## _cr05232 ## Source details **Canonical URL:** [_cr05232](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2005/_cr05232.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2005/_cr05232.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2005/_cr05232.pdf.json) --- ### I. Creating Fiscal Space — macro context, fiscal stance, and expenditure composition - Macro vulnerabilities and context - Total public debt declined to about 60 percent of GDP at end-2004. - Surging inflation and a large external current account deficit that is increasingly debt-financed make the economy more vulnerable to shocks. - Serbia represents about 93 percent of the union GDP (the Union of Serbia and Montenegro). - Fiscal performance and near-term stance - Recorded consolidated general government overall balance improved by 3 percentage points of GDP in 2004 compared to 2003, achieving a first primary surplus for the first time in 2004. - Fiscal stance in 2005 will be further strengthened by 1.5 percent of GDP. - For 2005, the surplus of the consolidated general government is targeted at 1.2 percent of GDP. - Size of government and revenue burden - Tax burden is about 45 percent of GDP. - Revenues are on average about 5 percentage points of GDP higher than in neighboring countries, while expenditures are 3 percentage points higher. - Government outlays reached 45 percent of GDP in 2004. - Composition of spending (2004–05) - Wage bill: 10 percent of GDP. - Entitlements: 19 percent of GDP. - Subsidies: 3 percent of GDP. - Capital spending: 3 percent of GDP. - Non-discretionary spending comprise over two-thirds of total spending. - Expenditure-side policy directions (bulleted) - Reduce the size of government and reorient expenditure priorities to support economic growth. - Cut questionable and low-efficiency spending to generate savings for structural reform and to create a better social safety net (for example, unemployment benefit). - Raise capital spending. - Forcefully tackle expenditure rigidities, including further downsizing of government employment and measures to keep the wage bill under control while allowing limited wage decompression. - Improve public expenditure management, focusing on human resource capacity and treasury organization. ### II. Public sector employment, union-level reforms, and pensions - Public sector employment reform - Wage bill in Serbia and Montenegro is higher than most countries in the region. - 2005 civil service reform: trim government employment by tackling overstaffing at lower grades and allow some wage decompression. - Some 3,000 workers or 1.5 percent of government employment will be laid off in 2005. - Defense ministry manpower reduction in 2005: 15 percent (over 6,000 personnel, mainly civilian employees) through attrition, early retirement, and separations. - Pension system: current burden and reform measures - Pension outlays among the highest in the region; require large annual budget transfers. - Doubling of the dependency ratio since 1990 to about 0.8 in 2004. - Average pension payment is about 70 percent of average net wages of the economy. - With relatively flat collection from contributions, an annual transfer of 5 percent of GDP is needed to finance the cash deficit for the pension fund for workers. - Payroll contributions currently at 22 percent of gross wages; cover about two thirds of pension expenditures. - Arrears in pension payments range from one month in the FE up to two years for pensioners in the FF. - Reforms enacted and planned - 2001 reforms: raised retirement age by 3 years; shifted indexation to 50:50 wage/price; reduced minimum pension to 20 percent of gross average wage; rationalized pensionable base. - Government-agreed measures phased in over a 4-year period: - Replace quarterly with annual indexation. - Index post-retirement pension benefits only to inflation by delinking from wages. - Increase the retirement age. - Administrative priority: with World Bank assistance, select competent agency to set up a national registry of pensioners. - Pension reform recommendations (short–medium term) - Introduce a tax/contribution on pension benefits (example cited: 10 percent). - Reduce indexation post-entitlement while maintaining pre-entitlement Swiss formula temporarily. - Shift periodicity of indexation to annual from quarterly. - Increase and equalize gradually statutory retirement age for men and women to 65, and raise minimum retirement age. - Eliminate accelerated pensionable service credits and the 15 percent additional pensionable years of service benefit for women. - Improve contribution collection and individual contributor database to reduce evasion. ### III. Pension projections, reform options, and fiscal impact (Fund for Employees, FE) - Baseline FE projection highlights (selected rows reproduced exactly) - Real GDP growth (percent): 2006 4.8; 2007 5.0; 2008 5.2; 2009 5.4; 2010 5.4; 2025 5.0; 2050 5.0. - Inflation (percent change): 2006 9.0; 2007 6.0; 2008 4.0; 2009 3.5; 2010 3.2; 2025 3.1; 2050 3.0. - Real wage growth (percent): 2006 4.8; 2007 5.0; 2008 5.2; 2009 5.4; 2010 5.4; 2025 5.0; 2050 5.0. - Total fertility rate (per 100 women): 2006 172.6; 2007 173.1; 2008 173.6; 2009 174.1; 2010 174.6; 2025 181.9; 2050 194.2. - Disability rate: Expected to fall by 26 percent by 2050. - Survivors: Expected to fall by 32 percent by 2050. - Overall balance (percent of GDP): 2006 -5.2; 2007 -5.5; 2008 -5.7; 2009 -5.8; 2010 -6.0; 2025 -6.9; 2050 -0.9. - FE projected path and drivers - FE balance expected to worsen to about 6.9 percent of GDP in the 2020s then improve during the 2030s to a deficit of about 1 percent of GDP by 2050. - Pattern driven by adverse demographics in first 20 years, longer-term improvement from reduced replacement ratio and fading disability overhang from 1990s. - Selected reform savings (Fund for Employees, 2006─50; Percent of GDP) — IMF staff estimates (exact table values) - Tax on pension benefits of 10 percent: 2006 1.0; 2007 1.0; 2008 1.0; 2009 1.0; 2010 1.0; 2025 1.2; 2050 1.0. - No indexation until Q3 2008—CPI thereafter: 2006 1.2; 2007 2.0; 2008 2.5; 2009 2.7; 2010 2.8; 2025 3.9; 2050 3.8. - Gradual shift in indexation coefficient to CPI: 2006 0.0; 2007 0.2; 2008 0.4; 2009 0.6; 2010 0.9; 2025 3.5; 2050 3.7. - Change to annual indexation: 2006 0.5; 2007 0.7; 2008 0.7; 2009 0.7; 2010 0.8; 2025 2.0; 2050 3.0. - Eliminate 15 percent benefit for women: 2006 0.0; 2007 0.1; 2008 0.1; 2009 0.1; 2010 0.2; 2025 0.7; 2050 0.8. - Reduce early retirement by 40 percent: 2006 0.0; 2007 0.1; 2008 0.1; 2009 0.2; 2010 0.2; 2025 0.5; 2050 1.0. - Authorities’ reform package (FE, 2006─50) — projected financial effects (exact values) - Baseline overall balance (percent of GDP): 2006 -5.2; 2007 -5.5; 2008 -5.7; 2009 -5.8; 2010 -6.0; 2025 -6.9; 2050 -0.9. - Authorities’ reform scenario: 2006 -4.6; 2007 -4.6; 2008 -4.6; 2009 -4.6; 2010 -4.6; 2025 -2.2; 2050 4.7. - Savings (Authorities’ reform vs Baseline): 2006 0.6; 2007 0.8; 2008 1.1; 2009 1.2; 2010 1.5; 2025 4.6; 2050 5.6. - Interpretation: Short-term savings limited (about 0.6 percent of GDP in 2006); average annual savings 2006–2010 about 1 percent of GDP; medium/long-term largest source of savings is shift to full CPI indexation. ### IV. State and Socially Owned Enterprises (SSOE) — 2004 deficits, quasi-fiscal activity, and policy - SSOE role and 2004 findings - SSOEs employ roughly the same number as the budget sector and play a dominant economic role. - Aggregate SSOE deficit rose from an estimated 1.2 percent of GDP in 2003 to 1.6 percent of GDP in 2004. - Most increase in financing occurred via private sector supplier credits to NIS and EPS which rose by 0.6 percent of GDP. - Cash deficits of NIS rose from 0.3 percent of GDP in 2003 to 1.0 percent of GDP in 2004. - Estimated cash deficit of EPS rose from 0.6 percent of GDP in 2003 to 0.7 percent of GDP in 2004. - Quasi-fiscal items and implicit subsidies - Government increased net implicit consumer subsidies to energy users by about U.S. dollars 179 million (0.9 percent of GDP) from 2003 to 2004 due to suspension of diesel and gasoline wholesale price formula. - NIS benefited from free crude from Vojvodina oilfields without paying royalties: U.S. dollars 111 million in 2003 and U.S. dollars 145 million in 2004. - Data shortcomings and recommended analytical extensions - Coverage and quality gaps: incomplete SSOE coverage; above-the-line data limited; corporate balance sheets often unaudited. - Financing data weaknesses: non-cash debt settlements and reschedulings not identified. - Recommendation: obtain transparent audited above-the-line reporting for SSOEs and extend consolidated financing data including deposits, non-cash transactions, and reschedulings. - Analytical extension: distinguish current versus investment spending; assess whether administered prices reflect full enterprise costs. - Policy recommendations for SSOEs and privatization - Rapid privatization of SSOEs is "first best" to reduce quasi-fiscal activities and improve governance. - Develop clear exit strategy for enterprises that cannot be sold; use strengthened bankruptcy process where appropriate. - Require audited financial statements consistent with IAS for all large SSOEs. - Ensure SSOEs pay taxes and do not accumulate arrears to energy suppliers, tax administration, or employees. - NIS should pay royalties for extracting oil. - Strengthen NBS monitoring of SSOE financing operations (deposits, loans, foreign debt). ### V. Public Expenditure Management (PEM), Treasury Single Account (TSA), and debt management - Treasury and TSA coverage - Treasury organized around a TSA; commitments registration and monthly cash limits introduced in 2004. - About 3,500 indirect budget users (estimated ~30 percent of the republican budget) remain outside TSA coverage and are executed by the Public Payment Agency (PPA). - PPA records revenues, revenue sharing, and services extrabudgetary funds and local budgets. - Recommended PEM reforms - TSA should comprehensively cover all indirect budget users and be the sole account to receive all government inflows and fund operations. - Close own revenue accounts of indirect budget users currently serviced separately by PPA (examples: court fees, road tolls, hospital services). - Reduce foreign currency revenue accounts of indirect budget users and fold them into the TSA. - Merge treasury and PPA operations per March 2005 action plan; centralize IT and networks. - Record all stages of government transactions: approved appropriations, intra-year changes, commitments, obligations, and cash discharge. - Unify all investment programs into a single budget framework; appoint lead ministry to coordinate all foreign-financed investment programs. - Build MOF budget planning unit to implement a 3-year rolling budget, including costing of current-year new programs. - Treasury capacity and debt management - Treasury to expand into active cash and debt management: continuous forecasting of inflows/outflows to TSA, guide cash releases, manage liquidity, coordinate borrowing with monetary policy. - Urgent need to widen and deepen human resources; serious understaffing noted. - Early passage of draft Law on Public Debt is a priority. - Short-term: debt management services remain shared with NBS; debt recording function retained in NBS and could be formalized under service-level agreement. - Focus on maturity structure, currency composition of public debt, and use of privatization proceeds to retire expensive debt. ### VI. Tax policy, social contributions, and revenue administration - Tax policy and recent changes - Serbia transformed a complex tax system into a fairly modern, pro-growth system; exemptions greatly reduced. - 2004: elimination of 3.5 percentage point wage bill tax; reduction of CIT and PIT on self-employed from 14 percent to 10 percent; harmonization of social security tax bases and contributions. - 2005: elimination of financial transaction tax; introduction of two-rate VAT (standard 18 percent, reduced 8 percent). - Proposed conditional 2005 changes (subject to revenue performance): reduce remaining PIT on salaries from 14 percent to 10 percent (harmonize PIT and CIT at flat 10 percent); abolish tax on share holding to promote capital markets. - Caution: revenue-losing tax reforms may loosen fiscal stance and pressure the external current account deficit if not compensated. - Social security taxes and prudent strategy - Employers and employees pay combined additional 12.3 percent for health insurance and 1.5 percent for unemployment insurance; total payroll tax 35.8 percent. - Social taxes make up about one-third of total wage cost. - Even with current rates, significant annual budget transfers needed to finance cash deficits of social funds. - Recommendation: address expenditure side of social funds before reducing social contribution rates. - Revenue administration reforms and targets - New legal framework: Law on Tax Procedures and Administration (2003) and creation of Serbian Tax Administration (STA). - Law on Cash Registers (2004) to rein in gray economy; Law on Customs Administration (2004) harmonized taxpayer IDs. - STA established a large taxpayer unit (LTU) and improved enforcement; self-assessment introduced for personal and corporate profit taxes. - VAT implementation in 2005 was smooth after about 2 years of preparation. - Target segmentation for tax collections: - Aim for 70–75 percent of all tax collections from the largest 500–800 taxpayers. - Second segment (small- to medium-sized taxpayers) ~20–25 percent of revenue. - Small taxpayer segment not expected to generate more than 5 percent of revenue. - Current performance: STA collects about 50 percent of total revenue from its largest taxpayers. - STA institutional and HQ needs - Build STA capacity in strategic and operational national planning; development of national programs; technical advice to operational units; national performance targets and monitoring. - Strengthen HQ functions and clarify MOF vs STA roles: MOF leads tax policy design; STA leads administrative design, implementation, legislative drafting, data provision, and costing. - Administrative cost and compliance reduction recommendations - Amend by-laws to: - Provide filing and payment of payroll tax monthly for large employers and quarterly for others. - Require employees with a single source of income to file an annual tax return only if income exceeds a high threshold. - Review and simplify tax return forms. - Raise VAT registration threshold from 2 million dinar/year to 5 million dinar/year. - Make LTU criterion selective and define a large taxpayer by turnover only (2004 amendment added turnover but left other criteria intact; current multi-criteria practice deemed unnecessarily complicated). ### VII. Inflation determinants, model-based findings, and short-term forecasts - Main empirical findings - Main short-run determinant: the exchange rate. - Other important factors: incomes and the output gap have strong impacts; broad money growth is not statistically significant in the analysis. - 12-month RPI rose to 17.5 percent in May 2005; RPI lowest level: 7.5 percent in January 2004. - Local governments raised municipal service prices end-2004 contributing about 3 percentage points to inflation. - Introduction of VAT in January 2005 added a one-time boost to inflation. - Short-term model (2001–04) pass-through (Box 1 exact estimates) - Exchange rate pass-through: 0.72 within 12 months; 0.88 in 24 months. - Nominal incomes pass-through: reached 0.4 in first 2–3 months; declined to 0.12 in 12 months. - Broad money (M2) impact: insignificant or negative. - Industrial production (IPI) pass-through: strongly negative, reaching -0.26 in the first year. - Modified short-term model (core vs non-core) - Exchange rate → core inflation: 0.89 within 12 months. - Exchange rate → noncore residual: 0.37 within 12 months. - Industrial production → core inflation: -0.36 within 12 months. - Medium-term model (1997–2004) - Exchange rate (ER): 0.39 within 12 months. - Nominal incomes (INCOME): 0.29 within 12 months. - Broad money (M2): not significant. - Forecast scenarios for 2005 (Short-Term Model: Forecast for 2005 — exact table) - Scenario 1: ER 5; Inflation (RPI) 10; Nominal Incomes (INCOME) 9; Broad Money (M2) 28. - Scenario 2: ER 10; Inflation (RPI) 12; Nominal Incomes (INCOME) 14; Broad Money (M2) 21. - Scenario 3: ER 15; Inflation (RPI) 13; Nominal Incomes (INCOME) 19; Broad Money (M2) 15. - Policy implications from model and diagnostics - Cointegration of exchange rate and inflation in short-term model reflects managed exchange rate policy and yields higher pass-through (about 0.7) vs medium term (~0.4). - Rising import share (from 26 percent in 1997 to 53 percent in 2004) supports higher pass-through. - High euroization (65 percent foreign currency deposits at end-2004) may have contributed to strong pass-through. - Suggested monetary-policy constraint: nominal devaluation over 2005 should not exceed 5 percent to keep RPI inflation in single digits at end-2005 (model calculation). ### VIII. Euroization — prevalence, risks, and policy toolkit - Key facts and scope - Financial euroization prevalent: more than two thirds of total deposits denominated in euro; at least two thirds of loans either euro-denominated or indexed to euro. - Share of foreign currency–denominated deposits rose from 54 percent in 2002 Q3 to 67 percent at end-2004. - Remittances increased from 11 percent in 2002 to 14 percent in 2004 (relative to GDP). - Share of foreign currency–denominated and indexed loans in total loans rose to 70 percent. - Main costs and channels of vulnerability - Borrowers without euro cash flow are exposed to exchange rate depreciation → higher NPLs and provisioning needs. - High euroization reduces monetary policy effectiveness and promotes “fear of floating.” - Liability euroization limits central bank Lender of Last Resort capacity in foreign currency despite NBS gross foreign exchange reserves of US$4.47 billion and coverage ratio of 150 percent over deposits as of April 2005 (deposit base US$2.97 billion; US$1.2 billion held as required reserves). - Foregone seigniorage estimate: about 0.6 percent of GDP (assuming €3 billion cash and 3½ percent assumed interest rate on euro-denominated German bonds). - Policy strategy and recommendations - Pillars: sound macro policies to lower inflation; tighter macro-prudential measures to internalize euroization externalities; micro-prudential measures to minimize banking risks. - Macroeconomic: ensure positive real interest rates on dinar deposits; more proactive monetary policy and repo operations; consider inflation-targeting over time. - Macro-prudential: reduce remuneration on required reserves for household FX deposits to zero; increase SRRs on FX enterprise deposits and banks’ foreign borrowing; consider marginal SRRs or price-based capital controls only as last resort. - Micro-prudential: higher provisioning