63566 Belarus Public Expenditure Review Fiscal Reforms for a Sustainable Economic Recovery Volume 1 Report No. 63566-BY Belarus Public Expenditure Review Fiscal Reforms for a Sustainable Economic Recovery Volume 1 August 15, 2011 Poverty Reduction and Economic Management Unit (ECSPE) Europe and Central Asia Region Bel arus Public Expenditure Review Currency and Equivalent Units (as of August 15, 2011) Currency Unit = Belarusian Rubel (BYR) US1$=5,028 BYR Government Fiscal Year January 1 – December 31 Acronyms and Abbreviations AgGDP Agricultural Gross Domestic Product MOF Ministry of Finance BCSE Belarus Currency and Stock Exchange MOLSP Ministry of Labor and Social Policy BoP Balance of Payment MOT Ministry of Taxes CAPEX Capital Expenditures MTEF Medium Term Expenditure Framework CEBS Central European and Baltic States NBRB National Bank of the Republic of Belarus CHP Combined Heat and Power NCRC Notional Defined Contribution CIS Commonwealth of Independent States NPL Non-Performing Loans Organization for Economic Co-operation and CIT Corporate Income Tax OECD Development CP Cost-to-Price ratio OPEX Operating Expenditures CPI Consumer Price Index PAB Pension Assessment Base CRF Capital Recovery Factor PAYG Pay-As-You-Go CU Customs Unit PCI Pulverized Coal Injection DPL Development Policy Loan PEFA Public Expenditure Financial Accountability European Bank for Reconstruction and EBRD PER Public Expenditure Review Development ECA Europe and Central Asia PFM Public Financial Management ER Exchange Rate PFR Public Finance Review EU European Union PIT Personal Income Tax FAO Food and Agricultural Organization PPG Public and Publicly Guaranteed FDI Foreign Direct Investment PPP Purchasing Power Parity FX Foreign Exchange PROST Pension Reform Options Simulation Toolkit GASP Government System for Social Assistance QFD Quasi-Fiscal Deficit GDL Government Directed Lending R&D Research and Development GDP Gross Domestic Product RAPS Republican Agricultural Producers Support GOB Government of Belarus SOE State-Owned Enterprise HBS Household Budget Survey SPF Social Protection Fund HOBs Heat-Only Boilers SPS Sanitary and Phyto-Sanitary IFI International Financial Institution TCM Thousand Cubic Meters ISRA Institute for Systemic Research in Agriculture TFP Total Factor Productivity LITS Life in Transition Survey TOT Terms of Trade LPI Logistics Performance Index TPP Thermal Power Plant LSFO Low Sulfur Fuel Oil VAT Value Added Tax MOFA Ministry of Food and Agriculture WTO World Trade Organization MOE Ministry of Economy Vice President Philippe Le Houerou Country Director Martin Raiser Sector Director Yvonne Tsikata Sector Manager Benu Bidani Task Team Leader Gallina A. Vincelette Co-Task Team Leader Sebastian Eckardt ii Fiscal Reforms for a Sustainable Economic Recovery Contents Acknowledgments viii Preface ix Executive Summary x Chapter 1. Why is fiscal reform pertinent in Belarus? 1 A. Macro-economic vulnerabilities–Implications for fiscal policy 2 B. Deepening structural reforms–Implications for fiscal policy 9 C. Fiscal challenges 15 Chapter 2. Agriculture: Hardening budget constraints 18 A. Institutional framework of the agricultural sector 19 B. The fiscal cost of agricultural support 21 C. Composition of spending 23 D. Impact of fiscal support on sector performance  28 E. Implications of international trade integration on the restructuring on the agricultural support system 31 F. Policy options and fiscal and social impact 33 Chapter 3. Energy services: Achieving financial viability 37 A. Energy market structure  38 B. The fiscal cost of energy services  39 C. Impact of energy subsidies  44 D. Fiscal and social impact of the planned energy tariff reform  47 E. Policy options and their fiscal and social impact 52 Chapter 4. Social assistance: Improving targeting 55 A. The fiscal cost of social assistance—Spending trends and composition 56 B. Effectiveness of social assistance spending 62 C. Targeting efficiency of social assistance spending  63 D. Policy options and their fiscal and social impact 66 Chapter 5. Pensions: Fostering sustainability 70 A. Long term fiscal trends in the pension system 71 B. The pension system—Current system features and reform options 73 C. Policy options and their fiscal and social impact 84 List of References 88 Annexes 89 iii Bel arus Public Expenditure Review List of Figures Figure 0-1. The fiscal cost of underpriced energy has grown rapidly xiv Figure 0-2. The population dependency rate will increase steadily… xvii …while moving the pension system into deficit  xvii Figure 0-3. Fiscal impact of different pension reform measures xviii Figure 1-1. Energy subsidies on imports from Russia are in decline 3 Figure 1-2. Economic growth is driven by terms of trade gains 3 Figure 1-3. Expenditure growth with moderate deficits 3 Figure 1-4. Fiscal imbalances emerged with the crisis… 4 Figure 1-5. …but Belarus’ fiscal position compares well to regional peers 4 Figure 1-6. Growth has rebounded, but is threatened by unsustainable external imbalance 4 Figure 1-7. Public sector wage hikes boosted domestic demand 4 Belarus’s cyclically-adjusted fiscal balance started to deteriorate prior to the 2008 global financial crisis  5 Figure 1-8. CAD deficit is increasingly financed by debt inflows… 6 …reflected in increased debt service burden  6 Mounting directed lending programs and bank recapitalization are associated with deteriorating state Figure 1-9.  budget 7 Figure 1-10. Belarus’ public expenditure level continues to be high… 9 Figure 1-11. …creating a high tax burden 9 Figure 1-12. Statutory tax rates in Belarus are high 11 Figure 1-13. The composition of general government expenditures 12 Figure 1-14. State support remains extensive 13 Figure 1-15. Belarus and ECA average PEFA scores* 15 Figure 2-1. Food exports have grown and now contribute positively to the trade balance 19 While budgetary spending on agriculture has remained stable, tax collections from agricultural Figure 2-2.  enterprises have declined 21 Figure 2-3. RAPS represents the largest source of annual budgetary spending on agriculture 23 Figure 2-4. Amber box measures absorb the lion share of spending 24 Figure 2-5. Input subsidies dominate amber box measures 25 Figure 2-6. Green box measures have increased, but still remain small 26 Figure 2-7. Debt of agricultural enterprises is ballooning 27 Figure 2-8. Farm subsidies disproportionately benefit small farms 28 Figure 2-9. Agricultural labor and capital productivity is considerably lower than the rest of the economy 30 Figure 2-10. Returns on capital in agriculture have declined while credit flows have risen 30 Figure 2-11. With farm production costs on the rise, revenues are stagnating 31 Figure 2-12. Over a half of the farms are not profitable 31 Figure 2-13. A WTO accession of another CU party may induce an accelerated reduction in subsidies… 33  …which would generate fiscal savings  33 Figure 3-1. Institutional structure of the energy sector in Belarus 38 Figure 3-2. Belarus imports natural gas below the European price… 40 …but the price gap is closing  40 iv Fiscal Reforms for a Sustainable Economic Recovery Figure 3-3. Residential tariff increases have lagged behind rising production costs 40 Figure 3-4. Cost recovery levels of residential services have declined sharply 41 Figure 3-5. The fiscal cost of underpriced energy has grown rapidly 42 Figure 3-6. The cost of under-priced energy is financed through direct budgetary subsidies and cross-subsidies 43 Figure 3-7. Energy subsidies have kept household utility expenditures low… 44  …but energy subsidies disproportionately benefit higher income groups 44 Figure 3-8. Imbalanced tariff structure imposes implicit tax on the real sector 45 Figure 3-9. Cross-subsidies squeeze net profits of Belenergo 46 Figure 3-10. Rising fuel costs crowd out O&M and depreciation charges 46 Figure 3-11. Investment needs and plans in the energy sector 46 Natural gas price scenarios  48 Figure 3-12. Required residential tariff increases are substantial, but so would be the fiscal savings 49 Figure 3-13. Higher tariffs will impact poverty 50 Figure 3-14. Simulated impact of the tariffs increase on population groups 51 Belarus redistributes a larger share of GDP through social assistance programs than most countries in Figure 4-1.  the region 56 Cost drivers of social assistance programs: Spending on some programs has expanded but has fallen Figure 4-2.  on others 59 Figure 4-3. Child-related benefits have risen as a share of total social assistance spending 59 Income-tested programs absorb a small share of social assistance spending and are among the lowest Figure 4-4.  in the region 61 Figure 4-5. Social assistance programs have contributed to Belarus’ strong social outcomes 62 Figure 4-6. Privileges are not a cost-effective way to alleviate poverty 62 Figure 4-7. Social assistance is reaching the poor, but targeting can be improved 63 Figure 4-8. Overall social assistance benefits are progressive 64 Figure 4-9. Benefits incidence of childcare programs 64 Figure 4-10. Privileges are poorly targeted 65 Figure 4-11. Benefit distribution by quintile of household consumption per capita, 2009 65 Figure 4-12. Privileges vary in size and targeting efficiency 66 Figure 4-13. Social assistance benefits are an important income component for the poorest households 66 Figure 5-1. The demographic environment will change dramatically 71 Figure 5-2. The population dependency rate will increase steadily… 72 …changing the ratio of contributors and beneficiaries in the pension system  72 Figure 5-3. The pension system will move into deficit… 72 …and replacement rates will decline  72 Figure 5-4. The retirement age is low…. 74 Figure 5-5. …but many pensioners continue to work 74 Figure 5-6. Retirement age reform scenarios: Average pension/average wage 76 Figure 5-7. Retirement age reform scenarios: Projected pension system balance 76 v Bel arus Public Expenditure Review Figure 5-8. Benefit indexation scenarios: Average pension/average wage 78 Figure 5-9. Benefit indexation scenarios: Projected pension system balance 78 Figure 5-10. Belarus’ pension contribution rate is high in regional comparison 79 Figure 5-11. Reform package scenario: Average pension/Average wage 80 Figure 5-12. Reform package scenario: Projected pension system balance 80 Figure 5-13. Income redistribution within the pension system 81 Figure 5-14. Notional accounts: Average pension/average wage 83 Figure 5-15. Notional accounts: Pension system balance 84 Figure 5-16. Income redistribution under notional accounts 84 Figure 5-17. Fiscal impact of different reform measures 85 Figure 5-18. Impact of different reform measure on average replacement rates 86 List of Boxes Box 1-1. Pro-cyclical fiscal policy in Belarus 5 Box 1-2. Types of quasi-fiscal activities 7 Box 1-3. Strengths and weaknesses of Belarus’ PFM system 15 Box 2-1. WTO classification of domestic support to agriculture 24 Box 2-2. Classification of agricultural support measures in CU 32 Box 3-1. Billing residential consumers for energy services 41 Box 3-2. Natural gas price scenarios 2011–2014 48 Box 4-1. Family allowance benefits in Belarus 57 Box 4-2. Benefits under targeted social assistance (GASP) 61 Box 5-1. PROST projections: Key macroeconomic and demographic assumptions  73 Box 5-2. Treatment of working pensioners in the EU countries  75 Box 5-3. International experience with notional accounts 83 vi Fiscal Reforms for a Sustainable Economic Recovery List of Tables Table 0-1. Options for fiscal consolidation % of GDP xi Table 0-2. Fiscal support to agriculture in Belarus xiii Table 1-1. Main macroeconomic indicators, 2003–2010 2 Table 1-2. Gross external debt, as of January 1st 2011 6 Table 1-3. Structure of tax revenues (percent of total)  10 Table 2-1. Fiscal support to agriculture in Belarus 22 Table 2-2. International comparison of fiscal support for agriculture 23 Table 2-3. Cumulative yield growth, percent 29 Table 2-4. Cross country comparison: yields vs. fertilizer use 29 Table 2-5. Fiscal and social impact of proposed reform measures 36 Table 3-1. GoB’s target household tariffs levels and budget subsidy to compensate for cross-subsidization 47 Table 3-2. Implied cumulative tariff increases under different price scenarios 50 Table 3-3. Fiscal and social reform impact of proposed measures 54 Table 4-1. Overview of social assistance programs and design features 58 Table 4-2. Social protection program impact on poverty and inequality  62 Table 4-3. Projected impact of abolishing privileges  67 Table 4-4. Projected impact of proposed reform in children-related benefits 69 Table 4-5. Fiscal and social impact of proposed reform measures 69 Table 5-1. International comparison: Degree of redistribution within the pension system 81 Table 5-2. Advantages and disadvantages of notional defined contribution account systems 82 Table 5-3. Projected medium term fiscal savings of different pension reform options (percent of GDP) 85 Table 5-4. Fiscal and social impact of proposed reform measures 87 vii Bel arus Public Expenditure Review Acknowledgments  his report was prepared by a multi-sector team, led by Gallina A. Vincelette and Sebastian Eckardt T (ECSPE), and comprising: Marina Bakanova, Svetlana Budagovskaya, Mariam Lomaia, Ruslan Piontkivsky, Maryna Sidarenka, Victor Sulla (ECSPE), Matthias Grueninger and Felicia Pricop (ECSS1), Pekka Salminen, Alexander Sharabaroff (ECSSD), Sergi Biletsky, Katerina Petrina, and Asta Zhviniene (ECSHD). Anna Sokol and Christopher S. Humphrey diligently edited the Russian and the English version of the report, respectively. Larysa Hrebianchuk and Maureen Itepu provided excellent support during the preparation process. The team is grateful to Photo.ByMedia.net for supplying the photographs for the cover and to Irina Oleinik for advising the team on the designs of this report.  he peer reviewers were Wolfgang Fengler (Lead Economist, AFTP2), Ivailo Izvorski (Lead Economist, T EASPR) and Marijn Verhoeven (Lead Economist, PRMPS). The team benefited from the guidance of Yvonne Tsikata, Benu Bidani, Bill Dorotinsky and Pablo Saavedra (ECSPE), and Martin Raiser (Country Director, ECCU2). The team appreciates the comments from World Bank and IMF colleagues made during internal presentations and discussions.  he team is grateful to the the Ministry of Finance, the Ministry of Taxes, the Ministry of Economy, the T Ministry of Labor and Social Protection, the Ministry of Energy, the Ministry of Utilities, the Ministry of Agriculture, the National Statistics Committee, the National Bank and other agencies for their collaboration and input during the preparation of this report. The team is particularly grateful to Mr Vladimir Amarin, the First Deputy Minister of Finance, and Mr Maxim Ermolovich, Head of Budget Policy Department for facilitating the overall coordination of the work from the Government side. viii Fiscal Reforms for a Sustainable Economic Recovery Preface n early 2010, the Ministry of Finance of Belarus requested sequenced and targeted advice from the I Bank on the fiscal reform options available across priority areas of the budget. Responding to the authorities’ request, a programmatic (two-phase) Public Expenditure Review was agreed upon. The first phase focuses on providing: (i) an overall picture of the general government budget structure and key sources of fiscal pressure; (ii) reform options (with fiscal impact) in pensions, social assistance, housing and communal services, and subsidies in the agriculture sector. This report delivers the analysis and recommendations of the first phase. The issues covered by the second phase will be decided in response to the evolving reform agenda of the Government. The second phase will also consolidate the reform options and bring them into a consistent macro-fiscal framework, assessing different implementation scenarios.  he macroeconomic environment worsened significantly since the original request for this work. T Loose fiscal and monetary policies during the second half of 2010 (in the run-up to the presidential election) exacerbated pre-existing external imbalances and other macroeconomic vulnerabilities. As pressures stemming from the widening current account deficit, low level of foreign exchange reserves and uncertainty about securing external financing mounted, a de-facto system of multiple exchange rates took hold. Multiple exchange rates persisted even after the May 2011 announcement of the National Bank of the Republic of Belarus of the new targeted exchange rate, due to lack of credibility. A macroeconomic adjustment plan is now being discussed. n the context of economic developments over the last three years, a macroeconomic adjustment I program is needed, hand-in-hand with a program of structural reforms that includes a significant fiscal component. While exchange rate and domestic demand adjustment will help in the short run, the large external imbalances also require a structural transformation of the economy over the medium term. A program of structural reform would: (i) harden budget constraints and restructure the SOE sector; (ii) implement investment climate reforms to enable FDI and enlarge the private sector; and (iii) reduce the government footprint in the economy while strengthening safety nets. These three pillars encompass a significant fiscal reform agenda. Thus, the PER offers options for reform to deal not only with fiscal consolidation, but also to tackle the deeper structural transformation needed in Belarus’ economy. The policy recommendations aim to reduce the public sector footprint and deliver a more efficient and smaller public sector in Belarus, with appropriate mitigation measures for the most vulnerable. ix Bel arus Public Expenditure Review Executive Summary Looming macroeconomic and structural challenges Belarus faces significant fiscal challenges, stemming from the structural features of its economic model. While the Government has maintained moderate budget deficits for most of the 2000s, its size and interventions have constrained the development of the private sector. The state-driven economic model—based on incomplete restructuring of the enterprise sector and broad-based income redistribution—manifests itself in the substantial fiscal costs of various transfers and subsidies to households and enterprises. Despite a significant fiscal consolidation over the past three years, general government expenditure in Belarus continue to account for about 44 percent of GDP. This is about 3 percentage points higher than the regional average and about 5 percentage points above the contries with similar per capita income. The resulting size and structure of the tax burden undercuts the competitiveness of the economy. Quasi-fiscal activities are widespread, including underpriced utility services to households and directed lending to state-owned enterprises. Directed lending—which totaled more than 8 percent of GDP in 2010—and the growing recapitalization expenditures for state- owned banks threaten the integrity of the fiscal and overall macroeconomic stance in Belarus. As of mid-2011, under a severe exchange rate and balance of payments crisis, the Government needs to pursue a carefully sequenced strategy of macroeconomic adjustment coupled with a strong program of structural reforms. Expansionary monetary and fiscal policies boosted growth in 2010, but a fast-growing current account deficit brought about substantial need for external financing. To restore external balance, a macroeconomic adjustment is urgently needed to contain domestic demand through a combination of monetary tightening, including cuts in directed lending programs, depreciation of the exchange rate and fiscal consolidation. On the fiscal side, the brunt of the adjustment falls on expenditures, given the already heavy tax burden in Belarus. The short-term tightening may be achieved by increased nominal tax collection due to rising inflation, capital expenditures cuts, freezes of public sector wages and cuts in Government housing construction programs and subsidies to enterprises in the agriculture sector. Along with the imminent expenditure containment to address the current macroeconomic challenges, deeper fiscal reforms will have to focus on a sustainable reduction of the size of the public sector. This will require expenditure rationalization, starting with inefficient subsides to consumers and producers and addressing future aging-related imbalances in the pension system, to allow for a sustainable reduction in the heavy tax burden that constrains Belarus’ competitiveness. These reforms need to be accompanied by improved targeting of social protection systems to mitigate the social impact of accelerated economic restructuring. This programmatic Public Expenditure Review (PER) proposes fiscal policy options to simultaneously tackle the need for fiscal consolidation as well as the structural challenges with significant fiscal linkages. This programmatic work has three objectives. First, it seeks to identify options for achieving x E xecutive Summary Fiscal Reforms for a Sustainable Economic Recovery Table 0-1. Options for fiscal consolidation % of GDP 2012 2013 2014 2015 Short-Term Fiscal Austerity Measures* Savings from Fiscal Tightening -1.7% -1.2% -0.9% -0.4% Capital Expenditures -1.0% -0.5% -0.5% 0.0% Goods and Services -0.3% -0.3% 0.0% 0.0% Public Sector Wage Restraint -0.4% -0.4% -0.4% -0.4% Wages and Salaries -0.3% -0.3% -0.4% -0.3% Social Protection Fund Contributions -0.1% -0.1% -0.1% -0.1% Structural Realignment of Selected Budget Expenditures** Rationalization of Agricultural Subsidies (Reduction of Amber Box Spending -0.80% -1.10% -1.10% -1.50% to 10 % of Gross Agricultural Output) Rationalization of agricultural subsidies (savings under baseline path -0.30% -0.50% -0.60% -1.10% envisaged under customs union) Rationalization of agricultural subsidies (savings under accelerated -0.50% -0.60% -0.50% -0.30% reforms, additional to the baseline path ) Utility Tariff Reform (Baseline Import Price Scenario for Gas***) -1.00% -1.60% -1.80% -2.20% Increase Belenergo’s industrial heat tariffs to cost recovery levels -0.06% -0.10% -0.11% -0.14% Eliminate preferential electricity and gas tariffs granted by Ministry of -0.30% -0.50% -0.60% -0.70% Economy resolutions on an ad hoc basis. Increase residential heat, electricity and gas tariffs to cost recovery -0.60% -1.00% -1.10% -1.40% Rationalization of Social Assistance Privileges -0.40% -0.40% -0.40% -0.40% Eliminating regressive privileges -0.20% -0.20% -0.20% -0.20% Eliminating progressive privileges (additional savings) -0.10% -0.10% -0.10% -0.10% Scaling up of Targeted Social Assistance 0.80% 0.87% 0.89% 0.94% Planned increase to mitigate adjustment effects 0.50 0.50 0.50 0.50 To mitigate impact of reductions in privileges 0.20% 0.20% 0.20% 0.20% To mitigate impact of utility tariff reforms (for households below the 0.10% 0.17% 0.19% 0.24% poverty line) Pension Reform Package (Elimination of Early Retirement Options, Increase -0.30% -0.50% -0.60% -0.80% in Retirement Age, Price Indexation, 24% Contribution Rate ) Potential Fiscal Impact of Proposed Reforms -3.37% -3.90% -3.95% -4.34% Source: World Bank staff estimates. Note: *Assumptions are based on Government plans for the medium term: 1) capital expenditure and goods and services Expenditures: Short-term retrenchment of 1 percent of GDP in 2012, followed by a return to 2010 levels by 2015. 2) Wage restraint: Real wages growth in line with GDP growth. ** Measures to support a structural realignment of expenditures and the associated estimates of their fiscal impact are based on chapters 2-5 of this report.  *** Cost coverage ratios for provision of energy services by tariffs for household as approved by the GoB medium-term programs. The estimates of fiscal and social impact of utility  tariff reforms under different gas price scenarios are presented in Chapter 3 of this report. The figures presented in this table refer to the baseline scenario; actual savings would be higher if the gas price increases more rapidly. E xecutive Summary xi Bel arus Public Expenditure Review savings in selected areas of the budget to help the macroeconomic adjustment, while facilitating the implementation of structural reforms in the economy. Second, it quantifies the fiscal impact of expenditure rationalization. Third, it takes a novel approach in systematically incorporating redistributional and social impact analyses of proposed reforms, especially on the poor.1 The analysis and recommended policies in this first installment of the programmatic work focus on four areas: rationalizing agricultural support; achieving financial viability of the energy sector; improving targeting of social assistance; and fostering sustainability of the pension system. The options for fiscal reform proposed in this report are estimated to generate average net fiscal savings of 3.9 percent of GDP annually over the medium term. To realize the medium-term (2012–15) fiscal savings, however, structural realignment of the expenditure program should begin, starting with inefficient subsidies to consumers and producers in the energy, agricultural and other sectors of the economy and introducing parametric changes in the pensions system. The overall envelope of energy subsidies to households is projected to decline more moderately in real terms in 2011 as significant tariff increases would be needed to offset the large exchange rate impact on energy import prices. But the pace of tariff adjustments would need to accelerate in 2012 and 2013, accompanied with scaling up targeted social assistance to protect low income households from the adverse effects. Subsidies to agricultural enterprises could be cut moderately in nominal terms in 2011, but the pace of reduction would need to accelerate in 2012-2014 to provide fiscal space for reductions in the tax burden. Given the implied structural changes in the expenditure program, the fiscal impact of these reforms extends beyond the four-year projection period. The envisaged pension reforms would put the system on a long-term sustainable path, mitigating the impact of demographic changes. Utility tariff reforms and the rationalization of agricultural subsidies, while containing immediate spending pressures associated with rising energy prices, would also restore long-term financial viability of these sectors. Overall, these expenditure reforms would reduce the size of the public sector, allowing for a fiscally sustainable reduction in the tax burden while safeguarding the fiscal space needed for public investment and protecting vulnerable groups from adverse effects of the crisis. A summary of the analysis and recommendations in each of the four budget areas is presented below. Rationalizing agricultural support The unreformed and underperforming agricultural sector of Belarus is a burden on the state budget. The sector contributes 8 percent to GDP while absorbing as much as 5 percent of GDP in the form of state support. Fiscal support to the agricultural sector, including tax concessions and budgetary expenditures, is substantial, accounting for 67 percent of agricultural GDP. The result of this extensive support is worrisome: the sector is largely unprofitable, with enterprises registering soaring levels of debts, often government-guaranteed, resulting in further pressure for obtaining state support (Table 0-2). 1 It should be noted that the social impact analysis presented here does not quantify the impact of the macroeconomic adjustment that the country will be pursuing in the short run to stabilize its economy. xii E xecutive Summary Fiscal Reforms for a Sustainable Economic Recovery Table 0-2. Fiscal support to agriculture in Belarus Total fiscal support* to agriculture as share of: 2005 2006 2007 2008 2009 GDP 5.0 5.5 5.5 4.7 5.2 Gross Agricultural Output 25.4 27.9 29.7 24.1 26.7 Agriculture GDP 63.8 69.5 73.7 59.1 66.7 Sources: Belstat, MoF, MoAF, ISRA. Note: * Fiscal support includes budget expenditures and tax privileges. Agricultural subsidies absorb the largest share of spending, while expenditures on non-distorting public goods are limited. Subsidies have increased as a result of growing input prices and growing expenditures on interest subsides and called guarantees for directed lending activities in the sector. Capital expenditures are down from a peak of almost 11 percent of total agricultural spending to 5 percent in 2009–10. Using the WTO classification of agricultural support, amber box measures (market- and trade-distorting state support) account for the largest chunk of spending, while the share of market-neutral green box measures remains low. Agricultural policy aims to support producers and protect consumers by keeping food prices low, but current policies distort incentives and undermine the financial viability of the sector. Farm production costs have risen quickly as a result of increased prices for fertilizers and other farm inputs, while price controls and artificially low farm gate prices have depressed farm revenues. Productivity of labor and capital in agriculture is much lower than in other sectors of the economy and profitability has been declining, with about half of Belarus’ farms being unprofitable enterprises. Continuous state support and soft loans have avoided the needed restructuring of non-viable enterprises and contributed to poor and declining farm performance. The reform agenda in agriculture requires a set of inter-related fiscal and structural reforms, along three pillars: 1. Reduce state controls and regulations in agriculture. Structural reforms would engender a gradual withdrawal of state control and support along the agricultural value chain, including phasing out production and employment targets as well as wage and price controls (of both outputs and inputs), increasing managerial autonomy and a possible privatization of farm assets. 2. Increase and diversify non-distorting measures (green box). A part of the large amber box spending should be shifted to the green box. Such measures would strengthen the long-term competitiveness of Belarus agriculture. 3. Reduce distorting subsidies (amber box). The system of providing permanent subsidies on key farm inputs, such as fertilizers and seeds, distorts producer incentives and may lead to adoption of inefficient production technologies. The focus could shift to “market smart” subsidies on products that exhibit comparative advantages so that once public support is withdrawn enterprises remain viable and competitive. E xecutive Summary xiii Bel arus Public Expenditure Review Economic liberalization and restructuring of the agricultural sector will help increase economic benefits and reduce the demand for state support. Liberalization will help facilitate labor reallocation from agriculture toward economic sectors with higher returns and from rural to urban areas. The current expensive support system for agriculture has been designed with the social objective of keeping sectoral employment. Reducing support/subsidies and sectoral liberalization will unavoidably release agricultural labor. Adjustment assistance payments, retirement packages and specialized re-training could be used to mitigate the social impact of reforming the sector. Achieving financial viability of energy services The energy market is profoundly affected by Figure 0-1. The fiscal cost of underpriced energy rising prices for energy imports from Russia. has grown rapidly Belarus generates about 97 percent of its percent of GDP 2.0 electricity from natural gas-based thermal 0.28 power plants. While still below the European 1.5 0.27 0.34 market prices, import prices of natural gas 0.10 0.20 0.28 0.16 0.06 more than tripled over the past five years. 1.0 0.18 0.09 0.12 0.09 0.12 0.08 0.06 0.11 0.05 1.10 The fiscal cost of underpriced energy amounts 0.5 0.87 0.84 0.57 0.63 to about 2 percent of GDP. Despite nominal 0 increases, energy tariffs for residential 2005 2006 2007 2008 2009 consumers have remained flat in real terms operating losses by service line Residential heat Non-residential heat Residential and agricultural electricity since 2005, while operating losses of utility Residential natural gas Preferencial energy tariffs companies have risen to 1.7 percent of GDP Sources: MoF, Belenergo, ZhKH, and World Bank staff estimates. in 2009. Residential heat services are the largest source of losses, accounting for 1.1 percent of GDP, while non-residential heat adds another 0.06 percent of GDP. Belenergo’s operating losses on electricity sales to residential consumers account for 0.28 percent of GDP. Beltopgas incurs losses equal to 0.27 percent of GDP on its residential gas sales. In addition, about 130 industrial consumers benefit from preferential electricity, heat and gas tariffs, which add 0.28 percent of GDP to the total cost of energy subsidies. Energy subsidies are financed through direct budget subsidies and—increasingly—a complex system of cross-subsidies from profits of non-residential energy sales. Explicit subsidies from local government budgets to ZhKH only cover a small fraction, 0.3 percent of GDP or about 15 percent of the total public financing provided to the sector. The lion’s share of the cost was shifted to cross-subsidies between different service lines and consumers of utility companies. Cross-subsidies, mostly from electricity and gas sales to non-residential consumers, have more than doubled in real terms, rising from 0.8 percent of GDP in 2005 to 1.7 percent of GDP in 2009. Energy subsidies are poorly targeted and largely inequitable, discourage energy efficiency and deprive the sector of the financing needed for routine maintenance and investments. The top 30 percent households in the income distribution receive 45 percent of the overall subsidy for utilities, compared xiv E xecutive Summary Fiscal Reforms for a Sustainable Economic Recovery to 15 percent for the lowest 30 percent of households. Moreover, low energy prices distort price signals to households leading to energy inefficient consumption patterns, while cross-subsidies between commercial and residential consumers impose an implicit tax on the business sector, adding to an already large tax burden. With rising energy costs, the Government plans to move to cost-reflective energy pricing by 2014 and generate fiscal savings in the range of 1–2.6 percent of GDP annually.2 The required tariff increases to achieve these cost recovery targets are substantial. Residential heat tariffs would have to be increased most aggressively by 112–256 percent by 2014, depending on import fuel prices. The planned range of the cumulative tariff increases (in real terms) for natural gas and electricity is between 110 and 233 percent and from 63 to 108 percent, respectively. With no real increases in residential tariffs, the estimated size of budgetary and cross-subsidies for 2011–14 would be in the range of 1.9–3.9 percent of GDP annually, depending on natural gas prices.3 The analysis highlights the need to comprehensively reform energy policy to reduce the fiscal burden and build a reliable energy sector capable of addressing the country’s needs. The reform options coming out of the analysis are in three major areas: 1. Reform tariffs to improve the financial viability of energy service provision and reduce the fiscal burden from under-pricing. While tariff reforms are sensitive, experience from other countries shows that they can be both politically feasible and fiscally efficient if accompanied by countervailing measures to compensate the loss in purchasing power, in particular of the poor, through targeted social assistance (discussed below). 2. Make financing transparent for tariffs below cost recovery and explicitly provision for them in the budget as direct subsidies of the end-user price. This would reduce fiscal risks and expose resources devoted to lowering energy prices to the same expenditure prioritization process as other competing expenditure programs. 3. Strengthen the regulatory environment of the energy sector to engender a climate conducive to private investments and competition. Priority reforms should aim at enabling energy companies to raise commercial financing and improve operational, managerial and financial efficiency. Better targeting of social assistance spending Belarus operates one of the most extensive social assistance systems in the region, with total spending equal to 2.7 percent of GDP and reaching about half of the population. In 2009, Belarus spent BYR 3.6 trillion (US$3.1 billion at PPP-adjusted exchange rate) on various social transfers and 2 Estimates based on four alternative scenarios for natural gas price: (i) World Bank’s Global Economic Prospects projections; (ii) /(iii) 20 percent reduction/ increase compared to (i); (iv) priced equally to low sulfur fuel oil. 3 To encourage efficient consumption of energy resources and prevent substantial growth in energy tariffs, the GoB also considers the introduction of differentiated tariffs for gas, electricity and water. Details of the proposal were not available at the time of drafting of this report. E xecutive Summary xv Bel arus Public Expenditure Review income support programs. This spending level as a share of GDP is about 67 percent more than the regional average. Social assistance programs absorb over 6 percent of consolidated general government spending. More than 80 percent of this spending is financed through the Social Protection Fund (SPF), with the remainder coming from the state budget. Categorical transfers dominate social assistance programs, with income-tested programs accounting for negligible share of spending. Social assistance spending comprises 14 main programs with varying objectives, resource commitments and eligibility criteria. The largest programs by spending size are various child-related allowances and allowances for temporary disability and medical treatment. In addition, about 32 categorical privileges and subsidies are extended to war veterans, disabled and other groups of citizens for various services ranging from public transportation to housing and utilities. The share of income-tested programs to support low-income families remains small at only about 0.05 percent of GDP in 2009, significantly below the regional average. Only four out of 14 programs apply income-related eligibility criteria. Although social assistance programs have broad coverage among the poor, targeting is weak as benefits, especially privileges, continue to reach many beneficiaries in the higher deciles of the income distribution. Belarus’ social assistance programs have broad coverage among the poorest households: 82 percent of the poorest 10 percent of households report receiving some type of benefit. However, with over 38 percent of household in the richest decile receiving some type of social assistance, significant benefits flow toward non-poor beneficiaries. Less than 10 percent of total social assistance spending benefits the population below the poverty line. Targeting of privileges is weak and leakages to non-poor households are pronounced. The richest 20 percent of the income distribution receive 20 percent of the benefits of privileges, while the poorest 20 percent receive only 27 percent. The average per capita transfer received from privileges by the higher income deciles is 36 percent higher than for the poorest quintiles. Only subsidies for school lunches and pre-school education are well-targeted, reaching the low-income spectrum of the population. Larger programs, such as public transportation, sanatorium treatments and medicines, tend to be regressive. Together, regressive programs account for 61 percent of spending on privileges. To improve targeting efficiency and reduce the cost of social assistance spending, realignment of spending from untargeted to targeted programs is needed. A set of policy options can help achieve this goal: 1. Further rationalize categorical privileges. A shift in privilege programs from rights-based to needs- based eligibility will generate fiscal savings in the range of 0.23 to 0.35 percent of GDP, depending on the scope of the reductions. Initial cuts could target regressive privileges, while those with progressive benefit incidence could be maintained. 