Report No. 29123-RO Romania Restructuring for EU Integration--The Policy Agenda Country Economic Memorandum (In Two Volumes) Volume II: Main Report and Annexes June 2004 Poverty Reduction and Economic Management Unit Europe and Central Asia Region Document of the World Bank CONTENTS ACRONYMS AND ABBREVIATIONS........................................................................................ 2 1. STABILITY, GROWTH, AND INTEGRATIONWITH THE EU .................................. 1 A.MACROECONOMIC STABILITY ...................................................................................................... 1 A.1. Ovewiew.............................................................................................................................. 1 A.2. Fiscal Policies ..................................................................................................................... 3 A.3. Incomes Policy, Exchange Rate. and Prices ....................................................................... 4 A.4. Implicationsfor External Sustainability.............................................................................. 6 B.ROMANIA'S C. GROWTH PERFORMANCEREGIONALPERSPECTIVE.............................................................. IN 9 AND INTER-SECTORAL RESTRUCTURING ..................................................................... 14 D.RISKSAND VULNERABILITIES .................................................................................................... 18 D.1.Budgetary Stability............................................................................................................ 18 0.2. Fiscal Sustainability.......................................................................................................... 20 D.3.Nonpayment. Arrears. and Barter Settlement................................................................... 23 2. BUILDINGONTRADE INTEGRATION ....................................................................... 25 A.INTRODUCTION .......................................................................................................................... 25 B.DYNAMICS FOREIGN c.FEATURESTHE OF TRADE:CRITICAL ROLEOFEUMARKETS ........................................... 26 OF CURRENT WAVE OF STRONG EXPORT PERF~RMANCE .................................. 28 C.I. Outstanding Dynamics of EU-Oriented Exports: TheEarlier Phase............................... 28 C.2. Levers ofExport Growth................................................................................................... 29 C.3. Acceleration in Imports Have Kept up with Export Growth............................................. 31 C.4. Conclusion......................................................................................................................... 32 D.PATTERNSOFINTEGRATIONINTOTHEEU:IMPLICATIONSFORTHE SUSTAINABILJTYOFEXPORT PERFORMANCE 33 D.1. FDIInflows and Trade...................................................................................................... ................................................................................................................................ 33 D.2. Factor Intensities of EU-oriented Exports: The Gap betweenEndowments and Factor Intensities .................................................................................................................................. 36 0.3.Exports to E U Sunrise Markets: Is the Change in Factor Intensities Under Way?..........38 D.4Participation in InternationalNetworks of Production andDistribution ......................... 39 D.5. The Challenge of the Impeding Change in the TradeRegimefor Textiles and Clothing44 E.NETWORK OF PREFERENTIAL TRADEAGREEMENTS: POTENTIAL FORDISTORTIONS ..................46 E.1.Applied MFN Tariff Rates in ComparativePerspective.................................................... 48 E.2.Preferential Partners: Presenceand Conditions in Access to Romanian Markets...........49 E.3. Tariff Barriers in 2002: TheExtent of ReverseDiscrimination........................................ 49 F.IMPLICATIONSTARIFFRATES ONINDUSTRIALPRODUCTS OF THE ALIGNMENTOFMFNAPPLIED WITH THE EUCET .......................................................................................................................... 52 F.1.A "Shadow Customs Union" with the EURestricted toIndustrial Products ...................53 G.F.2. Summary: Costs and Benefits of Industrial TariffHarmonization with E U CET.............57 CONCLUSIONS AND POLICY RECOMMENDATIONS ...................................................................... 58 3. RESTRUCTURINGTHE ENTERPRISESECTOR ....................................................... 61 A.ECONOMIC RESTRUCTURING REFORMS: ROMANIAINA REGIONALPERSPECTIVE ..................... 62 A.1. Privatization Unfolds Slowly. Farfrom Finished.............................................................. 62 A.2. Privatization Modes Affect Outcomes ............................................................................... 66 A.3. Hard Budget Constraints. Yet to Come............................................................................. 67 A.4. Domestic Competition Slow to Come................................................................................ 72 B.GROWTH THE ENTERPRISE IN SECTOR ........................................................................................ 74 B.1. Sales Growth of Enterprises .............................................................................................. 74 B.2. TotalFactor Productivity .................................................................................................. 75 C.THEREFORMAGENDA ............................................................................................................... 76 C.1.Privatizing to OutsideStrategicInvestors......................................................................... 76 C.2.Licensing Regimes............................................................................................................. 76 C.3.BusinessEntry ................................................................................................................... 77 C.4.Labor Regulations ............................................................................................................. 78 C.5.Bankruptcy ......................................................................................................................... 79 C.6.EfJiciency and Predictability of Commercial Contracts ................................................... 79 4. IMPLEMENTINGAGRICULTURAL TRANSFORMATION ..................................... 82 A.AGRICULTURE AND FOOD THE ECONOMY IN AND SECTORAL PERFORMANCE ............................ 82 B.STATUSOFSECTORAL REFORMS................................................................................................ 84 B.1.DistortiveAgricultural Support Policies with High Levels of Subsidization .................... 85 B.2. Light and Inconsistent Taxation in theAgricultural Sector .............................................. 87 B.3.Fluctuating TradePolicies ................................................................................................ 88 B.4.Incomplete Transition in theFarming Sector.................................................................... 89 B.5.Agroprocessing Industry Unpreparedfor E UEntry......................................................... 92 B.6. Commodity and Factor Markets Not Adapted to the Needs of Privatized Agriculture.....92 C.B .7. Urban-Rural Gap: SigniJicant Rural SocialProblems..................................................... 94 SELECTING AN APPROPRIATE COURSEFORAGRICULTURAL POLICY.......................................... 96 C.1.Shqt in Policiesfor Increased Competitivenessandfor the Introduction of CAP............96 C.2. SimulatingFuturePolicy Optionsfor Romania's Agricultural Sector ............................. 98 C.3. The Welfare Impact of Introducing theEU CommonAgricultural Policy........................ 99 C.4. TheImpact of Romania's Current Agricultural Policy................................................... 101 D.DCONSOLIDATION INTHE FARMINGSECTOR AND REDUCING EXCESS AGRICULTURAL LABOR .102 0.2. Consolidation of Small-Scale Farming Sector and Reducing Labor in Agriculture ......102 .1. Completion of Farm Restructuring and Farm Privatization.......................................... 103 E.ACCELERATING NECESSARY INAGROPROCESSING, SUPPLY, AND THE REFORMS INPUT MARKETING .................................................................................................................................. 105 F. CREATING EUCONFORMING INSTITUTIONS AND REGULATORY SYSTEMS............................... 106 G. OVERCOMINGOBSTACLES TORURALDEVELOPMENT-REDUCING POVERTY ............108 RURAL 5. ENHANCINGLABORMARKETADJUSTMENT ...................................................... 110 A.PARTICIPATION, EMPLOYMENT, NON-EMPLOYMENT AND ....................................................... 111 ROMANIA...................................................................................................................................... B.JOBDESTRUCTION, CREATION, ANDJOBLESSNESS........................................................... 113 JOB 118 C EMPLOYMENT . DISTRIBUTION CHARACTERISTICS ............................................................. AND 120 D. WAGES AND PRODUCTIVITY ............................................................................................. 123 128 F.CONCLUSIONS AND RECOMMENDATIONS ................................................................................. E.LABOR MARKET POLICIES........................................................................................................ 130 6. FROMQUASI-FISCAL FINANCING TO EFFICIENT FINANCIAL INTERMEDIATION ................................................................................................................. 135 A.INTRODUCTION ........................................................................................................................ 135 B.THESTRUCTUREOFTHEFINANCIALSYSTEM........................................................................... c.THEROLEOF 135 ....................................................................................... 138 D.THESOUNDNESSOFTHEBANKINGSECTOR............................................................................. THE BANKINGSECTOR 142 E.REGULATORY AND SUPERVISORYISSUES ................................................................................. 145 G. CONCLUSIONS.......................................................................................................................... 147 7 . THE QUASI-FISCAL CHALLENGE: ENERGY SECTOR REFORM ..................... 149 A.THEENERGY SECTOR AND STATE SUPPORT............................................................................ 149 B.POWERPRICINGAND PAYMENTSDISCIPLINEINROMANIA...................................................... 152 B.1. Pricing and Payments Discipline .................................................................................... 152 B.2.Rulesfor allocating sector revenues between generation. transmission. and distribution.154 B.3. ThePower Sector Social Safety Net ................................................................................ 156 C. GASSECTORFINANCIALVIABILITY AND AFFORDABILITY ....................................................... 157 C.1. Gas Pricing and Affordability ......................................................................................... 157 C.2. TheDirect Affordability Impact of Gas Price Increases................................................. 