for FX-denominated/indexed lending; require banks’ boards to implement FX-risk policies; require borrower hedging assessment and explicit borrower disclosure of depreciation scenarios; strengthen NBS data gathering on FX exposures. - International experience: de-dollarization is usually a by-product of credible anti-inflationary policies and fiscal adjustment; forced conversions generally unsuccessful. ### IX. External competitiveness and export performance — diagnosis and policy - External imbalance summary - Current account deficit before grants hovered at 12–13 percent of GDP in recent years; trade deficit around 23–25 percent of GDP. - SM exports averaged about 22 percent of GDP (2001─04) vs regional average of 38 percent of GDP. - Export composition and direction - Exports concentrated in intermediate products (about 60 percent), commodities, and agricultural products. - Main destination: Bosnia and Herzegovina, followed by other former Yugoslav republics; EU3 export share losing ground to FY republics in 2000–04. - Price competitiveness and REER - On average in 2004 CPI- and RPI-based REER indices around 25 percent lower than 1997 average; compared to 2001 levels, indices significantly higher (by 19 and 34 percent respectively). - Tentative conclusion: real exchange rate does not seem excessively appreciated; risk in 2005 from still high inflation with a stable nominal exchange rate. - Non-price competitiveness - Weak export performance linked more to non-price factors (investment climate, governance, infrastructure) than to exchange rate level. - Policy recommendations: - Deepen and accelerate privatization and structural reforms to improve investment climate. - Liberalize trade regime and conclude regional and EU trade agreements; accession to WTO important. - Be cautious with export-specific subsidies; prioritize economy-wide reforms. ### X. Statistical appendix and selected macro-fiscal data highlights (exact values preserved) - Selected macro indicators and fiscal figures (exact values as presented) - GDP (in millions of U.S. dollars): 1999 $10,214; 2000 $8,603; 2001 $11,576; 2002 $15,528; 2003 $20,665; 2004 $23,996 (Est.). - GDP (in billions of dinars): 1999 193; 2000 382; 2001 772; 2002 998; 2003 1,189; 2004 1,401. - Real GDP growth (annual percent): 1999 -18.0; 2000 5.0; 2001 5.0; 2002 5.5; 2003 3.8; 2004 2.7. - Retail price inflation (annual average percent): 1999 42.1; 2000 69.9; 2001 91.1; 2002 21.2; 2003 11.3; 2004 9.5. - General government total revenue (Serbia, in billions of dinars): 2004 592.7; total expenditure and net lending 593.0; overall balance -0.3. - National Bank of Serbia gross foreign reserves (millions of U.S. dollars, end period): 1999 289; 2000 516; 2001 1,169; 2002 2,280; 2003 3,557; 2004 4,302. - Current account balance (in percent of GDP): 1999 -7.5; 2000 -3.9; 2001 -4.6; 2002 -8.9; 2003 -7.3; 2004 -13.1. - External debt (year-end, percent of GDP): 1999 10.7; 2000 11.4; 2001 11.9; 2002 11.8; 2003 14.3; 2004 14.9. - Monetary aggregates (M2 end-period percent change): 1999 67.6; 2000 61.4; 2001 104.9; 2002 52.7; 2003 27.5; 2004 30.3. - Banking sector: share of foreign currency deposits in total deposits rose to 67 percent at end-2004; share of FX-denominated/indexed loans in total loans rose to 70 percent. *Source: _cr05232 - References.* ### References.............................................................................................................. ### _cr05232 - References.............................................................................................................. ### I. Creating Fiscal Space: A Reform Agenda for the Fiscal Sector in Serbia — Key Findings and Policy Directions - Macro context and vulnerabilities - Total public debt declined to about 60 percent of GDP at end-2004. - Surging inflation and a large external current account deficit that is increasingly debt-financed make the economy more vulnerable to shocks. - Serbia represents about 93 percent of the union GDP (the Union of Serbia and Montenegro). - Fiscal performance and near-term stance - The recorded consolidated general government overall balance improved by 3 percentage points of GDP in 2004 compared to 2003, achieving a first primary surplus for the first time in 2004. - The fiscal stance in 2005 will be further strengthened by 1.5 percent of GDP. - For 2005, the surplus of the consolidated general government is targeted at 1.2 percent of GDP. - Data regarding overall fiscal deficits in the early 2000s should be interpreted with care; the sharp deterioration in 2002 was due mainly to bringing onto the fiscal accounts certain unrecorded quasi-fiscal activities of earlier years. - Size of government and revenue burden - Tax burden is about 45 percent of GDP. - Revenues are on average about 5 percentage points of GDP higher than in neighboring countries, while expenditures are 3 percentage points higher. - Government outlays reached 45 percent of GDP in 2004. - Composition of spending and rigidities (2004–05) - Wage bill: 10 percent of GDP. - Entitlements: 19 percent of GDP. - Subsidies: 3 percent of GDP. - Capital spending: 3 percent of GDP. - Non-discretionary spending (transfers to households such as pensions, social entitlements, medical and unemployment benefits, as well as government wages) comprise over two-thirds of total spending. - Quasi-fiscal activities have generally been under control, but failure to pass through high international oil prices in a state-owned oil company has undermined some consolidation. - Preliminary work indicates operating deficits of socially-owned enterprises have declined, but in 2004–05 this may be more than offset by worsening performance of the state-owned oil company. - Policy recommendations (expenditure side) - Reduce the size of government and reorient expenditure priorities to support economic growth. - Cut questionable and low-efficiency spending programs to generate savings for structural reform and to create a better social safety net (such as unemployment benefit) for those affected by reform. - Raise capital spending. - Forcefully tackle expenditure rigidities, including further downsizing of government employment and measures to keep the wage bill under control while allowing limited wage decompression. - Improve public expenditure management, with a focus on human resource capacity and organization of the treasury. - Tax policy and administration - Major policy changes in the next 2–3 years are not essential given prior improvements to the tax system. - Measures should be taken to improve tax compliance and rein in the gray economy to better mobilize resources. ### II. Section II — The Serbian Pension System: Issues and Reform Options (structure and topics) - Main subsections listed in the source: - A. Main Parameters of the Serbian Pension System - B. Current Financial Problems, Recent Reforms, and Outstanding Issues - C. Reform Options - D. Conclusion - Box referenced: - II.1. Benefits for Both Old Age and Disability Pensions are Based on a Point System Formula - Key data and projections mentioned in the source list (text tables and figures relevant to pensions): - Figure II.1. Contributors and Pension Beneficiaries in the Fund for Employees, 1986–2004. - Text Tables: - II.1. Baseline Projections of the Finances of the Fund for Employees, 2006-50. - II.2. Statutory and Early Pensionable Ages, 2004. - II.3. Social Security Contribution Rates, 2004. - II.4. Fund for Employees: Financial Savings from Selected Reform Options, 2006–50. - II.5. Fund for Employees: Financial Savings from Authorities’ Plan, 2006–50. ### III. Section III — Serbia: Deficits of State and Socially Owned Enterprises in 2004 — Findings and Policy - Main subsections listed: - A. Introduction and Main Findings - B. SSOE Performance in 2004 - C. Shortcomings and Possible Extensions - D. Policy Recommendations - Key text tables and figures referenced: - III.1. State and Socially Owned Enterprises in Serbia. - III.2. Overall Fiscal Deficit, 2003–04. - III.3. Deficit Financing for the 10 Largest Socially Owned Enterprises. - III.4. SSOEs Subsidies and Arrears, 2003–04. - Summary notes from chapter context: - State-owned enterprises, particularly the state-owned oil company, remain a drag on overall economic performance. - Operating deficits of socially-owned enterprises have declined, but performance is uneven across enterprises. ### IV. Section IV — Inflation Determinants in Serbia — Analysis and Model-Based Findings - Main subsections listed: - A. Introduction - B. Inflation Developments in Serbia - C. Brief Review of Studies on Inflation in the Region - D. Model Specifications and Output - E. The Results - F. Model Forecast of Inflation - Boxes and figures: - Box IV.1. Summary of the Model Estimates. - Figures: contribution of components to 12-month RPI growth and dinar depreciation rate (2002–05); main indicators of inflation (2002–05); short- and medium-term model variables and fits; impact of electricity price increases (2002–05); model forecast for 2005. - Text tables on model pass-throughs and short-term forecast for 2005: - IV.1. Short-Term Model: Pass-Through from Explanatory Variables to RPI Inflation. - IV.2. Short-Term Model: Pass-Through from Explanatory Variables to the Core Inflation and Noncore Inflation Residual. - IV.3. Medium-Term Model: Pass-Through from Explanatory Variables to RPI Inflation. - IV.4. Short-Term Model: Forecast for 2005. ### V. Section V — Euroization in Serbia: Macroeconomic, Prudential, and Policy Implications - Main subsections listed: - A. Introduction - B. Recent Developments and International Comparison - C. Explanations for High and Rising Euroization - D. Macroeconomic and Prudential Implications - E. Policy Issues - Boxes and figures: - Box V.1. Dollarization: Key Concepts. - Box V.2. De-dollarization and Managing the Risks of High Dollarization: The International Experience. - Figures: Euroization of deposits and loans (2003–05); liability euroization international comparisons; liability euroization in Serbia and selected transition economies (1999–2004); Serbia interest rates (2003–05); impact of euroization on money multipliers (2001–04); euroization and lender of last resort capability of the NBS (2003–05). ### VI. Section VI — Serbia and Montenegro: Export Performance and External Competitiveness - Main subsections listed: - A. Introduction - B. Export Performance in Historical Perspective - C. Price Competitiveness Factors - D. Non-Price Competitiveness Factors - E. Measures to Improve Competitiveness - F. Conclusions - Figures and tables: - Figures include nominal exchange rates (1994–2005), real effective exchange rates (1995–2004), proxies of ULC and transition indicators, and wage comparisons. - Text tables: - VI.1. Serbia and Montenegro: Regional Comparison of Trade Flows, 2000–04. - VI.2. Serbia: Direction of Exports, 1997–2004. - VI.3. Serbia: Composition of Exports and Imports, 1997–2004. ### Statistical Appendix and Supporting Data - Extensive statistical appendix covering 32 tables with indicators for 1996–2004 (selected ranges vary by table), including: - Selected Economic and Financial Indicators, 1999–2004. - Selected Social Indicators, 1996–2003. - GDP by sector, employment, labor costs, prices, retail inflation rates, enterprise counts and liquidation, general government fiscal operations, pension fund beneficiaries, public sector debt, monetary survey, bank rates, balance of payments, composition and destination of trade, and external debt stock (various years as specified per table). - Specific table highlights include: - Table 14. Pension Fund, Number of Beneficiaries of Pension and Resources for Pension Payments of Serbia, 1999–2004. - Table 21. Balance Sheet of the National Bank of Serbia, 1999–2004. - Table 27. Balance of Payments, 1999–2004. - Table 32. Stock of External Debt, 1999–2004. *Source: _cr05232 - References..............................................................................................................* ### 6. Public sector employment reform is essential for placing the wage bill on a ### _cr05232 - 6. Public sector employment reform is essential for placing the wage bill on a sustainable path. ### Public sector employment reform and the wage bill - The wage bill in Serbia and Montenegro is higher than most countries in the region. - A first step taken in 2005: civil service reform to trim government employment by tackling overstaffing at the lower grades and allow some wage decompression in the civil service. - Some 3,000 workers or 1.5 percent of government employment will be laid off this year. - Further employment cuts in the general government, especially in the health and education sector, may be needed in 2006 and beyond, pending ongoing assessment of the public sector reform, with assistance from an EU agency and the World Bank. ### Union-level expenditure reform (defense and broader union budget) - In 2005, the defense ministry will reduce its manpower by 15 percent (over 6,000 personnel, mainly civilian employees) through attrition, early retirement, and voluntary and forced separations. - Expected benefits: - Improve structure of the union budget to better balance wage and nonwage expenditures. - Create space to fund better training programs and, if deemed essential, some hardware acquisitions as the military consolidates operations and bases. ### Pensions: current burden and reform measures - Pension outlays are among the highest in the region and are expected to remain significant as a percent of GDP in the short to medium term. - Reforms enacted in 2001 included: - Raised the retirement age by 3 years for both men and women. - Shifted the pension indexation formula to the average of wage and price indexation (50:50). - Reduced the minimum pension to 20 percent of gross average wage. - Rationalized the pensionable base. - Short-term savings from 2001 reforms were relatively modest; a more significant impact is expected in the long run. - Current pension system requires large annual budget transfers to finance its cash deficit. - Key statistics and projections: - Doubling of the dependency ratio (pensioners to contributors) since 1990 to about 0.8 in 2004. - Average pension payment is about 70 percent of average net wages of the economy. - With relatively flat collection from contributions, an annual transfer of 5 percent of GDP is needed to finance the cash deficit for the pension fund for workers. - Dependency ratio expected to worsen in the medium term; pension fund deficit projected to deteriorate further absent corrections. - Government-agreed measures to be phased in over a 4-year period: - Replace quarterly with annual indexation. - Index post-retirement pension benefits only to inflation by delinking it from wages. - Increase the retirement age. - Administrative reforms: - With World Bank assistance, a high level work group set up to tackle deficiencies in pension administration. - Priority: select a competent agency to set up a national registry of pensioners. ### Subsidies and enterprise support - Explicit subsidies to enterprises and agriculture should be rolled back further in coming years. - Subsidies as a share of GDP declined to under 3 percent of GDP in 2005. - Rationale: - Further cuts needed to harden budget constraints on enterprises, accelerate restructuring of companies with positive net worth, and force liquidation of those with negative net worth. - Specific recommendations: - Reform plan for the state-owned railway company should be a priority to avoid subsidy cuts causing arrears to the private sector. - Reexamine agricultural subsidies (about 0.5 percent of GDP), including poorly targeted price support (for example, milk), with a view to downsizing. - Policy-based lending for the agriculture sector should be phased out as soon as feasible. - Note: Implicit subsidies, such as tax arrears, are discussed elsewhere. ### Public Expenditure Management (PEM) — treasury, TSA, and PPA - Progress made: Serbia now has a treasury system organized around a Treasury Single Account (TSA) and introduced commitments registration and monthly cash limits in 2004. - Coverage gaps and fiscal risk: - TSA can execute budgets of all direct budget users, but about 3,500 indirect budget users (estimated at about 30 percent of the republican budget) remain outside TSA coverage and are executed by the Public Payment Agency (PPA). - PPA also records revenues, shares revenues (such as personal income tax) between republican and municipal levels, and services extrabudgetary funds and local budgets. - Recommended reforms: - TSA should evolve to comprehensively cover all indirect budget users and be the sole account to receive all government inflows and fund all operations. - Close own revenue accounts of indirect budget users currently serviced as separate accounts by the PPA (examples: court fees, road tolls, hospital services). - Reduce foreign currency revenue accounts of indirect budget users and fold them into the TSA. - Institutional integration: - An action plan issued in March 2005 to merge treasury and PPA operations. - Present treasury IT system is custom-made and not easy to network; taking over PPA infrastructure and network for accounting and processing is logical to centralize expenditure management. - Recording and control: - All stages of government transactions should be recorded: approved appropriations, intra-year changes, commitments, obligations, and cash discharge. - Persistent payment arrears indicate need for full general ledger coverage of all expenditure stages. - Investment budgeting: - Unify all investment programs into a single budget framework. - Foreign loan-financed capital investment programs are not effectively integrated with the budget; inadequate coordination leads to excessive foreign-financed investment programs and pressures on external debt ceilings. - Appoint a lead ministry to coordinate all investment programs financed from outside the budget. - Medium-term budgeting: - Annual Budget Memorandum provides 3-year projections, but outer year expenditure projections often do not account for new programs. - MOF should prioritize building the budget planning unit to implement a 3-year rolling budget in the near future, including accurate costing of current-year new programs for future years. - Treasury capacity and debt management: - Treasury must expand functions into active cash and debt management, preparing financial plans and forecasting inflows/outflows into the TSA continually. - Treasury should guide budget execution, determine cash releases, manage liquidity with government borrowing in coordination with monetary policy. - Urgent need to widen and deepen human resource base of the treasury due to serious understaffing. - Early passage of a long-delayed draft Law on Public Debt should be a priority. - Treasury should concentrate on policy development for maturity structure, currency composition of public debt, and use of