2. To mitigate the adverse social impacts of categorical privileges rationalization and utility tariff increase on the poorest, targeted social assistance program benefits could be increased. The former would require less than 0.2 percent of GDP, and the later would cost between 0.3 to 0.7 percent of GDP. xvi E xecutive Summary Fiscal Reforms for a Sustainable Economic Recovery 3. Expand the use of income thresholds for child benefits to better support the low-income population. The total amount saved by reducing benefits to higher income households, equal to 0.2 percent of GDP, would be redistributed to households receiving similar children-related benefits in the lowest quartile of the income distribution. Under the proposed fiscally neutral reform, poverty is projected to fall by 1 percentage point. Fostering sustainability of the pension system Belarus’ aging population will put growing pressure on the pension system. As a consequence, the pension fund is projected to move from surplus into a structural deficit as early as 2014, while income replacement rates for future pensioners would decline (Figure 0-2). In addition to demographic pressures, cost containment of the pension system is needed to provide space to reduce contribution rates and ease the tax burden on labor, thereby strengthening competitiveness. Figure 0-2. The population dependency rate will  …while moving the pension system into deficit increase steadily… number of old-age people for each 100 working age people in percent of GDP 150 16 14 125 12 10 100 8 6 75 4 2 50 0 2008 2018 2028 2038 2048 2058 2068 2000 2010 2020 2030 2040 2050 2060 2070 Base case (increasing fertility, decreasing mortality) Constant fertility, decreasing mortality Contribution revenue Pension expenditure Constant fertility & mortality Source: World Bank staff projections. The pensions system in Belarus is a conventional single pillar pay-as-you-go scheme. The key characteristics of the system are: low retirement age (55 for women and 60 for men), continued participation of many pensioners in the labor market, relatively lax early retirement conditions, pension benefits indexed to nominal wages, heavily compressed pension grid and high contribution rates. The key challenge for the pension system in Belarus is to secure its long-term fiscal sustainability given the projected demographic changes, while allowing for a fiscally responsible reduction in contribution rates. Specific reform options could include: (i) gradually increasing the retirement age for both genders to 65; (ii) changing benefit indexation from wage to price indexation; (iii) implementing a reform package designed to allow for a fiscally responsible reduction in contribution rates;4 and (iv) 4 The reform package presented in Chapter 5 comprises: i) eliminate early retirement option; ii) increase retirement ages for women by six months a year and for men by three months a year until both genders reach age 65 in year 2030; iii) index pension benefits to the higher of consumer price increase or nominal wage growth minus 3 percent; and iv) gradual decrease the pension contribution rate from 29 percent in 2010 to 24 percent in 2015. E xecutive Summary xvii Bel arus Public Expenditure Review introducing notional accounts to strengthen Figure 0-3. Fiscal impact of different pension the link between individual benefits and reform measures contributions. While the size of fiscal savings percent of GDP % of insured wage bill 6 21.9 potentially generated by the four reform options varies, all would generate significant surpluses 4 14.6 to the pension fund, providing fiscal space for 2 7.3 possible rate reductions. In the medium term, 0 0 all options would mitigate the adverse impact -2 -7.3 of the changing demographic environment -4 -14.6 (Figure 0-3). Average annual fiscal savings (the -6 -21.9 estimated difference between the base case 2008 2018 2028 2038 2048 2058 2068 scenario and the reform scenario) over the next Base case 65/65 plus price indexation NDC Price indexation only 65/65 Reform package decade range between 1.3 and 3.4 percent of Sources: MoF, Belenergo, ZhKH, and World Bank staff estimates. GDP. Increasing the retirement age to 65 years for both men and women, combined with indexing pension benefits to inflation, would generate the largest fiscal savings, followed by introducing notional accounts. xviii E xecutive Summary Fiscal Reforms for a Sustainable Economic Recovery Chapter 1. Why is fiscal reform pertinent in Belarus? B elarus faces significant fiscal challenges, stemming from the structural features of its economic model. While the Government has maintained a broadly balanced budget for most of the 2000s, its quasi-fiscal activities threaten the integrity of the fiscal stance in the country. The public sector is deeply involved in the economy through government-directed lending (GDL) programs, high public investment, sizeable subsidies and transfers under a heavy tax burden. Belarus’ state-driven economic model—based on incomplete restructuring of the enterprise sector and broad-based income redistribution—manifests itself in the substantial fiscal costs of various forms of state support. Just in the last five years, the stock of government-directed lending programs has risen exponentially, from 3 percent of GDP in 2005 to almost 25 percent of GDP in 2010. Transfers under the Social Protection Fund and other social welfare programs (over 14 percent of GDP in 2010) along with steep nominal increases in public wages continue to absorb significant fiscal space. The large footprint of the state in the economy continues to soften budget constraints for economic agents and delay the restructuring of energy, agricultural and other sectors of the economy. As of mid-2011 in a deteriorating macroeconomic environment, the government-dominated economic model is under considerable pressure to rationalize fiscal policy. Expansionary monetary and fiscal policies boosted growth in 2010. The fast-growing current account deficit brought about substantial needs for external financing. Given the deterioration of political relations with Europe and the USA after the presidential elections in end-2010, retaining external investor confidence will require urgent macroeconomic adjustment based on fiscal and monetary policy tightening as well as increased exchange rate flexibility. In the context of these imbalances, public finances have become less leveraged, requiring significant expenditure cuts and substantial tightening of credit policy to contain domestic demand. At the same time, public expenditure rationalization could become an important structural compliment to the short-term fiscal tightening to facilitate the macroeconomic adjustment. Expenditure rationalization, starting with inefficient subsides to consumers and producers and addressing future aging-related imbalances in the pension system, will allow for a sustainable reduction in the heavy tax burden that constrains Belarus’ competitiveness. 1 Bel arus Public Expenditure Review A. Macro-economic vulnerabilities–Implications for fiscal policy Underpriced energy inputs from Russia and a favorable external environment helped to maintain relative macroeconomic and fiscal stability until 2008. Underpriced energy inputs from Russia and positive terms of trade (ToT) gains were at the core of Belarus’ economic growth, fuelling an average real GDP growth rate of 8.3 percent in 2001–2008 (Table 1-1). The imputed size of the average annual Russian oil and gas subsidy for 2001–2008 was equal to 14.5 percent of Belarus’ GDP. Combined with the surge in export prices for refined oil products, energy subsides contributed to cumulative ToT gains amounting to 34 percent between 2000 and 2008, which in turn fuelled strong economic growth during that period. Table 1-1. Main macroeconomic indicators, 2003–2010 2003 2004 2005 2006 2007 2008 2009 2010 Real GDP (change in percent) 7.0 11.4 9.4 10.0 8.6 10.2 0.2 7.6 CPI, e.o.p. (change in percent) 25.4 14.4 8.0 6.6 12.1 13.3 10.1 9.9 Terms of Trade (change in percent) 0.0 2.2 12.3 3.8 -2.5 11.0 -11.1 1.9 Current Account Balance -2.4 -5.2 1.4 -3.9 -6.7 -8.7 -13.1 -15.5 (percent of GDP) Foreign Exchange Reserves 0.5 0.5 0.9 0.7 1.6 0.9 2.0 1.2 (in months of imports of G&S) Net FDI (in billions of US$) 0.2 0.2 0.3 0.4 1.8 2.1 1.8 1.3 General Government Balance -1.4 0.0 -0.7 1.4 -0.6 0.0 -0.7 -1.8 (percent of GDP) PPG Debt (percent of GDP) 10.4 8.9 8.3 8.8 11.6 13.7 22.3 23.5 of which: Domestic 5.9 5.7 5.7 6.5 6.4 6.8 5.8 5.7 External 4.5 3.2 2.6 2.3 5.2 6.9 16.5 17.8 Memo: Nominal GDP (in billions of US$) 17.8 23.1 30.2 37.0 45.3 60.8 49.2 54.6 GNI per capita (US$, Atlas method) 1,610 2,160 2,760 3,470 4,240 5,380 5,560 5,560 Sources: National statistics, IMF, and World Bank. On the fiscal front, the strong economic performance prior to 2008 translated into buoyant revenue growth, supporting a pro-cyclical expansion of expenditures. The average annual growth of consolidated general government expenditures per capita was 11.3 percent in real terms between 2003 and 2008, exceeding output per capita growth for the period (9.9 percent). Consolidated general government expenditure grew from 46 percent of GDP in 2004 to 49.6 percent of GDP in 2008, while revenues soared from 46 percent of GDP to 51 percent in the same period on the back of strong growth. The expansion of expenditures was driven by subsidies and transfers, followed by capital expenditures and expansionary income policies in public sector wages and social transfers. Modest cash deficits masked the pro-cyclical weakening of the fiscal position. While the headline budget was approximately 2 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery in balance, Belarus entered the global financial crisis with an estimated structural deficit of 3.4 percent of GDP in 2008 (Box 1-1). Figure 1-1. Energy subsidies on imports from Figure 1-2. Economic growth is driven by terms Russia are in decline of trade gains in percent of GDP GDP growth 20 14 R =0.6051 2 18 12 16 2004 2008 14 10 2006 2005 12 2007 8 10 2003 8 6 2010 6 4 4 2 2 0 0 2009 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 -20 -10 0 10 20 Gas subsidy Oil subsidy terms of trade (perchent change) Sources: MoF and World Bank staff projections. Sources: MoF and World Bank staff projections. The global financial crisis was transmitted to Belarus through a plunge in trade partners’ export demand and prices, with exports falling by 35 percent in 2009, and through reduced subsidies on energy imports from Russia. Growth plummeted to 0.2 percent in 2009. Financing constraints in international capital markets reduced the availability of foreign borrowing to finance the growing current account deficit. Faced with a balance of payment crisis, Belarus requested support from the IMF (the program expired in March 2010) and undertook a macroeconomic adjustment through a combination of an early (albeit moderate) realignment of the exchange rate and tight fiscal policies. Fiscal consolidation played a key role in the macroeconomic adjustment of 2009, but structural fiscal issues remained unaddressed. The slowdown in external trade and the real sector placed downward pressures on revenues. Consolidated revenues contracted from 51 percent of GDP in 2008 to 45.6 percent of GDP in 2009, and further to 42 percent in 2010 (Figure 1-3). Revenues from profit tax, oil-related duties and VAT were the most affected. The Government restrained public wages, increased tariffs in electricity, gas and heating, delayed investment projects, saved on goods and Figure 1-3. Expenditure growth with moderate services and concluded 2009 with a fiscal deficits deficit at below 1 percent of GDP (Figure 1-4). in percent in percent 60 5 Unlike its regional peers, Belarus managed to 51 4 49.1 47.6 49.5 49 49.6 refrain from running high budget deficits during 50 45.9 47.7 46 46 47.4 48 45.7 46.4 3 43.8 42 the global financial crisis (Figure 1-5). However, 40 37.6 40.5 2 1.4 1.4 1 the economy remained dependent on external 30 0.4 0 0.0 -1 factors, including regional export demand 20 -0.7 -0.7 -1.7 -1.8 -2 (from the CIS, especially Russia), energy 10 -2.9 -3 imports from Russia, and ToT fluctuations in -4 0 -5 oil products and fertilizers. As Russia gradually 2011 2003 2004 2005 2006 2007 2008 2009 2010 planned started to move towards market-based pricing Revenues Expenditures Balance (right axis) Sources: MoF, SPF, World Bank ECA regional tables, and World Bank staff estimates. of its energy exports to Belarus, import prices Chapter 1. Why is fiscal reform pertinent in Belarus? 3 Bel arus Public Expenditure Review for oil and gas increased sharply. As a result, ToT declined by over 11 percent year-on-year in 2009, contributing to rising trade and current account deficits. Figure 1-4. Fiscal imbalances emerged with the Figure 1-5. …but Belarus’ fiscal position crisis… compares well to regional peers in percent in percent 2011-2.9 0.1 20 planned -3.0 -1.8 2010 -2.6 0.8 15 -0.7 2009 -1.8 1.1 2008 0.0 1.4 10 1.3 0.4 2007 -0.6 1.1 5 1.4 2006 0.4 1.0 -0.7 0 2005 -1.7 1.0 0.0 2004 -0.3 0.4 -5 -1.7 2003 -1.0 0.0 -10 -3 -2 -1 0 1 2 2001 2003 2005 2007 2009 2011 Consolidated general government balance Consolidated state balance (republican & local) Max Min Belarus CEBS CIS Social protection fund balance Sources: MoF, SPF, World Bank ECA regional tables, and World Bank staff estimates. Sources: MoF, SPF, World Bank ECA regional tables, and World Bank staff estimates. Growth rebounded rapidly to 7.6 percent in Figure 1-6. Growth has rebounded, but is 2010, initially on the back of recuperating threatened by unsustainable external imbalance export demand, but later increasingly fuelled in percent 15 by domestic demand (Figure 1-6). After a 11.4 9.4 10.0 10.2 10 8.6 7.6 sharp contraction in 2009, industrial output 7.0 5 grew by 11.3 percent, reflecting economic 0.2 0 recovery in major export markets, most -5 importantly Russia. But in the second half of -10 2010 domestic demand accelerated, fuelled -15 by expansionary fiscal and monetary policies -20 and credit expansion (including GDL) in the run 2003 2004 2005 2006 2007 2008 2009 2010 up to the presidential elections in December Real GDP (change in percent) Current account balance (percent of GDP) Sources: Belstat and World Bank staff estimates. 2010. This was reflected in the accelerated rate of gross fixed investment, which grew by 15.1 percent, while expansionary income Figure 1-7. Public sector wage hikes boosted policies drove household consumption up by domestic demand 10.3 percent. in thousand BYR 1,500 Expansionary fiscal policies in the run-up 1,300 to the presidential elections in 2010 also 1,100 contributed to the short-term recovery. Despite higher-than-expected revenue in 2010Q4 900 stemming from the new arrangements under 700 the Customs Union, the consolidated budget 500 deficit for 2010 increased from a planned 01/09 05/09 09/09 01/10 05/10 09/10 01/11 1.5 percent of GDP to 2.6 percent of GDP. The Month average nominal wage Monthly average real public sector wage Sources: Belstat and World Bank staff estimates. widening deficit accommodated: (i) revenue 4 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery Box 1-1. Pro-cyclical fiscal policy in Belarus The cyclically-adjusted balance shows the underlying fiscal position, eliminating cyclical effects. We statically estimate the permanent trend and cyclical component of GDP by applying the Hodrick-Prescott filter to annual GDP data for 1995-2010, extending it with projections to 2015 to mitigate end-of-series bias. Cyclically-adjusted revenue is calculated from actual tax revenues adjusted according to the ratio of potential output to actual output and the estimated revenue elasticity (1.7). We assume expenditures are inelastic, taking into account labor market rigidities and the absence of comprehensive unemployment benefits and other automatic stabilizers on the expenditure side in Belarus. While Belarus has maintained moderate cash balances, the cyclically-adjusted balance began to deteriorate prior to the crisis, reflecting a pro-cyclical policy stance. Following the growth acceleration starting in 2004, GDP appears to have been above potential between 2006 and 2008 (see left-hand figure). While the Government maintained prudent cash deficits during this period, the nominal budget figures mask an underlying weakening fiscal situation. The headline budget was approximately in balance, but the underlying structural deficit (taking into account the large positive output gap) dropped to 3.4 percent of potential GDP in 2008, thus adding to the economic overheating. This was a result of steep expenditure increases fueled by buoyant revenues in the run-up to the crisis. The size of the underlying structural deficit improved in 2009, reflecting a significant but pro-cyclical fiscal retrenchment at the time when the crisis struck and the output gap turned negative. The structural balance started to weaken again in 2010, reflecting loosened fiscal policy in the context of a nascent economic recovery (see right-hand figure).  elarus’s cyclically-adjusted fiscal balance started to deteriorate prior to the 2008 global B financial crisis GDP, 2005 constant prices, trillion BYR in percent percent of GDP 100 14 5 12 3 80 10 1 60 8 -1 40 6 -3 4 20 -5 2 0 0 -7 2000 2002 2004 2006 2008 2010 2003 2004 2005 2006 2007 2008 2009 2010 GDP growth (right axis) Actual GDP Trend GDP Output gap Cyclically adjusted balance Actual balance Sources: MoF, Belstat and World Bank staff estimates. shortfall due to changed conditions for importing oil from Russia since January 2010 (1.9 percent of GDP); (ii) increased support to agriculture due to adverse weather conditions in the summer of 2010 (0.3 percent of GDP); (iii) wage and pensions increases in excess of budgeted for 2010 (0.9 percent of GDP) (Figure 1-7); and (iv) recapitalization of the three largest major state-owned banks (1.3 percent of GDP) and called guarantees (both central and local) on bank loans (0.6 percent of GDP). The rise in the annual cost of state bank recapitalization is particularly worrisome as it de facto serves as compensation to the banks for their losses incurred because of the Government’s lending policies. Chapter 1. Why is fiscal reform pertinent in Belarus? 5 Bel arus Public Expenditure Review After being reduced from 2.4 percent of GDP in 2008 to less than 0.04 percent of GDP in 2009, bank recapitalization expenditures grew again to 1.3 percent of GDP in 2010. The recovery of 2010 was also accompanied by rising inflation and soaring external imbalances. Inflation picked up in the second half of 2010 and accelerated to 18 percent year-on-year in January–May 2011, on the back of currency devaluation and hikes in food prices and utility tariffs. The pressure from producer prices was also high, with PPI growth in January–May 2011 at close to 25 percent year-on- year. On the external front, the current account deficit expanded to close to 16 percent of GDP in 2010. The trade deficit widened despite rebounding export demand; imports surged, driven by Table 1-2. Gross external debt, growing domestic demand in the second half as of January 1st 2011 of 2010. US$, million percent of GDP Financing of the current account deficit Gross external debt 28,512 52.2 relied on significant foreign borrowing, more Total long term 15,719 28.8 than doubling external debt over the past two General Government 10,058 28.4 years. The fragility of Belarus’ external position is exacerbated by the financing structure of Monetary authorities 588 1.1 the current account, relying heavily on debt- Banks 3,059 5.6 creating inflows with low and declining FX Other sectors 2,013 3.7 reserves (Figure 1-8). FDI inflows are negligible Total short term 12,794 23.4 (2.4 percent of GDP in 2010). The Government successfully placed debut Eurobonds of Monetary authorities 1,267 2.3 US$1 billion in August 2010, followed by Banks 2,693 4.9 additional issuances of US$250 million in Other sectors 8,020 14.7 December 2010 and US$800 million in January Intercompany lending 814 1.5 2011. Increased borrowing coupled with the economic downturn, growing fiscal deficit and Memo: BYR depreciation led to an increase in the Forex, % of short term debt 39.3 stock of gross external debt, which reached Source: World Bank staff calculations, based on official data. Figure 1-8. CAD deficit is increasingly financed  …reflected in increased debt service burden by debt inflows… in percent of GDP in percent of GDP in percent in percent 20 5 5.0 25 4.5 15 0 4.0 20 3.5 10 -5 3.0 15 2.5 5 -10 2.0 10 1.5 0 -15 1.0 5 0.5 -5 -20 0 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2005 2006 2007 2008 2009 2010 Remittances FDI Portfolio investments Trade credits Loans CAD (rhs) PPGXDS/revenues of republican budget PPGXDS/XGS XDS/XGS (rhs) Sources: NBRB, Belstat, and World Bank staff calculations. 6 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery 52.2 percent of GDP (Table 1-2) in 2010, 18 percent of which is external PPG debt. PPG Box 1-2. Types of quasi-fiscal activities debt service expenditures increased more than 3.5 times over the last two years as a share Operations related to the financial system of republican budget revenues, and servicing ••Subsidized lending this debt exerts increased fiscal pressures on ••Under-remunerated reserve requirements a thinly-stretched government budget (Figure ••Credit ceilings 1-8). ••Rescue operations Low explicit budget deficits in Belarus Operations related to the exchange and provide only a partial view of the overall trade systems fiscal situation and fiscal risks. A number of ••Multiple exchange rates important activities—tax expenditures, quasi- ••Import deposits fiscal activities and contingent liabilities—need ••Deposits on foreign asset purchases to be taken into account to estimate the true overall fiscal balance and its sustainability. ••Exchange rate guarantees While a detailed review of these activities is the ••Subsidized exchange risk insurance subject of the next phase of the programmatic ••Nontariff barriers PER, this chapter provides some general Operations related to the commercial information about the direction and potential enterprise sector scope of such activities. ••Charging less than commercial prices GDL programs are a large and growing quasi- ••Provision of noncommercial services (e.g., social services) fiscal activity in Belarus. Such loans are disbursed through state-owned commercial ••Pricing for budget revenue purposes banks, and borrowers under these programs ••Paying above commercial prices to pay lower than market (interbank) interest suppliers rates. An interest rate subsidy is recorded in the Source: IMF Manual on Fiscal Transparency, 2007. budget as opposed to the compensation which banks receive on the difference between the market and the compensated rate. Moreover, Figure 1-9. Mounting directed lending programs often loans under GDL programs carry central and bank recapitalization are associated with deteriorating state budget and local government guarantees. In 2009– percent of GDP 2010, GDL grew by 60 percent on average 28 per annum, when increasingly binding budget 24 25.5 constraints forced the Government to rely 20 18.6 16.5 16.6 20.1 more on quasi-fiscal resources of commercial 16 banks (Figure 1-9). More than 60 percent of 12 11.4 12.7 10 all banking credits granted in Belarus by the 8 11 6.6 end of 2010 originated within GDL programs. 4 3.9 5.5 About 70 percent of all directed credits support 0 commercial and investment activities of the 2005 2006 2007 2008 2009 2010 Directed lending Bank recapitalization enterprise sector, while the rest provides Russian energy subsidy Directed lending, stock financing for new housing construction and thus Source: World Bank staff calculations, based on official data. Chapter 1. Why is fiscal reform pertinent in Belarus? 7 Bel arus Public Expenditure Review largely benefits households. In 2010, off-budget GDL financing (gross flow) accounted for 8.5 percent of GDP, while the stock of GDL ballooned from 4 percent of GDP in 2005 to 25.5 percent of GDP in 2010. GDL not only contributes to the growing external imbalances but also distorts the allocation of capital in the economy. The rapid expansion of credit has boosted domestic demand, undermining the needed external adjustment. Concerns about these programs relate not only to their costs, but also to the quality of lending and the future stability of the banking sector in an environment where banks have very limited autonomy for independent credit decisions. The high default rate of directed credits indicates large inefficiencies in credit allocations. Enterprise arrears (mostly SOEs) to commercial banks have increased rapidly since 2006 and accounted for over 15 percent of GDP at end-2010. Moreover, the distinction between public and publicly guaranteed (PPG) debt and state-owned enterprise (SOE) debt is unclear in Belarus, blurring the distinction between explicit and implicit liabilities. In early 2011, macroeconomic performance worsened sharply following the credit and fiscal expansion of 2010. As pressures from loose macroeconomic policies built up in early 2011, the National Bank of the Republic of Belarus (NBRB) saw its FX reserves diminish to alarming levels and resorted to a number of administrative measures. The NBRB initially relied on monetary policy tightening and restrictions on the FX market to revive dwindling supply of foreign exchange.5 The effectiveness of these measures was limited, triggering a loss in BYR convertibility on current account operations and the establishment of a multiple exchange rate regime (including official, over-the-counter in exchange bureaus, interbank and black market rates—although there are now significant pressures on exchange bureaus and interbank transactions to operate at the official rate).6 Multiple exchange rates persist even after the NBRB announced the new targeted exchange rate that depreciated the BYR from 3,155 per dollar to 4,930. Public finances in Belarus are under immense pressure in the context of the large macroeconomic adjustment needs, necessitating structural reforms and further fiscal retrenchment. The key elements of the macro-stabilization program will include significant tightening of monetary and fiscal policies as well unification of the exchange rate at a level consistent with expected external financing. At the same time, the continuation and deepening of structural reforms has become even more urgent. Since large fiscal and quasi-fiscal operations are a reflection of the state-led economic model, fiscal policy changes must go hand in hand with structural economic reforms aimed at enhancing competitiveness and long- term growth prospects. However, the timeframe available for the implementation of these policies has been markedly reduced. 5 The former included increasing the refinancing rate and other reference rates as well as reserve requirements. The latter included: (i) temporarily banning residents from purchasing FX on the domestic market to finance investment imports exceeding €50,000; (ii) ceasing FX support to banks to meet the FX demand for citizens; (iii) introducing an import priority list according to which foreign currency will be distributed on the Belarus Currency and Stock Exchange (BCSE), with payment for medicine, energy resources and foreign loans as priorities; and (iv) allowing the exchange rate on the over-the-counter market to deviate from the official rate within the band of +/-10 percent (as compared to former +/-2 percent limit). The official band remained unchanged (+/-8 percent to central parity) and allowed for BYR devaluation vis-à-vis a basket of currencies during 2011Q1 by about 6.2 percent. 6 In a statement made on March 31, the NBRB announced a moratorium on the introduction of any further measures while acknowledging as temporary the lack of BYR convertibility on current account operations. 8 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery B. Deepening structural reforms–Implications for fiscal policy Structural reform efforts intensified in 2009 with the aim of reorienting the economy to a more private sector-led model. The Government began to gradually ease administrative controls. General ceilings on monthly price increases have been eliminated while administrative price controls and restrictions on retail trade margins were abolished (except for a limited number of goods)—a key policy action supported by the World Bank’s 2009 development policy loan (DPL). The Government has signaled its intentions to stop enforcing quantitative economic targets, including those for output and employment, starting with companies that do not receive public financial support and in which the Government has a minority share. This reorientation has profound fiscal policy implications for the size and structure of both public expenditures and revenues. The size of the public sector in Belarus remains large compared to other middle-income transition economies. The Government undertook a significant fiscal consolidation in response to the 2008 crisis in order to achieve macroeconomic stabilization and to counteract growing external imbalances. In 2010, the consolidated general government expenditure-to-GDP ratio contracted by more than 5.8 percentage points of GDP compared to 2008. However, Belarus continues to redistribute about 43.8 percent of GDP through central and local government budgets and the social protection fund (Figure 1-10). This is 3 percentage points higher than the regional average and 4.7 percentage points above the predicted value corresponding to its per capita income. Figure 1-10. Belarus’ public expenditure level Figure 1-11. …creating a high tax burden continues to be high… general government expenditure to GDP ratio percent of GDP 55 2 R =0.1977 50 50 BIH HUN 45 UKR MNE EST SVK 40 45 BLR SRB POL LTV CZE 35 MDA 40 GEO BGR LTU 30 HRV ROM RUS SVN 25 35 KGZ ALB AZE 20 UNK 30 MAC TUR 15 TJK ARM 10 25 5 KAZ 20 0 0 10,000 20,000 30,000 ARM GEO ROM KGZ SVK LTU RUS LVA BGR POL HRV SRB CZE UKR EST BIH SVN HUN BLR per capita GDP (current US$) Sources: IMF Government Financial Statistics, World Development Indicators, and World Bank staff estimates. The resulting size and structure of the tax burden undercuts the competitiveness of the economy and the Government’s objective of sustaining rapid economic growth. Including the contributions to the SPF, the tax-to-GDP ratio is now at 39.6 percent of GDP, substantially lower than the pre-crisis average of 44 percent (2004–2008), but still the highest among emerging markets in the region (Figure 1-11). This heavy overall tax burden, the high marginal rates of direct taxation (including profit tax and other fees) and the complicated set of tax exemptions undermine investment growth (including FDI) and competitiveness. Chapter 1. Why is fiscal reform pertinent in Belarus? 9 Bel arus Public Expenditure Review Reducing the tax burden and simplifying the tax system Belarus’ tax system is based on consumption and income taxes, which make up 47.1 and 24.2 percent of 2010 consolidated general government revenues, respectively  (Table 1-3). The standard VAT rate is 20 percent, and the profit tax rate is 24 percent. A flat personal income tax rate of 12 percent (or 15 percent for self-employed) is applied to wage income, profits of self-employed and capital income accruing to individuals. In addition, social security contributions, accounting for 33.6 percent of consolidated general government revenues, are levied on wages at 29 percent (1 percent paid by the employee and 28 percent paid by the employer) and at 35 percent on the income of individual entrepreneurs. Belarus has made significant progress in simplifying its tax system since 2005 (Annex 1.2). Major distortive and inefficient taxes—such as turnover taxes, local sales tax on goods and services (in the Table 1-3. Structure of tax revenues (percent of total) Shares, percent 2004 2005 2006 2007 2008 2009 2010 2011* Total revenues 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Tax revenues 94.2 95.3 93.0 93.2 91.7 88.4 92.0 90.9 Corporate and income tax 21.2 21.0 21.2 20.3 21.9 20.6 24.2 23.3 o/w: PIT tax 8.4 8.0 8.6 8.4 8.2 9.2 11.0 11.0 Profit tax 9.7 10.1 10.8 10.4 11.7 9.9 11.4 11.7 CIT 1.4 1.4 0.6 0.4 0.8 0.5 0.6 0.6 Tax on territory development 1.7 1.4 1.2 1.1 1.1 1.0 1.2 eliminated Property tax 5.7 4.9 4.4 4.1 3.9 3.5 3.8 3.6 Taxes on goods and services 38.4 47.3 50.8 45.8 40.1 42.6 47.1 47.6 o/w: VAT 22.9 25.2 25.4 23.6 22.3 25.9 33.3 34.3 Excise 6.7 5.8 9.7 8.3 7.6 7.8 8.9 9.6 Foreign activity tax revenues 6.6 7.2 7.1 17.1 20.8 17.1 11.8 13.0 Other taxes and duties 1.7 12.6 7.2 5.9 4.9 4.6 5.0 3.4 Non-tax revenues 5.8 4.7 7.0 6.5 8.1 11.4 7.8 8.8 Non-repayable transfers 0.0 0.0 0.0 0.3 0.2 0.2 0.2 0.3 SPF revenues 32.5 31.6 32.0 30.7 28.4 34.6 40.3 37.1 Memo GDP, trillion BYR 50.0 65.1 79.3 97.2 128.8 137.4 163.0 200.0 Tax burden, % 42.2 45.7 45.8 46.8 47.6 41.8 39.6 35.7 Source: MoF. Note: *Budgeted. 10 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery presence of the national VAT) and the local development fee of 3 percent on net corporate profit—have been abolished. In addition, a flat personal income tax with statutory tax rate of 12 percent was introduced, as in other countries in the region. These reforms culminated in the adoption of a new Tax Code in 2009 (in effect from January 1, 2010), consolidating previous tax legislation comprised of numerous laws, presidential edicts and decrees, government resolutions and instructions—an important step in making tax legislation more transparent and predictable. However, the tax system is still characterized by a high degree of complexity, numerous tax exemptions and fragmented tax bases. There are, for example, at least 10 different taxes on income, including a personal income tax, profit tax, a single regime for agricultural producers and a simplified tax levied on qualifying organizations. Rates for some taxes, in particular the profit tax, vary across different enterprises depending on sector and/or social importance. Numerous tax exemptions, deferrals, rescheduling and forgiveness are applied to individual enterprises or whole sectors (for example, agricultural production), while the personal income tax also recognizes a large number of deductions. Reduced VAT rate (10 percent) is applied to main foodstuff, several goods for children and other goods and services. Utility services and medicine are exempted from VAT. All this narrows the tax base and can lead to inefficient resource allocation across the economy. The contribution of consumption taxes has been growing, especially after the 2 percentage- point increase in the VAT rate in 2010. Despite numerous exemptions, VAT revenues are now the primary source of tax revenue. The current VAT rate of 20 percent is at the same level as in most peer countries. The tax burden on the business sector, in particular profit taxation, remains high. Profit tax revenue contributes around 10 percent of the tax revenue, higher than in most neighboring countries (Figure 1-12). Statutory tax rates in Belarus are higher than in most countries in the region, affecting competitiveness. For example, the profit tax standard statutory rate is 24 percent. Prior to 2011, an additional local development fee applied to the same base net of the profit tax, thus raising the effective profit tax rate to 26.3 percent. Figure 1-12. Statutory tax rates in Belarus are high Payroll tax rate Corporate income tax rate percent percent 50 30 45 1 40 4 25 35 3 25 20 30 0 5 9 15 20 18 25 0 15 10 20 40 38 24 25 15 28 31 10 19 20 20 20 20 20 20 26 24 15 10 20 17 18 22 17 20 5 10 5 0 0 0 GEO RUS KAZ MDA LVA LTU SRB TUR HRV BLR UKR POL MDA SRB LVA POL HRV GEO KAZ LTU RUS TUR BLR UKR Employer contribution Employee contribution Sources: Various country reports. Chapter 1. Why is fiscal reform pertinent in Belarus? 11 Bel arus Public Expenditure Review In addition, enterprises in several sectors pay mandatory contributions to various innovation funds—a highly distortionary and investment-unfriendly taxation practice. SOEs and enterprises with joint state and private ownership make mandatory contributions to 26 national and seven regional innovation funds to finance research and development expenditures of companies. Private companies also contribute to the innovation funds, but on a voluntary basis. The contribution rates range from 0.25 percent up to 19 percent of the cost of production and services of these companies. Contributions differ from one enterprise to another and are determined by the fund each enterprise belongs to. Collected contributions in 2009 totaled about 1.3 percent of GDP. The three biggest innovation funds account for over 60 percent of revenues from contributions, and are administered by the Ministry of Energy, the Ministry of Architecture and Construction and the Ministry of Communication. The tax burden on labor is also significant. While the revenue generated by the personal income tax is lower than in other countries, the overall tax burden on labor is substantial due to the high social security contribution rate, which is essentially levied on the same tax base. Taken together payroll taxes reach maximum marginal rates of over 40 percent. Gradual and steady increases in pension payments, driven by wage indexation of benefits, have not allowed a reduction in the high marginal rate of payroll taxes, discouraging entrepreneurs and encouraging under-reporting of income. The heavy reliance on taxation of labor and capital is harmful to growth. Further tax reforms should aim at eliminating remaining tax exemptions and distortions while focusing on reductions in statutory rates of major direct taxes, such as the profit tax and payroll taxes. Realigning the composition of expenditures The size and composition of expenditures reflect Belarus’ state-driven economic model, which is based on incomplete restructuring of the enterprise sector and broad-based income redistribution. This manifests itself in the substantial fiscal costs of various forms of state support, which have softened budget constraints and delayed restructuring in energy, agricultural and other sectors of the Figure 1-13. The composition of general government expenditures Functional expenditure composition Economic expenditure composition in percent of GDP in percent of GDP 2011 11.4 13.5 5.5 3.9 1.9 0.6 3.3 0.3 2011 -1.15 20.2 5.9 5.9 7.9 1.3 planned planned 2010 12.6 14.3 5.1 4.0 2.4 0.7 3.4 1.3 2010 -0.05 21.6 8.4 6.1 7.1 0.7 2009 10.3 13.5 5.2 4.2 3.7 0.9 4.1 4.4 2009 24.2 8.1 6.3 6.7 0.32 0.8 2008 14.2 12.6 5.7 4.2 2.4 2.2 2.9 5.5 2008 23.5 10.1 6.7 6.7 2.03 0.6 2007 12.3 13.4 5.7 4.5 2.7 2.7 4.2 3.6 2007 23.4 8.5 7.0 8.0 1.83 0.4 2006 11.0 13.5 6.9 5.0 2.5 3.1 4.4 1.4 2006 22.0 9.6 7.2 8.2 0.26 0.4 2005 14.6 13.5 6.9 5.0 2.1 3.1 1.8 1.1 2005 21.6 9.4 7.7 8.1 0.86 0.4 2004 14.5 13.1 6.7 4.9 2.3 2.8 1.5 0.2 2004 20.2 8.7 8.0 7.8 0.52 0.5 2003 16.2 13.3 6.5 4.9 2.6 2.6 1.4 0.2 2003 20.8 8.6 8.4 7.8 0.92 0.5 0 10 20 30 40 50 -10 0 10 20 30 40 50 Other Education Housing & communal services Agriculture Subsidies & transfers Goods & services Net lending Social policies Health Transport Fuel & energy Capital expenditures Wages & salaries Interest Sources: MoF, SPF. 12 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery economy (Figure 1-13). Public investments (10.1 percent of GDP in 2008) were a major component in gross capital formation. Transfers under the SPF and other social welfare programs (14 percent of GDP in 2010) together with steep nominal increases in public wages broadly distributed ToT gains, contributing to low poverty and inequality but fuelling domestic consumption and absorbing significant fiscal space in the years leading up to the crisis. The fiscal cost of state support for the enterprise sector has been increasing steadily since 2004. The overall cost of state support increased from 7.6 percent of GDP in 2004 to 10.4 percent of GDP in 2008, largely driven by subsidies growth, which rose from 3 to 4.3 percent of GDP during this period (Figure 1-14).7 The total cost of state support, including tax privileges, declined in 2009 during the global crisis, but remained high at almost 9 percent of GDP. Moreover, the reduction in the direct budget financing of SOE support in 2009 was accompanied by the expansion in directed credit programs, under which immediate costs were shifted from the budget to the banking system. In addition to Figure 1-14. State support remains extensive high fiscal costs, the existing system distorts in percent of GDP the business environment and undermines 2010 competition. All enterprises face a high level 0.1 2.2 0.2 3.8 0.6 1.26 2009 of taxation, while allocation of state support 0.3 3.1 0.2 3.7 0.2 0.04 2008 remains unequal and discretionary. According 0.2 3.3 0.2 3.7 0 2.35 to the database managed by the Ministry 2007 0.6 4 0.1 3.8 0.1 0.57 of Economy, more than 3,000 enterprises 2006 0.1 3.6 0 3.6 0.2 0.95 receive state support every year. An economic 2005 0 3 0.1 3.6 0.1 1.06 system with lower statutory taxes, fewer tax 0 1 2 3 4 5 6 7 8 9 10 exemptions and fewer subsidies would be Other programs Tax benefits Interest rate subsidies Budget subsidies Called guarantees Bank recapitalization more supportive of enterprise restructuring Sources: MoF, SPF. and productivity growth in the longer term. Over 10 percent of total budget outlays are expenditures on agriculture, accounting for 2.6 percent of GDP on average in 2003–2010. Overall expenditure on agriculture is significantly larger, however, as the sector remains the primary recipient of state support through tax concessions. Most of the programmatic state support in Belarus benefits agricultural producers who operate under a set of sector-wide tax exemptions and reduced tax burden, in addition to the massive explicit budget subsidies. The total cost of government support to agriculture, including budget, off-budget resources and tax concessions, is estimated at over 5 percent of GDP. Housing and communal services also receive a notable amount of budget resources and cross- subsidization. Public expenditure on housing and communal services—including energy, waste collection, housing maintenance, etc.