragmented by the existence of a rural doctor program, a vertical program running parallel to the UHC Program. Primary care providers are paid a fixed capitation without adjusting for patient risk,48 which creates incentives for the primary care doctors to push patients towards hospital care. Meanwhile, hospitals are paid on the basis of activity that creates incentives to pull patients towards ambulance and inpatient care, and towards emergency care in particular (WHO 2016)(Appendix 3 describes the payment mechanisms in detail). Table 2.8 summarizes the incentives facing providers and patients. This set of provider payment mechanisms and incentives explains why Georgia has some of the highest Cesarean Section rates in the region. Table 2.8: Incentives Facing Patients and Providers by Type of Care Type of Care Patient Incentives Provider Incentives First Contact Care Ambulance care Easy access, free of charge, quick solution More patients mean more revenue; refer to health problems, entry point for hospital patients to hospital to minimize risk. care. Rural doctor Easy access, limited scope of services, low Narrow scope of care, easy to refer patients trust, free of charge, entry point to the next to the next level, good performance is not level of care. rewarded, blurred role with family doctors. Primary care Easy access, mostly free of charge, Limited care delivery options, easy to refer to entry point to specialists, limited scope of the next level or to specialists in primary care services and services unclear, low trust, (in some cases this increases income), good relatively low cost for patients who have to performance is not rewarded, blurred role pay out-of-pocket. with rural doctors, very limited options to prescribe. Hospital Easy access, wide scope of care, high trust, More patients mean more revenue, easy and emergency emergency cases are free of charge, easy reasonable to refer to inpatient settings. department access to inpatient care, free medicines. 48 Fixed capitation pays a physician a fixed amount for each enrolled person assigned to them, per period of time, whether or not that person seeks care. Risk adjustment differentiates capitation rate for a patient taking into account her or his health status. For instance, elderly populations are likely to have greater health needs and incur higher costs, which would justify higher capitation rates for the elderly. 47 Next Level Care Specialist Wide scope of care, high trust, complicated More patients means more revenue from the outpatient and to navigate, copayment varies based on SSA and from copayments, good performance inpatient care patient, UHC Program and provider is not rewarded, revenue is gained by characteristics, free medicines. hospitalizing patients, categorizing patients as emergency cases and treating private patients. Vertical Targeted service package, free of charge, More patients means more revenue, but programs free medicines. providers have fewer opportunities to increase prices or collect copayments. Source: WHO (2016). 2.44. Weak primary care, inadequate coverage of outpatient drugs and a generally fragmented service delivery system mean that effective management of NCDs is severely compromised. Effective management of chronic conditions like NCDs requires a well-integrated health system centered around primary care, and effective coverage of essential drugs. The experience of Western Europe and North America in reducing NCD-related mortality in the past century highlighted the importance of early diagnosis and testing for key risk factors (for example, hypertension, cholesterol) and the treatment and management of those risk factors. 2.45. Much can be done to prevent and treat CVD with a package of highly cost-effective beta- blockers to control hypertension, statins for cholesterol reduction and thrombolytic for heart attack and stroke. These interventions require: (i) a strong primary care system that can screen and diagnose patients, manage the conditions appropriately and follow up as needed; (ii) adequate coverage of outpatient drugs so that patients adhere to the drug treatments that are prescribed; and (iii) good coordination between primary care and hospital services so that patients are referred appropriately to hospitals and there is adequate follow-up. While rates of awareness and treatment of hypertension are relatively high in Georgia, the proportion of people with controlled hypertension is much lower (Figure 2.19). Figure 2.19: Control of Hypertension – International Comparison Source: Smith and Nguyen (2013). 48 High cost of pharmaceuticals generates pressures on the UHC 2.46. A disproportionately large share of total health spending (40 percent) is on pharmaceuticals. This is very high by international standards–on average, only 17 percent of health spending in OECD countries is on medicines, and a typical range for middle-income countries is between 20 and 30 percent. Overall, two-thirds of OOP spending in Georgia is on medicines (MoLHSA 2016). Limited coverage of outpatient medicines in the public sector means that patients have little choice but to pay out-of-pocket for drugs at pharmacies. Furthermore, low-cost generic medicines are generally less available in retail pharmacies, as compared to more expensive originator products. This may influence purchasing decisions, skewing consumption towards higher-priced medicines. In addition, the fact that medicines are free when provided by inpatient services under UHC also incentivizes patients to utilize costly inpatient care. 2.47. Expanding the UHC drug benefit would help to defray OOP costs of medicines incurred by individuals and improve financial protection, especially for the poor and other vulnerable groups. At current spending levels, however, replacing OOP spending on medicines with public financing would be way beyond the means of the health financing system. Georgia will need to find a way to provide a pharmaceutical benefit that is sufficient to be seen as meaningful by the population, but still affordable to the state. F. THE WAY FORWARD 2.48. Going forward, the government needs to ensure the financial sustainability of the UHC Program in order to maintain the progress made to date and further deepen coverage and financial protection. To achieve these higher-level health system goals, it is critical that efforts are made to increase efficiency and obtain better value for money from current spending. In this section, we discuss three mechanisms through which to improve efficiency of public spending on health. 2.49. First, the SSA needs to be empowered so that it continues to move towards active purchasing, so it is capable of making more effective use of available public funds and is more accountable. A systematic analysis of the purchasing situation carried out by the WHO for MoLHSA in 2015 (WHO 2016) made the following recommendations: i. Establish clear goals for the SSA, ways of monitoring its performance and mechanisms for regular reporting to the public. ii. Redesign the SSA’s internal organizational structure to enable different program teams to work together to tackle overlaps and inefficiencies, and to encourage the development of new and innovative ways of improving purchasing. iii. Develop the SSA’s analytical capacity to support day-to-day operations. This would involve further improvement of the IT systems and a strengthening of staff analytical skills. Introduction of a regular management reporting system would foster a culture of using data and evidence on a regular basis. iv. Optimize administrative procedures and reduce bureaucracy (paperwork) by minimizing activities that do not add value. 