privatization proceeds to retire expensive debt. - Enhance analytical capability of the debt unit. - Short-term: debt management services should remain a shared responsibility with the National Bank of Serbia (NBS); debt recording function should be retained in the NBS and could be formalized under a service level agreement. - Focus on integrating cash and debt management policy and operations. ### Tax policy - Simplification and orientation: - Serbia transformed a highly complex tax system into a fairly modern, pro-growth tax system; exemptions have been greatly reduced. - Policy focus: reduce taxation on labor income and increase tax burden on consumption, leading to a steady fall in the direct/indirect tax ratio. - Key reforms and rates: - 2004 developments included elimination of the 3.5 percentage point wage bill tax, reduction of the corporate income tax rate (CIT) and the personal income tax (PIT) on the self-employed from 14 percent to 10 percent, and harmonization of social security tax bases and uniformatizaion of contribution rates. - 2005: elimination of a highly distortionary financial transaction tax and introduction of a new 2-rate VAT (standard rate 18 percent, reduced rate 8 percent) replacing a cascading retail sales tax. - Note: recent discussion to expand the reduced VAT list (for example, to tickets for sports and entertainment and new apartments) would weaken the VAT structure. - Proposed further tax changes in 2005 (conditional on revenue performance): - Complete transformation to reduce the remaining PIT (on salaries and wages) from 14 percent to 10 percent to harmonize PIT and CIT at a flat rate of 10 percent. - Proposal to abolish tax on share holding to promote capital market development and securities trading. - Caution: Further revenue-losing tax reforms, even if sound, might need to wait for an improvement in the external balance; without compensating measures, further tax cuts would loosen fiscal stance and pressure the external current account deficit. - Social security taxes: - Social taxes make up about one-third of total wage cost. - These taxes fund pay-as-you-go pension funds (workers, farmers, self-employed), health fund, and unemployment fund. - Even with current rates, significant annual budget transfers are needed to finance cash deficits of these funds. - Recommended prudent strategy: address expenditure side of social funds before reducing social contribution rates to avoid weakening fiscal consolidation. ### Revenue administration - Institutional and legal improvements: - New legal framework: Law on Tax Procedures and Administration (2003) and creation of Serbian Tax Administration (STA) under the ministry of finance. - STA took over tax collection responsibilities from the former Federal Payment Agency. - Law on Cash Registers passed in 2004 to rein in gray economy. - Law on Customs Administration enacted in 2004 harmonizing taxpayer identification numbers for customs and tax administration. - Administrative reforms and milestones: - Reorganization under STA included establishment of a large taxpayer unit (LTU) and improved tax enforcement strategy, contributing to stronger revenue receipts. - Self-assessment introduced for personal and corporate profit taxes. - Payroll tax administration simplified via harmonization of contribution bases and clarification of common definitions. - VAT implementation in 2005 was smooth after about 2 years of preparation. - Tax revenue collection as a percent of GDP has grown steadily since 2000. - Further strengthening and segmentation: - Clarify tax policy role of STA: MOF to lead tax policy design; STA to lead administrative design of policy proposals and implementation issues, legislative drafting, data provision for revenue analysis and forecasting, and costing policy proposals. - Further taxpayer segmentation recommended: refine service and compliance strategies; treat medium-sized and small taxpayers and specific industries (e.g., financial services) separately. - Target metrics for STA: - Aim for 70–75 percent of all tax collections from the largest 500-800 taxpayers (broadly consistent with international norms). - Second segment (small- to medium-sized taxpayers) to contribute around 20-25 percent of revenue. - Small taxpayer segment not expected to generate more than 5 percent of revenue. - Current performance: STA collects about 50 percent of total revenue from its largest taxpayers. - To align with international practice, make LTU criterion more selective and define a large taxpayer by turnover only (2004 amendment apparently failed to emphasize turnover as sole criterion). *Source: _cr05232 - 6. Public sector employment reform is essential for placing the wage bill on a* ### 28. Headquarters functions would also need to be strengthened. While the STA has ### _cr05232 - 28. Headquarters functions would also need to be strengthened. While the STA has ### Headquarters functions and organizational needs - Managers lack operational HQ experiences and have concentrated efforts largely on the development of instructions for branch offices. - STA capacity needs to be built up in: - strategic and operational planning on a national level; - development of national programs; - provision of technical advice and guidance to operational units; - establishment of national performance targets and measurement systems; - monitoring and evaluation of field offices. - Operational offices would eventually need to be structured along lines similar to the functions in the HQ. ### Recommendations to reduce compliance and administrative costs of the tax system - Amend by-laws to: - provide for filing payroll tax returns and paying payroll tax on a monthly basis for large employers and on a quarterly basis for others; - require employees with a single source income to file an annual tax return only if their income exceeds a high threshold; - review the tax return forms with a view to reducing their number and complexity; - raise the VAT registration threshold from 2 million dinar per year to 5 million dinar per year. - Note on LTU identification criteria: - The 2004 amendment to the LTU decree added turnover as a criterion for identifying a large taxpayer, but left other criteria (such as level of taxes paid) intact. - Current practice of using more than one criterion to determine a taxpayer for the LTU is described as unnecessarily complicated and not in line with best international practice. ### The Serbian pension system: overview and recent deficits - The pension system has been running large deficits and has become a drag on the budget. - In 2004: - the pension system deficit amounted to 5.4 percent of GDP and was financed by equivalent budget transfers. - The Fund for Employees (FE) ran a deficit of about 5 percent of GDP. - The Farmers’ Fund (FF) posted a deficit of 0.6 percent of GDP. - The Fund for the Self Employed (FSE) posted a surplus of 0.2 percent of GDP. - Payroll contributions: - currently at 22 percent of gross wages; - covered only about two thirds of pension expenditures. - Arrears in pension payments range from one month in the FE up to two years for pensioners in the FF. ### Chapter objectives and structure - The chapter: - discusses causes of current problems; - presents preliminary results of projected finances of the FE; - suggests options for reducing pension outlays; - provides preliminary estimates of the impact of the authorities’ recent reform package on the FE finances. - Division into sections: - Section A: description of main parameters of the Serbian pension system. - Section B: background for current financial problems, recent reforms and main issues. - Section C: reform options, chosen reform package with preliminary financial implications. - Section D: brief conclusion. ### A. Main parameters of the Serbian pension system - System type: - The Serbian Pension system is a pay-as-you-go system. - Key parameters: - contribution rates, eligibility criteria for old age, disability and survivor pensions, benefit formula (Box 1), and indexation rules. - Contribution rates and bases: - Contribution rates from employers and employees are currently 22 percent for old age, disability, and survivors’ insurance. - Maximum contribution base is five times the average monthly gross wage. - Minimum contribution base is 40 percent of average monthly gross wage. - Employers and employees pay a combined additional 12.3 percent for health insurance and 1.5 percent for unemployment insurance, making the total payroll tax 35.8 percent of wages. - Additional contributions are required for specified professions/groups that receive additional months of service credit (up to 18 months of service credited for every 12 months of contributions depending on job type). - Eligibility criteria (selected points): - Retirement ages: 63 for men and 58 for women with 20 years of service; 65 and 60 with 15 years of service. - Minimum retirement age: 53 with 40 years of service for men and 35 years for women. - Any individual meeting eligibility for two or more pension types is entitled to only one pension, at their choice. - Disability pensions: only for complete loss of capacity to work; immediate eligibility if work-related injury; otherwise requires minimum years of contributions depending on age. - Survivor pensions: minimum of 5 years of contributions before survivors qualify; survivor benefit rates depend on number of survivors (70 percent for one survivor, 80 percent divided between two survivors, 90 percent for three survivors, 100 percent for four or more survivors); survivors’ eligibility ages and conditions described. - Indexation: - The general point value (GPVt) is indexed by the “Swiss formula”: average of the inflation rate and nominal wage growth. - Indexation implemented quarterly. - Minimum and maximum pensions and post-retirement pensions are indexed by the same combination (50 percent inflation and 50 percent nominal wage growth) and periodicity. ### Box 1 — Point system formula for old age and disability pensions (selected mechanics) - Pension formula: - Pit = PPi * GPVt - GPVt was set at about 1.8 percent of the value of the average gross wage in 2004. - Personal points: - PPi = PCi * PSi - PCi is the personal coefficient (sum of ratios of individual contribution wage to economy’s average wage for each year, divided by the number of years). - An individual earning the average wage accumulates one point per year for the first 40 years; between 40 and 45 years earn only half a point; beyond 45 years earn no points. - All women are given an additional 15 percent service credit with respect to PSi. - Example: - A male earning the average wage every year and contributing for 41 years: PCi = (40*1 + 1*0.5)/41; PPi would round to 40.0—(40.5/41)*40.5—since PS i would be 40.5. - If retired in 2004, replacement rate would have been 72 percent of the average gross wage (40*1.8). - Minimum and maximum pensions: - Minimum pension set at 20 percent of average gross wages in 2002. - Maximum pension set implicitly to 170 GPVt by capping PCi at 4 and PSi at 42.5. - Neither contributions nor benefits are taxed. - Disability pensions determined like old age pensions but with increased length of service (PSi) according to specified formula. - Survivor pensions calculated on the basis of at least 20 years of contributions regardless of actual years of contribution; other family members can claim differences under specified conditions. ### B. Current financial problems, recent reforms, and outstanding issues - Financial unsustainability: - Sharp increase in dependency ratio during the 1990s. - Employees fund dependency ratio went from about 0.4 in 1990 to about 0.8 in 2004. - Causes: - Decline in number of contributors due to disruptions to economic activity in the 1990s and higher informalization. - Increase in beneficiaries due to system maturity and lax eligibility criteria for disability pensions. - Disability pensioners currently make up 30 percent of all pensioners—roughly three times levels seen in most other countries. - Disability system used to retire early individuals who would not have qualified for retirement until normal retirement age. - Reforms introduced: - 2001 reforms: reduced contribution rate from 32 percent to 19.6 percent while broadening gross wage definition; increased statutory retirement age by 3 years with comparable increase in minimum retirement age; changed indexation from wages to Swiss formula implemented quarterly; tightened disability eligibility mainly by eliminating partial disability. - 2003 reforms: introduced point system making accrual rates more uniform; changed pension calculation wage from best ten consecutive years to full career average. - Contribution rate increases: to 20.6 percent in 2003 and to 22.0 percent in 2004. - Expected impacts and timing: - 2003 World Bank Public Expenditure Review estimated 2001 reforms would reduce the 2005 deficit by 1.6 percent of GDP, though actual reduction likely less due to abnormal inflow of pensioners before legislative changes. - World Bank medium-term estimates: deficit could be reduced by up to 4 percent of GDP by 2010 and up to 7 percent of GDP by 2020. - New disability benefits reduced from 21,408 in 2002 to 10,746 in 2004. ### C. Projections and fiscal outlook for the Fund for Employees (FE) - Summary of projected pattern: - FE balance expected to worsen slightly up to about 6.9 percent of GDP in the 2020s and begin improving during the 2030s, reaching a deficit of about 1 percent of GDP by 2050. - Pattern driven by increasing dependency ratio due to adverse demographics dominating first 20 years; longer-term improvement due to significant reduction in replacement ratio and decreasing dependency pressures as disability overhang from the 1990s phases out. - Baseline projections table (selected rows reproduced exactly as presented): - Real GDP growth (percent): 2006 4.8; 2007 5.0; 2008 5.2; 2009 5.4; 2010 5.4; 2025 5.0; 2050 5.0. - Inflation (percent change): 2006 9.0; 2007 6.0; 2008 4.0; 2009 3.5; 2010 3.2; 2025 3.1; 2050 3.0. - Real wage growth (percent): 2006 4.8; 2007 5.0; 2008 5.2; 2009 5.4; 2010 5.4; 2025 5.0; 2050 5.0. - Total fertility rate (per 100 women): 2006 172.6; 2007 173.1; 2008 173.6; 2009 174.1; 2010 174.6; 2025 181.9; 2050 194.2. - Disability rate: Expected to fall by 26 percent by 2050. - Survivors: Expected to fall by 32 percent by 2050. - Overall balance (percent of GDP): 2006 -5.2; 2007 -5.5; 2008 -5.7; 2009 -5.8; 2010 -6.0; 2025 -6.9; 2050 -0.9. - Sources: National authorities; World Bank; Schwarz (2005); and IMF staff estimates and projections. ### Outstanding systemic issues - Contribution evasion and compliance: - Significant evasion of contributions is a main problem. - Less than 60 percent of the labor force is contributing to the pension system. - Lack of an individual database on contributors hampers enforcement. - Contributions collected by tax authorities are largely transmitted through employers; employers are required to file an M-4 form annually reporting individual contributions, but many neglect to do so. - Result: neither tax authorities nor pension funds have reliable records on individual contributors. - Retirement ages and early retirement: - Regular retirement ages are low relative to European average: - European average statutory retirement age: 63.8 for men and 61.2 for women. - Serbia: 63 for men and 58 for women. - Minimum retirement age of 53 for both men and women is among the most generous in Europe. - Early retirement prevalence: - As of end-2003, about 15 percent of the stock of FE pensioners retired before the normal retirement age (58 for women and 63 for men). - More than 35 percent of disability pensioners have retired early. - For new entrants in 2003: - 65 percent of new male and female old age retirees retired before the normal retirement ages. - For disability pensions in 2003, 47 percent of men and 28 percent of women retired early. *Prepared by Alejandro Simone; sources and figures as presented in the original text.* ### 14. Accelerated pensionable service for specific professions is facilitating early ### _cr05232 - 14. Accelerated pensionable service for specific professions is facilitating early ### Accelerated pensionable service and implicit subsidies - For some professions, including dangerous professions such as mining, a worker contributing for 12 months can be credited for up to 18 months of contributions. - This accelerated crediting: - Allows individuals in these professions to retire early as their pensionable service requirement is met faster. - Produces insufficiently compensated higher benefits because early retirement increases the duration over which benefits are paid. - Acts as an implicit subsidy from the government, which finances the deficit of the system through transfers, to specific activity sectors. - 2003 statistics: - Old age and disability retirees under accelerated pensionable service represented about 14.4 percent of the total number of old age and disability retirees in the FE. - They represented about 13.2 percent of new old age and disability retirees. - Recommendation: Companies in these sectors should be responsible for paying the higher cost their activities are generating. ### Replacement rates and indexation - Current replacement rates: - The average old age pension benefit represents about 50 percent of gross wages. - The ratio of the average pension benefit to the average wage net of contributions and taxes is about 80 percent of average net wages. - International standards are about 40 percent. - Contributing factors: - Pensions are not taxable while the worker earning the average wage does have to pay taxes, contributing to the relatively high effective replacement ratio. - Indexation concerns: - The current indexation of benefits includes a wage component and is unaffordable given the overhang of pensioners from the 1990s and very fast growth in real wages typical of transition countries. - The current quarterly periodicity of indexation exacerbates the costs of partial wage indexation; before 2001 indexation was monthly. - The high frequency ensured pensions tracked prices and wages closely but is no longer desirable because it is not affordable. - Long-term effect of pre-entitlement partial wage indexation: - Indexing the initial entitlement to an average of nominal wage growth and inflation will continuously reduce the replacement rate over time. - Under the current scheme, the level of the average replacement rate will be halved to slightly above 20 percent of the average gross wage by 2050. - Therefore, partial indexation to wages should be a temporary measure until an appropriate replacement rate is attained. - Risks if replacement rates reach very low levels: increased political pressure to raise benefits and increased incentives for evasion due to low rates of return on contributions. ### Reform strategies and prioritization - In a pay-as-you-go pension system, four basic