Navigating Fiscal Realities for Equitable Growth in Georgia © 2025 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Website: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, links/footnotes and other information shown in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this publication is subject to copyright. Because the World Bank Group encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution — Please cite the work as follows: World Bank. 2025. Navigating Fiscal Realities for Equitable Growth in Georgia. Washington, DC: World Bank Group. Translations — If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations — If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content — The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. Any queries on rights and licenses, including subsidiary rights, should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; email: pubrights@worldbank.org Cover and other pictures: Designed and generated by Team Design at Midjourney. Report and Cover design: Team Design. ii Contents Abbreviations vii Acknowledgments viii EXECUTIVE SUMMARY 2 Policy Context 2 Key Findings of the Fiscal Incidence Analysis 4 Policy Implications 8 CHAPTER ONE: Impact of Fiscal Policy on Poverty and Inequality 12 1.1. Fiscal Policy Context 13 1.2. Net Payers and Receivers of the Fiscal System 17 1.3. Impact of Fiscal Policy on Poverty 18 1.4. Inequality Impact of Fiscal Policy 21 1.5. Progressivity of the Fiscal System 25 1.6. Marginal Contributions to Inequality and Poverty Reduction 26 1.7. Cost Effectiveness of Fiscal Interventions 27 1.8. Incidence of Fiscal Interventions 28 1.9. Limitations of the Analysis 29 CHAPTER TWO: Equity and Gender Aspects of Fiscal Policy Reforms 32 2.1. Impact of Personal Income Tax 32 2.2. Impact of Indirect Taxes 34 2.3. Pensions and Redistribution 35 2.4. Indirect Subsidies 36 2.5. Education Spending 39 2.6. Gender Equity in Fiscal Policy 43 Conclusions and Key Recommendations 57 References 59 APPENDIX A: Data and Methodology 62 APPENDIX B: Sensitivity Analysis on Pensions: Pensions as Deferred Income Scenario 77 APPENDIX C: Sensitivity Analysis on Model Assumptions 82 iii Figures Figure ES.1 Net cash and fiscal position of households after taxes and transfers 4 Figure ES.2 Marginal contribution of each fiscal instrument, measured at consumable income 6 Figure ES.3 Cost effectiveness of transfers, at consumable income 7 Figure 1.1 CEQ income concepts in the Pensions as Government Transfers scenario for Georgia 16 Figure 1.2 Net fiscal position of households after taxes and transfers 18 Figure 1.3 Poverty headcount from market to consumable income 18 Figure 1.4 Poverty headcount from market to consumable income by region 19 Figure 1.5 Poverty headcount from market to consumable income by household category 20 Figure 1.6 Poverty impact of fiscal policy in UMIC and ECA countries (percentage points) 21 Figure 1.7 Impact of fiscal policy on inequality, measured by the Gini coefficient (PGT scenario) 22 Figure 1.8 Impact of fiscal policy on inequality (measured by the Gini coefficient) by region 22 Figure 1.9 Impact of fiscal policy on inequality (measured by the Theil index) 23 Figure 1.10 Inequality from market income to final income by household category 23 Figure 1.11 Redistributive impact of fiscal policy – SDG 10.4.2 24 Figure 1.12 Progressivity of selected fiscal policy instruments, Georgia 2022 25 Figure 1.13 Marginal contribution to inequality reduction of each fiscal instrument, in Gini points, measured at consumable income 26 Figure 1.14 Marginal contribution to poverty reduction of each fiscal instrument, in percentage points, measured at consumable income 27 Figure 1.15 Cost effectiveness of taxes and transfers, at consumable income 28 Figure 1.16 Absolute incidence: shares of total taxes and benefits paid/received by households, ranked by market income 29 Figure 2.1 Incidence of personal income tax 33 Figure 2.2 Personal income tax scenarios 34 Figure 2.3 Incidence of indirect taxes and consumption 34 Figure 2.4 Headcount poverty before and after pensions and TSA, by age groups 35 Figure 2.5 Incidence of indirect subsidies, by quintile 36 Figure 2.6 Incidence of indirect subsidies, by type and by quintile 37 Figure 2.7 Net enrollment rates 40 Figure 2.8 Incidence and concentration share of education spending 41 Figure 2.9 Marginal contribution of education spending to inequality reduction, by level 41 Figure 2.10 Concentration of education spending, by quintile, by region 42 iv Figure 2.11 Population pyramid for Georgia, 2024 44 Figure 2.12 Interaction of the fiscal system with gender and intrahousehold dynamics 45 Figure 2.13 Household typologies by region 48 Figure 2.14 Household typologies, by household care categories 49 Figure 2.15 Poverty and inequality by household categories 50 Figure 2.16 Health expenditure per capita by gender and age group, 2022 51 Figure 2.17 Female labor force participation 52 Figure 2.18 Distribution of individuals that are unemployed, by reason (percent), Georgia 2022 52 Figure 2.19 Gender gap by income concept 54 Figure 2.20 Gender gap by income concept, urban versus rural areas 54 Figure 2.21 Gender gap by income concept by quintile 54 Figure A.1 CEQ income concepts in the Pensions as Government Transfers scenario for Georgia 64 Tables Table 1.1 Georgia’s government revenue, 2022 14 Table 1.2 Georgia’s government expenditure, 2022 15 Table 2.1 Household typologies 46 Table 2.2 Household typologies in Georgia 46 T2.2 Panel A Share of household type in income groups and location 46 T2.2 Panel B Distribution of household types across income and location 47 Table 2.3 Pre-fiscal poverty, by household typologies 48 Table 2.4 Employment by age groups and sex (percent) in Georgia, 2022 53 Table A.1 Regional effective VAT rates (percentages) 71 Table A.2 Per capita value of education benefit (GEL) 75 Table B.1 Impact on inequality and poverty 77 Table B.2 Marginal contributions of each fiscal intervention to inequality and poverty (for disposable, consumable, and final income) 78 Table C.1 Regression results of probit 83 Table C.2 Results of the sensitivity analysis on macrovalidation 84 Table C.3 Results of the sensitivity analysis on poverty and inequality 85 v Boxes Box 1.1 Gaps in consumption captured in the Georgia HIES survey 30 Box 2.1 Water and sanitation spending in Georgia 38 Box A.1 Tax and transfer allocation methods 63 vi Abbreviations CEQ Commitment to Equity MC Marginal Contribution CIT Corporate Income Tax MoES Ministry of Education, Science COVID-19 Coronavirus Disease 2019 and Youth ECA Europe and Central Asia MoF Ministry of Finance EU European Union MoILHSA Ministry of Internally Displaced People from Occupied Territories, E-CEQ Engendered Commitment to Equity Labor, Health, and Social Affairs EUROMOD Tax-Benefit Microsimulation NBG National Bank of Georgia Model for the European Union PDI Pensions as Deferred Income FIA Fiscal Incidence Analysis PFM Public Financial Management GDP Gross Domestic Product PGT Pensions as Government Transfers GEL Georgian Lari PIT Personal Income Tax GeoStat National Statistics Office of Georgia PMT Proxy Means Test GGGI Global Gender Gap Index PPP Purchasing Power Parity HIES Household Income and SDG Sustainable Development Goal Expenditure Survey SVHD Socially Vulnerable Household IDP Internally Displaced People Database IMF International Monetary Fund TSA Targeted Social Assistance IO Input-Output Matrix UHCP Universal Health Care Program IPL International Poverty Line UMIC Upper-Middle-Income Country KI Kakwani Index UN United Nations LFS Labor Force Survey VAT Value Added Tax LMIC Lower-Middle-Income Country WB World Bank Vice President Antonella Bassani Country Director Rolande Simone Pryce Regional Director Asad Alam Global Director Luis-Felipe Lopez-Calva Practice Manager Ambar Narayan Task Leaders Saida Ismailakhunova vii Acknowledgments This report has been prepared under the Georgia Poverty Reduction program by Saida Ismailakhunova (Senior Economist, TTL, EECPV), Beenish Amjad (ETC, EECPV), Maynor Cabrera (Consultant, EECPV), and Ken Simler (Consultant, EECPV). The team is thankful to Alan Fuchs (Lead Economist, EMNPV) and Mariano Ernesto Sosa (ETC, EPVGE) for previously conducted work. The financial support provided by Sweden is highly appreciated. The team acknowledges the valuable support of GIZ for conducting the gender analysis. The team appreciates the insightful advice and guidance provided by the Peer Reviewers: Matthew Wai-Poi (Lead Economist, EEAPV), Mariam Dolidze (Senior Economist, EECM1), and Yeon Soo (Senior Economist, EPVGE). The report was prepared under the guidance of Rolande Simone Pryce (Country Director, ECCSC) and Ambar Narayan (Practice Manager, EECPV). The team would like to express special gratitude for the advice and guidance provided by Miguel Eduardo Sanchez Martin (EFI Program Leader for the South Caucasus, ECCDR) and Obert Pimhidzai (Lead Economist, EECPV). Administrative support was provided by Armanda Carcani (Program Assistant, EECPV) and Tatuli Gongadze (ETT, ECCGE). The analysis has been conducted in close coordination with government counterparts and development partners, with earlier presentations and workshops used to support a dialogue on poverty and welfare issues faced by Georgia. The team is grateful to the Ministry of Finance and National Statistics Office of Georgia (GeoStat) for data that was drawn upon for the analysis. Any errors and omissions remain the responsibility of the authors. viii Executive Summary Policy Context | Key Findings of the Fiscal Incidence Analysis | Policy Implications | 1 Executive Summary Policy Context Georgia has made impressive economic progress While Georgia’s record of growth and poverty over the past two decades, achieving strong and reduction is commendable, substantial resilient economic growth and development in challenges remain. As of 2022, almost half of the the face of numerous external and domestic population (47.7 percent) was living below the shocks. Through sound macroeconomic UMIC international poverty line of US$6.85 management, regulatory reforms, and (2017 PPP) per day, while 15 percent was below strengthened institutions, Georgia graduated to the lower-middle-income country (LMIC) line of upper-middle-income country (UMIC) status in US$3.65 per day. Progress on reducing income 2015 and now aspires to high-income country inequality has been uneven, with the Gini index status and accession to the European Union (EU). increasing from 36.2 in 2004 to 39.6 in 2011 before From 2014 to 2023, Georgia’s average annual gross falling back to 33.5 in 2022. 1 Rural poverty and domestic product (GDP) growth of 4.8 percent inequality are considerably higher than in urban areas, while rural residents also have poorer surpassed the average among countries in the access to public services such as healthcare, Europe and Central Asia (ECA) region and schools, water, and electricity. Gender inequality UMICs. In recent years, this robust GDP growth is also pronounced: despite being more educated has been complemented by social reforms in than men on average, women are much less likely labor policy, healthcare, and pensions. to be engaged in the formal labor market, in large part because they shoulder greater Recent fiscal policy in Georgia has been responsibility for caring for children and the characterized by moderate deficits and low debt elderly. Women with formal employment tend to levels. Although Georgia’s ratio of revenue to earn less than men, and women are less likely to GDP trails that of peer countries, its tax-revenue- start a business. to-GDP ratio of 24 percent compares favorably to ECA and UMIC averages. As a share of GDP, To meet these challenges and continue to build public expenditure in Georgia was 31 percent in on its development momentum, Georgia must 2022, among the lowest in the ECA region, yet it also strategize for demographic change. One of compares well with UMICs and structural peers. the most important challenges is the financing of Although the ratio of total public expenditure to the social protection system so that it can GDP is relatively low, Georgia allocates a larger sustainably meet the needs of an aging share of public expenditure to social assistance, population. Georgia’s old-age dependency ratio is fiscal subsidies, and capital expenditures than the 23 percent and will increase as the share of the averages in the ECA region and among structural elderly population grows and the share of and aspirational peers. working-age people declines. This will strain the pension system, which has played a dominant 1 Poverty and inequality statistics from Geostat survey data on World Bank Poverty and Inequality Platform https://pip.worldbank.org/country-profiles/GEO 2 role in reducing income inequality and keeping This Fiscal Incidence Analysis (FIA) report is households from falling into poverty, even designed to serve as a baseline for the though Georgia’s pension replacement rate of evaluation of current fiscal policy and potential 16.5–18.5 percent is the lowest in the region. The policy reforms to enhance fiscal equity and aging population will have broader impacts on sustainability. The Commitment to Equity public expenditures, especially health spending as (CEQ) methodology helps in understanding the burden of noncommunicable chronic diseases better which population subgroups benefit the associated with aging grows rapidly. most from public expenditures and which subgroups are paying the taxes that finance those To finance the social spending needed to sustain expenditures. It traces changes in the improvements in living standards, it is distribution of household incomes step by step necessary to further strengthen the tax base and from market income (before taxes and transfers) generate more fiscal space. This can be achieved to final income and therefore identifies changes by improving both domestic revenue in income inequality and poverty attributable to mobilization and the efficiency of public different fiscal policy interventions such as taxes, spending. The additional fiscal space will permit transfers, and social expenditures. The equity expansion of social expenditures and targeted aspect of fiscal policy is critical to informing social assistance (TSA) transfers to enhance fiscal Sustainable Development Goals Indicator 10.4.2, which measures the redistributive impact of equity across income classes and regions. In fiscal policy. The FIA helps identify spending particular, there is scope for continued priorities based on their contributions to rationalization of value added tax (VAT) reducing poverty and inequality and examines exemptions and for improvements in the VAT’s how fiscal space can be created without efficiency and transparency, in line with EU increasing the burden on poor and vulnerable standards and international good practices. households. These tax reform areas align with the policy recommendations in the World Bank’s Public Finance Review for Georgia 2024. Public expenditure efficiency reforms — such as moving from generalized subsidies in the water, gas, and transport sectors to targeted social transfers — could help improve fiscal sustainability. Spending efficiency initiatives would serve to reinforce higher revenue collection. Furthermore, expenditure reforms to improve the accessibility and quality of public health and education services could have long- term intergenerational impacts in reducing poverty, inequality, and spatial disparities. In turn, enhanced health and education services will provide the necessary investment in human capital that will support Georgia’s future inclusive growth and prosperity. 3 Key Findings of the Fiscal only the top 10 percent are net payers. Because of the large roles of old-age pensions and public Incidence Analysis health expenditure (orange and green bars, respectively), the redistributive impact of fiscal policies is strongest in households that comprise Overall, Georgia’s fiscal policies for the year 2022 elderly members but no children. For pensions, were progressive, serving to reduce income these impacts are concentrated in the poorer inequality and poverty. The left panel of Figure 0.1 quintiles, whereas health benefits are spread quite summarizes how tax and social spending increase evenly across the income quintiles. or reduce aggregate household income at different points in the income distribution. The stacked bars The poorest quintile of the Georgian population has show each of the main fiscal instruments, and the extremely low market income, making them highly dashed lines show the net effects of taxes and dependent on government transfers. The right transfers on household incomes. The red dashed panel of Figure ES.1 displays taxes and benefits as a line shows the aggregate value of cash-based share of market income by income quintile. In the interventions such as pensions, transfers, and poorest quintile, the final incomes are about seven subsidies minus taxes paid for each quintile, which times larger than market incomes, with almost half is the net cash position. That line crosses the of that difference attributable to non-contributory horizontal axis between the third and fourth pensions. Direct transfers also contribute quintile, meaning that the poorest 60 percent of significantly to lifting incomes among the bottom 20 the population are net beneficiaries of the tax and percent. The impact is much more muted in the transfers system, while the richer 40 percent are richer quintiles because the taxes paid and benefits net payers. The gray dashed line shows the same received are relatively small fractions of market calculations with non-cash benefits such as public income, compared to the multiples of market health and education services included. In this case, the bottom 90 percent are net receivers and income observed in the poorest quintiles. Figure ES.1: Net cash and fiscal position of households after taxes and transfers Net fiscal position, by quintile (GEL million) Net fiscal position, by quintile (percent of market income) Direct transfers Pensions Direct transfers Pensions 3000 Direct taxes Subsidies Direct taxes Subsidies 800 Indirect taxes Health Indirect taxes Health Education Education 2000 Net cash Fiscal position 600 Net cash Fiscal position GEL, million 1000 400 Percent 0 200 -1000 0 -2000 -200 1 2 3 4 5 1 2 3 4 5 Per capita market income quintiles Per capita market income quintiles Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: The figure divides the population into ten equal-sized groups (deciles) ranked by market income (income before taxes and transfers, or pre-fiscal income). For each decile, the stacked bars show the amount by which taxes reduce household income (below the horizontal axis) and public spending increases income (above the horizontal axis). The net cash position, shown by the dashed red line, illustrates the aggregate sum of all cash-based interventions (all taxes, direct transfers, and indirect subsidies) for each decile. The net fiscal position is the aggregate sum of the cash-based interventions plus in-kind benefits from public expenditure such as health and education services and is shown by the dashed gray line. If the line is above the horizontal axis, that decile is a net beneficiary of the tax and transfer system, and a lines below the horizontal axis indicate net payers into the system. 4 Most direct transfers and most in-kind health and Gini index for market income was 44.7 but only education benefits are progressive relative to 24.6 for final income, which accounts for the consumable income, indicating the role of these redistributive impact of cash and in-kind transfers, interventions in reducing income inequality. pensions, subsidies, and taxes. The magnitude of The incidence of indirect subsidies is more equally inequality reduction attributable to fiscal policies distributed than market incomes, but their in Georgia is among the highest observed in the marginal contribution to inequality reduction is ECA region and among UMICs globally. For small. In absolute terms, most of the subsidy comparison, taxes and transfers (including in-kind benefits accrue to richer households. The incidence benefits) in UMICs reduce income inequality by an of indirect taxes (VAT and excise) contributes to average of 8.7 Gini points, versus 20.1 points in an increase in inequality, while the incidence of Georgia. On average, taxes and transfers in UMICs personal income tax (PIT) makes only a small have zero impact on poverty reduction, compared contribution to inequality reduction. to a reduction of 17 percentage points in Georgia. The CEQ analysis of Georgia’s fiscal policy — Most of the inequality-reducing impact of the particularly social expenditures, which include current fiscal system comes from public spending, publicly funded pensions — finds that fiscal policy with the largest reductions in poverty and has contributed to large reductions in absolute inequality coming from non-contributory old-age poverty and income inequality. While the pace of pensions. Old-age pensions reduce income poverty and inequality reduction in Georgia has inequality by 7.8 Gini points and the poverty slowed in recent years, the analysis shows that as of headcount by 18.6 percentage points (Figure ES.2). 2 2022, the taxes and transfers analyzed reduced the The corresponding figures for the vulnerable share of Georgians living below the official families transfer are 3.0 Gini points for inequality national poverty line from 32.2 percent to 15.2 and 5.4 percentage points for poverty. Public percent (from pre-fiscal to consumable income). health spending accounts for 4.3 Gini points of Transfers and other social expenditures also inequality reduction, while general education reduced inequality significantly. As of 2022, the accounts for a reduction of 3.2 Gini points. 