158 C.3. The Gas Price Impact on District Heating...................................................................... 158 C.4. Payments in the Gas Sector............................................................................................. 160 D.REGIONALENERGY MARKET: CHALLENGES AND OPPORTUNITIES FORROMANIA ..................160 E.ENERGY INDUSTRY INSTITUTIONAL REFORM ........................................................................... 161 E.1Industry Restructuring...................................................................................................... 161 E.2 Private Sector Participation ............................................................................................. 161 F.INDEPENDENT REGULATION ..................................................................................................... 162 H.INVESTMENTINTHEPOWER SECTOR....................................................................................... 165 G.MARKET RULES........................................................................................................................ 163 H.l.TheFrameworkfor Investment ....................................................................................... 166 H.2. Contracting Parties......................................................................................................... 168 I.CONCLUSION............................................................................................................................. 168 8. ENVIRONMENTAL STANDARDS:UPGRADING AND CONTAINING THE COSTS ........................................................................................................................................ 170 A.INTRODUCTION B.BENEFITS MEETNG ENVIRONMENTAL OFTHEACQUIS INROMANIA..................170 ........................................................................................................................ c.COSTS OF THE PART 172 OFUPGRADING ENVIRONMENTAL STANDARDS ............................................................. 174 C.1. Estimates of Total Costs and Costs by Year.................................................................... 174 C.2. The Financing Gap after EU Grants Have Been Taken into Account ............................ 176 D.KEYISSUES SECTOR......................................................................... C.3. TheEnvironment Fund.................................................................................................... 179 INTHEENVIRONMENTAL 180 D.1. Ways in Which the Budgetary Burden Can Be Reduced and More Internal Resources Mobilized................................................................................................................................. 180 0.2. Institutional and Administrative Reforms ....................................................................... 184 E.CONCLUSIONS.......................................................................................................................... 184 ANNEX 1: FISCAL SUSTAINABILITY ANALYSIS ........................................................... 187 ANNEX 2: MODELING OF AGRICULTURAL POLICY .................................................. 189 ANNEX 3: IMPACT OF CHANGES INAGRICULTURAL POLICY .............................. 191 ANNEX 4: EARNINGS FUNCTION FOR MALES AND FEMALES IN2000 ................192 ANNEX 5: NON-BANKINGFINANCIAL SERVICES ........................................................ 194 ANNEX 6: THE ATHENS MEMORANDUM ....................................................................... 199 ANNEX 7: EUDIRECTIVES: IMPLEMENTATION COSTSAND ASSUMPTIONS 201 .... STATISTICAL ANNEX ........................................................................................................... 205 ANNEX 1:FISCAL SUSTAINABILITY ANALYSIS.. ......................................................... 187 ANNEX 2: MODELING OFAGRICULTURAL POLICY ................................................. 189 ANNEX 3: IMPACT OF CHANGES INAGRICULTURAL POLICY ............................. ,191 ANNEX 4: EARNINGS FUNCTION FOR MALESAND FEMALES IN2000 ................192 ANNEX 5: NON-BANKINGFINANCIAL SERVICES ........................................................ 194 ANNEX 6: THE ATHENS MEMORANDUM ....................................................................... 199 ANNEX 7: EUDIRECTIVES: IMPLEMENTATION COSTSAND ASSUMPTIONS ....201 STATISTICAL ANNEX ........................................................................................................... 205 Table 1.1: Populationby Area Table 1.2: Population by Gender Table 1.3: Populationby Age Group and Gender Table 1.5: Labor Market (ILO methodology) Table 2.1: Gross Domestic Product by Expenditure Table 2.2: Gross Domestic Product by Expenditure, as percent of GDP Table 2.3: Gross Domestic Product by Expenditure(constant 1998 prices) Table 2.4 Gross Domestic Product by Sector Table 2.5: Gross Domestic Product by Sector, as percent of GDP Table 2.6: Gross Domestic Product by Sector (constant 1998 prices) Table 2.7: Financingof Investment Table 2.8: Industrial Output Indices, by Activities Table 2.9: Structure of Industrial Output by Activities Table 2.10: Evolutionof the Privatization Process Table 2.11:Private Sector Share inTotal GDP and inValue Added by Branches Table 3.1: Balance o f Payments Table 3.2: Exports byMajor Commodities Table 3.3: Imports by Major Commodities Table 3.4: Direction o f Trade inExports Table 3.5: Direction o f Trade inImports Table 3.6: ForeignDirect Investment (Subscribed Capital) Table 3.7: ForeignDirect Investmentby Country of Origin Table 4.1: Consolidated General Government Budget Table 4.2: Consolidated General Government Budget (percent of GDP) Table 4.3: SummaryTable: GFS Consolidated General Budget Table 4.4: Summary table: GFS Local Budgets Table 4.5: ConsolidatedMonetary Survey Table 4.5: ConsolidatedMonetary Survey Table 5.1: Prices Table 5.2: Administered Prices Table 5.3: Average Net Nominal Monthly Salary Earnings, by Activities Table 5.3: Real Wage Indices Table 5.4: Interest Rate Table 5.5: Exchange Rates Table 6.1: External andDomestic Debt Table 6.2: Selected External Debt Indicators Map IBRD28710R1 1. STABILITY, GROWTH,AND INTEGRATIONWITHTHE EU 1.1 Until' recently, Romania was one of the poorest performing economies among Central Eastern Europe countries (CEEC).2 Today, Romania is at a crossroads. A decade o f transition was marked by limited structural reforms, lack o f financial discipline resulted in unsustainable macroeconomic policies which ended in two major economic crises, hyperinflation, and substantial decline in standards o f living. In the last three years, however, Romania has made remarkable progress by restoring growth, macroeconomic stability, and poverty has started to decline. Romania would need to build up on these recent improvements in economic performance, strengthen its macroeconomic framework, and accelerate implementation o f structural reforms. This chapter examines macroeconomic stability, risks and vulnerabilities, growth, and progress inRomania's integrationwith the EuropeanUnion(EU). 1.2 Romania is pursuing a broad reform program, including institutional, governance, and economic restructuring reforms, which are anchored on its process for EU accession. The institutional and governance reforms are being made in the context o f negotiations and implementation o f the various chapters o f the acquis communautaire and are being supported by the World Bank's Programmatic Adjustment Loan, currently being developed, and by programs o f other development partners (see the Section 1 o f the Summary Report o f this study). To complement these efforts, this study focuses on selected policies o f the real sector as Romania integrates with the EU and the world markets. Indeed, recent progress indicates that reforms are tuminga comer-the challenge is to make consistent and sustained progress in the institutional, govemance, and economic restructuring reform agendas. A. MACROECONOMIC STABILITY A.l. Overview 1.3 The legacy o f the socialist years has weighted heavily over the transition years in Romania. Policies in Romania during the 1990s were dominated by entrenched interest groups embedded in state-owned-enterprises (SOE), a pervasive lack o f financial discipline in their relations with the financial sector and with public administration which resulted inunsustainable macroeconomic policies. As a result, structural reforms and institutional change were also delayed. Romania experienced two major economic crises one in 1991/1993, and another one in 1997/1998 (see Table 1.1). In the 1991/1993 crisis, annual inflation rate reached 256 percent, debt to Gross Domestic Product (GDP) ratio increased from 3 percent in 1990 to 16 percent in 1993, and accumulated GDP contraction was about 30 percent. Inthe 1997/1999 crisis, inflation reached above 150percent per year from a level o f about 35 percent per year in 1995/96, debt to 'Thischapter 1 was preparedby Juan Carlos Ginarte, Stella Ilieva, and Rosalinda Quintanilla (ECSPE). Central Eastern European Countries (CEEC) is defined here to include Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Lithuania, Latvia, Poland, Romania, Slovenia, Slovak Republic. 2 GDP ratio reached 40 percent, and accumulated GDP contraction was about 15 percent. These events exacerbated the effects o f the collapse o f output on broad segments o fthe population. Table 1.1 Selected Economic Indicators 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 20022003p GNI per capita (Atlas method, US$) 1,240 1,190 1,270 1,470 1,590 1,520 1,520 1,580 1,680 1,710 1,850 (Percent) Total Povertyrate(headcount, %) 25.4 20.1 30.3 30.8 33.2 35.9 30.6 28.9 ExtremePovertyrate (headcount, %) .. ..9.4 6.3 11.2 11.3 12.5 13.8 11.4 10.9 Unemploymentrate 8.2 10.4 10.9 9.5 6.6 8.9 10.4 11.8 10.5 8.8 8.4 7.3 RealGDP growth -8.8 1.5 3.9 7.1 3.9 -6.1 -4.8 -1.2 2.1 5.7 5.0 4.9 CPI inflation (p.a.) 210.4 256.1 136.7 32.3 38.8 154.8 59.1 45.8 45.7 34.5 22.5 15.1 (Percent of GDP) Investment 31.4 28.9 24.8 24.3 25.9 20.6 17.7 16.1 19.5 22.6 23.1 ForeignDirect Investment 0.3 0.3 1.1 1.2 0.7 3.5 4.8 2.9 2.8 3.0 2.5 2.7 Gross DomesticSavings 20.8 24.0 22.5 18.6 17.6 13.5 9.7 11.4 14.5 13.9 15.0 Total revenuesand grants 38.7 34.0 32.0 31.4 29.6 28.9 29.9 31.9 31.4 30.5 29.7 30.9 Total expenditure andnet lending 43.7 34.4 34.7 34.8 33.9 34.2 35.3 35.5 35.4 33.7 32.4 33.6 Overall fiscal balance -4.6 -0.4 -2.2 -3.4 -4.8 -5.3 -5.4 -3.6 -4.0 -3.3 -2.6 -2.3 Current accountbalance -8.0 -4.5 -1.4 -5.0 -7.3 -6.0 -6.9 -4.0 -3.7 -5.5 -3.4 -5.8 Extemal debt 18.5 18.2 23.5 27.2 23.8 25.9 27.8 29.6 30.7 30.2 Public andpublicly-gtd.Debt 1/ .. 11.8 12.9 19.4 22.9 21.1 24.2 23.6 24.6 25.5 Interest payment 3.8 4.7 5.3 4.8 3.8 3.0 2.9 Official reserves (mo. of imports) 0.2 0.1 1.1 0.4 0.5 2.1 1.3 1.6 2.1 2.9 3.9 3.5 Broadmoney/GDP (percent) 30.8 22.3 21.4 25.3 27.9 24.6 24.8 24.6 23.0 23.2 24.7 24.4 Source: World Bankand IMF databases; RomaniaPublic ExpenditurCsReview (2002) andPovertyAssessment (2003) 1.4 Inthe last three years, however, there hasbeen aremarkable turnaroundbothinterms of the magnitude and quality o f macroeconomic adjustment and growth (see Table 1.1). GDP growth has been reestablished to around 5 percent per year, led primarily by investment and exports rather thanconsumption. Occasionally, thought, aggregate consumption surfaces as a key driver o f growth, reflecting sporadic reflationary pressures related to high wage increases in SOEs. The inflation rate has declined faster than expected by the macroeconomic stabilization program to about 14percent per year in2003, the lowest level since the start o f transition. Fiscal deficits have been cut to about half compared to 1997 levels but the current account gap, estimated at 5.8 percent of GDP in2003 i s high.Budgetaryconsolidation in2003 was largerthan expected by the program. A stronger than expected external performance contributed to the level o f official reserves reaching a record level o f over US$7 billion in2003-an impressive increase o f US$3.5 billion since 2000. The banking sector is on firmer footing and direct lending by the central bank has been eliminated. International market sentiment towards Romania has continued to improve- Standard&Poors upgraded Romania's sovereign currency risk from B+ to BB- in February 2003. Improvements in macroeconomic performance and growth have contributed to reductions inpoverty levels (Box 1.1). 