reform strategies can be combined: - Raising contribution rates. - Improving collection enforcement of contributions. - Reducing replacement rates. - Reducing the system dependency ratio. - Timing and impact: - Increasing contribution rates and reducing benefits to current and soon-to-be retirees typically produce significant short-term improvements. - High dependency ratio and collection enforcement problems are unlikely to be significantly affected in the short term; these can be improved in medium to long term by restricting pension eligibility for future retirees and improving collection enforcement. - Authorities are working with the World Bank to create a database for contributors and strengthen collection enforcement. - Given current labor taxation and evasion: - Tightening benefits seems the main reform path to generate short term savings. - Key comparative figures: - Pension contribution rate: 22 percent (Serbia) versus 24 percent average OECD level. - Overall payroll taxation: 35.8 percent of gross wages (Serbia) versus 36 percent average OECD level. - Continuing to increase pension contribution rates risks worsening the business environment and increasing incentives for evasion, possibly reducing collections. ### Recommended benefit-reduction options (short to medium term) - Introduce a tax/contribution on pension benefits: - Example: A 10 percent tax on current pensions could be seen as a “solidarity” contribution and would reduce unnecessary tax preferences for pensions. - Reduce indexation of pension benefits post entitlement while maintaining the current pre-entitlement indexation temporarily: - Options for post-entitlement indexation range from freezing indexation completely to a gradual convergence to CPI indexation. - Pre-entitlement: the current Swiss formula can be kept temporarily since it gradually reduces the replacement rate, but ultimately pre-entitlement indexation will need to shift back to wages in the medium to long term to avoid deterioration below an adequate replacement rate. - Shift periodicity of indexation to annual from quarterly: - Keeping pensions without indexation for a longer period in an environment of significant positive inflation can effectively reduce pension costs. - Increase and equalize gradually the statutory retirement age for men and women at 65 years, and raise gradually the minimum retirement age: - The current minimum retirement age of 53 years is among the lowest in Europe and exacerbates early retirement prevalence and cash flow problems. - Eliminate accelerated pensionable service credits and the 15 percent additional pensionable years of service benefit for women: - These provisions increase pension costs by facilitating early retirement and do not internalize the full cost to the labor market. ### Projected savings from selected reforms (Fund for Employees, 2006─50; Percent of GDP) - Selected reforms and IMF staff estimates (Table 4): - Tax on pension benefits of 10 percent: - 2006: 1.0 - 2007: 1.0 - 2008: 1.0 - 2009: 1.0 - 2010: 1.0 - 2025: 1.2 - 2050: 1.0 - No indexation until Q3 2008—CPI thereafter: - 2006: 1.2 - 2007: 2.0 - 2008: 2.5 - 2009: 2.7 - 2010: 2.8 - 2025: 3.9 - 2050: 3.8 - Gradual shift in indexation coefficient to CPI: - 2006: 0.0 - 2007: 0.2 - 2008: 0.4 - 2009: 0.6 - 2010: 0.9 - 2025: 3.5 - 2050: 3.7 - Change to annual indexation: - 2006: 0.5 - 2007: 0.7 - 2008: 0.7 - 2009: 0.7 - 2010: 0.8 - 2025: 2.0 - 2050: 3.0 - Eliminate 15 percent benefit for women: - 2006: 0.0 - 2007: 0.1 - 2008: 0.1 - 2009: 0.1 - 2010: 0.2 - 2025: 0.7 - 2050: 0.8 - Reduce early retirement by 40 percent: - 2006: 0.0 - 2007: 0.1 - 2008: 0.1 - 2009: 0.2 - 2010: 0.2 - 2025: 0.5 - 2050: 1.0 ### Authorities’ reform package and projected fiscal impact (Fund for Employees, 2006─50) - Authorities’ measures: 1. Gradually increase retirement ages for women from 58 to 60 and from 63 to 65 for men in equal annual increases in a four years period. 2. Change periodicity of indexation to an annual basis starting in October 2005. 3. Phase in a shift to price indexation in four years for pensions, with equal annual increases in the inflation weight of the indexation formula. 4. Maintain the Swiss formula for calculating initial benefits. - Projected financial effects (Table 5; Percent of GDP): - Baseline scenario: - 2006: -5.2 - 2007: -5.5 - 2008: -5.7 - 2009: -5.8 - 2010: -6.0 - 2025: -6.9 - 2050: -0.9 - Authorities’ reform scenario: - 2006: -4.6 - 2007: -4.6 - 2008: -4.6 - 2009: -4.6 - 2010: -4.6 - 2025: -2.2 - 2050: 4.7 - Savings (Authorities’ reform scenario vs Baseline): - 2006: 0.6 - 2007: 0.8 - 2008: 1.1 - 2009: 1.2 - 2010: 1.5 - 2025: 4.6 - 2050: 5.6 - Interpretation: - Short-term savings from the authorities’ package are limited to about 0.6 percent of GDP in 2006. - Average annual savings per year from 2006 to 2010 are likely to be about 1 percent of GDP. - In the medium to long term, the shift to full CPI indexation is expected to be the largest source of savings, contributing to much larger improvements by 2050. *Source: _cr05232 - 14. (IMF staff report).* ### 26. Although past reforms and the latest package will significantly improve the ### _cr05232 - 26. Although past reforms and the latest package will significantly improve the ### Pension system outlook and fiscal projections - Past reforms and the latest package will significantly improve the finances of the system in the medium to long term, but will not be sufficient to attain financial balance in the short term. - Current projections: - The FE would continue to run a deficit of 4.6 percent of GDP in the period 2006-2010. - The deficit would reach levels of 3 percent of GDP by 2020. - Unless additional measures are taken to reduce the benefits of current, or soon to be, retirees, financial balance will be achieved only in the long term. - Main drivers and constraints: - The large increase of the dependency ratio during the 1990s is the main cause of the system’s current financial problems. - Politically difficult measures (examples cited) that could reduce the cost of the pensioner overhang more rapidly include suspending indexation or taxing pension benefits. - Without such measures, it will take a long time for the costs of the 1990s overhang to work through and disappear. - Early retirement and recommended reforms: - Early retirement has been a reason for increased costs in many European pension systems; Serbia shares similar demographic trends and is likely to experience similar pressures. - Recommended measures to reduce pension costs in the medium to long term: - Increase the minimum retirement age. - Eliminate the accelerated pensionable service for certain professions and groups. - Eliminate the 15 percent additional pensionable years of service benefit for women. - These measures should be taken as soon as possible to reduce pension costs. *References cited in source: Government of Serbia, 2004; Martineau, Jean-Noel, June,2004; Schwarz, Anita, May 2005; World Bank, 2005; World Bank, February, 2003.* --- ### SSOE deficits in 2004 — summary of main findings - Purpose: assess extent of deficits in the non-budget public sector (state- and socially owned enterprises, SSOE) and their macro-fiscal implications. - Context and drivers: - SSOEs in Serbia employ roughly the same number as the budget sector and play a dominant role in the economy. - Contributing factors to rising deficits include reduced budget subsidies, incomplete passthrough of international energy prices to consumers, and intensified privatization efforts that may have increased moral hazard for managers and unions to boost expenditures and debt before expected privatization and associated debt write-offs. - Data and methodological note: - Analysis relies on cash deficits estimated from the financing side because above-the-line data are not available. - Despite severe data limitations, higher SSOE deficits may have contributed to widening macro-imbalances in 2004. ### SSOE performance and key statistics (2003–04) - Aggregate and sectoral deficits and financing: - SSOE deficit rose from an estimated 1.2 percent of GDP in 2003 to 1.6 percent of GDP in 2004. - Most of the increase in financing occurred in the form of private sector supplier credits to NIS and EPS which rose by 0.6 percent of GDP. - Total public sector deficit (Table 2 summary): 4.4 percent of GDP in 2003 and 1.7 percent of GDP in 2004 (table lists components and memorandum items). - Memorandum item: GDP in billions of dinars: 1,095 (2003) and 1,290 (2004). - Company-level and sectoral notes: - The overall operating deficits of the largest 10 socially owned enterprises declined from 0.7 percent of GDP in 2003 to 0.4 percent of GDP in 2004 on an unconsolidated basis (adjusted for loan write-offs for Zastava and Matroz). - Cash deficits of NIS rose from 0.3 percent of GDP in 2003 to 1.0 percent of GDP in 2004. - Estimated cash deficit of EPS rose from 0.6 percent of GDP in 2003 to 0.7 percent of GDP in 2004. - A handful of SSOEs (NIS, EPS, Post, Telekom, Aircontrol) accessed bank financing or other private credit in 2004; many others could not due to weak balance sheets. - Non-cash debt settlements and reschedulings are typically not identified in available financing data. - Subsidies, arrears, and quasi-fiscal activity: - Total budgetary subsidies and indirect support via payment arrears vis-à-vis EPS and NIS declined from 2.3 percent of GDP in 2003 to 1.8 percent of GDP in 2004. - Arrears to energy companies moved from 0.7 percent of GDP in 2003 to 0.6 percent of GDP in 2004; NIS arrears 0.5 to 0.2; EPS arrears 0.2 to 0.3. - The government increased net implicit consumer subsidies to energy users by about U.S. dollars 179 million (0.9 percent of GDP) from 2003 to 2004 due to changes and a temporary suspension of the price setting formula for diesel and gasoline wholesale prices. - The change in pricing practices protected domestic consumers from higher prices but likely contributed to deterioration in NIS accounts; this impact was cushioned by an increase in the economic value of an implicit subsidy allowing NIS to exploit Vojvodina oilfields without paying royalties to the budget. - Labor cost controls: - NIS and EPS better controlled their wage bill in 2004, which rose by 8 percent (broadly in line with the indicative program target), while nominal GDP growth is estimated at almost 18 percent. ### Data shortcomings and suggested extensions - Major data limitations that may cause underestimation of the consolidated public sector deficit: - Coverage: The socially owned enterprise coverage is incomplete and above-the-line data is not available for 2004; detailed financing data compiled only for the largest 10 socially owned companies. For the remaining 50–60 large socially owned companies under restructuring aggregate loan data is available, but not deposits. - Quality and comparability: Data quality for large state-owned companies is poor; corporate balance sheets and income data are not audited, accounting practices change frequently, and time lags for above-the-line data are large (as of March 2005, 2004 above-the-line data submitted only for NIS and EPS, except wage bill data for the 8 largest state-owned enterprises). - Financing data weaknesses: non-cash debt settlements and rescheduling are typically not identified and required adjustments are not made. - Exclusions: Municipally owned and operated companies (mainly water and transport utilities) are excluded; these depend on local subsidies and have suffered from underinvestment. - Recommended analytical extensions: - Additional micro-economic analysis distinguishing current versus investment spending to assess sustainability; anecdotal evidence suggests maintenance and investment spending have been systematically suppressed, temporarily lowering deficits but risking unsustainability as capital stock deteriorates. - Assess whether administered prices reflect true enterprise costs (including adequate investment and maintenance costing); follow pricing examples like diesel and gasoline pricing in Annex 1 and extend to other utilities to evaluate distortive administered pricing schemes and the absence of full-cost pricing principles. ### Policy recommendations (summary of implications) - For the pension system: - Implement measures to reduce costs associated with the 1990s pensioner overhang more rapidly if short-term financial balance is desired. Options include suspending indexation or taxing pension benefits (noted as politically difficult). - Reduce incentives for early retirement through: - Increasing the minimum retirement age. - Eliminating accelerated pensionable service for certain professions and groups. - Eliminating the 15 percent additional pensionable years of service benefit for women. - For SSOE and broader public sector finances: - Strengthen financial discipline in SSOEs and accelerate reform/privatization where feasible while managing moral hazard risks tied to pre-privatization debt accumulation. - Ensure transparent and audited above-the-line data reporting for state-owned enterprises to allow consolidated fiscal assessment. - Improve collection of consolidated financing data (including deposits, non-cash debt transactions, and reschedulings) across large socially owned companies under restructuring. - Extend data collection and performance monitoring to municipally owned utilities (water, transport) and incorporate their fiscal effects into consolidated public sector assessments. - Adopt and enforce full-cost pricing for utilities to eliminate implicit subsidies and reduce quasi-fiscal activities; evaluate administered tariffs for distortions and investment/maintenance shortfalls. - Address implicit consumer subsidies and pricing rules (example: diesel and gasoline wholesale pricing) to limit quasi-fiscal losses in companies like NIS and EPS. *Prepared by Harald Hirschhofer and Janko Guzijan, with the assistance of Dejan Maljkovic; sources cited include Privatization Agency; Ministry of Finance; National Bank of Serbia; Company balance sheets; and IMF staff estimates.* ### 12. Progress in reducing the broad public sector deficit requires: Privatization; ### 12. Progress in reducing the broad public sector deficit requires: Privatization; ### Policy recommendations and required measures - Rapid privatization of socially-owned enterprises (SSOE) is identified as the "first best option" to achieve structural reform in the SSOE sector and reduce quasi-fiscal activities. - Expected benefits: improve incentives, reduce governance problems, introduce new management techniques, technologies, and fresh financial resources. - If new investors are large international listed companies, transparency would be strengthened because such companies are required to submit audited financial reports to the public. - Recent efforts noted: speed-up privatization of socially-owned companies and spin-off of non-core activities of the state owned utilities; recommendation to extend privatization to core areas (recently announced for the refineries). - Develop a clear exit strategy for enterprises that cannot be sold to private investors. - The newly strengthened bankruptcy process is an option under certain conditions; government is increasing use of this instrument. - Increase transparency and accountability to address governance problems and improve public understanding of SSOE performance. - Recent requirements for some state enterprises to publish audited financial statements consistent with International Accounting Standards (IAS) should be broadened to all large SSOEs. - Any privileged treatments by the owner, the tax authorities, and direct subsidies need to be publicly known and costed. - Specific recommendation: NIS should pay royalties for extracting oil. - The NBS could strengthen monitoring of financing operations of SSOE, especially deposits, loans and foreign debt (supplier credits). - Strengthen financial discipline to reduce SSOE quasi-fiscal operations and return them to profitability. - SSOEs should pay taxes like any other companies and be subjected to regular audits. - Large socially owned enterprises should not be permitted to accumulate arrears towards energy suppliers (NIS and EPS), the tax administration, or employees. - Recent efforts by energy suppliers to improve collections need to be intensified. - If unprofitable enterprises are considered politically and socially important, the bills and taxes should be paid from the state budget. ### Annex 1 — Implicit consumer subsidies on diesel and gasoline: key findings - Market distortions: - Administrative price setting and import restrictions on oil derivatives; pricing adjustment mandated only if (1) international price for Ural crude changes by more than 2.5 percent in dinar terms, and (2) more than 30 days have elapsed since the last adjustment. - Government decree gives NIS the exclusive right to import oil derivatives and protects domestic refineries. Existing private retailers must either have crude imports processed by NIS refineries, or buy derivatives from NIS. - Policy inconsistency produced an implicit consumer subsidy when world prices rose: - An implicit tax on consumers of about SRD 5.7 billion in 2003 was replaced by an implicit consumer subsidy of about SRD 4.8 billion in 2004. - Resulting benefit to consumers: estimated SRD 10.5 billion (0.8 percent of GDP). - From April–October 2004 the government did not apply the oil price formula; combined actions (not applying formula and lowering refining margin from U.S. dollars 45 per ton to U.S. dollars 34 per ton) are estimated to have lowered NIS income by some SRD 4.5 billion or U.S. dollars 70 million over April–October. - Cushioning effect for NIS: - NIS received an implicit subsidy in the form of free crude produced in the Vojvodina fields; the budget does not receive royalties from this exploitation. - Benefit to NIS from this practice: U.S. dollars 111 million in 2003, and U.S. dollars 145 million in 2004; production volumes are declining considerably. ### Annex 1 — Key tables and figures (selected exact values as presented) - Table A2. Difference Between Domestic Prices and International Benchmark Multiplied by Consumption Volumes, 2003─04 (In billions of dinars) - Gasoline: 2003 = -1,8 ; 2004 = 2,9 ; Change = 4.7 - Diesel: 2003 = -3,9 ; 2004 = 1,9 ; Change = 5.8 - Total: 2003 = -5,7 ; 2004 = 4,8 ; Change = 10.5 - Table A3. NIS Value of Crude Produced, 2002─04 - Domestic production (in 000's tons): 2002 = 690.2 ; 2003 = 677.7 ; 2004 = 656.2 - (Change in percent): 2002 = ... ; 2003 = -1.8 ; 2004 = -3.2 - Domestic production (in 000's barrels): 2002 = 5,107 ; 2003 = 5,015 ; 2004 = 4,856 - International price for Ural (U.S. dollar/barrel): 2002 = 23.7 ; 2003 = 27.0 ; 2004 = 34.8 - Production costs (U.S. dollar/barrel): 2002 = 5 ; 2003 = 5 ; 2004 = 5 - Implicit value of crude produced (in millions of U.S. dollars): 2002 = 95.7 ; 2003 = 110.5 ; 2004 = 144.7 - Implicit value of crude produced (in billions of dinars): 