49 v. Consider introducing one contract/agreement per provider for the UHC Program and all of the vertical programs. This would enhance the negotiating power (leverage) of the SSA in relation to providers. 2.50. Second, Georgia’s health service delivery system needs to be reoriented towards greater emphasis on primary care, reduced reliance on acute hospital and emergency care, and better coordination between primary and hospital care. Such a reorientation would ensure that, for most chronic diseases, the onset of the disease is prevented or secondary prevention is provided early on; and treatment, where needed, is provided early, thus avoiding the more acute and catastrophically expensive treatments that are needed later in the disease. Coordination of care across providers reduces duplication of tests, and a strong primary care function based on a longstanding doctor-patient relationship ensures continuity of care and better patient case management. Experience from countries such as Germany, Denmark, segments of the United States, and Turkey have shown that a strong primary health care system is critical in achieving good health outcomes. 2.51. Key elements of this reorientation of service delivery are: i. Significantly expanding access to essential drugs in primary care. This involves not only expanding the outpatient drug benefit in the UHC Program but also increasing the availability, affordability and prescribing of generic medicines. As long as access to outpatient drugs remains limited, patients will face strong incentives to go directly to hospitals and/or emergency services when ill. ii. Enhancing quality and efficiency of primary care. The rural doctor program needs to be integrated into the UHC Program in order to reduce fragmentation of primary care. The incentives facing primary care providers need to be strengthened so that primary care providers take more responsibility for patient care, particularly for patients with multiple, chronic conditions (WHO 2016). This includes adjusting the capitation rate for risk or patient needs and introducing a performance-based component to SSA reimbursement for primary care. iii. Enhancing quality and efficiency of hospital care. The current method of paying hospitals is complex, contributes to waste and compromises transparency. The SSA has already taken steps to standardize the payment rates and could use its leverage as single purchaser further in this area. The SSA should also introduce quality and access standards as part of agreements with hospitals participating in the UHC Program and vertical programs (WHO 2016). iv. Strengthening coordination of care. As a starting point, it would be useful to examine: (i) the extent to which services are being delivered in appropriate care settings; and (ii) whether there is adequate coordination and continuity of care across care settings. Having identified the areas where care coordination fails the most, MoLHSA could incrementally strengthen care coordination by aligning incentives of patients and providers with the right objectives and putting in place the right infrastructure and tools, especially electronic data management systems to pursue this goal further. 50 2.52. Third, ensuring value-for-money of medicines provided in the public sector is another key reform area. Current reimbursement and purchasing systems contribute to inefficient spending. While medicines provided under MoLHSA’s vertical programs are centrally procured, medicines for inpatient care are purchased directly by hospitals, likely with large variations in prices. In the absence of price controls and uniform reimbursement rates, the UHC Program’s outpatient benefit also pays for medicines at variable prices. 2.53. Adopting a reference pricing system can improve the efficiency of spending on pharmaceuticals in the public sector. Two forms of reference pricing, which are not mutually exclusive, can be considered. First, external reference pricing (ERP) uses price information from other countries to derive a benchmark for the purposes of setting or negotiating prices at home.49 A complement to ERP is the use of internal reference pricing (IRP), which can help to foster price competition and harmonize prices within drug classes. IRP involves setting a single price to be paid by public payers by comparing prices of equivalent or similar products in a given chemical, pharmacological, or therapeutic group (Ruggeri and Nolte 2013).50 2.54. EU countries commonly use IRP as a means to regulate generic drug prices. A good example is from Germany, where the same reimbursement ceiling is applied to all generic equivalents in a drug class. Patients pay the difference between the reimbursement amount and the manufacturer’s list price. After this reimbursement method was introduced, only 7.5 percent of all products included in reference price groups were priced above the reference price (OECD 2008). Likewise in Greece, patients choosing originator over generic drugs are required to pay the difference. The quality of drugs would, however, need to be monitored and enforced strictly in order for such measures to be successful. 2.55. Over the medium term, the government needs to consider expanding outpatient drug coverage in the UHC Program. Expanding the UHC drug benefit would help to defray OOP costs of medicines incurred by individuals and improve financial protection. In practice, a gradual expansion of the drug benefit needs to be considered, starting with a limited subset of patients and products, for example, high burden diseases, vulnerable groups, or high-cost medicines. Given that CVD ranks at the top for both BoD and mortality, and the relatively low cost of treating hypertension, one option would be to start by expanding the benefit for first line antihypertensive drugs. Depending on the fiscal space available, the drugs could be provided at a higher reimbursement rate, or free of charge, to a broad majority or all patients with CVD. To ensure that there is minimal leakage or gaming of the benefit, a track-and-trace system that enables electronic prescriptions needs to be in place to monitor prescribing, dispensing, and adherence. 49 ERP methodologies differ widely across countries–for example, variations in product and country reference baskets, and different algorithms used to determine the reference price for a given product. ERP information is also usually combined with other policy considerations before prices are set (Ruggeri and Nolte 2013). 50 Putting in place an effective IRP system in Georgia will require a process for defining reimbursement levels. This will need to be informed by better pharmaceutical market data, and will require a multistakeholder approach to ensure buy-in and participation in the system. Generally, countries that use IRP adopt a series of steps to determine a price that will be paid for through public financing sources. Data available in Georgia today is limited to median unit prices for selected originator medicines and their lowest price generic equivalent. A better understanding of all market price segments for each therapeutic group, per dose or course of treatment would help to inform policy making on using IRP to establish reimbursement rates in the public sector. Given the oligopoly situation in Georgia’s pharmaceutical sector, any successful referencing system would need to include all the major players. Excluding a party with significant market share would skew the data, and could also result in nonparticipation by that company. A multi-stakeholder forum could help to bring all key players on board and mitigate objection to a price referencing system. 