2 These are the marginal contributions of the tax and public held constant. These graphs highlight a subset of taxes and expenditure regime — that is, the individual impact of the spending that have the largest positive or negative marginal policy intervention on inequality and poverty reduction when contributions. the impacts of all other fiscal interventions in the model are 5 Figure ES.2: Marginal contribution of each fiscal instrument, measured at consumable income a) to inequality reduction, in Gini points* b) to poverty reduction, in percentage points** Transport subsidy 0,0 Water subsidy 0,0 Value added tax (VAT) -3,4 Education—vocational 0,0 Poverty reducing Value added tax (VAT) 0,0 Inequality reducing Property tax 0,0 Personal income tax (PIT) -2,5 Excises 0,0 Electricity subsidy 0,0 Loss of breadwinner transfers 0,1 Gov. contributions to pensions Gas subsidy 1,9 0,1 Education—tertiary 0,1 IDP transfers 0,2 Other transfers 0,3 Other transfers 2,9 Personal income tax (PIT) 0,4 Gas subsidy 0,8 Education—preschool 0,9 Targeted social allowance (TSA) 5,4 Targeted social allowance (TSA) 3,0 Education—general 3,2 Health 4,3 Old-age pensions 18,6 Old-age pensions 7,8 0 2 4 6 8 10 -10 -5 0 5 10 15 20 Marginal effect on inequality (Gini points) Marginal effect on poverty (percentage points) Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: IDP transfers means internally displaced persons transfers *Tertiary education, Government contributions to pensions, loss of breadwinner transfers, electricity subsidy, excises, property tax, VAT, vocational education, water subsidy, and transport subsidy have zero marginal effects on redistribution and were not included in the chart. **Excises, property tax, loss of breadwinner transfers, transport subsidy, electricity subsidy, government contributions to pensions, and internally displaced people (IDP) transfers do not have significant marginal contributions changing poverty levels and were not included in the chart. Old-age pensions currently make the largest internally displaced people (IDP) transfers, and marginal contribution to reducing poverty and electricity subsidies. TSA transfers are also the inequality, but TSA transfers are more cost most cost effective for reducing inequality, effective. Looking forward, more consideration followed by loss of breadwinner transfers, with should be given to expanding spending on those the lowest impact per unit on inequality found for interventions that have a greater poverty- and transport subsidies and government contributions inequality-reducing impact per Lari spent than to pensions. This analysis also shows that the other public spending. Figure ES.3 compares the dominant role of pensions in reducing poverty cost effectiveness 3 of transfers and subsidies, and inequality seen in Figure ES.2 is attributable showing that TSA transfers have more than primarily to the higher budget allocation to double the impact on poverty and inequality than pensions and secondarily to the cost effectiveness pensions per unit of expenditure. Pensions have of pensions, which is about average among the the same effect on poverty reduction per unit of interventions analyzed. expenditure as loss of breadwinner transfers, 3 Cost effectiveness is calculated by dividing marginal can be accomplished for each unit of currency spent on an contributions by the budget amount captured in the survey. instrument, such as GEL or a point of GDP. Essentially, this shows how much Gini or poverty reduction 6 Figure ES.3: Cost effectiveness of transfers, at consumable income TSA transfers 7 12 Loss of breadwinner transfers 3 7 Pensions 2 5 IDP transfers 2 5 Electricity subsidy 1 5 Gas subsidy 1 4 Transport subsidy 3 Gov. contributions to pensions 0 1 Water subsidy 2 0 0 2 4 6 8 10 12 Inequality / poverty cost effectiveness by unit of GDP Inequality Poverty Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: The cost effectiveness index measures how well a fiscal policy instrument reduces inequality or poverty relative to the budget allocated as a share of the GDP. The higher the value, the greater the cost-effectiveness of the fiscal policy measure. Low spending levels on non-pension transfers have little impact on reducing inequality. Indirect have hindered fiscal policy’s contribution to subsidies are untargeted, and a disproportionate reducing poverty among the large share of share of the absolute benefits flow to richer households that do not include an elder. At households. The two richest income deciles receive consumable income (including public pension 47 percent of total indirect subsidies compared to and other transfers), the poverty rate for 7.4 percent for the poorest two deciles, with the households with both elders and children is 17 domestic transport and gas subsidies primarily percent, and only 7 percent for households with responsible for the skewed distribution favoring elders but no children. At market income those richer households. Indirect subsidies also show an rates would be a little higher than 40 percent. In urban bias, as urban areas capture almost 99 percentage points, that is more than triple the percent of the transport and water subsidies. limited poverty reduction experienced for households with children but no elders (29 On the revenue side, the current tax system has percent at market income to 22 percent at little effect on income inequality and accounts consumable income) and households with neither for an increase in the poverty headcount. elders nor children (15 percent at market income Notably, both direct and indirect taxes in to 13 percent at consumable income). Higher Georgia have a poverty-increasing effect, with poverty rates among households with children, the VAT and personal income tax (PIT) coupled with low spending on education, are estimated to increase the poverty headcount ratio barriers to human capital accumulation and work by 3.4 percentage points and 2.5 percentage against Georgia’s need to improve labor force points, respectively. In other countries, the PIT participation and productivity. impact on poverty is often neutral because incomes below the poverty line or slightly above Taken together, indirect subsidies — most notably the poverty line are typically below the minimum those on water, electricity, transport, and gas — income threshold for paying PIT. 7 Although fiscal policies have helped narrow the disaggregated by gender to examine the divide between rural and urban areas, the differential impact of fiscal policies on women regional differences in poverty and inequality and men. The analysis showed that in urban areas remain large. When estimating poverty using the tax and transfers system reverses the gender market income, 40.8 percent of the rural income gap, such that women are favored in population lives below the poverty line, terms of post-fiscal income. However, in rural compared to 26.8 percent of the urban areas fiscal policy interventions have less impact population. After taxes and transfers, the rural on gender income inequality, largely because of poverty rate falls to 20.2 percent, while urban rural women’s lower engagement in formal poverty is reduced to 11.9 percent. A similar employment. Furthermore, fiscal policies albeit more subtle difference is observed for targeted at children are not sufficient to offset inequality, with fiscal policies — especially direct the high incidence of poverty among households cash transfers and in-kind health transfers — with children. The results strongly indicate that reducing the rural Gini index from 49.0 to 26.1, childcare and elderly care constrain mostly while the urban Gini index falls from 41.4 to 23.3. women’s participation in economic life. Women experience “time poverty” linked to the burden of Fiscal policy is also reducing the gender gap, 4 unpaid domestic responsibilities and little or no but only in urban areas. The FIA was use of childcare services. Policy Implications those services, while eliminating spatial disparities in both access and quality. • Increase the overall amount spent on The FIA provides useful guidelines for a coherent transfers and shift the spending allocation set of fiscal policies that will better position away from untargeted subsidies and more Georgia to overcome demographic and other toward targeted social assistance, especially challenges to achieve a more prosperous, for the non-elderly population. equitable, and sustainable growth trajectory. • Adopt measures to facilitate greater While Georgia’s progress to date has been participation of women in economic life as admirable, strategic adjustments are needed to workers or business owners to address pressing challenges such as an aging counterbalance the shrinking working-age population, a shrinking labor force, stagnant population and foster gender equity. productivity growth, and persistent poverty and inequalities in opportunity and outcomes. The • Implement reforms of the pension system to core components of the recommended fiscal ensure its sustainability as the elderly share reforms may be summarized as follows: of the population continues to grow. • Enact fiscal reforms to create more fiscal space • Promote greater equality of opportunity by to finance these initiatives. Among other increasing investment in human capital things, this includes VAT reforms, through scaling up funding for education and rationalization of subsidies, and improvements healthcare and improving the quality of in public expenditure efficiency. 4 The gender gap is calculated by classifying households as Gender gap = (average female-majority household income – male-majority or female-majority (or “no majority” in the case average income of male-majority) / average income of male- of equal numbers of males and females, and then computing: majority household 8 Over the longer term, sustained reductions in 40 percent of women 5 are not engaged in the labor inequality — both between and within rural and market, despite having higher college and urban areas — depend on promoting greater university educational attainment than their male equality of opportunity. This depends in turn on counterparts. Among women in the labor force, a ensuring equitable investments in human capital, large share is in informal employment. The main especially in health services and education. The reasons for lower participation are lower female analysis shows that although education spending is wages and women’s greater responsibilities for progressive, Georgia needs to make its education unpaid domestic work, childcare, and elder care. system more effective in preparing students for an The estimated economic cost of the gender gap in active labor market by improving the quality of labor force participation is 11 percent of GDP, thus education and implementing policies and measures to close the gap are pro-growth, programs that ensure fair access to quality especially in the context of a shrinking working- education across all regions of the country. age population. Policies such as additional Investing in human capital is a prerequisite for the subsidies for childcare services and wider provision labor productivity growth that is needed for of flexible work arrangements can help boost Georgia to continue to improve living standards in women’s participation in economic life. the face of an aging population. To ensure that the human capital investments pay off, measures are Continuing pension reforms toward a funded needed to address high unemployment (especially pension system is crucial to alleviate the fiscal among the youth), encourage female employment, pressures associated with an aging population. and decrease labor informality. These are good for Ensuring the long-term fiscal sustainability of the economic growth and could improve the worker- government-funded pension scheme will require to-retiree ratio and contribute to improved fiscal structural changes — especially in the labor market performance. — that could counteract the negative effects of population aging. Changes in the retirement age — To address persistent poverty and inequalities, gradually phased in over many years for the more needs to be spent on social policies, and it current working-age population — could also help needs to be spent more efficiently. Georgia’s social on this front, as would measures to increase the spending trails that of its economic and regional fertility rate to reshape current demographic peers and is much lower than spending in the EU trends. These measures will contribute to ensuring Member States. The FIA reveals that TSA is the fiscal sustainability, and need to be reinforced by most cost-effective mechanism for reducing reforms that improve the efficiency of social poverty and inequality in Georgia, but its impact is spending. relatively small because the budgetary outlay as a share of total outlay is relatively small. Clearly, To finance these pro-growth and pro-equity policy old-age pensions and targeted social assistance are measures in a sustainable manner, both important pillars of Georgia’s social safety complementary fiscal reforms are needed to net, but some rebalancing is needed to improve increase fiscal space. Relevant policies include poverty and inequality outcomes as well as the measures to increase domestic revenue efficiency of public spending. mobilization as well as to improve public expenditure efficiency. Enhanced revenue Policies to increase women’s labor force mobilization would support the dual objectives of participation are needed to sustain economic improved fiscal sustainability and greater fiscal growth and achieve greater gender equity. Nearly equity if additional revenues are used to 5 https://data.worldbank.org/indicator/SL.TLF.ACTI.FE.ZS?locations=GE 9 compensate targeted households with direct Finally, the design and implementation of fiscal transfers. A prime example is VAT exemptions policies should continue to consider different that are intended to alleviate the burden on poorer dimensions including fiscal sustainability, households, but their broad nature means that in economic efficiency, equity, and externalities. absolute terms, the benefits are concentrated Enhanced fiscal equity considerations matter among richer households. Further rationalization because economic growth alone is not sufficient to of VAT exemptions can bring in sizeable revenues eliminate poverty and inequality. The fiscal system that are currently foregone, which could in turn be can support this process through the right mix of used to bolster spending on transfers that investments and redistribution via taxation and contribute more to poverty and inequality social expenditures. As the country explores the reduction than the VAT exemptions do. There is mix of taxation and public expenditure policies to scope for increasing revenue by raising the health ensure fiscal sustainability, it is important that the taxes on alcohol and tobacco products, which are design and implementation of reforms maintain a currently lower in Georgia than in peer countries. focus on equity. Rationalization of indirect subsidies is a critical step for enhancing fiscal efficiency and sustainability while also promoting greater equity. The current subsidy regime, which is largely untargeted, represents a significant fiscal burden that contributes nothing to fostering equity or reducing poverty and inequality. Furthermore, some indirect subsidies introduce price distortions that encourage overconsumption of items such as fossil fuels, water, and other natural resources. Evaluating the effectiveness of subsidies in achieving poverty reduction and redistribution objectives is essential. Rationalizing subsidy allocations to prioritize areas with high social and economic impact will help ensure fiscal sustainability and efficiency. 10 C H A P T E R O N E Impact of Fiscal Policy on Poverty and Inequality Fiscal Policy Context | Net Payers and Receivers of the Fiscal System | Impact of Fiscal Policy on Poverty | Inequality Impact of Fiscal Policy | Progressivity of the Fiscal System | Marginal Contributions to Inequality and Poverty Reduction | Cost Effectiveness of Fiscal Interventions | Incidence of Fiscal Interventions | Limitations of the Analysis | 11 C H A P T E R O N E Impact of Fiscal Policy on Poverty and Inequality Georgia’s fiscal performance has been strong, This chapter examines Georgia’s fiscal policy resulting in the country’s swift transition from and its impact on poverty and inequality. Fiscal lower-middle-income country (LMIC) status policy can be instrumental in reducing poverty upon independence in 1991 to upper-middle- and inequality. Fiscal incidence analysis (FIA) is a income country (UMIC) status in 2018. Over the method to allocate the burden of taxes and the past two decades Georgia has exhibited monetized value of government expenditures to remarkable resilience and stability, with per estimate the incidence of taxes and benefits and capita growth rates above the averages for both their impact on inequality and poverty. This FIA UMIC and the Europe and Central Asia region. is based on the Commitment to Equity (CEQ) Fiscal policy has been prudently managed, methodology, 6 which is a well-established supported by a robust public finance framework for analyzing fiscal incidence. The management system and fiscal rules. analysis examines the impact of the government’s taxes, public social spending programs, and Although Georgia has made notable gains in subsidies on different income groups using data income growth and poverty reduction over the from the Household Integrated Economic Survey past decade, certain challenges remain. Despite (HIES) and macro-fiscal administrative records progress, Georgia faces significant infrastructure for fiscal year 2022. The CEQ approach serves as and human capital investment needs. While it a foundation for developing more complex aspires to become a high-income country and microsimulation models that can be used to accede to the European Union (EU), poverty analyze the potential distributional effects of remains high at 47 percent according to the future policy reforms. These models are UMIC international poverty line (IPL) and at 15 important for designing evidence-based fiscal percent according to the LMIC IPL. Inequality policies that prioritize equity throughout the rose in the 2000s and hovered around the World policy cycle, from the planning stage to Bank’s new high-inequality country threshold of implementation. For further details regarding the 40 (Gini) at the start of the 2010s but has since data and methodology used in this analysis, see fallen back to the mid-30s. The country is aging Appendix A. and faces an increasing fiscal burden from pensions, increased health spending, and an erosion of the tax base. 6 The CEQ methodology aims to address four key questions: inequality and poverty? What is the impact of fiscal reforms How much income redistribution and poverty reduction is that change the size and/or progressivity of a particular tax or being accomplished through fiscal policy? How equalizing and benefit? The methodology has been extended to and adapted pro-poor are specific taxes and government spending? How for over 85 low- and middle-income countries over the past effective are taxes and government spending in reducing decade. For methodological details, see Lustig (2018). 12 1.1. Fiscal Policy Context A strong macroeconomic framework levels are sustainable and, on the decline, leading to a contained level of deficit and spending. However, the low efficiency of public investments remains a Over the past two decades, Georgia has demonstrated robust economic growth despite challenge. Similarly, while Georgia’s spending levels facing significant challenges. This period has been on education and healthcare are relatively contained characterized by strengthened institutions, compared to its peers, there is scope to improve productivity-boosting deregulation initiatives, and their efficiency within the current system (Public sound monetary and fiscal policies fostering an Financial Review of Georgia, World Bank, environment of low inflation and public debt. Thanks forthcoming). to these factors, Georgia surpassed its peers in terms of per capita gross domestic product (GDP) growth over the past decade. From 2014 to 2023, Georgia’s Tax system overview average per capita growth rate of 4.8 percent outpaced the averages of both the UMIC grouping and the Georgia's revenue administration has shown Europe and Central Asia (ECA) region. The country stability and resilience, with taxes comprising much of its revenue mix. The revenue-to-GDP ratio has even maintained its economic growth in the face of significant challenges such as the 2008-armed has remained relatively stable over the years, conflict with Russia, 2008–09 global financial crisis, averaging 26 percent. Despite a revenue-to-GDP 2014–15 global oil price shock, coronavirus disease ratio below that of peers, Georgia’s tax collection 2019 (COVID-19) pandemic, and Russia's invasion of level is in fact above that of peer countries. The Ukraine since 2022. revenue mix is largely tax-based, with a nontax- revenue-to-GDP ratio of around 2.0–2.4 percent of The Government’s Strategy 2030 emphasizes robust GDP, and consists of profits paid by the National institutions, effective fiscal management systems, Bank of Georgia (NBG), interest and rental revenues and a skilled workforce to support the aspirations of from state property, penalties, and other non- the country. Amid heightened global uncertainty, a classified proceeds. prudent strategy is imperative to maintain fiscal discipline. The near-term outlook for Georgia hinges Tax revenue remained stable in fiscal year 2022, with on robust financial management, significant human notable contributions from taxes on goods and capital investment, and labor market development to services and the personal income tax (PIT). About facilitate growth. 13.3 percent of GDP stems from indirect taxes levied on consumption, production, and trade activities, with a major contribution from the value-added tax Fiscal policy overview (VAT) (10.3 percent of GDP), followed by excise taxes (1.9 percent of GDP) (Table 1.1). Collection of VAT While commendable in maintaining reasonable and excise taxes (on tobacco and alcohol) is lower than deficits and low debt levels, Georgia's fiscal policy that of peer countries. Within direct taxes, the PIT exhibits weaknesses in supporting the efficient use represents the largest source of revenues (7.0 percent of public resources and ensuring optimal outcomes of GDP). This fiscal incidence analysis (FIA) based on from budget allocations. Fiscal policy in Georgia is CEQ methodology captures an equivalent of 18.4 supported by a robust public finance management percent of direct taxes (PIT and property tax) and 16.8 system and fiscal rules. Overall, Georgia’s fiscal percent of indirect taxes (VAT and excises). Details on policy has been largely countercyclical. Public debt each tax are provided in Appendix B. 13 Table 1.1: Georgia’s government revenue, 2022 Government revenue Amounts Total (% of Included in the Captured in (GEL, millions) (GEL, millions) GDP) CEQ analysis CEQ (as % of (yes/no) budget) Total revenue and grants 19,377.6 26.8 n.a. 15.4 Revenue 19,111.2 26.4 n.a. 15.6 Tax revenue 17,386.0 24.1 yes 17.2 Direct taxes of which 7,567.8 10.5 yes 18.4 Personal income tax 5,034.1 7.0 yes 27.2 Profit tax 1,930.2 2.7 no n.a. Property tax 603.5 0.8 yes 3.0 Indirect taxes of which 9,589.1 13.3 16.8 VAT 7,452.9 10.3 yes 19.6 Excise taxes on petroleum, tobacco, and alcoholic beverages 1,374.0 1.9 yes 11.1 Other excises 636.2 0.9 no n.a. Custom tax 126.0 0.2 no n.a. Other taxes 229.1 0.3% no n.a. Nontax revenue 1,725.2 2.4% no n.a. Grants 265.9 0.4% no n.a. Source: World Bank staff calculations based on official administrative data Composition of public expenditures corresponded to Economic Affairs (Table 1.2). 8 In contrast, the government's social expenditures stood at 15.5 percent of GDP, explained mostly by Public expenditure composition indicates prioritization for spending on social protection and spending on social protection, which had the economic affairs. In fiscal year 2022, total general second-highest spending with expenditures government expenditures in Georgia reached 29.0 amounting to 6.7 percent of GDP. In-kind transfers percent of GDP, 7 below those of peers (averaging 38 amounted to 7.5 percent of GDP, with almost equal percent). Within government expenditures, non- distribution to the health and education sectors. The social expenditure represented the second-largest government’s expenditures on subsidies amounted to outlay (almost 13.6 percent of GDP), half of which 2.5 percent of GDP. 7 8 General government expenditures based on data from the Economic affairs include transport, energy, agriculture, multi- Ministry of Finance, Georgia. purpose economic projects, and so forth. 14 Table 1.2: Georgia’s government expenditure, 2022 Government spending Amounts Total (% of Included in Captured in (GEL, millions) (GEL, millions) GDP) the analysis CEQ (as % of (yes/no) budget) Total expenditure 21,144 29.0 n.a. 34.3 Primary government spending 20,383 28.0 n.a. 35.6 Social spending 11,268 15.5 n.a. 65.6 Social protection 4,903 6.7 yes 71.2 Non-contributory pensions 2,902 4.0 yes 88.3 Social assistance excluding pensions 2,002 2.7 yes 46.3 Targeted social assistance 1,165 1.6 yes 37.8 Internally displaced population 94 0.1 yes 62.6 Other social programs 743 1.0 yes 57.5 In-kind transfers 5,491 7.5 yes 60.7 Education of which 2,362 3.2 yes 59.0 Pre-school and general 1,495 2.1 yes 86.7 Vocational education 86 0.1 yes 11.4 Higher education 147 0.2 yes 59.9 Other education 634 0.9 no n.a. Health 2,119 2.9 yes 91.5 Housing and urban 1,011 1.4 no n.a. Non-social spending 9,876 13.6 yes 9.0 Subsidies of which 1,828 2.5 yes 24.1 Public utilities 470 0.6 yes 83.4 All other spending* 8,048 11.0 no n.a. Source: World Bank staff calculations based on official administrative data Note: *Other spending includes general public services and environmental protection Over the past decade, spending has shifted toward functions such as education, healthcare, and social capital projects, with reduced allocations for wages protection is relatively low compared to peers in and salaries but increased investments in GDP terms. Lack of unemployment benefits is infrastructure. Social benefits and subsidies surged another reason for low social protection spending. in response to the pandemic. While Georgia For example, the EU-27 countries spent 4.3 percent outspends peers in terms of spending allocated to of their total spending on unemployment benefits in social assistance and fiscal subsidies, spending on 2019, while Georgia does not have an unemployment health and education has remained relatively insurance system in place. contained. Georgia’s spending on welfare-enhancing 15 Impact of fiscal policy conducted using 2022 micro and macro-fiscal administrative data. Aside from its impact on GDP growth and This analysis estimates the impact of both taxes macroeconomic stability, fiscal policy affects and transfers on welfare. The analysis measured the distribution of income across households and individuals. Governments use fiscal policy to the impacts of taxes and transfers on poverty (both generate revenue to finance public spending, headcount and gap) based on the 2022 national which affects income redistribution, living poverty line (per adult equivalent real standards, and poverty levels. Fiscal interventions consumption) at the different CEQ income include direct and indirect taxes as well as, on concepts (Figure 1.1). 9 To begin, consumption the expenditure side, direct transfers, (recorded in the household survey) was equated contributory pensions, subsidies, and provision with the CEQ disposable income concept, and the of health and education. The redistributive rest of the CEQ income concepts were calculated potential of these policies affects the income based on disposable income (consumption) equality and poverty of households and minus/plus the relevant taxes/transfers. 