2002 = 5.6 ; 2003 = 6.4 ; 2004 = 8.5 - Table A1. Stock of Banking Loans to the Public Enterprises and Socially Owned Enterprises (In billions of dinars, end period) — as presented in source - 200220032004 - Public enterprises 4.310.911.0 - NIS 3.43.16.7 - EPS 0.31.81.1 - PTT 0.05.62.2 - ZTP 0.40.10.5 - RTS 0.20.40.5 - JAT 0.00.00.0 - Airport 0.00.00.1 - Telekom 0.00.00.4 - Socially owned enterprises 6.47.76.2 - HIP Petrohemija 1.22.53.1 - RTB Bor 0.00.00.0 - 14. Oktobar 0.80.90.1 - Matroz 0.00.00.0 - Zastava 1.71.60.3 - Azotara Pancevo 2.32.32.4 - MIN 0.30.20.3 - Industrija stakla Pancevo 0.10.10.1 - (Source: National Bank of Serbia.) ### IV. Inflation determinants in Serbia — main findings (chapter summary) - Main short-run determinant: the exchange rate. - Other important factors: incomes and the output gap have strong impacts; broad money growth is not statistically significant in the analysis. - Data quality issues weaken conclusions and leave room for further research. - Recent inflation dynamics: - 12-month retail price inflation (RPI) rose to 17.5 percent in May 2005. - RPI lowest level: 7.5 percent in January 2004. - Local governments raised prices for municipal services at end-2004 contributing about 3 percentage points to the inflation rate. - Introduction of value added tax (VAT) in January 2005 resulted in an additional one-time boost to inflation. - Review of other studies and estimated exchange rate pass-through: - Serbian National Bank (NBS) estimated pass-through during 1997–2004 at 0.32; comparable impact of growth of money supply (M1) at 0.35 on prices (NBS, 2004a). - Another NBS model estimated pass-through from exchange rate to core inflation for 2001–04 as 0.4. - Regional estimates range between 0.2 and 0.7. - Romania (Gueorguiev, 2003): 0.7 in the medium term, reduced to 0.45 in recent years. - Russia (Stavrev, 2003): 0.5–0.7. - Belarus (Vacher, 2004): 0.3–0.4. - Slovakia (Kuijs, 2001): one percent devaluation induces a “rapid” increase of prices of 0.2 percent. - Balassa-Samuelson effect estimates for transition economies in Central Europe vary widely, between 0.2 and 4.0 percentage points per annum. - Modeling approach and data: - A medium-term (1997–2004) VAR and short-term (2001–04) VEC models were constructed. - Inclusion of a cointegrating equation between exchange rate and inflation for the short-term model to capture the dinar/euro exchange rate as a nominal anchor in 2001-04. - Model variables included: RPI (or subcomponents), exchange rate (dinar/euro), nominal incomes (seasonally adjusted proxy averaging wages and retail sales indices), industrial production (HP-filter residual of seasonally adjusted IPI), and broad money (seasonally adjusted M2 including bank deposits in foreign currencies for short-term model). - Other variables (wages, retail sales, world oil prices) had insufficient explanatory power for inclusion in basic models. - Short-term models without cointegration did not produce statistically robust results and were not used. *Source: _cr05232 - 12. Progress in reducing the broad public sector deficit requires: Privatization; (IMF staff text).* ### Box 1. Summary of the Model Estimates ### Box 1. Summary of the Model Estimates ### Short-term model (2001-04) — key pass-through estimates and dynamics - Main determinant of inflation: pass-through from the exchange rate. - Exchange rate pass-through: - 0.72 within 12 months. - 0.88 in 24 months. - Nominal incomes pass-through: - Reached 0.4 in the first 2-3 months. - Declined to 0.12 in 12 months. - Broad money (M2) impact: either insignificant or negative. - Industrial production (IPI) pass-through: strongly negative, reaching -0.26 in the first year. - Model note: pass-through estimates largely as expected, except broad money impact (requires special explanation in Section E). - Table 1 monthly values (selected months preserved exactly): - Month 1: ER 0.45; INCOME 0.00; M2 0.00; IPI 0.00 - Month 2: ER -0.17; INCOME 0.41; M2 -0.03; IPI -0.09 - Month 3: ER 0.41; INCOME 0.32; M2 -0.03; IPI -0.19 - Month 6: ER 0.52; INCOME 0.26; M2 -0.06; IPI -0.17 - Month 12: ER 0.72; INCOME 0.12; M2 -0.07; IPI -0.26 - Month 24: ER 0.88; INCOME 0.10; M2 -0.08; IPI -0.29 ### Modified short-term model — core inflation and non-core residual dynamics - Using core and non-core inflation indicators instead of total RPI revealed transmission nuances. - Pass-through from exchange rate to core inflation: - 0.89 within 12 months. - Pass-through from exchange rate to noncore residual: - 0.37 within 12 months. - Incomes: - Stronger pass-through to noncore residual. - Very strong only in first 2–3 months; 0.12 within 12 months. - Industrial production: - Negative pass-through remained high, especially for core inflation: -0.36 within 12 months. - Non-core residual → core inflation: - Strong negative impact: -0.36 within 12 months. - Reverse impact: smaller and negative. - Table 2 monthly values (selected months preserved exactly): - Pass-through to Core Inflation — Month 1: ER 0.36; INCOME 0.00; IPI 0.00; NONCORE_RPI 0.00 - Pass-through to Core Inflation — Month 12: ER 0.89; INCOME 0.06; IPI -0.36; NONCORE_RPI -0.36 - Pass-through to Noncore Inflation Residual — Month 1: ER -0.07; INCOME 0.00; IPI 0.00; CORE_RPI -0.23 - Pass-through to Noncore Inflation Residual — Month 12: ER 0.37; INCOME 0.12; IPI -0.21; CORE_RPI -0.12 - Model note: numerical estimates here considered less reliable than basic short-term model but useful to identify transmission channels. ### Medium-term model (1997-2004) — estimates and caveats - Exchange rate remains key factor in medium term, but incomes influence strengthens. - Medium-term pass-through estimates: - Exchange rate (ER): 0.39 within 12 months. - Nominal incomes (INCOME): 0.29 within 12 months. - Broad money (M2): not significant. - Industrial production (IPI): not significant (positive 0.04 in month 12 per Table 3). - Modified medium-term model (using core inflation): - ER → core inflation: 0.47 within 12 months. - Caveat: major structural breaks and time-series distortions of 1999-2001 imply deviations of quantitative estimates can be quite significant; calls for more sophisticated model design. - Table 3 monthly values (selected months preserved exactly): - Month 1: ER 0.15; INCOME 0.00; M2 0.00; IPI 0.00 - Month 3: ER 0.33; INCOME 0.20; M2 -0.13; IPI -0.04 - Month 6: ER 0.37; INCOME 0.29; M2 -0.10; IPI 0.04 - Month 12: ER 0.39; INCOME 0.29; M2 -0.09; IPI 0.04 ### Structural explanations and interpretation (Section E) - Managed exchange rate policy and cointegration: - Cointegration of exchange rate and inflation in short-term model captures managed exchange rate policy and yields higher pass-through (about 0.7 in recent years vs about 0.4 in medium term). - Rising import share: - Share of imports doubled from estimated 26 percent in 1997 to 53 percent in 2004. - Sets 0.4–0.5 as the “natural” lowest threshold for reasonable pass-through estimates for 2004-05. - Euroization: - Share of foreign currency deposits in total bank deposits reached 65 percent at end-2004. - High euroization may have contributed to strong pass-through and may prevent decline in pass-through unless exchange rate flexibility increases. - Anecdotal price-formation evidence: - For many items (commodity and consumer/intermediate goods without Serbian substitutes) pass-through from international to domestic prices can be close to 1. - Many marketable services quoted in euros with settlements in dinars at current exchange rate. - Administered prices account for about one-third of the RPI basket (excluding oil); periodic adjustments of administered prices extend pass-through length but do not eliminate strength. - Income channel: - Low short-run impact of incomes on inflation could be due to exchange rate absorbing other impulses and availability of imports to meet increased demand. - Time pattern: incomes effect very strong in first 2–3 months then declines quickly. - Money growth channel: - Low impact of money growth on inflation may be due to high euroization. - Models produced negligibly small negative estimates of broad money growth (M2) impact on prices. - Difference in dynamics between dinar-denominated reserve money (has not grown rapidly) and overall money supply (increased with growth of euro-deposits). - Industrial production: - Strong negative impact of changes in industrial production on inflation in short term, up to minus 0.3 within 12 months, indicating rising output can alleviate inflation pressures. - Supply response weak; inefficient companies faced capacity constraints and adjusted mainly via prices. - Noncore residual → core inflation interaction: - High negative relationship can reduce the impact of administered price changes on overall inflation by lowering effective demand for other goods and services. - Example: electricity price hikes in July 2002–2004 coincided with smaller increases or declines in other prices (Figure 6); seasonality cannot fully explain effect. ### Model forecast and scenarios (Section F and Table 4) - Basic short-term model capability: - Captures main trend and can produce sound forecasts for a year ahead but cannot precisely forecast volatile monthly inflation. - Model failed to capture inflation accelerations in June-July 2004 and December 2004-January 2005 driven by one-time/transitory factors (oil derivative and electricity price rises; municipal price increases; VAT introduction). - Model devaluation–inflation sensitivity: - Nominal devaluation of the dinar over 2005 should not exceed 5 percent to keep RPI inflation in single digits at end-2005 (model calculation). - For a 10 percent nominal devaluation of the dinar over 2005, model forecast inflation of about 12 percent, consistent with nominal incomes growth of about 15 percent and a 21 percent increase in broad money. - Table 4 — Short-Term Model: Forecast for 2005 (Percent change during the year) — preserved exactly: - Scenario 1: Exchange Rate (ER) 5; Inflation (RPI) 10; Nominal Incomes (INCOME) 9; Broad Money (M2) 28 - Scenario 2: Exchange Rate (ER) 10; Inflation (RPI) 12; Nominal Incomes (INCOME) 14; Broad Money (M2) 21 - Scenario 3: Exchange Rate (ER) 15; Inflation (RPI) 13; Nominal Incomes (INCOME) 19; Broad Money (M2) 15 ### Technical notes and data choices (Technical Appendix) - Inflation measure: - Model uses RPI as main measure rather than CPI. - Rationale: RPI is broader and historically considered by government and mass media as main indicator; used in Fund programs with Serbia. - From medium-term perspective RPI and CPI broadly similar; short-run deviations matter for modeling. - Core inflation index: - Constructed by staff with NBS Research Department. - Excludes food, tobacco, energy and oil products, and communal, transport, and other services with administered prices — together about half of RPI basket. - NBS core inflation index filters out administered prices but does not exclude food; choice of core index affects quantitative estimates but not overall conclusions. - Incomes proxy: - Proxy for disposable incomes calculated as simple average of wage and retail sales indices. - Rationale: wage data may not fully capture household purchasing power (remittances, privatization/share sales, bonds for past savings losses, social transfers). - Short-term models with original wage index produced a “wrong-signed” coefficient for wages, confirming data inconsistency. - Retail sales index alone could be used but officially reported retail sales for 2004 could have an upward bias due to compulsory... *Source: Model estimates.* ### introduction of cash registers at the beginning of the year. ### _cr05232 - introduction of cash registers at the beginning of the year. ### Seasonal adjustments - Three variables with strong seasonal patterns were seasonally adjusted prior to the model exercises: Broad money (M2), industrial production (IPI), and proxy for incomes (INCOME). - Two key variables were not seasonally adjusted because seasonality was unclear or unstable: the exchange rate (ER) and inflation (RPI). Reasons given: - Exchange rate policy and the high share of administered prices in the RPI basket may explain unstable seasonality. - Seasonal adjustment of ER and inflation risked loss of important information on interdependencies (for example, the strength of the cointegrating relation declined significantly between the seasonally adjusted exchange rate and inflation). ### Variables and their modifications (as used in models) - RPI: Retail Price Index. - CORE_RPI: Core inflation index calculated by the staff; excludes the prices for food, tobacco, energy, and oil products, communal, transportation and other regulated services. - NONCORE_RPI: Index of prices for non-core components (residual) of the RPI basket; includes the prices for food, tobacco, energy, and oil products, communal, transportation and other regulated services. - ER: Nominal exchange rate of dinar to euro (dinars per 1 euro). - INCOME: Proxy for nominal disposable incomes of the households, calculated as simple average of wage and retail sales indices. - M2: Broad money, includes the bank deposits nominated in foreign currencies. - M: Proxy for broad money in the medium-term model that combined the available data on money aggregates into single data series. - IPI: Industrial production index. - IPI_SA_CYCLE: Hodrick-Prescott filter’s residual of the seasonally adjusted industrial production index (IPI). - C: Constant. - Notation: - _SA indicates seasonal adjustment. - _DIFF or D( ) indicates first difference of the basic variable. - (-1) or (-2) indicates first or second lag of the basic variable. - All variables were expressed as natural logarithms of their indices (with December 2003 = 100) prior to inclusion in the model exercises. ### Short-Term Model: Vector Error-Correction Estimates (Sample (adjusted): 2001M03 2004M10; Included observations: 44 after adjustments) - Cointegration restrictions: B(1,3)=0, B(1,4)=0, B(1,5)=0. - Convergence achieved after 5 iterations. Not all cointegrating vectors are identified. - LR test for binding restrictions (rank = 1): Chi-square(3) 3.058842; Probability 0.382642. - Cointegrating Eq: CointEq1 coefficients: - RPI(-1) -21.41783 - ER(-1) 19.76170 - INCOME_SA(-1) 0.000000 - M2_SA(-1) 0.000000 - IPI_SA_CYCLE(-1) 0.000000 - C 6.364260 - Error correction (coefficients on CointEq1 for D(...) ): - D(RPI): 0.003232 (standard error (0.00135); t-statistic [ 2.39126]) - D(ER): -0.000313 ((0.00062); [-0.50459]) - D(INCOME_SA): 0.021712 ((0.00455); [ 4.76765]) - D(M2_SA): 0.004225 ((0.00488); [ 0.86645]) - D(IPI_SA_CYCLE): 0.008704 ((0.00429); [ 2.02977]) - Selected lagged and contemporaneous coefficients (examples): - D(ER(-1)) on D(RPI): -0.711961 ((0.28965); [-2.45804]) - D(ER(-1)) on D(ER): 0.412164 ((0.13288); [ 3.10190]) - D(INCOME_SA(-1)) on D(RPI): 0.143375 ((0.04710); [ 3.04409]) - D(IPI_SA_CYCLE(-2)) on D(ER): -0.073850 ((0.02244); [-3.29058]) - Model fit and diagnostics: - R-squared (for equations): 0.743830 (RPI), 0.633113 (ER), 0.631108 (INCOME_SA), 0.454595 (M2_SA), 0.369681 (IPI_SA_CYCLE). - Adj. R-squared: 0.655772, 0.506995, 0.504301, 0.267113, 0.153008 respectively. - Sum sq. resids: 0.002294, 0.000483, 0.026043, 0.029856, 0.023089. - S.E. equation: 0.008467, 0.003884, 0.028528, 0.030545, 0.026861. - F-statistic: 8.447011, 5.020019, 4.976931, 2.424731, 1.706175. - Determinant resid covariance (dof adj.): 4.04E-19; Determinant resid covariance: 8.22E-20. - Log likelihood: 654.6286; Akaike information criterion -26.80130; Schwarz criterion -24.16556. - Impulse-response results are reported in Figure A2 as "Responses to Cholesky One-Standard-Deviation Innovations" for 24 periods for each variable pair (graphs included in source). ### Modified Short-Term Model: Vector Error-Correction Estimates (Sample (adjusted): 2001M03 2004M10; Included observations: 44 after adjustments) - Cointegration restrictions: B(1,4)=0, B(1,5)=0. - Convergence achieved after 14 iterations. Not all cointegrating vectors are identified. - LR test for binding restrictions (rank = 1): Chi-square(2) 18.66546; Probability 0.000088. - Cointegrating Eq: CointEq1 coefficients: - CORE_RPI(-1) -11.48962 - NONCORE_RPI(-1) -7.534333 - ER(-1) 18.99598 - INCOME_SA(-1) 0.000000 - IPI_SA_CYCLE(-1) 0.000000 - C -0.741001 - Error correction (coefficients on CointEq1 for D(...) ): - D(CORE_RPI): 0.002416 ((0.00104); [ 2.33378]) - D(NONCORE_RPI): 0.002776 ((0.00322); [ 0.86210]) - D(ER): -0.000862 ((0.00076); [-1.13360]) - D(INCOME_SA): 0.026202 ((0.00597); [ 4.39026]) - D(IPI_SA_CYCLE): 0.008001 ((0.00549); [ 1.45872]) - Selected lagged coefficients (examples): - D(NONCORE_RPI(-1)) on D(CORE_RPI): -0.101635 ((0.05413); [-1.87769]) - D(ER(-1)) on D(NONCORE_RPI): -1.499056 ((0.52953); [-2.83091]) - D(INCOME_SA(-1)) on D(CORE_RPI): 0.078887 ((0.02796); [ 2.82100]) - D(IPI_SA_CYCLE(-1)) on D(CORE_RPI): -0.084139 ((0.03254); [-2.58577]) - Model fit and diagnostics: - R-squared: 0.722595 (CORE_RPI), 0.562007 (NONCORE_RPI), 0.671435 (ER), 0.622760 (INCOME_SA), 0.385888 (IPI_SA_CYCLE). - Adj. R-squared: 0.627237, 0.411447, 0.558491, 0.493083, 0.174787 respectively. - Sum sq. resids: 0.000801, 0.007754, 0.000432, 0.026632, 0.022495. - S.E. equation: 0.005003, 0.015566, 0.003676, 0.028849, 0.026514. - F-statistic: 7.577711, 3.732782, 5.944843, 4.802411, 1.827976. - Determinant resid covariance (dof adj.): 3.52E-20; Determinant resid covariance: 7.17E-21. - Log likelihood: 708.2850; Akaike information criterion -29.24023; Schwarz criterion -26.60449. - Impulse-response results are reported in Figure A3 as "Modified Short-Term Model: Responses to Cholesky One-Standard-Deviation Innovations" for 24 periods for each variable pair (graphs included in source). ### Medium-Term Model: Vector Autoregression Estimates (Sample (adjusted): 1997M04 2004M10; Included observations: 91 after adjustments) - Dependent variables in differences: RPI_DIF, ER_DIFF, INCOME_SA_DIFF, M_SA_DIFF, IPI_SA_CYCLE_DIFF. - Selected coefficients (lag structure up to (-2)): - RPI_DIF(-1) coefficients on equations: 0.593587 (RPI_DIF) ((0.11880); [ 4.99640]), 0.455830 (ER_DIFF) ((0.27368); [ 1.66554]), 0.491676 (INCOME_SA_DIFF) ((0.22185); [ 2.21627]), 0.244750 (M_SA_DIFF) ((0.16559); [ 1.47806]), -0.059828 (IPI_SA_CYCLE_DIFF) ((0.34410); [-0.17387]). - INCOME_SA_DIFF(-1) on RPI_DIF: 0.185002 ((0.06483); [ 2.85381]). - IPI_SA_CYCLE_DIFF(-1) on INCOME_SA_DIFF: 0.354866 ((0.07951); [ 4.46329]). - M_SA_DIFF(-1) on RPI_DIF: -0.160950 ((0.08235); [-1.95442]). - Constant terms (C): - RPI_DIF: 0.007622 ((0.00547); [ 1.39405]) - ER_DIFF: 0.024783 ((0.01260); [ 1.96761]) - INCOME_SA_DIFF: 0.022968 ((0.01021); [ 2.24961]) - M_SA_DIFF: 0.017041 ((0.00762); [ 2.23612]) - IPI_SA_CYCLE_DIFF: 0.014736 ((0.01584); [ 0.93051]) - Model fit and diagnostics: - R-squared: 0.457715 (RPI_DIF), 0.095105 (ER_DIFF), 0.300216 (INCOME_SA_DIFF), 0.145359 (M_SA_DIFF), 0.142172 (IPI_SA_CYCLE_DIFF). - Adj. R-squared: 0.389929, -0.018006, 0.212743, 0.038529, 0.034944 respectively. - Sum sq. resids: 0.058931, 0.312742, 0.205497, 0.114486, 0.494373. - S.E. equation: 0.027141, 0.062524, 0.050682, 0.037830, 0.078611. - F-statistic: 6.752390, 0.840809, 3.432095, 1.360652, 1.325880. - Determinant resid covariance (dof adj.): 3.69E-14; Determinant resid covariance: 1.94E-14. - Log likelihood: 791.0663; Akaike information criterion -16.17728; Schwarz criterion -14.65973. - Impulse-response results are reported in Figure A4 as "Medium-Term Model: Responses to Cholesky One-Standard-Deviation Innovations" and "Accumulated Response" for up to 12 periods for each variable (graphs included in source). *Source: _cr05232 - introduction of cash registers at the beginning of the year.