51 2.56. In the short term, reducing duplicate coverage among those who have both PHI and UHC coverage and further strengthening SSA’s purchasing capacity would generate efficiency savings that can be used more productively within the sector. Using these savings to improve access to lower- cost, generic medicines at the outpatient level would help to reduce the reliance on expensive, hospital and emergency services. 2.57. In the medium to long term, Georgia is likely to find that its low level of public spending on health is a source of inefficiency in and of itself. Additional health spending (fiscal space permitting) would be needed to bring Georgia up to the level of other upper-middle-income countries, improve health outcomes, equity and financial protection. Health care costs will increase given the rising burden of NCDs and an aging population. The current service delivery structure is not appropriate for dealing with these demographic and epidemiological cost drivers in a cost-effective manner. Reorienting care away from hospitals and towards greater use of cost-effective primary care services will require significant additional investment in the medium to long term. 52 References Armstrong, J., Paolucci, F., McLeod, H., & van de Ven, W. (2010). Risk Equalization in Voluntary Health Insurance Markets: A Three Country Comparison. Health Policy, 98, 39-49. Dukes, Maurice Nelson Graham, ed. Drugs and money: Prices, affordability and cost containment. Vol. 15. IOS press, 2003. Fan, V., & Savedoff, W. (2014, March). The Health Financing Transition: A Conceptual Framework and Empirical Evidence. Washington DC: Center for Global Development. Hall, J., Lourenco, R., & Viney, R. (1999). Carrots and Sticks – the Fall of Private Health Insurance in Australia. Health Economics, 8, 653-660. Hsiao, W. C. (1995). Abnormal economics in the health sector. Health Policy, 32, 125-139. Retrieved from https://cdn1.sph. harvard.edu/wp-content/uploads/sites/100/2012/09/hsiao_1995_-_abnormal_economics_in_the_health_sector.pdf. Hsiao, W. C. (1995). Abnormal Economics in the Health Sector. Health Policy, 32(1-3), 125-139. Lu, J.-F., & Hsiao, W. C. (2003). Does Universal Health Insurance Make Health Care Unaffordable: Lessons from Taiwan. Health Affairs, 22(3), 77-80. Maynard, A., & Dixon, A. (2002). Chapter Five – Private Health Insurance and Medical Savings Accounts: Theory and Experience. In E. Mossialos, A. Dixon, J. Figueras, & J. Kutzin (Eds.), Funding Health Care: Options for Europe. Buckingham: Open University Press. Mossialos, E., & Thomson, S. (2002). Voluntary Health Insurance in the EU: A Critical Assessment. International Journal of Health Services, 32(1), 19-88. O’Donnell, O., van Doorslaer, E., Wagstaff, A., & Lindelow, M. (2008). Analyzing Health Equity Using Household Survey Data: A Guide to Techniques and Their Implementation. Washington DC: World Bank. Retrieved from https://openknowledge. worldbank.org/handle/10986/6896. OECD (2008). Health Policy Studies: Pharmaceutical Pricing Policies in a Global Market. OECD. (2015). Health at a Glance 2015. Paris: OECD Publishing. Pharmaceutical Pricing Policy (Chapter 9), Management Sciences for Health, 2012. http://apps.who.int/medicinedocs/ documents/s19585en/s19585en.pdf (accessed October 28, 2016). Ruggeri, K., and E. Nolte. “Pharmaceutical pricing: The use of external reference pricing”, Cambridge: RAND Corporation; 2013. Sagan, A., & Thomson, S. (2016). Voluntary health insurance in Europe. WHO Regional Office for Europe and the European Observatory on Health Systems and Policies. Sagan, A., & Thomson, S. (2016). Voluntary health insurance in Europe: role and regulation. WHO Regional Office for Europe and the European Observatory on Health Systems and Policies. Sapelli, C. (2004). Risk Segmentation and Equity in the Chilean Mandatory Health Insurance System. Social Science & Medicine, 58, 259-265. Seiter, Andreas. (2010). A practical approach to pharmaceutical policy. World Bank Group. Seiter, Andreas. “Georgia - Current Challenges in Pharmaceutical Policy”, World Bank Group, June 2013 (unpublished). Smith, O., & Nguyen, S. (2013). Getting Better: Improving Health System Outcomes in Europe and Central Asia. Washington DC: World Bank. Thomson, S. (2016, July 22). Medicines and Financial Protection. Tbilisi. Thomson, S., & Mossialos, E. (2004). What Are the Equity, Efficiency, Cost Containment and Choice Implications of Private Health 53 Care Funding in Western Europe? . Copenhagen: WHO Regional Office for Europe, Health Evidence Network Synthesis Report. Thomson, S., Busse, R., Crivelli, W., van de Ven, W., & Van de Voorde, C. (2013). Statutory health insurance competition in Europe: a four country comparison. Health Policy, 109, 209-225. Retrieved from http://www.healthpolicyjrnl.com/ article/S0168-8510%2813%2900015-8/abstract. Transparency International Georgia, “The Georgian Pharmaceutical Market”, Tbilisi, 2012. http://www.transparency.ge/ sites/default/files/post_attachments/The%20Pharmaceutical%20Market%20in%20Georgia_June%202012l_0.pdf (accessed October 22, 2016). van de Ven, W., Beck, K., Buchner, E., Schokkaert, F., Schut, F., Shmueli, A., & Wasem, J. (2013). Pre-conditions for efficiency and affordability in competitive healthcare markets: Are they fulfilled in Belgium, Germany, Israel, Netherlands and Switzerland. Health Policy, 109, 226-245. Retrieved from http://www.healthpolicyjrnl.com/article/ S0168-8510%2813%2900012-2/abstract. van Ginneken, E., Swartz, K., & Van der Wees, P. (2013). Health insurance exchanges in Switzerland and the Netherlands offer five key lessons for the operations of US exchanges. Health Affairs, 32(4), 744-752. Retrieved from http://www.ncbi. nlm.nih.gov/pubmed/23569055. WHO. (2016). Active purchasing for universal health coverage in Georgia: situation analysis and options for improvement. Barcelona: WHO Barcelona Office for Health Systems Strengthening. WHO. (2016). Making progress towards universal health coverage in Georgia. Briefing prepared by the WHO Barcelona Office for Health Systems Strengthening, Barcelona. WHO and World Bank. (2015). Tracking Universal Health Coverage: First Global Monitoring Report. WHO, World Bank and USAID. (2016). Summary of selected results from the Georgia Health Utilization and Expenditure Survey: 2010 and 2014. Final report prepared for the Ministry of Labour, Health and Social Affairs by the WHO Regional Office for Europe, USAID and the World Bank. Woolhandler, S., Campbell, T., & Himmelstein, D. (2003). Costs of Health Care Administration in the US and Canada. The New England Journal of Medicine, 349, 768-775. World Bank. (2016). World Development Indicators. Wouters, O., & MacKee, M. (2016). Private Financing of Health Care in Times of Economic Crisis: A Review of the Evidence. Global Policy, Special Issue Article. World Bank (2012). Georgia Public Expenditure Review: Managing Expenditure Pressures for Sustainability and Growth. World Bank (2014). Georgia Public Expenditure Review: Public Expenditures in Georgia: Strategic Issues and Reform Agenda. World Bank (2015). Georgia Public Expenditure Review: Selected Fiscal Issues. Funke, Michael. “Determining the Taxation and Investment Impacts of Estonia’s 2000 Income Tax Reform.” Finnish Economic Papers 15, no. 2 (2002): 102-109. Masso, Jaan, Jaanika Merkull, and Priit Vahter. “Shift from gross Corporate Income Taxation to distributed Corporate Income Taxation: Are there effects on firms?” Journal of Comparative Economics 41 (2013): 1092-1105. Masso, J., Meriküll, J., & Vahter, P. (2011). Gross profit taxation versus distributed profit taxation and firm performance: effects of Estonia’s corporate income tax reform. The University of Tartu Faculty of Economics and Business Administration Working Paper, (81-2011). OECD (2010). Tax Expenditures in OECD Countries Staehr, K. (2014). Corporate Income Taxation in Estonia. Is It Time to Abandon Dividend Taxation?.Tallinn University of Technology (TUT) Economic Research Series, (1). 