10 Market individuals. income is pre-fiscal income — that is, before the effects of taxes and transfers — for the baseline Recent fiscal challenges make it imperative to scenario of Pensions as Government Transfers assess the redistributive impact of fiscal policy (PGT). The income concepts result from by conducting fiscal incidence analysis (FIA). subtracting taxes and adding transfers to pre-fiscal The FIA helps in gaining a deeper understanding income. Disposable income includes direct taxes of the overall impact of Georgia’s fiscal policy on and transfers effects, while consumable income addressing poverty and inequality. This section considers direct transfers, direct taxes, indirect presents the results of the FIA for Georgia taxes, and subsidies. Figure 1.1: CEQ income concepts in the Pensions as Government Transfers scenario for Georgia - Direct transfers + Indirect subsidies + In-kind transfers Pensions, TSA, Veteran, Electricity, gas, Education, health Disability, others transportation, water Disposable income = Pre-fiscal income Net market Consumable Official consumption Final income = Market income income income from HIES + Direct taxes - Indirect taxes Personal income tax, VAT, excise tax property tax Source: Adapted from Lustig (2022) 9 The poverty headcount is measured as the share of population measured at final income since in-kind health and education below the national poverty line in Georgia. For the fiscal benefits are not part of monetary poverty measurement. incidence analysis, the impacts of taxes and transfers on 10 The poverty headcount at disposable income (13.3 percent) is poverty were measured as the difference between the poverty not equal to the official poverty headcount in Georgia, which is headcount at market income (pre-fiscal income) and the 15.6 percent. According to the CEQ methodology, disposable poverty headcount at consumable income (post-fiscal income income cannot be lower than direct transfers and pensions. excluding in-kind benefits). The poverty headcount was not 16 1.2. Net Payers and Receivers of the Fiscal System11 The bottom five deciles of the population are net The total fiscal position of households indicates cash recipients of the fiscal system, with the most that the bottom 90 percent are net receivers, significant contribution from pensions. To assess while only households in the top 10 percent are which segments of Georgia’s population experience net payers to the fiscal system. The total fiscal cash gains or losses due to taxes and transfers, this position (gray dashed line) includes all cash- study calculated the net cash position of based interventions plus in-kind benefits (such as households according to their market income education and health benefits) valued at the decile (Figure 1.2) by dividing the population into government’s cost of provision, which includes five quintiles ranked by market income (pre-fiscal in-kind benefits from health and education. income). 12 For each quintile, the stacked bars show Bottom 90 percent are the net receivers of the the incidence of the fiscal intervention with fiscal system whereas top 10 are contributing to respect to market income. The net cash position the system. (red dashed line) shows the aggregate sum of all cash-based interventions (all taxes, direct transfers, The first quintile has exceptionally high net and indirect subsidies) for each quintile. The fact benefits compared to market income. The that quintiles 1–3 are net receivers of the fiscal combination of low market income and a heavy system is consistent with the poverty-reducing reliance on pensions and transfers resulted in a impact, and the fact that the net fiscal position is nearly fourfold increase in revenue after more positive for the poorer deciles is consistent accounting for cash tax benefits. The second with the inequality-reducing impact. The net cash quintile also has a net favorable fiscal position at position provides a more accurate measure of almost 40 percent, decreasing to 8 percent for the households' purchasing power, making it more third quintile (Figure 1.2, left panel). The effective in reflecting the impact of fiscal policy on remaining quintiles have a negative budgetary poverty. In simpler terms, it shows whether the position. When factoring in the net tax effect, the government has enabled individuals to afford benefits relative to market income increase even goods and services beyond their initial market more for the low-income quintiles. income (Lustig 2022). 11 Net cash position by decile is calculated by aggregating the relative incidence of taxes and direct transfers (excluding in-kind benefits) with respect to pre-fiscal income (market income). When the net cash position is positive (negative), it means that the decile is a net receiver (payer) in the fiscal system. 12 The size of the effects represents the overall picture by decile, but there could be further heterogeneity of the effects within deciles (Lustig 2022). 17 Figure 1.2: Net fiscal position of households after taxes and transfers Net fiscal position, by quintile (GEL million) Net fiscal position, by quintile (percent of market income) Direct transfers Pensions Direct transfers Pensions 3000 Direct taxes Subsidies Direct taxes Subsidies 800 Indirect taxes Health Indirect taxes Health Education Education 2000 Net cash Fiscal position 600 Net cash Fiscal position GEL, million 1000 400 Percent 0 200 -1000 0 -2000 -200 1 2 3 4 5 1 2 3 4 5 Per capita market income quintiles Per capita market income quintiles Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and CEQ Methodology 1.3. Impact of Fiscal Policy on Poverty The national poverty headcount decreased by 17.1 playing a key role in the reduction of poverty (18.6 percentage points as a result of fiscal policy. The percentage points). Both indirect taxes and subsidies combined effect of pensions and direct transfers contributed to a poverty increase of 1.9 percentage minus payment of direct taxes remained the most points, from disposable to consumable income significant fiscal policy drivers for poverty reduction (Figure 1.3). Although the budget allocation to from market to disposable income (18.9 percentage subsidies remained at 2.5 percent of GDP, slightly points). Pensions, which have a significant share in below the allocation for health (2.9 percent of GDP), the budget allocation (4.0 percent of GDP), are its role in poverty reduction is insignificant. Figure 1.3: Poverty headcount from market to consumable income Poverty by income concept (PGT Scenario) (2022) Poverty gap by income concept (PGT scenario) (2022) 40 20 18 35 32,2 18 30 16 14 25 12 Percent Percent 20 15,6 10 8 15,2 13,3 8 15 6 4 10 3 4 5 2 0 0 Market income Disposable Consumable Official poverty Market Income Market Income Disposable Consumable income income line plus pensions Income Income Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and CEQ Methodology 18 When pensions were measured as deferred income by geography (Figure 1.4). To analyze this, the (the Pensions as Deferred Income (PDI) scenario) poverty headcount ratios were compared between the effect of fiscal policy on poverty was small. market income and consumable income for the Total poverty decreased 3.5 percentage points, which urban and rural regions. The results showed that the suggests that the old-age pensions (in the PGT poverty headcount at market income (pre-fiscal scenario) explain almost all poverty reduction in income) was 1.5 times larger in rural areas relative to that scenario. This is likely because individuals in urban areas, and the decrease in the poverty deciles 1 and 2 receive most of their income from headcount due to the combination of taxes and old-age pensions. The sensitivity analysis based on social expenditure items modeled was higher in rural the PDI scenario is described in Appendix B. areas (decrease of 20.7 percentage points) compared to urban areas (decrease of 14.7 percentage points). The impact of fiscal policy on poverty reduction One of the main reasons for the greater impact on varies by region, with taxes and social expenditure rural areas is that the proportions of pension accounting for greater poverty reduction in rural benefits and direct transfers for rural areas are areas compared to urban areas. The analysis higher at 43 percent and 47 percent, respectively, disaggregated the poverty headcount impacts by compared to the proportion of the rural population region to assess the differential impacts of the in Georgia at 39 percent. combination of taxes and social expenditure items Figure 1.4: Poverty headcount from market to consumable income by region Poverty by income concept by region (PGT Scenario) 45 40,8 40 35 30 26,6 25 20,1 Percent 20 17,6 15 10,5 11,9 10 5 0 Urban Rural Urban Rural Urban Rural Market income Disposable income Consumable income Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and CEQ Methodology The fiscal system reduces poverty the most in “elders and children,” “only elders and no child,” households with “only elders and no children.” and “no elders and no child.” The results showed The effects of the fiscal system vary by gender that in Georgia, the households with “only elders categorization and intrahousehold composition. 13 and no children” had the highest level of poverty Different categories of households were used to at market income (43 percent). The fiscal system estimate the results for five different typologies, has reduced the poverty headcount the most in including “single household,” “only children,” that same household category mainly because of 13 Gender is “a normative social construct defining and positive and negative incentives for broad-based compliance differentiating the roles, rights, entitlements, responsibilities, with those roles, responsibilities, and obligations” (E-CEQ and social obligations of women and men; it also provides 2022). 19 the impact of pensions and healthcare public children. Looking at only the bottom 40 percent spending (Figure 1.5). In comparison, the poverty of the population by household type, elderly reduction impact for households with “only individuals or those living with elderly family children” was only 7 percentage points, indicating members benefit significantly from the system, that the effects of old-age pensions were higher while people in other kinds of households receive compared to social transfers that support far fewer benefits. 14 Figure 1.5: Poverty headcount from market to consumable income by household category Poverty headcount (by household category) 50 41 42 40 29 27 30 23 20 22 Percent 18 17 20 14 15 13 12 13 8 7 9 7 10 6 6 0 Single household Only children Elders and children Only elders and no child No elders and no children Market income Market income plus pensions Disposable income Consumable income Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: [1] Official national poverty headcount does not match poverty at Disposable Income level. [2] Poverty is not calculated at Final Income level as per the methodology of valuing of in-kind transfers to health and education at the average cost of provision. [3] “Single household” is defined as household with single member; “only children” is household with children but no elders; “elders and children” is household with both elders and children; “only elder and no child” is household with elders (above age 65) but no children; and “no elders and no children” is household with couple but no dependents. The poverty reduction impact of fiscal policy in lower for UMICs. ECA countries included in the Georgia is comparable to that of UMICs but lower sample have a high share of older population, and than that of ECA countries. Poverty reduction was their average expenditure in old-age pensions as a compared using international poverty lines for share of GDP is almost double that of UMICs while UMICs and per capita estimations of income the share of social protection is similar. In the concepts. 15 The poverty reduction effect for Georgia context of ECA countries, Georgia does well in (4.3 percentage points) was close to UMICs (4.1 poverty reduction, considering that the sum of its percentage points) and below the average of ECA direct transfers and pensions is the lowest of all countries for which data is available (6.5 percentage ECA countries in the sample. Among UMICs, points). For countries with higher poverty reduction Georgia is only below countries with higher social than Georgia, the effect of pensions, direct transfers, protection expenditures such as Argentina, Brazil, and direct taxes is greater for ECA countries and and Jordan (Figure 1.6). 16 14 Among the bottom 40 percent of the population, households Georgia is estimated using per adult equivalent. Using per with only elders and no children experienced a reduction in capita estimates and a higher poverty line results in lower poverty (from market to consumable income) which is 1.9 poverty reduction than using the national poverty line with times greater than the average poverty reduction for adult equivalent. households with elders and children and for the households 16 An estimated 7.4 percent for ECA countries, using CEQ with only children. Institute Data Center information for Argentina, Colombia, 15 The results for Georgia use US$6.85 (2017 PPP), and the results China, Mexico, Tajikistan, Türkiye, and Ukraine and using for other countries use US$5.50 (2011 PPP). National poverty in Robayo-Abril and Cabrera (2024) for Bulgaria. 20 Figure 1.6: Poverty impact of fiscal policy in UMIC and ECA countries (percentage points) ECA countries UMICs 30 25 25 20,3 20 18,8 20 15 11,9 Percentage points Percentage points 15 17,3 3,0 4,3 7,9 10 10 9,9 -0,9 -0,2 0,6 1,3 1,8 5 5 4,3 -3,4 0 0 -0,5 0,2 -5 -5,8 -5 -10 -10 -15 Subsidies and indirect taxes Subsidies and indirect taxes Pensions, direct transfers and direct taxes Pensions, direct transfers and direct taxes Net cash impact Net cash impact Source: CEQ Institute Data Center; Cojocaru et al. (2019); De La Fuente & Cabrera (2023); Lara Ibarra, et al. (2023); World Bank (2023); Robayo-Abril and Cabrera (2024); Rodriguez and Wai-Poi (2021); and Rodríguez-Castelan et al. (2023) 1.4. Inequality Impact of Fiscal Policy17 The FIA results show that the combination of bigger than both, with a marginal effect on taxes and social expenditures in Georgia inequality of 7.8 versus 7.5 Gini points, contributes to reducing inequality. As measured respectively. Indirect subsidies had a small by the Gini coefficient, inequality was reduced impact on inequality reduction from disposable by 20 Gini points (from 44.7 at market income to to consumable income (0.5 Gini points) despite 24.6 at final income) (Figure 1.7). Most of the high spending by the government (2.5 percent of inequality reduction occurred from market GDP). The comparative effect of in-kind income to disposable income, suggesting that transfers on inequality reduction is higher than direct transfers — particularly pensions — are indirect subsidies, although there is no the main drivers of inequality reduction. The significant difference in budget allocation pensions share in government allocation was (education at 3.2 percent of GDP, health 2.9 higher than the shares of health and education, percent of GDP, and subsidies 2.5 percent of and its impact on inequality reduction was also GDP). 17 Equity impact is measured by the Gini index on a scale from 0 to 100, with 0 meaning perfect equality and 100 meaning perfect inequality. For the fiscal incidence analysis, the impacts of taxes and transfers on inequality were measured as the difference in the Gini coefficient between market income (pre-fiscal income) and the Gini at final income (post-fiscal income after including all taxes, transfers, and in-kind benefits from health and education). 21 Figure 1.7: Impact of fiscal policy on inequality, areas (49.0 Gini points compared to 41.4 Gini measured by the Gini coefficient (PGT scenario) points, see Figure 1.8). A larger number of households with pensioners have near-zero 50 44,7 market income, leading to higher inequality 45 within households with elders in rural areas. 40 32,6 32,1 Higher inequality in rural areas could also be due 35 to an improvement in income of a specific 30 24,6 population segment. This was evident from the Gini index 25 20 recent increase in median income and the 15 prevalence of employment as a source of income 10 within rural households (Bakradze et al. 2023). 5 However, there has been insufficient 0 improvement in other segments of the Market Disposable Consumable Final population, such as the poor. Disaggregated income income income income analysis by region suggests that the fiscal system has reduced inequality more in rural areas from Source: World Bank staff calculations based on HIES 2022, fiscal market income to final income than in urban administrative data, and the CEQ Methodology areas (22.9 Gini points compared to 18.1). The higher inequality reduction in rural areas is In the PDI scenario, the estimated inequality driven by direct transfers and in-kind transfers reduction was significatively smaller. The (mainly health). decrease was smaller (12.2 Gini points) because unlike in the PGT scenario, pension transfers Fiscal policy contributes to reducing urban-rural were not considered in the PDI scenario (see inequality gaps. The Theil index decomposes Appendix B for details). The main drivers of total inequality into within-group and between- inequality reduction in the PDI scenario were in- group components for rural and urban areas. kind transfers and social assistance transfers. According to the decomposition, almost all of the difference is due to within-group inequality, Inequality is greater in rural areas than in urban and a meager share of total inequality is between areas, and fiscal policy has a larger impact in groups. Fiscal policy has helped in reducing rural areas. Inequality at the market income within-group inequality and disparity between level is higher in rural areas compared to urban groups (Figures 1.8 and 1.9). Figure 1.8: Impact of fiscal policy on inequality (measured by the Gini coefficient) by region Inequality by income concept by region (PGT scenario) 60 49,0 50 41,4 39,8 40 34,4 30,8 30,3 Percent 30 26,1 23,3 20 10 0 Urban Rural Urban Rural Urban Rural Urban Rural Market income PGT Disposable income Consumable income Final income Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 22 Figure 1.9: Impact of fiscal policy on inequality (measured by the Theil index) Inequality decomposition by area 0,35 0,30 0,25 Theil index 0,20 0,15 0,10 0,05 0,00 Market income PGT Disposable income Consumable income Final income Within-group Between-group Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Analysis of the inequality reduction impact category, the fiscal system reduced that based on household categories indicates that inequality by half to 24 Gini points, the largest households with “only elders and no children” inequality reduction across all household types. have benefited the most from the fiscal system. This is mainly attributable to the impact of Categorization of households using the pensions (Figure 1.10). The magnitude of household composition typologies indicated that inequality reduction among households with the households with “only elders and no children” adults and children was about the same, from 45 had the highest level of inequality at market Gini points to 20 Gini points. income level (50 Gini points). For this household Figure 1.10: Inequality from market income to final income by household category Gini coefficient (by household category) 60 50 50 45 42 40 40 35 34 32 32 31 32 31 30 29 28 Percent 28 28 30 27 24 22 20 20 10 0 Single household Only children Elders and children Only elders and no child No elders and no children Market income PGT Disposable income Consumable income Final income Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 23 The impact of fiscal policy in reducing inequality more significant impact. Georgia is also one of in Georgia is among the highest for UMICs and in the countries with the highest pre-fiscal Gini the ECA region. Results show that Georgia's fiscal coefficient. Among UMICs, Georgia has achieved system successfully reduced inequality. Georgia the most significant reduction in inequality. The ranks 9th out of 98 countries on Sustainable primary factor behind this decline in inequality Development Goal (SDG) 10.4.2, 18 which is the pension effect, akin to countries like indicates the redistributive impact of its fiscal Argentina, Bulgaria, and South Africa when policy (Figure 1.11). These inequality impacts are combining the effects of pensions and social driven by pensions and direct transfers. assistance transfers. This is because a sizable Compared to other countries in the ECA region, number of households in Georgia depend the difference between the Gini coefficient primarily on old-age pensions for their income, before and after fiscal policy in Georgia is among resulting in their pre-fiscal income being low or the highest, with only Bulgaria experiencing a near zero. Figure 1.11: Redistributive impact of fiscal policy — SDG 10.4.2 18,0 16,0 14,0 Georgia; 12,1 12,0 SDG 10.4.2 10,0 8,0 6,0 4,0 2,0 0,0 Source: United Nations Sustainable Development Goals (SDGs) Indicator 10.4.2. https://sdgs.un.org/goals (https://sdgs.un.org/goals/10/targets/10.4.2), compiled by the World Bank using information from CEQ Institute, OECD, and World Bank 18 Sustainable Development Goal (SDG) 10.4.2 is the officially coefficient). More details on the indicator can be found at accepted indicator for assessing government effectiveness in www.sdgdata.gov.uk/10-4-2/. Notably, estimations for most tackling inequality within the SDG framework. It is calculated countries are based on pre-pandemic data, which may not under CEQ methodology as the difference between pre-fiscal fully capture the increased social spending and transfers and post-fiscal income inequality (as measured by the Gini observed during and after the pandemic. 24 1.5. Progressivity of the Fiscal System19 Progressivity analysis of the fiscal system indicates benefits to fiscal income for poorer households — that lower-income households tend to receive a that is, positive Kakwani, lower than Gini larger amount of direct transfers, while higher- coefficient). Some indirect subsidies (transport and income earners do not necessarily pay a higher gas) are progressive, but this classification is mainly proportion of taxes. To assess progressivity, the due to their less unequal distribution than Kakwani Index (KI) — an aggregate indicator of disposable income. However, the absolute incidence relative progressivity — was calculated for each fiscal of these subsidies reveals that they are concentrated intervention. The findings show that all indirect among affluent households. In addition, most in- taxes and VAT result in lower payments relative to kind health and education benefits are progressive prefiscal income, with negative Kakwanis in Figure except for tertiary education, which is progressive 1.12 which indicates that they are regressive. PIT but not pro-poor as shown in Figure 1.12. 20 and the excises are considered neutral, as their Kakwani values are close to zero, meaning they are It should be noted that progressive taxes are not paid in the same proportion as prefiscal income. allowed according to the Georgian Constitution and Most direct transfers are pro-poor (higher per capita Organic Law on fiscal rules. A regime of minimum benefits for the poorest households, with a Kakwani tax-free incomes for low earners also requires careful higher than the market income Gini coefficient), considerations about tradeoffs and potential and only pension contributions paid by the negative incentives for loopholes, double accounting, government are only progressive (a higher ratio of and as consequence inefficiencies of a tax system. Figure 1.12: Progressivity of selected fiscal policy instruments, Georgia 2022 Progressivity as measured by Kakwani Progressivity as measured by Kakwani VAT PIT All indirect taxes Indirect taxes excises Gov. contributions to pensions Regressive Progressive Transport subsidy All direct taxes Regressive Progressive Education—tertiary Property tax Pro-poor All indirect subsidies Pro-poor Education—vocational Gas subsidy Health Health Education—preschool IDP transfer Education—general Education—total Electricity subsidy Old-age pensions Old-age pensions Other transfers Loss of breadwinner transfers Water subsidy All direct transfers TSA -0,5 0,0 0,5 1,0 1,5 -0,5 -0,3 -0,1 0,1 0,3 0,5 0,7 0,9 Kakwani Index Kakwani Index Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: Yellow highlighted area is Gini and indicates if transfers are pro-poor or not 19 Progressivity is measured by the Kakwani Index (KI), which compares the concentration coefficient of a tax/transfer (with respect to a reference income) to a Gini coefficient measured over the same reference income. A positive KI means that the fiscal intervention is more equally distributed than is the reference income, a negative KI means that the fiscal intervention is less equally distributed than is the reference income, and a KI close to zero indicates that the distribution of the fiscal intervention is approximately as equal as the distribution of the reference income. Results are typically considered significant when the KI is higher than 0.10 in absolute value. The KI for taxes is calculated as the difference between the concentration coefficient of the tax and the Gini of the prefiscal income concept (market income). The KI for a transfer is calculated as the difference between the Gini of the prefiscal income concept and the concentration coefficient of the transfer. In both cases, a positive KI means that a tax or transfer is progressive, a negative KI means that it is regressive, and a zero KI means neutral (Lustig 2018). When a concentration coefficient is negative for transfers, the Kakwani is higher than the prefiscal Gini and the transfers are pro-poor, so the per capita benefit is higher for lower-income households. 