3 A.2. Fiscal Policies 1.5 In the last few years, Romania has made important progress in fiscal adjustment and consolidation (see Table 1.2). Monetary financing o f quasi-fiscal expenditures via directed credits to agriculture and other loss-making sectors-and main contributors to Romania's two major crises in the 1990s-were halted in 2000. The National Bank o f Romania (NBR)direct financing o f fiscal deficits ended in 2000. Between 1998 and 2003, Romania implemented a fiscal adjustment o f about 3.0 percentage points of GDP-the overall fiscal deficit hadreached a peak o f 5.4 percent o f GDP in 1998 and by 2003, preliminary estimates indicate that the overall fiscal deficit was 2.4 percent o f GDP. On the whole there have been important qualitative dimensions in Romania's fiscal consolidation of the last few years. The fiscal adjustment of 1998-2003 resultedmostly by lowering public expenditures to GDP from 35.3 to 32 percent with the strongest expenditure adjustment taking place in 2000-2003. At the same time, public revenues first increased between 1998 and 1999 to address unsustainable macroeconomic imbalances and later declined from 31.8 to 29.6 percent o f GDP in 2000-2002. Declines in tax revenues to GDP were supported by important adjustments on the public expenditure side. In addition most o f the adjustment on the expenditure was driven by reductions o f current public expenditures--from around 32 percent o f GDP in 1998-99 to around 29 percent of GDP in 2002-while publiccapital expenditures remainednearly unchanged. Box 1.1Macroeconomic Stability, Growth and Poverty Poverty and inequality are inextricably linked to macroeconomic stability and economic growth. Countries that experience severe macroeconomic distress tend to suffer steep drops in employment and real incomes, with the burdeno f economic shocks often falling disproportionately on lower-income groups. Whereas high-income groups generally have diversified sources o f income, the poor tend to make a living from employment, thereby making it relatively difficult to hedge against a decline inreal wages due to unanticipated inflation. Furthermore, the concentration o f spending patterns on basic necessities can severely weaken the ability of the poor to adapt their consumption behavior to changes in relative prices. It is essential, therefore, that governments pursue policies supportive o f macroeconomic stability, growth, employment, and low inflation, while maintaining an appropriate safety net to protect the most economically vulnerable segments o f society. Key findings o fthe Poverty Assessment (2003) ai include that: At 28.9 percent in 2002 total poverty rate has decreased significantly since 2000 due to the remarkable combination o f disinflation and strong growth. Extreme poverty has also decreased inrecent years declining to nearly 11percent in2002. Consumption inequality decreased between 1995 and 2002, and then subsequently increased slightly, though primarily in urban areas. However, the level inequality remains comparable to that o f other EU accession countries. Average consumption per adult equivalent was most volatile between the sub-periods 1995 to 2000 and 2000 to 2002 for self-employed persons outside o f agriculture. This was followed by employed wage earners, pensioners, and the unemployed. Most likely due to subsistence living, self-employed persons in agriculture experienced below average contraction and expansion o f consumption inthese two periods. Most o f the change inboth extreme and total poverty due to recession and economic recovery is associated with intra-sectoral effects (i.e., shifts within occupational activities such as wage employment, self- employment, and pensioners) rather than population shifts across sectors. Population's shifts across occupational categories accounted for respectively 30 percent and 24 percent o f extreme and total poverty changes inthe sub-period 1995 to 2000, butpractically no part o f the shifts in2000 to 2002. Changes in extreme urban poverty accounted for 56 percent and 71 percent, respectively, o f the total change inextreme poverty inthe two sub-periods. Total urban poverty represented 55 and 64 percent of the total changes inpoverty rates. - Romania:PovertyAssessment, VolumesI and2, Reportnumber26169-RO, TheWorldBank,2003. a/ 4 1.6 Highest priority on the fiscal side is to enforce financial discipline on the enterprise sector andthe bankingsector. Inthe case o fRomania, this implies eliminating the widespread practice o f nonpayment, arrears, and barter settlement i s central together with completing the restructuring of the bankingsector. Economic restructuring, including privatization or liquidation o f large SOEs, and implementing hard budget constraints on the enterprise sector and strict incomes policy remain as a major challenge to safeguard stability, growth, and integration. Upgrading skills is also needed to secure the benefits o f European and global integration. Romania faces important challenges in education reform (Section 5 in Volume 1)-the most recent TMISS international math and science assessment shows that Romania scores are below international average and declined between 1995 and 1999. Public expenditures to GDP in education in Romania i s about 3.2 percent, which is low compared to CEECs average o f 4.4 percent. In fact, implicit subsidies and tax arrears relative to GDP are each higher than the resources allocated to the education sector. This is indicative o f some of the trade-offs inpublic expenditure policies. Table 1.2 Consolidated General Government (percentage of GDP) 1998 1999 2000 2001 2002 Total revenue and grants 29.7 31.9 31.4 30.5 29.7 Current revenues 29.7 31.8 31.2 30.4 29.6 olw Tax revenues 27.8 30.1 29.4 28.3 27.6 Total expenditure 35.1 35.5 35.4 33.7 32.4 Current expenditure 31.8 32.6 31.9 30.5 29.2 Goods and services 11.4 12.6 12.5 12.1 12.0 Wages and salaries 5.0 4.8 5.5 5.0 4.9 Other (materials and operating exp.) 6.4 10.5 7.1 7.0 7.1 Interest 4.7 5.3 4.9 3.9 3.0 Subsidies andtransfers 14.8 14.7 14.5 14.5 14.0 Subsidies 1.7 1.7 2.2 2.1 2.0 Transfers 13.1 13.0 12.3 12.4 12.1 Capital expenditure 3.6 2.8 3.1 3.2 3.2 Lending less repayments 0.6 0.1 0.4 0.1 0.1 Overall balance -5.4 -3.6 -4.0 -3.3 -2.6 Financing 5.6 3.3 3.5 3.2 2.3 Domestic 3.9 1.6 -0.2 0.9 0.3 External 0.0 0.4 2.8 1.7 1.5 Privatization 1.8 1.3 0.6 0.3 0.4 Bank asset recovery 0.0 0.0 0.3 0.2 0.1 Discrepancy 0.2 -0.3 -0.5 -0.1 -0.3 Memorandum: Primary balance("A of GDP) -0.7 1.7 0.8 0.6 0.4 Source: Ministry of Finance. A.3. Incomes Policy, Exchange Rate, and Prices 1.7 In recent years, incomes policies, and exchange rate and monetary policies have been conducive to reducing Romania's inflation rate-inflation has declined from 46 percent in 1999 to 14.1percent in 2003. Romania has been exercising greater wage discipline in SOEs which in 5 turn has contributed to lowering of inflation rate. Though there has been a surge in 2003, generally wage growth has been constrained in recent years to rates that are more in line with productivity improvements, particularly since 2001. Inthe past, excessive wage growth in SOEs, and the means used to finance it, have been a key source of monetary growth and of inflati~n.~ Real wages grew at an average monthly rate of 1.4 percent in 2000 to 2002 compared with 1.8 percent in 1995 to 1999. As shown in Figure 1.1, the ebbs and flows of the growth rate of nominal wages have often presaged changes in inflation. Inturn, lower inflation reinforced this trend by reducing the cost-of-living adjustment needed to maintain real wage rates. Adding to this virtuous circle has been the slowdown in exchange rate adjustment that usually has been needed to maintain external competitiveness. The rate of nominal exchange devaluation has decreasedsteadily from 73 percent in 1999to 14 percent in2002.4and 9.8 percent in2003. Figure 1.1Wages, ExchangeRate, and Prices (12-month percentage change) YO 200 0 ' Nominalwage -CPI j 1 -~ CPI -- Lei/USD Source: World Bank staff estimates based on SIMA information. 1.8 Substantial progress inprivatization o f large SOEs, elimination of non-payment practices and corresponding barter settlement of arrears, and stronger foundations of fiscal policies would be needed for Romania to start moving towards a framework o f inflation targeting (see section D.3 below). The dynamics o f Romanian inflation are analyzed indetail inDavidMoore, "Inflation inRomania-Developments and Determinants," in The Road to EU Accession, IMF, 2001. The conclusion o f the paper is that increasing unit labor costs and exchange rate depreciation have been the primary determinants o f inflation since the early 1990s. Growth o f monetary aggregates and credit no doubt were an important sources o f inflationary pressure, but ultimately the medium-term path o f inflation was determined by policy decisions regarding incomes policy and exchange rate adjustment. 4 Nikolay Gueorguiev in "Exchange Rate Pass-Through in Romania", IMF Working Paper WPlO3l130 o f 2003, estimates that historically 27 to 43 percent o f the exchange rate change has passed through to consumer prices, and to 59 to 72 percent o f producer prices. Vis-&vis a 60140 percent basket o fEuro and USD. 6 A.4. Implications for External Sustainability 1.9 Fiscal consolidation, more supportive incomes policy, lower inflation, and slower real exchange rate appreciation, contributed to the improvements inextemal sustainability in 1999 to 2002. An improved level o f national savings and in the balance between national savings and investment resulted in a decline o f the external current account deficit by over 2 percentage points o f GDP in 2002, following a deterioration in 2001 (Table 1.3). Most o f the improvement inthe external current account deficit, over 80 percent, was due to a decline inthe trade deficit. This reflected declining import demand due to moderation in the growth o f domestic consumption together with relatively strong growth o f exports. In dollar terms, export growth more than doubled to 22 percent in2002 relative to 2001, primarilydue to strongperformance in manufactures including textiles, footwear, and (see Chapter 2). Exports contributed 5.3 percentage points to the average GDP growth rate o f 4.2 percent in 2000 to 2002, while the negative contribution of imports fell from 9 to 5 percent. More recently, large extemal capital inflows and rapidwage growthhave fueled strong demand growth and a widening o f the current account balance to over 6 percent o f GDP in2003. Table 1.3 Balanceof Pavments.1999-2002 (% of GDP) 1999 2000 2001 2002 Current account -4.1 -3.9 -5.6 -3.4 Trade account -3.5 -4.6 -7.5 -5.7 Exports 23.8 28.1 28.7 30.3 Imports 27.4 32.7 36.1 36.0 Services account -2.4 -1.7 -1.o -1.1 Unrequited transfers, net 1.8 2.3 2.9 3.4 Capital account 6.0 7.4 7.9 8.1 Direct investment and capital transfers 3.1 2.9 3.2 2.6 Portfolio investment -0.2 0.1 0.2 0.3 Public borrowing, net 2.4 2.8 2.8 Private borrowing, net 2.0 1.4 1.4 Trade credit, net -0.1 0.4 1.1 oiw short-term, net 0.0 0.5 1.o Net errors and omissions 0.7 2.5 -2.0 Overallbalance 1.8 4.2 4.8 2.6 Financing -1.8 -4.2 -4.8 -2.6 Net foreign assets o f NationalBank, net -2.4 -3.0 -4.9 -4.3 Net foreign assets o f dep. money bks, net 0.6 -1.3 0.1 1.7 Source: InternationalMonetaryFund. 