54 Annexes Annex 1. Policy Recommendations of Previous PERs (2012, 2014, 2015) PER (2012) Progress as of end- Policy Area Reform Option Sequencing 2016 Maintain fiscal discipline by reducing overall fiscal deficit not followed (deficit Short/ from 3.6 percent of GDP in 2011 to 2 percent by 2015, through widened to 2.8 percent Medium term greater than one-for-one consolidation of expenditures of GDP by 2015) Macro-Fiscal In event of upside growth scenario from higher FDI and capital Not followed (expendi- inflows, deepen fiscal adjustment by saving Short/ tures above the targets additional revenues, thus sticking to expenditure targets rather Medium term at 32 percent of GDP than deficit targets by 2015) Ensure that capital expenditures make a greater contribution Partially followed to expenditure consolidation going forward than they have to (capital spending date, particularly as private investment picks up increased to 7 percent Expenditure Short/ of GDP by 2015, social Composition Manage expenditure consolidation through greater Medium term pressures not con- selectivity in capital expenditures, containing medium term tained) social expenditure pressures, and enhancing sustainability of road investment program (greater detail on each below) Not followed Selectively target further significant one-time pension benefit (practice of one-time Short term increases ad-hoc pension increases maintained) Pensions In progress (draft pen- Explore mechanism to index basic pension benefit Medium term sion law developed) Develop voluntary savings for retirement by undertaking In progress (draft pen- diagnostic of impediments and policy initiatives in the areas of Medium term sion law developed) tax treatment, opt-out vs. opt-in, and financial instruments Partially followed (TSA Improve TSA coverage of poor through improved outreach and proxy means formula proxy-means formula adjusted, however Short term Social Assis- Prioritize TSA over universal programs (e.g., pensions and parallel social tance energy vouchers) for any additional social spending increases programs launched, i.e. HMR law) Link TSA recipients to employment opportunities and human Not achieved but Medium term capital investments efforts are underway Reduce high out-of-pocket pharmaceutical spending by Not achieved but Short term promoting use of generics efforts are underway Address weaknesses in supply of medical care, with focus on Not followed Medium term Health primary care Not followed, universal Expand state-funded health insurance coverage with focus on Medium term program introduced lower deciles on population from 2013 (UHC) 55 Further improve content and presentation of capital budget by In progress (PIM extending coverage, demonstrating consistency in financial Short term guidelines developed information, and deepening non-financial information and approved in 2016) In progress (PIM Introduce a systematic preliminary assessment and project Short term guidelines developed identification process and approved in 2016) Capital In progress (PIM Budgeting Develop uniform methodological guidance on appraisal tech- Short term guidelines developed niques and approved in 2016) In progress (PIM Strengthen strategic guidance Medium term guidelines developed and approved in 2016) Develop systematic comprehensive capital budgeting system Achieved Medium term guidelines Increase routine and periodic maintenance expenditures in line Not followed Short term with HDM recommendations As EW Highway investment winds down, phase in rebalancing Not followed of outlays for rehabilitation and new construction to reduce Medium term Road Sector backlog of secondary network in poor condition Develop a viable system for attending to needs of local road Not followed Short term network by initiating rehabilitation of target subset Put in place institutional arrangements to improve road Short/ Not followed expenditure efficiency Medium term PER (2014) Volume 1 Policy Area Option for Consideration Sequencing Macro-Fiscal Consolidate public expenditures and increase revenues Medium term Not followed (sub-national at the sub-national level, with a target to reduce the revenues remain low, fiscal fiscal deficit from 3.8 percent of GDP in 2014 to 2.3 deficit set to widen to 4.1 percent in 2017. percent by 2017) Not achieved, but in progress (new civil service law approved, enactment Establish rule-based compensation policies and is postponed to June 2017, regulations for public employees and civil servants to secondary legislation, law on contain increases in salaries and bonuses. remuneration etc are being drafted) In Progress (PIM guidelines approved, implementation to be launched in 2017). Followed (Public Investments Strengthen public investment management and is at 6.6 percent of GDP in external oversight. 2016 and is expected to increase to 9 percent by 2020) Sustain capital expenditures at 6 percent of GDP. 56 Policy Area Option for Consideration Sequencing Social Limit the medium-term growth of the basic pension Medium-term In progress (draft pension law Protection and benefit to no more than the rate of inflation. developed) Health not achieved (needs increased Strengthen the capacity of the Social Sector Agency. Medium-term with the new major function of the single purchaser assigned) Not achieved Enhance the UHC package by covering pharmaceuti- Short-term cals for at least the lowest decile of the population. Education Preschool Education: Use the facilities of existing underutilized primary schools to hold classes for preschool education, train Short-term Not followed primary school teachers to teach preschool children, and target children of disadvantaged backgrounds and ethnic minorities to enroll in preschools. Medium-term Partially followed (regulation Improve the quality of service delivery by making the and monitoring mechanisms Ministry of Education and Science (MES) responsible approved for pre-school for setting and monitoring quality standards. institution, implementation in progress) Not monitored General Education: Medium-term Increase the number of working hours of teachers to bring it closer to OECD standards. This would result Not monitored in a decline in the number of teachers and the savings could be used to improve the salary scale but with improvements in teacher quality. Medium-term Achieved partially (starting salary increased, gradient Increase the class size. This again would require a remain steep) Teacher quality reduction in the number of teachers. Medium-term remains weak Raise starting salaries of teachers without making the In progress (ADB supports the gradient steeper to improve teacher quality. effort) Vocational Education and Training (VET): Medium-term Add more general education content to VET programs. Medium-term Encourage public private partnerships in the financing and delivery of vocational education. Short-term Reduce drop-out rates from VET programs by providing career guidance to help pupils make better informed choices. 