20 A transfer is pro-poor when the per capita benefit declines with pre-transfer income, so it is lower in absolute terms for higher pre-transfer income. When a transfer is only progressive, benefits as a share of pre-transfer income decline with income. 25 1.6. Marginal Contributions to Inequality and Poverty Reduction21 Pensions and direct transfers are the most and health spending — contributed significantly significant contributors to reducing both to inequality reduction (7.8 and 4.3 Gini points, poverty and inequality. To assess which fiscal respectively). Pensions and direct transfers interventions have a greater individual impact in remained the most significant fiscal policy reducing poverty and inequality, the study drivers for poverty reduction (18.6 percentage estimated their marginal contributions (MCs), 22 points and 5.4 percentage points, respectively). considering the progressivity, size, and coverage The impact of indirect subsidies remained low to of an instrument or group of instruments. neutral in poverty and inequality reduction Government transfers — particularly pensions (Figures 1.13 and 1.14). Figure 1.13: Marginal contribution to inequality reduction of each fiscal instrument, in Gini points, measured at consumable income Transport subsidy 0,0 Water subsidy 0,0 Education—vocational 0,0 Value added tax (VAT) 0,0 Property tax 0,0 Inequality reducing Excises 0,0 Electricity subsidy 0,0 Loss of breadwinner transfers 0,1 Gov. contributions to pensions 0,1 Education—tertiary 0,1 IDP transfers 0,2 Other transfers 0,3 Personal income tax (PIT) 0,4 Gas subsidy 0,8 Education—preschool 0,9 Targeted social allowance (TSA) 3,0 Education—general 3,2 Health 4,3 Old-age pensions 7,8 0 1 2 3 4 5 6 7 8 9 Marginal effect on inequality (Gini points) Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 21 Marginal contributions (MCs) summarize the individual impact of a tax or transfer in reducing poverty and inequality when the impact of all other fiscal interventions in the model is held constant. For example, the MC to inequality reduction is calculated as the difference between the Gini coefficient of the reference income (consumable income in this case) relative to the Gini coefficient of the reference income minus (plus) the specific tax (transfer). The procedure is similar for calculating the MC to poverty reduction. A positive MC means that the fiscal intervention contributes to reducing poverty or inequality; a negative MC means that the fiscal intervention contributes to an increase in the poverty or inequality indicator. MCs depend on the size, progressivity, and coverage of fiscal interventions. 22 For the calculation of MC to inequality reduction, the following formula was used: Gini of consumable income minus Gini of consumable income including the specific tax or transfer (minus for tax and plus for transfer). A similar formula is used for the calculation of MCs to poverty reduction, based on the poverty difference without and with the tax or transfer. 26 Figure 1.14: Marginal contribution to poverty reduction of each fiscal instrument, in percentage points, measured at consumable income Poverty increasing Value added tax (VAT) -3,4 Personal income tax (PIT) -2,5 Excises -0,1 Water subsidy 0,0 Direct taxes property tax 0,0 Loss of breadwinner transfers 0,2 Transport subsidy 0,2 Electricity subsidy 0,2 Gov. contributions to pensions 0,3 IDP transfer 0,4 Gas subsidy 1,9 Other transfers 2,9 Targeted social allowance (TSA) 5,4 Old-age pensions 18,6 -5,0 0,0 5,0 10,0 15,0 20,0 Marginal effect on poverty (percentage points) Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Taxes, both direct and indirect, had neutral to elements of fiscal policy constant, both direct no impact on inequality reduction but increased and indirect taxes contributed to poverty poverty. Relative to other government increases, with the highest impact coming from expenditures, indirect subsidies including water, the VAT (3.4 percentage points) and PIT (2.5 electricity, and transport had an insignificant percentage points). The impact of indirect impact on inequality reduction (Figure 1.13). In subsidies on poverty reduction remained low to terms of poverty reduction, holding other neutral (Figure 1.14). 1.7. Cost Effectiveness of Fiscal Interventions Cost effectiveness analysis 23 indicates that direct poverty reduction as loss of breadwinner transfers, transfers are the most effective instrument for IDP transfers, and electricity subsidies. Water reducing poverty and inequality. TSA transfers subsidies and government contributions to pensions show the highest effectiveness in terms of reducing have a lower impact on poverty reduction. The poverty and inequality. Considering the amount effect on inequality is also highest for TSA transfers, simulated in the analysis as a share of GDP, TSA followed by loss of breadwinner transfers, with the transfers have more than double the impact on lowest impact per unit on inequality found for poverty and inequality than pensions by unit of transport subsidies and government contributions to expenditure. Pensions have the same effect on pensions (Figure 1.15). 23 Marginal contributions are often linked with a "cost in the survey. Essentially, this shows how much Gini or poverty effectiveness" indicator. Cost effectiveness is calculated by reduction can be accomplished for each unit of currency spent on dividing marginal contributions by the budget amount captured an instrument, such as GEL or a point of GDP. 27 Figure 1.15: Cost effectiveness of taxes and transfers, at consumable income TSA transfers 7 12 Loss of breadwinner transfers 3 7 Pensions 2 5 IDP transfers 2 5 Electricity subsidy 1 5 Gas subsidy 1 4 Transport subsidy 3 Gov. contributions to pensions 0 1 Water subsidy 2 0 0 2 4 6 8 10 12 Inequality / poverty cost effectiveness by unit of GDP Inequality Poverty Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: The cost effectiveness index measures how well a fiscal policy instrument reduces inequality or poverty relative to the budget allocated as a share of the GDP. The higher the value, the greater the cost-effectiveness of the fiscal policy measure. 1.8. Incidence of Fiscal Interventions24 Taxes are paid mostly by the top quintile, often the wealthier individuals who gain the whereas direct transfers are received mostly by greatest benefits from indirect subsidies. The two the bottom quintile. For PIT and VAT, the top richest deciles receive a larger share of indirect quintile pays 46 percent of total PIT modeled; subsidies — including domestic transport, gas, and these results are mostly explained by the fact that water subsidies — compared to the two poorest richer households have higher taxable income and deciles (27.4 percent versus 19.2 percent). Transport larger consumption. However, the relative subsidies mostly benefit the top income quintiles incidence indicates that the burden of PIT falls on of the urban population, and gas subsidies tend to the bottom quintile mainly due to the flat rate benefit higher-income earners and urban structure of the tax. VAT incidence is also households as they spend more on this service. In disproportionately high in relative terms for the contrast, electricity subsidies tend to benefit the bottom quintile. Property tax is the most lower quintiles of the income distribution, progressive of all taxes, but it is a small component especially in urban areas, and water subsidies of total tax collection. For total direct transfers, benefit lower-income urban households. Indirect most of the distribution of benefits is received by subsidies are not as efficient in assisting the poor poorer deciles. compared to social transfers. Total in-kind health benefits are distributed more progressively, with 23 While direct transfers and pensions provide percent versus 18 percent to the bottom and top crucial support to vulnerable populations, it is quintiles, respectively (Figure 1.16). 24 Absolute incidence is calculated as the total taxes (transfers) that each decile pays (receives) as a share of total taxes (transfers) modelled. In the case of social protection direct transfers, since absolute incidence is based on the aggregate transfers of each decile, it does not allow for distinguishing whether coverage or generosity is driving the results. 28 Figure 1.16: Absolute incidence: shares of total taxes and benefits paid/received by households, ranked by market income 100,0 90,0 80,0 70,0 60,0 Percent 50,0 40,0 30,0 20,0 10,0 0,0 Direct Pensions Direct taxes Subsidies Indirect taxes Health Education Market transfers income Top 20% de population ranked by market income 21–80% of the population ranked by market income Bottom 20% of the population ranked by market income Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 1.9. Limitations of the Analysis The CEQ Assessment framework for FIA is a defense, or debt interest payments — due to static and retrospective accounting exercise methodological limitations. This means the without behavioral, lifecycle, or general incidence results represent an accounting-based equilibrium effects. Major challenges include the first approximation of a true counterfactual. 28 In failure to capture top-income households 25 and practice, there are no standard errors calculated the exclusion of some interventions from the that would allow a statistical assessment of the model — such as the corporate income tax allocations made to individuals and households. 29 (CIT), 26 public expenditure on infrastructure, 27 Nevertheless, the CEQ Assessment framework 25 Household surveys also do not capture some low-income water infrastructure (publicly provided water), and local households very well, such as institutional populations roads. The methodology for including infrastructure (prisons, old-age care facilities, youth care facilities) and expenditure in fiscal incidence models is currently under households without a domicile address (informal housing). development by the CEQ Institute. 28 26 CIT is better modelled with tax administrative data. Meaning distributional (static) impacts, not causal impacts. Furthermore, data on CIT cannot help in the allocation of CIT 29 There are, however, CEQ Assessment procedures for burdens to households in the standard CEQ FIA framework. assessing the statistical significance of the estimated impact 27 Examples of the relevant infrastructure investments that are of a fiscal policy (or set of fiscal policies) on poverty, missing and that could affect households differently in the inequality, and other indicators. income distribution could be connectivity infrastructure, 29 for FIA provides a standard methodology (which accounts, which could be due to missing top- also enables international comparisons) for income households in the survey, methodological estimates of the impact of fiscal policy on differences, data collection challenges, and poverty, inequality, and social welfare more conceptual issues. This problem has been found generally. Besides, the HIES only covered 27.0 in other countries (see Box 1.1). percent of the total consumption in national Box 1.1: Gaps in consumption captured in the Georgia HIES survey The household consumption measurement using national accounts and household surveys often exhibits discrepancies. These differences can be attributed to various factors. One significant factor contributing to the disparity between national accounts and household surveys in measuring household consumption is unit nonresponse (Hlásny and Verme 2021). Consequently, national accounts show higher per capita household consumption measures than household surveys. Prydz et al. (2022) estimated the gap between per capita survey income and per capita household final consumption expenditures (HFCE), after compiling a data set of 2,095 national household survey means for 166 countries from 1967 until 2019, together covering countries that accounted for 97 percent of the world population in 2017. The authors found that the average HFCE estimated from household surveys was 20 percent lower than the average HFCE estimated from national accounts. For ECA countries, the household survey reported HCFE that were 27.5 lower than the HFCE from national accounts while middle-income countries reported HCFE that were 35.2 percent lower than national accounts. Fiscal incidence analysis for Georgia using CEQ methodology for the year 2022 has the limitation of low macrovalidation (share of the totals in the survey versus the administrative accounts). Analysis indicates that this low macrovalidation stems from the gap in the consumption measured in national accounts compared with the household’s survey. The gap in Georgia is considerable, reaching 72.5 percent lower than the HCFE. This has several implications, including overestimation of the welfare impact of certain interventions. One such example is how the reported pensions, compared to national accounts, tend to overestimate the impact of pensions in reducing poverty. The discrepancies between household consumption measurements using national accounts and household surveys can arise from unit nonresponse, underreporting of consumption by wealthy households, differences in capturing income streams, and inconsistencies in data sources and measurement methods. It can be linked to the underreporting of consumption in wealthy households (Arndt and Mahrt 2017). Additionally, missing or underreporting of the top incomes can contribute to the differences observed between estimates of consumption expenditure in national accounts and household surveys (Anand and Segal 2015). Furthermore, national accounts may better capture income streams from owner-occupied housing and government-provided public goods than household surveys, leading to discrepancies in GDP-based measures (Pinkovskiy and Sala-i-Martin 2014). Understanding these differences is crucial for policymakers and researchers to interpret and compare consumption data accurately. Source: World Bank's estimations based on HIES 2022 30 C H A P T E R T W O Equity and Gender Aspects of Fiscal Policy Reforms Impact of Personal Income Tax | Impact of Indirect Taxes | Pensions and Redistribution | Indirect Subsidies | Education Spending | Gender Equity in Fiscal Policy | 31 C H A P T E R T W O Equity and Gender Aspects of Fiscal Policy Reforms This chapter presents some key policy simulations, specific analyses for each fiscal intervention and gender effects of fiscal system. The results of the FIA point to some key areas of reform that can further improve the redistributive impacts and reduce poverty and inequality. These reforms can also help increase fiscal space and equity-enhancing investments. Implementation of these reforms would enable Georgia to raise more revenue from certain tax handles as well as improve the efficiency and targeting of government transfers. This chapter also explores the effects of the fiscal system in promoting gender equality while reducing poverty, vulnerability, and exposure to livelihood risks. Based on the Georgia CEQ 2022 results, possible areas of reform and parameters of simulations along with Engendered CEQ (ECEQ) results are presented below. 2.1. Impact of Personal Income Tax The PIT is paid mainly by the top two richest incidence of the PIT). 30 Moreover, income tax deciles but is not progressive. These results are progressivity is undermined by lower rates offered mostly explained by richer households with higher on income earned from property (rental and/or taxable income and consumption. However, sale). VAT incidence is also disproportionately relative incidence indicates that a higher PIT high in relative terms for the bottom 20 percent. burden is falling on the bottom 20 percent of the The (recurrent) property tax is the most population due to the flat rate structure of the tax progressive of all taxes, but it is a small component (see Figure 2.1 for the absolute and relative of total tax collections. 30 This analysis may not capture well the passive income, and household reporting of labor income or consumption, causing the difference might be due to some inconsistencies in an overestimation of the PIT for the bottom quintile. 32 Figure 2.1: Incidence of personal income tax PIT incidence (% of pre-fiscal income) Concentration share of PIT (% of budget) -17 47 45 -18 50 -16 45 40 (%) Share of budget Share of Market income -14 -11 35 -12 -11 -11 -11 30 25 25 -10 25 -8 20 16 16 15 10 10 -6 10 -4 5 2 3 -2 0 0 1 2 3 4 5 1 2 3 4 5 Per capita Market income quintiles Per capita Market Income quntiles Market income PIT Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Georgia’s PIT regime implies a higher rate on low- introducing an income exemption equivalent to one- to-average income earners with a formal job and third of the poverty line (GEL 800 annually), with a implies a lower rate on high-income earners. higher rate of 30 percent for high-income earners. High-income earners in Georgia are likely to benefit disproportionately from the policy of Including an exemption for low earners would offering a sharply reduced rate of 5 percent on help prevent the impact of PIT in increasing income from residential property (rental or sale) poverty. However, reducing tax rates would also while maintaining the same flat tax rate of 20 result in a loss of tax collection. Three scenarios percent regardless of how high the wage income could be considered as policy options (Figure 2.2): earned is. Certain categories of passive income are 1. Lowering the tax burden for all deciles, with a subject to lower tax rates compared to active higher effect on the bottom income earners, income. Since high-income earners are more likely would reduce poverty by around 0.6 to own, rent, and sell residential property than percentage points. low-income earners, they are more likely to benefit 2. Making high-income earners pay more would from this reduced tax rate. reduce poverty by the same amount as in the first scenario but with a higher impact on Implementing progressivity in the PIT brackets can inequality. reduce the effects of PIT both on poverty and on poor households. The analysis examined three 3. Lowering the threshold for exemption from options for increasing progressivity. The first option PIT and increasing the rate for top earners involved tax relief for low-income households, with would have a lower poverty impact but a an exemption (GEL 1,100 annually) equal to half of bigger decrease in inequality compared to the the poverty line. The second option introduced an second scenario. additional bracket for high-income earners (those In terms of tax collection, the cost could be earning less than the 90th percentile of wage reduced by 15 percent of tax collection in the first income), at a higher rate of 25 percent. The third scenario, around 10 percent in the second, and scenario aimed to minimize tax collection loss by approximately 5 percent in the third. 33 Figure 2.2: Personal income tax scenarios Change in total payments by decile Effects on poverty and inequality Change in total payments by decile (percent) Poverty and inequality change with respect to 5 baseline scenario 0 0,8 Percentage point and Gini point -5 0,6 -10 Percent change -15 0,4 -20 0,2 -25 -30 0,0 1 2 3 4 5 6 7 8 9 10 Total Scenario 1 Scenario 2 Scenario 3 Per capita market income decile Scenario 1 Scenario 2 Scenario 3 Poverty reduction (percentage points) Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: Scenario 1 simulates an exemption for the poorest households. Scenario 2 includes the same exemption but also introduces an additional tax bracket for high-income earners. Scenario 3 offers the same options as scenario 2 but with a higher tax rate for high-income bracket earners. 2.2. Impact of Indirect Taxes The bottom prefiscal income quintile pays a usually consists of individuals with very low higher percentage of their income in indirect incomes, especially if pensions and direct transfers taxes compared to their total income. This is due are not taken into account. When these are to the fact that VAT is the most important tax, factored in, households increase their ability to and the lowest income group pays a higher portion spend, but they also pay more VAT. The same of their income for this tax. The higher percentage applies to specific consumption taxes to a lesser paid by the lowest income group is because it extent, see Figure 2.3, left panel. Figure 2.3: Incidence of indirect taxes and consumption Indirect taxes (% of prefiscal income) Consumption and VAT (distribution by quintile) 50 40 80 68 70 72 64 67 40 60 30 Percent Percent 40 20 14 12 23 11 10 20 19 18 13 13 13 16 20 12 12 10 2 1 1 1 1 0 - 1 2 3 4 5 1 2 3 4 5 Per capita Market income quintiles Per capita Market income quintiles VAT Excise Taxable Exempted Informal Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 34 VAT payments are reduced because of informality informal sales outlets are more common among and exemptions for the poorest households. The lower-income households. This becomes clear lower-income population has a higher proportion when consumption is divided into three categories: of exempt consumption and makes more purchases those that pay VAT, those that are exempt from from informal places. According to Figure 2.3, VAT, and those that involve purchases from right panel, tax exemptions and purchases from informal places. 2.3. Pensions and Redistribution The aging population poses long-term fiscal risks Pensions and TSA are the largest drivers of associated with increased pension obligations. poverty reduction. Pensions transfers have a According to the latest United Nations (UN) significant impact on reducing inequality and estimates, the old-age dependency ratio for poverty (7.8 Gini points and 18.6 percentage Georgia is about 23 percent — meaning that for points, respectively). In particular, the effects of every elderly person, there are just over three pensions are noteworthy for reducing poverty persons of working age. Peer countries that are among the elderly population, as seen from the more advanced in the aging process such as Croatia blue area in Figure 2.5. The poverty-reducing and Estonia have considerably higher dependency impact of TSA is significantly larger for children, ratios. as indicated by the orange area in Figure 2.4. Figure 2.4: Headcount poverty before and after pensions and TSA, by age groups 70 60 50 40 Percent 30 20 10 0 Age groups TSA effect on poverty Pension effect on poverty Pre-fiscal poverty Poverty after pensions Poverty after pensions and TSA Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: Averages computed by age group using national poverty line 35 The fiscal costs of pensions in the future will population between ages 15–44 are saving for their depend on actual contributions to the private retirement, with the majority of active system. Increasing the contribution period and contributors being in the 40–44 age group. It finding an adequate level of individuals’ indicates that a considerable portion of the contributions to private pension plans are critical. population will depend on non-contributory These contributions are important because the pensions to sustain themselves in their old age. public sector supplements these pensions to Therefore, it is crucial to encourage the working ensure adequate replacement rates for retirees. population to participate more actively in the Analysis reveals that only 38 percent of the pension system. 