—increased in the past decade to reach over 3 percent of GDP in 2010. Rising expenditure in the energy sector was driven by price increases of gas and oil imports from Russia, and pressures will continue if Russia further reduces subsidies on energy supplies. In addition to explicit budgetary subsidies, there are also sizable and growing cross-subsidies in the energy sector. 7 See also World Bank (2010). Chapter 1. Why is fiscal reform pertinent in Belarus? 13 Bel arus Public Expenditure Review Under-priced residential gas, electricity and heat supplies are cross-subsidized from profits on gas, heat and electricity sales to non-residential consumers, amounting 1.7 percent of GDP. Public expenditure on social policies accounted for an average of 13.3 percent of GDP in 2003-2007, reflecting the generous redistributive nature of the economic model in Belarus. These expenditures declined only marginally during the 2008 crisis, and picked up again in 2009–2010. In 2010, social expenditures reached 14.3 percent of total public spending. Going forward, aging-related pressures will put pension expenditures on an upward trajectory, moving the balance of social protection into a structural deficit. The budget composition also reflects Belarus’ economic model: •• Subsidies and transfers are the largest component of expenditures, accounting for 22 percent on average in 2003–2011. More than 12 percent of these are transfers under the SPF and other social welfare programs; the rest is almost equally distributed between consumer and producer subsidies. •• High levels of capital expenditures—on average over 9 percent of GDP p.a. in 2003–2008— have strongly contributed to growth. However, capital expenditures took the hardest hit during the economic downturn and the fiscal consolidation. The 2011 budget stipulates further reductions in capital expenditures, to less than 7 percent of GDP. •• Pro-cyclical wage policy has allowed wages and salaries to grow in step with an expanding economy prior to 2008. While 2009 saw some wage restraint, the trend of growing wages continued in 2010 with wage and pension increases totaling a cumulative 50 percent over the year, including a 30 percent increase effective November 1, 2010. This will have a carryover effect on the 2011 budget, raising wage expenditures as a share of GDP to almost 8 percent, i.e. to the pre-crisis level. •• Net lending has also been expansionary, totaling around 2 percent of GDP in 2007-2008 and on top of the already large GDL. Net lending contracted with the post-crisis fiscal consolidation. Belarus has a robust public financial management (PFM) system, well-placed to accommodate proposed realignments of expenditures. Belarus has an effectively functioning fiscal and budget management system, which has enabled the Government to finance and execute a budget delivering public services to the general population while supporting the macro-fiscal discipline and the substantial fiscal consolidation undertaken in 2009. During the last decade the Government launched a number of reforms aimed at increasing the efficiency of public finance, including introducing program budgeting, bringing most extra budgetary funds within the budget, and improving the quality of public investment spending. In a number of PFM dimensions Belarus fairs well vis-à-vis its regional peers (Box 1-3). 14 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery Box 1-3. Strengths and weaknesses of Belarus’ PFM system General Assessment. The 2009 PEFA assessment concluded that Belarus has an effectively functioning fiscal and budget management system. Furthermore, the Government has launched a number of reforms aimed at increasing efficiency of public finance, including introducing program budgeting, bringing most extra-budgetary funds within the budget, and improving the quality of public investment spending. However, a number of PFM weaknesses remain, especially in strategic allocation, transparency and accountability of public spending. Scoring. Out of 28 scores, Belarus received 10As, 8Bs, 7Cs, 3Ds. Belarus scores relatively well on budget credibility, comprehensiveness, classification, orderliness of budget process, taxpayer registration and collection and many aspects of budget execution. While the budget does not have a particularly strong policy or strategic focus, the strong control framework ensures a high degree of compliance with prevailing laws and regulations. Despite substantial progress in credibility, transparency and unreported quasi-fiscal activities, public access to key fiscal information, taxpayer obligations and liabilities, public procurement, recording and management of cash balances and quality and timeliness of financial reporting, the overall picture is mixed and a number of improvements are pending. The Figure 1-15. Belarus and ECA average PEFA most challenging areas are multi-year fiscal scores* planning, formula-driven intergovernmental 4 Credibility of the budget fiscal transfers and internal and external 3 audits. External scrutiny Comprehensiveness and audit 2 and transparency Benchmarking: Belarus and ECA average. 1 Belarus scores above ECA regional averages Belarus 0 ECA on five of the six PEFA dimensions (Figure 1-15). Belarus scores less well for policy- Accounting, recording and reporting Policy based budgeting based budgets, reflecting primarily that other countries in the region started introducing Predictability and control in budget execution medium-term expenditure frameworks earlier Note: *Comparisons are made based on PEFA reports for 14 ECA economies. and are further along in implementation. Sources: World Bank (2009a). Belarus: Public Expenditure and Financial Accountability (PEFA). Public Financial Management Assessment, April 2009, Report No. 48239 – BY; PEFA reports for 14 ECA countries (www.pefa.org). C. Fiscal challenges In a precarious macroeconomic environment, Belarus currently faces fiscal pressures stemming from the need to simultaneously combat serious macroeconomic imbalances and deepen structural reforms. Economic policies in the short term will have to adequately address a number of risks intensifying the fiscal policy challenges in Belarus: •• Restoring confidence in the currency. With soaring demand for foreign exchange in early 2011, net international reserves plummeted in an effort to support the exchange rate. If steps are not taken to unify the exchange rate and tighten monetary and fiscal policies, multiple currency regimes Chapter 1. Why is fiscal reform pertinent in Belarus? 15 Bel arus Public Expenditure Review will continue, external financing needs will exacerbate and macroeconomic vulnerabilities will increase. •• Mobilizing external financing has become highly uncertain following the elections. The cooling of political relations with Europe and the USA after the presidential elections in end-2010 has increased the risks facing investors. •• Particular risks are associated with the growth prospects of key trading partners (mainly Russia) and the international prices of fertilizers and oil products. If external demand does not recover, the current account deficit will continue to widen. Its financing is likely to continue to rely on rapid debt accumulation, including public debt. Servicing the debt will shrink the fiscal envelop currently available for generous government programs, including GDL. Launching a macroeconomic adjustment is urgently needed to attain external stability, especially in view of a precarious reserves position and a rapidly falling exchange rate. The adjustment will need to feature a combination of fiscal and monetary tightening to contain domestic demand, including cuts in GDL programs and a unified and flexible exchange rate. On the fiscal side, the brunt of the adjustment will fall on expenditures, given the already heavy tax burden. The fiscal adjustment will need to rely on cuts across the board (subsidies and transfers, capital expenditures and expenditure on goods and services) as well as on a speedy increase of utility tariffs to cost recovery levels. The short-term contraction of discretionary spending will help to ease pressing macroeconomic imbalances. The Government has revised its 2011 government deficit target (excluding the social protection fund) downwards from an initially 2.9 percent of GDP to 1.5 percent of GDP. The contraction will be achieved through: i) increased nominal tax collections driven by rising inflation and to a lesser extend policy changes,  including rate increases in excise rates (equal to about 1 percent of GDP); ii) spending cuts in capital expenditures (of about 1 percent of GDP) by delaying capital projects in  road construction and other areas (strategic investments, for example in improving energy efficiency will be protected); and iii) increases in social assistance spending to mitigate the impact of the adjustment on vulnerable  households (equal to about 0.5 percent of GDP). n addition, public wage restraint and cuts in Government housing construction and agricultural machinery I programs should be considered to deepen the fiscal adjustment. Over the medium term, a structural realignment of expenditure would need to begin with the 2012 budget preparation. A gradual rationalization of inefficient subsidies in energy, agricultural and other sectors of the economy is essential to reduce the overall size of the public sector and create fiscal space for fiscally sustainable tax reductions, public investment and protection of vulnerable groups from adverse effects of the crisis. 16 Chapter 1. Why is fiscal reform pertinent in Belarus? Fiscal Reforms for a Sustainable Economic Recovery Addressing these challenges will require moving ahead with fast and comprehensive fiscal reforms, encompassing both the structure of expenditures and revenues. This Public Expenditure Review provides analysis and recommendations aimed at supporting a realignment of public expenditures covering key areas of the budget. The agriculture and energy sectors absorb large and growing subsidies (Chapter 2 and 3). Further reforms to improve the targeting of social assistance spending are critical to allow for an expansion of benefits to mitigate the impact of structural reforms, including proposed cuts in consumer subsidies, on the most vulnerable while containing costs of these programs (Chapter 4). Finally, pension reforms are needed to ensure long-term sustainability of the SPF (Chapter 5). Chapter 1. Why is fiscal reform pertinent in Belarus? 17 Bel arus Public Expenditure Review Chapter 2. Agriculture: Hardening budget constraints T he unreformed and underperforming agricultural sector of Belarus is a burden on the state budget. The sector contributes 8 percent to GDP while absorbing as much as 5 percent of GDP in form of state support. Fiscal support to the agricultural sector, including tax concessions and budgetary expenditures, is substantial, accounting for 67 percent of agricultural GDP. The result of this extensive support is worrisome: the sector is largely unprofitable, with enterprises registering soaring levels of debts, often government guaranteed, putting further pressure for obtaining state support. Agriculture in Belarus is heavily regulated, with price controls and state interventions through the entire value chain dominated by large state-owned farms. The state sets production targets, supplies agricultural inputs at subsidized prices, controls wages and procures most agricultural output at regulated farm gate prices. The bulk of state support to agriculture takes the form of subsidies directed predominantly to underperforming enterprises on inputs such fertilizers and pesticides as well as growing subsidies to compensate for bank losses. Only a small portion of support goes to the provision of public agricultural services. The overarching objective of agricultural policy in Belarus is to support agricultural producers and protect consumers by keeping food prices low, but current interventions distort incentives and undermine the sector’s financial viability. Farm production costs have risen quickly as a result of increased prices for fertilizers and other farm inputs, while price controls and artificially low farm gate prices have depressed farm revenues. Productivity of labor and capital in agriculture is much lower than in other sectors of the economy and profitability has been declining, with about half of Belarus’ farms not generating profit. State support and soft loans have avoided the needed restructuring of non-viable enterprises and contributed to poor and declining farm performance. A reorientation of the existing policy framework is needed to improve agricultural competiveness and enable further trade integration. In addition, distortions to production and trade are likely to become 18 Fiscal Reforms for a Sustainable Economic Recovery issues in Belarus’ WTO negotiations and any future bilateral/multilateral trade negotiations, thus jeopardizing the country’s export prospects and slowing trade integration. A reformed policy framework could ensure sustainable long-term agricultural growth through stronger emphasis on productivity and market competitiveness, resulting in a smaller but more transparent and efficient support envelope for the sector. A three-pronged approach would include changes in both regulatory and fiscal policies: (i) liberalize agricultural prices for both inputs and outputs and reduce state intervention in farm management decisions; (ii) reallocate and subsequently reduce distorting “amber box” subsidies; and (iii) increase and diversify non-distorting “green box” measures.8 Economic liberalization and restructuring of the agricultural sector will help increase economic benefits and reduce the demand for state support.9 Liberalization will help facilitate labor reallocation from agriculture toward economic sectors with higher returns and from rural to urban areas. The current expensive support system for agriculture has been designed to keep sectoral employment. The reform of the system leading to reduced support/subsidies and sectoral liberalization will unavoidably release agricultural labor. Adjustment assistance payments, retirement packages and specialized re-training could be used to mitigate the social impact of reforming the sector. A. Institutional framework of the agricultural sector Accounting for some 8 percent of output and close to 10 percent of employment, agriculture continues to play an important role in the Belarusian economy, but it has not been a primary driver of economic growth. The sector has grown steadily since 2000, with cumulative growth of close to 40 percent. However, compared to the overall GDP growth of 50 percent over the same period (2000–2009), the growth in agriculture has been slower and more volatile. The agro-processing industry produces approximately 4–4.5 percent of GDP, with an average annual growth rate of 7.8 percent. Figure 2-1. Food exports have grown and now contribute positively to the trade balance Agriculture and agro-processing play an in US$ m 3,500 increasingly important role in Belarus’ 3,000 foreign trade. Agricultural exports accounted 2,500 for 7.9 percent of total Belarus’ exports 2,000 1,500 on average during 2001–08. The share of 1,000 agricultural exports in total exports increased 500 to 10.7 percent in 2009 and almost 13 percent 0 in 2010. During this period Belarus turned -500 -1,000 from a net food importer to a net food exporter 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 (Figure 2-1). The positive balance in agricultural Agricultural export Agricultural import Agricultural net export Source: World Bank staff estimates from Belstat data. trade accounted for almost US$1.9 billion, Note: *HS 1-24. 8 For an explanation of amber and green box measures, see Box 2.1 below. 9 Structural issues in the agricultural sector will be discussed in greater detail in the forthcoming Belarus Country Economic Memorandum. Chapter 2. Agriculture: Hardening budget constraints 19 Bel arus Public Expenditure Review or 0.7 percent of GDP. However, two-thirds of the agricultural export growth was due to the price factor, which may not be sustainable going forward. Belarus has benefited from the recent food price increase (terms of trade in agricultural goods improved by 9.6 percent in 2010) but it also became more vulnerable to food price volatility. High concentration of food exports also intensifies the risks. The primary destination of Belarus agro-food exports is Russia, accounting for 88 percent of total agricultural exports in 2010. The share of Russia is especially high in the most important export group, dairy products (over 90 percent in 2010), and close to 100 percent for exports of meat products. Belarus agriculture is dominated by large state-owned farms. Some 88 percent of total agricultural land is operated by over 1,600 state agricultural enterprises, with an average size of about 4,700 hectares and contributing about 70 percent to gross agricultural output. Private (peasant) farms in Belarus own a negligible 1.2 percent of agricultural land. There are approximately 2,000 such farms, with an average size of about 50 hectares. The rest of agricultural land (11.2 percent) belongs to over one million household plots with an average size of about less than a hectare. Large commercial farms produce more capital-intensive products such as grains, sugar beets and animals, while small households specialize in labor-intensive products such as potatoes, vegetables and fruits. The sector is heavily regulated, with price controls and state interventions through the entire value chain. Notwithstanding progress in liberalizing prices in the past few years,10 agricultural and food prices for “socially important” goods remain highly regulated, including through fixed prices, price ceilings and capping of trade margins.11 At the farm gate level, the state determines the prices for agricultural products sold to state procurement organizations, and these account for the majority of agricultural sales in Belarus. For all farm gate sales beyond the procurement network, the state sets recommended prices based on a state-set formula, agreed by the buyers and sellers, and then reported to and agreed with the local agricultural committees. The state also sets production targets, supplies agricultural inputs at subsidized prices and controls wages. By deciding on specific input uses, selecting and procuring the inputs, providing them at preferential prices, etc., the state exercises tight controls, greatly reducing farmers’ managerial autonomy. While the current policy and regulatory framework aims to support agricultural producers and protect consumers by keeping food prices low, it distorts incentives and undermines financial viability of the sector. Price controls distort market signals, undermining incentives of producers and leading to inefficient allocation of resources within the sector. Especially, since production costs have been rising over the past years due to price increases for fertilizers and other farm inputs, price controls and artificially low farm gate price, have depressed farm revenues. While this has cushioned the impact of rising production costs on food prices, it has imposed a cost on the rest of the economy in the form of subsidies, credit and tax concessions funneled to the sector. 10 This included reducing the list of socially important goods and creating a closed limited list of goods with regulated trade margins. Decree 72 from February 25, 2011 further liberalized prices by: (i) eliminating the right to introduce mandatory ceilings on price increases by government resolution; (ii) reducing the list of goods and services with prices (tariffs) subject to regulation; and (iii) repealing the mandatory justification of price calculation of goods and services (with the exception of goods and services subject to price regulation). The list of regulated prices, however, remains rather long and agricultural and food prices remain the least liberalized. 11 The list of socially important goods includes the following food items: baking flour, bread and bakery products (excluding scone bakery products, ring- shaped and dried crust products, cakes, pies, donuts); milk (excluding sterilized, baked and goat milk), kefir, sour cream, curds, powdered baby milk, canned meats for children, meat (beef, pork), sugar, artificial sweeteners for patients with diabetes mellitus, fresh potatoes, fruit and vegetable products (per latest amendments by the GOB Resolution 752 from June 8, 2009). 20 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery B. The fiscal cost of agricultural support Budget expenditures in agriculture are large, accounting for 7.9 percent of total public expenditures, and roughly half agricultural GDP. In real terms, expenditures for agriculture increased by 50 percent from 2005 to 2009. In 2009, public expenditures for agriculture accounted for about 4 percent of GDP but a staggering 53 percent of agricultural GDP (left side, Figure 2-2). Agriculture receives the most budget support of all economic sectors, taking up to 40 percent of budget resources earmarked for the national economy. Figure 2-2. While budgetary spending on agriculture has remained stable, tax collections from agricultural enterprises have declined in billion BYR, 2005 constant prices in percent in billion BYR, 2005 constant prices in percent 4,500 70 250 5 4.4 4,000 58.7 60 54.0 3,500 53.4 200 3.7 4 47.2 45.8 50 3,000 3.1 2,500 40 150 2.6 2.4 3 2,000 3,811 30 227 214 3,579 3,455 100 189 197 2 1,500 3,161 175 2,416 20 1,000 50 1 500 10 0 0 0 0 2005 2006 2007 2008 2009 2005 2006 2007 2008 2009 Ag budget expenditures % of ag budget spending in ag GDP Tax collection from ag enterprises Tax collection as % of ag GDP Sources: Belstat, MoF, ISRA. Note: Tax collections excluding payroll taxes of employees in the agricultural sector. While budget support has been on the rise, tax collections have declined, increasing net transfers to the sector. State subsidies/transfers to the sector have grown in absolute terms over the last five years, but tax collections (excluding payroll taxes) have stayed constant, thus registering a decline in real terms (right side, Figure 2-2). Taxes and fees paid by farms to the budget, mostly under a high concessional tax regime for farms (discussed below), have declined to 2.5 percent of agricultural GDP, due to stagnant revenues of agricultural enterprises. Therefore, net transfers paid from the state budget to agricultural enterprises are quite significant and have been growing, accounting for 49 percent of agricultural GDP (2009). The size of net transfers to the sector is four times larger than the amount of net profits generated in the sector in 2009, compared to 2.6 times in 2005 (see Appendix 2.1). Apart from significant annual budgetary outlays, the sector benefits from a reduced tax burden due to a simplified highly concessional tax regime. In 2000, the Government introduced a unified agricultural tax of 2 percent on gross turnover, which replaced all taxes paid by agricultural producers except for the VAT and payments to the Social Protection Fund (SPF) and the Innovation Fund. In addition, agricultural producers enjoy a reduced VAT rate of 10 percent (vs. the regular rate of 20 percent) and a reduced SPF rate of 30 percent (vs. a regular 34 percent).12 Although the unified tax scheme is not compulsory, most producers (98.7 percent as of 2009) chose to pay this tax, with the remainder paying regular taxes (corporate profit tax, income tax, property tax and a range of local taxes). 12 Institute for Systemic Research in Agriculture (ISRA) of the National Academy of Sciences of R.B. Chapter 2. Agriculture: Hardening budget constraints 21 Bel arus Public Expenditure Review Tax concessions in agriculture result in foregone revenue equal to about 1 percent of GDP  (see Appendix Table 1). They have been estimated to account for about 60 percent of the total amount of tax payments that would otherwise have been made under the regular tax regime.13 The estimated amount of annual tax concessions for agriculture adds approximately one-third to the budgetary support, thus raising the total fiscal support to agriculture to 7,104 billion BYR in 2009, which is over 5 percent of country’s GDP and 67 percent of agricultural GDP (Table 2-1). Table 2-1. Fiscal support to agriculture in Belarus 2005 2006 2007 2008 2009 Total fiscal support to agriculture* (billion BYR) 3,263 4,334 5,370 6,046 7,104 Budget expenditures in agriculture (billion BYR) 2,416 3,369 4,277 4,678 5,680 Tax privileges for agriculture (billion BYR) 847 965 1,092 1,369 1,423 Agricultural budget expenditures as share of: Total budgetary expenditures 7.7 9.0 9.0 7.3 8.9 GDP 3.7 4.3 4.4 3.6 4.2 Gross Agricultural Output 18.8 21.7 23.6 18.7 21.4 AgGDP 47.2 54.0 58.7 45.8 53.4 Net profits in agriculture 315 357 428 259 462 Total fiscal support to agriculture as share of: GDP 5.0 5.5 5.5 4.7 5.2 Gross Agricultural Output 25.4 27.9 29.7 24.1 26.7 AgGDP 63.8 69.5 73.7 59.1 66.7 Net profits in agriculture 426 459 538 335 577 Source: Belstat, MoF, MOAFA, ISRA. Note: * Fiscal support includes budget expenditures and tax privileges. In international perspective, the level of fiscal support to agriculture in Belarus is considerably higher than in other countries  (Table 2-2). For example, the level of transfers to agriculture as a share of GDP in the highly-subsidized economies of the EU-27 as well as in the USA and Russia is four to five times lower than in Belarus. The share of these transfers in agricultural value added is approximately 10 percent in Russia, 30 percent in Ukraine, and 49 percent in the EU-27—all below the Belarus level of close to 60 percent. Transfers to agriculture in Belarus calculated per hectare of agricultural land—at US$232 per hectare14 in 2009—are lower only than the EU-27 transfers (US$556 per hectare), but are higher than per-hectare agricultural transfers in all other comparators. 15 13 The general tax regime includes the following taxes: corporate profit tax (24 percent), property tax (1 percent), VAT (20 percent), local taxes (max 3 percent), payments to the RAPS Fund (1 percent), contributions to the Innovation Fund (0.25 percent), contributions to the Social Protection Fund (34 percent). For more details see Appendix Table 1. 14 ISRA. 15 Transfers to agriculture per hectare of agricultural land in: Russia (US$18/ha), Ukraine (US$46/ha), Australia (US$6/ha), and the USA (US$214/ha). Data refer to 2009 for Belarus and 2007 for other countries. 22 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery Table 2-2. International comparison of fiscal support for agriculture Agriculture value Fiscal support to Fiscal support, % Fiscal support per ha of added, % GDP agriculture, % GDP agriculture value added agriculture land, US$/ha Belarus 7.78 4.15 67 232 Australia 2.41 0.32 13 6 Canada 2.20 0.52 24 102 EU-27 1.23 0.60 49 556 USA 1.08 0.46 42 214 Ukraine 7.55 2.25 30 46 Russia 4.76 0.45 10 18 Source: 2009 World Bank Policy Note, MoAF and ISRA for Belarus data. Note: Data for Belarus as of 2009; for the rest of the countries, data is as of 2007 or latest available year. Budgetary allocations for agriculture in Figure 2-3. RAPS represents the largest source Belarus presently come from three sources: of annual budgetary spending on agriculture  republican budget, earmarked Republican in percent 100 Agricultural Producers Support Fund (RAPS) and local budgets (Figure 2-3). The RAPS is the 80 largest source of annual budgetary spending 60 on agriculture, accounting for over 50 percent of the total on average during 2005–10. Local 40 budget spending accounted for about one-third 20 of total agricultural spending on average during 0 the same period, and the rest came from the 2005 2006 2007 2008 2009 2010 general republican budget. Local budgets RAPS fund Republican budget Sources: MoF, MoAF, ISRA. C. Composition of spending The composition of agricultural spending reflects interventionist agricultural policies, with subsidies absorbing the largest share of spending and limited expenditures on non-distorting public goods. Subsidies represent the largest category of agricultural expenditures. Subsidies have been increasing as a result of rising input prices as well as growing expenditures on interest subsides and called guarantees for directed lending in the sector. Capital expenditures fell to 5 percent of total agricultural spending in 2009–10, after reaching almost 11 percent in previous years (Figure 2-4). Using the WTO classification of agricultural support, “amber box” measures (market and trade distorting state support) account for the largest chunk of spending, while the share of market neutral “green box” measures has been growing but remains low (Box 2.1 and Figure 2-4). Chapter 2. Agriculture: Hardening budget constraints 23 Bel arus Public Expenditure Review Box 2-1. WTO classification of domestic support to agriculture Amber Box. All domestic support measures considered to distort production and trade (with some exceptions) fall into the amber box, which is defined in Article 6 of the Agriculture Agreement as all domestic supports, except those in the blue and green boxes. These include measures to support prices and subsidies directly related to production quantities. Minimum levels of amber box supports are permitted under the WTO: 5 percent of agricultural production for developed countries and 10 percent for developing countries. Blue Box. This includes amber box measures with conditions designed to reduce distortions. Any support that would normally be in the amber box is placed in the blue box if the support also requires farmers to limit production. At present there are no limits on blue box subsidies. Green Box. Green box subsidies, defined in Annex 2 of the Agriculture Agreement, must not distort trade, or at most cause minimal distortion (paragraph 1). They have to be government- funded (not by charging consumers higher prices) and must not involve price support. They tend to be programs that are not targeted at particular products, and include direct income supports for farmers that are not related to current production levels or prices. They also include environmental protection and regional development programs. Green box subsidies are allowed without limits provided they comply with the policy-specific criteria set out in Annex 2. Source: http://www.wto.org/english/tratop_e/agric_e/agric_e.htm Amber box measures account for the largest chunk of agricultural support Agricultural support relies heavily on amber Figure 2-4. Amber box measures absorb the lion box measures that distort agricultural share of spending markets and trade, and which are subject in percent to reduction commitments under the WTO. Amber box support measures in Belarus are 2010 88 12 dominated by five major categories, accounting for 86 percent of total agricultural expenditures 2009 93 7 and about 40 percent of agricultural GDP on average during 2008-2010 (Figure 2-4):16 2008 92 8 1. Interest rate subsidy, called guarantees 0 10 20 30 40 50 60 70 80 90 100 and financial aid:  This is the largest and Amber Box Green Box Sources: MoF, MoAF, ISRA. fastest-growing category, accounting for more than 42 percent of amber box measures in 2010, in line with increasing government directed lending (discussed in Chapter 1) and growing expenditures on called guarantees and financial aid provided by both republican and local budgets. 16 A detailed list is provided in Appendix Table 3. 24 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery 2. Subsidies for purchasing agricultural inputs:  The subsidies cover inputs such as fertilizers, pesticides, seeds and fuels, including repayment of arrears for these inputs. The subsidy accounted for 28 percent of total amber box measures in 2010. Unlike other transition countries, the state in Belarus is involved directly in farm management by deciding on specific input uses, selecting and procuring the inputs, providing them at preferential prices, etc., thus greatly reducing farmers’ managerial autonomy. Payments to farms are based on centrally-set levels of input use, approximately half being distributed in proportion to land and the other half in proportion to output. 17 The respective shares of the three sources of funding (own farm revenues, state transfers/subsidies and state- guaranteed preferential credits) are also established at the central level and vary by input type.18 3. Subsidies for leasing agricultural machinery: This large subsidy accounted for almost 12 percent of total amber box spending in 2010. The state-led machinery leasing program has been operating for many years and has helped replace most of the agricultural machinery inventory in the country. 4. Product-specific measures: These include Figure 2-5. Input subsidies dominate amber box subsidies to support crops (such as grain measures and flax fiber) and animal production. in percent of GDP 4.0 Since 2008 the share of this subsidy fell 3.5 0.3 from more than 23 percent to 9 percent of 3.0 0.2 0.5 total subsidies. This is a result of the 2009 2.5 0.8 0.1 0.3 decision to decrease state procurement 2.0 1.5 1.2 prices for agricultural products and cut 1.5 1.2 0.4 some programs in the State Program for 1.0 0.2 0.3 0.5 1.0 the Revitalization and Development of Rural 0.7 0.8 0 Areas 2005–2010 (SRDP). Price support 2008 2009 2010 has been kept mainly to stimulate flax fiber Subsidies for mineral fertilizers, plant protection agents, fuel Subsidies for purchase of agricultural machinery and equipment production, as well as to encourage rural Interest rate subsidy, called guarantees, financial aid Product specific measures households to sell some of their products Other to downstream state-owned enterprises. Sources: IRSA and MoFA. Subsidies for defaulted payments on bank loans have been growing rapidly, a worrying sign of the deteriorating financial situation of the sector. Steep credit growth to farms, much of which is guaranteed under the GDL programs, has led to expanding subsidies as the Government steps in and makes repayments. This not only raises issues of moral hazard and incentives in allocation of credit within the sector and the economy as a whole, but also creates potentially large contingent liabilities. Fast growing subsidies for the compensation of bank losses indicate that preferential borrowing coupled with other instruments such as state-financed guarantees and debt write-offs are not only costly support measures, but also highly unsustainable. Declining profitability and growing agricultural enterprise debt, which are discussed in more detail below, are revealing indicators of this trend. 17 See Serova, 2009. 18 For example, in 2010 the Government projected that the cost of mineral fertilizers and pesticides would be paid 40 percent from the resources of agricultural enterprises, 30 percent from the local budgets and 30 percent from the RAPS Fund. Fuels supplied to farms in 2010 need to be paid for entirely from own enterprise resources, possibly by use of short-term preferential bank loans. Source: Decree of Council of Ministers 1699, December 26, 2009. Chapter 2. Agriculture: Hardening budget constraints 25 Bel arus Public Expenditure Review Input subsidies for fertilizers and pesticides as well as subsidies for the compensation of bank losses have driven up the cost of state agricultural support. The hefty growth of these subsidy categories is a reflection of a mix of factors, at the core of which lies the growing gap between the artificially maintained prices for agricultural inputs (fertilizers and pesticides, but also bank loans) and the market rates of inputs in the absence of state interference in agricultural markets. In addition to the “price gap” factor, unrestricted input subsidies tend to stimulate excessive use of subsidized inputs. For example, over the last five years mineral fertilizer application nearly doubled (from 113 kg per hectare in 2004 up to 205 kg per hectare in 2009), although crop yields have not responded proportionately. Agricultural machinery support is also a large expenditure item, and it should be evaluated to assess options regarding continuation, phase out or rapid termination. This category includes subsidies for leasing agricultural machinery, repayment of arrears on bank credits for the purchase of machinery and compensation of bank losses related to provision of preferential credit. The state-led machinery leasing program has been operating for many years and has helped replace most of the agricultural machinery inventory in the country. A thorough cost-benefit evaluation of the program would be timely, considering the fiscal and economic costs involved versus the machinery availability level in the sector. Programs to address a critical asset gap need a clear time span and should terminate as soon as they accomplished their intended purpose. Green box measures have been growing, but remain low and lack diversification Green box support has been growing over the past years, but remains low in Belarus (Figure 2-6). Green box measures are increasingly recognized as the most effective way to support sustainable agricultural growth.19 Green box measures provide lasting support to agriculture in a Figure 2-6. Green box measures have increased, way that does not distort trade and generate but still remain small conflicts with trade partners. Moreover, green in percent of GDP 0.4 box measures provide a vehicle for investments 0.35 0.07 in the know-how, technology, management 0.3 skills and infrastructure that Belarus agriculture 0.25 0.11 0.07 0.1 needs to be competitive in a global economy. 