57 Policy Area Option for Consideration Sequencing State Owned Establish an inventory of SOEs to get a consolidated Short-term Achieved (DPL supported Enterprises picture of the government’s fiscal position. action) Establish a clear dividend policy for SOEs. Short-Term Dividend policy not established Mainstream quasi-fiscal operations like Medium-term In progress (Fiscal risks annex provision of subsidized utility services and investment developed and attached to the projects undertaken by SOEs into the budget to annual budget law, increase transparency and distributive capacity of the commitment to monitor government. quasi-fiscal operations is reflected) Intergovern- Improve reporting of subnational finances to increase Short-term Achieved (LGs are fully mental fiscal transparency and fiscal management of SNGs. transferred to the single relations treasury account, all transactions are monitored at the regular basis, quality of data improved substantially) Not followed Revise the formula for the calculation of the Medium-term equalization grant to make it simple and to better Partially achieved reflect the overall pool of resources and expenditure (property tax for vehicles has needs of SNGs. been enacted from January Remove the suspension of taxes on movable property Medium-term 1, 2017; Income threshold and lower the income threshold for exemptions on for immovable property tax immovable property. maintained) Partially followed (PIT revenue sharing mechanism introduced, but sharing principals deviate from the recommended one) Medium-term Allow SNGs to share the tax base for certain taxes (such as the personal income tax) and be able to set their own tax rates on top of the national tax rate applied to the same tax base. Both taxes (the central and local components) could be administered by the central tax administration. 58 PER (2014) Volume 2: Diagnostics of PIM System In good progress: PIM guideline approved consistent with the PER recommendations; implementation to be launched from 2017 System Component Recommendations Priority Investment guidance, 1. Improve the forward-looking, strategic guidance for public Short to medium project development & investment. term preliminary screening 2. Institute formal preliminary screening for project concepts Short term on the basis of a completed project profile. Formal project appraisal 3. Introduce regulated requirements for project preparation Short term and appraisal so that there is a rigorous national procedure with a clear allocation of roles and responsibilities that applies to all public capital investment projects. 4. Develop and disseminate formalized technical guidance Short term on methodologies for assessing projects, tailoring it to the nature and scale of the project proposals. Independent review of 5. Consider options for institutional arrangements for inde- Short to medium appraisal pendently reviewing project feasibility studies and appraisal term decisions and take a decision taken on the most appropriate. Project selection and 6. Introduce more specific criteria for selection of projects for Short term budgeting budget funding, including a stringent criterion regarding adequate preparation and a positive appraisal decision. 7. Consider taking further steps towards implementing fully Short term the recommendations contained in PER-1 for improving the information content of the capital budget annex. Project implementation 8. Some simple project implementation guidelines should be Medium term introduced for operational staff. Predictability in the 9. Reduce the use of non-competitive procedures, by Medium term availability of funds for addressing some of the legal loopholes and applying more commitment of capital restraint in the use of executive decrees. expenditures 10. Explore ways of paying more attention to ensuring quality Short to medium in procurement while not jeopardizing the very significant term progress that has been made in cleaning up procurement processes using the price-based e-procurement system. Effectiveness of internal 11. Begin to develop the capabilities of internal audit units in Short to medium controls and internal major investing ministries to undertake performance audits term audit of capital investment projects. Project adjustment 12. Initiate a system for undertaking fundamental reviews of Long term major projects with problems, including externally financed projects. Facility operation 13. Institute a systematic survey of usage on opening, at least for Medium term major projects. Basic completion review 14. Make an analytical project completion report a requirement Short to medium and evaluation of the PIM system. term 15. Introduce systematic evaluation of the impacts of completed Long term projects. 59 PER (2015) Policy Area Option for Consideration Sequencing Fiscal Incidence - Remove VAT exemptions that are not equity Short-to Medum-term Not followed enhancing Followed Not followed - Eliminate or adjust threshold of the personal income tax - Channel poverty reduction programs through Not followed (UHC is in the Targeted Social Assistance (TSA) program place covering entire at central and local government levels for better population, poverty reduction results. kindergarten services remain free for all) - Introduce incentives for wider use of private services rather than public services for more affluent population, such as private health insurance and private kindergarten service. Agriculture - Develop agricultural insurance to address Medium-term Followed programs agricultural risks. Unknown Unknown - Institutionalize monitoring and evaluation (M&E) component Not followed (no major changes or reforms - Develop information management system via introduced in the the Agency of Public Registry (NAPR) cadaster. mechanisms of agriculture subsidies) - Design future agriculture support programs with improved targeting, high quality in- put, clear timeline, complementary advisory services, and M&E components. Subnational - Develop guidelines for the selection and Medium-term Partially achieved expenditures implementation of investment projects, (nation-wide PIM gradually decentralizing decision-making on guidelines developed, project selection and funding allocation, devel- including for the LGs, op accountability mechanisms, and capacity Implementation to be building in relevant skills for local officials. launched in 2017) - Develop national standards for quality and Not followed spending per child for preschool and put in place mechanisms to secure equitable preschool funding using conditional grants, mandating specific budget provision, or both. 60 Annex 2. Policy and Institutional Context I. Evolution of Health Financing Policy in Georgia, 1995 - 2015 1995 – 2004: Social Health Insurance A publicly financed social health insurance scheme was introduced in 1995. The State Health Fund (the State Medical Insurance Company from 1996) was established to pool payroll contributions (3% employer and 1% employee contribution rates, with state transfers on behalf of pensioners, unemployed, children, etc). In parallel, municipal health funds received per capita transfers from municipal budgets. In 1997, these 65 municipality funds were organized under the umbrella of 12 regional funds to address regional disparities. The publicly financed benefits package included nine state-level programs and five municipality-level programs with a highly complex design and low awareness among the population of their entitlements. The programs’ development was not in line with population health needs but driven by political priorities instead. They also suffered from underfinancing – for example, the state program accounted for only 5% of total spending on health – which led to an increase in out of pocket payments. The social health insurance scheme was terminated in 2004 due its poor performance and the government’s wish to lower taxation to boost economic growth (WHO, 2016). 