2.4. Indirect Subsidies Indirect subsidies are untargeted and have neutral indirect subsidies are obtained by the top deciles. to no impact in reducing poverty and inequality. The total share of indirect subsidies received by the Relative to other government expenditures, indirect top quintile is larger than the share received by the subsidies — including water, electricity, gas, and bottom quintile (27.5 percent versus 19.2 percent) transport — have an insignificant impact in (Figure 2.5). The concentration of subsidies in the reducing inequality. Their impact on poverty also richest quintile is explained by the domestic remained largely neutral. Most of the benefits of transport subsidy and gas subsidy. Figure 2.5: Incidence of indirect subsidies, by quintile Concentration share of indirect subsidies (% of 35 30 budget) 27,5 30 Share of Market income (%) 25 25 20,2 19,2 Share of budget (%) 20 17,2 20 16,0 15 15 10 10 5 5 - - 1 2 3 4 5 1 2 3 4 5 Per capita market income quintiles Per capita market income quintiles Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 36 Significant regional disparities can also be seen in — goes to the richest two quintiles of the urban subsidies. Significant regional disparities can also population (Figure 2.6). The poor and residents of be seen in subsidies. Almost 99 percent of the rural areas and the mountainous region do not benefits of the transport and water subsidies flow substantially benefit from the gas subsidy. Rural to urban areas. Of the total transport subsidy households received one-third of this subsidy, and budget, 65.9 percent goes to the richest two the poor received only 18 percent. That is equal to quintiles in urban areas. Water subsidy only covers the share of gas subsidy benefits the entire rural urban areas, but the poorest population receives population receives. Electricity subsidies are the most of the benefits in a sector with significant only ones that benefit poorer households more challenges (see Box 2.1). A considerable share of significantly, although the share is consistently the gas subsidy — 33.6 percent of the total budget higher in urban areas. Figure 2.6: Incidence of indirect subsidies, by type and by quintile Gas subsidy (% of the benefits by quintile Transport subsidy (% of the benefits by and area) quintile and area) 25 45 41,4 19,6 40 20 35 14,0 30 15 24,5 Percent Percent 11,7 25 10,6 10,3 20 10 8,2 14,9 6,3 6,3 6,9 15 6,0 10,5 8,8 5 10 5 - - - - - 0 0 1 2 3 4 5 1 2 3 4 5 Per capita market income quintiles Per capita market income quintiles Urban Rural Urban Rural Electricity subsidy (% of the benefits by Water subsidy (% of the benefits by quintile quintile and area) and area) 25 23,1 70 59,3 60 20 50 15 13,9 Percent Percent 40 10,0 9,1 8,6 30 10 7,2 7,9 7,67,5 18,0 5,1 20 5 10,2 8,8 10 3,4 - 0,3 - - - 0 0 1 2 3 4 5 1 2 3 4 5 Per capita market income quintiles Per capita market income quintiles Urban Rural Urban Rural Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 37 Box 2.1: Water and sanitation spending in Georgia Despite abundant water supply in Georgia, the population faces challenges in its access and quality. The main reasons for worsening water quality are irrigation, melioration of salty soils, wastewater, and improperly arraigned reservoir caves. The design and structure of government spending on the water and sanitation sector in Georgia aims at supporting residential consumers and adhering to Sustainable Development Goal 6 (ensure access to water and sanitation for all). The share of Georgians connected to public water supply systems is increasing, whereas access to basic sanitation services is stagnating (see United Nations data dashboard). Water resources in Georgia are unequally distributed and are mainly concentrated in the western part, whereas eastern regions often experience water shortages. As of 2022, about 69 percent of the population had access to safely managed drinking water services, while only 24 percent had access to sanitation services (UNICEF, WASH). The government of Georgia has shown commitment to making basic drinking water, hygiene, and sanitation (WASH) services accessible to all of the population, particularly residents of rural areas. By 2025, it plans to expand water and sanitation services to more than 360,000 people. Standard 24-hour water supply is planned for all cities in Georgia by 2025, and from 2030, full access to standard water supply will be provided to all parts of the country (Georgia, 2022). The water supply system is deregulated through issuance of licenses, while tariffs are determined by the Government. As of 2022, nine water supply licensees operate in the water supply sector of Georgia; one is a state-owned company, five are municipal companies, and three are privately owned companies. Out of the total population with access to drinking water, as of 2022, 68.1 percent of the population was supplied with water-by-water supply licensees, while 31.9 percent was supplied by local self-government entities. Tariff calculation for water supply licensee enterprises is regulated by the Georgian National Energy and Water Supply Regulatory Commission (GNERC). For the government, water and sanitation spending involves expenditures on subsidies as well as capital expenditures. As of 2022, total government spending on water amounted to GEL 340.9 million, of which subsidies spending was GEL 81.5 million (23.9 percent) (Georgia budget, 2022). Remaining amounts were allocated to grants and capital transfers to the other municipal budgets and self-governing units. For the sanitation sector, spending was GEL 27.4 million, of which GEL 21.7 million was for subsidies. Distributional analysis using HIES 2022 shows the disparities in access to water and sanitation facilities. For access to water supply, upper deciles have relatively more access to water compared to the poor (61.6 percent for the bottom decile, compared with 88.6 percent for the top decile) (Figure B2.1.1). Significant disparities exist by city in access to water facilities, with Tbilisi residents reporting 100 percent availability compared with Samegrelo-Zemo Svaneti at only 43 percent (Figure B2.1.2 below). Similar disparities exist for both water and sanitation facilities by region, whereby 91 percent of urban residents have own flush toilets compared with only 9 percent of rural residents (Figure B2.1.3 and Figure B2.1.4). Actual benefits reaching poor and rural residents are only a small share of the total amount. Figure B2.1.1: Population with water supply system Figure B2.1.2: Population with water supply system installed in dwelling, by decile (percent of decile 120 installed in dwelling, by region (percent of region 100 100 population) 100 population) 86 88 72 72 73 78 80 62 65 Percent 80 54 60 43 60 40 Percent 20 40 0 20 0 Per capita market income deciles Region 38 Figure B2.1.3: Main source of potable and sanitary- Figure B2.1.4: Type of toilet used by the household, hygienic water supply, by area (percent of the area by area (percent of the area population) population) Rural 49 13 Rural 9 71 Urban 97 0 Urban 91 6 0 20 40 60 80 100 - 20 40 60 80 100 Percent Percent The water supply system installed in the dwelling Own flush toilet connected to the sewerage system The water system tap in the yard or vicinity Shared flush toilet connected to the sewerage system The well in the yard or vicinity Flush latrine not connected to the sewerage system (connected to the river, chan Natural spring in the yard or vicinity Pit latrine periodically cleaned or finally filled up and buried Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES 2.5. Education Spending Georgia’s education system benefits appear to be higher education, depending on the region, area, relatively less unequal for the general education gender, and quintile. While boys outnumber girls in level compared to preschool and higher education. preschool enrollment, females show greater In general education, which includes primary and participation in higher education. Despite variations secondary education, there are minimal differences in the quality of education, such differences are less between the bottom and top quintiles for the net pronounced than those observed in OECD enrollment rate (Figure 2.7). 31 However, net countries. 32 enrollment disparities are evident in preschool and 31 32 Net enrollment rate represents the percentage of boys and In Georgia, students from higher socio-economic backgrounds girls within the official age group who are enrolled in a specific (the top 25 percent) scored 65 points higher in mathematics level of education. than their disadvantaged peers (the bottom 25 percent). This difference is smaller than the average gap of 93 points observed across OECD countries (OECD 2023). 39 Figure 2.7: Net enrollment rates By region By area 100 95 95 95 93 93 100 94 92 90 90 80 80 67 70 70 60 60 56 49 51 49 51 47 Percent Percent 50 45 50 41 41 39 40 34 40 30 22 30 23 20 20 10 10 0 0 Preschool General Higher Preschool General Higher Education level Education level Tbilisi Shida Kartli Shida Kartli Adjara A.R. Other Urban Rural By quintile By gender 120 100 94 93 90 100 95 95 93 93 91 80 70 80 60 59 48 Percent 46 Percent 60 51 54 50 44 50 51 42 45 43 38 40 40 28 30 22 20 20 10 0 0 Preschool General Higher Preschool General Higher Education level Education level Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Female Male Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Note: Net enrollment rate is the ratio of the number of children enrolled in that level of education to the total population of children in that age group Education spending is concentrated in the poorest 31 percent of benefits. In relative terms, education households. The high concentration of education spending is 89 percent of income for the bottom 20 spending in poorer households is for general percent of the population, driven mainly by general education. General education benefits are pro-poor, education spending (Figure 2.8). In contrast, tertiary and it had a higher budget than the other education and vocational education benefits are concentrated levels in 2022 (see Table 1.2). The poorest 40 percent among the top 40 percent of the population at 54.1 of the population receives 47 percent of general percent and 49.3 percent of the respective budgets. education benefits, while the richest 40 percent gets 40 Figure 2.8: Incidence and concentration share of education spending Incidence of Education (% of pre-fiscal Concentration share of education (percent of 100 income) budget) General Preschool Vocational 33 80 35 Tertiary Vocational Tertiary 30 30 Preschool 24 24 25 60 24 23 22 Percent Percent General 25 21 21 2120 19 19 20 16 40 15 14 15 12 10 10 7 20 5 0 0 1 2 3 4 5 1 2 3 4 5 Per capita Market income quintiles Per capita Market Income quintiles Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Most of the education inequality reduction impact followed by preschool education. Vocational and is driven by general education. The impact of higher education have non-significant effects on education spending on reducing inequality is most reducing inequality in Georgia (Figure 2.9). significant for general education (3.2 Gini points), Figure 2.9: Marginal contribution of education spending to inequality reduction, by level Gini points 0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 General 3,2 Education levels Preschool 0,8 Higher education 0,1 Vocational 0,0 Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology Significant improvements are needed to reduce total budget going to urban areas instead of rural the regional disparities, particularly in preschool areas (Figure 2.10). 33 The budget share of urban and tertiary education. Regional disparities exist residents for tertiary and preschool education is for the distribution of benefits at the preschool 75.2 percent and 63.5 percent, respectively. and tertiary education levels, with most of the 33 Most of the population lives in urban areas, where there are proportion of children in rural areas attend public schools at also a higher number of children. However, a greater the general education level. 41 Figure 2.10: Concentration of education spending, by quintile, by region Preschool education (% of the benefits by General education (% of the benefits by quintile and area) quintile and area) 20 16 17,3 13,6 13,3 18 14 11,912,2 12,2 16 12,8 12 14 12,5 10,0 10,8 10,9 10 8,9 8,8 12 10,0 Percent Percent 9,3 10 8 6,6 6,4 5,6 8 6 6 3,6 3,4 4 4 2 2 0 0 1 2 3 4 5 1 2 3 4 5 Per capita market income quintiles Per capita market income quintiles Urban Rural Urban Rural Tertiary education (% of the benefits by Vocational education (% of the benefits by quintile and area) quintile and area) 20 16 17,3 13,6 13,3 18 14 11,912,2 12,2 16 12,8 12 14 12,5 10,0 10,8 10,9 10 8,9 8,8 12 10,0 Percent Percent 9,3 10 8 6,6 6,4 5,6 8 6 6 3,6 3,4 4 4 2 2 0 0 1 2 3 4 5 1 2 3 4 5 Per capita market income quintiles Per capita market income quintiles Urban Rural Urban Rural Source: World Bank staff calculations based on HIES 2022, fiscal administrative data, and the CEQ Methodology 42 2.6. Gender Equity in Fiscal Policy Gender equity in fiscal policy is critical for gender impacts because the burdens and benefits inclusive and sustainable growth, poverty created by fiscal policies accrue to activities reduction, and welfare. The fiscal system affects rather than to individuals or their characteristics. women and men differently at the micro level Household budget surveys do not reveal through different channels, behaviors, and individual preferences, economic, financial, and interactions. 34 The fiscal system may in turn be social needs. In reality, fiscal policies have affected by gender relations and existing structural heterogenous impacts on individuals’ welfare and gender inequalities in the economy and society. decisions. Policies such as incentives for labor Gender gaps can arise in the outcomes and force participation that influence time allocation opportunities enjoyed by men and women across for child and elderly care and policies such as several dimensions, including education, earnings, legal restrictions or non-legal barriers to women’s access to productive assets, political access to education, healthcare, financial services, representation, and bargaining power within the and labor force participation affect each member household. Gender gaps affect individual well- of the family differently. A growing consensus being at the micro level, as well as economic has emerged among researchers that the family institutions, inclusive growth, and poverty utility function and aggregation of consumption reduction at the aggregate level. 35 Promoting and income do not accurately represent the gender equality can therefore play an important consumption behavior of individuals within the role in boosting economic productivity and unit household. 38 It is important to highlight the growth, enhancing economic resilience, and gender gaps and disparities existing within the reducing overall income inequality. 36 households. Standard FIA treats the household as a unitary Before examining the roles of gender in income entity where all its members have equal (per distribution and fiscal incidence, it is important to capita) access to income and consumption, or understand the broader age and gender they are weighted by an externally given composition of Georgia’s population. The parameter and income and consumption are demographic pyramid for the year 2024 shows the expressed in adult equivalent units. The CEQ total population of 3.69 million in 2024, which has methodology 37 estimates the combined impact of declined from 4.93 million in 1994 (Figure 2.11). Of the taxes and transfers on the household’s income the total population, 48 percent are male, and 52 and welfare at different stages at an aggregate percent are female. The ratio of the males to level. CEQ assessments typically do not assess females is declining with age, most notably for the 34 Gender is “a normative social construct defining and pro-poor are specific taxes and government spending? How differentiating the roles, rights, entitlements, responsibilities, and effective are taxes and government spending in reducing social obligations of women and men; it also provides positive inequality and poverty? What is the impact of fiscal reforms and negative incentives for broad-based compliance with those that change the size and/or progressivity of a particular tax or roles, responsibilities, and obligations” (E-CEQ 2022). benefit? The methodology has been extended to and adapted for over 85 low- and middle-income countries over the past 35 World Bank (2012). decade. For methodological details, see Lustig (2018). 36 38 Hsieh et al. (2019), IMF (2013), IMF (2018). For example, Phipps and Burton (1995, 1996, 1998), Browning et al. (1994), Findlay and Wright (1996), Lundberg and Pollak 37 The CEQ methodology aims to address four key questions: (1996), Cherchye, Demuynck and De Rock (2011), Thomas How much income redistribution and poverty reduction is (1990), and Lee (2007). being accomplished through fiscal policy? How equalizing and 43 over-80 age group, of which only 35.0 percent are demographic change occurring in Georgia. The males. Population aging (that is, the increase in the share of population age 60 and above as a population above age 60 in absolute terms, both as percentage of total population has increased from relative to the younger population and as a 16.2 percent in 1994 to 22.9 percent in 2024, 60 proportion of the total population) is a significant percent of which are women. 39 Figure 2.11: Population pyramid for Georgia, 2024 Population Pyramid Georgia, 2024 85+ 34 13 80-84 48 22 75-79 64 35 70-74 103 65 65-69 124 89 60-64 140 111 55-59 121 107 50-54 117 110 Age group 45-49 119 115 40-44 125 122 35-39 137 137 30-34 134 136 25-29 109 118 20-24 103 113 15-19 95 108 10-14 112 122 5-9 130 139 1-4 86 92 0 19 21 Population (in thousands) Female Male Source: National Statistics Office of Georgia An Engendered CEQ (E-CEQ) Georgia has been characteristics. It analyzed various gender conducted to explore the effects of the fiscal inequalities such as those in employment, system in promoting gender equality while entrepreneurship, and other opportunities created reducing poverty, vulnerability, and exposure to by the fiscal policy as presented in Figure 2.12. livelihood risks. The analysis reviewed the The E-CEQ methodology recommends classifying interaction of the fiscal system with gender and households into different typologies to highlight intra-household resource dynamics. The gender dimensions and potential gender methodology applied is by Fuchs Tarlovsky and disparities. It aims to explore different access Gonzalez Icaza (2023). The E-CEQ Georgia points by identifying and quantifying the fiscal leveraged the CEQ Georgia 2022 analysis to benefits and burdens experienced by women and estimate fiscal policy impacts at different income men. The basic framework for constructing the levels on households according to their gender main typologies for the E-CEQ Georgia is composition, age, and employment-based summarized in Table 2.1. 39 National Statistics Office Georgia, data on population by age and sex from 1994 to 2024. 44 Figure 2.12: Interaction of the fiscal system with gender and intrahousehold dynamics Gender disparities and Household incomes and interactions with fiscal system intrahousehold dynamics Households as suppliers of Market income labor, taxpayers, and income Gross wages and salaries, income from capital, private recipients transfers, income from own production • Decision to supply labor in formal & informal markets • Household assets and access Market income + Pensions to inflows Contributory pensions treated as deferred income • Eligibility to SA/SP affected by demographics and Direct transfers Direct taxes and E.g. Direct social contributions reproductive decisions assistance (SA), E.g. PIT, other • Time allocation for care & non-contributory Disposable income taxes, housework as input for labor pensions income SSCs participation Households as consumers of goods & services, subsidy recipients, and taxpayers Indirect subsidies Indirect taxes • Consumption needs affected E.g. Subsidies on E.g. VAT, excises, by gender & demographics food, energy, customs utilities, etc. • Eligibility may be affected by demographic and Consumable socioeconomic type Households as caregivers, investors in assets, and risk managers Payments for In-kind transfers • Perceived returns and public services willingness to invest in E.g. Education and E.g. Co-payments, health human capital accumulation fees • Health needs affected by gender and age • Time allocation for care and Final income housework (Care needs of young children and elderly; potential help from elderly) Source: ECEQ methodology adapted from Fuchs et al. (2023) and Lustig (2022) 45 Table 2.1: Household typologies Typology Household types Demographics only (as a proxy for needs and opportunities) (T1) Gender majority A. Majority of female adults B. Majority of male adults C. Equal share Demographics + income (proxy for intra-household roles and bargaining power) (T2) Gender of household A. (Labor) Earners are a female majority earners in labor markets B. (Labor) Earners are a male majority C. Equal number of male and female labor earners D. No (labor) earners in the household (T3) Gender sustaining A. Larger share of female earnings (>60 percent of household labor earnings) household earnings B. Larger share of male earnings (>60 percent of household earnings) C. No majority share (<60 percent share of household earnings) D. No earners in the household Source: ECEQ methodology adapted from Fuchs et al. (2023) and Lustig (2022) Analysis suggests that households sustained earners, while 34.9 percent are in households with mainly by male-earned labor incomes are the only male earners (Panel B). Households in which most predominant category of households in more than 60 percent of earnings in the households Georgia. The distribution of household typologies are being contributed by males are in the majority, across income groups and across urban and rural with a share of 40.6 percent of total households. locations is described in Table 2.2 below. The Only one-fifth of households are sustained mostly purely demographic types under Typology 1 (T1) of (>60 percent) by female-earned labor incomes. This the majority of adults are evenly distributed across is consistent with the fact of low female labor force income groups and locations. Only 23.0 percent of participation and therefore a low female share in Georgians live in households with only female total earning. Table 2.2: Household typologies in Georgia Panel A: Share of household type in income groups and location Household typology All Quintile Quintile Quintile Quintile Quintile Bottom- Top-60 Urban Rural 1 2 3 4 5 40 Gender of household adult members (T1) Members are female majority 100.0 19.7 19.2 21.0 20.0 20.0 38.9 61.1 57.6 42.4 46 Members are male majority 100.0 19.6 19.1 21.3 18.5 21.4 38.8 61.2 51.2 48.8 Equal number of females and males 100.0 20.4 21.2 18.2 20.7 19.5 41.6 58.4 47.7 52.3 Gender of household earners (T2) Earners are female only 100.0 16.9 21.6 23.6 17.0 20.8 38.6 61.4 55.2 44.8 Earners are male only 100.0 14.0 19.9 19.7 23.4 23.0 33.9 66.1 51.2 48.8 Earners are both female and male 100.0 14.1 17.3 20.2 24.4 24.0 31.4 68.6 35.5 64.5 No labor earnings 100.0 48.5 22.6 14.4 9.0 5.5 71.1 28.9 79.1 20.9 Gender sustaining household earnings (T3) Female (>60% earnings) 100.0 12.8 20.6 24.8 18.8 23.1 33.4 66.6 63.8 36.2 Male (>60% earnings) 100.0 8.1 18.4 19.7 26.9 26.9 26.5 73.5 52.0 48.0 No majority 100.0 6.0 11.1 20.4 27.4 35.1 17.1 82.9 59.0 41.0 No labor earnings 100.0 6.0 11.1 20.4 27.4 35.1 17.1 82.9 59.0 41.0 Panel B: Distribution of household types across income and location Household typology All Quintile Quintile Quintile Quintile Quintile Bottom- Top-60 Urban Rural 1 2 3 4 5 40 Gender of household adult members (T1) Members are female majority 37.6 37.0 36.1 39.6 37.6 37.5 36.6 38.3 41.5 33.3 Members are male majority 21.1 20.8 20.2 22.6 19.6 22.5 20.5 21.6 20.7 21.6 Equal number of females and males 41.3 42.2 43.7 37.8 42.8 40.0 42.9 40.2 37.8 45.1 Gender of household earners (T2) Earners are female only 23.0 19.4 24.8 27.2 19.6 23.8 22.1 23.5 24.3 21.5 Earners are male only 34.9 24.4 34.7 34.5 40.8 39.9 29.5 38.4 34.2 35.6 Earners are both female and male 26.8 18.9 23.2 27.3 32.8 32.1 21.0 30.7 18.3 36.2 No labor earnings 15.3 37.3 17.3 11.1 6.9 4.2 27.3 7.4 23.3 6.7 Gender sustaining household earnings (T3) Female (>60% earnings) 19.2 12.4 19.7 23.9 18.1 22.1 16.1 21.3 23.5 14.6 Male (>60% earnings) 40.6 16.5 37.3 40.1 54.6 54.4 26.9 49.7 40.5 40.8 No majority 6.4 1.9 3.6 6.6 8.8 11.2 2.7 8.9 7.3 5.5 No labor earnings 33.7 69.2 39.3 29.4 18.5 12.2 54.3 20.1 28.8 39.2 Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES 47 The male-majority household is the predominant compared to 8 percent) and urban areas (20 category in the overall population, the poor percent compared to 14 percent) (Figure 2.13). population, and urban areas. The proportion of Households without labor income are more male-majority households is significantly higher prevalent in urban areas than in rural ones. than the female majority in rural areas (13 percent Figure 2.13: Household typologies by region Types of households by region (as percentage Household demographics (percent) of total) 25 40 33,4 20 35 31,1 30,0 20 30 16 24,0 14 14 25 22,8 23,7 15 13 21,2 Percent Percent 11 20 8 13,8 10 15 10 5 2 5 0 0 Female Male No majority No labor Female Male No majority No labor majority majority income majority majority Urban Rural Population Poor Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES Households with no labor earnings are the proportion of households where women earn the poorest in all the categories (T3 in Table 2.3). As majority of income is about 18.8% for poor