0.2 0.06 0.06 0.06 Green box support increased by over one 0.15 0.06 0.06 0.03 third in 2010 compared to previous years and 0.1 0.05 0.1 0.1 0.09 reached 12 percent in total budgetary support 0 to agriculture in 2010, from 7–8 percent in 2008 2009 2010 previous years. The increase was due to the Chernobyl effect mitigation Agricultural research, education and professional training addition of a new expenditure item—capital R&D, anti-flood and soil fertility monitoring Other investments for building drainage systems, Capital investments for construction of melioration systems Sources: IRSA and MoFA. 19 The worldwide trend in agricultural policy is towards increasing the use of green box measures, as the cost and inefficiencies associated with the distorting amber box measures become evident. In international perspective, the share of green box measures in AB+GB support increased from 50 percent in the mid-1990s to roughly 70 percent in 2003; for the EU, this share increased from under 30 percent to almost 50 percent. Source: 2009 World Bank Policy Note. 26 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery as well as the allocation of more funds to agricultural research and education and to budgetary organizations (for detailed breakdown of agricultural expenditures see Appendix 2.4). The composition of green box measures is not very diversified, and basic public services aimed at supporting competitiveness of the sector are limited. The main elements of support are land improvement measures aimed at liquidating the consequences of the Chernobyl event (22 percent in 2010), agricultural research, education and training (16 percent), other R&D and anti-hail measures (17 percent) and funding of budgetary organizations (18 percent) (Figure 2-6). Other categories included the compensation of crop losses in 2008 (15 percent) and capital investments for building drainage systems in 2010 (17 percent).20 It is increasingly recognized that the provision of public goods that fall within the green box category is the best way to help agriculture grow in a sustainable manner. Phasing out amber box measures could be slow because subsidies provide soft budget constraints to agricultural producers. However, Belarus’ experience shows that direct and indirect subsidies to agriculture do not necessarily lead to increased productivity and efficiency, on the contrary, the sector continues to shrink and the share of loss-making farms grows. This has also been the case in many EU countries where after decades of costly market price supports agriculture had remained largely inefficient and highly dependent on continuous support. Today, major agricultural exporting nations in the world (such as Australia, Canada, the EU and the USA) have already made an irreversible move towards green box measures, encouraged by the need to reduce budget spending on agriculture and the WTO process. Moreover, green box measures provide an ideal vehicle for bringing know-how, technology, management skills, and investment in infrastructure, including quality infrastructure. Belarus’ farmers could greatly benefit from these public services to boost their competitiveness in a global economy. Rising indebtedness of farms poses additional fiscal risks In addition to growing budget support, Figure 2-7. Debt of agricultural enterprises is mounting liabilities of agricultural ballooning enterprises—which presently account for trillion BYR, current prices in percent 35 300 over 250 percent of agricultural GDP—pose 258 30 250 a growing fiscal risk. Over the last five years, 221 25 the debt stock of agricultural enterprises has 162 161 19.1200 20 153 quintupled to close to BYR 32 trillion. Total 123 13.5 150 15 debt comprises two major elements: (i) debt to 8.8 100 10 5.6 suppliers; and (ii) credits and loans. The latter 4.2 12.7 50 5 2.3 7.7 9.9 grew by 400 percent over 2005–2010, largely 4.0 5.3 6.2 0 0 under government-directed lending programs 2005 2006 2007 2008 2009 2010 discussed in Chapter 1 (Figure 2-7). These soft Debts to suppliers Debts on credits and loans Debt stock to Ag GDP Source: Belstat. loans have cushioned the impact of depressed 20 See full list of green box support measures in Appendix Table 3. Chapter 2. Agriculture: Hardening budget constraints 27 Bel arus Public Expenditure Review farm revenues but will likely need to be supported by larger subsidies in the future. As discussed in the previous section, loss coverage for loan defaults is a growing expenditure item, and at current levels it is likely that substantial sector liabilities will not be sustainable and will eventually be shifted to the budget. The current practice of repaying farm debts with public funds has led to a situation in which poor performers have been continuously swallowing larger amounts of state support (needed to cover their increasing debts and arrears) than good performers. Subsidies predominantly benefit smaller farms 21 The distribution of subsidies across agricultural enterprises in Belarus is less skewed than the EU. Over the period 2003–2007, half of the commercial farms in Belarus received between 18 and 23 percent of the total budget support to the sector, and the remaining half received between 77 and 82 percent. While the distribution is skewed, it is much less so than the 20-80 ratio often cited in connection with the EU farm subsidies (i.e. 20 percent of the farms collect roughly 80 percent of the subsidies), or the ‘7-75’ ratio measured for budget support to livestock producers in Ukraine.22 Figure 2-8. Farm subsidies disproportionately Farms with lower revenue are the prevailing benefit small farms beneficiaries of preferential treatment in the subsidies, cumulative, % subsidy allocation. Looking at the cumulative 100 distribution of subsidies per cumulative farm 80 revenue, the smallest farms, generating 60 20 percent of the revenue, together receive about 35 percent of all subsidies (Figure 40 2-8). At the other end of the distribution, the 20 28 largest farms—or 2 percent of farms that generate 20 percent of revenue—receive only 0 0 10 20 30 40 50 60 70 80 90 100 about 7 percent of all subsidies. revenues, cumulative, % Source: World Bank (2009b). D. Impact of fiscal support on sector performance Despite extensive fiscal support for agriculture, sector performance is poor and deteriorating. Stagnating yields, agricultural labor productivity and fixed capital productivity significantly below the rest of the economy, falling enterprise profitability and ballooning debts are all indicators that the existing agricultural support system is not producing the intended impact. 21 Since an update of the Belarus Farm Database could not be obtained, this section is based on the analysis provided in the 2009 World Bank Policy Note covering the period up to 2007. Provided the Government objectives and instruments of agricultural state support have not changed since then, there is every reason to believe that the situation depicted for the period 2003-2007 has remained much the same in 2008-2010. 22 World Bank (2009). 28 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery Agricultural productivity has stagnated. Table 2-3. Cumulative yield growth, percent Yields, a partial measure of productivity, have Product 2000–04 2005–09 marked considerable increases over the last Grains 47 16 decade, but yield growth slowed down after Potatoes 42 -2 2005. Ten-year data series show that, yield Vegetables 44 18 growth was significantly stronger in the first Sugar beet 34 24 half of the last decade than in the second for all Flax fiber 50 17 major crop categories, suggesting stagnating productivity and diminishing marginal returns Milk 33 24 Source: Belstat. (Table 2-3). Belarus yields are lower than yields of some countries in the region, despite higher fertilizer application. A cross-country comparison shows that fertilizer use in Belarus is five to six times higher than in neighboring Ukraine and Romania, but also higher than in Poland, Czech Republic and even Germany (Table 2-4). In terms of yield performance, however, Belarus is far behind Germany for all crops, but also behind Poland and Czech Republic for most crops. This is a particularly negative outcome in the light of increasing input subsidies, especially on fertilizers. Table 2-4. Cross country comparison: yields vs. fertilizer use Fertilizer use Yields (kg/ha), 2007 (tons per ha), average 2005–09 Cereals Potatoes Vegetables Sugar beet Flax fiber Belarus 218 3,045 19,760 25,059 39,368 648 Ukraine 33 2,750 13,421 15,388 29,964 513 Romania 45 2,830 14,068 13,946 31,683 2,500 Czech Rep 163 4,774 25,961 16,329 54,642 3,064 Germany 194 6,742 41,742 31,484 62,041 - Poland 213 3,155 18,749 25,566 47,483 261 Sources: WB/DDP time series for fertilizer use data and FAO/FAOSTAT-agriculture for yield data. Both labor productivity and fixed capital productivity in agriculture have remained below the rest of the economy, suggesting low efficiency of public spending and misallocation of resources within the economy. In 2009, the capital productivity in agriculture was only a half of the capital productivity in the rest of the economy, while in 2000 the gap was smaller, at 35 percent (Figure 2-9). Capital productivity in the rest of the economy has grown faster than in agriculture, despite the fact that fixed capital investments in agriculture grew much faster than in the rest of the economy.23 Despite the massive allocation of capital into agriculture, labor productivity has continually lagged behind the rest of the economy by about 20 percent. This has also led to considerable increases in the capital/ labor ratio in agriculture (from 23 to 33 million BYR per worker), while the same ratio in the rest of the economy remained pretty much unchanged. Presently the capital-labor ratio in agriculture is about a 23 The share of fixed capital investments in agriculture in total FCI increased from 6.8 percent in 2000 to 18.2 percent in 2009. Chapter 2. Agriculture: Hardening budget constraints 29 Bel arus Public Expenditure Review third higher than in the rest of the economy (Appendix 5). In addition, the rates of return on capital have rapidly declined in the last decade, in contrast to the rising capital inflow to the sector (Figure 2-10). The fact that capital invested in agriculture is only half as productive as in other sectors suggests a low efficiency of public spending, and also a misallocation of resources within the economy—the same amount of resources invested in other sectors would have generated significantly higher returns. Figure 2-9. Agricultural labor and capital Figure 2-10. Returns on capital in agriculture productivity is considerably lower than the rest of have declined while credit flows have risen the economy in percent in percent in percent 90 25 100 80 20 80 70 15 60 60 10 40 50 5 20 40 0 0 2000 2004 2005 2006 2007 2008 2009 2001 2002 2003 2004 2005 2006 2007 2008 2009 Labor productivity Capital productivity RORK Credit/VA, RHS Source: Belstat. Source: Belstat. Profitability, soft budget constraints and farm restructuring In the aggregate, profitability of farms has halved from 8 to 4 percent between 2005 and 2009 due to farm costs rising faster than revenues. Farm costs have risen by 55 percent and currently exceed agricultural GDP by as much as 70 percent. Fiscal support to agriculture—accounting for approximately 50 to 60 percent of total farm revenues for 2005–2009—increased in volume by approximately 40 percent. Revenues are limited by the farm gate price regulation policy on outputs, preventing producers from benefiting from rising food market prices. As a result, net profits have remained stagnant in real terms and decreased as proportion of agricultural GDP from 15 percent in 2005 to 12 percent in 2009 (Figure 2-11). These two factors, (i) restrained prices of agricultural output that hold back farm revenues and (ii) growth of input prices compensated through increasing subsidies and loans, are the key drivers eroding the sustainability of the agricultural support system in Belarus. Soft budget constraints have delayed restructuring and are at the core of poor and declining farm performance. Almost half of Belarus farms operate at a loss. Under the current practice of subsidizing farms and repaying debts of loss-making enterprises with public resources, farms have little incentive to show good performance. Continuous state support and soft loans have avoided the needed restructuring of non-viable enterprises in the sector. In spite of mergers of loss- with profit-making farms, the number of poorly performing farms is overwhelming.24 These government-led mergers have pushed downward the average share of loss-making enterprises over the last decade—from 42 percent in 2000 down 24 Loss-making enterprises are those that have negative net profits, which in turn are made up of four elements: profits from sold products and services, profits from operational revenues and costs, profits from non-sale revenues and costs, and taxes and fees from profits. 30 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery Figure 2-11. With farm production costs on the rise, revenues are stagnating trillion BYR, current prices percent of ag GDP 20 17.9 180 169 157 18 160 152 152 150 15.3 16 140 14 120 11.4 12 100 9.5 10 7.8 80 69 68 8 5.9 6.8 64 57 64 6 5.0 60 4.3 4 3.3 40 1.8 15 18 2 0.8 0.9 1.0 1.2 20 15 14 12 0 0 2005 2006 2007 2008 2009 2005 2006 2007 2008 2009 Total costs of ag enterprises Total fiscal support to ag Net profits of ag enterprises Total costs of ag enterprises Total fiscal support to ag Net profits of ag enterprises Sources: Belstat; ISRA for data on total fiscal support to agriculture. Figure 2-12. Over a half of the farms are not to 16 percent in 2004 and to a negligible profitable 2 percent in 2009—masking structural issues 2010 51.9 30.5 8.3 6.7 1.4 1.2 in the sector.25 The number of unprofitable 2009 49.8 32.5 7.7 7.9 1.1 1.0 enterprises climbed from 33 percent of the 2008 32.8 28.6 16.0 14.8 5.1 2.7 total in 2004 to 52 percent in 2010 (Figure 2007 51.6 24.8 11.5 8.5 2.2 1.4 2-12). The number of enterprises with low 2006 44.8 26.3 11.1 11.9 3.6 2.3 profitability (from 0 to 5 percent) also increased 2005 35.6 28.4 12.8 13.9 5.8 3.5 during this period, thus pushing up the number 2004 33.0 28.3 15.4 13.3 6.1 3.9 of agricultural enterprises with negative or low 0 10 20 30 40 50 60 70 80 90 100 profitability from 61 to 82 percent in a six-year percent of total number of agricultural enterprises period (2004–2010). In 2010, only 1.2 percent Loss-making Profitability 0–5% Profitability 6–10% Profitability 11–20% Profitability 21–30% Profitability over 30% of agricultural enterprises had profitability Source: ISRA. levels over 30 percent. This impedes the agricultural sector from finding and strengthening its real competitive advantages, reduces enterprise efficiency, inhibits agricultural incomes and ultimately hampers growth of the sector. E. Implications of international trade integration on the restructuring on the agricultural support system International trade integration will require a substantial reduction in amber box support measures. Under the recently adopted (January 2011) Customs Union (CU) agreement on unified rules for agricultural state support, Belarus will have to keep its amber box support under strict control and gradually reduce input and production subsidies. The agreement binds the type and level of support that the three members countries—Belarus, Kazakhstan and Russia—are allowed to provide to their agricultural sectors. 25 The total number of large farms shrunk significantly over the last decade: 2,626 in 2000; 2,311 in 2004; and 1,745 in 2009. Chapter 2. Agriculture: Hardening budget constraints 31 Bel arus Public Expenditure Review The CU agreement provides specific targets for reducing trade-distorting support measures. The following rules apply to the three types of support measures (Box 2-2): (i) no limits on non-trade- distorting support measures; (ii) highly trade-distorting measures are not allowed; and (iii) trade- distorting measures shall not exceed 10 percent of gross agricultural output (GAO). While Belarus does not presently use highly trade-distorting export subsidies, the total amount of category trade-distorting subsidies—i.e. amber box support—accounted for approximately 21 percent of GAO in 2005–09, and, based on preliminary estimates, 17–18 percent in 2010. A six-year transition period has been set for Belarus that allows for gradual reduction of trade distorting measures by 1 percentage point annually to 10 percent in 2016. Box 2-2. Classification of agricultural support measures in CU •• Non-trade-distorting measures  are those that generally fall under the green box and include the bulk of general agricultural services and direct payments to producers that are de-coupled from production volumes (to include structural adjustment assistance). •• Highly-trade-distorting measures  are considered export subsidies for agricultural inputs and outputs. •• Trade-distorting measures  are those that do not fall under either of the above two categories. Source: CU agreement on unified rules for state support to agriculture, adopted January 12, 2011. The subsidy reduction may have to accelerate if one of the members of the CU decides to move more swiftly to WTO accession. The CU agreement stipulates that in case any of the parties join the WTO, the subsidy ceilings will be determined based on WTO commitments. Russia and Kazakhstan have smaller measures of support to the agriculture sector as percent of GDP compared to Belarus. Moreover, Russia and Kazakhstan export structures depend less on agriculture. Thus, these countries may have a faster and less painful accession to WTO in terms of reducing input and production subsidies to meet the required level of support. In this case, Belarus would have to cut subsidies more drastically, potentially moving towards WTO requirements in three rather than seven years (left panel, Figure 2-13). An accelerated reduction of subsidies would significantly benefit the fiscal consolidation process while creating a more conducive environment for agricultural privatization and modernization. If subsidy reductions are accelerated to achieve WTO targets by 2014, this would lead to significant fiscal savings, in the range of half a percent of GDP (right panel, Figure 2-13). While this would require adjustment, the agriculture sector and its processing industries are less connected to the core manufacturing sectors of the economy. This implies that hardening budget constraints more swiftly in the sector would not cause drastic decreases in output and employment. For the same reasons, agriculture and food processing industries are fertile terrain for privatization and FDI, which in turn would support a faster reduction of subsidies whilst bringing new technology and know-how to the sector. 32 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery Figure 2-13. A WTO accession of another CU  …which would generate fiscal savings party may induce an accelerated reduction in subsidies… percent of gross agricultural output percent of GDP 17 1.8 16 1.6 1.4 0.58 0.49 15 1.2 0.51 14 1.0 13 0.8 0.34 0.6 1.1 1.2 12 0.4 0.8 11 0.2 0.5 10 0 2010 2011 2012 2013 2014 2015 2016 2017 2011 2012 2013 2014 CU agreement Accelerated path Additional fiscal savings Baseline fiscal savings Source: World Bank staff estimates. Note: Fiscal savings are estimated over the baseline spending of 17 percent of GAO. F. Policy options and fiscal and social impact The high fiscal cost coupled with low effectiveness and low sustainability of the current support system in agriculture underscore the need to reorient the existing policy framework. The need for reform is reinforced by the CU agreement and Belarus’ potential WTO accession scenario in the medium term and in view of Russia’s potential WTO accession in the short term. Reforms would aim at ensuring sustainable long-term agricultural growth through stronger emphasis on productivity and market competitiveness and a smaller but more transparent and efficient support envelope for the sector. The core of a reform agenda would require changes in both regulatory and fiscal policies. A three-pronged approach is suggested, with the following inter-related measures: 1. Reduce state controls and regulations. Fiscal reforms in the agricultural support system cannot be undertaken in isolation, but are inextricably linked to structural reforms in the sector. Structural reforms would encapsulate a gradual withdrawal of state controls along the agricultural value chain, including phasing out production and employment targets, wage and price controls (of both outputs and inputs), increasing managerial autonomy and a possible privatization of farm assets. •• Liberalize agricultural prices for both inputs and outputs. Reducing state control and regulation of agriculture, accompanied by subsequent phasing out of subsidies for inputs and outputs, would diminish distortions to agricultural incentives and improve the responsiveness of agricultural enterprises to market signals. •• Withdraw the state from direct farm management. To take full advantage of price liberalization and market signals, it is crucial that farm managers are allowed and encouraged to make independent decisions regarding their operations, in terms of types and quantities of crops/ animals to be grown, input volumes to be applied, technology to be used, etc. Chapter 2. Agriculture: Hardening budget constraints 33 Bel arus Public Expenditure Review •• Privatization. While more market-oriented management of farms can be introduced independent of the specific ownership model, privatization might eventually become desirable, taking into account the significant value of the agricultural assets that have been built up. 2. Increase and diversify non-distorting support measures. There is considerable scope for increasing and diversifying Belarus’ green box spending for agriculture by reorienting a part of the large amber box resource envelope. Such measures would strengthen the long-term competitiveness of Belarus agriculture through developing rural extension services and upgrading food safety systems. •• Consider introducing immediate adjustment assistance to counter-balance the short-term negative effects of price liberalization and cuts in amber box subsidies. Compensatory adjustment assistance could come in the form of de-coupled payments aimed at reducing production-related distortions. As long as these payments do not distort trade, they are not subject to limitations, as specified by the new CU agreement. However, since they have a temporary role—to offset the short–term adverse effects of reforms on agricultural producers— it is important to ensure that they are gradually phased out over time. •• Strengthen agricultural extension services. Producing agricultural and food products for both the domestic market, which involves competing with imported products, as well as for highly demanding export markets (including the increasingly sophisticated Russian market) requires a whole set of new knowledge and skills. Advisory services could assist with this. •• Modernize Belarus’ food safety system. A reform of the quality, food safety and animal and plant health management systems will be necessary to maintain access to profitable market segments in CIS countries as well as increase access to new export markets, especially in the EU. WTO accession also implies a transition from the current food safety and quality system to WTO SPS-based systems, which is likely to become the dominant standard for the CIS markets once Russia’s WTO accession is completed. Anticipating this shift, Belarus needs to develop a strategy and action plan based on its capacities, geography, product mix, market opportunities and health risks. As a small country with a simple export package, Belarus should selectively adopt international standards for products with export potential. Agricultural advisory services would have an important role to play in this transition process by raising awareness and educating producers, and perhaps also food handlers and consumers. 3. Reduce distorting amber box subsidies. These actions should be closely linked and informed by the results of the reform processes undertaken under the liberalization agenda. •• Reduce interest rate subsidies provided to agricultural enterprises and tighten credit access to farms. Budgetary compensations of bank losses have registered the most spectacular growth, from 77 billion BYR in 2005 up to 1,018 billion BYR in 2010, thus becoming one of the most costly support measures. •• Phasing out agricultural machinery support is another potential area of cost saving. Programs to fill a critical asset gap need to have a transparent time span and terminate as soon as 34 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery they accomplish their intended purpose. Therefore, a thorough cost-benefit evaluation of the machinery leasing program would be timely, considering the fiscal costs involved and the existing stage of machinery availability in the sector. •• Replace permanent input subsidies with smart subsidies. The system of providing permanent subsidies on key farm inputs, such as fertilizers and seeds, distorts the incentives of producers and may lead to inefficient production technologies. It should be phased out in step with the transition to a more market-based sector framework. The focus could shift to “market smart” subsidies on products that exhibit comparative advantages so that once public support is withdrawn enterprises remain viable and competitive. The following measures may be considered: (i) subsidizing—over a limited and explicitly time period—the adoption of new agricultural technologies (embedded in seeds, seedlings, animal breeds); and/or (ii) subsidizing capital investments in market infrastructure needed for building efficient and competitive value chains. The latter could include the fresh horticulture value chain that requires a wide range of modern facilities (field coolers, cold storages, packing houses, refrigerated transportation, etc.) or other produce-specific value chains that have proven production-related comparative advantages. The fiscal impact from implementing the proposed reform measures is expected to be positive, resulting in a smaller but more transparent and efficient support envelope for agriculture. In the medium to long term, the rationalization of agricultural support would result in a smaller but more sustainable overall support envelope with better targeted green box support measures supplemented by downsized and more efficient amber box measures (Table 2-5). In addition, accelerating the reduction of amber box measures to achieve WTO levels would generate savings in the range of half a percent of GDP annually over the coming seven years. The provision of de-coupled adjustment assistance to mitigate the impact of reduced subsidies on agricultural enterprises would be smaller than the size of subsidy reductions, and therefore the net fiscal impact is expected to be positive. The proposed reforms are likely to cause negative social impact initially, which is expected to turn positive over time. Price liberalization coupled with reduction/removal of a number of subsidies and hardening budget constraints for enterprises will likely have a negative social impact in the short run. However, the effect could be mitigated through immediate adjustment assistance for producers and new public services in agriculture that will smooth the transfer to market-based agriculture. Over time, as enterprises learn to operate under market conditions and reduced state controls and regulations, the expected medium- to long-term social impact is expected to turn positive. The positive social impact is expected to come from more independent decision making at the farm level concomitant with the state support through both “public good” services (as part of green box support) and market-smart subsidies (part of amber box support). At the same time, the social impact of possible restructurings of under-performing enterprises could be mitigated though both compensations (de-coupled payments, retirement packages) and specialized training. Chapter 2. Agriculture: Hardening budget constraints 35 Bel arus Public Expenditure Review Table 2-5. Fiscal and social impact of proposed reform measures Proposed reform measures Fiscal impact Social impact Reduction of state controls and regulations in agriculture Rising food prices could lead to Liberalize agricultural prices (inputs and Neutral consumption shocks to poorer outputs) population Higher degree of farm management autonomy Neutral Neutral Privatization and restructuring could Privatization Positive privatization proceeds lead to job losses in some farms Increase and diversify green box measures Develop agricultural extension and advisory Moderate cost Neutral services Modernize the quality, food safety and animal Neutral/moderate Cost Neutral/positive and plant health management systems Structural adjustment assistance (decoupled payments, producer retirement programs, Adjustment assistance would mitigate regional assistance programs, re-training Transition cost negative impact on rural employment programs) to offset the unfavorable impact of induced by farm restructuring reforms (1) and (3) Accelerated reduction of amber box subsidies Year 1: 0.38 percent of GDP Year 2: 0.57 percent of GDP Neutral Year 3: 0.51 percent of GDP 36 Chapter 2. Agriculture: Hardening budget constraints Fiscal Reforms for a Sustainable Economic Recovery Chapter 3. Energy services: Achieving financial viability B elarus’ energy sector is characterized by a high degree of state ownership, strong vertical integration, administered tariffs, growing operating losses and cross-subsidization. While low tariffs have kept household expenditures on utility services in Belarus below 10 percent of household income—much lower than in most countries in the region—they have undermined the financial viability and sustainability of the sector, especially since energy costs are on the rise with the phasing out of under-priced gas imports from Russia. Growing financial losses incurred by utility companies are compensated through a complex, non-transparent system of direct subsidies and cross-subsidies between non-residential and residential consumers as well as between energy services. Taken together, the sector drains fiscal resources to the tune of about 2 percent of GDP annually. While imposing a rising fiscal cost, energy subsidies are untargeted and largely inequitable, disproportionately benefitting the higher income population that has better access to and higher per capita consumption of energy. In addition, persistent under- pricing has deprived energy suppliers of the financing needed to carry out routine maintenance and investments in capital asset replacement, reducing efficiency and reliability of energy supply. Low energy prices also distort price signals to households, leading to inefficient energy consumption patterns, while cross-subsidies between non-residential and residential consumers impose an implicit tax on the industrial and business sector, adding to an already high tax burden. Responding to growing fiscal pressures, the Government has adopted an ambitious plan to gradually increase residential tariffs for natural gas, heating and electricity to full cost recovery levels by 2013 and 2014. The chapter emphasizes the need for tariff reforms to improve commercial viability of energy services provision and to reduce the fiscal burden resulting from under-pricing. However, such measures should be accompanied by scaling up targeted safety nets to offset adverse impact on poor households as well as financial and technical assistance for end- use efficiency to soften the social impact (discussed in Chapter 4). 37 Bel arus Public Expenditure Review A. Energy market structure Generation, transmission and distribution of energy services in Belarus are dominated by state-owned enterprises. The heat and electricity markets comprise two main participants: Belenergo and ZhKH. Both are state-owned, vertically integrated monopolies, catering to residential, industrial and other consumers. Belenergo’s main business is electricity and heat generation, transmission and distribution, as well as electricity and heat retail mainly in big cities such as Minsk and oblast capitals. ZhKH’s main business is heat supply, gas distribution, running water supply, sewage and water treatment. ZhKH is involved in heat production, transmission, distribution and retail through its subsidiaries located in big cities and smaller towns (where Belenergo is not present). Independent producers account for less Figure 3-1. Institutional structure of the energy sector in Belarus Ministry of Oblast Economy executive committee Direct subsidy from local budgets Cross-subsidy from electricity to heat sales Industry Industry Ministry of Ministry of Energy Other Energy Other ZhKH Consumers Consumers BelTopGas Belenergo Households Households Energy Council of efficiency Ministers department Setting of energy tariffs Heat sales, Gcal Eelctricity sales, kWh Gas sales, TCM Direct or cross-subsidy, US$ Approval of heat losses Sources: World Bank, MoEconomy, MoEnergy, MoZhKH, and Energy Efficiency Department. 38 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery than 1 percent of the energy market, indicating that private sector participation is low by international standard and below its potential in Belarus. The natural gas market comprises two main players: Beltransgas, which operates transmission and storage facilities, and Beltopgas, which operates gas distribution and retail sales. The energy market continues to be heavily regulated, but there is no independent energy regulator in Belarus. Currently, energy regulation, including tariff-setting,26 is performed by the Ministry of Economy for Belenergo and BelTopGas27 (except for energy sales to households) and oblast executive committees for ZhKH (except for energy sales to households) (Figure 3-1). Tariffs for household energy sales (heat, electricity and natural gas) are set by the Council of Ministers. Annual household income growth is the main determinant of tariffs, and as such energy tariffs for households cannot exceed the relative growth of household income. In addition to the Council of Ministers, Ministry of Economy and oblast executive Committees, the Energy Efficiency Department indirectly participates in tariff-setting by approving norms for technical heat losses. B. The fiscal cost of energy services The energy market is profoundly affected by rising prices for energy imports from Russia. The energy mix is heavily reliant on natural gas. Belarus generates about 97 percent of its electricity from natural gas- based thermal power plants.28 Natural gas is primarily imported from Russia, close to 18–19 thousand cubic meters (TCM), annually. The import price continues to be below the European market prices, but Russia’s drive to introduce market-based prices for its energy exports to CIS countries have gradually closed the price gap (Figure 3-2).29 The nominal border price of natural gas more than tripled over the past five years to about US$ 240 per TCM in the first quarter of 2011. The implicit subsidy provided to the Belarusian economy through these underpriced energy supplies30 has fallen from an estimated 9 percent of GDP in 2005 to 4.6 percent of GDP in 2009, with significant impacts throughout the economy. Despite nominal increases, energy tariffs for residential consumers have remain flat in real terms since 2005, lagging behind rising production costs. Energy production costs have increased substantially, driven by increases in the border price of natural gas, the most important energy input. While nominal tariff increases have taken place each year, in real terms residential tariffs have remained relatively 26 A cost-plus methodology is used for tariff-setting. The method uses direct costs, indirect costs and fixed costs related to the production and sale of energy services, and converts them into per unit costs, in principle applying a profit margin. However, in full economic recovery terms, this margin has been negative under the current practice of under-pricing energy services. 27 Belenergo computes energy “tariff level declarations”, which have to be approved by Regional Executive Committees before they are submitted for Ministry of Economy’s approval. 28 International Energy Agency. 29 According to a contract signed between Belarus and Russia in December 2006, Belarus is contracted to pay the European market price for Russian natural gas by 2011, although no export duty is applied to Russian gas sold to Belarus. Since then gas prices were negotiated annually between Russia and Belarus. The latest 2011 first quarter gas price was agreed at about US$240 per TCM. 30 Economic border price for natural gas for Belarus was estimated based on the average European natural gas border price, net-backed to Belarus using an economic (full cost recovery) price for natural gas transmission. Consequently, an implicit subsidy was estimated as a product of the difference between the economic and actual natural gas border prices multiplied by the volume of natural gas consumed by Belarus. Chapter 3. Energy services: achieving financial viabilit y 39 Bel arus Public Expenditure Review Figure 3-2. Belarus imports natural gas below  …but the price gap is closing the European price… price of natural gas 2008, US$/TCM US$/TCM current prices in percent 500 450 90 450 400 83.6 80 78 400 350 70 70 350 300 65.9 60 300 250 50.9 50 250 200 200 40 150 150 33.8 30 100 100 20 50 50 10 0 0 0 BLR ROM UKR MDA LVA DEU SVK LTU CZE HUN AUT 2005 2006 2007 2008 2009 2010 Economic price of natural gas Belarus border price Imputed subsidation rate (rhs) Sources: International Energy Agency, World Bank staff estimates. flat and not kept pace with rising costs. Tariffs for non-residential consumers, on the other hand, have gradually increased roughly in line with production costs between 2005 and 2009, and generally cover operating costs (with the exception of heat services provided by Belenergo). Non-residential energy tariffs for heat, electricity and gas are above cost recovery levels. Except for non-residential heat sales by Belenergo, tariffs for most non-residential consumers are above cost Figure 3-3. Residential tariff increases have lagged behind rising production costs ZhKH Heat Belenergo Heat constant US$ 2009 per GCal constant US$ 2009 per GCal 80 35 70 30 60 25 50 20 40 15 30 20 10 10 5 0 0 2005 2006 2007 2008 2009 2005 2006 2007 2008 2009 Average-weighted Residential Non-residential Heat production cost Electricity Natural Gas constant US$ 2009 per c/kWh constant US$ 2009 per TCM 12 200 180 10 160 8 140 120 6 100 80 4 60 2 40 20 0 0 2005 2006 2007 2008 2009 2005 2006 2007 2008 2009 Average-weighted Residential Non-residential Production cost Sources: Ministry of Energy, Ministry of ZhKH. 40 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery recovery level, and non-residential tariff increases for electricity, gas and have reflected rising production costs (Figure 3-3). Box 3-1. Billing residential consumers for energy services Heat. There is a one-part residential tariff for heat in Belarus (e.g. BYR/Gcal). Almost all households have a building-level heat meter installed and are billed according to the amount of heat consumed. In the case of multi-apartment buildings, the total heat consumed by the building is then allocated among apartments based on apartment size. Due to the district heating infrastructure configuration in Belarus, most of heat consumers can not individually regulate their heat consumption. Heat regulation is done on the supply side, taking into account ambient temperature. Electricity. There is a one-part residential tariff for electricity in Belarus (e.g. BYR/kWh). Almost all households have individual electric power meters installed. Billing is based on the amount of electricity consumed. The reluctance to pass through increases in production costs to residential consumers led to a deterioration of cost recovery levels for residential heat, gas and electricity tariffs (Figure 3-4): •• Heat:  Residential district heat has the Figure 3-4. Cost recovery levels of residential lowest cost recovery rate. Residential heat services have declined sharply tariffs have not increased in real terms percent 140 since 2005 despite heat supply costs 120 109 more than doubling in real terms for ZhKH 100 93 and increasing 1.6-fold for Belenergo. 