1.10 The positive trade developments in recent years reflect greater competitiveness and growing export market share in Europe made possible by an improved policy environment and new opportunities associated with EUintegration. Better macroeconomic performance, including lower inflation and faster productivity growth in tradeables particularly exports (see Chapter 2), have been critical for containing real exchange rate appreciation. Following the 15 percent devaluation o f the real effective exchange rate in 1999, the annual rate o f appreciation averaged 4.6 in 2000 to 2002, andjust 2.1 percent in the last two years. The latter is less than one-half of 7 the growth rate o f GDP, for which one-half could be a reasonable estimate of the rate o f growth oftotal factor productivity.6 1.11 Enhanced budgetary discipline and better trade performance have translated into more favorable savings-investment balances for growth and external current account stability. As shown inTable 1.4, gross domestic savings (GDS) increased from 11.4 percent o f GDP in 1999 to 17.3 percent o f GDP in 2002, reflecting relatively moderate, though uneven, consumption growth. In2003, the savings rate is preliminarily estimated to have been 16.9 percent of GDP, a slight decrease resultingmainly to slightly lower non-government savings. The increase inGDS and a rise innet foreign transfers totaling 1.7 percentage points of GDP in 1999 to 2002 have increased national savings by nearly 8 percentage points o f GDP over this period. Government savings accounted for about 20 percent o f this increase, while the remaining 80 percent i s due to greater levels o f private saving. Along with foreign savings o f about 3.5 percent o f GDP, this has made it possible to finance a 7 percentage point of GDP increase in gross domestic investment, over 90 percent of which originated from the private sector. Table 1.4 Savine and Investment Balances 1999 2000 2001 2002 Gross domestic savings 11.4 14.2 14.9 17.3 Net transfers from abroad 0.6 1.4 2.1 2.3 National savings 11.9 15.6 17.0 19.7 Government -0.9 -1.0 -0.1 0.6 Non-government 12.8 16.6 17.1 19.1 Gross domestic investment 16.1 19.4 22.5 23.1 Government 2.8 3.0 3.1 3.2 Non-government 13.3 16.4 19.4 19.9 Foreign savings 4.1 3.9 5.5 3.4 Source: IntemationalMonetary Fund. 1.12 The recent downward trend infiscal and external deficits has helped to contain the rate o f increase o f external and public indebtedness-total external debt has increased from around 26 to 34 percent o f GDP between 1999 and 2002 (Table 1.5). Public and publicly-guaranteed debt shows sustained increase o f 1.9 percentage points o f GDP since 2000 reaching 26 percent o f GDP in2002. While stocks o f external andpublic debt ofRomania compare favorably with debt levels in developing and transition economies this rate o f debt accumulation is not likely to be ~ _ _ _ _ _ See section on growth below. 7 There is no standard set o f benchmarks for measuring debt sustainability. However, there have been studies to determine critical values for indebtedness indicators beyond which a country would most likely default. D. Cohen in "Growth and Extemal Debt: A New Perspective on the African and Latin American Tragedies" (CEPR Working Paper No. 1753) estimates critical points that could help to determine the sustainability of both public and total external debt. His benchmarks are net present value o f debt service/GDP o f 50 percent, NPVIXGS o f 200 percent, and NPVIrevenue o f 290 percent. Countries whose ratios exceed these benchmarks were found to have a risk of rescheduling external debt exceeding 60 percent. The World Bank provides a set o f thresholds for classifying middle-income countries by the degree o f indebtedness. Countries whose N P V o f total external debt service to GNP ratio between48 and 80 percent and net present value o f debt service to export inthe range o f 132 to 220 percent are considered by the Bank to be moderately indebted. If any one o f these two ratios exceeds the top the range, then a country is considered "severely indebted". When both ratios are below the bottom o f the benchmark ranges, the 8 sustainable in the longer term. Further fiscal consolidation and structural reforms are needed to moderate the rate of debt accumulation andpreserve low levels o f indebtedness. 1.13 Total external debt service is below 20 percent of exports o f goods and services, and has been declining. This reflects a lengthening of the maturity structure of the external debt, thereby helpingto reduce principal payments as well as stronger export performance. Inaddition, interest payments have been coming down due to declining global interest rates and a reduction in Romania's country risk premium, which came down from nearly 800 basis points in 1999 to about 300 basis points in 2003. The capital market's increasing confidence in the Romanian economy undoubtedly reflects in part the growing accumulation of foreign exchange reserves. Gross official reserves of the National Bank o f Romania have increased from 2.1 months of imports in 1999 to nearly 3.5 months in2002.8 At the same time, the ratio o f official reserves to short-term debt, an indicator o f vulnerability to suddencapital flow reversal, has risen from 6 in 1999 to about 14 at the end o f 2002 (Figure 1.2). Table 1.5 Indicatorsof PublicandExternalIndebtedness,1999 2002 - 1999 2000 2001 2002 In % of GDP: Public and publicly guaranteed debt 1/ 24.2 23.6 24.6 25.5 olw external 16.1 17.4 18.6 20.0 Total external debt 25.6 28.9 31.1 34.3 In % of exports of goods and services: Total external debt 92.5 87.1 92.1 96.8 Total external debt service 29.3 16.2 19.4 21.6 olw principal 24.3 11.9 15.0 17.7 interest 5.0 4.3 4.4 3.9 In % of general government revenue: Public and publicly guaranteed debt 90.5 75.4 80.6 86.2 General govt. interest payments 16.6 15.5 12.7 10.2 Memorandum: Variable rate ext. public debt (manUS$) 3100 3771 4109 4861 % total public external debt 49.8 54.2 53.2 51.0 Short-term debt (% oftotal external) 4.2 3.5 3.3 3.0 Gross official res. inmos of imp. of GNFS 2.1 2.5 3.2 3.9 Spread of benchmark bonds (bps, eop) 780 406 421 282 Treasury billrate (eop) 104.8 59.4 38.4 17.4 11Excludes IMFpurchases andtemporary borrowing fromthe State Treasury. Source: Ministry of Public Finance; InternationalMonetary Fund; World Bank staff calculations. country i s classified as "less indebted". Finally, the Maastrict treaty establishes a public debt benchmark of 60 percent o f GDP for participation inEuropeanMonetary Union. The buildup o f reserves is accounted for, in part, by trade developments and strong capital inflows discussed below. 9 Figure 1.2 Official Reserves to Short-term Debt I Official Reservesto Short-termdebt I i I 16 1200 1000 800 600 400 200 0 95 96 97 98 99 00 01 02 I i 0OfficialReserveslShort-termdebt+Short-term debt Source:World Bankdatabases, SIMA. B. ROMANIA'S PERFORMANCE REGIONAL IN PERSPECTIVE 1.14 With some exceptions, Central Eastern European Countries (CEEC) have maintained a solid record o f performance. Economic growth o f Romania compares well with that o f other countries in the region. Overall fiscal and external current account deficits also compare favorably with other CEEC. Better macroeconomic management has contributed to these improvements. 1.15 Nevertheless, as illustrated inTable 1.6, challenges remain inRomania if it is to catch up with solid performers in the region. There are weaknesses in the foundations o f Romania's macroeconomic and growth performance. While inflation rate in Romania has declined from 45.8 to 14.1 percent between 1999 and 2003, current inflation rates greatly exceed those of other countries inthe region whose inflation rates have been inthe single digit levels since late 1990s. Although fiscal deficit o f Romania does not seem high relative to the other countries, further fiscal adjustment and stricter incomes policy are needed to sustain gradual but sustained reduction o f inflation and create fiscal space for priority investments. Similarly, while the external current account deficit does not seem high compared to other countries in the region, unlike these, the financing of the deficit is increasingly provided by relatively mobile capital and less by foreign direct investment (FDI). This exposes the economy to the risk o f sudden capital flow reversal and could ultimately hurt growth. Much o f this capital represents the borrowings o f commercial banks that are on-lent to domestic companies and households inforeign currency at relatively high domestic interest rates. Although the prudential and supervisory framework has been strengthened since the banking crisis challenges remain in the basic infrastructure o f the financial sector (see Chapter 6), and transfer o f currency mismatch to enterprises and household balance sheets increases the vulnerability o f these entities to sudden exchange rate adjustment, andmay create large contingentpublic liabilities. 10 Table 1.6 RomaniaandCEECs: SelectedEconomicIndicators.2002 Interest rates Overall External Broad Private Average Growth Inflation GDP (Avg. long Fiscal CAB Money Sector FDI(1995- (Yo) (Yo) term) Balance output ( O h2002) (Yo of (YO) (Yoof GDP) (Yoof GDP) (Yo of GDP) of GDP GDP) Romania 5.0 22.5 35.2 a/ -2.8 -4.8 23.7 65 2.7 Bulgaria 4.8 5.8 9.8 -0.6 -4.4 43.3 75 4.3 Croatia 5.2 1.7 12.8 -4.8 -7.1 66.0 60 4.5 Czech Rep. 2.0 1.8 3.6 bl -0.9 -5.3 75.5 80 7.4 Estonia 5.6 3.6 7.9 1.2 -12.8 42.6 80 5.1 Hungary 3.3 5.3 -9.5 -4.0 47.2 80 4.2 Latvia 6.1 1.8 -2.5 -7.8 36.5 70 5.5 Lithuania 5.9 0.3 6.8 -1.2 -4.4 29.3 75 4.0 Poland 1.3 1.9 -6.6 -3.9 64.6 75 2.9 Slovak Rep. 4.4 3.3 7.7 bl -7.4 -8.3 65.3 80 5.4 Slovenia 3.2 7.4 -1.9 -1.8 55.5 65 1.9 Notes: ai Bank lending(weightedaverage); biInterbankrate. Source: IntemationalFinancialStatistics, IMF; World Bank database SIMA. 1.16 At 24 percent o f GDP, the level o f monetization o f the Romanian economy is about half the average level o f other countries inthe region. The banking sector provides most the financial intermediation, holding about 90 percent o f the total assets inthe financial system. As a result o f important reforms in the banking sector since the banking crisis, today the liquidity o f banks is high, the banking sector seems to be well capitalized and profitable. However, in contrast to other countries inthe region where the banking sector i s mostly private-median o f state owned banking i s 4.6 percent o f total assets-state owned banks in Romania hold nearly 40 percent o f the bank assets and all deposits, and 32 percent o f loans. These assets are, however, concentrated in only two state banks, both earmarked for privatization. The low level of financial intermediation reflects weaknesses in the basic infrastructure needed for financial sector development-the banking sector in particular-the still dominant role o f state owned banks, and the heavy reliance on non-payment, accumulation o f arrears, andrelated barter settlement o f these as the most important source o f financing (see Chapter 6). 