expected, households with no labor earners are households and 20.0% for non-poor households, overrepresented among the poor: 45.4 percent of accounting for approximately one-fifth of all the poor households have no labor earnings. The households in both cases. Table 2.3: Pre-fiscal poverty, by household typologies Household typology Non-poor Poor All Gender of household adult members (T1) Members are female majority 38.8 36.9 37.6 Members are male majority 19.6 22.1 21.1 Equal number of females and males 41.6 41.1 41.3 Gender of household earners (T2) Earners are female only 21.5 23.8 23.0 Earners are male only 41.2 31.0 34.9 Earners are both female and male 32.8 23.2 26.8 No labor earnings 4.4 21.9 15.3 48 Gender sustaining household earnings (T3) Female (>60% earnings) 20.0 18.8 19.2 Male (>60% earnings) 55.5 31.7 40.6 No majority 10.1 4.2 6.4 No labor earnings 14.4 45.4 33.7 Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES Poorest households have a greater burden of dependents and elderly. Poorest households have dependents both under age 5 and 6-14 years of more care responsibilities for the elderly as well as age. Household care categorization reflects that dependents (Figure 2.14). the richest households have the smallest burden of Figure 2.14: Household typologies, by household care categories Household care responsibility categorization, Household care responsibility categorization, by quintile 50 by category 50 40 40 Percent Percent 30 30 20 20 10 10 0 0 1 2 3 4 5 Dependents and Dependents, no Elderly but no No dependants, elderly elderly dependants no elderly Dependents and elderly Dependents, no elderly Elderly but no dependants No dependants, no elderly 1 2 3 4 5 Household care responsibility -dependants, by category 30 25 20 Percent 15 10 5 0 Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Per capita Market income quintiles Dependants under 5 years Dependants 6-14 years Dependants both under 5 and 6-14 No dependants Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES 49 The results indicate that although households with poverty and inequality reduction thanks mainly to care responsibilities for elders are poorest, fiscal the fiscal system (largely pensions and healthcare). policy has been effective in mitigating that impact. Social transfers for children have a smaller impact The analysis shows that households with only elderly on poverty and inequality reduction compared to members and no children experience the greatest pensions for the elderly 40 (Figure 2.15). Figure 2.15: Poverty and inequality by household categories Headcount poverty (by household category) Gini index (by household category) 45 60 40 50 35 30 40 Gini index 25 Percent 30 20 15 20 10 10 5 0 0 Single Only Elders and Only elders, No elders / Single Only Elders and Only elders, No elders / household children children no child no children household children children no child no children Per capita market income quintiles Per capita market income quintiles Market income Consumable income Market income Consumable income Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES The health needs of the Georgian population sensitive needs of healthcare by age. Figure 2.16, vary by gender. Health expenditures are which shows health expenditure per capita important components in many Georgian disaggregated by gender and by age group, households’ overall consumption bundles. A indicates that females over 60 years of age have substantial share of health transfers occurs higher per capita spending on health than males. through government provision, therefore The higher average life expectancy of females allocating this consumption across individuals may be the reason behind the per capita differentiated by gender and age is potentially difference. In the 15–24 age bracket, women important to gain a more accurate picture of the receive more funding than males, which could overall incidence impact of healthcare spending. stem from the reproductive healthcare needs of It is also important to understand the gender- women. 40 See Figures 1.4 and 1.7 and Section 1.1. 50 Figure 2.16: Health expenditure per capita by gender and age group, 2022 700 650 600 GEL 550 500 450 400 Age group Female Male Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES Low female labor force participation results in a of women are not engaged in the formal labor smaller share of households being sustained market despite having better education (college mainly by female-earned labor income. While and university levels) than males. The economic Georgia has made substantial progress in cost of the gender gap in labor force inclusion, significant gaps in labor participation participation is estimated to be 11 percent of persist. Georgia’s ranking on the Global Gender GDP per capita. 42 The gender gap affects Gap Index (GGGI) 41 worsened from 55 in 2022 to entrepreneurial activities and the private sector: 76 in 2023, lagging behind most European in 2019, women represented only 30 percent of countries. Female labor force participation in newly registered business owners. Male Georgia has been stagnant for the last two participation in employment activities is 1.7 decades, in contrast to increasing rates in other times higher than that of females. 43 ECA countries (Figure 2.17). Almost 60 percent 41 42 Developed in 2011 by the World Economic Forum, the GGGI UNDP (2023). measures economic participation, educational attainment, 43 health and survival, and political empowerment. Link Georgia Time Use Survey 2021–22. https://www.weforum.org/publications/global-gender-gap- report-2023 51 Figure 2.17: Female labor force participation Labor Force, female (% of total labor force) 54 52 as % of total labor force 50 48 46 44 42 40 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2020 2021 2022 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Europe and Central Asia Georgia Armenia United Kingdom France Kazakhstan UMIC Source: World Bank Development Indicators (2023) The gender wage gap, unpaid domestic work, the women’s share being 88.3 percent. On and care responsibilities remain the major average, women's participation in unpaid reasons for the low participation of females in caregiving work is 31.0 percent, compared to 14.5 employment (Figure 2.18). Average wages paid percent for males. Among parents, 80.0 percent to women are 36.2 percent lower than those paid of mothers participate in childcare activities, to men. 44 Of Georgia’s total population, 66 twice the participation rate of fathers at 37.6 percent engage in unpaid domestic work, with percent. Figure 2.18: Distribution of individuals that are unemployed, by reason (percent), Georgia 2022 70 66,1 61,5 61,2 60 49,8 50,2 50 38,5 38,8 40 Percent 33,9 30 20 10 0 Old Ill or disabled Studying Care responsibility and other Reasons for staying out of the labor force Male Female Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES 44 ILO (2021). 52 The analysis highlights working age women face ages 25–34 at 52.1 percent. This share is high high unemployment and under-employment compared to men, particularly for ages 25–34 mainly due to domestic care responsibilities. (70.3 percent are employed) (Table 2.4). Of the Unemployment is high among young women ages respondents, 61.2 percent of the females were 15–24 at 81.6 percent of survey respondents and unemployed due to care responsibilities. Table 2.4: Employment by age groups and sex (percent) in Georgia, 2022 Category Ages 15-24 25-34 35-44 45-54 55-64 65+ Total Male Unemployed or unactive* 69.1 29.7 26.0 24.5 27.3 54.4 37.7 Employed 30.9 70.3 74.0 75.5 72.7 45.6 62.3 Female Unemployed or unactive 81.6 52.1 38.2 29.8 42.2 66.4 51.2 Employed 18.4 47.9 61.8 70.2 57.8 33.6 48.8 Total Unemployed or unactive 74.8 40.9 32.2 27.3 35.8 61.8 45.0 Employed 25.2 59.1 67.8 72.7 64.2 38.2 55.0 Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES Note: *The survey does not have a question on whether the household head was “looking for a job,” in the absence of which it was assumed that students and differently abled were unactive due to that reason All fiscal interventions decrease the gender gap for consumable income, which indicates a but only in urban areas. The gender gap was relatively lesser effect of fiscal interventions — calculated as the difference in income between particularly direct taxes and direct transfers — on male- and female-majority households. 45 The female-majority households in rural areas. This is analysis indicates that significant spatial partly due to less engagement of female-majority inequalities exist across regions and that the taxes households in formal employment in rural areas and transfers have been effective in reducing the (Figures 2.19 and 2.20). Gender gaps by quintile gender gap in favor of female-majority households are highest for the 1st and 2nd quintiles mainly due in urban areas. However, the gender gap remains to low engagement with the formal labor force negative in rural areas at all income levels except (Figure 2.21). 45 Gender gap is calculated as follows: gender gap = (average female majority household income – average income of male majority) / average income of male majority household 53 Figure 2.19: Gender gap by income concept 6 4 2 0 Percent -2 -4 -6 -8 Market income Market income plus pensions Disposable income Consumable income Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES Figure 2.20: Gender gap by income concept, urban versus rural areas 10 5 0 Percent -5 -10 -15 -20 Market income Market income plus pensions Disposable income Consumable income Gender gap—urban Gender gap—rural Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES Figure 2.21: Gender gap by income concept by quintile 25 20 15 10 Percent 5 0 -5 -10 Market income Market income plus pensions Disposable income Consumable income Quintile 2 Quintile 3 Quintile 4 Quintile 5 Quintile 1 Source: World Bank staff calculations based on CEQ Georgia 2022 and HIES 54 The results from E-CEQ Georgia indicate that the participation in economic life. Fiscal policies circumstances of women undermine their targeted at children are not currently sufficient productivity and inclusion in formal labor to offset the high incidence of poverty among markets, and women are more likely to have households with children and are hampering lower-paying jobs. The analysis focused on women’s participation in economic life. Women typologies defined by demographic and income experience “time poverty” linked to the burden of compositions. The results suggest that childcare unpaid domestic responsibilities and little or no and elderly care constrain mostly women’s use of childcare services. 55 Conclusions and Key Recommendations 56 Conclusions and Key Recommendations The design and implementation of fiscal policies Including an exemption for low earners would should continue to consider different dimensions help prevent the PIT from increasing poverty. including fiscal sustainability, economic However, reducing tax rates would also result in a efficiency, externalities, and equity. Enhanced loss of tax collection. In order to drive meaningful fiscal equity consideration matters because reform, it is essential to enhance tax economic growth alone is not sufficient to administration and minimize tax compliance eliminate poverty and inequality. The fiscal system expenses to prevent tax evasion and ensure can support this process with the right mix of efficient tax collection that does not adversely investments and redistribution through taxation affect the poor. Compensatory measures to and social expenditures. As the country explores mitigate the effects of PIT on lower-income the mix of taxation and public expenditure policies households could also be explored. to ensure fiscal sustainability, it is important that the design and implementation of reforms Pension reforms are critical for obtaining fiscal maintain a focus on equity. space and managing the fiscal burden of an aging population. Supporting the long-term fiscal Fiscal policy reforms are needed to protect fiscal sustainability of the government-funded pensions space, either through domestic revenue scheme will require structural changes, especially mobilization or public expenditure efficiency, to in the labor market, to counteract the negative be able to expand social expenditure and effects of population aging. Addressing high redistribution measures while managing unemployment, particularly among youth, increasing demands for old-age pensions. encouraging female employment, boosting labor Domestic revenue mobilization reforms — for productivity, and decreasing informality could instance, the PIT reform — could align the contribute to improved fiscal performance and objectives of fiscal sustainability with fiscal equity collect more funds for pension funds when the if additional revenues are used to compensate population is active. Several policy reforms could targeted households with social protection benefit countries such as Georgia undergoing transfers. There are several policy options for population aging, including postponing the generating additional tax revenues, including retirement age and employing older people for shifting the tax mix toward greater reliance on longer. Increasing the fertility rate would also help indirect taxes — in particular, revisiting VAT modify demographic trends. Another policy exemptions (for example, on imported vehicles). direction that could play an important role in Increasing health taxes on tobacco and alcohol and supporting fiscal sustainability is improving the raising the income tax rate on property income efficiency of social spending. and other passive income can also expand the fiscal space. Meanwhile, public expenditure efficiency In the coming years, it will be important to reforms could align fiscal sustainability and fiscal consider how pensions are adjusted to meet the equity if government spending is targeted and growing fiscal burden and protect the old-age progressive. population from falling into poverty. Despite 57 efforts to introduce employer- and employee- sector to achieve SDG 6 and provide equal access funded pension schemes, the government's role in to the entire population is critical. providing pensions remains vital. Population aging is associated with a greater need for redistribution, In terms of advancing gender equality objectives, posing a burden on fiscal policy, placing upward gender-responsive practices should be pressure on pensions and age-related spending, and institutionalized in public policies. The burden of having adverse effects on tax bases and the unpaid domestic care responsibilities falls structure of public revenues. disproportionally on women. Adopting structural fiscal measures such as government support for On the spending side, while education spending is early childcare and education as well as active progressive, significant improvements are needed labor market programs is anticipated to have to reduce regional disparities, particularly in significant impacts on women’s economic preschool and tertiary education. The education participation. Policies could include additional system needs to be more effective in preparing subsidies and government financial support for students for the demands of the modern workforce, childcare and early education, and more public increase the quality of education, and implement childcare services. More flexible work policies and programs that ensure fair access to arrangements for the female labor force, longer quality education across all regions of the country. paid maternity leave, and active labor market In this regard, improving the quality of education — programs could also encourage women to join the particularly in rural areas — is critical to reduce labor market. This approach will promote women's disparities. Promoting rural development programs empowerment, enhance their participation in the with increased focus on education, healthcare, and economic and social spheres, and foster sustainable employment opportunities can support economic growth. development in rural Georgia. In light of the poverty levels among the rural To enhance fiscal efficiency, Georgia should population, policies addressing rural poverty and reassess its indirect subsidy allocations. Subsidies development are an important priority. The FIA should be evaluated for their effectiveness in results show a significant difference in spending achieving poverty reduction and redistribution levels on education and subsidies, favoring urban objectives. Rationalizing subsidy allocations and residents. To better address rural poverty, mitigation measures are key areas with high social Georgia should enhance targeted development and economic impact that will help to ensure fiscal programs that improve access to education, sustainability and efficiency. In this regard, quality of education, healthcare, and employment allocation of subsidies to the water and sanitation opportunities in rural areas. 58 References 1. Anand, S., and P. Segal. 2015. “The global distribution of income.” 937-979. https://doi.org/10.1016/b978-0-444-59428-0.00012-6 2. Arndt, C., and K. Mahrt. 2017. “Is inequality underestimated in Mozambique? Accounting for underreported consumption.” https://doi.org/10.35188/unu-wider/2017/379-0 3. Aziz, O., N. Gemmell, and A. Laws. 2013. “The Distribution of Income and Fiscal Incidence by Age and Gender: Some Evidence from New Zealand.” April 30, 2013. Victoria University of Wellington Working Paper in Public Finance No. 10/2013, Available at SSRN: https://ssrn.com/abstract=2375926 or http://dx.doi.org/10.2139/ssrn.2375926 4. Bakradze, N., S. Basilidze, and N. Nurashvili. 2023. Household income and income inequality in Georgia (2013-2022). PMC Research Center. https://pmcg-i.com/app/uploads/2023/09/Household-Income-and-Income-Inequality-in-Georgia- 2013-2022-ENG.pdf 5. De La Fuente, A., and M. Cabrera. 2023. Poverty and Distributional Impact of Fiscal Policy in Dominican Republic. World Bank. https://documents.worldbank.org/en/publication/documents- reports/documentdetail/099454211202324336/IDU0a89bf1940a00093e30adef81507a10 6. Fuchs Tarlovsky, A., and M. Gonzalez Icaza. 2023. Assessing the Effects of the Fiscal System on Gender Disparities in Armenia: Pilot-Study for the Engendered CEQ Methodology. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/39903 7. Fuchs Tarlovsky, A., D. Sharma, and M. Gonzalez Icaza. 2020. Informality in Georgia a micro-level assessment. Washington, DC: World Bank. 8. Georgia: Public Budget. 2022. “Performance report budget 2022, Chapter 14. By function.” Ministry of Finance. https://www.mof.ge/en/5642 9. Parliament of Georgia. 2010. Tax Code of Georgia. https://matsne.gov.ge/en/document/view/1043717 10. Hlásny, V., and P. Verme. 2021. “The impact of top incomes biases on the measurement of inequality in the United States.” Oxford Bulletin of Economics and Statistics 84(4): 749-788. https://doi.org/10.1111/obes.12472 11. Jirasavetakul, L., and C. Lakner. 2019. “The distribution of consumption expenditure in Sub-Saharan Africa: the inequality among all Africans.” Journal of African Economies 29(1): 1-25. https://doi.org/10.1093/jae/ejz016 12. Lara Ibarra, G., M. Cabrera, O. Canozzi Conceicao, and R. Campante Cardoso Vale. 2023. Poverty and Inequality Implications of Fiscal Policies: The Case of Brazil. WPS 10495. Policy Research Working Paper. World Bank. https://documents.worldbank.org/en/publication/documents- reports/documentdetail/099423406222337139/IDU07d8faca60e1940410d093030b2726e376e29 59 13. Lustig, N., ed. 2022. Commitment to equity handbook: Estimating the impact of fiscal policy on inequality and poverty. CEQ Institute and Tulane University and Brookings Institution Press. 14. OECD (Organisation for Economic Co-operation and Development). 2023. Georgia | Factsheets | OECD PISA 2022 results. OECD. https://www.oecd.org/publication/pisa-2022-results/country-notes/georgia-09138858 15. Pinkovskiy, M., and X. Sala-i-Martin. 2014. “Lights, camera,...income! Estimating poverty using national accounts, survey means, and lights.” SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2423393 16. Prydz, E., D. Jolliffe, and U. Serajuddin. 2022. “Disparities in assessments of living standards using national accounts and household surveys.” Review of Income and Wealth 68(S2). https://doi.org/10.1111/roiw.12577 17. Robayo-Abril, M., and M. Cabrera. 2024. Reassessing Welfare Impacts of Bulgarian Fiscal Policy through a Child Poverty Perspective. WPS 10657. Policy Research Working Paper. Washington, DC: World Bank. https://doi.org/10.1596/1813-9450-10657 18. Rodriguez, L., and M. Wai-Poi. 2021. Fiscal Policy, Poverty and Inequality in Jordan. The Role of Taxes and Public Spending [Policy summary]. World Bank. 19. UNECE. 2022. “Brief Report on The Progress Under the Protocol on Water and Health.” UNECE. https://unece.org/sites/default/files/202204/Georgia_summary_report_5th_cycle_26Apr22_ENG.pdf 20. Wai-Poi, M., M. Sosa, and P. Bachas. 2024. Taxes, Spending and Equity: International Patterns and Lessons for Developing Countries [Unpublished working paper]. World Bank. 21. World Bank. 2023. Rising Strong: Peru Poverty and Equity Assessment. https://www.worldbank.org/en/country/peru/publication/resurgir-fortalecidos-evaluacion-de- pobreza-y-equidad-en-el-peru 22. World Bank. 2012. World Development Report 2012: Gender Equality and Development. © World Bank. http://hdl.handle.net/10986/4391 License: CC BY 3.0 IGO. 23. World Bank. Forthcoming. “Georgia: Fiscal Policy for Inclusive Growth. Public Finance Review.” Washington, DC: World Bank Group. 60 Appendix A 61 A P P E N D I X A Data and Methodology a. Concepts and methodology National Statistics Office of Georgia (GeoStat). The HIES 2022 is a 12-month survey covering 13,621 households. It has A fiscal incidence analysis measures how the Government’s taxes and social expenditures information on income, employment, affect households’ welfare. The goal of such an consumption, health, and education. analysis is to answer questions regarding the • Tax legislation and comprehensive impact of taxes and social expenditures on administrative data on direct and indirect poverty and inequality, the progressivity of taxes taxes, tax rates, tax collection, and and social expenditures, and which households government budget execution data from the are net payers or net receivers in the fiscal MoF and World Bank teams. system. By developing detailed and parametrized • Social protection data from the main social microsimulation models for fiscal incidence programs on budget, beneficiaries, and analysis, the analysis creates a platform that can allocation rules from the Ministry of be used not only to assess the distributional Internally Displaced People from Occupied impacts of existing fiscal policies but also to Territories, Labor, Health, and Social simulate the ex-ante distributional impact of Affairs (MoILHSA) and the World Bank potential policy reforms. These models can Social Protection team. therefore be particularly useful for informing • Data to calculate the subsidy value for evidence-based fiscal policy design, looking at domestic electricity distribution and fuel equity considerations both before and after from the Ministry of Economy and implementation. Sustainable Development and the World Bank Energy team. b. Data sources • Education expenditures and student enrollment from the Ministry of Education, The fiscal incidence analysis in Georgia used the Science and Youth (MoES) and World Bank following data sources: Education team. • The most recent nationally representative • Health expenditure and health visits from household survey, the 2022 Georgia the Ministry of Labor, Health and Social Household Income and Expenditure Survey Defense and the World Bank Health team. (HIES), which is a multi-purpose survey on • The 2020 Labor Force Survey (LFS) on labor household budget and living standards used informality and public sector employees, to estimate official poverty statistics for also implemented by GeoStat. Georgia. The HIES was implemented by the 62 c. Commitment to Equity (CEQ) Tax-benefit Microsimulation Model for the European Union (EUROMOD), cover fewer fiscal framework interventions – for instance, in-kind benefits are typically excluded. The CEQ methodology has been The 2022 fiscal incidence analysis for Georgia was implemented in over 70 countries, which facilitates developed using the Commitment to Equity (CEQ) the production of results that are internationally methodology, 46 which provides a systematized comparable. framework for determining the impacts of the fiscal system on poverty and inequality. A CEQ The first step of fiscal incidence analysis is Assessment is a rigorous and standardized fiscal allocation of the main taxes and transfers to incidence methodology that permits systematic households in the survey. The main modeling analysis of several fiscal interventions that affect techniques in this report include direct households’ welfare (direct taxes, indirect taxes, identification, inference, imputation, and direct transfers, indirect subsidies, and in-kind simulation. These concepts are described below in benefits from health and education) and simulates Box A.1. Typically, fiscal interventions such as direct how fiscal systems work in practice. CEQ uses a taxes, indirect taxes, and indirect subsidies are common framework developed by the CEQ Institute modeled based on simulation, combining tax rules and presented in the CEQ Handbook (Lustig 2022). and characteristics of individuals’ employment and Other fiscal incidence methodologies, such as the consumption patterns, respectively. Box A.1: Tax and transfer allocation methods The CEQ methodology allocates each tax and transfer by drawing from a set of methods described in detail in Chapter 3 of the CEQ Handbook 2022, namely: • Direct identification: The survey asks whether a tax (transfer) was paid (received), and the amount. • Inference: Whether a tax (transfer) was paid (received), or the amount, is determined based on administrative data on the taxes (transfers) and other relevant secondary information self-reported by the household. • Imputation: Taxpayers (beneficiaries) are directly identified in the survey, but the amount of a tax (subsidy) paid (received) is imputed based on program rules. • Simulation: Both the identification of taxpayers (beneficiaries) and the amount of a tax (subsidy) paid (received) are estimated based on program rules. • Prediction: A regression model is applied to predict who receives benefits or pays taxes, and/or the amount. Source: Adapted from Chapter 3 of Lustig (2022) 46 The Commitment to Equity project (CEQ) is led by Nora Lustig at Tulane University. The latest CEQ Handbook was published in 2022 and is available online. More information at: http://commitmentoequity.org/publications-ceq-handbook 63 The building block of fiscal incidence analysis is the construction of income concepts. Conceptually, for each household, the analysis starts with “market income” or “market income plus pensions” as the pre-fiscal income (any private income that excludes any government intervention). 