80 75 Consequently cost recovery rates for district 60 60 heat declined from about 75 percent (2005) 45 45 40 37 to about 36.8 percent (2010) for Belenergo 21 20 and from about 45 percent (2005) to 0 about 21 percent (2010) for ZhKH. While 2005 2006 2007 2008 2009 2010 there is no difference in residential heat Heat ZhKH Heat Belenergo Electricity Natural Gas Sources: Ministry of Energy, Ministry of ZhKH. tariffs, Belenergo’s production costs are roughly half of ZhKH’s. This substantial difference is explained in part by economies of scale and Belenergo’s use of efficient combined heat and power plants. In addition, the greater level of wear of ZhKH’s major capital assets results in reduced heat generation efficiency and higher losses in heat transmission and distribution network (17–20 percent vs. 10–12 percent in Belenergo). Chapter 3. Energy services: achieving financial viabilit y 41 Bel arus Public Expenditure Review •• Electricity: While electricity sales to non-residential consumers are profitable, residential services operate with a growing loss.31 The supply costs of electricity increased 1.7-fold in real terms between 2005 and 2009. While non-residential electricity tariffs have increased in step with rising production cost, increases in residential tariffs have lagged, increasing only about 1.2-fold. Cost recovery rates declined from 93 percent in 2005 to 60 percent in 2010. •• Gas:  Cost recovery rates for gas supplies to households, which generated positive operating margins in 2005, rapidly declined to 45.3 percent in 2010. The supply cost of natural gas more than tripled in last five years. Non-residential tariffs kept pace with this steep rise, but residential tariffs increased at slower pace, leading to a marked decline in cost recovery. Under-priced energy supplies to residential Figure 3-5. The fiscal cost of underpriced energy consumers impose a growing fiscal cost, has grown rapidly equal to about 1.7 percent of GDP in 2009. percent of GDP 2.0 Due to declining cost recovery, cumulative 0.28 operating losses of utility companies have 1.5 0.27 0.34 more than doubled in real terms, rising from 0.10 0.20 0.28 0.16 0.06 0.76 percent of GDP in 2005 to 1.7 percent 1.0 0.18 0.09 0.12 0.09 0.12 of GDP in 2009, nearing US$0.84 billion in 0.08 0.11 0.06 0.05 1.10 2009 at the end-2010 exchange rate (Figure 0.5 0.87 0.84 0.57 0.63 3-5). Residential heat services are the largest 0 source of losses, accounting for 1.1 percent 2005 2006 2007 2008 2009 of GDP (2009), while non-residential heat adds operating losses by service line Residential heat Non-residential heat Residential and agricultural electricity another 0.06 percent of GDP (2009). ZhKH Residential natural gas Preferencial energy tariffs accounts for losses equal to 0.47 percent of Sources: MoF, Belenergo, ZhKH, and World Bank staff estimates. GDP on residential heat sales, and Belenergo accounts for losses equal to 0.67 percent of GDP. In 2009, Belenergo’s operating losses on electricity sales to residential consumers account for 0.3 percent of GDP. Beltopgas incurs losses equal to 0.27 percent of GDP (2009) on its residential gas sales. In addition, some privileged industrial consumers receive subsidies equal about 0.3 percent of GDP (2009) in the form of energy sales at preferential tariffs. Some 130 industrial consumers benefit from preferential electricity, heat and gas tariffs below cost recovery granted by Ministry of Economy resolutions on an ad hoc basis. In addition, 106 non-residential consumers were granted deferment of arrears for energy services in 2009, equal to 0.01 percent of GDP to the total cost of subsidized energy. 31 The financial cost recovery level is calculated by Belenergo based on the “alternative way of electricity supply” cost allocation method (in Belarus this is also referred to as “economic” method), which assumes that the same amount of electricity generated by a combined heat-and-power (CHP) plant would be generated by a condensing power plant (which has lower total efficiency compared to CHPs). Therefore, under this cost allocation method the benefits of combined heat-and-electricity generation are attributed to heat. Thus, compared to the energy (or “physical”) cost allocation method, the share of operating expenditures (OPEX) and capital expenditures (CAPEX) attributed to heat is lower under the “economic” method. The “economic” method results in cross-subsidization from electricity to heat sales. Total cost of heat production includes fuel costs, transportation expenses, operation and maintenance expenses, salaries, amortization (depreciation) and payments to the Innovation Fund. 42 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery The cost of under-priced energy provision Figure 3-6. The cost of under-priced energy is equals 2 percent of GDP, financed by direct financed through direct budgetary subsidies and budget subsidies and—increasingly—through cross-subsidies a complex system of cross-subsidies from percent of GDP 2.0 profits of non-residential energy sales. Explicit 0.28 budgetary subsidies from local government 1.5 0.27 0.34 budgets to ZhKH only cover a small fraction, 0.10 0.20 0.30 0.16 0.09 0.3 percent of GDP or about 15 percent of the 1.0 0.25 0.27 0.08 0.03 0.12 0.03 total public financing provided to the sector. 0.20 0.18 0.12 0.08 0.04 0.05 While budgetary subsidies have increased 0.5 0.08 0.10 0.96 0.74 0.67 0.44 0.41 about 60 percent in real terms to cover growing 0 operating losses, the lion’s share of the cost 2005 2006 2007 2008 2009 was shifted to cross-subsidies between financing sources by operating losses CS from non-residential electricity sales (Belenergo) different services lines and consumers of utility CS from non-residential heat sales (ZhKH) companies.32 Cross-subsidies have almost CS from non-heat sales (ZhKH) Direct subsidy from local government (to ZhKH) tripled in real terms, rising from 0.6 percent CS from non-residential gas sales CS from non-residential consumers of gas, electricity, heat of GDP in 2005 to 1.7 percent of GDP in 2009 Sources: MoF, Belenergo, ZhKH, and World Bank staff estimates. (Figure 3-6). Note: CS=cross-subsidy Direct budgetary subsidies to ZhKH account for only 0.3 percent of GDP, a very small portion of the total losses on residential heat sales. ZhKH receives direct budgetary subsidies from local governments covering operating losses on a range of communal services, including heating, warm water, garbage collection and housing maintenance. Subsidies to compensate for the financial net loss incurred in the heat supply segment from ZhKH to households reached 0.3 percent of GDP during 2009, after gradual increase over the past five years. Despite recent increases, the subsidies provided to ZhKH have only been sufficient to compensate for about three-quarters of the financial losses incurred in heat sales to households, with the remainder financed through cross-subsidies from non-residential consumers of heat and other communal services. Cross subsidies provide more than 85 percent of the total subsidies to the sector. •• Cross-subsidies between different business lines and customer segments within ZhKH amount to about 0.2 percent of GDP. ZhKH resorts to cross-subsidies from non-residential consumers of heat as well as from non-heat-related services to cover losses on residential heat sales after the local government budget subsidy is taken into account. In 2009 non-residential consumers of heat supplied by ZhKH provided about US$40 million of cross-subsidies to residential consumers of heat, while an additional US$45 million of cross-subsidies came from ZhKH’s other service lines.33 •• Cross-subsidies from Belenergo’s non-residential electricity sales have grown rapidly, amounting to about 1 percent of GDP in 2009. Unlike ZhKH, Belenergo does not receive any budgetary 32 Cross-subsidies are transfers (made outside the government's budget) from a profitable to loss-making segment of the energy enterprise in question. 33 Based on the data received from ZhKH, it was estimated that after collecting heat-related revenues as well as receiving direct budget subsidies from local governments, there still was a financial net loss on heat sales. This financial net loss was assumed to be covered from non-heat-related revenues received by ZhKH. Chapter 3. Energy services: achieving financial viabilit y 43 Bel arus Public Expenditure Review subsidies and fully relies on cross-subsidies to cover its operating losses. Non-residential (mainly industrial) consumers of electricity cross-subsidize residential consumers, with non-residential tariffs set higher than the financial cost recovery level.34 Cross-subsidies from non-residential consumers more than doubled in real terms in five years, growing from 0.4 percent of GDP in 2005 to about 1 percent of GDP (or US$471 million) in 2009. •• Cross-subsidies from Beltopgas non-residential gas sales total 0.3 percent of GDP. Beltopgas covers its operating losses on residential gas sales from revenues on non-residential gas sales. While there were no cross-subsidies in 2005, they have grown rapidly from 0.1 percent of GDP in 2007 to 0.3 percent of GDP in 2009. •• Cross-subsidies from gas, electricity and heat sales equal 0.29 percent of GDP to cover losses on sales at preferential tariffs for privileged non-residential consumers. Beltopgas, Belenergo and ZhKH cover operating losses on the sales of heat, electricity and gas to privileged non-residential client through profits in other service lines. C. Impact of energy subsidies Extensive energy subsidies have a number of adverse effects. Besides creating large and growing fiscal implications, they distort price signals to households and lead to inefficient energy consumption patterns and add to the costs for industry undermining its competitiveness. Financed primarily through cross-subsidies, they are also largely outside the resource prioritization and control processes of the budget framework. As such they remain shielded from competition with other expenditure programs. They also impose an implicit tax on the real sector, adding to an already large tax burden while depriving utility companies of the financing needed for routine maintenance and investment. Figure 3-7. Energy subsidies have kept  but energy subsidies disproportionately benefit … household utility expenditures low… higher income groups Household utility expenditures share in total consumption Energy subsidies percent of household consumption percentage of total subsidies 20 25 18 16 20 14 12 15 10 8 10 6 4 5 2 0 0 MDA MNE MKD ARM HUN BLR ROU UKR HRV GEO BGR KGZ KAZ TUR SRB 2009 AZE ALB POOR 2 3 4 5 6 7 8 9 RICH TJK POL EST LTU BIH LTV income decile Sources: HBS 2010, ECA databank, World Bank staff estimates. 34 For most electricity consumers, the tariff is set above cost recovery level, as defined by Belenergo (see above). The only customer groups for which electricity tariff was set below cost recovery were residential consumers (continuously over 2005–2010), and agricultural consumers (2005–2007). 44 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery While under-pricing has ensured affordability of energy supplies to households, these universal consumption subsidies are highly regressive. Household expenditures on utilities in Belarus are among the lowest in the Europe and Central Asia region. On average, Belarusian households spend less than 7.5 percent of total their consumption on utilities share on utilities, among the lowest in the region in 2009 (left-hand panel Figure 3-7). Heating expenditures represent 69 percent of households’ energy expenditures, electricity 23 percent and gas only 8 percent. However, poorer households spend a larger share on energy relative to the top deciles. In 2009, the energy expenditure of the lowest income quintile was 10 percent while energy expenditures of the rich was approximately 6 percent (right-hand panel Figure 3-7). Given this pattern of utility spending, poorer households are more vulnerable to changes in utility prices, an important issue that needs to be considered in energy pricing and that will be analyzed in more detail below. Energy subsidies are poorly targeted, disproportionately benefiting higher income groups. Energy subsidies are strongly regressive. While higher income groups tend to spend a lower share of their income on utilities, in absolute terms their consumption of energy is higher (because high income groups tend to occupy larger living areas). As a result, 45 percent of the overall subsidy for utilities flows to the top 30 percent of households in the income distribution, compared to 15 percent for the lowest 30 percent of households. People in urban areas receive 82 percent of the subsidy, and Minsk gets the highest share among the regions. Employed people, pensioners and small families receive the majority of the subsidy, while poor big families and other groups of the population at the lower end of the income distribution get a smaller portion. At the same time, the cross-subsidies from non-residential users to residential customers add to an already large tax burden on the business sector. About 1.4 percent of GDP is transferred through cross-subsidies from non-residential users of electricity, heat and gas to residential customers. This imposes a direct cost on the business sector, especially those in energy-intensive sectors, undermining their competitiveness. Over the past five years, non-residential electricity tariff increased by 60 percent in real terms, outgrowing GDP growth. As a result, Belarus has one of the most unbalanced tariffs structures in the region. The ratio of industrial to household tariffs for both electricity and gas remains above 1.0, despite the higher cost of serving households (Figure 3-8). Figure 3-8. Imbalanced tariff structure imposes implicit tax on the real sector Ratio of industrial-to-residential electricity tariffs 3.0 2.5 2.0 1.5 1.0 0.5 0 HUN ARM POL EST ROU HRV GEO RUS SVK TUR BGR KAZ AZE LTU LVA ALB SRB MDA BIH MKD KGZ BLR UKR Source: Energy Regulators Regional Association. Chapter 3. Energy services: achieving financial viabilit y 45 Bel arus Public Expenditure Review Figure 3-9. Cross-subsidies squeeze net profits Under-pricing has undermined the financial of Belenergo position of utility companies, constraining US$ million (current prices) percent GDP their ability to meet routine maintenance and 450 1.2 1.1 capital requirements. The hefty margins that 400 1.0 1.0 utility providers realized during period of low 350 300 0.7 0.8 gas prices (until 2007) allowed them to finance 250 0.6 0.6 investments nearly exclusively from cash flows. 200 394.0 150 298.3 320.1 339.5 0.4 0.4 During 2006–2009, Belenergo invested about 100 174.9 0.2 US$2.15 billion in modernizing its capital 50 assets. This has allowed for reducing the wear 0 0 2005 2006 2007 2008 2009 of major capital assets from 60.7 percent in Belenergo net profits 2005 to 51.2 percent in 2009.35 However, Source: Belenergo. as a result of falling electricity consumption by industry in 2009 (which had provided a Figure 3-10. Rising fuel costs crowd out O&M and depreciation charges Heat (Belenergo) Electricity (Belenergo) percent percent 100 100 80 80 60 60 40 40 20 20 0 0 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010 Fuel Costs Transport O&M, Repairs Salaries Depreciation Payments to innovation fund Sources: Belenergo, ZhKH, World Bank staff estimates. major source of income for Belenergo) and an Figure 3-11. Investment needs and plans in the increased need to finance cross-subsidies to energy sector residential consumers, Belenergo’s financial US$ billion (current prices) net profit declined sharply from US$396 12 million in 2006 to US$174.9 million in 2009 10 Investment gap USD 3.5 billion (nominal), or from 1.1 percent of GDP to 8 less than 0.4 percent of GDP (Figure 3-9). In 6 11.0 Planned investment addition, contributions to the Innovation Fund, 5.3 4 2010–15 operations and maintenance expenditure 2 Actual investment and depreciation charges levied to recoup 2.2 0 2006–09 amortization and financing cost of investments Required Planned & actual investment investment (2006–15) have been crowded out by rising fuel costs Sources: Ministry of Energy, World Bank staff estimates. (Figure 3-10). 35 See Ministry of Energy, State Program of the Development of Belarusian Energy System (2011-2015). 46 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery This deteriorating financial situation is aggravated by the existing investment gap in the energy sector. Historic and planned investment levels in the heating and electricity sector (Belenergo) fall short of the estimated investments needs (as of 2005) for the rehabilitation and modernization. The Government’s own energy strategy places investment needs in the range of about US$1 billion annually up to 2015, leaving an estimated financing gap of about US$3.5 billion (Figure 3-11).36 D. Fiscal and social impact of the planned energy tariff reform With energy costs expected to rise, the importance of energy prices reflecting costs is well recognized in Belarus. To restore financial viability of the sector, the State Program of the Development of Belarusian Energy System (2011-2015), which was proposed in August 2010, calls for a comprehensive tariff reform and gradual elimination of cross-subsidization in natural gas supply by 2013, and in electricity and heat supply by 2014 (Table 3-1). If implemented as planned, these reforms will generate significant fiscal savings, but they will require steep increases in tariffs that will hurt especially the poorer households. The following section estimates the required tariff increases, the expected fiscal savings and the social impact of the proposed reforms under different gas price scenarios (Box 3-2).37 Table 3-1. GoB’s target household tariffs levels and budget subsidy to compensate for cross-subsidization 2011 2012 2013 2014 Cost recovery levels of tariffs for natural gas (percent) 62.9 80.7 100 100 Cost recovery levels of tariffs for electricity (percent) 67.0 77.1 89.6 100 Cost recovery levels of tariffs for heat (percent) 53.4 69.2 84.6 100 Cross-subsidization (mln US$) 537 476 242 0 Source: Ministry of Energy. The devaluation of the exchange adds further to cost pressures in the energy sector, reinforcing the need for tariff adjustments but likely requiring a softening of the adjustment path to achieve full cost recovery. Since the energy sector is heavily dependent on natural gas imports, recent steps by the 36 For the purposes of this chapter, modernization is defined as replacement of obsolete capital assets, without expanding installed capacity of the system. Required investments in capital asset modernization (electricity and district heating infrastructure, Belenergo) were estimated as of 2005, assuming 60.7 percent of capital assets were obsolete. International prices for energy equipment (HoB, CHP, T&D networks) were used. Capital recovery factor (CRF) was estimated at 10 percent discount factor, assuming 20 years lifetime of investments, and constituted 0.117. The energy cost allocation method was used in the case of CHPs. The data for historic and planned investments in capital assets modernization was obtained from the Ministry of Energy (See State Program of the Development of Belarusian Energy System for 2011–2015). 37 The estimates make the following assumptions: • Constant 2009 US$ terms. • Constant cost structure of heat and electricity supply for 2011-2014 as reported by Belenergo and ZhKH. Considering significantly worn-out state of major capital assets, allowing for adequate recovery of investments needed for asset replacements through internal cash revenues (tariffs) is recommended. • Transmission and distribution costs of natural gas supply at US$40 per TCM. Transmission costs at US$10 per TCM. • Efficiencies: 90percentfor heat generation; 38 percent for electricity generation at condensing power plants; 35 percent for electricity generation at combined-heat-and-power plants operating in condensing moe; 73 percent for total electricity and heat generation at combined-heat-and-power plants operating in co-generation mode.  T-cost of electricity supply was assumed at 0.5 US¢/kWh; D-costs – 1.0 US¢/kWh. Chapter 3. Energy services: achieving financial viabilit y 47 Bel arus Public Expenditure Review Box 3-2. Natural gas price scenarios 2011–2014 n order to estimate the fiscal and social impact of the proposed tariff reforms, four alternative I scenarios with different assumption about the import price of natural gas are run: Base Case: World Bank’s Global Economic Prospects projections for the average European natural gas price were used. The prices were net-backed to Belarus using an economic (full cost recovery) price for natural gas transmission. Low Case: 20 percent reduction in natural gas border prices, compared to the Base Case. High Case: 20 percent increase in natural gas border prices, compared to the Base Case. Base Case Competing Fuel: World Bank’s Global Economic Prospects projections for the average crude oil price and historic relative prices for Low Sulfur Fuel Oil (LSFO) (Rotterdam, Holland) were used. Natural gas was assumed to be priced equally to LSFO (in calorific value equivalent). This pricing approach has been used by some natural gas suppliers in a number  Natural gas price scenarios of countries, where natural gas was priced NG border price US$/TCM, 2009 constant prices based on the price of a competing fuel. 350 In the conditions of Belarus, both thermal 300 power plants (TPPs) and heat-only boilers 250 (HoBs) could be operated using natural gas 200 or LSFO/mazut. 150 100  nder all price scenarios, gas prices are U 50 expected to peak in 2011 and 2012 and 0 slowly decline, primarily due to the projected 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 decline in international crude oil prices. Base case Base competing fuel Low case High case NBRB and the subsequent devaluation of the exchange rate imply a further and substantial reduction of cost recovery levels. At the same time, given the substantial shocks households will have to absorb in view of the devaluation, rising inflation and declining real incomes significantly narrow the space for tariff adjustments this year. As the estimations in the following sections, which are based on an official exchange rate of 3,050 BYR/US$, will show, the implied tariff adjustments of the stated cost recovery targets were already ambitious, but will become socially unsustainable in view of the cost increase implied by the exchange rate devaluation. While the objective to achieve cost reflective tariff structures is reinforced by these latest developments, it may take more than four years to attain this objective. The required tariff increases to achieve the stated cost recovery targets are substantial (upper panels Figure 3-12). Given low initial cost recovery levels, residential heat tariffs would have to be increased most aggressively: from 112 to 256 percent by 2014. The range of the cumulative tariff increases for natural gas and electricity is from 110 to 233 percent and from 63 to 108 percent, respectively (Table 3-2). All ranges are in real terms and nominal increases would have to be higher to take into account inflation. 48 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery Figure 3-12. Required residential tariff increases are substantial, but so would be the fiscal savings Heat tariffs Electricity tariffs Natural gas tariffs US$/Gcal US$/KWh US$/TCM 70 14 400 60 12 350 50 10 300 250 40 8 200 30 6 150 20 4 100 10 2 50 0 0 0 2009 2011 2013 2015 2009 2011 2013 2015 2009 2011 2013 2015 Base case Base competing fuel Low case High case Subisidy need without tariff reform (BelEnergo & ZhKH) Fiscal savings generated by tariff reforms (BelEnergo & ZhKH) percent of GDP percent of GDP 5 3.5 3.0 4 2.5 3 2.0 2 1.5 1.0 1 0.5 0 0 2011 2012 2013 2014 2011 2012 2013 2014 Base case Base competing fuel Low case High case Subsidy need with tariff reform (ZhKH) Subsidy need with tariff reform (BelEnergo) percent of GDP percent of GDP 0.7 2.5 0.6 2.0 0.5 0.4 1.5 0.3 1.0 0.2 0.5 0.1 0 0 2011 2012 2013 2014 2011 2012 2013 2014 Base case Base competing fuel Low case High case GoB's plan Sources: World Bank staff estimates. Belenergo, ZhKH, MoF. Note: All tariff projection in 2009 US$. Chapter 3. Energy services: achieving financial viabilit y 49 Bel arus Public Expenditure Review Table 3-2. Implied cumulative tariff increases under different price scenarios Base Optimistic Pessimistic Base Case Competing Fuel Scenario Scenario HH Electricity Tariffs, USc/kWH 63 108 39 86 HH Heat Tariff, US$/Gcal 161 256 112 211 HH Natural Gas, US$/TCM 152 233 110 194 Source: World Bank staff estimates. The proposed tariff reform would generate large fiscal savings. Assuming no real increases in residential tariffs, the estimated amount of subsidies (both budgetary and cross-subsidies) needed to compensate for the gap between the cost of energy supply and revenues from energy sales to households for 2011–2014 is in the range of US$4.3–8.9 billion (in 2009 constant prices) depending on the natural gas price scenario, equal to 1.9–3.9 percent of GDP annually.38 Subsidy needs would peak in 2011 at 2.2–4.3 percent of GDP and then slowly decline (middle-left panel, Figure 3-12). Following the proposed tariff increases, subsidy needs would sharply decline over the period and reach zero for Belenergo (once full cost recovery level is achieved in 2014) and 0.2–0.4 percent of GDP for ZhKH (lower panels of Figure 3-12). This would generate fiscal savings in the range of US$2.4–6 billion, or 1 and 2.6 percent of GDP on average annually. Fiscal savings would increase rapidly to their peak level by 2014 of 1.4–3.3 percent of GDP (middle right-hand panel, Figure 3-12). While generating large fiscal savings, tariff reforms would affect household welfare, especially among the poor. The distributional impact of the increases outlined above is estimated based on household monetary loss due to tariff increase.39 The loss of each household is estimated as a function of the level of the corresponding tariff, share of the utility in total consumption, the proposed increase in the tariff and consumption elasticity to the increase in prices (Figure 3-13). Figure 3-13. Higher tariffs will impact poverty Poverty projections Loss as percent of GDP poverty rate percent of GDP 7.5 3.5 3.0 7.0 2.5 6.5 2.0 6.0 1.5 1.0 5.5 0.5 5.0 0 2010 2011 2012 2013 2014 2011 2012 2013 2014 Base case Base competing fuel Optimistic scenario Pessimistic scenario Source: World Bank staff estimates based on HBS. Baseline poverty rate is estimated 5.4 in 2009. The scenarios are based on -0285 consumption elasticity for all components. 38 While there is a government plan to eliminate cross-subsidization for Belenergo (Table 3-1), the authors are not aware of such plan for ZhKH. We assume that the tariff adjustments needed to achieve full cost recovery for Belenergo are uniformly applied to ZhKH. 39 See Annex 3.2 for the methodology used for the estimation of the social impact of tariff increases. 50 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery Tariff increases to cost recovery levels, with no mitigating measures, would increase the poverty rate 1.4–1.9 percentage points over four years, using the Belarus national poverty line. As a result of the proposed reform, up to 190,000 people could fall into poverty and an additional 120,000 in moderate poverty and vulnerability, keeping other economic parameters including household consumption constant. The total monetary loss of the population due to the proposed residential tariff increase is between 2.1 to 2.9 percent of GDP, directly mirroring the estimated fiscal savings resulting from tariff increases. As result of the proposed reforms, the share of household energy consumption is expected to rise to 12.4 percent in 2013, which is approximately average in ECA region now, and to 16 percent in 2014 (at cost recovery level). Given the regressive incidence of energy subsidies, most of the fiscal savings are generated from withdrawing the general consumption subsidy from middle class and rich households that can afford cost recovery tariffs. A disproportionate amount of the energy subsidy is flowing to higher income groups. Only 2 percent of the total subsidies flow to the poor, while 98 percent are captured by households above the poverty line. Conversely, the income losses following tariff increases will primarily accrue to richer households, expected to account for between 2 and 2.8 percent of GDP (Figure 3-14). Figure 3-14. Simulated impact of the tariffs increase on population groups Distribution of energy subsidy share in total population 100 82.4 80 64.1 60 43.8 41.1 34.4 31.7 35.5 40 21.2 25.6 26.4 27.0 26.7 15.3 20.2 17.6 13.3 12.9 14.0 10.5 13.9 9.8 17.7 11.7 19.6 20 8.9 3.0 6.2 3.4 1.7 0.8 0.6 1.2 0 Rural Urban Brest Vitebsk Gomel Grodno Minsk Minsk-Oblast Mogilev Basic General secondary Special secondary Tertiery None/ primary Employee Self-employed Unemployed Retired Student OLF HH with children 0–17 years HH with elders 60+ Q1 Q2 Q3 Q4 Q5 1 2 3 4 5+ Quintile Region Household size Head education Labor force Comp. Source: World Bank staff estimates, based on HBS 2009. While the aggregate income loss of poor households is relatively small, poorer households are particularly vulnerable to the adverse impact of tariff increases. Poor households spend a larger share of their income on utilities, which would be expected to reach 18 percent under cost-reflective tariff structures. Almost 80 percent of the increase in utility spending would come from the proposed increase in heating tariffs. This would cause a significant reduction in overall household welfare, and low-income families should therefore be supported through the social protection system to compensate for the income loss resulting from tariff increases, as envisaged in the Government’s energy strategy. The loss of the lowest 25 percent of the income distribution is between 0.2 and 0.5 percent of the GDP. In other words, in order to compensate the lowest 25 percentile of the population for the loss due to the proposed reform, taking in the consideration the efficiency of Belarus’ social protection, a targeted transfer of 0.3 to 0.7 percent of GDP would be required in order to mitigate the impact of higher tariffs on the poor. Chapter 3. Energy services: achieving financial viabilit y 51 Bel arus Public Expenditure Review E. Policy options and their fiscal and social impact Reform of the energy sector remains a key priority of the Government. Soaring energy prices combined with under-pricing of domestic energy and a sector structure that favors large and in some cases loss- making government-owned enterprises—financed through an opaque system of indirect subsidies— makes for a difficult reform environment. Long-standing underinvestment in the sector has led to significant level of wear, resulting in reduced energy supply efficiency and reliability. Fiscal pressure for reform is rising, but price hikes are unpopular and politically difficult to implement. However they are unavoidable if growing financial liabilities and fiscal crisis are to be avoided and the sector is to be put back on a financially sustainable and self-reliant path. Within these constraints, the Government is making progress towards structural reforms. A number of policy options emerge from the analysis presented in this chapter. First, the analysis reinforces the need for comprehensive tariff reforms to improve commercial viability of energy service provision and to reduce the fiscal burden resulting from under-pricing. While tariff reforms are sensitive, experience from other countries shows that they can be both politically feasible and fiscally efficient if accompanied by countervailing measures to compensate the loss in purchasing power, in particular of the poor, through targeted social assistance (discussed in Chapter 4). Additional options include: ••  Increase Belenergo’s industrial heat tariffs to cost recovery levels. Estimated fiscal savings of this measure would be about 0.06–0.1 percent of GDP. ••  Gradual phase out of preferential electricity and gas tariffs granted by Ministry of Economy resolutions on an ad hoc basis. Estimated fiscal savings of this measure would be about 0.3 percent of GDP. ••  Consider two-part tariffs for heat and natural gas. This results in a more transparent cost structure, specifically regarding capital costs, and typically improves financial viability of energy utilities in the long run. ••  Consider progressive utility tariff structures, which not only could have positive distributive impacts, but also encourage efficiency savings at the margin.40  •• Consider establishing a formula-driven tariff setting mechanism. The formula should comprise the following elements: 1) OPEX (which should automatically adjust for rising input prices), 2) depreciation (reflecting long-term marginal costs41), 3) allowed return42 and 4) adjustment factor (which should allow tariffs to gradually catch up to cost recovery levels). The parameters should be determined in a transparent manner to provide adequate revenues to the utility to meet its cost of service delivery. 40 To encourage efficient consumption of energy resources and prevent substantial growth in energy tariffs, the GoB also considers the introduction of differentiated tariffs for gas, electricity and water. Details of the proposal were not available at the time of drafting this report. 41 Depreciation = asset base / average asset life. 42 Allowed return = asset base * weighted average cost of capital in financing the assets. 52 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery The required revenue should include adequate return on investment and incentives for the utility to operate and maintain its network and asset base in a socially optimal manner. Second, where tariffs below cost recovery are deemed desirable, financing should be made more transparent and explicitly provisioned in the budget as direct subsidies of the end-user price. This would reduce fiscal risks and expose resources devoted to lowering energy prices to the same expenditure prioritization process as other competing expenditure programs. Third, further progress will require strengthening the regulatory environment to generate a more conducive climate for private investments and competition in the sector. Priority reforms should be aimed at enabling energy companies to raise commercial financing and improve operational, managerial and financial efficiency. The current institutional environment, while providing stability, suffers from constraints in this regard. Furthermore, tariff setting and regulatory arrangements are complex and lack the autonomy needed to ensure prudent regulation. As a consequence, energy companies have limited access to external finance and lack economic incentives to improve efficiency. While not the focus of this report, several institutional reform options exist and were discussed in detail in a World Bank Energy Policy Note in 2006. The fiscal impact of the proposed reforms would be positive, and could be quite significant depending on developments in the import price of natural gas. The estimated fiscal savings generated by the proposed tariff increases would amount to 1.4–3.3 percent of GDP, depending on year and gas price scenario. Since under current practice a large share of the subsidies is covered through cross-subsidies, these savings would primarily accrue in the form of increased net profits of Belenergo (accounting for about 0.9–2.1 percent of GDP) needed to maintain adequate level of investments in operation and maintenance and capital asset replacement. To a lesser extent budgetary savings for local governments would be generated through reduced subsidy needs for ZhKH, amounting to 0.5–1.2 percent of GDP. Chapter 3. Energy services: achieving financial viabilit y 53 Bel arus Public Expenditure Review Table 3-3. Fiscal and social reform impact of proposed measures Policy Recommendations Estimated Fiscal Impact Estimated Social Impact Gradually adjust residential energy tariffs in 2011–2014 Total savings in the range Increase in poverty rate of to cost recovery levels, as stated in the State Program for of 1.4–3.3 percent of GDP 0.35 percent annually, in the Development of Belarusian Energy System. annually 2011–14, of which the absence of mitigation 0.9–2.1 percent of GDP in measures (see below) reduced cross-subsidies and 0.5–1.2 percent of GDP in reduced budgetary subsidies Introduce well-targeted safety nets for the poor as well as Costs in the range of Full mitigation of poverty financial and technical assistance for end-use efficiency 0.3–0.7 percent of GDP impact to soften the social impact (also discussed in Chapter 4). Increase Belenergo non-residential heat tariffs to cost Savings in the range of Neutral recovery levels (currently at about 93 percent). 0.06–0.1 percent of GDP Consider two-part tariffs for heat and natural gas. This Further assessment needed Neutral results in more transparent cost structure, specifically regarding capital costs, and typically improves financial viability of energy utilities in the long-run Consider progressive electricity tariff structures, which Further assessment needed Neutral not only have positive distributive impacts, but also encourage efficiency savings at the margin. Gradually phase out of preferential electricity and gas Savings of about 0.3 percent Neutral tariffs granted by Ministry of Economy resolutions on an of GDP ad hoc basis. 