1.17 FDIinflows are expected to increase, but linger below potential, reflecting a weak but improving business environment. Foreign direct investment attracted by Romania crossed the $10bn threshold inSeptember 2003, cumulatively from the beginning o f the transition. However, at around $430, per capita FDI remains several times lower than in neighboring accession countries. Both the volume and the composition of the FDIinflows are linked with the progress achieved in privatization and the perception o f the investors vis-a-vis the attractiveness o f the business environment. So far, FDI has been concentrated primarily on the labor-intensive, export-oriented industries, while greenfield investment has been limited. 1.18 The low levels o f FDIflows to Romania compared to other countries inthe region places Romania at a disadvantage in terms o f its development prospects compared to other countries in the region. Solid performers in the region have leveraged economic restructuring, financing needs, and resulting net job creation through FDI-initially a large share o f FDIwas related to privatization and as the policy framework improved green field FDI increased in importance. The low levels o f FDI in Romania reflects the slow progress in economic restructuring o f large SOEs and weaknesses in its policy framework. The share o f the private sector in the economy has remained low at 67 percent o f GDP compared to an average o f other countries in the region o f 76 percent o f GDP. Implementation o f fundamental structural reforms and further fiscal consolidation are needed to improve the size and performance o f the rivate sector and reorient the role o f the state towards the provision o f basic public goods in a predictable policy ! framework. Economic restructuring, including privatization o f large SOEs, remains as a major challenge in Romania to safeguard macroeconomic stability, and enable the country's growth potential and meaningful net employment opportunities. 1.19 Romania's strong trade performance i s resulting in its solid integration with EU, sub- regional and global markets (Table 1.7). Trade integration is progressing faster compared to other economic dimensions o f integration discussed above. While total trade relative to GDP i s low compared to other CEECs, both the share o f imports intotal imports andthe share o f exports intotal exports frodto the EUand sub-regional markets are at the higher end o fthose observed inother CEECs. This indicates that trade integrationwith the EUis playing a significant role in economic restructuring and productivity gains in the tradable sectors. Romania i s establishing strong economic links in this highly competitive market. While FDI levels are low, there i s evidence that these flows are having a strong pull effect in Romania's trade performance given the strong links o f these flows to small and medium enterprises (see Chapter 2). Limited domestic financing to small and medium enterprises (SMEs) does not seem to be holding back these companies from achieving productivity and quality levels required to compete in EU markets. Which in turn, this implies that the benefits o f FDI, and consequently policies that affect FDI flows, o f incorporating local producers into global production networks are far more important and have more impact in terms o f economic restructuring and productivity gains than other policy dimensions such as limited domestic financing to SMEs. Measures to improve the policy framework that affect FDI flows across the board include securing macroeconomic stability and structural reforms to improve investment climate. 1.20 Romania's impressive trade performance is indicative o f the potential o f this economy. This potential, however, remains mostly untapped due to the protracted implementation o f price liberalization and structural reforms. Figure 1.3 shows two indicators o f progress o f reforms for CEEC: one for the initial phase and one for the second phase o f reforms. Romania's ratings o f the initial phase o f reforms is good, other CEEC have continued to advance. While Romania had a better rating initial phase o f reforms than Bulgaria in 1999, by 2003 Bulgaria's rating surpassed that o f Romania. Romania has made some progress inthe second phase o f reforms between 1999 and 2003. However it is inthe second phase o freforms where Romania has laggedbehind other CEEC the most. Romania's rating in second phase o f reforms in 2003 is only marginally comparable where Bulgaria, Latvia and Lithuania second phase o f reforms were back in 1999. If Romania i s to catch up with other CEEC in second phase reforms, structural reforms in enterprise restructuring and large scale privatizationwould need to be implemented decisively. Basic public goods include education, primary health care services, a social safety net targeted to the poorest 8 segments of its population. 12 Table 1.7 Romania and CEEC: Selected Indicators of Integration, 2002 Imports Imports Imports from Exports to EU Exports to FDIfrom plus from EU CEEC-10 (YOof total CEEC-10 EU Exports to (YOofTotal (YOof total exports) (YOoftotal (YOof GDP(Yo) imports) imports) exports) total FDI) Romania 69.3 61.6 5.1 65.7 11.3 61.1 Bulgaria 83.4 54.3 5.2 55.5 10.5 69.1 Croatia 113.9 56.9 50.6 71.2 Czech Rep. 68.1 57.4 2.4 65.4 1.3 84.1 Estonia 125.2 65.5 2.3 69.4 6.4 Hungary 108.9 56.2 9.8 72.6 10.0 80.2 Latvia 78.5 51.2 3.1 52.5 21.5 Lithuania 96.9 42.6 0.3 43.9 0.7 Poland Slovak Rep. 81.3 Slovenia 96.3 67.6 9.3 57.7 5.8 85.6 Note: FDIshares are for 2001; all other indicators are 2002 data. Source: World Bank staff estimates based on UNCOMTRADE and World Bank database SIMA. 1.21 Based on GDP per capita at PPP, Romania i s the poorest country with the second largest population among the CEEC. Inaddition to makingprogress inpoverty reduction-poverty rates have declined from a peak level o f 13.8 in 2000 to 10.9 in 2002-Romania's social indicators have also been improving (Table 1.8). Adult illiteracy, infant mortality as well as mortality o f under-5 years o f age, school enrollment and life expectancy at birthhave improved since the start o f transition. These improvements are remarkable taking into account that Romania, besides beingone o f the poorest countries inthe region, it seems to have started the transition ina more precarious condition in terms o f social indicators and without as an effective social protection system as other CEECs. Direct comparison o f social indicators, irrespective o f differences inper capita incomes, show that Romania, on the whole, was andremains worse o f f than other CEECs. Improvements o f social indicators over time have been slower in Romania compared to other CEECs-macroeconomic instability, slow progress in structural reforms, and wide fluctuations ineconomic growth havecontributedto the slow paceofimprovementsinsocial indicators. 1.22 However, differences inRomania's social indicators to other countries inthe region seem to be explained almost entirely by differences in per capita income, except in the case o f mortality rates. Thus improvements inper capita income, that is stability and growth, inRomania seem to have a large pay o f f in the improvement of social indicators. This suggests that the potential impact on social indicators o f rising per capita income supported by a fiscally sustainable and well targeted social safety net i s likely to be large. Reorienting public expenditures away from fiscal support to the enterprise sector and towards sustainable and well targeted social public expenditures would improve growth prospects and social indicators. 13 Figure 1.3 Selected Indicators of Progress of Reforms: CEECs Initial Phaseof Reforms II Second Phase of Reforms EBRDIndex I EBRD Index $lovak Rep. Hungary Poland Estonia Lithuania I Czech Rep. Latvia I 1 Hungary Poland Czech Rep. Lithuania Slovenia Latvia ~ Estonia Slovak Rep. Croatia Slovenia Bulgaria Croatia Romania PA!!!!! ~- Bulgaria Romania 3.00 3.50 4.00 4.50 , 1.oo 2.00 3.00 4.001 Source: EBRD Transition Indicators; Indicators for initial phase of reforms are calculated as unweighted averages of indicators for: small-scale privatization; price liberalization; trade and foreign exchange system. Indicators for second phaseof reforms are calculated as unweighted averages of indicators for: large-scaleprivatization; governance and enterprise restructuring; competition policy; banking reform and interest rate liberalization; securities markets and nonbank financial institutions; infrastructure reform. Table 1.8 Selected Social Indicators, CEECs GDP per Mortality Mortality rate, capita, PPP Adult Illiteracy School enrollment, rate, infant under-5 (per Life expectancy at (US$) ( Y O ) secondary (% net) (per 1,000) 1,000) birth,total (years) 2002 1990 2002 1990 2000 1990 2001 1990 2001 1990 2002 Bulgaria 6909 2.8 1.4 63.3 87.6 14.8 14 18.7 16.0 71.4 71.8 Croatia 9967 3.1 1.5 63.2 79.0 lb 10.7 7.0 12.5 8.0 72.2 73.8 Czech Republic 15148 86.1 la 87.1 la1 10.8 4.0 12.4 5.0 71.7 75.0 Estonia 11712 0.2 0.2 82.3 /h 82.8 12.4 11 17.2 12 69.5 70.6 Hungary 13129 0.9 0.6 74.8 87.2 I C 14.8 8.0 16.8 9.0 69.3 71.8 Latvia 8965 0.2 0.2 76.8 Id 74.4 13.7 17.0 18.1 21.0 69.3 70.4 Lithuania 10015 0.7 0.4 80.5 l e 88.6 10.3 8.0 13.5 9.0 71.3 72.7 Poland 10187 0.4 0.3 75.8 90.9 19.3 8.0 21.9 9.0 70.9 73.8 Romania 6326 2.9 1.7 72.8 if 79.6 26.9 19.0 35.7 21.0 69.7 70.0 Slovak Republic 12426 74.9 11.96 8.0 14.1 9.0 70.9 73.3 Slovenia 17748 0.4 0.3 88.6lg .. 8.4 4.0 10.2 5.0 73.3 75.9 Source: World Bank database (SIMA). Notesla data 1993 and la1 1995; Ib 1997; I C 1999; Id 1993; l e 1994; If 1993; lg 1997; 14 Figure 1.4 Overall Fiscal Balance, Current Account, and Growth, 1992-2002 (Yo GDP) I (% GDP) (Yo) 0.0 ("/.) 8 0 I 1 6 0 I I -2.0 4.0 ' -4.0 2.0 , 0.0 -6.0 -2.0 -4.0 i -8.0 -6.0 I ~ ~ I -8 0 1-10.0 -100 I -;E--Fiscalbalance t-Current account , I I-GDPgrowth 1 l-GDPgrowth -1 I Source: World Bank staff estimates based on SIMA. c.GROWTHAND INTER-SECTORALRESTRUCTURING 1.23 Structural weaknesses and macroeconomic instability have been very damaging to growth. Figure 1.4 shows the evolution o f economic growth and the internal and external imbalances inRomania and the first-wave accessioncountries. InRomania, these variables have exhibited wide swings reflecting cycles of growth in demand. Before the banking crisis, fiscal and quasi-fiscal stimulus acceleratedconsumption growth and growth began to recover from the collapse at the start of transition, but with practically no recovery o f investment to sustain it. At the same time, the strong increase in demand put considerable pressure on prices and the real exchange rate. As a result, the current account deficit widened to unsustainable levels, necessitating a sharp contraction in demand and output. After the crisis, the renewed reform effort resulted in lower fiscal and external deficits that provided a stronger basis for sustained growth. In the CEECs, by contrast, growth has generally not exhibited a pronounced cyclical pattern despite a sharp deterioration of fiscal and external balances in the second half of the 1990s. These countries have maintained relatively strong investment growth, as well as increasing consumption. 1.24 The implication of output volatility inRomania is that on average growth has been quite low in comparison with the advanced Central European reformers. Between 1990 and 2002 the economy grew at an average rate of about -1 percent. The growth rate inthe "first wave" o f EU accession countries, by contrast, has averagedaround 1.5 percent." As a result, Romania has not fully recovered the output lost inthe beginning of the 1990s, while the accessioncountries have, on average, grown their economies almost one-third inthe last 12 years. loThis includes all accession countries, except Cyprusand Malta. 