47 Then taxes are subtracted and transfers are added to obtain households’ final income or post-fiscal income (after government’s taxes and transfers). After all the taxes and transfers are modeled, the CEQ methodology calculates different income concepts for each household to assess how fiscal policy affects households’ income at various stages of redistribution. Once completed, the distributional impacts of any tax or transfer decisions on the welfare of a country’s households can be estimated (Figure A.1). Figure A.1: CEQ income concepts in the Pensions as Government Transfers scenario for Georgia - Direct transfers + Indirect subsidies Pensions, TSA, Electricity, gas, + In-kind transfers Veteran, Disability, transportation, Education, health others water Disposable income = Pre-fiscal income Net market Consumable Official Final income = Market income income income consumption from HIES + Direct taxes - Indirect taxes Personal income VAT, excise tax tax, property tax Source: Adapted from Lustig (2022) Building a fiscal incidence model based on the CEQ methodology requires understanding how a country’s fiscal system works and allocating the taxes and social expenditure across households and individuals, using microdata from a representative socioeconomic household survey. In countries where consumption is a more reliable measure of welfare than income (as in Georgia), the analysis starts by setting disposable income equal to the consumption welfare aggregate and adds or removes certain fiscal interventions to calculate the other income concepts. Given that disposable income already includes certain interventions of the fiscal system, the analysis can move “backward” to remove fiscal interventions until reaching “pre-fiscal income,” a measure of households’ total income from private sources only. Also, the analysis works “forward” from disposable income, adding additional fiscal interventions until reaching “post-fiscal income” (income once all taxes and transfers modeled here have been included). Figure A.1 displays the different steps of the income construction process. 47 Pre-fiscal income could be equivalent to market income or market income plus pensions depending on the pension scenario chosen for the baseline. 64 In the case of Georgia, this fiscal incidence analysis used the Pensions as Government Transfers (PGT) scenario from the CEQ framework. In this scenario, pensions are treated as a government transfer – a tool for redistribution from one segment of society to another. In the PGT scenario, the pre-fiscal income concept is market income. 48 The alternative to the PGT scenario is the Pensions as Deferred Income (PDI) scenario, in which pensions are treated as saved private income. In the PDI scenario, the pre-fiscal income concept is market income plus pensions (see sensitivity analysis under PDI scenario in Appendix B). A CEQ Assessment has two scenarios for treating pensions. As mentioned earlier, the baseline scenario treats pensions as a government transfer, a tool for redistribution from one segment of society to another by taxing individuals through pension contributions and transferring benefits through pension income. An alternative scenario treats pensions as saved private income (as a result from previous savings) spent later in the form of PDI. Both scenarios present extremes, and reality is always some combination of the two. This analysis uses PGT as the main scenario because in Georgia, old-age pensions remain mainly non- contributory. Actual pension transfers are funded by tax collection, and the Georgian pension system redistributes from tax-paying individuals to pension recipients. Nonetheless, treating pensions solely as a government transfer means that those relying solely on a pension will appear to have zero market income, thus overestimating the proportion of the poor (Lustig 2022). Different post-fiscal income concepts are used for calculating the impact of fiscal interventions on poverty and inequality. For calculating the total effect on inequality reduction, the post-fiscal income is the final income concept — that is, the total effect is the difference in inequality levels between market income and final income. However, when calculating total poverty reduction, the post-fiscal income is the consumable income concept — that is, the total effect is the difference in the poverty rate between market income and consumable income. One reason for this is that in-kind health and education benefits are allocated at the average government cost of provision, which provides little information about how the transfers improve a household’s purchasing power or their monetary poverty status. Another reason is that the publicly financed portions of health and education services consumed by households are generally not included in either the consumption aggregate or the consumption-based poverty line. 48 As a result, pension income is subtracted from disposable income along with the direct transfers to get to net market income, and pension contributions are added with the direct taxes to reach market income. 65 1. Limitations 2. Methodologies for Allocating Taxes and The CEQ Assessment framework for fiscal Transfers in Georgia incidence analysis is a static and retrospective accounting exercise without behavioral, lifecycle, or general equilibrium effects. Major This section describes in detail the allocation challenges include the failure to capture top- methodologies used to model each tax and income households 49 and the exclusion of some transfer included in the analysis. interventions from the model — such as the corporate income tax (CIT), 50 public expenditure on infrastructure, 51 defense, or debt interest a. Coverage of fiscal incidence payments — due to methodological limitations. analysis This means the incidence results represent an accounting-based first approximation of a true The 2022 fiscal incidence analysis in Georgia counterfactual. 52 In practice, there are no modeled the core fiscal interventions from the standard errors calculated which would allow a country’s tax system and public social statistical assessment of the allocations made to expenditure. On the tax side, the study modeled individuals and households. 53 Nevertheless, the PIT, property tax, value added tax, and excise CEQ Assessment framework for fiscal incidence duties. Together, the total direct and indirect taxes analysis provides a standard methodology (which included for potential allocation to individuals and also enables international comparisons) for households in the microdata represented about estimates of the impact of fiscal policy on poverty, GEL 15.1 billion, equivalent to 77.9 percent of total inequality, and social welfare more generally. government tax revenues and 20.9 percent of the country’s GDP in fiscal year 2022. Some categories of tax revenues were excluded from the analysis due to data limitations. On the social expenditure side, the analysis covered the main social protection programs (pensions, TSA, disability), indirect subsidies (domestic electricity, gas, and transport), health expenditure, and education expenditure (preschool and general, vocational, and higher). Altogether, the total social expenditure included for potential allocation to individuals and 49 Household surveys also do not capture some low-income water infrastructure (publicly provided water), and local households very well either, such as institutional populations roads. The methodology for including infrastructure (prisons, old-age care facilities, youth care facilities) and expenditure in fiscal incidence models is currently under households without a domicile address (informal housing). development by the CEQ Institute. 52 50 CIT is better modelled with tax administrative data. Meaning distributional (static) impacts, not causal impacts. Furthermore, data on CIT cannot help in the allocation of CIT 53 There are, however, CEQ Assessment procedures for assessing burdens to households in the standard CEQ FIA framework. the statistical significance of the estimated impact of a fiscal 51 Examples of the relevant infrastructure investments that are policy (or set of fiscal policies) on poverty, inequality, and missing and that could affect households differently in the other indicators. income distribution could be connectivity infrastructure, 66 households in the microdata represented GEL 9.7 Due to limitations of the HIES data and fiscal billion, which accounted for about 93.3 percent administrative data, the fiscal incidence model of federal government social expenditures, 36.7 excluded taxes such as CIT, which has a percent of government total expenditures, and substantial share in government revenue. The 13.8 percent of GDP in fiscal year 2022. analysis also excluded infrastructure expenditure, defense, debt interest payments, and small social The total expenditures and taxes actually programs (without clear allocation rules). allocated in the microdata using the current incidence model produced a tax-to-expenditure ratio that was higher than the ratio observed in b. Assessment of fiscal interventions the administrative data for the same taxes and expenditures. The ratio of total taxes to total The analysis followed the CEQ Framework public transfers 54 was 42 percent. The reason (Figure A.1 above) for the definition of income behind this discrepancy is that the allocation of concepts. Given that the official welfare taxes in the survey depends on correct aggregate in Georgia is based on consumption information about income and consumption expenditure, the calculation of income concepts patterns. However, the HIES only covered 27.0 starts with disposable income, which for this percent of the total consumption in national report is defined as the official consumption accounts, which could be due to missing top- aggregate available in the HIES 2022. The CEQ income households in the survey, methodological income concepts above disposable income are differences, data collection challenges, and calculated backward: net market income equals conceptual issues. In contrast, the estimation of disposable income minus direct transfers, while public transfers performs better in the survey market income plus pensions equals net market model given that the allocation is based on income plus direct taxes. The CEQ income socioeconomic characteristics, and many concepts below disposable income are calculated transfers aim to cover the poorer segments of the in the standard way: consumable income equals population (which is better represented in the disposable income minus indirect taxes plus HIES). The redistributive effect of fiscal policies indirect subsidies, while final income equals may be overestimated due to the low share of consumable income plus in-kind benefits from coverage of taxes in relation to transfers — the health and education. share of taxes included in the CEQ (15 percent) is lower than the share of spending included (34 percent). This low share is partly because of the lower coverage of VAT related to low consumption coverage in the HIES. 54 Total taxes include direct and indirect taxes. Total transfers refer to direct transfers, indirect subsidies, and in-kind benefits from health and education. 67 3. Taxes and Contributions Direct Taxes and Contributions Personal income tax reporting secondary income or self-employment were randomly assumed to be fully formal or informal to match the informality rate in each of In fiscal year 2022, total PIT collection (under all categories) was 6.9 percent of GDP. This three population segments: Tbilisi, other urban constituted about 66.5 percent of all direct taxes areas, and rural areas. Informality rates were and 29.1 of total tax revenues in Georgia, making assumed to be 17 percent for Tbilisi, 36 percent it the second-largest source of tax revenue after for other urban areas, and 76 percent for rural VAT. PIT is calculated on gross personal income areas (Fuchs et al. 2020). Private transfers and with a flat tax rate of 20 percent and is deducted remittances were considered exempt from PIT. from domestically or locally sourced labor Major limitations of the model include the income, including income from employment, exclusion of second-job employees (who should benefits from employment, and income from self-report taxes voluntarily) and the missing businesses. Non-Georgian source income of information on top-income households in the residents is exempted from the tax. Self- HIES data. employed persons with an annual turnover of less than GEL 500,000 pay a 1 percent tax on their turnover, and the rate increases to 3 percent if Property tax the annual turnover exceeds GEL 500,000. Tax- exempt income includes grants, state pensions, Property tax is a local tax on immovable property, yachts, planes, helicopters, and academic scholarships, income from property motorcycles owned by an individual. Total obtained by inheritance, income received after divorce, lottery winnings (up to GEL 1,000), property tax collected in fiscal year 2022 was 0.8 income from blood donations, alimony, and percent of GDP, which constitutes about 7.9 income from the sale of a car if the property was percent of all direct tax collection in Georgia. owned for more than six months (Parliament of The tax is charged on both natural persons and Georgia, 2010). legal persons, but in the case of natural persons only when the family’s annual income is greater PIT was modeled based on the HIES data for than or equal to GEL 40,000. For legal persons, individuals who reported being paid using the the rate is 1 percent of the average residual value main sources of income as encapsulated in the of the assets. For real estate, these values are survey. From the HIES, annual income was indexed by a factor of 3 if the asset was purchased grossed up based on PIT. PIT was then before the year 2000, by a factor of 2 if the asset simulated, imputing statutory rates to labor and was purchased between 2000–03, and by a factor capital income reported in the 2022 HIES by of 1.5 if asset was purchased in the year 2004 or applying the applicable rates, exemptions, and later. The factor has been defined by the limitations. Primary labor income was assumed acquisition date of the property. For natural to be fully formal, while secondary labor persons, the rate is 0.5 percent to 0.2 percent for activities and self-employment were adjusted to family income between GEL 40,000–100,000, and reflect informality in labor markets. Workers it is 0.8 percent to 1.0 percent for family income above GEL 100,000. For the property tax on 68 agricultural land, the annual tax base per one The pension scheme is financed by employers, hectare varies from GEL 5 to GEL 100. The employees, and the government. Each employer annual tax base for non-agricultural land is GEL transfers 2 percent of the employee’s salary 0.24 per one square meter. income to the employee’s private pension account while paying the salary. In addition, each Property tax was simulated in the HIES based on employer transfers up to 2 percent of the the following variables: (1) property ownership employee’s salary income to the private pension and property type (residential and land), (2) the account on behalf of its employee. Self-employed average municipal tax rate in each region applied individuals transfer 4 percent of their annual to properties identified in each region, and (3) income to the individual pension account, self-reported house values used from the survey. representing both the employer and employee For the missing housing values in the survey, a contributions. The government transfers (1) 2 regression model controlling for house percent of the income of the participant to his or characteristics was used to estimate house value. her private pension account if the annual salary Based on these variables and the provincial tax of the participant/income of the self-employed rules, the property tax was calculated for each individual is less than GEL 24,000 or (2) 1 percent region. of the income of the participant if the annual Limitations for modeling this tax include (1) it salary of the participant/income of the self- was not possible to proxy tax evasion levels employed individual is between GEL 24,000– (informality) and (2) other types of property that 60,000. If the annual salary of an are taxed under property tax law such as yachts, employee/income of the self-employed individual planes, helicopters, and motor cars were excluded is greater than GEL 60,000, the government will from the analysis. only contribute to 1 percent of GEL 60,000 (meaning GEL 600). Contributions to old-age pensions Contributions were not included in the analysis as they are considered a non-public scheme, except As of January 2019, joining a pension scheme for the transfer made by the government became mandatory for all employees, with a contribution on behalf of the employees. The significant change in the pension system from model assumes that the government contribution being non-contributory to being contributory. works as a transfer to the households eligible to Citizens of Georgia as well as foreign citizens contribute to the pension system. The government with residence permits who are official employed contribution is capped according to law (1 percent are obliged to contribute 4 percent of their gross above GEL 24,000, 0 percent above GEL 60,000). earnings to the Pension Fund. Exempted Only formal workers identified as being under individuals include those who were above 60 retirement age were assumed to contribute to the years of age (55 years of age in the case of women) pension system. Only the contributions from before the law's enactment and self-employed government were included as a direct transfer individuals. from the government to households. 69 Indirect Taxes Value added tax In fiscal year 2022, total effective VAT VAT was simulated in the HIES based on the collections reached 10.2 percent of GDP, expenditures on goods and services reported making VAT Georgia’s largest tax on by households 55 in the HIES 56 and effective VAT consumption. This accounted for about 77.7 rates available at the regional level. 57 HIES percent of total indirect taxes and 43.0 percent provides information on the household’s place of total tax collection in the country. The of purchase for different items. Effective rates standard VAT rate in Georgia is 18 percent, but were calculated to account for informality multiple items are exempted, including: based on the place of purchase, considering financial services, medical services, import of those ”purchased in a street” as informal and goods for the use of diplomatic representative the rest (market or at a shop/store) as formal offices, baby foods and hygiene products, games purchases. Effective VAT rates were calculated for lotteries/casinos, import and/or supply of as the statutory rate adjusted by the average electric buses, provision of technical services to formality rate per region and urban/rural aircraft and ships conducting international sea setting (Table A.1). Effective VAT rates were passages, and supply and import of cars under applied to the value of consumption, except for National Commodity Nomenclature of Foreign the goods and services that are exempt from Economic Activities Code 8703. Taxable VAT. It was assumed that all purchases in persons include individuals registered for VAT, bazaars and street markets are informal and do importers and non-resident persons, producers, not collect or report taxes, while all purchases and/or importers of excisable goods. A person in shops are formal and fully taxed. is required to register as a VAT payer if that person carries out an activity and the total amount of taxable transactions during the previous 12 consecutive calendar months exceeds GEL 100,000. 55 The analysis focused on purchased expenditure (excluding health care, education, recreation and culture, farming home production, gifts, and salaries in-kind consumed). products, and inputs was obtained from consumption reported in section 4. Items exempted from VAT were 56 Reported consumption in section 3 of the HIES (diary of identified and separated: agricultural and farm produce, expenses, named Table Shinda 03) and section 4 of the HIES health services and goods, education services, books, (quarterly questionnaire of expenses, named Shinda 04) was financial services, and baby food. used to calculate the value of the consumption items affected 57 by indirect taxes. A standard unit value model was used to The analysis focused on purchased expenditure (excluding estimate the price of goods and services. Consumption of home production, gifts, and salaries in-kind consumed). The food was obtained from section 3. Spending on housing netting down of purchased expenditure was calculated as = (building materials, maintenance), utilities, fuel for heating, expenditure/[(1+gst rate)*(1+excise rate)*(1+all duty rates)]. furniture, appliances, clothes, footwear, textiles, transport, 70 Table A.1: Regional effective VAT rates (percentages) Region Urban Rural Region Kakheti 14.2 14.1 14.1 Tbilisi 16.9 16.9 Shida Kartli 16.5 15.6 16.0 Kvemo Kartli 14.5 15.5 15.0 Samtskhe-Javakheti 16.2 15.3 15.7 Adjara A.R. 16.8 17.0 16.9 Guria 16.8 16.7 16.7 Samegrelo-Zemo Svaneti 15.9 14.5 15.1 Imereti, Racha-Lechkhumi and Kvemo Svan 16.3 15.6 16.0 Mltskheta-Mtianeti 14.3 14.1 14.2 Racha-Lechkhumi & Kvemo Svanetj 16.9 13.2 14.1 Total 16.4 15.2 16.0 Source: World Bank staff calculations based on CEQ Georgia 2022, HIES, and Fuchs et al. (2020) Excise duties Excise duties were simulated in the HIES based on the expenditures on goods reported by In fiscal year 2022, total excise duties collections households. 58 Statutory excise rates were applied reached 2.8 percent of GDP. This contributed over mobile communications, fuel, alcohol, and around 20.9 percent of total indirect taxes. Excise tobacco. For the analysis, only direct effects were tax is imposed at various rates on certain goods modeled where statutory rates were applied for the such as alcohol and alcoholic beverages, tobacco, share of formal purchases of excisable items. means of transport, petrol and diesel fuel, and Indirect effects were not estimated due to lack of international call termination services in a mobile information in the input-output (IO) matrix. or fixed network in Georgia. The export of excisable goods and supply of Georgian goods for sale in a duty-free zone are exempt. The statutory Customs duties excise tax rate varies by good and is determined based on content (such as the percentage of alcohol In fiscal year 2022, total customs duty collections content for alcoholic beverages). The purpose of reached 0.2 percent of GDP. This constituted around imposing federal excise duties is not only to collect 1.3 percent of indirect taxes. The rate of customs tax revenues but also to discourage the use of some duties varies by good, ranging from 5 percent to 12 unhealthy products such as tobacco. percent. For motor vehicles, this tax depends on the age and engine capacity of the vehicle. Customs duties were not included in the analysis. 