54 Chapter 3. Energy services: achieving financial viabilit y Fiscal Reforms for a Sustainable Economic Recovery Chapter 4. Social assistance: Improving targeting B elarus operates one of the most extensive social assistance systems in the region, with total spending equal to 2.7 percent of GDP and reaching about half of the population. This substantial resource commitment has contributed to low levels of poverty and to low inequality in the country. However, the current system uses up significant fiscal resources and in the tight fiscal environment, pressure to contain sectoral spending, including social assistance, remains acute. At the same time, robust social safety nets are needed to preserve the significant gains in social development that Belarus has achieved, especially in view of a possibly protracted economic recovery and structural reforms that are likely to hurt the most vulnerable. Faced with these simultaneous challenges, the need to realize efficiency gains is urgent. While social assistance programs have broad coverage among the poor, targeting is weak. Benefits, and especially privileges, reach many beneficiaries in the higher deciles of the income distribution. As a result, only about 40 percent of social assistance goes to the poorest 20 percent of the population. Targeting efficiency is undermined by spending on a number of privileges extended to war veterans, disabled and other categorical groups of citizens which are poorly targeted and which continue to absorb about 0.5 percent of GDP in 2010 despite a rationalization of these programs over the past years. The share of income- tested programs to support low-income families has grown rapidly but remains small, accounting for only about 0.05 percent of GDP in 2009. Despite scaling up income-targeted benefits, the overall program mix has shifted away from income-tested programs toward categorically targeted ones. Only four out of 14 programs apply income-related eligibility criteria. Spending on income-tested programs fell from 9.4 percent of total sector spending in 2005 to only 6.3 percent in 2009. To improve efficiency, the analysis suggests a set of inter-connected recommendations aimed at reducing poorly targeted benefits and partially reallocating the savings to targeted programs. Going forward, a further realignment of spending from untargeted to targeted programs is needed to improve the impact and reduce the cost of social assistance expenditures. 55 Bel arus Public Expenditure Review A. The fiscal cost of social assistance—Spending trends and composition Belarus has a well-developed social assistance system, providing a wide range of benefits to a large part of the population. The mix of programs is similar to that of OECD countries, with an emphasis on family allowances such as childcare benefits and birth grants, non-contributory pensions and income support to people with disabilities. In addition, Belarus has an extensive system of categorical benefits extended to veterans, Chernobyl victims and a number of occupational groups. The Government has begun to rationalize categorical benefits while introducing a targeted social assistance program designed to provide income support to households below the poverty line. With total spending equal to 2.7 percent of GDP, Belarus redistributes a larger share of its GDP through its social assistance system than most countries in the region. In 2009, Belarus spent BYR 3.6 trillion (US$3.1 billion at PPP-adjusted exchange rate) on various social transfers and income support programs. This is about 67 percent (1 percentage point of GDP) more than the regional average (Figure 4-1). The spending level is closer to high-income OECD countries, which spend on average 2.5 percent of GDP. Social assistance programs absorb about 6.2 percent, more than transport (about 5 percent) but less than health (about 9 percent) and education (12 percent). More than 80 percent of this spending is financed through the Social Protection Fund, which is administered by the Ministry of Labor and Social Protection, with the remainder coming from the state budget. Figure 4-1. Belarus redistributes a larger share of GDP through social assistance programs than most countries in the region Social Assistance Spending in percent of GDP percent GDP, 2008 3.0 0.01 LVA 0.8 0.01 0.03 0.05 0.01 0.04 BGR 1.2 2.5 0.6 0.5 POL 1.4 1.0 0.8 0.8 0.7 LTU 1.5 2.0 EST 1.5 0.8 0.8 ROU 1.6 1.5 0.7 0.7 0.7 0.7 RUS 1.8 SRB 1.9 1.0 UKR 1.9 1.2 1.1 1.1 1.1 1.2 1.4 BLR 2.7 0.5 BIH 3.3 0 HRV 3.4 Jan-Sep 2005 2006 2007 2008 2009 2010 0 1 2 3 4 Child related births Disability related benefits Privileges Trageted social assistance (GASP) Sources: MoF, MoL, ECA regional database, and World Bank staff estimates. On the back of strong economic growth, social assistance spending increased 27 percent in real terms between 2005 and 2009, but remained stable as a share of GDP and of total public spending. Spending increases were driven by rising benefits under most programs, as the Government redistributed the gains of high growth through pro-cyclical income policy and social transfers. At the same time, the number of beneficiaries decreased substantially for some child-related benefits and for privileges, leading to a relative contraction of spending on these programs measured against GDP and total public spending (discussed in more detail below). 56 Chapter 4. Social assistance: improving targeting Fiscal Reforms for a Sustainable Economic Recovery While the level of social assistance spending has remained stable, the composition has shifted. Spending is composed of 14 main programs with varying objectives, resource commitments and eligibility criteria (Table 4-1).43 The largest programs in terms of spending are various child-related allowances (about 1.4 percent of GDP, 2010) and allowances for temporary disability and medical treatment (0.8 percent of GDP, 2010). In addition, about 32 categorical privileges (and subsidies) extend to war veterans, disabled and other groups of citizens for various services ranging from public transportation to housing and utilities, which together account for 0.6 percent of GDP in 2010, down from 1 percent in 2005. The share of income-tested programs to support low-income families has grown rapidly (26 percent in real terms between 2005 and 2010) but remains small, accounting for only about 0.05 percent of GDP in 2009. Box 4-1. Family allowance benefits in Belarus Family allowances: For a child younger than age 3, monthly allowances of 100 percent of the national average subsistence income level are paid to non-working women; 50 percent to women employed more than 50 percent or with a child in kindergarten. For a child aged 3 or older, 30 percent of the national average subsistence income level is paid for each child whose family income does not exceed 60 percent of the average national average subsistence income level for September of the previous year; 50 percent of the benefit is paid if family income is 61 percent to 80 percent of the average national average subsistence income level for September of the previous year. Maternity benefit: Employed women receive 100 percent of average monthly earnings; students on leave from employment receive 100 percent of the education grant; unemployed women receive 100 percent of the unemployment benefit. The monthly minimum benefit is 50 percent of the national average subsistence income level. The monthly maximum benefit is three times the national average wage in the month before childbirth and maternity leave. Benefits are paid for 126 calendar days (140 calendar days in the case of complications or multiple births). For the adoption of a child younger than 3 months, the benefit is paid for 70 days, starting from the date of the adoption. Prenatal care grant: A lump sum of the average sub¬sistence income level is paid during the first 12 weeks of pregnancy. Birth grant: A lump sum of 500 percent of the national average subsistence income level is paid for the first child; 700 percent for subsequent children. Additional grants are paid for multiple births and for fami-lies with several children. Sick childcare allowance: 100 percent of the salary of the parent who cares for the child is paid for up to 14 days for outpatient care; without limit for hospitalization. Disabled childcare allowance: 65 percent of the national aver¬age subsistence income level is paid. Source: http://www.ssa.gov/policy/docs/progdesc/ssptw/2010-2011/europe/belarus.pdf. 43 Most social benefits are regulated by Law 3563-XII of January 31, 1995 and / or by Law 1898-XII of October 30, 1992 and financed by the SPF. Public targeted social assistance (GASP) benefits are financed from the national budget and administered in accordance with Presidential Decree 458 of September 14, 2009 and relevant government decrees. Chapter 4. Social assistance: improving targeting 57 Bel arus Public Expenditure Review Table 4-1. Overview of social assistance programs and design features Program AU* Eligibility Basic formula Payment Amount** Other Adjustments Categorial Programs 100%*MSB if mother If single mother: works half time 1) B + 75%*B before 1.5 Child care benefit, for 50%*MSB if 1) mother Ind Categorical Monthly 193,800 years; children from 0 to 3 years works more than half a month; 2) at 1.5 years 2) B + 40%*B after 1.5 child starts at pre-school years Paid regardless of other Allowance for HIV-infected Ind Categorical B = 45%*MSB Monthly 98,984 child-related benefits children under 18 received 5*MSB for first child Child birth grant Ind Categorical 7*MSB for second One-time 1,591,820 (third...) child Allowance for goods of first necessity for children in Ind Categorical 2*MSB per child One-time 497,000 case of multiple births Pre-12th pregnancy week Ind Categorical MSB One-time 244,400 registration Paid as of the 30th week of pregnancy for 126 days; Categorical 100% of average wage to adopting parent - for 70 Maternity benefit Ind One-time n/a days If insured Not more than 3*AMW plus certain Min flat benefit is 50%*MSB categories Categorical Children under 14 staying at Benefit for taking care of home - 14 days covered sick children under 14 and in case of mother's or If treated in hospital: Ind 100% of average wage Term n/a under 5 - for the entire other guardian's illness, If insured of children under 3 and period; aged 5-14, disabled disabled children under 18 child - for the period when additional care is required Allowance for medical treatment and rehabilitation Ind Categorical 100% of average wage Term n/a of disabled children under 18 Categorial 80% of average wage for Granted for no longer than Allowance for temporary first six days, then - 100% 120 consecutive days disability or medical Ind Term n/a If insured or 150 days if repeated treatment and rehabilitation Not more than 3*AMW incidents in one year Allowance for social Cash or in kind; size of rehabilitation devices for Ind Categorial - One-time n/a benefit is determined on a disabled children under 18 case by case basis and disabled category 3 Income-tested programs Awarded for up to six B = MSB - Y per each months; may not exceed six GMI program (GASP) Fa Y < MSB Monthly 65,400 family member months in one year; may be in kind One-time social benefit to Awarded once over 12- low-income families in hard Fa Y < 150%*MSB B = up to 500%*MSB One-time 174,000 month period; may be in circumstances (GASP) kind Programs using mixed approach Categorial B + 40%*B if lone mother Child care benefit, for 1) Y < No benefit if able bodied children from 3 to 16 (18) Ind 60%*MSB 1) B = 30%*MSB Monthly 72,700 single mother or able- years old 2) Y < 2) B = 15%*MSB bodied father not working or 80%*MSB on unemployment benefit Categorial B + 40% of respective child Benefit for taking care of Ind No other B = 65%*MSB Monthly 161,535 care benefit (up to 3/from 3 disabled children under 18 income to 16 (18)) Benefit for taking care of Categorial B = 65%*MSB disabled category 1 or Ind No other B = 100%*MSB if two Monthly 161,535 seniors above 80 income more individuals in care Note: AU: assistance unit; Fa: family; Ind: individual beneficiary; MSB: national subsistence minimum budget, or poverty line; AMW: national average monthly wage; B: benefit. 58 Chapter 4. Social assistance: improving targeting Fiscal Reforms for a Sustainable Economic Recovery Spending on child-related benefits remains the largest spending category, absorbing almost half of the spending in the sector, but the allocation across different child-related benefits has changed (Box 4-1). Total spending on child-related benefits in 2009 amounted for BYR 1.7 trillion (US$1.5 billion at PPP-adjusted exchange rate), equal to 46 percent of the total spending in the sector. Spending has been stable and expanded in step with rising GDP. Figure 4-2. Cost drivers of social assistance programs: Spending on some programs has expanded but has fallen on others Child care for children under age of 3 Child care for children above age of 3 average benefit, in thousand Roubles average benefit, in thousand Roubles number of beneficiaries, in thousands number of beneficiaries, in thousands (2005 constant prices) (2005 constant prices) 350 1,600 350 700 300 1,400 300 600 250 1,200 250 500 1,000 200 200 400 800 150 150 300 600 100 400 100 200 50 200 50 100 0 0 0 0 2005 2006 2007 2008 2009 2010 2005 2006 2007 2008 2009 2010 Number of beneficiaries Average benefit Targeted social assistance Privileges average benefit, in thousand Roubles average benefit, in thousand Roubles number of beneficiaries, in thousands number of beneficiaries, in thousands (2005 constant prices) (2005 constant prices) 350 250 3,000 500 300 200 2,500 400 250 2,000 200 150 300 1,500 150 100 200 1,000 100 50 500 100 50 0 0 0 - 2005 2006 2007 2008 2009 2010 2005 2006 2007 2008 2009 Number of beneficiaries Average benefit Sources: MoL. World Bank staff estimates. While child-related benefits are the largest Figure 4-3. Child-related benefits have risen as spending category, the allocation has shifted a share of total social assistance spending percent of total social assistance towards benefits for children under 3 years 50 1 1 of age (Figure 4-3). Spending has been 1 1 1 40 1 steadily declining for childcare benefits for 15 14 17 18 23 13 children above the age of 3. While the average 30 benefit paid under this program increased 20 13 13 14 13 14 13 in line with other social assistance benefits, 5 10 6 6 6 7 7 this increase was more than offset by falling 1 1 3 4 4 4 8 7 5 numbers of beneficiaries (Figure 4-2), as 0 4 3 3 Jan-Sep income growth pushed more families above the 2005 2006 2007 2008 2009 2010 Child care benefit for children above 3 years old Child birth grant income eligibility threshold for these benefits Maternity benefit Multiple birth benefit Child care benefit for children up to 3 years old Other (80 percent of minimum subsistence income). Source: MoLSP. Chapter 4. Social assistance: improving targeting 59 Bel arus Public Expenditure Review During the same period, spending on childcare benefits for children under the age of 3 expanded as the client base of the program grew and average benefits increased. Spending on privileges dropped sharply as legal changes eliminated a large number of categorical transfers and the number of beneficiaries fell. Law No. 239-3 on Social Privileges, Rights and Guarantees, enacted in 2007, eliminated a large number of privileges.44 Prior to the law more than 50 categories of citizens were eligible for more than 100 different types of social benefits. Currently, 32 privileges extend to 27 categories of citizens. Reflecting these changes, the share of households receiving privileges dropped from 2.4 million households in 2007 to about 1.5 million in 2008. This, together with more modest increases in average benefits, generated significant fiscal savings, bringing down spending on privileges from 1 percent of GDP in 2005 to 0.5 of GDP in 2010. Despite these efforts to rationalize categorical transfers, their coverage remains broad, reaching almost half of the population and absorbing 20 percent of social assistance spending. Some of these privileges, such as privileged job retention, continued health care at the medical institutions of last employment and a subset of first-served rights, are rather notional. However, a subset of privileges for the purchase of various goods and services as well as land tax exemption or higher tax-free income for families with children can be quantified in terms of public expenditures or foregone tax revenues. Around 40 percent of households reported privileges and subsidies, averaging 35,600 BYR (US$32 at PPP-adjusted exchange rate) per household per month in 2009. Most frequently reported were privileges for public transportation, purchase of medicine and medical services and housing and utility services. As the analysis below will show, these benefits are particularly poorly targeted, underlining further opportunities for rationalization to free up resources to be spent more efficiently, including on programs ensuring support to those most in need. Spending on the income-tested program to support low-income families has grown rapidly but remains small, accounting for only about 0.03 percent of GDP in 2009. Spending on income-tested targeted social assistance program (Box 4-2) increased six fold in real terms, as coverage expanded and the number of beneficiaries increased from merely 57,000 in 2005 to nearly 206,000 in 2009. In 2006 application procedures were eased through the introduction of a one-window process for all targeted social assistance (GASP) benefits which, alongside the increase in the amount of benefit itself, resulted in rapid growth of beneficiaries. Despite these increases, coverage is low. Counting all households with per capita income below the monthly social benefits threshold of US$88, we estimate that 480,000 people (4.9 percent) would benefit from this type of program, which is more than double the actual number of recipients of monthly social benefits under GASP in the official statistics. Overall spending also remains low because benefits paid under the program remain relatively modest, despite the recent increases. While income-targeted benefits have been scaled up, the overall program mix has shifted away from income-tested programs toward categorically targeted ones. Spending on income-tested programs 44 The law eliminated benefits for the following categories: citizens disabled category 3; individuals accompanying WWII disabled category 3 and war invalids who fought in other states; citizens who were unfairly persecuted during the repressions of 1920-80s and later rehabilitated; certain categories of servicemen; employees sent to Afghanistan in the period from December 1979 to December 1989 who completed their term of contract or left earlier for good reason; family members other than parents and children of deceased in war or military action. 60 Chapter 4. Social assistance: improving targeting Fiscal Reforms for a Sustainable Economic Recovery Box 4-2. Benefits under targeted social assistance (GASP)  he system of targeted social assistance (GASP) was established in 2001 to provide relief or T income support benefits to the poor. Initially, the needs-based threshold was set at the level of 50 percent of the subsistence minimum (in January 2001), later on (in October 2001) it was lifted to 60 percent, and since December 2007 it was revised to 100 percent, providing full coverage to the population below the poverty line.  ASP benefits are financed from the national budget and administered in accordance with G Presidential Decree No. 458 and other relevant government decrees.  Benefits under this program include: (1) a monthly social benefit awarded for one to six months per year for families whose cumulative income from all sources (and after receiving all social transfers) is below the poverty line. (2) a one-time social benefit payable once per year if a beneficiary family finds itself in a difficult life situation (losing provider because of full disability, turning 80, inability to cater for oneself because of illness, natural disaster, fire, etc.).  oth GASP benefits are paid in cash and/or in kind. In case of the latter, social workers purchase B food, goods for children, medicine and pay for housing and utility services if the beneficiary is found to misuse the aid money and/or fails to take good care of children. fell from 9.4 percent of total sector spending in Figure 4-4. Income-tested programs absorb a 2005 to only 6.3 percent in 2009. Only four out small share of social assistance spending and of 14 programs apply income related eligibility are among the lowest in the region criteria, while all other programs are categorical percent of total social assistance 100 transfers. Reduced spending on child benefits for children above the age of 3 years, which 80 are only paid to families below 80 percent of 60 the minimum subsistence budget, accounts for this shift. The current share of spending 40 devoted to income-tested programs is among 20 the lowest in the region, significantly below the 0 regional average of 37 percent of total social LVA SRB EST BLR LTU UKR ROU BIH POL HRV ALB GEO ARM KGZ MKD UNK assistance spending (Figure 4-4). Source: ECA regional database. Going forward, the Program of Socio-Economic Development of Belarus for 2011-2015 (PSED) identifies the main social protection priorities for Belarus. Targeted social assistance programs will be a major element of state support for low-income households. The program aims at the creation of a unified support system by integrating income-related allowances for children over 3 years of age and free food for children under 2 years into the system of targeted social assistance programs which will simplify the procedure of receiving social assistance, its accessibility and relevance. Moreoever, the program focuses on improving transparency, streamlining administrative procedures, and increasing the effective use of budget funds. Chapter 4. Social assistance: improving targeting 61 Bel arus Public Expenditure Review B. Effectiveness of social assistance spending Social assistance spending has contributed Figure 4-5. Social assistance programs have to low levels of poverty and inequality in the contributed to Belarus’ strong social outcomes country. With 4.9 percent of the population Poverty headcount US$ 5 per day below the national poverty line and a Gini percent of population 35 Coefficient of 0.24, Belarus has one of the 30 lowest poverty rates and most equitable 25 income distributions in the region and for its 20 level of income (Figure 4-5). Social assistance 15 spending directly contributes to these positive 10 social outcomes. In the absence of all social 5 assistance programs, poverty headcount would 0 increase from 4.9 percent to an estimated HRV BLR HUN UKR LTU RUS BIH LVA POL MNE SRB BGR ROU 7.3 percent and the Gini coefficient would Source: ECA regional database. increase to almost 0.26 (Table 4-2). Table 4-2. Social protection program impact on poverty and inequality Poverty Poverty gap Gini coefficient headcount Actual 4.9 percent 0.9 percent 0.245 Without social assistance 7.3 percent 2.2 percent 0.259 Without social assistance benefits other than privileges 6.9 percent 1.9 percent 0.256 Without privileges 5.4 percent 1.1 percent 0.248 Source: HBS 2009. World Bank staff estimates. Effectiveness varies across programs, with Figure 4-6. Privileges are not a cost-effective privileges generally achieving lower impact. way to alleviate poverty Decomposing impact by program type, Impact on poverty gap privileges appear to have a smaller impact on both poverty and inequality than other social Poverty line assistance programs. Privileges reduce the US$ 120 poverty headcount by 0.5 percentage points, while social assistance programs contribute Poverty line a 2 percentage point reduction. These results US$ 88 are underpinned by a cost-benefit analysis (Figure 4-6) where we calculate the reduction 0 0.2 0.4 0.6 (in US$) in the poverty gap45 for US$88 Social assistance w/o privileges Privileges poverty line obtained for each dollar spent on Sources: HBS 2009. World Bank staff estimates. 45 The poverty gap measures the average distance of the poor from the poverty line. 62 Chapter 4. Social assistance: improving targeting Fiscal Reforms for a Sustainable Economic Recovery benefits under social programs, not including administrative cost. If privileges are excluded, social assistance is more cost-effective, with almost US$ 0.23 reduction in the poverty gap for each dollar spent in the program. The impact is relatively small because the poverty line is rather low. With the poverty line set at US$120, which would qualify 20 percent of population as poor, the cost-benefit ratio would be US$0.44 and US$0.26 respectively for social assistance without and with privileges. C. Targeting efficiency of social assistance spending Social assistance programs have wide coverage, reaching more than half of the population. An estimated 5.8 million beneficiaries receive benefits under the various programs, equal to 56.4 percent of the population. Coverage varies substantially between programs. The most extensive coverage is provided by privileges, which—despite the drop in the number of beneficiaries following the rationalization of the system in 2007—continue to reach 40 percent of the population. Families with children are also a sizeable recipient group of social benefits: over 700,000 families receive childcare benefits. In addition, over a half million people with disabilities received benefits in 2009, 6 percent of whom are children. A growing number of low-income households qualified for the income support under GASP. In 2009, 31.8 percent of the 200,000 GASP beneficiaries were still either one-parent families raising under-aged children, and 28.6 percent were families with many under-aged children, followed by 21.5 percent in other categories of families with children. Since 2008, the share of families with children as GASP benefit recipients has been falling, although beneficiaries have growth in absolute terms due to applications from other types of families and single individuals, including pensioners and disabled. Although social assistance programs have broad coverage among the poor, targeting is weak as benefits, especially privileges, continue to also reach many beneficiaries in the higher deciles of the income distribution. Belarus’ social assistance programs have broad coverage among the poorest household: 81.9 percent of the 10 percent poorest households and 68.6 percent of the 20 percent Figure 4-7. Social assistance is reaching the poor, but targeting can be improved Coverage Targeting accuracy share of poorest quintile covered (%) share of total social assistance to poorest quintile (%) 100 80 70 80 60 60 50 40 40 30 20 20 10 0 0 BIH GEO KGZ UNK ALB MKD ARM HRV LTU UKR SRB LVA BLR EST POL ROU BIH LVA EST BLR KGZ UKR POL ROU HRV ALB LTU MKD GEO SRB ARM UNK Source: ECA regional database. Chapter 4. Social assistance: improving targeting 63 Bel arus Public Expenditure Review poorest households reported receiving some type of benefit.46 This is high in regional comparison (left side panel of Figure 4-7). However, 38.4 percent of household in the richest are also receiving some social assistance benefit. Among the households below the poverty line, 75 percent receive benefits, suggesting that only about a quarter of the poor are not being reached with present social assistance programs. A staggering 93 percent of non-poor households receive some type of social assistance benefit, suggesting that coverage is higher among non-poor households. Figure 4-8. Overall social assistance benefits are As result, significant benefits flow to non- progressive poor beneficiaries, undermining targeting Social assistance efficiency. Social assistance spending is percentage of total expenditure progressive (Figure 4-8), only about 27 percent 20 of overall social assistance spending benefits 15 the poorest 10 percent of the population, and about 40 percent goes to the poorest 10 20 percent. Less than 10 percent of the total social assistance spending benefits the 5 population below the poverty line. This places Belarus in the middle of the benefit distribution 0 POOR 2 3 4 5 6 7 8 9 RICH of social safety nets in the region (right-hand income decile panel Figure 4-7). Sources: HBS 2009. World Bank staff estimates. Childcare benefits, the largest spending program, are well targeted. While not income-tested, childcare benefits for children under the age of 3 also have a fairly progressive distribution, due to the fact that young families tend to have lower incomes. In this case, the poorest 20 percent of the income distribution capture 37 percent of the benefits. Targeting efficiency of the income tested childcare benefits for children over the age of 3 is even higher: Almost 60 percent of the expenditures under this program flow to the poorest 20 percent of the beneficiaries (Figure 4-9). Figure 4-9. Benefits incidence of childcare programs Universal childcare benefits successfully reach the …means-tested benefits are even better targeted. low income brackets… childcare benefits under age of 3, percentage of total expenditure childcare benefits above age of 3, percentage of total expenditure 25 50 20 40 15 30 10 20 5 10 0 0 POOR 2 3 4 5 6 7 8 9 RICH POOR 2 3 4 5 6 7 8 9 RICH income decile income decile Source: HBS 2009, World Bank staff estimates. 46 To estimate such result, we ranked the sampled households from 2008 HBS by per capita household income net of benefits, thus approximating pre- transfer income. Then, we divide the population into 10 equal deciles, where decile 1 (D1) represents the poorest 10 percent of the country's population and decile 10 (D10) the richest 10 percent of population. 64 Chapter 4. Social assistance: improving targeting Fiscal Reforms for a Sustainable Economic Recovery Figure 4-10. Privileges are poorly targeted Targeting of privileges is weak and leakages to Privileges non-poor households tend to be pronounced. percentage of total expenditure The coverage of privileges is considerably 16 less progressive: 34 percent of the richest 10 14 percent of the population receive some form 12 of privilege. The weak targeting is even more 10 8 acute in terms of benefit incidence: the richest 6 20 percent of the income distribution receive 4 20 percent of the benefits of privileges, while 2 the poorest 20 percent receive only 27 percent 0 of the benefits (Figure 4-10). The average per POOR 2 3 4 5 6 7 8 9 RICH income decile capita transfer received from privileges by the Sources: HBS 2009. World Bank staff estimates. higher income deciles is 36 percent higher than for the poorest quintiles, according to HBS data. This general tendency masks differences across types of privileges. Only food subsidies and subsidies for pre-school education are well-targeted, reaching the low-income spectrum of the population. Higher coverage is observed in rural areas, among large families, and households with children. However, the larger programs, such as public transportation, sanatorium treatments and medicines, tend to be regressive (Figure 4-11 and Figure 4-12). The financially most significant benefits for sanatorium treatments, for example, accounts for almost 40 percent of the spending on privileges, equal to 0.2 percent of GDP. But less than 20 percent of this spending accrues to the poorest quintile of the population, according to HBS data. Public transport subsidies are also regressive: 85 percent of the benefits accruing to the urban population, covering only about 19 percent the poorest quintile of the population, which receive Figure 4-11. Benefit distribution by quintile of household consumption per capita, 2009 0.4 Telephone 0.4 Housing & utilities 0.4 Electricity 0.2 0.2 0.2 0 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 0.4 Fuel 0.3 Transport 0.3 Medicine 0.2 0.2 0.2 0.1 0.1 0 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 0.3 Health rehabilitation 0.4 Food 1 Preschool education 0.2 0.2 0.5 0.1 0 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Source: HBS 2009, World Bank staff estimates. Chapter 4. Social assistance: improving targeting 65 Bel arus Public Expenditure Review only about 6 percent of the expenditures. Finally, benefits on medicines are highly regressive and large, accounting for 16 percent of total social assistance spending with low coverage (12 percent) and low targeting efficiency (5 percent). Together, regressive programs account for 61 percent of Figure 4-12. Privileges vary in size and targeting the spending on privileges. efficiency coverage, poorest quintile, in percent 70 Social assistance transfers play an important 60 role in supporting the income of the poorest 50 Sanatorium households (Figure 4-12). On average social treatment 38.9 School 40 lunches 31.2 benefits account for 9.5 percent of the income Public 30 transportation Other 4.5 of the poorest 10 percent of the population, but 9.4 20 only for 0.7 percent of the richest 10 percent. 10 Among those families falling below the poverty Medicines 15.9 0 line, social assistance benefits account for 0 5 10 15 20 25 30 35 40 12.4 percent of the income. targeting efficiency, share of total expenditure to poorest quintile Source: HBS 2009, World Bank staff estimates. Note: Bubble size corresponds to size of the program as share of total spending on privileges. D. Policy options and their fiscal and social impact While social assistance programs have Figure 4-13. Social assistance benefits are an been key to reducing poverty and improving important income component for the poorest equality in Belarus, their resource households commitment is among the highest in the transfer as percentage of welfare agggregate 10 region. With the fiscal environment expected to remain constrained over the medium 8 term, pressure to contain sectoral spending, 6 including social assistance, will continue. At the same time, robust social safety nets are 4 needed to preserve the significant gains in 2 social development that Belarus has achieved, 0 especially in view of a possibly protracted POOR 2 3 4 5 6 7 8 9 RICH economic recovery and structural reforms that income decile are likely to hurt the most vulnerable. Faced Source: HBS 2009, World Bank staff estimates. with these simultaneous challenges, the need to realize efficiency gains is urgent. Over the past years the Government has already begun to take important steps to rationalize social assistance programs. This most importantly included the elimination of a large number of categorical social benefits that were poorly targeted and had limited impact on poverty, freeing up about 0.5 percentage points of GDP. At the same time, the Government established a means-tested program designed to provide income support to households below the minimum income. 66 Chapter 4. Social assistance: improving targeting Fiscal Reforms for a Sustainable Economic Recovery Building on these reforms, there are opportunities to further strengthen social assistance targeting, thereby enhancing impact and containing costs. A set of inter-connected recommendations follows from the analysis presented in this chapter. The general thrust of these reform measures would be to reduce poorly-targeted benefits and partially reallocate the savings to targeted programs. The proposed measures would generate net savings of about 0.1–0.15 percent of GDP, while adverse social impacts in terms of increased poverty gap would be neutralized since income shortfalls of the poorest deciles of the population would be offset by increases in targeted programs. If targeted social assistance is used to mitigate the adverse social impact of utility tariff increases, this would create additional fiscal cost of 0.3–0.7 percent of GDP, resulting an overall net increase in social assistance spending. However, this increase is more than offset by reduced energy subsidies, resulting in net savings. Rationalization of categorical privileges Categorical privileges can be rationalized. Eliminating, reducing or at most maintaining nominal values of untargeted benefits will generate fiscal savings. Privileges for World War II and Afghanistan war veterans as well as Chernobyl victims could be preserved. Initial cuts could target regressive privileges, while those with progressive benefit incidence could be maintained. A shift in privilege programs from rights-based to needs-based eligibility will generate fiscal savings in the range of 0.23 to 0.35 percent of GDP, depending on the scope of reductions. Table 4-3 illustrates the social and fiscal impact of two proposed measures: (i) full elimination of privileges, and (ii) elimination of poorly-targeted privileges. In both cases, benefits to war veterans and Chernobyl- affected people are preserved.47 Table 4-3. Projected impact of abolishing privileges Poverty impact Loss as percent of GDP (percentage points change) Total Official rate 1st Quartile Official rate 1st Quartile population Elimination of all privileges 0.41 1.13 0.35 0.02 0.09 Elimination of poorly-targeted privileges 0.16 0.66 0.23 0.01 0.04 Sources: HBS 2009, World Bank staff estimates.  poverty impact of the complete elimination of privileges is expected to result in a 0.4 percentage- 1. The point increase in poverty based on the official poverty line, and a 1.1 percentage-point increase for the population in the bottom income quintile. The income loss of the entire population is expected to be 0.35 percent of GDP, while the loss of the poor and vulnerable is expected to be less than 0.09 percent of GDP. 47 War veterans and Chernobyl-affected people represent approximately 20 percent of all programs beneficiaries. Chapter 4. Social assistance: improving targeting 67 Bel arus Public Expenditure Review 2.  Abolishing only ineffective, poorly-targeted programs would result in an income loss for the entire population of 0.23 percent of GDP, while the loss for the poor and vulnerable is expected to be less than 0.04 percent of GDP. Poverty is expected to increase by 0.16 percentage points based on the official poverty line and 0.66 percentage points among the first quintile of the distribution. Scaling up of targeted social assistance To offset negative social impacts of reductions in privileges, targeted social assistance could be scaled up. The proposed reduction in privileges would affect consumption levels of all beneficiaries, including those falling below the poverty line. In order to mitigate the adverse social impacts of these reforms, the benefits under the targeted social assistance program could be increased to offset the negative impact on household income of the poorest quintile of the population. As noted above, the elimination of privileges would result in an income loss equal to 0.09 percent of GDP among the population below the poverty. Based on the effectiveness of the current social assistance program, it is estimated that less than 0.2 percent of GDP is required to compensate low-income people for their loss due to the proposed abolishing of the subsidies and privileges. Half of this amount would likely be needed to compensate the poor and vulnerable for their loss in case of only ineffective poorly targeted programs are abolished. In addition, the adverse social impact of utility tariff increases proposed in Chapter 3 could be mitigated through targeted social assistance. The increases in utility tariffs would have direct income effects for all households in Belarus, including those below the poverty line. The social impact analysis of these reforms, which is presented in Chapter 3, indicates that raising utility tariffs to full cost recovery levels would result in an overall monetary loss to households in the range of 2 to 3 percent of GDP. The loss to the poorest 25 percent of the population is estimated at only 0.2–0.5 percent of GDP, a reflection of the highly regressive nature of the utility subsidies. Given the efficiency of the current social assistance program, a transfer of 0.3 to 0.7 percent of GDP is required to fully offset the negative social impact of the tariff increases for the poorest quarter of the population. This could be achieved through a heating allowance (flat amount) paid only to households with income lower than 1.5 time the monthly social benefit.48 Expand the income thresholds for child benefits Improving targeting of children-related benefits can be achieved by anchoring the benefits to household income. The efficiency of children-related benefits could be further improved by introducing income-related anchoring and making these already effective programs even more efficient in poverty reduction and support of the low-income population. Improved targeting is expected to reduce leakage of transfer expenditures to higher-income groups, thereby reducing program costs. 