15 7igure 1.5 Growth Drivers: Romania an1 EECs 2o, Romania: Contributionto Growth, YO CEECs*: Contribution to Growth, YO 15 8 6 15 10 4 10 5 2 0 5 0 -2 -5 4 0 -10 -6 -5 -8 1 - 1 0 -10 1992 1994 1996 1998 2000 2002 =Final consumption OGDl OExports Blmports yource:NIS andWorld Bank staffestimates. Source: WorldDevelopmentIndicators, Eurostat. 1.25 The decomposition of the growth by its expenditure components inFigure 1.5 shows that Romania's weaker performance interms o f investment growth compared to advanced reformers in CEECs. In the CEECs, the contribution to growth of these variables has nearly always exceeded the contribution from consumption. The advanced reformers in Central Europe experienced an average investment growth o f 7.1 percent in 1996 to 2002, while the growth rate inRomaniaaveraged only about 2 percent. This suggeststhat generallythere hasbeena superior environment for domestic investment and FDI in comparison with Romania and that macroeconomic instability inRomania duringthis period took a heavy toll. 1.26 Moreover, as a result o f their strong extemal positions they have been able to maintain relatively stable consumption growth and strong import demand. While inRomania arguments in favor o f a slow pace for reform often center on the need to protect household consumption, such spending has been unstable, often exhibiting a contraction. 1.27 There are also important differences between Romania and the CEECs in terms o f the sectoral transformation o f the economy. At first glance, it appears that the economy is evolving towards a modem structure o f production similar to pattems in the EU, as have other CEECs. Namely, the shares o f agriculture and industry intotal production have been declining, while the service economy has become an increasingly important source o f income (Table 1.9). To a large extent, this process o f transformation has been more pronounced inRomania. Yet, it is important to note that the country emerged from the system o f central planningwith relatively high shares o f agricultural and industrial output, and a small service sector. Agriculture's share o f production inthe early 1990s, for example, represented over three times the levelinfirst-wave countries. 16 Table 1.9 RomanianGrowth Performance in 1990 2002 - A Avprprro Ro-1 CnP Crnwth l%i Index of Real GDP in2002 1990-2002 1995-2002 (1990 = 100) Romania -0.9 1.3 91.9 Accession Countries 11 1.4 3.9 127.7 B. Structure of Production (YO)2/ 1990 1995 2001 Romania Agric, forestry & fishing 23.7 21.5 14.8 Industry 50.0 42.7 34.6 Services 26.3 35.8 50.6 Accession Countries 3141 Agric, forestry & fishing 7.5 6.5 3.7 Industry 45.9 43.6 38.3 Services 46.7 49.9 58.0 C. Structure of Employment (YO) 1990 1995 2001 Romania 51 Agric, forestry & fishing 29.0 34.4 41.4 Industry 43.5 33.6 27.2 Services 27.5 32.0 31.3 Accession Countries 3141 Agric, forestry & fishing 19.0 18.4 17.6 Industry 35.7 34.0 30.8 Services 45.3 47.6 51.5 Memorandum: Romania Access. Cts. Average growth in 2000 - 2002 (YO) 4.2 3.O NO~PP. l/Averagesarebasedonpopulationweightsof"firstwave"accessioncountries, excluding Cyprus and Malta. 2/ Share of GDP at factor cost. 3/ Excludes Cyprus, Malta, andthe Baltic countries. 4/ Observations for 1990 are weighted average ofthe earliest available year, which are as follows: Output data: Romania, Czech Rep., Slovenia - 1990;Hungary - 1991; Poland, Slovakia- 1992. Employmentdata: Romania, Slovenia 1990; Hungary, Poland - 1992; Czech Rep. - 1993; - Slovakia- 1994. 5/ Romania'semployment data i s for 2000. Source:National statistics reported in the 2002 Handbook o f Statistics ofthe Vienna Institute for Intemational Economic Studies; WE3 staff calculations. 1.28 In contrast with these countries, however, between the early 1990s and 2001 the agricultural share of total employment increased from around 29 percent to 41 percent, while the share of industrial employment dropped from 50 to 35 percent. In the accession countries, agricultural and industrial employment shares declined, respectively, around 1 and 5 percentage 17 points, suggesting that a normal process o f economic and social development i s occurring in these countries. l1 1.29 The implication o f these trends for growth potential cannot be understated. The incentive structure o f the economy has been shifting resources towards relatively unproductive sectors. In other CEECs market forces were relatively freer to allocate resources to efficient uses. In Romania, market incentives have been counteracted by persistence o f subsidies, price controls, and other factors driving resources in the opposite direction. With more labor being shifted to agricultural activity, national productivity levels could grow too slowly to significantly increase living standards, as the amount o f investment in capital (i.e., machinery and other modem equipment) available per worker to produce goods and services is unlikely to increase rapidly over time with such an allocation o f labor. 1.30 That resource shifts to agriculture are unlikely to improve labor standards may be seen from the findings of the recent Poverty Assessment. The rise inextreme and total poverty among agricultural workers, respectively, accounted for 14 and 9 percent o f the total increase inpoverty rates in 1995 to 2000. When poverty rates beganto fall in2000 to 2002, they represented only 12 and 7 percent o f the total change. Moreover, most o f the poverty change inboth sub-periods was due to changes in poverty inurban areas. Extreme and total poverty inrural areas were over 40 percent o f the overall increase inpoverty until 2000. The subsequent national-level reduction in poverty, rural areas accounting for only 30 to 35 percent o fthe improvement. 1.31 As shown in Table 1.10, productivity improvements have been crucial for growth.12 In the 1990s total factor productivity (TFP) growth has accounted for most, if not all o f economic e~pansion.'~ Between 1991 and 2002 the contribution from labor was negative as a result o f labor shedding. Due to periods o f contracting investment followed by weak capital accumulation, on average there was no contribution from capital. Growth was entirely due to more efficient use o f labor and capital, and likely the result o f the limited privatization and restructuring that has been achieved. Table 1.10 Accountingfor EconomicGrowth Average annual growth rates Contributions to growth (percentages) GDP Labor Capital Labor CaDital Total factor 1991-1998 -0.5 -1.8 -2.4 -1.2 -0.8 1.5 1999-2002 4.1 -0.5 3.5 -0.3 1.2 3.3 1991-2002 0.7 -1.5 -0.8 -1.0 -0.3 2.0 Source: World Bank staff estimates. 'I This estimate excludes data from Cyprus, Malta, and the Baltic states. l2 The numbers in Table 1.10 are based on the assumption that the capital share o f income is one-third., an 1.assumptioncommonly made ingrowth studies. For a brief overview o f growth theory and accounting se Appendix [ The appendix also presents the results of a sensitivity analysis o f the assumptions underlyingthe model used to generate the calculate the contributions. This analysis finds the estimates below to be quite robust to changes in the capital share o f income. l3 TFP growth is measured as a residual source of growth once labor and capital contributions have been accounted for. Generally speaking, the size of the physical capital stock is especially difficult to quantify. In the case o f Romania there i s also a large informal economy which may lead to inaccurate estimates of output and labor and capital input. TFP estimates, therefore, canbe highly sensitive to measurement error. 18 1.32 An important question to address is whether productivity growth can continue so strongly without faster investment growth, particularly that which is linked closely with FDI. Inthe long- run, efficiency can only increase from the adoption of new technologies, including managerial techniques and advanced capital goods. Hence the conclusion is once more that the incentive structure o f the economy needs to be changed to encourage greater domestic and foreign investment. This will also require reforms that encourage more domestic savings and further financial restructuring and a business climate that attracts more long-term foreign capital to improve the intermediation between savers and investors. This will lead to a reallocation o f resources to the most productive sectors o f the economy. 1.33 The findings o f this growth accounting suggest that Romania could grow at least 4.5 percent per annum. However, with a strong structural adjustment program the economy could potentially grow much faster. If investment in more efficient capital increases, the decline in employment stops, and total factor productivity at the pace experienced in recent years, then economy could potentially growth much faster. Assuming that the investment rate rises sufficiently, capital's contribution to growth could increase to 2.0 to 2.5 percent.I4 Employment growth could be at least zero inthe next few years as new job creation more strongly offsets new unemployment as a result o f industrial restructuring. Furthermore, it may be possible to sustain the 3 percent TFP growth o f the last few years due the current low level o f capital inten~ity.'~ Altogether, these suggest that Romania's medium-term growth potential is between 5.0 to 5.5 percent. D.RISKSANDVULNERABILITIES 1.34 Elimination o f heavy reliance o f nonpayment for goods and services, taxes, and social security contributions by SOEs, the government, and some entities in the private sector (including recently privatized companies) are at the core o f Romania's development prospects. Nonpayment, arrears, and barter settlement undermines stability and growth. Key risks to budgetary stability inthe medium term include inflexibility o f public expenditures, and changes ininterest rates and foreign exchange rates. The social security system is a long term source o f fiscal vulnerability. D.l. Budgetary Stability 1.35 In the medium-term, the budget is exposed to considerable market risk. In particular, spending could increase substantially if there were a sharp rise in interest rates and a strengthening o f the dollar. One way to see the interest rate sensitivity o f budgetary expenditure is to note that about 2.2 percentage points o f the 2.0 percentage point reduction in current spending was a result o f interest savings. While total spending declined 3.1 percentage points o f l4 The estimated change in capital's contribution to growth is taken from Thomas Harjes, "Romania's Growth Potential inthe MediumTerm", inRomania: Selected Issues and Statistical Appendix, IMF, January 2003. In a recent growth accounting exercise, it was determined that average total factor productivity growth inthe largest first-wave accession countries in 1991 to 1999 varied from less than 1 percent to 2.9 percent. See "Real Convergence to EU Income Levels: Central Europe from 1990 to the Long Term", by P. Doyle, L. Kujis, and G. Jiang, IMF Working Paper WP/01/146 o f 2001. Also, the Selected Issues paper referred to above projects a TFP contribution rate of2 to 3 percent, depending inversely on capital's share o f income. 19 GDP between 1999 and 2002, primary expenditure decreased only 0.9 percentage points during this period.16 Moreover, the primary surplus, a key determining factor o f debt sustainability, actually declined from 1.7 percent o f GDP in 1999 to 0.4 percent of GDP in2002 and turnedinto a deficit of 0.3 percent o f GDP in2003. 1.36 Much o f the decrease in interest payments is due to a shift away from domestic borrowing to relatively inexpensive foreign f~nding.'