2004: Introduction of Targeted Social Assistance and the Medical Insurance Program From 2004 to 2006, the government focused on the introduction of a targeted social assistance program based on means testing. This was extended to the health sector and the medical insurance program (MIP) was launched in 2007. The MIP targeted poor households, teachers, orphaned children and some others groups. The MIP covered a defined set of primary care benefits, emergency care, elective surgery, delivery and cancer treatment. These services were provided for free up to specified maximum amounts per year. The MIP covered essential drugs up to an annual coverage limit of 50-200 GEL, with user charges (50% of the price of the drug). The state provided immunization, dialysis, diabetes, TB, HIV and treatment for other infectious diseases and mental and other health services to all residents. In its early stages, the MIP was implemented by a single public purchaser. In September 2007, however, the government contracted out the MIP to private insurance companies (14 in total). The MIP led to an increase in population coverage from about 100,000 people in 2006 (mostly in Tbilisi) to 200,000 in September 2007 and over 700,000 by April 2008 (Transparency International, 2012). In 2008, the government moved to a voucher-based system which gave the beneficiary the right to choose one out of nine participating private insurance companies. The voucher-based system – a flat rate per person insured – gave insurance companies an incentive to be attractive to beneficiaries and to provide additional benefits (WHO, 2016). 61 2009: Voluntary Health Insurance In 2009, the government introduced the state voluntary health insurance (VHI) program to encourage non-MIP beneficiaries to enroll with private insurance companies. The VHI program targeted people aged 3-60 not covered by MIP and not already covered by private insurance. VHI covered primary, outpatient and inpatient care and emergency care up to 8000 GEL per year, with more generous benefits packages available at a higher premium. The VHI annual premium was 60 GEL (compared to the MIP annual premium of 180 GEL in 2009), of which two thirds were covered by the government and one third by the individual. The government contribution more than doubled per beneficiary (from 40.2 GEL to 96 GEL) if a private insurance company was able to attract more than 10,000 individuals. The target was to cover 300,000 to 500,000 individuals through the VHI program, rising to more than 1,000,000 by the end of 2009. However, numbers started to fall in 20106 and in July of that year the government abolished the state subsidy, causing numbers to fall even further (Transparency International, 2012). 2010: Private health insurance companies as purchasers at the regional level In 2010, the government divided the country into 26 medical regions, and the beneficiaries (about 900,000 – roughly 20% of the population) were assigned to the private insurance company responsible to their region of residence. Private insurers for each region were selected through public tender and granted a three-year contract. Annual premiums fell from 180 GEL in 2009 to 116-132 GEL, depending on region, and were not allowed to exceed 144 GEL. Private insurers were required to renovate hospitals in their region. Private insurers started competing to be the sole insurer in each region and MIP beneficiaries no longer had free choice of insurer. The private insurers enjoyed high profit margins – 18% in 2012 and 30-40% in earlier years (WHO, 2016). 2012: Expansion of MIP coverage In 2012, and just prior to the national elections, MIP coverage was extended to pensioners, disabled people, students and children under six years old. New beneficiaries were provided with the same benefits package as existing MIP beneficiaries but with user charges (10-20% of the service cost; 50% of the cost of essential drugs, with a maximum coverage amount of 100 Gel per year for drugs). 2013: Introduction of UHC Following the 2012 elections, the new government announced that all Georgians would be eligible for publicly financed coverage through the UHC program, which was introduced in February 2013 for those not covered by MIP or private health insurance. In its early phase, the UHC program provided a basic benefits package (primary and emergency care) to beneficiaries who registered with the primary care provider of their choice (WHO, USAID and World Bank, 2014). In the next phase in July 2013, the benefits package was extended to cover non-emergency care and some essential drugs. 62 Administration of the UHC program was handed over to the Social Security Agency (SSA) so as to move away from competitive private insurance model and the high administrative costs and profit margins associated with it. Some of the vertical programs were incorporated into the MIP and UHC program to reduce fragmentation. In April 2014, the transition to pool all publicly financed health programs in the SSA began. Since September 2014, the SSA has administered all health programs. The trends in coverage over time following the policy developments summarized above. Annex Table 1. Share (%) of the Population Covered by any Health Insurance, 2007, 2010 and 2014 Consumption quintiles 2007 2010 2014 Poorest 18.5 39.8 ***100.00 Second 14.8 31.1 ***99.98 Third 12.5 26.6 ***99.86 Fourth 14.4 23.7 ***99.80 Richest 8.6 26.2 ***99.90 Total 14.1 29.5 ***99.90 Source: (WHO, World Bank and USAID, 2016). Note: Statistical significance of difference from baseline is highlighted as follows: *** p<0.01; ** p<0.05; * p<0.1; quintiles always refer to household consumption Annex Figure 1. Georgia’s Path to UHC Source: MOHLSA 63 II. The Nuts and Bolts of the UHC Program Program objectives The UHC program is regulated by Government Decree 62. Its aims are to: • Ensure access to medical services for people who do not have private health insurance • Enhance geographical and financial access to primary health care • Increase outpatient services in order to rationalize costly and high-tech hospital services • Improve population health through better access to emergency and planned inpatient and outpatient services Program beneficiaries Georgian citizens, asylum seekers, internally displaced persons and persons with humanitarian status except prisoners are entitled to coverage under the UHC Program. Those who had private insurance on July 01, 2013 are not eligible UHC Program coverage, but can access a minimum package of services if they can demonstrate to the SSA that they no longer have VHI (about 1% of the population). In practice, eligibility is hard to verify as private insurers are not required to report their beneficiaries to MOLHSA or the SSA. The register of privately insured people has not been changed since July 2013. Program benefits The UHC Program benefits are described in Annex Table 2. The package of benefits essentially varies by the co-payment rate and case limits applied to different groups. The actual co-payment a provider pays depends on provider prices however, as shown in Annex Box 1. All Georgian citizens are also eligible to benefit from vertical programs covering a wide range of health services. The vertical programs are administered by the SSA separately from the UHC Program. Program coverage UHC Program coverage is based on the number of persons registered with a primary care provider. As Annex Table 3 shows coverage varies considerably by region, with the highest in large urban centers such as