48 Similar system is in place currently in Romania and is being introduced in Moldova. 68 Chapter 4. Social assistance: improving targeting Fiscal Reforms for a Sustainable Economic Recovery The proposed reform envisions a redistribution of child-related benefits from the population with income above the country’s median to beneficiaries in the lowest quartile of the distribution. The total saved amount obtained from reducing benefits to higher income households is redistributed to households receiving similar children-related benefits in the lowest quartile of the income distribution. The simulation takes into consideration the leakage of the proposed program, thus the benefits of the rich will not fall to zero following the redistribution.49 The proposed reform is fiscally neutral, but poverty is expected to fall by 1 percentage point, meaning close to 100,000 people will move above the official poverty line (Table 4-4). The decrease in poverty would be 30 percent higher for the lowest quintile of the income distribution instead of the official poverty threshold. Overall, 0.2 percent of the GDP would be redistributed from higher to lower income groups in the population. Table 4-4. Projected impact of proposed reform in children-related benefits Poverty impact Loss as (percentage point change) percent of GDP Official Rate 1st Quartile Total Redistribution of Children’s Allowances from upper 50p to -1.01 -1.31 0.2 lower 25p Source: HBS 2009, World Bank staff estimates. In sum, the proposed reforms would result in a slight increase in the fiscal cost of social assistance spending (primarily as a result of mitigating the impact of utility tariff increases) with positive results in reduced poverty. The social and fiscal impacts of the proposed measures are summarized in Table 4-5. Table 4-5. Fiscal and social impact of proposed reform measures Reform option Fiscal impact Social impact Further rationalization of categorical privileges: Savings of 0.23-0.3 0.16 percentage Eliminate or scale down untargeted benefits to generate fiscal savings. percent of GDP point increase in Initial cuts target privileges with regressive nature, while those with poverty headcount progressive benefit incidence are maintained. Privileges to World War II and Afghanistan war veterans as well as Chernobyl victims are preserved. Expand income thresholds for child benefit programs: Neutral 1 percentage point Redistribute child-related benefits from the population with income reduction in poverty above the country’s median to beneficiaries in the low quintile of the headcount distribution to improve targeted efficiency and reduce program leakages. Scale up targeted social assistance: Total costs of 0.43- 0.16 percentage (i) Compensate reductions in privileges through targeted social 0.84 percent of GDP point reduction in assistance. Costs of 0.13 and poverty headcount (ii) Compensate utility tariff increases through targeted social 0.14 of GDP assistance. Costs of 0.3 to 0.7 percent of GDP 49 The following four programs were included in the simulation: birth grants, parental care grants, child allowances for children less than 3 years old and child allowances for children age 3 or older. We did not take in the consideration children-related survival benefits and sick and disabled childcare allowances. Chapter 4. Social assistance: improving targeting 69 Bel arus Public Expenditure Review Chapter 5. Pensions: Fostering sustainability D uring the last decade the pension scheme of Belarus has benefited from a favorable demographic environment, but these conditions are now changing. The previously favorable demographics resulted in a relatively balanced budget and high pension benefits. However, the aging population means that the pension system will face growing financial challenges. The proportion of working age and retired people is already starting to reverse, and the population dependency ratio (the number of old age people for each 100 working age people) is projected to double from 57 in 2008 to 115 in 2050, assuming no change in fertility rates. As a consequence of these changes, the financial situation of the pension system is projected to deteriorate. The pension fund is project to move into a structural deficit as early as 2014, which is expected to grow steadily and reach 5 percent of GDP in four decades. In addition, the system would deliver increasingly lower income replacement rates for future pensioners while keeping contribution rates at current levels, increasing labor costs and undermining the competiveness of the economy. The key challenge for the pension system is how to secure long-term fiscal sustainability given the projected demographic changes, while allowing for fiscally responsible reduction in contribution rates. The chapter outlines specific reform options in this regard. First, effective average retirement ages could be raised. This may be socially feasible as an extremely high proportion of young retirees, especially those retired under the early retirement provisions, continue to actively participate in the labor market. Second, in order to further reduce costs and free up fiscal space for reductions in the contribution rate, a shift of pension indexation rules to inflation as opposed to wages could be considered. The chapter also examines positive and negative aspects of converting the current pension scheme into a system of individual notional defined contribution accounts, thereby linking benefits more closely to individual contributions. 70 Fiscal Reforms for a Sustainable Economic Recovery A. Long term fiscal trends in the pension system Historically, pension system finances in Belarus have been relatively stable, reflecting favorable demographic conditions. The Social Protection Fund (SPF) has registered small surpluses of 0.5–1 percent of GDP for most of the 2000s. Low birth rates and high mortality for the cohorts born 1940–1946 have meant very low numbers of pensioners and high numbers of contributors. Prior to 2008 the size of each new 22 year-old cohort was larger than the last one, bringing increasingly higher numbers of new contributors to the pension plan. In 2008, around 3.6 million individuals below retirement age paid contributions to the fund each month. In addition, there were 0.56 million working (and contributing) pensioners. On the other hand, about 1.7 million fully retired individuals received work-related pensions, in addition to 126,000 survivor pensions. However, demographic trends will begin to reverse sharply in the near future. The demographic environment is going to worsen substantially over the next 40 years (Figure 5-1). The population dependency ratio (i.e. the number of old age people for each 100 working age people) is projected to steadily increase, doubling from 57 in 2008 to 115 in 2050, assuming no change in fertility rates. This projected worsening of the demographic situation is mostly dictated by the size of the age cohorts alive today. Even if policies to increase the fertility rate would be successful, this would start relieving demographic pressures no earlier than in four decades and the dependency rate would peak at double its current level in 2050 (Figure 5-2). Figure 5-1. The demographic environment will change dramatically 2008 2075 96 96 88 88 80 80 72 72 64 64 56 56 48 48 40 40 32 32 24 24 16 16 8 8 0 0 Male Female Male Female Source: World Bank staff projections using the Pension Reform Options Simulation Toolkit (PROST). Note: The right hand side graph shows the projected shape of the population pyramid in the year 2075. The outer shape of the right hand side graph presents the situation where fertility rates gradually increase from 1.26 children per woman recently to the replacement rate of 2.06 children per woman by the end of the simulation horizon, which would eventually stabilize the size of the population of Belarus. The inner shape of the pyramid demonstrates how much the population of Belarus would contract if the fertility rates stay at the current level until the year 2075. Mirroring the aging of the population, the beneficiary-to-contributor ratio will increase dramatically. The number of contributors will decline while the number of pensioners will start to grow as the post- war baby boomers (1950/60s cohorts) reach retirement age. In 2020, the 60-year-old cohort will be 40 percent larger than it was in 2008. At the same time declining birth rates since the 1980s will mean that the numbers of contributors will fall. By 2020, the 22-year-old cohort will be half of its current size.50 50 This estimate ignores the effects of emigration, which may compound the problem even further. Chapter 5. Pensions: fostering sustainabilit y 71 Bel arus Public Expenditure Review Figure 5-2. The population dependency rate will  changing the ratio of contributors and … increase steadily… beneficiaries in the pension system number of old-age people for each 100 working age people 150 5,000 125 3,000 100 1,000 75 50 -1,000 2008 2018 2028 2038 2048 2058 2068 2078 2008 2018 2028 2038 2048 2058 2068 2078 Base case (increasing fertility, decreasing mortality) Constant fertility, decreasing mortality Old age Disabled Survivors Constant fertility & mortality All contributors Contributors excl. working pensioners Source: World Bank staff projections using PROST. The number of pension beneficiaries is projected to surpass the number of contributors by 2033. Without countervailing reforms to the pension system, this demographic trend will move the SPF into a structural deficit.51 Contribution revenue dedicated to the pension system is expected to stay relatively stable at around 8.5 percent of GDP, while expenditures will start to rise relentlessly in the next few years. An eventual reversal of this trend around 2050 is conditional on improved fertility rates (see below). At the peak of the fiscal strain around 2050, the pension system deficit is expected to reach 5-6 percent of GDP (Figure 5-3). Figure 5-3. The pension system will move into  …and replacement rates will decline deficit… percent of GDP in percent 16 45 14 42.5 12 10 40 8 37.5 6 35 4 2 32.5 0 30 2000 2010 2020 2030 2040 2050 2060 2070 2008 2018 2028 2038 2048 2058 2068 2078 Contribution revenue Pension expenditure Source: World Bank staff projections using PROST. Moreover, the replacement rate—defined as average pension divided by average wage—is expected to steadily decline. The average wage incorporates social transfer payments, which are not subject to social security tax. The replacement rate would be around 10 percent lower by the end of the period. Macroeconomic volatility of recent years is projected to be partially transmitted into the replacement rate in the next decade, while lengthening period of wage base assessment is expected to contribute 51 This assumes unchanged retirement ages of 55 and 60 for men and women, respectively, for the entire period. 72 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery to the reduction of the replacement rate in the medium to long run (Figure 5-3). Another Box 5-1. PROST projections: Key important factor contributing to the decline macroeconomic and demographic of the average replacement rate is the policy assumptions of price indexation of minimum pension guarantee level. This would lead to a gradual  he projections presented in this chapter T decline of minimum pension compared to use the macroeconomic assumptions for prevailing average wage with minimum pension real GDP and wage growth. Longer-term guarantee benefiting ever smaller proportion of assumptions allow for the gradual decrease pensioners. in real wage growth from 5 percent to 3.5 percent. However, even at this relatively optimistic wage growth rate, GDP growth would not exceed 3.5 percent in the outer B. The pension system—Current system years due to the contraction in the size of the working age population. GDP growth rate is features and reform options exogenous in the model, as the ratio of wage bill to GDP is held constant The pensions system in Belarus is a  nder the baseline scenario, the overall U conventional single pillar pay-as-you-go population is projected to decline from the 9.7 million in 2008 to about 7.3 million by (PAYG) scheme. The system is characterized by the end of the simulation period. At the same relatively low retirement age (55 for women and time the size of the population aged between 60 for men), wage indexation of benefits and 15 and retirement age of 55/60 is expected a benefit formula that serves redistributional to decrease by 42 percent, while the number purposes with a compressed pension grid and of population above pension age is expected weak linkages between individual contributions to increase by 19 percent. Fertility rates and benefits. Pension benefits are financed are assumed to gradually increase from through payroll taxes with a contribution rate 1.26 children per woman to the replacement of 29 percent, which is high by international rate of 2.06 children per woman by the end of standards and puts a substantial burden the simulation horizon, which would eventually on labor costs, thereby undermining the stabilize the size of the population of Belarus. competitiveness of the Belarusian economy. Mortality projections for the modeling period are taken from the World Bank’s Population Database and are consistent with the United Nations WPP 2002 demographic assumptions. Mortality rates are assumed The retirement age is low to decline and life expectancy at the age 60 and could be increased over the simulation period is expected to increase from 14 and 19 years to 23 and The current retirement age in Belarus is 28 for men and women respectively. This the lowest in the region and far below the estimate ignores the effects of emigration. retirement age of 65 years customary in the  ll projections are carried out using the World A OECD countries. In Belarus women retire at Bank’s Pension Reform Options Simulation age 55 and men at age 60, much lower than Toolkit (PROST). most countries in the region (Figure 5-4). The retirement age has not been adjusted to rising Chapter 5. Pensions: fostering sustainabilit y 73 Bel arus Public Expenditure Review life expectancy, leading to prolonged benefit periods that together with declining birth rates are the main source of fiscal pressure in the pension system.52 Moreover, the five year difference between retirement ages for men and women is especially difficult to justify, given that the life expectancy for women aged 55 is 23.5 years compared to only 14.5 years for men aged 60. In addition, the current system exhibits relatively lax early retirement conditions, which further add to the fiscal burden. Early retirees constitute about 4.2 percent of all pensioners, adding significantly to pension expenditures. Figure 5-4. The retirement age is low…. 66 64 62 60 58 56 54 52 50 BLR BIH RUS UKR MDA TKM KAZ TJK SRB LTU BGR ALB BOS HRV GEO MNE POL SVN AZE LVA SVK MKD ARM KGZ CSK EST UNK HUN TUR Men Women Source: World Bank, MoL. Note: The numbers quoted in the graph include old age and disability pensioners. Despite generous pension benefits, a large Figure 5-5. …but many pensioners continue to share of the pensioners in Belarus continues work to work. Like other countries Belarus limits 65–69, female 17% combined wage and pension income, but 60–64, female 30% at 130 percent of the average wage this 55–59, female 51% threshold is high, incentivizing continued workforce participation among pensioners (see Box 5-2 for a comparison with the EU). As 65–69, male 26% a result, about half of the newly retired men 60–64, male 47% and women (in the five year bracket following - 50,000 150,000 250,000 350,000 the statutory retirement age) continue to work Working pensioners Fully retired while receiving pension benefits (Figure 5-5). Source: World Bank, MoL. Note: Data include old age and disability pensioners. The numbers quoted in the graph Among early retirees an even higher share of (dark blue) represent percentage of working pensioners in total pensioners. 71 percent continues to work, underscoring the failure of the early retirement program to target people that are unemployable or of poor health. This demonstrates that a big portion of the population currently eligible for retirement benefits is able to work and is employable. 52 In addition, faced with changing demographic conditions and growing life expectancy, many countries around the world are raising their retirement age further. For example, the United States gradually increased its normal retirement age for social security from 65 to 66 between 2002 and 2009 and is increasing it again from 66 to 67 between 2020 and 2027. The United Kingdom announced in 2006 that they will gradually increase their state pension age from 65 in 2024 to 68 in 2046. Germany is gradually increasing its pension age from 65 in 2012 to 66 in 2024 and then to 67 in 2029. Denmark is increasing the age threshold for the public old-age pension from 65 in 2024 to 67 in 2027. Furthermore from 2025, the eligibility age will be directly linked to changes in life expectancy at age 60. Japan is increasing its age for access to the earnings-related component of its pension from 60 to 65 by 2025 for males and by 2030 for females. Increases in pension age that affect both men and women are being implemented in the Czech Republic, Greece, Hungary, Italy and Korea. 74 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery Box 5-2. Treatment of working pensioners in the EU countries In the majority of EU countries, early retirement programs do not allow or restrict combining wage and pension income (see table below). Exemptions are granted for low income retirees. In most EU countries, early retirement is viewed as a last resort option for people who are no longer employable. In contrast, early retirement in Belarus is viewed as a right based on past occupation or lifetime contribution to society. While disallowing or limiting accumulation of pension and wage income for early retirees is a sensible pension policy that helps to enforce statutory retirement ages, some countries have gone even further in trying to limit the accumulation of wage and pension for pensioners above regular pensionable age. Very few countries apply such policy and those that do view it as a more politically acceptable alternative to retirement age increase. However, such policies, especially if taken to more extreme levels, can deliver mixed results. For example, in an attempt to reduce pension expenditures in 2009, Latvia passed a law that cuts benefits of working pensioners by 70 percent. The immediate response was that 50 percent of the affected group quit the labor market. In this case, savings from the reduced benefits for the 50 percent of pensioners who decided to continue to work were likely matched by the loss of social insurance contributions and income tax revenue from the 50 percent who decided to quit. Overall, the fiscal effect was likely neutral, while the income of the affected group was greatly reduced. Early retirement Regular retirement Full pension Pension reduced Full pension Pension reduction allowed (percent of reduction) allowed (percent of reduction) Belgium (up to 100 percent income Belgium (up to 100 percent income Bulgaria Bulgaria tested) tested) France Denmark (30 percent income tested) Germany Denmark (30 percent income tested) Italy Germany (income tested) Estonia Greece (70 percent income tested) The Netherlands Spain (reduced if full time work) France Spain (reduced if full time work) Portugal Cyprus (taxed by SS contribution) Ireland Latvia Luxembourg (50 percent income Finland Italy Slovenia (100 percent income tested) tested) Slovakia Estonia (100 percent) Cyprus Sweden Greece (100 percent) Lithuania United Kingdom Czech Republic (100 percent) Luxembourg Ireland (100 percent) Malta Latvia (100 percent) The Netherlands Lithuania (100 percent) Austria Hungary (100 percent income tested) Poland Malta (100 percent) Portugal Austria (100 percent income tested) Romania Romania (100 percent) Finland Slovenia (100 percent income tested) Slovakia Sweden United Kingdom Source: EU MISSOC database (Social Protection Systems in Member States). Chapter 5. Pensions: fostering sustainabilit y 75 Bel arus Public Expenditure Review Given the low retirement age and the high labor force participation of pensioners, a rise in statutory retirement age would relieve fiscal pressure and is socially feasible. With the lowest retirement age in the region, Belarus arguably has room for increasing the time individuals spend in the labor force. A gradual increase in the retirement age would increase the number of contributors while the number of pensioners would fall, and those entitled to a benefit would receive it for a shorter period of time. This would increase pension revenues while containing expenditures thereby improving the balance of the pension fund. To demonstrate the impact on the pension system balance and benefit levels we model two options,53 both including the immediate elimination of the early retirement program: •• 63/63, women first: retirement ages for women are increased by six months per year until they reach parity with men at age 60 in 2020, after which retirement age is increased by six months per year for both genders until reaching 63 in 2026. •• 65/65, both genders together: retirement ages for women are increased by six months a year and those for men by three months a year until both genders reach age 65 in 2030. A gradual increase in the retirement age would relieve fiscal pressures and provide for increasing in pension benefits and—to a lesser extent—reducing contribution rates. Under the existing indexation regime of pensions and pension assessment base (100 percent to wage growth), the combination of increasing length of service and decreasing pension assessment base results in higher old age replacement rates by the end of simulation horizon by 5 to 10 percentage points (Figure 5-6). Consequently, the increase in replacement rates is accompanied by a significant improvement of the system’s fiscal balance (Figure 5-7). The reform would stabilize and even increase pension system surpluses for almost three decades and would open fiscal space for some reduction in pension contribution rates. However, without changes to benefit rules, the fiscal impact would be partially offset by growing pension benefits and in the long run the fiscal balance would still be expected to move into negative territory. Figure 5-6. Retirement age reform scenarios: Figure 5-7. Retirement age reform scenarios: Average pension/average wage Projected pension system balance in percent percent GDP % insured wage bill 55 6 21.9 50 4 14.6 2 7.3 45 0 0 40 -2 -7.3 35 -4 -14.6 30 -6 -21.9 2008 2018 2028 2038 2048 2058 2068 2078 2008 2018 2028 2038 2048 2058 2068 2078 Base case 63/63, women first 65/65, both genders together Base case 63/63, women first 65/65, both genders together Source: World Bank staff projections using PROST. Source: World Bank staff projections using PROST. Note: The right hand scale of Figure 5 7 indicates the size of the contribution rate change—decreased (if positive) or increased (if negative)—to put the PAYG system in fiscal balance. 53 Under both options, the earning profiles—average wage by age and gender—are assumed to be almost flat at ages 60-65 for both males and females, i.e., the average wage does not increase as people continue working until they reach the increased retirement ages. However, longer calculation periods reaching 35 years by 2029 contribute to decreasing overall pension assessment base. 76 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery Increasing the retirement age would not be expected to reduce employment opportunities for youth. It is sometimes argued that increasing the retirement age will lead to a reduction of employment opportunities for new entrants to the labor force and the currently unemployed. This “lump of labor” argument has been found unwarranted.54 The number of positions in the labor market is not fixed. Many studies suggest that increasing the proportion of working elderly does not crowd out the jobs of younger cohorts as their experience and skill sets are very different. These two groups of workers are not good substitutes. The creation of new firms often requires both skill sets, making young and older workers natural complements that can increase the productivity of the economy. Wage indexation of benefits has resulted in high pension benefits, but creates substantial spending pressures in the future Pension payment benefits are currently indexed to nominal wage growth, and as a result pension benefits and expenditure grew rapidly over the past decade. As a result of wage indexation, pensioners fully participated in the growth of real incomes and improved living standards experienced over the past decade, keeping replacement rates stable. However, going forward wage indexation will put further upward pressure on pension expenditures in an environment where demographic changes are expected to already threaten the fiscal sustainability of the pension system. A change to benefit indexation could generate substantial fiscal savings while preserving the purchasing power of pension benefits. Pension benefit indexation rules aim to avoid erosion of pension benefits. Stability of these rules is important for current and future pensioners as it can substantially reduce the uncertainty about their future income stream and allow better long-term planning. Therefore, there is a notable advantage to automatically index pension benefits based on observable macroeconomic indicators. Many countries have changed the indexation rule to link benefits to inflation rather than wage growth to contain pension expenditures. Inflation indexation helps maintain real purchasing power of pensions, but decouples pensions in payment from real income growth. Moving to price indexation has been a key step in pension reforms in many countries. To demonstrate the impact of price indexation on the fiscal balance and benefit level, we again model two different options: •• Benefit indexation to prices: Benefits for pensions in payment are adjusted for inflation starting in 2011. •• Benefit indexation to prices, combined with increase in retirement age: Benefits for pensions in payment are adjusted for inflation starting in 2011 combined with an increase in the retirement ages for women and men to reach age 65 in 2030. 54 For a recent empirical study of the impact of increased retirement age on youth employment covering 12 countries, see Gruber et al. 2009, Social Security Programs and Retirement Around the World: The Relationship to Youth Employment, NBER Working Paper 14647. The authors find no evidence that increasing the labor force participation of older persons reduces the job opportunities of young persons. Indeed the evidence suggests that greater labor force participation of older persons is associated with greater youth employment and with reduced youth unemployment. Chapter 5. Pensions: fostering sustainabilit y 77 Bel arus Public Expenditure Review Both options would generate consistent surpluses, thereby providing space for reducing contributions rates. Under both options pension expenditures are contained, generating consistent surpluses in the pension fund that would provide room for commensurate reductions in the contribution rate (Figure 5-8 and Figure 5-9). These surpluses are a direct result of assumed high real wage growth rates, which would push up contributions while real pension benefits would remain flat. Figure 5-8. Benefit indexation scenarios: Figure 5-9. Benefit indexation scenarios: Average pension/average wage Projected pension system balance in percent percent GDP % insured wage bill 45 6 21.9 40 4 14.6 2 7.3 35 0 0 30 -2 -7.3 25 -4 -14.6 20 -6 -21.9 2008 2018 2028 2038 2048 2058 2068 2078 2008 2018 2028 2038 2048 2058 2068 2078 Base case Price indexation only 65/65, plus price indexation Base case Price indexation only 65/65, plus price indexation Source: World Bank staff projections using PROST. Source: World Bank staff projections using PROST. Although moving to price indexation without increasing the retirement age will fiscally stabilize the pension system, it will substantially erode pension benefits. Projections show that when price indexation is applied without changes to the retirement age, pension fund surpluses are generated, but replacement rates decrease to around 20 percent. With the current retirement age, contribution periods are too short to accumulate pension rights and the discrepancy between growth rate of wages and growth rate of pensions would be applied over long retirement periods. While price indexation would maintain real pensions, their purchasing power would erode in relation to prevailing wages, which may be socially unacceptable. Combining price indexation of benefits with retirement age reform offers more balanced results. This option generates high surpluses of up to 4 percent of GDP, and in the long run replacement rates are comparable to those under base scenario, due to the higher average pension assessment base. However, in the short run, which under our assumptions is characterized by high real wage growth, replacement rates drop quite precipitously as wage growth outpaces growth in pension benefits. Limiting the difference between wage and benefit growth during the periods when both the economy and wages are growing fast will further enhance the reform package. Recent experience in Eastern European countries provides practical examples of how this can be done without reverting to old discretionary indexation practices. Hungary’s new pension law of 2009, for example, introduces general price indexation of benefits only when GDP growth is below 3 percent and gradually increases the indexation parameter to 50 percent price + 50 percent wage growth when GDP growth is above 5 percent. Serbia also introduced a similar law going into effect in 2012, according to which pensions will be indexed to prices when GDP growth is below 4percent, but will get an additional increase equal to GDP growth minus 4 percent when GDP growth is above 4 percent. These mixed indexation rules ensure 78 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery that pensioners continue to participate in improvement in living standards during periods of rapid real income growth. Reforms should aim for a fiscally sustainable reduction in contribution rates Generous pension benefits are financed through payroll taxes with a relatively high contribution rate of 29 percent. The contribution rate is charged on the insured wage bill, which is intended to cover all pension benefits including old age, disability and survivor insurance. Among the transition economies, only Romania, Hungary and Ukraine have a higher contribution rate than Belarus, which charges roughly twice as much as in Bulgaria (Figure 5-10). Pension reforms should therefore aim not only to resolve emerging aging-related fiscal imbalances but also to provide space for a fiscally responsible reduction in the contribution rates to lower the tax burden on labor and boost competitiveness of the Belarus economy. Figure 5-10. Belarus’ pension contribution rate is high in regional comparison 35 30 25 20 15 10 5 0 BGR MKD HRV MNE SRB ALB SVN MDA RUS LVA LTU POL CSK SVK BLR ROU HUN UKR Source: World Bank. We estimate the effects of one possible package that would reduce the contribution rate while maintaining benefit adequacy and balancing the pension fund. The package includes:  •• elimination of early retirement option; ••  increase of retirement ages for women by six months a year and for men by three months a year until both genders reach age 65 in 2030; ••  indexation of pension benefits to the higher of i) consumer price increase or ii) nominal wage growth minus 3 percent; and  •• gradual decrease of pension contribution rate from 29 percent in 2010 to 24 percent in 2015. These reforms would simultaneously provide fiscal space for decreasing social security contribution rate while maintaining adequate benefits and sustaining the pension system (Figure 5-11 and Figure 5-12). The sustainability goal is reached by eliminating early retirement and a gradual increase of Chapter 5. Pensions: fostering sustainabilit y 79 Bel arus Public Expenditure Review retirement age. All else being equal, contribution revenue will increase and pension expenditure will decrease. In addition, the indexation of benefits would maintain purchasing power of pensioners in the long run, while generating consistent surpluses needed to decrease the contribution rate almost immediately without significantly worsening the fiscal balance of the system. Figure 5-11. Reform package scenario: Figure 5-12. Reform package scenario: Average pension/Average wage Projected pension system balance in percent percent GDP % insured wage bill 50 6 21.9 40 4 14.6 2 7.3 30 0 0 20 -2 -7.3 10 -4 -14.6 0 -6 -21.9 2008 2018 2028 2038 2048 2058 2068 2078 2008 2018 2028 2038 2048 2058 2068 2078 Price indexation only Reform package Base case Reform package Source: World Bank staff projections using PROST. Source: World Bank staff projections using PROST. Redistribution through the pension system has contributed to income equality, but undermined the link between individual contributions and benefits The benefit formula implies a heavily compressed pension grid. The key components of the old-age pensions benefit formula are length of service and wages earned over a specified reference period. In 2010, the reference period was the last 26 years of service, of which wages earned during any 16 subsequent years are used to compute the pension assessment base (PAB). The length of the reference period as well as the calculation period is increased each year by one year until the calculation period reaches 35 years in 2029. The old-age benefit formula grants 55 percent of the individual PAB for the first 25 years of service for males and 20 years of service for females. Old-age pensioners accrue an additional 1 percent of PAB for each year of service above 25 years for men and 20 years for women, up to a limit of 20 percent of PAB. The benefit is reduced proportionally for pensioners with fewer years of service, but at least five years of service are required to get a pension. Furthermore, the reduced benefit cannot fall below 50 percent of the legal minimum pension. Finally, pension averages for higher earners are increasingly suppressed in order to ensure equality in pension benefits. The effect of this procedure results in the average PAB falling to around two-thirds of the average wage. As a result, the pension system in Belarus is characterized by a significant degree of redistribution. For example, an individual earning 0.5 of average wage can expect a pension equal to 0.3 of average wage, while an individual earning twice the average wage can expect a pension equal to 0.5 of average wage (left side, Figure 5-13). As well, lower income pensioners have higher effective replacement rates than those with higher incomes (right side, Figure 5-13). For example, minimum wage workers can expect 95 percent of their pre-retirement income replaced after retirement, while workers earning twice the average wage can only expect the replacement rate of 25 percent. 80 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery Figure 5-13. Income redistribution within the pension system gross relative pension level gross replacement rate 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 individual earnings, proportion of average earnings Source: World Bank staff projections, APEX model. Note: Numbers are for individuals with full careers. Put in international perspective, the degree of redistribution in Belarus pension system stands out. Gini coefficients—which measure inequality, with 1 being the most unequal and 0 being completely equal—for Belarus show that the degree of equality among pensioners (column 1, Table 5-1) is much higher than the degree of equality among workers (column 3, Table 5-1). The progressivity index takes the value of 100 for pension systems with completely flat benefits and a value of 0 for pure insurance schemes with no redistribution. Both Gini and progressivity measures suggest that Table 5-1. International comparison: Degree of the degree of redistribution within the redistribution within the pension system Belarus pension system is relatively high Progressivity and stands significantly above the OECD Pension Gini index Gini wage average. Sweden 55 26.4 -14.4 23.1 Hungary 33.6 0 33.6 Through these redistributive properties, Italy 23.3 1.8 23.7 the pension system has provided old-age Poland 29.2 4.1 30.5 income security and contributed to low Netherlands 24.3 5.7 25.7 poverty. Pension benefits are generous Finland 22.6 5.9 24 Spain 25.7 17.1 31.1 and fairly equally distributed in Belarus Germany 19.8 24.7 26.3 and as a result pensioners, on average, Norway 13.6 38.1 22 are less poor than non–pensioners. Using Japan 14.3 46 26.4 the national poverty rate, about 4 percent United States 16.1 50.8 32.7 of the retired population falls below the Belgium 10.2 52.6 21.6 poverty line compared with 6 percent for Czech Republic 8.8 65.5 25.5 Korea 10.2 65.5 29.6 the entire population. However, pockets of Belarus 11.2 67.5 34.5 poverty exist and close to 25 percent of Australia 8.1 70.1 27.2 the country’s poor are pensioners. United Kingdom 5.1 82.4 28.9 Ireland 0 100 29.6 New Zealand 0 100 27.7 OECD 18 16.2 39.8 27.2 Source: Pensions at a Glance 2009: Retirement-Income Systems in OECD Countries. OECD, 2009; and World Bank staff estimates for Belarus. 