* ### References ### _cr05232 - References ### References (selected) - Belaisch, Agnès, 2003, “Exchange Rate Pass-Through in Brazil,” IMF Working Paper 03/141 (Washington: International Monetary Fund). - Gueorguiev, Nikolay, 2003, “Exchange Rate Pass-Through in Romania,” IMF Working Paper 03/130 (Washington: International Monetary Fund). - Kuijs, Louis, 2001, “Monetary Policy Transmission Mechanisms and Inflation Modeling in Slovakia,” Chapter III of Slovak Republic: Selected Issues and Statistical Appendix, IMF Country Report No. 01/129 by Costas Christou and others (Washington: International Monetary Fund). - Mattina, Todd, 2004, “Understanding the Serbian Exchange-Rate Based Stabilization” (unpublished; Washington: International Monetary Fund). - Maliszewski, Wojciech, 2003, “Inflation in Georgia,” Chapter I of Georgia: Selected Issues and Statistical Appendix, IMF Country Report No. 03/347 by Paulo Neuhaus and others (Washington: International Monetary Fund). - Mihaljek, Dubravko; Klau, Marc, 2004, “The Balassa–Samuelson Effect in Central Europe: A Disaggregated Analysis,” in Comparative Economic Studies, No.46, pp. 63–94. - NBS, 2004a, “What Influences the Inflation Rate in Serbia?” Economic Review, National Bank of Serbia (July) – pp. 29–39. - NBS, 2004b, “Inflation – Measuring Methods,” Economic Review, National Bank of Serbia (October) – pp. 30–35. - Peiris, Shanaka J., 2003, “Inflation Dynamics in Vietnam,” Chapter I of Vietnam: Selected Issues, IMF Country Report No. 03/381 by Olaf Unteroberdoerster and others (Washington: International Monetary Fund). - Sommer, Martin, 2004, “Inflation in Slovenia: Recent Developments and Outlook,” Chapter I of Republic of Slovenia: Selected Issues and Statistical Appendix, IMF Country Report No. 04/149 by Jarkko Soikkeli and Martin Sommer (Washington: International Monetary Fund). - Stavrev, Emil, 2003, “The Pass-Through from the Nominal Exchange Rate to Inflation,” in Russian Federation: Selected Issues, IMF Country Report No. 03/146 by Angana Banerji and others (Washington: International Monetary Fund). - Tzanninis, Dimitri, 2001, “Modeling Inflation in Czech Republic: Short-Run and Long-Run Dynamics,” in Czech Republic: Selected Issues and Statistical Appendix, IMF Country Report No. 01/112 by Dimitri Tzanninis and Rachel van Elkan (Washington: International Monetary Fund). - Vacher, Jérôme, 2004, “Estimates of Exchange Rate Pass-Through,” in Republic of Belarus: Selected Issues, IMF Country Report No.04/139 by Veronica Bacalu and others (Washington: International Monetary Fund). - Xiao, Yuan, 2004, “Modeling Inflation in Latvia,” Chapter II of Republic of Latvia: Selected Issues, IMF Country Report No. 04/261 by Francesco Luna and others (Washington: International Monetary Fund). ### EUROIZATION IN SERBIA: MACROECONOMIC, PRUDENTIAL, AND POLICY IMPLICATIONS — Key findings (Sections A–D, as provided) - Scope and prevalence - Financial euroization is prevalent, with more than two thirds of total deposits denominated in euro, and at least two thirds of all loans either denominated in euro or indexed to the euro. - Payments euroization is common for big ticket items, and many prices are de facto linked to the dinar (SRD)/euro exchange rate. - Main costs and channels of vulnerability - With most new bank lending indexed to the euro, banks face indirect credit risk because borrowers without euro cash flow are vulnerable to unexpected exchange rate depreciation. - High euroization reduces the effectiveness of monetary policy and promotes a “fear of floating.” - Euroization compounds banking sector vulnerabilities and limits macroeconomic policy options. - Objective of the chapter - The chapter does not prescribe an approach directly aiming at reducing euroization; rather it makes a case for containing euroization and minimizing associated risks. - Organization: Section B — recent trends and international comparison; Section C — causes for limited use of the local currency; Section D — benefits, costs, and key risks; Section E — policy options for containing euroization and managing risks. - Key definitions (Box 1) - Financial dollarization: residents hold financial assets or liabilities in foreign currency-denominated instruments; asset substitution and liability dollarization defined. - Payments dollarization: use of foreign exchange as a medium of exchange. - Real dollarization: domestic prices and wages indexed to the exchange rate. - Official dollarization vs de facto dollarization (bicurrency system). - Modern theories: dollarization as risk hedging in a bicurrency system; degree of dollarization tends to increase with inflation; financial dollarization emerges ahead of payments dollarization at lower levels of inflation. - Recent developments and statistics (Section B) - Share of foreign currency-denominated deposits in total deposits rose from 54 percent in 2002 Q3 to 67 percent at end-2004. - Remittances increased from 11 percent in 2002 to 14 percent in 2004 (relative to GDP). - The share of foreign currency-denominated and foreign currency-indexed loans in total loans to the nongovernment sector has risen to 70 percent. - Pass-through from exchange rate changes to inflation estimated at 0.7 within a 12-month period (Chapter IV). - The share of administrative prices in the RPI basket is 40 percent. - Payments euroization is largely limited to high value items such as real estate, cars, and high-end consumer goods; property transactions and rents typically priced and settled in euro. - Even compared to regions with a long legacy of dollarization, Serbia’s financial euroization is high: in 2001 South America had close to 60 percent of deposits in foreign currency; Serbia already had the highest level of liability dollarization among comparator economies. - Explanations for high and rising euroization (Section C) - Historical loss of confidence from the 1990s, including: freezing of foreign currency deposits in the early 1990s and subsequent securitization into bonds with below-market coupons; hyperinflation in 1993. - Negative real interest rates on dinar deposits have been present since 2003. - Since early 2004, derived real interest rates on euro deposits in Serbia have exceeded those on dinar deposits. - Monetary tightening since mid-2004 was insufficient to prevent real deposit interest rates from falling when headline and core inflation more than doubled during the 12-month period through April 2005. - If a positive real deposit rate of 2 percent is assumed to be sufficient to encourage dinar holdings, actual monetary policy rates in Q4 2004 and Q1 2005 should have been higher by at least 400 basis points. - The share of NPLs in total loans at end-2004 was 22 percent. - D-mark banknote conversion to euro-denominated deposits estimated to have increased the stock of euro deposits by around €2–3 billion. - Market entry of foreign banks preferring euro-denominated transactions and the relatively smooth depreciation of the dinar since early 2003 encouraged euro-denominated foreign borrowing and euro-indexed lending. - Improvements in Serbia’s EU accession prospects may have supported the preference for euro denomination. - A hysteresis and ratchet effect: past instability and network externalities have likely perpetuated euroization despite macroeconomic stabilization. - Interest rate and policy transmission observations (Section C and D lead-in) - Real interest rates on total household and enterprise deposits have been continuously negative since early 2003. - Calculation notes: real interest rates are based on a six-month rolling average for monthly RPI data, using the past four and the next two months. - The monetary policy interest rate was the interest rate on NBS bills through end-January 2005, when issuance was replaced with repo operations. - Statutory reserve requirements (SRR) noted: currently at 47 percent for foreign currency-denominated household deposits and 26 percent for other foreign currency-denominated deposits and commercial banks’ foreign borrowing; and a 20 percent withholding tax for nonresidents (10 percent for nonresidents in countries that concluded a double-taxation agreement with Serbia). Italic: Source: _cr05232 - References. ### 14. In the context of an uncertain political environment and an incomplete ### _cr05232 - 14. In the context of an uncertain political environment and an incomplete ### Overview and benefits of euroization - Euroization has supported remonetization and intermediation amid an uncertain political environment and incomplete transition. - Confidence in the banking system has increased markedly due to progress in macroeconomic stabilization and the market entry of foreign banks. - Continued rise in euro-denominated deposits has facilitated remonetization of the economy, creating network benefits that are welfare enhancing. - Rapid growth in bank credit has contributed to economic growth. - Euroization has allowed portfolio diversification in a nascent capital market: - The T-bill market remained small with short-term maturities due to declining government financing needs. - Issuance of commercial bonds and papers, and stock market turnover, are very modest. - The largest market is for euro-denominated Frozen Foreign Currency Deposit Bonds (FFCDs). - Risk diversification arising from euroization has improved agents’ risk minimization strategies. ### Costs and vulnerabilities of euroization - Increased credit risk from borrower exposure to exchange rate risk: - A large share of euro-indexed lending is to borrowers without sufficient foreign currency-denominated cash flow; the export sector accounted for only 17 percent of GDP in 2004. - A sizeable, unexpected depreciation, or surge in exchange rate volatility, would raise debt service burdens, increasing the NPL ratio. - Second-round effects, including a potential credit crunch from economic slowdown, would further deteriorate loan portfolio quality. - Depreciation would reduce collateral values, forcing increased provisioning. - Possible flight to quality—transfer of euro-denominated deposits to foreign banks—could undermine vulnerable domestic banks. - Impact on banks’ open foreign exchange positions: - Footnote notes the impact appears remote because the Serbian banking system generally runs a long foreign exchange position and would realize a net dinar balance-sheet gain from a depreciation. - Constraints on monetary policy effectiveness: - The NBS has been increasingly unable to control the expansion of broad money and credit. - Tightening since mid-2004 produced a moderate decline in reserve money, but strong broad money growth continued as expansion was driven by euro-denominated or euro-indexed assets and liabilities. - To compensate for weak interest rate transmission, the NBS relied heavily on statutory reserve requirements (SRRs) despite introducing repo operations in January 2005. - SRRs are already high, limiting scope for further increases given adverse consequences for financial intermediation. - Limits on exchange rate policy: - High euroization can generate “fear of floating,” making authorities reluctant to adopt more flexible exchange rate arrangements. - Reduced seigniorage revenues: - Assuming about €3 billion in cash circulate in Serbia (about 16 percent of GDP), replacing an equivalent amount of dinar-denominated cash, the foregone seigniorage revenue would amount to about 0.6 percent of GDP (16 percent of GDP times 3½ percent assumed interest rate on euro-denominated German bonds). - Footnote clarifies this estimate represents the minimum of foregone seigniorage and could be higher if money is created with backing of domestic assets. - Systemic liquidity risks from high liability dollarization: - Euro deposits are only partially covered by liquid euro assets; full deposit withdrawals create high liquidity risk not offset by required reserves on foreign-currency deposits. - A surge in Serbia’s country risk could induce depositors to convert euro deposits into cash euro or transfer them abroad; foreign banks could cut credit lines and freeze exposure. - Without liquid euro assets abroad backing liquid euro liabilities, banks may run out of liquid euro-denominated reserves and fail to pay euro liabilities as they fall due. - Constraints on the central bank’s Lender of Last Resort (LOLR) function: - In April 2005, gross foreign exchange reserves of the NBS adequately covered commercial banks' foreign currency-denominated deposits with a coverage ratio of 150 percent. - Gross foreign exchange reserves amounted to US$4.47 billion; the deposit base was US$2.97 billion, of which US$1.2 billion were held at the NBS as required reserves. - Thus, gross foreign exchange reserves can cover a deposit run of the remaining free deposits in the amount of US$1.77 billion. - Any sudden drawdown of foreign exchange reserves during a bank run could erode confidence in the exchange rate regime and contribute to a “twin crisis” where banking and exchange rate crises interact. ### Policy issues and recommended strategy - Main policy challenge: minimize risks emanating from euroization. - Key pillars to mitigate risks: - (i) Sound macroeconomic policies to contain euroization at least at the current level. - (ii) Tighter macro-prudential measures to internalize negative externalities of euroization for society. - (iii) Micro-prudential measures to minimize inherent banking-sector risks. - Greater exchange rate flexibility in a supportive macro environment could reduce euroization by creating a two-way risk in the foreign exchange market. - Containing or reducing euroization is reinforced by the medium-term goal of EU and EMU accession. ### Macroeconomic policy recommendations - Build policy credibility—especially by lowering inflation—to contain euroization; sustained implementation of sound macroeconomic policies is first best. - Policies to contain domestic demand and accelerate structural reforms are imperative to reduce the current account deficit to sustainable levels; otherwise demand for dinar assets and the exchange rate will remain sensitive to investor sentiment. - Strengthen the monetary policy framework to enhance confidence in the dinar: - Ensure positive real interest rates on dinar deposits through more proactive monetary policy, including stepping up repo operations to raise monetary policy interest rates during rising inflation. - Reinforce interest rate transmission via institutional measures: the ceiling and floor of the interest rate corridor should be market determined, achieved either by linking the reference discount rate to the market-determined repo rate or by sufficiently frequent adjustments of the discount rate by the Monetary Board. - Strengthen the regulatory and supervisory framework and increase competition to alleviate concerns about higher interest rates increasing NPLs; more competition should lower spreads and allow higher deposit rates without equivalent increases in lending rates. - Consider establishment of an alternative nominal anchor such as inflation targeting over time; successful de-dollarization in other countries has been associated with credible inflation targeting regimes. ### Macro-prudential recommendations - Internalize social costs of euroization by charging higher costs on euro transactions relative to dinar ones as a second-best market intervention: - Reduce remuneration on required reserves in households’ foreign currency deposits to zero. - Increase SRRs on foreign currency-denominated enterprise deposits and commercial banks’ foreign borrowing. - If euro-indexed lending continues at an unsustainable pace, consider: - Marginal SRRs on foreign currency-denominated deposits that rise if deposits exceed thresholds, effectively limiting the credit multiplier for additional deposit taking. - Capital controls (e.g., price-based controls requiring deposits of short-term inflows at the central bank at sub-market remuneration) only as last resort due to short-term effectiveness and distortion risks. - Credit ceilings are costly given ongoing structural change and market share variations; they would complicate a welcome market development. ### Micro-prudential recommendations - Improve regulatory and supervisory framework to ensure appropriate assessment and management of risks from asset and liability euroization: - Higher provisioning requirements for foreign exchange-denominated and foreign exchange-indexed lending to raise cost of euro loanable funds relative to dinars. - Higher general provisioning on foreign exchange-denominated and foreign exchange-indexed loans as a flexible solvency buffer convertible into specific provisions without triggering capital adequacy issues. - Require banks’ boards of directors to issue written policies and procedures to monitor and manage credit risk from borrowers’ exchange rate exposure, with internal controls and audits to ensure compliance. - Require banks to assess extent of hedging by borrowers with euro-denominated or euro-indexed debt and to evaluate debt-service capacity under various depreciation scenarios. - Inform borrowers explicitly about servicing risks for euro-indexed loans, possibly by using a range of hypothetical depreciation rates for the dinar/euro exchange rate. - Strengthen NBS data-gathering to include detailed information on the stock of euro-denominated and euro-indexed loans and the extent to which respective debtors are hedged. - Implement corrective measures against banks with weak risk monitoring or management regarding foreign currency lending. ### International experience (Box 2 summary) - De-dollarization is difficult; successful cases are typically by-products of macroeconomic stabilization. - Studies identify few successful de-dollarizers; forced conversion experiences have generally been unsuccessful. - Voluntary de-dollarization tends to follow credible anti-inflationary policies, fiscal adjustment, and sometimes inflation targeting within a flexible exchange rate regime. - Institutional measures to support local currency demand (e.g., market-determined deposit rates, inflation-indexed bonds) can help. - Until macro stability is achieved, policy should focus on managing solvency and liquidity risks from dollarization via prudential measures such as: - Differential SRRs on foreign currency deposits. - Higher liquidity requirements on foreign currency deposits. - Higher provisioning or capital charges on foreign currency lending. *Source: National Bank of Serbia data; and IMF staff calculations.