Tbilisi and Imereti. This suggest that access to a primary care provider affects the population’s ability to enroll in the UHC Program. Annex Table 2. Summary of UHC Benefits and User Charges, 2015 User charges Type of Benefit Former Beneficiaries All Others Former MIP of The Programme for Veterans (Previously Beneficiaries Pensioners etc Uninsured) Planned outpatient Free Free Free Free care Outpatient specialist Free Free Free 30% co-payment visits 64 User charges Type of Benefit Former Beneficiaries All Others Former MIP of The Programme for Veterans (Previously Beneficiaries Pensioners etc Uninsured) Covered up to 100 GEL Covered up to 50 per year for pensioners GEL per year (200 Covered up Essential drugs (50 GEL for children 0-5 GEL for pensioners) to 50 GEL per Not covered (around 50) years) year 50% co-payment 50% co-payment Diagnostic tests Free Free Free 30% co-payment (basic lab tests) Free for most Diagnostic tests (ultrasound, ECG, Free Free 30% co-payment 10-20% co-payment for x-ray) CT scans Covered up to 500 Covered up to Covered up to 500 Normal Delivery Covered up to 500 GEL GEL 500 GEL GEL Covered up to 500 GEL C-section Covered up to 800 Covered up to Covered up to 800 GEL 800 GEL GEL 20% co-payment Covered up to 15,000 GEL per case Covered up to 10% co-payment Covered up to Covered up to 15,000 GEL per case Elective surgery (pensioners) 15,000 GEL 15,000 GEL per case per case 30% co-payment 20% co-payment (chil- dren 0-5, disabled people, students, teachers) Covered up to 15,000 GEL per year Covered up to Covered up to Covered up to 10% co-payment Chemo-, hormone 12,000 GEL 12,000 GEL per year 12,000 GEL per year (pensioners) and radio therapy per year 20% co-payment 20% co-payment (children, students, teachers, disabled people) Emergency Free Free Free Free outpatient care 10% co-payment Covered up to (pensioners) 15,000 GEL per case Emergency inpatient Free Free care 20% co-payment (chil- 30% co-payment dren 0-5, disabled people, students, teachers) 65 Annex Box 1. Example of Payment Rules for Planned Surgery and Less Urgent Emergency Care The SSA tariff for episode X is 1000 Gel. Provider A did not participate in the MIP. Its price for episode X is 1500 GEL. The SSA pays 700 GEL (70% of the maximum tariff) and the partient (previusly uninsured) pays a co-payment of 300 GEL (30% of the maximum tariff). The patient also pays an additional 500 GEL (the difference between the maximum tariff and the provider’s price) – 800 GEL in total. Provider B did participate in the MIP, where its price for episode X was 500 GEL. The current maximum tariff for this episode for this provider is therefore 550 GEL (500 GEL plus 10%). The provider can charge a higher price, eh 1000 GEL. In that case, the SSA pays 385 GEL (70% of the maximum tariff) and the patient pays 165 GEL (30% of the maximum tariff). The patient also pays an additional 450 GEL (the difference between the maximum tariff and the provider’s price) – 615 GEL in total. Source: (WHO, 2016) Annex Table 3. UHC Program Coverage by Region, 2016 (First Six Months) Region Population Covered % Population Covered Racha-Lechkhumi 16,048 51 Mtskheta-Mtianeti 61,163 65 Samtkhe Javaketi 110,300 69 Sida kartli 186,756 71 Kakheti 238,297 75 Samegrelo Zemo Svaneti 251,896 76 Kvemo kartli 330,630 78 Guria 90,076 80 Adara 292,734 87 Imereti 484,949 93 Tbilisi 1,082,976 97 Total 3,145,825 85 Annex 3. Breakdown of Public Health Expenditures Annex Table 4. Breakdown of UHC Program by Type of Spending, 2013 – 2015 2016 UHC program sub-components 2014 2015 (H1) Av- Av- Av- erage erage erage Total cost Total cost Total cost # of cases # of cases # of cases cost per cost per cost per case case case (GEL) (GEL) (GEL) Cardiac surgery 2,355 13,421 5,699 3,256 19,244 5,910 1,966 11,929 6,068 Emergency inpatient care 166,062 211,582 1,274 236,456 357,241 1,511 139,313 203,190 1,459 Chemo-Hormone and radio 31,654 24,197 764 44,376 22,521 508 23,955 11,403 476 therapy 66 Elective Surgery (except cardio 71,408 60,936 853 110,233 99,157 900 56,377 50,773 901 surgery) Obstetrics and caesarean sections 48,924 28,364 580 50,508 29,932 593 23,081 13,929 603 Emergency outpatient care 467,308 33,556 72 717,486 59,126 82 387,872 32,872 85 Planned ambulatory care 7,913,760 48,093 6 10,107,374 60,080 6 6,295,968 35,106 6 basic drugs 5,706 26 5 4,424 21 5 3,202 13 4 Total 8,707,177 420,175   11,274,113   6,931,734 359,215   647,322 Source: MOHLSA Annex Table 5. Breakdown of Spending on Vertical Programs Vertical program (in million GEL) 2013 2014 2015 2016 (plan) Ambulance medical services 15,15 29,66 31,96 33,25 Dialysis and kidney transplantation 22,14 25,13 30,53 32,00 Rural doctors 11,29 20,38 23,92 26,00 Referral (individual care) 17,92 19,69 19,69 20,00 Mental health 14,57 15,09 16,16 15,30 TB diagnosis and treatment 8,65 8,43 15,29 14,00 Immunisation 5,97 4,43 11,17 14,28 Infectious disease management 1,26 7,38 8,39 8,00 Hepatitis C management 0,00 0,00 6,06 20,63 Diabetes 4,86 5,75 7,82 8,10 HIV/AIDS prevention and treatment 3,14 4,10 5,78 8,42 Maternal and child health 4,91 6,05 6,40 7,00 Rare diseases 3,82 4,21 5,64 6,00 Necrology 3,99 4,19 4,23 5,00 Early detection of diseases and screening 1,46 1,48 1,56 2,00 Palliative care of oncological patients 2,35 1,41 1,62 3,10 Safe Blood 0,82 1,07 1,34 1,65 Paediatric Oncology and haematology 1,67 1,63 1,27 1,70 Medical screening for army recruits 1,15 0,90 0,79 1,00 Post diploma education 0,00 0,03 0,31 0,87 Disease surveillance 0,65 0,92 0.61 1,70 Prevention of occupational disease 0,27 0,27 0,27 0,27 Health promotion 0,00 0,00 0,14 0,40 Financial support for medical facilities 0,00 4,95 Total 126,06 167,15 200,95 229,80 Share (%) of public spending on health 29% 29% 25% 51 The increase in 2014 was the result of private insurance handing over the program to the SSA. 52 The increase in 2014 was the result of private insurance handing over the program to the SSA. 67 Annex 4. Health Service Delivery System Ownership of Health Facilities All health care providers have been regulated under commercial law since 1997. Health care providers, both hospitals and primary care providers are largely private. However, there has been some nationalization of health facilities in rural and mountainous areas in recent years to ensure access to services where private, for-profit providers would not operate. Primary Care Primary care is provided through primary care facilities and a parallel rural doctor program. The Primary Health Care Masterplan (2003) aimed to consolidate 750 primary care facilities outside of Tbilisi into 549 facilities each serving 30,000 people. The rural doctor program covers 1.1 million people living in rural areas. While rural doctors can prescribe medicines, their prescriptions do not qualify for reimbursement through the UHC Program. People living rural areas are also registered with a primary care center in the nearest rayon in order to obtain access to specialist care and prescriptions covered by the UHC Program. The share of rural doctor visits has been declining over time Annex Table 6. Primary care is reimbursed on a capitation basis. The capitation payment is calculated taking into account expected rates of services use. The payment is not adjusted annually and has remained the same since the beginning of the UHC Program. The payment does not take into account age, patient characteristics or regional variations in need (WHO, 2016). The monthly capitation rate is 1.93 GEL, of which 0.86 GEL is for a family doctor and 1.07 GEL for primary specialists and diagnostics. For patients who are also covered by a rural doctor, the rate is 1.07 GEL. Annex Table 6. Primary Care Visits, 2009 – 2015, millions 2009 2010 2011 2012 2013 2014 2015 Total number of visits 7.9 8.4 8.6 10.5 12.2 13.1 13.2 Including: Visits to doctor 7.4 7.9 7.7 8.5 11.0 11.9 12.1 Of which, rural doctors 1.6 1.6 1.5 1.4 1.5 1.6 1.4 Home care 0.4 0.4 0.4 0.3 0.3 0.2 0.5 Source: NCDC, 2016. 