58 The analysis focused on purchased expenditure (excluding self-production, gifts, and salaries in-kind consumed). 71 4. Social Expenditure and Subsidies Direct Transfers and Pensions Social protection programs included in the analysis were TSA; loss of breadwinner, disability, and veteran’s pensions; and assistance to IDP. All of these programs were modeled based on direct identification in the HIES 2022, which directly identifies cash transfers received by the household or individual. Public pensions Targeted Social Assistance (TSA) Georgia provides non-contributory pensions to The TSA is Georgia’s largest social protection citizens above age 60 who permanently reside in program. Total spending on this program the territory of Georgia or who are foreigners amounts to about 1.6 percent of GDP, residing in Georgia for the past 10 years, with approximately 10 percent of all social spending in the aim of providing means of subsistence. In Georgia. It provides monthly cash transfers to 2006, the Georgian Parliament introduced the poor households identified based on a Proxy Law on State Pensions, which resulted in the Means Test (PMT) assessment, which elimination of the contributory pensions system approximates a household’s level of welfare and and the implementation of a flat-rate basic poverty status using a set of indicators empirically pension with three components: old age, correlated with monetary welfare. It also includes disability, and survivor’s pension. The universal a Child Benefit Program as well as a university pension system in Georgia relies mainly on tuition fee waiver for students from households general government revenue. The financing with a PMT score below 150,000. mechanism has posed fiscal sustainability risks due to demographic aging. Pension spending is Based on the Socially Vulnerable Household the largest expenditure in social protection in Database (SVHD), the authorities calculate a PMT Georgia, with total expenditure amounting to 4.0 score which then determines families' eligibility percent of GDP and 27.8 percent of the for specific social programs. The benefit levels government’s total social spending. Total differ depending on the welfare score of the coverage of pensions as of fiscal year 2022 stood household and the number of household members. at 819,140 individuals (22 percent of the total TSA program coverage grew from 427,000 population) or 617,946 households (56 percent of individual beneficiaries (about 120,000 total). The pension law was amended in 2018 and households) as of December 2019 to 528,900 made the pension scheme mandatory for all individuals (about 129,200 households) as of May employees, except for those men who were above 2022 to respond to the increased poverty and 60 years of age and those women who were above vulnerability related to the COVID-19 pandemic 55 years of age before the law came into effect on (Carraro et al. 2022). The value of the child benefit August 6, 2018. The pension scheme is financed is GEL 100 per month per child, and the eligibility by employer, employee, and government (see threshold is GEL 120,000. previous paragraphs describing contributions to old-age pensions). 72 Internally displaced population Other social assistance programs Internally displaced people (IDP) benefits are Several other benefits are provided to the poor another “categorical” (means-tested) benefit and vulnerable population through several for persons living in high mountainous regions. small-scale social protection programs such as The expenditure remained at less than 0.1 the disability program and the veterans percent of GDP in 2022. The total number of program. These programs are mostly targeted at individual beneficiaries in 2022 was 116,905, specific sections of the population, with which is about 44,808 households. eligibility criteria being defined by the specifications of each. The total outlay of these programs is about 1 percent of GDP. Indirect Subsidies In fiscal year 2022, the government spent about 2.5 percent of GDP on subsidies. Georgia has several types of indirect subsidies, including subsidies on electricity, gas, water, and urban transport. The government provides these subsidies for the consumption of these utilities by individuals and households. Subsidies more than doubled in nominal terms since 2015 to reach more than 3 percent of GDP in 2022, reflecting an increase in transfers for public city transportation, agriculture, preschool education, waste collection and processing, water supply, culture, as well as other general economic and commercial activities. Electricity subsidy between the cost of electricity per unit supplied and the actual electricity price paid by the domestic consumers) 61 times the electricity The analysis explored the electricity subsidy by taking the difference between the cost of quantity used by the domestic consumers. The electricity per unit supplied and the actual price analysis of electricity subsidies focused on the paid by domestic consumers. 59 To calculate the households that pay a tariff below the electricity direct effects of the domestic electricity subsidy, 60 cost of supply. the following steps were taken: (1) the amount of households’ expenditure on electricity based on the “electricity charges” reported in the HIES Natural gas subsidy household expenditure module was identified, (2) the quantity of electricity consumed was estimated The analysis focused on the government subsidy by dividing total expenditure on electricity by the provided for natural gas consumed by monthly price of electricity in 2022, and (3) the households. To calculate the direct effects of the amount of the direct effect of the subsidy was natural gas subsidy on households’ welfare, the calculated as the subsidy amount (difference following steps were taken: (1) the average subsidy 59 60 The average electricity cost for domestic consumers was According to Lustig (2018), calculating direct effects is estimated at 17.4 GEL/kwh for Tbilisi and 17.7 GEL/kwh for the sufficient if household energy subsidies are provided for rest of the country. domestic consumption only. Source: https://www.energo-pro.ge/en/service/electricity- 61 tariff/23 In the model, the households that are paying above the electricity cost are not considered to be subsidy recipients. 73 unit (difference between final consumer price and households’ expenditure on water based on the prescribed price) for natural gas was calculated for “water charges” reported in the HIES household fiscal year 2022, based on available macro-fiscal expenditure module was identified, (2) the share data — according to the IMF, the average subsidy of consumption not paid by the consumer was unit for natural gas in 2022 was 0.26 per GEL spent; determined using survey responses, and (3) the (2) the quantity of natural gas consumed by each subsidy was considered to be the amount household was estimated in the HIES by dividing estimated in step 2. total household expenditure on gas by the final consumer price; and (3) the direct effect of the gas subsidy was calculated as the subsidy unit Transport subsidy multiplied by the quantity of gas consumed by each household. The indirect effects could not be The government also provides transport subsidy. calculated because this requires a disaggregation by To calculate the direct effects of the transport product of the “petroleum sector” in the IO matrix, subsidy, the following steps were taken: (1) the which is not available. beneficiary population in the survey was identified as senior citizens and beneficiaries of the TSA program, (2) the amount spent per household on Water subsidy public transportation in the regions receiving the benefit (Tbilisi) was estimated, and (3) the subsidy The analysis explored the water subsidy being was determined assuming that the household provided to domestic consumers. To calculate reports only 40 percent of the expenditure and the direct effects of the domestic water subsidy, that the subsidy is equal to 60 percent of the total the following steps were taken: (1) the amount of transport tariff. In-kind education benefits Public expenditures on education executed by the central and municipal governments in fiscal year 2022 were estimated at 3.3 percent of GDP. Georgia’s 2016 Early Childhood Education and Care law outlines that preschool education should be accessible for all children. In Georgia, education governance is characterized by a dual structure that combines centralization and decentralization. The Ministry of Education and Science (MoES) plays a pivotal role in shaping education strategy, policy, curriculum development, school standards, textbooks, and funding allocation. Functional classification of the education system is organized in four levels: 1. Preschool education falls under the Georgia, state-funded public institutions are jurisdiction of the municipalities. fully covered, while private institutions offer 2. General education funding is implemented both state-funded and self-funded programs. through a voucher program. The vouchers are 4. Higher education institutions receive the primary income source for schools, used to funding for specific priority fields pay for teachers' salaries and maintenance costs. (agricultural sciences, education, 3. Vocational education is fully funded by the engineering, natural sciences, social sciences, government, while private institutions offer humanities, and interdisciplinary fields), state-funded and self-funded options. In fully covering tuition fees. 74 The in-kind benefits from access to public were assumed to be the ones with the higher education were modeled based on direct expenses up to the share of private students as identification of beneficiaries in the HIES and reported by GeoStat at each education level. Per imputation of average benefits (measured at the capita in-kind education benefits are shown in average government cost of providing these Table A.2. services). The analysis covered preschool, general education, vocational education, and higher Table A.2: Per capita value of education benefit (GEL) education. The education model allocated per- capita cost at each education level to families with Level Per capita cost individuals that reported to be studying at each Preschool 2,149 respective level. The HIES does not contain enough information to differentiate between General 1,888 students going to private and public schools, so Vocational 2629 education expenditure data was used to differentiate between students attending public Higher 1178 schools and private schools. Private school students Source: World Bank (forthcoming) In-kind health benefits Public expenditures on health in fiscal year 2022 totaled 2.9 percent of GDP. Georgia’s spending on health is implemented through the Universal Health Care Program (UHCP). Health services are stratified by income and other priority groups. A broad set of services is provided for the priority groups (children, pensioners, disabled, veterans). A more limited group of benefits is offered to the rest of the population. MoILHSA has oversight of the health system. The analysis used the “insurance approach” to MoILHSA is responsible for developing and allocate in-kind public health benefits. The implementing national health care policy and model inputted health spending on a per capita strategy, drafting and enforcing health care laws basis, calculating per capita spending for the and regulations, setting up and overseeing population of each priority group and the general national public health programs, advocating for population. As a first step, health spending was adequate allocations from the budget for health divided across the different health programs, care programs, and regulating health care which benefit many priority groups at different professions, health care facilities, and the coverage rates. Those coverage rates were used to pharmaceutical market. allocate each program’s budget across all priority groups, calculating a budget allocation program with these restrictions: (1) 100 percent of the program budget is spent, and (2) the share allocated to any individual in each priority group is proportional to the coverage rate. 75 Appendix B 76 A P P E N D I X B Sensitivity Analysis on Pensions: Pensions as Deferred Income Scenario In the CEQ methodology, pensions are treated in two different ways: Pensions as Government Transfers (PGT) and Pensions as Deferred Income (PDI). The analysis presented in the report used PGT as the baseline scenario, while this section presents the sensitivity analysis of results based on PDI. The PDI scenario implies: (1) the contributions to private social security schemes are considered mandatory savings (of income) rather than a tax on income, and (2) pension income for pension recipients is treated as current “market income plus pensions.” Hence, public pensions in this analysis are treated as deferred income (embedded in the pre- fiscal income concept “market income plus pensions”). Results for PDI Scenario Table B.1: Impact on inequality and poverty Market Disposable Consumable Final income plus income income income pensions Inequality Gini 0.367 0.326 0.321 0.246 Theil Index 0.224 0.184 0.179 0.109 90/10 6.626 4.431 4.264 2.860 Poverty (using Headcount Index 18.7% 13.3% 15.2% n.a. national Poverty gap 7.7% 3.2% 3.2% n.a. moderate poverty line) Squared poverty gap 4.7% 1.2% 1.4% n.a. Source: World Bank's estimations based on HIES 2022 77 Table B.2: Marginal contributions of each fiscal intervention to inequality and poverty (for disposable, consumable, and final income) Size (with Concentration Kakwani Redistributive Poverty respect to Coefficient Coefficient effect reduction effect original Marginal Marginal income)1 contribution contribution (national moderate poverty line) Disposable Income 0.9694 Direct transfer: loss of breadwinner (per capita) 0.0011 -0.4390 0.8062 0.0007 0.0017 Direct transfer: vulnerable families (per capita) 0.0292 -0.6881 1.0553 0.0281 0.0514 Direct transfer: internally displaced persons (per capita) 0.0039 -0.0685 0.4357 0.0015 0.0035 Direct transfer: others (per capita) 0.0212 -0.2357 0.6030 0.0088 0.0242 Direct transfer: government contributions to pensions (per capita) 0.0060 0.3416 0.0257 0.0002 0.0018 All direct transfers excluding contributory pensions 0.0613 -0.3877 0.7549 0.0411 0.0791 All direct transfers including contributory pensions 0.0613 -0.3877 0.7549 0.0411 0.0791 Direct taxes: personal income tax (per capita) -0.0908 0.3913 0.0240 0.0028 -0.0224 Direct taxes: property tax (per capita) -0.0012 0.7266 0.3593 0.0005 0.0000 All direct taxes -0.0920 0.3957 0.0284 0.0033 -0.0224 All direct taxes and contributions -0.0920 0.3957 0.0284 0.0033 -0.0224 Consumable Income 0.8931 Direct transfer: loss of breadwinner (per capita) 0.0011 -0.4390 0.8062 0.0008 0.0016 Direct transfer: vulnerable families (per capita) 0.0292 -0.6881 1.0553 0.0301 0.0537 Direct transfer: internally displaced persons (per capita) 0.0039 -0.0685 0.4357 0.0015 0.0038 Direct transfer: others (per capita) 0.0212 -0.2357 0.6030 0.0094 0.0290 Direct transfer: government contributions to pensions (per capita) 0.0060 0.3416 0.0257 0.0003 0.0031 All direct transfers excluding contributory pensions 0.0613 -0.3877 0.7549 0.0441 0.0868 Direct taxes: personal income tax (per capita) -0.0908 0.3913 0.0240 0.0028 -0.0253 78 Direct taxes: property tax (per capita) -0.0012 0.7266 0.3593 0.0005 0.0000 All direct taxes -0.0920 0.3957 0.0284 0.0032 -0.0253 All direct taxes and contributions -0.0920 0.3957 0.0284 0.0032 -0.0253 Indirect subsidy: electricity (per capita) 0.0023 -0.0016 0.3688 0.0005 0.0022 Indirect subsidy: gas (per capita) 0.0236 0.1589 0.2084 0.0037 0.0186 Indirect subsidy: water (per capita) 0.0001 -0.3847 0.7519 0.0000 0.0000 Indirect subsidy: transport (per capita) 0.0032 0.3500 0.0173 0.0000 0.0018 All indirect subsidies 0.0292 0.1662 0.2011 0.0043 0.0216 Indirect taxes: VAT (per capita) -0.0954 0.3396 -0.0276 0.0003 -0.0340 Indirect taxes: excises (per capita) -0.0101 0.4100 0.0427 0.0005 -0.0015 All indirect taxes -0.1055 0.3464 -0.0209 0.0009 -0.0372 All taxes -0.1975 0.3693 0.0021 0.0041 -0.0551 Final Income 1.1138 Direct transfer: loss of breadwinner (per capita) 0.0011 -0.4390 0.8062 0.0004 Direct transfer: vulnerable families (per capita) 0.0292 -0.6881 1.0553 0.0179 Direct transfer: internally displaced persons (per capita) 0.0039 -0.0685 0.4357 0.0010 Direct transfer: others (per capita) 0.0212 -0.2357 0.6030 0.0067 Direct transfer: government contributions to pensions (per capita) 0.0060 0.3416 0.0257 -0.0001 All direct transfers excluding contributory pensions 0.0613 -0.3877 0.7549 0.0267 Direct taxes: personal income tax (per capita) -0.0908 0.3913 0.0240 0.0071 Direct taxes: property tax (per capita) -0.0012 0.7266 0.3593 0.0004 All direct taxes -0.0920 0.3957 0.0284 0.0074 All direct taxes and contributions -0.0920 0.3957 0.0284 0.0074 Indirect subsidy: electricity (per capita) 0.0023 -0.0016 0.3688 0.0003 Indirect subsidy: gas (per capita) 0.0236 0.1589 0.2084 0.0019 Indirect subsidy: water (per capita) 0.0001 -0.3847 0.7519 0.0000 Indirect subsidy: transport (per capita) 0.0032 0.3500 0.0173 -0.0001 All indirect subsidies 0.0292 0.1662 0.2011 0.0021 Indirect taxes: VAT (per capita) -0.0954 0.3396 -0.0276 0.0054 Indirect taxes: excises (per capita) -0.0101 0.4100 0.0427 0.0009 79 All indirect taxes -0.1055 0.3464 -0.0209 0.0064 All taxes -0.1975 0.3693 0.0021 0.0134 All taxes and contributions -0.1975 0.3693 0.0021 0.0134 In-kind health transfers: total (per capita) 0.1284 -0.0373 0.4046 0.0347 Net health transfers 0.1284 -0.0373 0.4046 0.0347 In-kind education transfers: pre- school (per capita) 0.0165 -0.1867 0.5539 0.0058 In-kind education transfers: general (per capita) 0.0693 -0.2214 0.5887 0.0252 In-kind education transfers: vocational (per capita) 0.0006 0.0531 0.3142 0.0001 In-kind education transfers: tertiary (per capita) 0.0058 0.1035 0.2638 0.0010 Net education transfers 0.0923 -0.1928 0.5601 0.0325 Source: World Bank's estimations based on HIES 2022 80 Appendix C 81 A P P E N D I X C Sensitivity Analysis on Model Assumptions CEQ analysis is a data-intense approach and relies on several assumptions to model taxes and transfers. Some of these assumptions are critical and altering them can have significant effect on the results. This section presents the sensitivity analysis based on alternate scenarios for two such key assumptions for CEQ Georgia 2022. The assumptions made included (1) informality in personal income tax (PIT) collection and (2) use of the direct identification method to identify beneficiaries from the survey. The following sections elaborate on both assumptions, provide an alternate scenario, present the results of the sensitivity analysis, and discuss their implications for the overall findings. 1. Informality and Personal Income Tax CEQ Georgia includes modeling of the impact of PIT on poverty and inequality. The macrovalidation (ratio of PIT collection from the household survey with administrative data) was very low. In the baseline scenario, the share of formal employment in Tbilisi and urban and rural areas outside of Tbilisi was considered. However, this assumption was applied only to the second job and the self-employed respondents. The sensitivity analysis extends this assumption to the first job, as well, using the same statutory tax rates. 2. Increasing TSA Coverage CEQ Georgia includes modeling of the impact of social transfers, including the main social transfer program of targeted social assistance (TSA). To estimate the benefits, the direct identification method was used to identify the beneficiaries of TSA. The results using the direct identification method had macrovalidation of about 82 percent. Due to the lack of information on the Proxy Means Test (PMT) formula, variables, and their weights, it was difficult to simulate the composition of the beneficiaries and the total disbursement of the program. There was a coverage gap, with the survey capturing around 82 percent of the total beneficiaries. The sensitivity analysis was conducted for the number of beneficiaries to match the total from the admin accounts with the survey. A probit model was estimated to be used to select the new beneficiaries and increase the total to 118,000 households. The model was P (Beneficiary of TSAT) = f(children under 16, children over 16, urban household, family size, region, age of household head, working age woman and working age man). The results of the regression are provided in Table C.1. 82 Table C.1: Regression results of probit Probit regression Number of obs = 9,789 LR chi2(17) = 3963.05 Prob > chi2 = 0.0000 Log likelihood = -2666.1085 Pseudo R2 = 0.4264 tsa_tra_dummy Coefficient Std. err. z P>z [95% conf. interval] under16 1.561182 .0476518 32.76 0.000 1.467786 1.654577 older16 .7365746 .0282407 26.08 0.000 .6812238 .7919254 urbanorrural .4217039 .046396 9.09 0.000 .3307695 .5126383 familysize -.6861009 .0343609 -19.97 0.000 -.7534471 -.6187547 regno 1 -.1443233 .0977868 -1.48 0.140 -.3359819 .0473353 2 -.2791584 .0931373 -3.00 0.003 -.461704 -.0966127 3 -.016612 .1054669 -0.16 0.875 -.2233232 .1900993 5 -.6660101 .1064262 -6.26 0.000 -.8746015 -.4574186 7 -.1828625 .1067724 -1.71 0.087 -.3921325 .0264076 8 -.2347075 .0944361 -2.49 0.013 -.4197988 -.0496162 9 -.3908391 .0951068 -4.11 0.000 -.577245 -.2044331 10 -.3701873 .0941848 -3.93 0.000 -.554786 -.1855885 11 -.554642 .0953705 -5.82 0.000 -.7415647 -.3677194 13 .5608429 .079252 7.08 0.000 .4055119 .7161739 hage .0106514 .0018698 5.70 0.000 .0069866 .0143162 working_age_woman .4226504 .0400337 10.56 0.000 .3441858 .501115 working_age_man .3691626 .0396552 9.31 0.000 .2914397 .4468854 _cons -2.613349 .1567854 -16.67 0.000 -2.920643 -2.306056 Source: World Bank's estimations based on HIES 2022 and MOF data The new beneficiaries were selected from non-recipients, getting the highest probability from the probit model. The benefits were applied according to the number of children under age 16 and over age 16. Due to the lack of information on total disbursements for this program, sensitivity analysis was only conducted for the number of beneficiaries. 83 Results For the sensitivity analysis of PIT, total tax collection was reduced by around GEL 477 million, around 65 percent of the PIT tax collection estimated in the baseline scenario. For the TSA sensitivity analysis, the direct transfers disbursement amount increased by GEL 13 million, or around 3 percent more of the estimated benefits (See Table C.2). Table C.2: Results of the sensitivity analysis on macrovalidation CEQ CEQ Actual Actual Ratio estimation estimation (GEL (% of household (GEL (% of millions) consumption, millions) disposable national income) accounts) [1] [2] [3] [4] [1]/[3] Personal income tax Original 1,371 9.4 5,034 9.7 0.27 Sensitivity analysis 894 6.1 5,034 9.7 0.18 TSA vulnerable families* Original 440 3.0 1,165 2.3 0.38 Sensitivity analysis 453 3.1 1,165 2.3 0.39 Source: World Bank staff calculations based on HIES and MOF data Note: *Actuals is not exactly TSA program and corresponds to family and children transfers from Social Protection function Sensitivity analysis showed the enhanced poverty reduction impact of TSA and the insignificant impact on inequality. The significant impact on poverty reduction was observed in the TSA alternative scenario (See Table C.3). Analysis of TSA sensitivity showed lower poverty rates, as TSA transfers would provide benefits to low-income households. Despite the smaller amounts compared to the PIT, the new TSA had a greater effect because it targeted the lower-income deciles. Under the PIT alternate scenario, some poor households would no longer be subject to the tax as they are considered informal, and the impact on poverty is insignificant. Inequality reduction is minimal in both alternative simulations, and it is negligible compared to the baseline in the PIT scenario. There is no statistically significant difference in the inequality reduction in the TSA scenario. 84 Table C.3: Results of the sensitivity analysis on poverty and inequality Inequality reduction Poverty reduction From market to From market to From market to consumable consumable final income income income Changing PIT assumptions 0.126 0.201 17.1 Changing TSA assumptions 0.132 0.207 18.8 Original/Baseline 0.126 0.201 17.1 Source: World Bank's estimations based on HIES 2022 Results indicated that the sensitivity analysis approach for TSA has a potential impact on the results, but this is not the case for the PIT alternative simulation. When it comes to PIT simulation, it is assumed that informality only affects a second job. However, the estimates from the survey are much lower than the actual tax collection. The TSA alternative simulation produces a more poverty-reducing impact than that of the baseline. It is unclear whether the alternative simulation includes all beneficiaries and whether the benefits are given to all children in the household. If the coverage is higher than what is reported by the survey, more poverty reduction is expected. 85 86