~ Payments were also kept low by the pronounced decline in global interest rates in recent years.l8 However, given the historically low level o f rates in recent years, it is uncertain whether low interest payments will continue to support low fiscal deficits.19 In2002, for example, the effective rate o f interest on foreign debt paid by Romania was 11.4percent. Had this been 3 percentage points higher, the additional cost to the budget would have been 0.4 percent o f GDP, excluding the impact o f higher borrowing to finance the additional expenditure and any effect on domestic rates. This makes it important to maintain the reform drive to reduce domestic interest rates further and to ensure access to cheap foreign savings.20 1.37 The social security system is a key long-term source o f fiscal risk. The pension system deficits, in particular, are likely to grow further without more reform. Since financing gaps are mainly covered by the State budget, overall government deficits and public indebtedness will tend to rise. Inaddition, the bigger these programs become, the less the flexibility o f the budget to adjust to adverse shocks. Arguably, at least 90 percent o f total government expenditure i s non- discretionary. This assumes that quick spending cuts would not be easily made to wages and salaries, social transfers, budgetary subsidies to ailing state-owned companies, interest payments. It also assumes that public investment is prioritized and that at least half o f non-wage spending on goods and services would not be easily reduced. 1.38 Despite recent reforms, including a two-year increase inthe statutory retirement age and a widening o f the contribution base to include the self-employed, farmers and the unemployed, the pension system remains unsustainable.21 The dependency ratio (number o f pensioners to number o f contributors to the system) is close to 1. As it stands, the deficit o f the pension fund is about 1percent o f GDP. Currently, the system does not generate arrears to pensioners, but it is dependent on transfers from the state budget to close financing gaps. l6Primary expenditure equals total expenditure less interest payments. The share o f external public debt intotal public debt has increased from 58 percent to 74 percent between 1999 and 2002. Foreign interest costs compare favorably with domestic funding. The effective rate o f interest paid on foreign loans, definedas total interest payment denominated inlocal currency divided by the stock o f debt, was 11.4 percent in2002. This includes the effect o f currency depreciation. By contrast, the effective rate on domestic loans was 23.7 percent. Lower rates have affected the cost o fnew borrowing and existing obligations, which increasingly are contracted at variable rates. A t 51 percent in2002, the share o f floating rate obligations intotal external public debt has increased over 5 percentage points since 1995. l9 Six-month U S dollar LIBOR, for example, has decreased from an average rate o f 5.6 percent in 1990 to 2000 to about 1percent in2003. 2o Another way inwhich financing costs have been reduced is through temporary borrowing from the Ministryof Public Finances' treasury account at below market rates. 21 A comprehensive discussion o f the issues is provided the 2002 Romania Public Expenditure and Institutional Review. The fiscal section o f the 2001 IMF Selected Issues report provides a summary discussion. 20 1.39 Without further reform system imbalances will continue to grow and pose a serious threat to macroeconomic sustainability, particularly if the economy grows slowly. The urgency for reform i s increasing due to the projected slowdown o f population growth and steady aging o f the population, which will increase the demands on the system, while contributions tend to decline. Inaddition, there is aneedto reduce highpension contribution ratesto support business initiative and employment creation. D.2. FiscalSustainability 1.40 Given the fiscal risks and challenges confronted by Romania, it is important to ask what level of fiscal deficit would be sustainable, what level o f adjustment might be required in the face o f external shocks or a slowdown in economic growth, and whether macroeconomic adjustment mechanisms follow afiscal dominant or a monetaly dominant regime. 22 1.41 A simple, simulation exercise has been conductedfor this report to calculate an indicative level o f the sustainable primary surplus under alternative conditions (see Annex l).23 Inthis analysis, the primary surplus is said to be sustainable if the initial level o f public debt is maintained for given levels o f GDP growth, real interest rates on public debt, and m ~ n e t i z a t i o n . ~ ~ 1.42 Table 1.11shows the size o f the primary surplus needed to keep the debt to GDP ratio at the current level under alternative growth and interest rates. Ifthe growth rate decelerates by 2 percentage points, the primary surplus will have to more than double relative to the current level to stabilize the debt ratio. Similarly, in the event o f an increase in interest rates by 200 basis points (Le., closer to long-run levels), the primary surplus would need to increase about 1 percentage point o f GDP so that fiscal sustainability is preserved. Table 1.11ImpliedPrimarySurplus(percent of GDP) Interest rate (YO) Real growth rate (YO) 6Yo 8Y O 10% 12% 14% 16% 0.0% 1.o 1.5 2.0 2.6 3.1 3.6 2.0% 0.3 0.8 1.4 1.9 2.4 2.9 4.0% -0.3 0.2 0.7 1.2 1.7 2.2 6.0% -0.8 -0.3 0.2 0.7 1.2 1.6 Source: World Bank staff estimates. 1.43 The base-case simulation is based on the assumption o f a strong policy effort that includes faster and deeper structural reform. This, inturn, implies certain levels o f GDP growth, inflation andinterest rates. Assuming a relatively favorable external environment is reasonable to 22For the theoretical underpinnings see Sargent, Thomas J. and Neil Wallace, 1981, "Some Unpleasant Monetarist Arithmetic" Federal Reserve Bank of Minneapolis Quarterly Review Fall 1981 (Reprinted inRational Expectations and Infation, New York: Harper Collins College Publishers, 1993). 23 For a description o f this methodology see R. Anand and S. van Wijnbergen, "Inflation and the Financing of Government Expenditure: an Introductory Analysis with Application to Turkey", World Bank Economic Review, March 1989. 24O f course, incurring temporarily high deficits and debt levels may be appropriate insome circumstances, and it is unlikely that a country should try and maintain a stable debt-to-GDP ratio at all times. Therefore, this analysis tends to be more useful inguiding policy decision over medium-and long-termhorizons. 21 expect growth o f at least 4 percent over the next 3-5 years. The Government's intention to reduce inflation to single digits and lead Romania to economic and monetary union implies medium- term inflation o f about 2.0 percent per annum, the European Central Bank's target rate. As the economy becomes more integratedwith global capital markets and financial discipline increases, it is reasonable to assume that domestic interest rates and external borrowing costs will decline. Inthis analysis, average financing costs are reduced conservatively to 8 to 10 percent from the current effective nominal rate o f about 14percent. Finally, the public debt to GDP ratio is held at the current level o f about 27 percent, andthe monetary base at around 5.0 percent of GDP. Based on these assumptions we estimate seigniorage revenue o f about 0.3 percent per annum and a debt-stabilizing primary balance o f 0.2 to 0.8 percent o f GDP, suggesting that levels between 0.4 and 1 percent o f GDP seem to be sustainable provided further strengthening o f the adjustment mechanism used to achieve it.. 1.44 The choice o f macroeconomic adjustment mechanism determines a weak or a robust foundation o f fiscal sustainability. Fiscal sustainability analysis presumes fiscal policies can be relied on to balance the government's intertemporal budget and monetary policies can focus on implementing a disinflation strategy. In this context, it i s important to examine the extent to which Romania's adjustment mechanism resembles a monetary or a fiscal dominant regime. 25 In a monetary dominant regime (or Ricardian regime) the primary deficit is the policy adjustment instrument, which is adjusted to variations in the real value o f government debt thus ensuring that fiscal policy is used to balance the government's intertemporal budget without subordinating monetary policy to achieve this balance. In this case monetary policy variables-domestic interest rates, the exchange rate, the domestic price level-are not determined by fiscal considerations but by monetary policy objectives. Altematively in afiscal dominant regime (or Non-Ricardian regime) the level o f the primary surplus i s chosen independently, often due to exogenous political processes, of the level o f expected real interest payments and real net public debt, the balance of the government's intertemporal budget is achieved by subordinating monetary policy to fiscal needs. 1.45 As in other emerging market economies, Romania faces the challenge o f consolidating disinflation and ensuring financial stability while opening up the capital account. Additionally, it must align its policies with the goals o f EU accession and prepare for the eventual adoption o f the euro. This will require Romania to have a macroeconomic policy mix that moderates exchange rate swings and lowers inflation rates to single digits. In particular, monetary policy will need to become functionally independent o f fiscal needs, i.e. efforts are needed to ensure a monetary dominant regime for macroeconomic adjustment, so that the authorities can deliver disinflation securely and provide a stable exchange rate against the euro under an appropriate inflation target.26 1.46 Figure 1.5 summarizes the result o f a test o f fiscal dominance over monetary policy 25 The empirical analysis presented here is based o n the tests o f fiscal dominance for the United States conducted by M. Canzoneri, R. Cumby, and B. Diba in"Is the Price LevelDeterminedby the Needs ofFiscal Solvency" in The American Economic Review, December 2001. The sources o f the quarterly data for Romania are the National Bank and the Ministry of Public Finance. 26L a w 101o f 1998 established the central bank's legal independence. Functional independence refers to the ability to operate effectively irrespective of fiscal policy. 22 covering the period fiom the first quarter o f 1995 to the second quarter o f 2003. It is based on vector autoregression (VAR) analysis, examining the time-series properties o f the quarterly ratios to GDP o f public liabilities and the consolidated general government primary surplus. As shown by the solid line inthe chart, however, even with a tendency for positive increases inthe primary surplus to persist over time there is little evidence that the ratio o f liabilities to GDP tends to decline over the same period. Thus there is little evidence supporting a view that Romania's macroeconomic adjustment mechanism is monetary dominant.27 Furthermore, this finding is relatively insensitive to changes in the sample period, the chronological ordering o f effects between surpluses and liabilities, and the inclusion o f discount factors and nominal GDP inthe VARs to examine whether thebehavior ofliabilities isconditionedbythese variables. Figure 1.5 Response of the Ratio of Public Liabilities to GDP to a One Standard Deviation Increase in the Ratio of the General Government Surplus to GDP Response of SURPGDP to SURPGDP Response of LIABGDP to SURPGDP .06, I .04' .02 ``\\\- -.01 , , , , / I / 1 2 3 4 5 6 7 8 9 10 Source: World Bank staff estimates. 