68 Hospitals Hospitals are predominantly privately owned. A small number of single-profile hospitals (e.g. for TB) have remained in public ownership. Hospitals range in size (Annex Table 7). The majority of hospitals operate with fewer than 100 beds. Payment for hospital care is case-based and payment rules vary depending on provider characteristics and type of care provided (WHO, 2016). The general rule is that if a provider participated in MIP, the SSA tariff would not exceed that paid under MIP by 10 percent. However, new providers can submit their own prices, which has led some legal entities to close and open as a new entity in order to charge a higher price. There are two categories of emergency care – urgent and non-urgent, as well as a separate category, critical and intensive care, the tariffs for which are calculated differently. The maximum tariff the SSA will pay for each case is calculated monthly based on prices submitted by providers. Data from the preceding month is used in each case. Annex Table 7. Number of Hospitals and Hospital Beds by Regions, 2015 Beds per 100000   Number of Hospitals Number of Beds Population Adjara 19 1160 345.5 Tbilisi 131 6465 581.9 Kakheti 16 463 145.4 Imereti 30 1843 345.6 Samehrelo-Zemo Svaneti 20 580 175.7 Shida Kartli 10 439 166.5 Kvemo Kartli 18 718 168.8 Guria 6 127 112.3 Samtskhe-Javakheti 9 397 247.2 Mtskheta-Mtlianeti 4 105 111.2 Racha-Lechkhumi and Kvemo svaneti 4 90 283.9 Ministry of Defence 1 174 Ministry of Corrections 2 269 Georgia 270 12830 345.2 Referral System Primary care has been designated as the first point of contact in the health. A referral from a primary care provider is needed to secure an appointment with a specialist or go to hospital. In practice, patients can see a specialist without a referral by paying a higher fee. 69 Annex 5. The Role of Private Health Insurance in Georgia No country relies on PHI to provide health coverage for the majority of the population due to well- established market failures associated with voluntary coverage and due to the substantial regulatory capacity and high transaction costs involved in addressing these failures (Hsiao W. C., 1995). For similar reasons, very few countries allow public or private insurers to compete with each other to offer publicly financed health coverage. Georgia abolished this option in 2013. The international experience suggests that effective competition among multiple public or private insurers is technically much more complex and demanding for government than having a single purchaser (Thomson, Busse, Crivelli, van de Ven, & Van de Voorde, 2013); (van de Ven, et al., 2013); (van Ginneken, Swartz, & Van der Wees, 2013). Purchaser competition is no guarantee of stronger performance and often results in higher administrative costs. Many countries do, however, allow PHI to play a limited role alongside publicly financed health coverage. In Europe, for example, PVHI may cover (Sagan and Thomson 2016):  people not eligible for publicly financed health care – a very small and shrinking number of countries  services excluded from publicly financed health coverage – mainly dental care for adults  co-payments for publicly financed health care – a small number of countries  people wanting to benefit from access to faster treatment, better amenities in hospital or privately provided services – most countries, with markets tending to be smaller in poorer countries What Role Does PHI Play in Georgia? PHI mainly offers people access to services that are not covered by the UHC program (e.g. some outpatient drugs and dental care). Since 2013, the number of people covered by PHI has declined sharply (Annex Table 9). In 2015, half a million people (14 percent of the population) had PHI coverage, mostly through corporate policies provided by employers. This includes around 214,000 state employees (about 6% of the population) who have PVHI paid for by the Ministry of Defense and the Ministry of Internal Affairs in addition to being covered by the UHC program. The MOLHSA estimates that about 40 percent of all PVHI beneficiaries are eligible for the UHC Program because they belong to an entitled group (pensioners, for example). The MOLHSA also estimates that demand for PVHI has risen since 2015 as people look for ways to cover expenses not covered by the UHC Program (WHO, 2016). Annex Table 8. Number of People Covered by Private Insurance Companies, 2013-2015 VHI beneficiaries 2013 2014 2015 Government coverage (Ministry of defence and Ministry of 1,467,453 975,810 214,197 Internal Affairs) Voluntarily privately insured 491,885 535,505 308,440 Total 1,959,338 1,511,315 522,637 Source: MOLHSA, 2016 70 What is the Evidence on PVHI as a Health Financing Mechanism Globally? PHI accounts for a relatively small share of private health expenditures globally: only 15 percent in 2012, compared to 77% from household out-of-pocket payments (OOPs). PVHI does not contribute much to private spending in most countries in Europe either (Annex Figure). The extent to which PVHI helps achieve key health system goals varies depending on the role it plays in relation to publicly financed health coverage and on the regulatory environment and regulatory capacity. In Europe, PVHI is systematically biased in favour of better off people (Sagan & Thomson, 2016). It is generally associated with inequitable access to health care, does not contain costs or increase efficiency, and can undermine the financial stability of publicly financed health coverage (Mossialos & Thomson, 2002) (Thomson & Mossialos, 2004). These challenges to health system performance tend to grow as the role of PVHI grows. They are also greater in contexts where boundaries between PVHI and publicly financed health coverage are not enforced, where capacity for careful oversight of the PVHI market is lacking and where PVHI is subsidised by the state. Countries such as Australia and Chile have had similar experiences with PVHI (Armstrong, Paolucci, McLeod, & van de Ven, 2010) (Hall, Lourenco, & Viney, 1999) (Sapelli, 2004). Finally, if PVHI is regarded as a means to reduce OOP spending, the evidence contradicts this, as shown in Annex Figure for Europe. Careful oversight is needed to ensure PVHI is aligned with health system goals and does not undermine health system performance. The transaction costs involved can be high and regulation cannot correct all shortcomings (Woolhandler, Campbell, & Himmelstein, 2003) (Hsiao, 1995). In addition to regulation, public subsidies are often needed to ensure PVHI is affordable for those who need it. However, such subsidies do not represent an efficient or equitable use of public money because they tend to favour better off people. Annex Figure 2. VHI Share of Private Health Spending in the European Region, 2014 80 60 % 40 20 0 Source: Thomson (2016) using NHA data from the WHO Global Health Expenditure Database Note: national estimates suggest PHI accounted for 10% of private spending on health in 2014 71 Annex Figure 3. Relationship between PVHI and OOP Share of Total Health Spending, 2012 R2 = 0.05 VHI % Out-of-pocket payments % Source: (Thomson S. , 2016) using NHA data from the WHO Global Health Expenditure Database 72 5A, Nino Ramishvili St. Tel.: (995 32) 291 30 96 www.worldbank.org/georgia World Bank Europe and Central Asia