* ### 1. The large external imbalances of Serbia and Montenegro (SM) have raised ### The large external imbalances of Serbia and Montenegro (SM) have raised questions about external competitiveness ### Overview and key findings - The current account deficit before grants has hovered at 12–13 percent of GDP in the last few years, driven by a trade deficit of around 23–25 percent of GDP, which was only partly compensated by an exceptional level of remittance inflows. - The chapter concludes that SM’s relatively weak export performance is related more to non-price factors than to price factors, such as the level of the exchange rate. - Limited availability and quality of data and exceptional historical factors constrain the scope of the analysis. ### Export performance in historical and regional perspective - SM exports of goods and services averaged about 22 percent of GDP (2001─04), compared to a regional average of 38 percent of GDP; only Albania had a lower average in the region. - Imports of goods in percent of GDP were close to the regional average (Table 1), while imports of goods and services were significantly below the average, even though they still amount to 50 percent of GDP. - Persistently lower growth rate of exports compared to the growth rate of imports is a major concern. - Historical events that weighed on export performance: - International isolation during the Milosevic years and UN and European Economic Community sanctions in the first half of the 1990s. - Breakup of the former Yugoslavia and the fragmentation of vertically integrated enterprises (e.g., Zastava group). - NATO bombings during the Kosovo war causing large damage to physical infrastructure and factories. - Non-membership in major multilateral organizations (IMF, WTO) negatively affected trade relations and external financing. - A gravity model confirms that SM trade has significantly been sub–potential given its geographic position and access to the sea. ### Direction of exports (selected patterns) - The main destination of Serbian exports since 1997 has been Bosnia and Herzegovina, followed by other former Yugoslav republics, indicating persistence of pre-existing trade flows despite dissolution and wars. - Principal importing countries also include Italy, Germany, France, and Russia. - In the last four years (up to 2004) the share of exports to the EU3 has been losing ground to exports to FY republics, suggesting possible loss of competitiveness in quality-oriented European markets. - SM exports to industrial countries as a share of total exports of former Yugoslav republics were very low until 1996 and then had a strong increase following the end of the sanctions; relative share increased to around 10 percent in the years up to 2004. ### Composition of exports and imports - SM has not significantly changed the pattern of specialization inherited from Yugoslavia: exports are concentrated in intermediate products, commodities, and agricultural products. - Exports (1997─2004 averages, in percent of total): - Intermediate goods: 61 percent (97-00 average), 57 percent (00-04 average); specific annual figures include 64, 67, 55, 56, 53, 54, 55, 65 for 1997–2004 respectively. - Capital goods: 6 percent (97-00 average), 10 percent (00-04 average); specific annual figures 5, 6, 7, 8, 9, 8, 15, 7 for 1997–2004 respectively. - Consumption goods: 33 percent (97-00 average), 33 percent (00-04 average); specific annual figures 31, 27, 38, 36, 38, 38, 30, 28 for 1997–2004 respectively. - Imports (1997─2004 averages, in percent of total): - Intermediate goods: 69 percent (97-00 average), 60 percent (00-04 average); specific annual figures 68, 68, 68, 70, 67, 61, 57, 56 for 1997–2004 respectively. - Capital goods: 15 percent (97-00 average), 19 percent (00-04 average); specific annual figures 13, 15, 16, 16, 16, 20, 19, 21 for 1997–2004 respectively. - Consumption goods: 16 percent (97-00 average), 21 percent (00-04 average); specific annual figures 19, 17, 15, 14, 17, 19, 24, 23 for 1997–2004 respectively. - Intermediate products consistently represented around 60 percent of exports since 1997, with a decrease after 2000 and strengthening in 2004 (notably steel and rubber). ### Price competitiveness: exchange rates, REER, and wages - The exchange rate collapsed after the Kosovo war and took approximately two years (end-2000 to end-2002) to stabilize following the exchange rate-based stabilization program of end-2000; selecting 1999 or 2000 as base years without considering these exceptional circumstances can lead to erroneous conclusions about misalignment. - Best candidate base years for REER comparisons: 1997 (and to a lesser extent 2001), though both years have limitations due to exceptional factors and data issues. - Real exchange rate indicators: - On average in 2004 both the CPI-based and the RPI-based REER indices are around 25 percent lower than the average 1997 values. - Compared to the 2001 levels, both indices are significantly higher (by 19 and 34 percent, respectively). - Tentative conclusion: the real exchange rate does not seem excessively appreciated; there may be space for real appreciation in coming years consistent with a Balassa-Samuelson effect from productivity increases. - Caveat: these relatively favorable REER levels were being threatened in 2005 by still high inflation coupled with a rather stable nominal exchange rate. - Studies of equilibrium exchange rate: - Mattina (2004) concluded that the exceptionally large real appreciation starting in 2000 was mostly due to inflation inertia and adjustment to 1999-2000 disequilibrium conditions rather than to growth in domestic demand, and that the end-2003 exchange rate was still below the equilibrium level. - Unit labor costs (ULC) and wages: - Statistical base for ULC is weak; yearly net wages in proportion to per capita GDP are used as a rough proxy for ULC for cross-country comparisons. - In 2003 the value of this ULC proxy in SM was 0.96 percent. - Comparators: 1.24 percent average of all transition countries; 1.30 percent value of the countries in the region. - Lower-than-average USD wages are the main source of the difference. - The ULC proxy for SM is higher than in more successful neighboring countries such as Bulgaria, Romania, and Slovenia (with the exception of Croatia). ### Non-price competitiveness and policy implications (summary from chapter framing) - The chapter indicates that non-price factors (investment environment, governance, and other structural issues) are important contributors to weak export performance; detailed analysis of these factors appears in the chapter’s third section (not reproduced here). - Improving the trade balance and export performance is identified as a key priority for policy makers; an assessment of export competitiveness can inform the exchange rate policy debate, which has shifted between current account and inflation objectives in recent years. *Prepared by Carlo Sdralevich. Sources: National authorities; IMF staff estimates; National Bank of Serbia.* ### 12. ULCs in SM are consistent with its transition stage. Relating the proxy ULC to the ### 12. ULCs in SM are consistent with its transition stage. Relating the proxy ULC to the ### Wage and unit labor cost (ULC) findings - Relating the proxy ULC to the stage of transition, as measured by the average of the EBRD eight transition indicators (excluding infrastructure), shows that SM is below the level of countries at similar points in the transition process (Figure 4), although the absence of a clear correlation between the two variables in the universe of all the transition countries raises doubts as to the meaningfulness of this observation. - A strongly positive correlation is apparent between the EBRD transition indicator average and monthly USD wages, with more advanced countries displaying higher proxy ULCs (Figure 5). SM seems to be in line with this relationship, if anything with slightly high wages in relation to its transition stage. - A related staff study plotting gross USD wages against GDP per employee for transition countries (including Balkan countries) shows a strong increasing relationship between gross wages and productivity, and that, against this fitted relationship, SM is undervalued by around 10 percent. ### Recent trends in real and nominal wages - CPI-deflated real wages for SM remained around the levels of 1997 until the end of 2001, when they began an increasing trend. - In Serbia: - Overall real wages and manufacturing sector wages (in euro terms) appreciated starting in 2001 but flattened out in the course of 2004. - Wages in the manufacturing sector followed the same pattern (Figure 6). - The wage trend is common to both the so-called economy (private, socially owned enterprises and state-owned enterprises) and non-economy sector (public sector excluding state-owned enterprises). - Implication: SM’s relative competitiveness up to 2003 in a static comparison of wages and proxy-ULCs across countries could have been eroded by rising wages and would need to be compensated by a corresponding increase in productivity. ### Labor quality and demographic effects - The still low wage level is accompanied by a highly skilled work-force; average education level is relatively high, similarly to other countries from the former Yugoslavia. - SM suffered an extensive brain-drain during the 1990s, concentrated particularly in the highest-educated layers, which halved the number of PhDs and Masters-level employees in the period 1995-2000 (Jovicic et al. 2001). ### Non-price competitiveness factors and investment climate - Price competitiveness does not sufficiently explain the poor SM export performance; causes likely lie in non-price factors: - Poor quality of the investment climate can increase costs of doing business, affect competitiveness of SM enterprises, and work against creation of new export-oriented enterprises and FDI. - SM inherited a weak productive base and began the transition process with a significant lag due to international isolation. - Evidence on the investment climate: - Mixed but overall less than satisfactory assessments, particularly in perception of investment climate and quality of governance. - FIAS “Doing Business” data: - Difficulty of starting a business in SM is in line with Balkan and FRY countries, though minimum capital required (as percentage of per capita GNI) is significantly higher. - Closing a business is particularly expensive but not longer than in comparable countries. - Labor flexibility seems better than average in easiness of hiring and low cost of separation; firing is as difficult as the average Balkan and transition countries. - Cost of registering property is faster than average but more expensive. - Enforcing contracts is much more difficult, with an average of 1,000 days for procedure, almost double the average for other countries in the region. - Other indicators: - SM scores low on the EBRD Transition Indicators compared to regional peers, and has the lowest share of private sector in GDP, together with Bosnia and Herzegovina. - World Economic Forum ranks SM at the 89th position on the 2004–05 Growth Competitiveness Index (GCI) and 85th on the Business Competitiveness Index (BCI) out of 104 countries. - The 2004 Corruption Perceptions Index (CPI) by Transparency International ranks SM at the 97th place, with the same score as Macedonia and the lowest ranking among former Yugoslavia countries. - Local surveys (Jefferson Institute 2003) highlight constraints: lack of business sophistication (e.g., marketing), insularity of enterprises, lack of credit access, inadequately protected trade regime, inadequate infrastructures and technology, and excessively volatile regulatory environment. - FDI: - Non-privatization FDI in SM is much below potential; in 2003 SM ranked 10th in terms of FDI stock among 14 South Eastern Europe countries, clustered in the second-last performing group with Bulgaria and Slovenia. - Only Bosnia and Herzegovina, Albania, FYR Macedonia, and Moldova performed worse. - Demekas et al. (2004) find SM has a non-privatization FDI potential much larger than actual performance; implication is that better policies would help increase FDI attractiveness. ### Policy recommendations to improve competitiveness - Focus on non-price factors while maintaining control over real exchange rate dynamics: - Continue the process of liberalization of the trade regime and conclude “open” trade agreements regionally and with the EU. - SM has concluded six free trade agreements with Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Moldova, and Romania, and is working on an agreement with FYR Macedonia. - SM has a free trade agreement with Russia and has obtained EU trade preferences under the Autonomous Trade Preferences regime for Western Balkans; special agreements exist in sugar and textiles. - Accession to WTO is an important forthcoming step; Serbia and Montenegro are preparing separately. - Note: simple and weighted average tariffs in Serbia are expected to be 8.7 percent and 6.3 percent respectively after the ongoing (as of June 2005) realignment of tariffs. There remain some non-tariff barriers, notably the ban on the import of refined oil derivatives in Serbia. - Address the quality of the business environment: - Accelerate the transition process by intensifying privatizations, improving provision of services and infrastructures, and increasing efficiency of the financial sector. - Be cautious with export-specific initiatives (export insurance agencies, subsidized credit, export processing zones) because they can increase distortions and induce rent-seeking; economy-wide reforms are generally more effective. - The Union government launched the Serbia and Montenegro Export Credit Agency (SMECA) with World Bank help; the Serbian government is setting up an export-promotion agency that will insure exports and provide subsidized credit to exporting companies. - Improve the image of SM abroad: - Government established an Export and Investment Promotion Agency (SIEPA). - Most significant progress likely to come from advances in political relations with potential investor countries. ### Conclusions - Export performance in SM has been disappointing and is likely more related to non-price than price competitiveness factors. - Roots of competitiveness problems trace to economic isolation, sanctions, conflict in the 1990s, pre-existing economic crisis, and chronically low investment. - Competitiveness depends on non-price factors underlying the investment climate and capacity constraints due to low past investment and inherited economic structure, rather than primarily on real exchange rate or wage levels. - Best policy approach: - Accelerate and deepen economic reforms to unleash private investment and normalize trade relations. - Contain dynamics of the real exchange rate through anti-inflationary macroeconomic policies and appropriate exchange rate management to preserve any price advantage while allowing real incomes to rise in line with productivity improvements. *Source: IMF staff analysis as presented in the chapter text.* ### References ### _cr05232 - References ### References - Christie, Edward, 2002, “Potential Trade in Southeast Europe: a Gravity Model Approach,” WIIW Working Papers No. 21, March 2002. - Demekas, Dimitri G., Balasz Horvath, Elina Ribakova, and Yi Wu, 2004, Foreign Direct Investment in Southeastern Europe: How (and How Much) Can Policies Help?, (draft), International Monetary Fund. - Jefferson Institute, 2003, Competitiveness of the Serbian Economy, Jefferson Institute, Belgrade, Serbia. - Jovicic, Milena, Miroslav Zdravkovic, and Radmila Dragutinovic Mirovic (2001): Federal Republic Yugoslavia: Trade Potentials and Comparative Advantages. - Mattina, Todd, 2004, Understanding the Serbian Exchange Rate-Based Stabilization, (unpublished), European Department, International Monetary Fund. - Popovic-Avric, Snezana, 1999,“Foreign Trade, 1990–1998,” Yugoslav Survey No. 3. - Popovic-Avric, Snezana, 2000, “Competitiveness of Yugoslav Exports and Courses of Actions Towards Improving It,” Yugoslav Survey No. 3. - World Bank, 2004, Republic of Serbia: An Agenda for Economic Growth and Employment, Poverty Reduction and Economic Management Unit, Europe and Central Asia Region. - World Economic Forum, 2004, Global Competitiveness Report 2004–05. ### Key statistics (selected, Serbia and Montenegro, 1999–2004) - GDP (in millions of U.S. dollars): 1999: $10,214; 2000: $8,603; 2001: $11,576; 2002: $15,528; 2003: $20,665; 2004: $23,996 (Est.). - GDP (in billions of dinars): 1999: 193; 2000: 382; 2001: 772; 2002: 998; 2003: 1,189; 2004: 1,401. - Real GDP (real growth rate, annual percentage change): 1999: -18.0; 2000: 5.0; 2001: 5.0; 2002: 5.5; 2003: 3.8; 2004: 2.7; (final line shows 7.2 for 2004 elsewhere—source lists multiple lines; primary table reports "2.7.2"? The table lists "Gross domestic product -18.0 5.0 5.0 5.5 3.8 2.7 7.2".) - Industrial production (annual percent change): 1999: 4.4; 2000: 11.1; 2001: 0.0; 2002: 1.7; 2003: -2.7; 2004: 7.5. - Retail prices (annual average, percent): 1999: 42.1; 2000: 69.9; 2001: 91.1; 2002: 21.2; 2003: 11.3; 2004: 9.5. - Unemployment rate (percent, ILO from 2004 onwards, excluding Montenegro): 1999: 13.3; 2000: 12.1; 2001: 12.2; 2002: 13.3; 2003: 14.6; 2004: 18.5. - Average net wage (in euros per month): 1999: 8; 2000: 17; 2001: 102; 2002: 149; 2003: 175; 2004: 192. - Average net monthly pay per employee (dinars, Total): 1999: 1,309; 2000: 2,588; 2001: 5,545; 2002: 9,196; 2003: 11,486. - Average net monthly pay per employee (dinars, 2004 Est. in Table 6/other table): 2004 Dec. listed as 17,131 dinars (Table 6 shows monthly dinar values and euros). - Exports of goods (f.o.b., in billions of U.S. dollars): 1999: 1.7; 2000: 1.9; 2001: 2.0; 2002: 2.4; 2003: 3.1; 2004: 4.2. - Imports of goods (c.i.f., in billions of U.S. dollars): 1999: -3.3; 2000: -3.7; 2001: -4.8; 2002: -6.3; 2003: -7.9; 2004: -11.7. - Trade balance (in billions of U.S. dollars): 1999: -1.6; 2000: -1.8; 2001: -2.8; 2002: -3.9; 2003: -4.9; 2004: -7.4. - Current account balance after grants (in billions of U.S. dollars): 1999: -0.8; 2000: -0.3; 2001: -0.5; 2002: -1.4; 2003: -1.5; 2004: -3.1. - Current account balance (in percent of GDP): 1999: -7.5; 2000: -3.9; 2001: -4.6; 2002: -8.9; 2003: -7.3; 2004: -13.1. - External debt (year-end, percent of GDP): 1999: 10.7; 2000: 11.4; 2001: 11.9; 2002: 11.8; 2003: 14.3; 2004: 14.9. - Gross official reserves (in billions of U.S. dollars): 1999: 0.3; 2000: 0.5; 2001: 1.2; 2002: 2.3; 2003: 3.6; 2004: 4.3. - Gross official reserves (months of imports of goods and services): 1999: 0.9; 2000: 1.2; 2001: 2.4; 2002: 3.1; 2003: 3.3; 2004: 3.7. - Money