55 Negative progressivity index reflects regressive benefit formula in Sweden as low and high income workers receive higher effective replacement rates than middle income workers. Chapter 5. Pensions: fostering sustainabilit y 81 Bel arus Public Expenditure Review However, the pension benefit formula has undermined the link between individual contributions and benefits. To strengthen the link between contributions and benefits the introduction of notional defined contribution (NDC) accounts is another reform option considered by policy makers in Belarus. The main design features of NDC retirement accounts are as follows: •• Benefit formula. The retirement benefits paid from the NDC pension system are calculated based on an affiliated worker’s accumulated lifetime contributions plus interest. This notional interest is usually determined by the growth of pension scheme revenue or insured wage bill. At the time of the retirement “notional account” is customarily converted into the annuity based on the observed life expectancy of current pensioners. •• Financing mechanism. Just as in the conventional PAYG schemes, the benefits are financed by contributions from currently insured workers. A buffer fund is usually recommended to manage surpluses and deficits that arise when current revenues do not match current expenditures. The buffer fund sometimes may have to be quite large to be able to absorb prolonged periods of pension deficits. This financing mechanism is fundamentally different from funded defined contribution accounts, where individual contributions are invested into the assets of contributor’s choice and then sold at the time of retirement. Table 5-2. Advantages and disadvantages of notional defined contribution account systems Advantages Incentives to participate. NDCs mimic the functioning of a simple savings account and are much more intuitive to understand than many conventional PAYG systems. This should, at least theoretically, strengthen the incentives of people to contribute to the system and to postpone their retirement. However, experience suggests that these incentives are not sufficient to noticeably expand pension system coverage or increase effective retirement age. Most likely this lack of behavioral changes is due to very high personal discount rates that people apply in their daily financial decisions. Reduced government involvement. By introducing the NDC system the government gives up its ability to influence benefit levels and fiscal situation of the pension scheme. The two major parameters governing pension system— notional interest rate and annuity factors—are normally based on observable statistics, like growth of wage bill and life expectancy of the elderly. This might be viewed as a positive feature, given how politically difficult it is to adjust pension parameters like retirement age, contribution rate or benefit formula in conventional PAYG system. Disadvantages Lack of redistribution. NDCs function like savings accounts, constraining redistribution from high to low income individuals, although explicit government contributions to individual accounts of subsidized worker categories are possible. Without any redistribution, members with short careers or low wages are very likely to experience extremely low benefit levels and may need to rely on additional poverty prevention mechanisms in retirement. Risk of volatility. Notional interest rate can experience significant fluctuations, especially in rapidly changing demographic, macroeconomic and institutional environments. This volatility can induce wide differentials between the replacement rates of two generations born a few years apart. Such volatility could arise from sharp or prolonged shocks to fertility rates, changes to the degree of labor market formalization, effective retirement age, wage growth, inflation level and pension contribution rates to NDC accounts. Complicated transition rules. Introduction of NDCs requires well thought-through transitional arrangements. Will all current workers be “converted” into the new system? If so, what is a fair and fiscally sustainable conversion formula? If not, what is the cut-off age? Will the chosen cut-off age introduce different outcomes for two cohorts born one year apart? 82 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery Table 5-2 summarizes the principal advantages and disadvantages of NDCs, while Box 5-3 discusses international implementation experience. Box 5-3. International experience with notional accounts Latvia (NDCs since 1996). The real notional interest rate, tied to the growth of wage bill, averaged 17 percent between 2005-2009 due to rapid formalization of economy, increasing retirement age and high wage growth. This led to 69 percent growth in average newly assigned pension from 2005 to 2009, compared to real wage growth of 28 percent during the same period. Instead of building up a buffer fund from 2005-2009 surpluses, most resources were used to introduce additional pension supplement and increase generosity of pension indexation, in part to soften the effects of replacement rate volatility. Due to the financial crisis of 2009, financing of the pension system burdened with new pension supplements became difficult and deficits emerged earlier than planned. Negative real notional interest rates are foreseen in the next three years, continuing the experience of volatile replacement rate outcomes. Kyrgyzstan (NDCs since 1997). The concept of NDCs was distorted at implementation as no interest is paid on past contributions, resulting in low replacement rates and reduced incentives to contribute. This was partially due to low administrative capacity and the absence of reliable system of records. The system uses an ad hoc annuity mechanism not related to observable life expectancies. Russia (NDCs since 2002). The notional interest rate in Russia is tied to the revenue growth of the NDC component of pension system. As the contribution rate has decreased and an increasing portion of pension contributions was diverted to the funded pension component, the revenue stream to the NDC scheme has slowed compared to wage growth. This resulted in comparatively low notional interest rates and, eventually, pensions. To remedy the situation, the basic pension was increased in an ad hoc manner to raise pensions from 26 percent of average wage to 40 percent in 2010. Such a large basic pension component completely undermines the strong link between contributions and benefits that NDC reform was attempting to introduce. Moreover, without the increase in retirement ages or further increases in basic pension, the system is projected to provide extremely low replacement rates in the future. In terms of fiscal impact, the NDC system Figure 5-14. Notional accounts: is attractive because it is self-balancing in Average pension/average wage the long run once stable macroeconomic in percent 45 and demographic conditions are reached. Under our assumptions for Belarus system 40 balance occurs around 2048 (Figure 5-15). 35 However, without an increase in retirement 30 ages the balance is achieved at the expense 25 of the low replacement rates in the range of 20 20-25 percent (Figure 5-14). Replacement 15 rates in the medium term could be improved 2008 2018 2028 2038 2048 2058 2068 2078 by adjusting the mechanism of accrued right Base case Notional Source: World Bank staff projections using PROST. Chapter 5. Pensions: fostering sustainabilit y 83 Bel arus Public Expenditure Review Figure 5-15. Notional accounts: conversion, but that would still not guarantee Pension system balance fiscal balance for the next four decades. It percent GDP % insured wage bill would be tempting to use projected medium- 6 21.9 term surpluses to lower the contribution rate, 4 14.6 but that would reduce replacement rates in the 2 7.3 long run even further. 0 0 -2 -7.3 However, the introduction of NDCs would -4 -14.6 reduce replacement rates. Under NDCs, the balancing of the fund would effectively -6 -21.9 2008 2018 2028 2038 2048 2058 2068 2078 be achieved by reducing pension liabilities, Base case Notional matching them to accumulated contribution at Source: World Bank staff projections using PROST. Note: The estimates presented here rely on a conversion of accrued rights in such a the time of retirement. While this balances the way that differences between pensions of new pre-reform and post-reform retirees are minimized. Furthermore, from the start of the reform all pensions are assumed to be indexed to prices, as otherwise initial pensions determined in an actuarially fair fashion pension fund, the average replacement rate would be extremely low in the long run. would drop sharply to less than 30 percent over the next two decades (Figure 5-14). By linking individual benefits to contributions, NDCs would also eliminate the redistributional properties of the pension system. In addition to reducing the average replacement rate, by linking pension benefits to individual contributions, NDCs would also eliminate the redistribution that occurs as a result of the current pension benefit formula. Under NDCs, pension benefits would reflect differentials in earnings while replacement rates would be flat and independent of income levels (Figure 5-16). The introduction of NDCs would therefore be expected to be harmful to poor pensioners and likely have adverse impacts on poverty rates among pensioners. Figure 5-16. Income redistribution under notional accounts gross relative pension level gross replacement rate 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 individual earnings, proportion of average earnings Gross pension level - PAYG Gross pension level - NDC (35 years of contributions) Sources: MoF, MoL, ECA regional database. World Bank staff estimates. C. Policy options and their fiscal and social impact The key challenge for the pension system in Belarus is how to secure its long-term fiscal sustainability given projected demographic changes, while allowing for a fiscally responsible reduction in contribution 84 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery rates. The chapter has provided a menu of reform options that would affect the long-term sustainability of the pension fund. Specific reform options include:  •• Increase the retirement age.  •• Change benefit indexation from wage to price. ••  Introduce a package of parametric reforms designed to allow for a fiscally responsible reduction in contribution rates, comprising: i) eliminate early retirement; ii) increase retirement ages for women by six months a year and for men by three months a year until both genders reach age 65 in 2030; iii) index pension benefits to the higher of consumer price increase or nominal wage growth minus 3 percent; and iv) gradually decrease the pension contribution rate from 29 percent in 2010 to 24 percent in 2015. ••  Introduce notional defined contribution accounts to strengthen the link between individual benefits and contributions. The fiscal savings generated by the different Figure 5-17. Fiscal impact of different reform reform options vary significantly (Figure measures 5-17 and Table 5-3). In the medium run, all percent of GDP % of insured wage bill 6 21.9 options would mitigate the adverse impact of the changing demographic environment, with 4 14.6 substantial improvements over the expected 2 7.3 deterioration of the fund balance under the 0 0 base case. In fact, all four options would -2 -7.3 generate significant surpluses to the pension -4 -14.6 fund, providing fiscal space for possible rate -6 -21.9 reductions. Estimated fiscal savings (the 2008 2018 2028 2038 2048 2058 2068 difference between base case and the reform Base case 65/65 plus price indexation NDC Price indexation only 65/65 Reform package scenario) over the next decade range between Sources: World Bank staff estimates. Table 5-3. Projected medium term fiscal savings of different pension reform options (percent of GDP) Period 2012 2013 2014 2015 2016 2017 2018 2019 2020 Average Reform package 0.3 % 0.5 % 0.6 % 0.8 % 1.2 % 1.7 % 2.1 % 2.5 % 2.9 % 1.3 % 65/65, both genders 0.7 % 1.0 % 1.4 % 1.7 % 2.0 % 2.3 % 2.6 % 2.8 % 2.9 % 1.8 % together Price indexation only 0.3 % 0.8 % 1.3 % 1.7 % 2.2 % 2.5 % 2.8 % 3.1 % 3.3 % 1.8 % NDC 0.6 % 1.2 % 1.7 % 2.3 % 2.8 % 3.3 % 3.7 % 4.0 % 4.3 % 2.4 % 65/65 plus price 1.0 % 1.8 % 2.6 % 3.3 % 4.0 % 4.6 % 5.1 % 5.5 % 5.8 % 3.4 % indexation Source: World Bank staff projections. Chapter 5. Pensions: fostering sustainabilit y 85 Bel arus Public Expenditure Review 1.3 and 3.4 percent of GDP on average annually. The increase of the retirement age to 65 years for both men and women combined with indexing pension benefits to inflation would generate the most significant fiscal savings, followed by the introduction of notional accounts. The options do vary in terms of their long-term impact. Most importantly, an increase in the retirement age, while preserving the balance over the medium term, would still imply a deterioration of the balance in the long run. A rate reduction would become affordable if the proposed saving measures are implemented. The savings generated by increasing the retirement age and changing indexation rules would make rate reductions affordable. The reform package modeled here assumes a 5 percent reduction in the pension contribution rate. The resultant shortfall in revenue would be more than offset by longer contribution periods and shorter retirement periods, with lower benefits implied by a later retirement and slower growth of pension benefits under price indexation. Even with a rate reduction of 5 percent, the fund would be expected to remain in surplus for the coming four decades and only then moderate deficits would emerge. The reform options also have differing impacts Figure 5-18. Impact of different reform measure on replacement rates. Changes to indexation on average replacement rates and the introduction of notional accounts would 60 balance the pension fund but erode pension 50 benefits relative to wage earnings (Figure 40 5-18). Among the options presented here, only 30 the increase in retirement age would have a 20 positive effect on pension replacement rates, 10 allowing them to remain at current levels if 0 combined with other options. Furthermore, by 2008 2018 2028 2038 2048 2058 2068 2078 allowing pension benefits to grow more rapidly Base case 65/65 plus price indexation NDC Price indexation only 65/65 Reform package during periods of high wage growth, as we Sources: World Bank staff estimates. model under the reform package, the wedge between pension benefits and real incomes can be contained, enhancing the social and political feasibility of proposed pension reforms. It is therefore important to structure a pension reform package that balances the objectives of fiscal sustainability, rate reductions and pension adequacy. While the specific reform options presented here represent only some of many possible scenarios, they indicate that the elimination of early retirement and a gradual increase of retirement age would increase contribution revenue while containing pension expenditure. In addition, the indexation of benefits (to nominal wage growth minus 3 percent in periods of high wage increases and to inflation otherwise) would maintain purchasing power of pensioners in the long run, while generating consistent surpluses needed to decrease contribution rate almost immediately without significantly worsening the fiscal balance of the system (Table 5-4). 86 Chapter 5. Pensions: fostering sustainabilit y Fiscal Reforms for a Sustainable Economic Recovery Table 5-4. Fiscal and social impact of proposed reform measures Policy Options Estimated Fiscal Savings Estimated Social Impact Positive impact on replacement rates 1st year 0.7 % of GDP 1st year 0 % 2nd year 1.0 % of GDP Increase the retirement age 2nd year 0 % 3rd year 1.4 % of GDP 3rd year 0 % 10th year 2.9 % of GDP Change benefits indexation Negative impact on replacement rates 1st year 0.3 % of GDP 1st year -0.3 % 2nd year 0.8 % of GDP Price indexation only 2nd year -1.8 % 3rd year 1.3 % of GDP 3rd year -4.2 % 10th year 3.3 % of GDP 10th year -14.5 % 1st year 1.0 % of GDP 1st year -1 % 2nd year 1.8 % of GDP 2nd year % 65/65 plus price indexation 3rd year 2.6 % of GDP 3rd year -4.2 % 10th year 5.8 % of GDP 10th year -14.5 % 1st year 0.3 % of GDP Negative impact on replacement rates 2nd year 0.5 % of GDP Package of reforms Short term -3.6 PP 3rd year 0.6 % of GDP Long term -0.8 PP 10th year 2.9 % of GDP 1st year 0.6 % of GDP Negative impact on replacement rates 2nd year 1.2 % of GDP Notional accounts Short term -10.4 PP 3rd year 1.7 % of GDP Long term -14.6 PP 10th year 4.3 % of GDP NDC will remove the redistributional properties of the pension system, which would negatively impact pension levels of poorer pensioners and positively the richer pensioners (other things being equal). Chapter 5. Pensions: fostering sustainabilit y 87 Bel arus Public Expenditure Review List of References Government of Belarus. (Ministry of Energy). 2010. State Program of the Development of Belarusian Energy System 2011–2015. Government of Belarus. Decree No. 1699, December 26, 2009. Government of Belarus. Decree No. 72, February 25, 2011. Government of Belarus. Resolution No. 752, June 8, 2009. Gruber et. al. 2009. Social Security Programs and Retirement Around the World: The Relationship to Youth Employment. NBER Working Paper No. 14647. International Monetary Fund. 2007. Manual on Fiscal Transparency. Republic of Belarus. Law No. 1898-XII, October 30, 1992. Republic of Belarus. Law No. 3563-XII, January 31, 1995. Republic of Belarus. Presidential Decree No. 458, September 14, 2009. Serova, E. 2009. Agricultural Policy Framework and State Support for Agriculture in the Republic of Belarus. FAO. World Bank. 2005. Belarus Country Economic Memorandum: Window of Opportunity to Enhance Competitiveness and Sustain Economic Growth. Report No.32346-BY, November. World Bank. 2009a. Belarus: Public Expenditure and Financial Accountability (PEFA). Public Financial Management Assessment. Report No.48239-BY, April. World Bank. 2009b. Belarus: Agricultural Productivity and Competitiveness. Impact of State Support and Market Intervention. Report No. 48335-BY. June. World Bank. 2010. Belarus: Industrial Sector Performance and Trends Before and During Global Crisis. Belarus Economic Policy Note No.1. Report No. 54371-BY. 88 List of References Fiscal Reforms for a Sustainable Economic Recovery Annexes ANNEX 1.1. Fiscal Indicators, 2005–2010 (in billion BYR) 2005 2006 2007 2008 2009 2010 2011 planned 2011 Q1 1. State (republican and local) budget Revenue 23,420 29,007 36,726 51,054 46,741 48,765 55,832 13,129 Personal income tax 1,882 2,480 3,081 4,183 4,305 5,381 6,149 1,681 Profit tax 2,377 3,141 3,837 5,993 4,608 5,580 6,549 2,223 VAT 5,909 7,365 8,670 11,399 12,083 16,226 19,138 3,493 Excises 1,368 2,826 3,046 3,901 3,637 4,350 5,371 900 Property tax 1,149 1,268 1,519 2,005 1,627 1,850 2,005 583 Customs duties 1,682 2,069 6,281 10,613 7,970 5,776 7,264 1,988 Expenditure (economic classification) 23,821 28,765 37,240 50,922 49,201 52,971 62,489 12,818 Wages and salaries 5,276 6,511 7,481 8,626 9,258 11,528 15,828 3,733 Social protection fund 1,473 1,795 2,041 2,347 2,471 3,079 4,343 1,014 contributions Goods and services 5,002 5,701 7,003 8,683 8,677 9,993 11,487 2,726 Interest 229 293 387 742 1,069 1,111 2,990 465 Subsidies and transfers 5,905 6,668 10,137 14,934 16,127 13,710 15,735 3,706 Capital expenditures 5,377 7,595 9,579 12,980 11,162 13,631 13,314 1,513 Net lending 559 203 612 2,609 437 (81) (2,623) (339) State Balance (400) 242 (515) 132 (2,460) (4,205) (6,657) 312 2. Social Protection Fund Revenue 7,405 9,288 11,422 14,716 16,196 19,677 20,691 5,737 Expenditure 6,735 8,491 10,386 12,996 14,694 18,377 20,494 5,259 Balance (cash) 669 797 1,035 1,720 1,502 1,300 197 477 3. General Government Revenue 30,825 38,294 48,147 65,770 62,936 68,442 76,523 18,866 Expenditure 30,556 37,256 47,627 63,918 63,895 71,348 82,983 18,077 Augmented General Government 269 1,038 521 1,852 (958) (2,905) (6,460) 789 Balance GDP, bn BYR 65,067 79,267 97,165 128,829 137,442 162,964 200,000 40,313 Annexes 89 Bel arus Public Expenditure Review 1. State (republican and local) budget Revenue 36.0 36.6 37.8 39.6 34.0 29.9 27.9 32.6 Personal income tax 2.9 3.1 3.2 3.2 3.1 3.3 3.1 4.2 Profit tax 3.7 4.0 3.9 4.7 3.4 3.4 3.3 5.5 VAT 9.1 9.3 8.9 8.8 8.8 10.0 9.6 8.7 Excises 2.1 3.6 3.1 3.0 2.6 2.7 2.7 2.2 Property tax 1.8 1.6 1.6 1.6 1.2 1.1 1.0 1.4 Customs duties 2.6 2.6 6.5 8.2 5.8 3.5 3.6 4.9 Expenditure (economic classification) 36.6 36.3 38.3 39.5 35.8 32.5 31.2 31.8 Wages and salaries 8.1 8.2 7.7 6.7 6.7 7.1 7.9 9.3 Social protection fund 2.3 2.3 2.1 1.8 1.8 1.9 2.2 2.5 contributions Goods and services 7.7 7.2 7.2 6.7 6.3 6.1 5.7 6.8 Interest 0.4 0.4 0.4 0.6 0.8 0.7 1.5 1.2 Subsidies and transfers 9.1 8.4 10.4 11.6 11.7 8.4 7.9 9.2 Capital expenditures 8.3 9.6 9.9 10.1 8.1 8.4 6.7 3.8 Net lending 0.9 0.3 0.6 2.0 0.3 (0.05) (1.3) (0.8) State Balance (0.62) 0.30 (0.53) 0.10 (1.79) (2.58) (3.33) 0.77 2. Social Protection Fund Revenue 11.4 11.7 11.8 11.4 11.8 12.1 10.3 14.2 Expenditure 10.4 10.7 10.7 10.1 10.7 11.3 10.2 13.0 Balance (cash) 1.0 1.0 1.1 1.3 1.1 0.8 0.1 1.2 3. General Government Revenue 47.4 48.3 49.6 51.1 45.6 42.0 38.3 46.8 Expenditure 47.0 47.0 49.0 49.6 46.5 43.8 41.5 44.8 Augmented General Government 0.41 1.31 0.54 1.44 (0.70) (1.78) (3.23) 1.96 Balance GDP, bn BYR 65,067 79,267 97,165 128,829 137,442 162,964 200,000 40,313 Source: MoF. 90 Annexes Fiscal Reforms for a Sustainable Economic Recovery ANNEX 1.2. Grounds for provision of state support Forms of state support* Grounds for provision of state support Deferral and (or) installment payment of ••risk of insolvency (bankruptcy) in case of a lump-sum payment of tax and taxes, fees, customs duties and interest interest arrears; ••delay of the budget funding including payment for the government order that has been completed; ••seasonal nature of production and (or) sale of goods (works, services); ••in case of damage caused as a result of a natural disaster, mad-made disaster or other force majeure Tax credit ••risk of insolvency (bankruptcy) in case of a lump-sum payment of tax arrears; ••seasonal nature of production and (or) sale of goods (works, services) Normative distribution of revenues ••difficult financial situation of a payer Financial support ••implementation of the budget programs expected to be funded from the republican budget; ••procurement of technological equipment and spare parts for the implementation of the investment projects that went through the comprehensive state expert examination and have received positive appraisal; ••implementation, following the decision of the President of the Republic of Belarus, of the activities of particular government significance; ••accomplishment of other goals determined by the President of the Republic of Belarus Provision of budget credits and loans ••formation of seasonal (temporary) of inventories in case of seasonal nature of production and (or) sale of goods (works, services); ••for the implementation of investment projects; ••for other purposes Reimbursement to legal entities of in case of the implementation of: the part of interest on the use of bank ••budget programs expected to be funded from the republican budget; credits ••investment projects awarded through tenders; ••energy efficient projects; ••business plans of the legal entity’s development Discounted prices (tariffs) on natural ••risk of insolvency (bankruptcy) if energy inputs are purchased at prices gas, electric and heat energy, petrol and (tariffs) which are set in accordance with the legislation diesel fuel Deferral and (or) installment repayment ••risk of insolvency (bankruptcy) in case of a demand of a lump-sum payment of arrears for consumed natural gas, of arrears for consumed natural gas, electric and heat energy electric and heat energy Source: Ministry of Economy. Note: *Forms are specified in the Decree of the President of the Republic of Belarus No. 182 dated 28 March 2006 Annexes 91 Bel arus Public Expenditure Review ANNEX 1.3. Progress in reducing of tax burden and tax system simplifications of, 2005–2011 Tax Measure 2005 Chernobyl tax (payroll tax) Rate reduced from 4% to 3% Local earmarked fees (transport fee, fee for maintenance Rate reduced from 5% to 4% of local infrastructure) (surcharge on profit) Two turnover taxes (Fee for earmarked local housing- Rate reduced from 1.15% to 0.9% investment funds and Fee for financing expenditures for maintenance and repair of housing stock) 2006 Two turnover taxes (Fee for earmarked local housing- Cancelled investment funds and Fee for financing expenditures for maintenance and repair of housing stock) Gasoline and diesel tax (10% of gasoline and diesel costs) Cancelled Tax on sales of motor fuel (20% of fuel price) Cancelled Local earmarked fees (transport fee, fee for maintenance Rate reduced from 4% to 3% of local infrastructure) (surcharge on profit) Tax on banks incomes (30% of revenues minus Cancelled expenditures) Local fee on individuals for crossing state border (no more Cancelled than 0.25 base rate) 2007 Excise on motor fuel Rates reduced on average by 2.27 times Chernobyl Tax (payroll tax) (3% of wage fund) Cancelled Mandatory contribution to State Employment Fund (payroll Cancelled tax) (1% of wage fund) Tax on dividends (15%) Cancelled Tax on incomes of insurance companies (30% of income Cancelled minus expenditures) Local fee for hunting and fishing (rates were set as Cancelled multiples of base rates) 2008 Special tax on incomes from operations with securities Rate reduced from 40% to 24% (40% of income minus expenditures) Tax on automobile roads users (1% of sales proceeds) Cancelled (turnover tax) Local users’ fees (surcharges on profit) Cancelled ••For conducting trade on the territory of the respective territorial-administrative unit; ••For construction on the territory of the respective territorial-administrative unit; ••For placement of outdoor advertising in foreign languages, as well as registered trademarks in foreign languages (rates rates were set as multiples of base rates) Contribution to the construction science development fund Cancelled (0.25% of cost estimates) 92 Annexes Fiscal Reforms for a Sustainable Economic Recovery 2009 Tax on purchasing of automobile transport vehicles Rate reduced from 5% to 3% Special tax on incomes from operations with securities Cancelled (24% of income minus expenditures) Simplified tax system Rates reduced: ••From 10% to 8% (w/o VAT); ••From 8% to 6% (with VAT); ••From 20% to 15% (gross revenue) Local sales tax Unified tax rate on sales of domestic and foreign goods at 5% (instead of earlier applied 5% and 15% respectively) Local tax on services Rate reduced from 10% to 5% Personal income tax Transition to flat tax rate of 12% (instead of progressive scale with the highest rate of 30%) 2010 Contribution to Agricultural Support Fund (turnover tax) Rate reduced from 2% to 1% Tax on purchasing of automobile transport vehicles (3% of Cancelled price of vehicle) Local sales tax (5% of sales) Cancelled Local user fees for parking in a specially equipped places Cancelled (rates rates were set as multiples of base rates) Value added tax Rate increased from 18% to 20% (to compensate for loss of revenues related to the cancellation of the taxes above) 2011 Contribution to Agricultural Support Fund (1% of sales Cancelled proceeds) (turnover tax) Local fee for territorial development (3% of profit) Cancelled (surcharge on profit) Local tax on services (5% of sales) Cancelled Environmental Tax Three payments within the tax are cancelled ••Oil processing fee ••Oil transit fee ••Fee on production and/or import of goods with 50% and more of volatile organic structures Fee for import on the territory of Belarus ozone-destroying Integrated into the Environmental Tax substances Fee for issuance of permits for transit of Belarusian Integrated in the State Duty transport vehicles via territory of foreign states Sources: WB Staff compilation on information from MoF, MoT, legal database. Annexes 93 Bel arus Public Expenditure Review ANNEX 2.1. Tax concessions in agriculture (2009, million BYR) Taxes and payments WITH Taxes and payments WITHOUT concession status concession status Estimate of tax Taxes and mandatory payments concessions Tax base Tax rate % Tax due Tax base Tax rate % Profit tax -52,946 24% - 2,247 -2,247 Property tax 48,453,638 1% 484,536 1,166 483,370 V.A.T 99,969,846 20% 883,554 -41,315 10% 924,869 Local taxes -52,946 max 3% - 553 -553 Payments to RAPS Fund and 9,969,846 1% 99,698 1284 98,414 Payments to Road Fund Innovation Fund 9,943,710 0.25% 11454 11454 0.25 0 Unified agricultural tax 196,093 2% -196,093 Payments to Social Protection Fund 288,879 34% 982,219 866,664 30% 115,555 TOTAL 2,461,461 1,038,146 1,423,315 Source: ISRA ANNEX 2.2. Forgone tax revenues from agriculture (2005–2009, million BYR) Share in total - Taxes/Payments 2005 2006 2007 2008 2009 2009, % Value-added tax 398,449 440,447 524,354 678,117 924,969 65% Property tax 272,229 336,964 394,810 428,958 483,370 34% Payments to RAPS Fund and 136,133 163,644 198,728 178,305 98,414 7% Payments to Road Fund Payments to Social Protection Fund 80,849 64,099 93,901 119,247 115,555 8% Local taxes and fees -1,149 -2,286 11,575 -553 0% Profit tax -9,218 -4,915 -3,325 126,848 -2,247 0% Unified agricultural tax -76,940 -91,401 -113,964 -174,379 -196,093 -14% Emergencies tax and fees into the 45,924 56,923 - - - - state employment fund TOTAL 847,426 964,612 1,092,218 1,368,671 1,423,415 100% Sources: World Bank (2009) for 2005-07 data; ISRA calculations for 2008-2009 data. 94 Annexes Fiscal Reforms for a Sustainable Economic Recovery ANNEX 2.3. Subsidies received and taxes paid by agricultural enterprises in Belarus (billion BYR) 2005 2006 2007 2008 2009 Revenues of agricultural enterprises 5,634 6,324 7,338 10,133 11,173 Costs of agricultural enterprises 6,646 8,058 9,421 13,193 14,365 Net Profits of agricultural enterprises 765 944 999 1,806 1,230 o/w Profits from sold products, goods and services 407 316 71 617 -22 Subsidies received by agricultural enterprises* 2,222 3,100 3,935 4,303 5,226 Taxes and fees paid by agricultural enterprises 227 228 226 267 261 Net subsidies received by agricultural enterprises** 1,995 2,872 3,709 4,036 4,965 Memorandum Share of net subsidies in revenues of ag enterprises, % 35 45 51 40 44 Share of net subsidies in costs of ag enterprises, % 30 36 39 31 35 Share of net subsidies in net profits of ag enterprises, % 261 304 371 224 404 Sources: WB Staff calculations on MoF, MoFA, Belstat and IRSA data. * Estimated as the difference between Total Budget expenditures on agriculture minus Green Box spending.  ** Calculated as subsidies minus taxes & fees.  Annexes 95 Bel arus Public Expenditure Review ANNEX 2.4. Composition of budget expenditures on agriculture based on WTO classification (billion BYR) 2008 2009 2010 Measures Republican Local Republican Local Republican Local Total budget budgets Total budget budgets Total budget budgets A. Green Box measures 387.6 212.5 175.1 402.2 242.1 160.1 625.9 425.2 200.7 Measures for the mitigation of 129.8 129.8 133.8 133.8 143.9 143.9 effects of Chernobyl accident Capital investments for construction of melioration 0.0 0.0 106.0 106.0 systems Compensation of crop losses 60.7 60.7 21.3 21.3 10.0 10.0 Measures for animal disease 0.0 0.0 13.4 13.1 0.3 control Agricultural research, education 38.2 38.2 86.6 86.6 102.5 100.1 2.4 and professional training Information system creation 3.6 3.6 2.5 2.5 8.3 8.3 Salary additions for young 0.0 1.0 1.0 9.9 9.9 professionals Carrying out of republican and 7.5 7.5 8.4 8.4 11.6 10.0 1.6 regional contests in agriculture Purchase of laboratory equipment for investigations on 0.0 3.0 3.0 7.8 7.8 animals, feed and products Re-equipment of maintenance 0.0 0.0 16.7 16.7 stations Measures on R&D, anti-flood and 75.9 75.9 80.1 80.1 105.4 83.2 22.2 soil fertility monitoring Other (budgetary organizations, 71.9 26.6 45.3 65.5 39.2 26.3 90.4 60.1 30.3 crop surveys), including: reduction in labor costs on potential bonding 0.0 0.0 0.4 0.4 measurement compensation of losses of veterinary & sanitation 0.0 0.0 0.5 0.5 disposal plants operation of State Technical Inspection, Data Processing 0.0 0.0 0.4 0.4 Center, etc. operation of budgetary 0.0 0.0 85.6 56.6 29.0 organizations expenditures for crop surveys 0.0 0.0 3.5 3.5 and agro-chemical services 96 Annexes Fiscal Reforms for a Sustainable Economic Recovery B. Amber Box measures 4,289.6 2,690.9 1,598.7 5,277.9 3,492.8 1,785.1 4,579.3 3,104.0 1,475.3 B1. Non-product specific 3,293.2 1,984.9 1,308.3 4,571.1 3,006.9 1,564.2 4,164.9 2,738.0 1,426.9 measures Soil fertility enhancement (mineral fertilizers, plant 690.6 597.3 93.3 1,117.4 839.5 277.9 1,121.8 783.8 338.0 protection agents) Subsidies to agricultural enterprises, peasant farms and agricultural consumption coops 0.0 56.8 56.8 0.0 for partial compensation of interest rates on to 2-to-10-year investment credits Increase of statutory funds of agro-food enterprises, purchase 0.0 3.5 3.5 0.0 of property or legal rights to property (equity shares) Capital investments for reconstruction and restoration of 268.0 268.0 254.6 254.6 234.5 234.5 melioration systems Partial or full compensation of agricultural input costs, 2,061.9 1,069.9 992.0 2,815.8 2,120.1 695.7 2,232.9 1,674.2 558.7 including: subsidies for fuel utilized to carry out seasonal 203.8 51.3 152.5 208.7 199.0 9.7 83.6 83.6 agricultural works subsidies for purchase of agricultural machinery and 318.0 318.0 560.2 560.2 542.4 526.0 16.4 equipment subsidies for plantation and maintenance of perennial 0.0 0.0 20.5 16.2 4.3 plantations subsidies for support of animal/poultry breeding 0.0 0.0 29.9 19.3 10.6 material other subsidies for support of animal sector (building of 0.0 0.0 88.5 34.5 54.0 farms, complexes, poultry factories) other (compensation of interest rates on bank loans, repayment of credits 1,540.1 700.6 839.5 2,046.9 1,360.9 686.0 1,468.0 1,078.2 389.8 guaranteed by the central government and the regional executive committees): compensation of interest 0.0 0.0 1,007.0 969.9 37.1 rates on bank credits National Rezervation 0.0 0.0 14.2 14.2 "Беловежская пуща" Annexes 97 Bel arus Public Expenditure Review repayment of budget loans 0.0 0.0 -12.1 -12.1 and borrowings GOB payments as guarantor 0.0 0.0 15.0 15.0 on bank credits cost reduction of feed quality determination and milk 0.0 0.0 1.7 1.7 quality sampling repayment of arrears on credits provided to organizations designated 0.0 0.0 216.9 68.2 148.7 to purchase agricultural machinery and supply it to ag producers on leasing terms purchase and cost reduction 0.0 0.0 59.2 23.0 36.2 of seeds repair and seasonal maintenance of tractors, harvesters, grain cleaning 0.0 0.0 51.1 51.1 and drying equipment, other agricultural machinery and equipment purchase of other inputs/ 0.0 0.0 32.5 32.5 commodities repayment of loans for machinery repairs, fuels and 0.0 0.0 82.5 82.5 oils, plant protection agents, seeds direct budget payments to producers (price supplements on products procured from the 49.7 49.7 43.8 43.8 45.5 45.5 population, support to peasant farms) other (subsidies and compensations from the regional 223.0 223.0 279.2 279.2 530.2 530.2 budget), including: financial aid 0.0 0.0 289.1 289.1 repayment of arrears on loans for fuels and oil and 0.0 0.0 176.1 176.1 loans to dairies reconstruction and repair of 0.0 0.0 35.0 35.0 industrial facilities cost reduction of compound 0.0 0.0 30.0 30.0 feed 98 Annexes Fiscal Reforms for a Sustainable Economic Recovery B2. Product specific measures 996.4 706.0 290.4 706.8 485.9 220.9 414.4 366.0 48.4 Subsidies for crop production, of 622.3 608.8 13.5 554.6 335.2 219.4 388.3 341.7 46.6 which: grains (harvesters, threshing 113.2 100.0 13.2 404.9 188.9 216.0 60.5 18.2 42.3 floors, plant construction) fiber flax 186.3 186.3 82.1 82.1 225.5 225.5 potatoes 31.0 31.0 1.3 1.3 4.3 4.3 other (sugar beet, phyto- based products, fruits, 291.8 291.5 0.3 66.3 64.2 2.1 98.0 98.0 vegetables) Fish 15.8 15.8 3.7 3.7 10.6 10.6 other types of animal production 55.0 55.0 98.2 98.2 0.0 other subsidies for support of 289.2 13.8 275.4 19.6 19.6 animal production subsidies for support of "elite" 0.0 25.1 25.1 seed production other subsidies for support of 14.1 12.6 1.5 5.6 4.1 1.5 15.5 13.7 1.8 crop production TOTAL SUPPORT (A+B) 4,677.2 2,903.4 1,773.8 5,680.1 3,734.9 1,945.2 5,205.2 3,529.2 1,676.0 Green Box, % 8.3 7.3 9.9 7.1 6.5 8.2 12.0 12.0 12.0 Amber Box, % 91.7 92.7 90.1 92.9 93.5 91.8 88.0 88.0 88.0 Sources: MOFA and IRSA. Annexes 99 Bel arus Public Expenditure Review ANNEX 2.5. Labor and capital productivity in Belarus: agriculture versus the rest of the economy 2000 2004 2005 2006 2007 2008 2009 GDP, billion 2000 BYR 9,134 11,969 13,094 14,403 15,642 17,237 17,272 Labor force, thou people 4,441.0 4,316.3 4,403.0 4,457.5 4,505.0 4,594.4 4,625.7 Fixed capital, trillion 2000 BYR 86.6 90.1 91.9 94.2 95.7 98.6 100.5 Share of agriculture in GDP, % 11.6 8.3 7.9 7.9 7.5 7.9 7.8 Share of agriculture in total employment, % 14.1 10.7 10.5 10.2 9.8 9.5 9.5 Share of agriculture in fixed capital, % 16.6 14.9 14.8 14.7 14.6 14.1 14.3 Labor productivity in agriculture, million 2000 BYR 1.69 2.15 2.24 2.50 2.66 3.12 3.07 / worker Labor productivity in rest of the economy, million 2.12 2.85 3.06 3.31 3.56 3.82 3.80 2000 BYR / worker Relative labor productivity in agriculture, % 80 76 73 76 75 82 81 Fixed capital productivity in agriculture, BYR 0.074 0.074 0.076 0.082 0.084 0.098 0.094 Fixed capital productivity in the rest of the economy, 0.112 0.143 0.154 0.165 0.177 0.187 0.185 BYR Relative fixed capital productivity in agriculture, % 66 52 49 50 47 52 51 Capital / labor ratio in agriculture, million 2000 BYR 22.96 29.07 29.42 30.46 31.65 31.84 32.72 / ag worker Capital / labor ratio in the rest of the economy, 18.93 19.89 19.87 20.07 20.11 20.36 20.58 million 2000 BYR / worker Sources: WB Staff calculations on Belstat data. 100 Annexes Fiscal Reforms for a Sustainable Economic Recovery ANNEX 3.1. Average heat tariffs (excl. VAT) and tariff cost recovery by consumer category, nominal US$/Gcal Average natural gas tariffs (excl. VAT) and tariff cost recovery by consumer category, nominal US$/thousand cm 2010 Consumer Category 2005 2006 2007 2008 2009 (Jan-Aug) Residential Organizations (Households) 61 67 81 116 99 95 109% 116% 73% 64% 54% 45% Industry 61 63 120 151 174 204 Average tariff 61 63 117 148 168 195 Average electricity tariffs (excl. VAT) and tariff cost recovery by consumer category, nominal US cent/kwh 2010 Consumer Category 2005 2006 2007 2008 2009 (Jan-Aug) Residential Organizations (Households) 3.60 4.12 5.02 6.09 5.39 5.46 93% 93% 80% 82% 74% 62% Industry 5.98 7.01 9.36 10.78 10.19 11.68 154% 158% 148% 145% 140% 133% Budget organizations 4.70 5.56 8.03 10.29 9.83 10.93 121% 125% 127% 138% 135% 125% Non-industrial consumers 6.30 7.48 9.92 11.29 10.77 12.16 163% 168% 157% 152% 148% 139% Average tariff 5.32 6.22 8.22 9.68 9.03 3.93 137% 140% 130% 130% 124% 45% Sources: Ministry of Energy, WB Staff calculations. ANNEX 3.2. Methodology for the estimation of the social impact of tariff increases The methodology used for simulating the welfare impact of the tariff increases was that developed by Freud and Walich in 1995. Based on “Rising Households’ energy prices in Poland: Who gains? Who loses? World Bank Policy Research Working Paper 1495”. The loss in welfare for each household is calculated as follows: Loss = Q0(p1-p0)(1+e*(p1-p0)/2p0) Where loss is the monetary loss of the households due to increase in the prices, Q is quantity consumed, p is price, e is the price elasticity of demand, the subscript 0 refers to the period before the price change, and the subscript 1 refers to the period after the price change. As quantities of the households’ utilities consumption are not available in the HBS, the loss is calculated using the expenditures on utilities (from the HBS) and the residential tariffs, as follows: Loss = EXP0*[(p1-p0)/ p0]*(1+e*(p1-p0)/2p0) Where EXP is households’ expenditure for the analyzed utility component. We used 3 alternative levels of elasticity (e) for the analysis: 0, -0.4 and -0.285. Annexes 101 Bel arus Public Expenditure Review ANNEX 4.1. Benefits’ coverage by selected groups of households, 2009 Well targeted privileges in percent 60 50 40 30 20 10 0 Rural Urban Brest Vitebsk Gomel Grodno Minsk Minsk-Oblast Mogilev Basic General Secondary Special Secondary Tertiery None/primary employee self-employed unemployed retired student OLF HH with children HH with elderly Q1 Q2 Q3 Q4 Q5 1 2 3 4 5+ Quintile Region Household Size Head Education Head Labor Force Poorly targeted privileges in percent 60 50 40 30 20 10 0 Rural Urban Brest Vitebsk Gomel Grodno Minsk Minsk-Oblast Mogilev Basic General Secondary Special Secondary Tertiery None/primary employee self-employed unemployed retired student OLF HH with children HH with elderly Q1 Q2 Q3 Q4 Q5 1 2 3 4 5+ Quintile Region Household Size Head Education Head Labor Force Source: Households’ Budget Survey, 2009 102 Annexes