supply (end-of-period percentage change): M1 1999–2004: 47.3; 85.2; 125.2; 79.8; 10.9; 8.0. M2 1999–2004: 67.6; 61.4; 104.9; 52.7; 27.5; 30.3. ### Social and demographic indicators (selected) - Population (Serbia and Montenegro): 1999: 10,577; 2000: 10,600; 2001: 10,617; 2002: 10,629; 2003: 10,633; 2004: 10,651. - Population (Serbia): 1999: 9,938; 2004: 9,993. - Population (Montenegro): 1999: 640; 2004: 658. - Population ages 65 and above (percent): 1999: 13; 2000: 13; 2001: 13; 2002: 13; 2003: 14; 2004: 14. - Life expectancy at birth (years): 1999: 72; 2000: 72; 2001: 72; 2002: 72; 2003: 73; 2004: 73. - Infant mortality rate (per 1000 live births): 1999: 15; 2000: 14; 2001: 14; 2002: 17; 2003: 13; 2004: (data gap in table). - Fixed line and mobile phone subscribers (per 1,000 people): 1999: (not listed); 2000: (not listed); 2001: 271; 2002: 349; 2003: 416; 2004: 489 (Table 2 entries show progression to 581 and comparator Visegrad-4 1,034). ### GDP by sector (selected, current prices in billions of dinars, 1999–2004) - Gross domestic product (current prices): 1999: 192.9; 2000: 381.7; 2001: 771.8; 2002: 899.8; 2003: 1,189.1; 2004: 1,400.6. - Manufacturing (current prices): 1999: 43.0; 2000: 79.3; 2001: 153.2; 2002: 166.9. - Transport, storage, and communication (current prices): 1999: 16.4; 2000: 25.5; 2001: 54.9; 2002: 78.0. - Financial intermediation (current prices): 1999: 10.8; 2000: 16.5; 2001: 39.7; 2002: 55.6. - Gross value added (basic prices): 1999: 178.3; 2000: 358.8; 2001: 702.4; 2002: 857.6. ### Fiscal operations (selected) - Serbia and Montenegro general government revenue (in percent of GDP, table header): Revenue series show "...36.7 38.9 42.8 42.7 45.2" and Expenditure "...37.6 40.3 47.3 46.0 45.5" (table reports ellipses for some years). - Table 11 (General Government Fiscal Operations, 2000–04, in billions of dinars): - Total revenue: 2000: 140.2; 2001: 300.4; 2002: 430.5; 2003: 507.9; 2004: 633.0. - Total expenditure and net lending: 2000: 143.6; 2001: 310.2; 2002: 475.9; 2003: 547.4; 2004: 637.1. - Overall balance (in billions of dinars): 2000: -3.4; 2001: -9.8; 2002: -45.4; 2003: -39.5; 2004: -4.1. - Foreign grants (in billions of dinars): 2000: 2.8; 2001: 5.4; 2002: 11.3; 2003: 2.7; 2004: 1.4. - Overall balance including grants: 2000: -0.6; 2001: -4.4; 2002: -34.1; 2003: -36.8; 2004: -2.7. - Serbian general government fiscal operations (Table 12, in billions of dinars): - Total revenue: 2000: 128.8; 2001: 277.1; 2002: 402.7; 2003: 474.9; 2004: 592.7. - Total expenditure and net lending: 2000: 129.6; 2001: 283.9; 2002: 445.3; 2003: 511.9; 2004: 593.0. - Overall balance: 2000: -0.8; 2001: -6.8; 2002: -42.6; 2003: -34.9; 2004: -0.3. - Overall balance including grants: 2000: -0.7; 2001: -3.4; 2002: -33.0; 2003: -33.0; 2004: 0.5. - Montenegro consolidated fiscal operations (Table 17, in millions of euros, 2001–04): - Total revenue: 2001: 339.6; 2002: 399.1; 2003: 456.7; 2004: 501.5; 2004 Est.: 558.2. - Total expenditure and net lending: 2001: 397.6; 2002: 443.6; 2003: 503.0; 2004: 542.1; 2004 Est.: 577.4. - Overall balance after grants: 2001: 0.2; 2002: -17.0; 2003: -22.8; 2004: -58.3; 2004 Est.: -47.5. ### Monetary and financial sector (selected) - National Bank of Serbia net foreign reserves (end-period, in millions of dinars): 1999: -22,549; 2000: -19,501; 2001: -133; 2002: 62,304; 2003 Dec.: 90,840; 2004 Dec.: 114,193. - Net foreign reserves (in millions of euros): 1999: -549; 2000: -332; 2001: -2; 2002: 1,044; 2003: 1,330; 2004: 1,448. - Reserve money (National Bank of Serbia, end-period in millions of dinars): 1999 Dec.: 9,467; 2000 Dec.: 19,845; 2001 Dec.: 41,361; 2002 Dec.: 69,324; 2003 Dec.: 69,996; 2004 Dec.: 76,969. - Currency in circulation: 1999: 6,688; 2000: 10,933; 2001: 25,324; 2002: 43,719; 2003: 42,979; 2004: 45,165. - Reserve deposits: 1999: 2,779; 2000: 8,912; 2001: 16,037; 2002: 25,605; 2003: 27,017; 2004: 31,804. - Monetary survey (M2, end-period, Serbia, in millions of dinars): 1999: 40,587; 2000: 59,396; 2001: 112,280; 2002: 186,706; 2003: 236,934; 2004 Dec.: 308,725. - Dinar-denominated M2 2004 Dec.: 132,249. - Foreign-currency deposits (non-frozen) 2004 Dec.: 176,476 (millions of dinars). - Deposit money banks' assets (Serbia, Table 23, in millions of dinars): 2002 Dec.: 259,222; 2003 Dec.: 299,200; 2004 Dec.: 428,624 (Est.). - Reserves and deposits at National Bank of Serbia (2004 Dec.): 99,526. - Foreign assets (2004 Dec.): 53,941. - Claims on other domestic sectors (2004 Dec.): 261,131. - Foreign liabilities (2004 Dec.): 80,067. ### Interest rates and banking (selected) - Commercial Bank weighted average lending rates (short-term credits, 1999–2004 monthly series): examples - 1999 Dec.: 3.27; 2000 Dec.: 5.04; 2001 Dec.: 2.55; 2002 Dec.: 1.54; 2003 Dec.: 1.23; 2004 Dec.: 1.23 (Table 24). - Deposit interest rates (weighted averages, selected): Short-term deposits of households (percent): 1999 Dec.: 1.38; 2000 Dec.: 3.10; 2001 Dec.: 3.76; 2002 Dec.: 2.82; 2003 Dec.: 1.25; 2004 Dec.: 1.03 (Table 25). - National Bank of Serbia discount rate (Table 26): 1999 Dec.: 2.00; 2000 Dec.: 2.00; 2001 Dec.: 1.30; 2002 Dec.: 0.77; 2003 Dec.: 0.73; 2004 Dec.: 0.69. ### Prices and inflation (selected) - Retail price index (Index 1995=100): 1999: 434; 2000: 737; 2001: 1408; 2002: 1706; 2003: 1898; 2004: 2078. - Consumer price index (Index 2000=100): 2000: 100; 2001: 189; 2002: 225; 2003: 246; 2004: 271. - Producer price index (Index 1997=100): 1999: 69; 2000: 69; 2001: 69; 2002: 76; 2003: 81; 2004: 88. ### Labor market, pensions, and social funds (selected) - Total employment (annual average, in thousands): 1999: 2,298; 2000: 2,238; 2001: 2,243; 2002: 2,207; 2003: 2,178. - Employment in manufacturing (thousands): 1999: 707; 2000: 668; 2001: 647; 2002: 594; 2003: 553. - Pension fund (Serbia) number of beneficiaries (in thousands): 1999: 1,254; 2000: 1,259; 2001: 1,276; 2002: 1,252; 2003: 1,252; 2004 Est.: 1,243. - Average pensions (in dinars): 1999: 1,132; 2000: 2,154; 2001: 4,665; 2002: 6,722; 2003: 8,102; 2004: 9,578. - Insurees (in thousands): 1999: 1,671; 2000: 1,597; 2001: 1,555; 2002: 1,459; 2003: 1,384; 2004: 1,580. - Dependency ratio (employee/pensioner): 1999: 1.32/1; 2000: 1.26/1; 2001: 1.22/1; 2002: 1.17/1; 2003: 1.11/1; 2004: 1.27/1. - Labor Market Fund (Unemployment Insurance of Serbia) average monthly number receiving unemployment benefits (in thousands): 1999: 45.1; 2000: 47.4; 2001: 51.2; 2002: 69.6; 2003: 90.9; 2004 Est.: 76.6. *Source: _cr05232 - References and Statistical Appendix (tables and figures as provided).* ### 10. Foreign assets.backed export credit (not repaid within 3 mos.) ### 10. Foreign assets.backed export credit (not repaid within 3 mos.) ### Interest rates, reserve requirements, and NBS asset classifications - Foreign assets-backed export credit (not repaid within 3 mos.): 4.00; 4.00; 2.60; 1.55 - 11. Mandatory subscription of NBY bills: 2.00; 2.00; 1.30 - a) paid within the same working day: 0.16 - b) paid during the following working day: 1.21; 1.15; 1.09; 1.02; 1.09; 1.05; 1.34; 1.30; 1.34; 1.65; 1.59; 1.65; 1.59; 1.65 - 13. Allocated required reserves: 1.05; 1.05; 0.50 - 14. Interest rate on the daily average dinar RR up to the level of calculated dinar RR: 0.28; 0.26; 0.25; 0.23; 0.25; 0.24; 0.25; 0.24; 0.25; 0.25; 0.24; 0.25; 0.24; 0.25 - 15. Interest rate on the daily average forex RR up to the level of calculated forex RR - a) on an allocated RR in USD: 0.02; 0.02; 0.02; 0.02; 0.02; 0.02; 0.02; 0.02; 0.02; 0.03; 0.03; 0.03; 0.03; 0.04 - b) on a allocated RR in EUR: 0.05; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03; 0.03 - c) on a allocated RR in CHS: 0.00; 0.00; 0.00; 0.00; 0.00; 0.00; 0.00 - 16. Interest rate on the banks' excess liquidity deposited with the NBS: 0.32; 0.26; 0.25; 0.23; 0.25; 0.24; 0.25; 0.24; 0.25; 0.25; 0.24; 0.25; 0.24; 0.25 - 17. NBS assets determined by separate law on transforming short-term loans to long-term loans: 0.97; 0.96; 0.97; 0.97; 0.97; 0.96; 0.90; 0.96; 0.93; 0.96; 0.93; 0.96; 0.96; 0.93; 0.96; 0.93; 0.96 - 18. NBS assets determined by separate decision of the governor: 0.66; 0.65; 0.66; 0.66; 0.66; 0.65; 0.61; 0.65; 0.63; 0.65; 0.63; 0.65; 0.65; 0.63; 0.65; 0.63; 0.65 *Source: National Bank of Serbia. Interest rate on the credit amount for daily liquidity maintenance granted to the banks on the basis of the pledged securities.* ### Balance of Payments highlights (Serbia and Montenegro, 1999–2004) - Trade balance: -1,619; -1,788; -2,834; -3,908; -4,886; -7,434 - Exports f.o.b.: 1,676; 1,923; 2,003; 2,412; 3,054; 4,219 - Imports c.i.f.: -3,295; -3,711; -4,837; -6,320; -7,941; -11,653 - Services (non-factor services, net): 228; 331; 417; 292; 336; 486 - Receipts: 471; 624; 740; 829; 1,130; 1,678 - Expenditure: -243; -293; -323; -537; -795; -1,192 - Net factor income: -411; 1; -26; -111; -222; -274 - Net interest: -411; 1; -26; -111; -222; -274 - Earnings: 435; 348; 627; 708; 1,081 - Payments 1/: -84; -42; -74; -173; -291; -354 - Unrequited private and official transfers, net: 668; 1,119; 1,915; 2,343; 2,777; 4,092 - Private remittances, net: 668; 848; 1,324; 1,719; 2,239; 3,510 - Inflows: 948; 1,132; 1,698; 2,089; 2,661; 4,129 - Outflows: -280; -284; -374; -370; -422; -620 - Current account balance, before grants: -764; -598; -1,119; -2,007; -2,533; -3,712 - (In percent of GDP): -7.5; -7.1; -9.7; -12.9; -12.3; -15.5 - Official grants: 0; 2; 7; 15; 9; 162; 45; 38; 58; 3 (note: values shown in sequence in source) - Foreign direct investment, net: 112; 251; 655; 621; 1,405; 1,028 - Foreign loans, net: -251; 803; 745; 378; 222; 2,011 - Medium and long term, net: 122; 132; 99; 379; 756; 1,588 - Disbursements: 292; 273; 324; 219; 742; 2,119 - Of which: Official creditors: 0; 22; 72; 53; 43; 46; 05; 37 (sequence as in source) - Amortization: -17; -14; -33; -43; -218; -530 - Short term, net: -37; -33; 75; 15; 86; 642; 3 (sequence as in source) - Other capital inflows: 304; 962; 989; 228; 129; 6 - Commercial banks, net: 0; 0; -274; -144; 312; 6 - Capital account balance: 117; 255; 894; 1,846; 2,539; 3,362 - Errors and omissions: 410; 267; 239; 320; 409; 432 - Overall balance: -237; 195; 605; 784; 952; 665 - Financing: 237; -183; -5,981; -855; -1,001; -3,905 - Net foreign assets (increase, -): 111; -246; -395; -816; -1,001; -711 - Central Bank, net: 111; -246; -395; -816; -1,001; -711 - Gross foreign reserves (increase, -): 111; -227; -523; -1,111; -1,277; -719 - Of which: IMF purchases: 0; 15; 21; 28; 29; 52; 76; 24; 3 (sequence as in source) - Gross foreign liabilities (increase +): 0; -191; 282; 952; 768 (sequence as in source) - Of which: IMF repayment: 0; 20; 0; 0; 0; -235 - Arrears (reduction, -): 126; 63; -5,587; -390; -3,194 - Residual gap: 0; ...; 5,377; 715; 93; 3,259 - Arrears settlement with creditors: 0; ...; 5,377; 390; 3,194 - Debt relief from creditors: 0; ...; 0; 32; 596; 5 - Memorandum items: - Current account balance, after grants: -764; -339; -528; -1,383; -1,513; -3,148 - (In percent of GDP): -7.5; -3.9; -4.6; -8.9; -7.3; -13.1 - Gross international reserve (in millions of U.S. dollars, end period): 289; 516; 1,169; 2,280; 3,557; 4,302 - (In months of prospective imports of goods and services): 0.9; 1.2; 2.4; 3.1; 3.3; 3.7 - Debt service, cash: 101; 56; 107; 183; 451; 972 - (In percent of GDP): 1.0; 0.6; 0.9; 1.2; 2.2; 4.1 - Principal: 171; 433; 432; 186; 75 - Interest: 84; 42; 74; 141; 233; 297 - External Debt: ...; 11,403; 11,948; 11,839; 14,303; 14,876 - (In percent of GDP): ...; 132.5; 103.2; 76.2; 69.2; 62.0 *Sources: Serbia and Montenegro authorities; and IMF staff estimates.* ### Exports composition (Serbia, 1999–2004) - Exports by broad category (in millions of U.S. dollars): Total: 1,369; 1,558; 1,721; 2,075; 2,755; 3,701 - Intermediate goods: 748; 879; 913; 1,130; 1,505; 2,421 - Capital goods: 102; 121; 157; 165; 427; 241 - Consumer goods: 518; 558; 651; 780; 824; 1,040 - Exports by detailed category (in millions of U.S. dollars): - Food and live animals: 288; 253; 271; 477; 499; 656 - Beverages and tobacco: 18; 13; 13; 17; 32; 56 - Raw materials, except fuel: 72; 111; 89; 107; 138; 208 - Mineral fuel and lubricants: 36; 45; 70; 76; 119; 93 - Animal and plant oils and fats: 8; 16; 18; 19; 17; 65 - Chemical products: 145; 145; 132; 168; 249; 394 - Products classified by materials: 373; 499; 505; 549; 691; 1,227 - Machines and transport devices: 180; 208; 241; 251; 570; 411 - Miscellaneous manufactured products: 213; 265; 358; 363; 457; 563 - Others: 36; 43; 45; 48; 43; 329 - Exports by broad category (in percent of total): Intermediate goods: 55; 58; 53; 54; 56; 65; Capital goods: 7; 8; 9; 8; 15; 7; Consumer goods: 38; 36; 38; 37; 30; 28 - Exports by detailed category (in percent of total): Food and live animals: 21; 11; 6; 12; 3; 18; Beverages and tobacco: 1; 1; 1; 1; 1; 2; Raw materials, except fuel: 57; 5; 5; 6; ...; etc. (percent breakdowns as listed in source) *Source: Serbia and Montenegro Statistical Office.* ### Imports composition (Serbia, 1999–2004) - Imports by broad category (in millions of U.S. dollars): Total: 2,881; 3,330; 4,261; 5,614; 7,473; 11,133 - Intermediate goods: 1,971; 2,330; 2,853; 3,404; 4,241; 6,245 - Capital goods: 475; 532; 671; 1,132; 1,452; 2,380 - Consumer goods: 435; 467; 737; 1,078; 1,780; 2,508 - Imports by detailed category (in millions of U.S. dollars): - Food and live animals: 225; 216; 331; 404; 473; 634 - Beverages and tobacco: 32; 50; 95; 104; 132; 162 - Raw materials, except fuel: 212; 208; 180; 190; 222; 349 - Mineral fuel and lubricants: 457; 662; 830; 902; 1,080; 1,637 - Animal and plant oils and fats: 7; 79; 16; 16; 18 - Chemical products: 477; 523; 659; 791; 1,012; 1,436 - Products classified by materials: 625; 715; 878; 1,176; 1,558; 2,181 - Machines and transport devices: 633; 720; 893; 1,454; 2,236; 3,543 - Miscellaneous manufactured products: 190; 212; 321; 507; 714; 1,049 - Others: 22; 176; 57; 130; 125 - Imports by broad category (in percent of total): Intermediate goods: 68.4; 70.0; 66.9; 60.6; 56.7; 56.1; Capital goods: 16.5; 16.0; 15.8; 20.2; 19.4; 21.4; Consumer goods: 15.1; 14.0; 17.3; 19.2; 23.8; 22.5 - Imports by detailed category (in percent of total): Food and live animals: 7.8; 6.5; 7.8; 7.2; 6.3; 5.7; Mineral fuel and lubricants: 15.9; 19.9; 19.5; 16.1; 14.5; 14.7; Machines and transport devices: 22.0; 21.6; 21.0; 25.9; 29.9; 31.8; (other percent breakdowns as listed in source) *Source: Serbia and Montenegro Statistical Office.* ### Destination of exports and origin of imports (Serbia, 1999–2004) - Destination of exports (Total, in millions of U.S. dollars): 1,368; 1,558; 1,721; 2,075; 2,755; 3,701 - Developed western countries: 602; 715; 851; 962; 1,397; 2,009 - Developing countries: 766; 843; 869; 1,114; 1,358; 1,692 - Italy: 148; 208; 293; 312; 349; 486 - Bosnia and Herzegovina: 287; 236; 240; 323; 416; 651 - Germany: 165; 175; 230; 242; 293; 373 - Republic of Macedonia: 170; 206; 174; 206; 331; 267 - Switzerland: 34; 27; 25; 19; 19; 23 - Russia: 73; 86; 80; 91; 132; 158 - Greece: 74; 75; 63; 82; 61; 120 - Hungary: 28; 53; 59; 74; 77; 123 - Great Britain: 21; 34; 38; 42; 54; 83 - France: 40; 42; 43; 60; 115; 157 - Other countries: 328; 416; 476; 626; 909; 1,260 - Destination shares (percent of total): Developed western countries: 44.0; 45.9; 49.5; 46.3; 50.7; 54.3; Developing countries: 56.0; 54.1; 50.5; 53.7; 49.3; 45.7 - Origin of imports (Total, in millions of U.S. dollars): 2,881; 3,331; 4,261; 5,614; 7,473; 11,134 - Developed western countries: 1,466; 1,596; 2,015; 2,880; 4,627; 6,871 - Developing countries: 1,415; 1,734; 2,246; 2,733; 2,846; 4,263 - Russia: 211; 305; 665; 777; 1,024; 1,401 - Germany: 385; 452; 547; 762; 1,057; 1,501 - Italy: 289; 345; 412; 566; 747; 1,065 - Greece: 8; 77; 61; 04; 130; 155; 209 (values as listed in source) - Hungary: 94; 112; 176; 248; 264; 321 - Romania: 105; 143; 168; 136; 135; 210 - Bulgaria: 147; 322; 152; 105; 134; 236 - Austria: 101; 101; 126; 160; 232; 293 - Slovenia: 15; 15; 10; 9; 19; 42; 45; 33; 7 (sequence as presented) - Bosnia and Herzegovina: 158; 148; 118; 129; 169; 243 - Other countries: 1,290; 1,312; 1,684; 2,406; 3,312; 5,319 - Origin shares (percent of total): Developed western countries: 50.9; 47.9; 47.3; 51.3; 61.9; 61.7; Developing countries: 49.1; 52.1; 52.7; 48.7; 38.1; 38.3 *Source: Serbia and Montenegro Statistical Office.* ### Stock of external debt (1999–2004) - Total external debt (in millions of U.S. dollars): 10,744; 11,419; 11,740; 11,839; 14,303; 14,876 - By creditor type (in millions of U.S. dollars): - Multilateral creditors: 2,422; 2,488; 2,706; 3,702; 4,968; 5,554 - IMF: 133; 152; 272; 565; 913; 962 - IBRD 1/: 1,718; 1,781; 1,840; 2,175; 2,607; 2,839 - IDA: 0; 0; 0; 168; 282; 462 - Eurofima: 135; 126; 129; 158; 167; 170 - IFC 2/: 130; 132; 175; 213; 288; 134 - EIB: 266; 257; 49; 114; 212; 313 - Eurofund: 31; 30; 28; 31; 35; 33 - MIB: 9; 10; 11; 0; 0 - EBRD: 0; 0; 2; 4; 4; 13; 128; 0 - European Community: 0; 0; 19; 92; 353; 313; 361 (sequence as in source) - Official bilateral creditors: 4,213; 4,619; 4,610; 3,295; 3,650; 3,923 - Paris Club 3/: 4,180; 4,129; 4,037; 2,690; 3,007; 3,227 - Other bilateral creditors: 334; 90; 573; 606; 436; 697 - Commercial creditors: 3,122; 3,158; 3,399; 3,822; 4,630; 4,400 - London Club 4/: 2,228; 2,267; 2,300; 2,442; 2,738; 1,164 - Other commercial creditors 5/: 685; 682; 914; 1,196; 1,708; 3,052 - Debt incurred in non-convertible currencies: 209; 209; 185; 184; 184; 184 - Short-term debt: 987; 1,153; 1,026; 1,020; 1,056; 999 - Debt related to unpaid imports of oil and gas 6/: 431; 490; 502; 513; 520; 240 - Other short-term debt: 556; 663; 524; 508; 536; 759 *Source: National Bank of Serbia.* *Notes from source (selected):* - 1/ Including exchange rate adjustments. - 2/ In July 2004, debt relief of about 75% on the debt stock of EUR 172.76 mln was granted to banks from the system of former Vojvodjanska banka Udruzena banka a.d., Novi Sad. Regarding to Montenegrobanka a.d., Podgorica and other obligors from Montenegro, a debt relief of due regular and late interest was granted. - 3/ For 2001, debt figure reflects Paris Club estimate determined during the November 2001 Paris Club session. - 4/ Excludes US$500 million in debt held by Yugoslav "connected parties". In negotiations during June/July 2004, debt relief on NFA-TDFA of about 62% was granted. - 5/ In 2004, former trade debt for oil import from China has been transferred from the category "Short-term debt" into "other commercial creditors". - 6/ Non-guaranteed overdue obligations (trade credits) owed to oil and gas enterprises in Russia and China. In November 2003, there was concluded Protokol on the debt relief between NIS - Novi Sad and SINOCHEM, China for oil import. The final agreement on the settlement of the debt was signed in October 2004. Negotiations are underway with Russia for gas import. *Source: National Bank of Serbia; Serbia and Montenegro Statistical Office; IMF staff estimates.* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2005/_cr05232.pdf_