1.47 Though officially a managed floating regime is ineffect, inpractice the NBR implements a crawling peg policy that guides the exchange rate in line with its (pre-announced) inflation target.28Inthe European Monetary Union (EMU), however, exchange rate stability becomes a matter o f common interest and members must avoid rates that are inconsistent with economic fundamentals, excessive fluctuation, and competitive devaluation. In the first two years o f membership, the ERM2 arrangement must be implemented.*' This regime is a essentially a managed floating arrangement that allows a maximum o f 15 percent fluctuation around a central parity with the Euro, a relationship which must be maintained for two consecutive years before the euro can be adopted. The European Central Bank, acting jointly with national banks, will 2'The liabilities comprise public debt and the monetarybase, while the government surplus is defined to include the primary balance and seigniorage. The dotted lines measures 2 standard deviations from the values predicted by the regression model and indicate the level o f statistical significance o f these predictions. The right-hand side o f the regression equations include the first and second lagged values o f the ratios to GDP o f surpluses and liabilities, a constant term, and a dummy variable to capture the seasonality o f GDP. 28Modest real appreciation o f the lei consistent with the Balassa-Samuelson effect o f lei is permitted under this policy. 29A detailed discussion o f the implications o f EUaccession for monetary and exchange rate policies is providedby "Monetary and Exchange Rate Regimes in the Central European Economies on the Road to EU Accession and Monetary Union", inThe Roadto EUAccession, IMF, 2001. 23 intervene ifnecessary to ensure that the 15 percent bandis observed. 1.48 Recently, the NBR has had more success in pursuing the disinflation strategy without major detraction from its principal objective^.^' As indicated in the PEP, the NBR intends to introduce a formal inflation targeting regime in 2005 intransition towards ERM2 provided that durable gains have been achieved in the process o f di~inflation.~~ The aim i s to establish the inflationtarget as a nominal anchor to guide inflationary expectations. But such a strategy would require that monetary policy to focus on the price stability and ignore fiscal objectives at the moment that capital mobility is likely to be increasing. For this, the NBRwill need support from appropriate incomes and fiscal policies, and greater progress in enforcing enterprise financial discipline. Without such support Romania would effectively have no exchange rate policy and the disinflation strategy would not be effective. In effect, the nominal anchor for the economy would be fiscal policy. In addition, it may be premature to adopt inflation targeting unless the macroeconomic adjustment mechanism i s driven by a monetary dominant regime since fiscal dominance would undermine sustainability o f an inflation targeting strategy. 1.49 Insum, while current levels of fiscal deficits seem sustainable provided external current account conditions improve, Romania's macroeconomic adjustment mechanism is not monetary dominant. Further fiscal consolidation is needed to strengthen the ability o f the government to rely on fiscal policies to balance the government's intertemporalbudget and secure independence o f the monetary policies from fiscal financing requirements. Consistency over time o f fiscal discipline would enable Romania to move to inflation targeting which requires that monetary policies not be dominated by fiscal financing requirements. D.3.Nonpayment,Arrears, andBarterSettlement 1.50 The recent reduction in measured fiscal deficits does not reflect the true level o f budgetary commitments and long-term risks to fiscal sustainability. Although Romaniahas made considerable progress in containing fiscal deficits and improving the allocation o f public resources, spending has been controlled in part through nonpayment to suppliers o f goods and services (Table 1.12). 1.5 1 Despite recent improvements in budget formulation and execution, including the reduction in the number o f budgetary and off-budget funds to streamline control over expenditure, total budgetary arrears to the economy increased from 0.8 percent o f GDP at the 30On several occasions during the past decade, severe pressure for accommodating budgetary liquidity shortages and quasi-fiscal deficits could not be withstood, and the result was exchange rate devaluation and high inflation. With the elimination o f subsidies directed sectoral credits and preferential exchange rates in 1997, and the end to central bank financing ofthe state budget, the dominance o f fiscal policy over monetary objectives has decreased. In 1999, the removal o f credit subsidies, however, fueled a liquidity crunch, and ultimately a solvency crisis in the country's two largest banks, Banca Agricola and Bancorex. This led to distressed borrowing and a bailout o f both institutions. The resulting injection o f liquidity had to be sterilized, which resulted in excess monetary contraction and a sharp increase in interest rates (sterilization refers to central bank operations to offset an undesired monetary expansionor contraction). 31A detailed discussion of the preconditions for inflation targeting in Romania is provided by S. Rosentuler, E. Iorga, W. Salater, D Sasu, and A. Codirlasu, "Direct Inflation Targeting: A New Monetary Policy Strategy for Romania", NationalBank of Romania Occasional Paper No. 1, October 2002. 24 end o f 2001 to 1.0 percent o f GDP in 2002.32About 70 percent o f these debts are owed by the State budget, with the local government and Health Insurance Fund accounting for most o f the rest.33 As it tends to weaken the finances o f enterprises and contributes to widespread use of nonpayment, it is clearly not sustainable for the government to continue accumulating payment arrears. Table 1.12 Government Arrears 2001 2002 2001 2002 (billion lei) (% of GDP) State budget 6352 10319 0.5 0.7 Suppliers 6276 10169 0.5 0.7 Other 76 149 0.0 0.0 Local budget 2189 2712 0.2 0.2 Suppliers 2001 2365 0.2 0.2 Other 188 347 0.0 0.0 Social Security Fund 64 16 0.0 0.0 Suppliers 56 6 0.0 0.0 Other 8 10 0.0 0.0 Health Insurance Fund11 902 2563 0.1 0.2 Suppliers 902 2563 0.1 0.2 Other 0 0 0.0 0.0 Total 9507 15609 0.8 1.0 Note: li Health Insurance Fundarrears exclude overdueobligation of healthcareprovidersto their suppliers. Source: World Bank staff estimates. 32 Estimates presented may underestimate government arrears due to information constraints, items and entities excluded from consolidated general government accounts. Intra-governmental arrears, such as contributions to social insurance funds, are negligible insize and cancelled inthe consolidation o f the various government budgets. 33The stocks public debt presented inthis report exclude budgetary arrears. 2. BUILDINGON TRADE INTEGRATION A. INTRODUCTION 2.1 In the last four years, Romania has made good progress in stabilization, growth, and poverty reduction.34 This progress i s due, in large part, to its solid performance in trade integration with the EU and global markets. To extend this performance throughout the economy, three areas o f reform are central: (i)accelerating private sector development; (ii) transforming the agricultural and food sector; and (iii) increasing labor market flexibility. 2.2 Romania's impressive foreign trade performance, particularly so over the last four years, cannot be attributed to a "catch-up" dynamic triggeredby the steep contraction intrade in 1989- 92, when the value o f exports fell almost 60 percent and that o f imports 40 percent (Kaminski 1993). This factor may explain a twofold increase in the value of exports between 1992 and 1995, but it fails to explain the remarkable export performance over the period 2000-02, with an average annual growth rate o f almost 18 percent. The double-digit growth in trade over this period has been particularly impressive, as it has taken place in the face o f a weak import demand in the EU, by far Romania's most important trading partner. Furthermore, it has taken place against the background o f relatively low FDI inflows, which raises questions as to the forces behindthe current export expansion. 2.3 Nor can this trade performance be explained by progress in the implementation o f second-generation' structural reforms. According to international assessments (e.g., EBRD 2003), these reforms have laggedbehind all of the other ten EU candidates. So has FDI, which has contributed to restructuring and increasing the international competitiveness o f many industrial sectors in the CEEC-1035. Indeed, FDI inflows have been responsible for improvements inthe export performance o f successfbl CEEC-10, countries. 2.4 In addition, Romania has the highest applied MFN tariff rates among the CEEC-10. Indeed, Romanian MFN applied tariff rates appear to be a vestige o f "import-substitution thinking." An average supplier from an MFN country faces the highest level o f reverse discrimination in the pan-European market for industrial products in Romania. Since domestic producers o f industrial products face duty-free competition from firms from the EU and other FTA partners, the wisdom o f maintaining highMFNtariff protection seems somewhat dubious. Butthis is notthe point. 2.5 Instead, it leads to questions about the gap between allegedly poor policies and undisputedly good outcomes. Why is there such a good performance with so little FDI and, 34This chapter was prepared byBartlomiej K.Kaminski, University ofMaryland, with the assistanceof FrancisNg, DECRG of the World Bank, and Manuela Unguru, Institute of World Economics, Romania. 35 The CEEC-10 includes: Bulgaria, Czech Republic, Estonia, Hungary, Lithuania, Latvia, Poland, Romania, Slovenia, and Slovak Republic. 26 apparently, such a poor investmentclimate? Why has the combination o f regional liberalization and multilateral import-substitution policy not undercut exports and imports? And, more important, can such an export performance be sustained under these circumstances? And what can be done to sustain foreign trade expansion? 2.6 This chapter seeks to provide answers to these questions. It argues that the investment climate has not discouraged foreign investors from establishing firms (admittedly in low tech areas) and that the MFNtariff policy has prevented excessive trade diversion, albeit at a cost to the quality o f the investment climate. Yet ifthe Romanian export performance is to be sustained, the hassle of conducting business and the MFNapplied tariffrates should be drastically lowered. 2.7 This chapter discusses these issues inmore detail inthe six following sections. Section B reviews developments in Romania's foreign trade over 1993-2002. Section C identifies distinctive features of the current wave o f strong export performance. Insection D, the patterns of Romania's integration into the EU are examined from the point of view of their implications for the sustainability of the country's export performance. Section E traces the distortions and examines the consequences o f high reverse discrimination due to the combination o f intensive regional liberalization and high MFN tariff rates, while Section F examines the consequences adopting of the EU common external tariff for industrial products. Section G provides a conclusion and outlines policy recommendations. B. DYNAMICS FOREIGNTRADE: OF CRITICAL ROLEOFEUMARKETS 2.8 Foreign trade performance has been a bright spot in Romania's otherwise bumpy transition from orthodox central planning. After experiencing a significant contraction in its place inthe world markets through the 1980s, which was further exacerbated by a steep decline intrade after the collapse of the Ceaucescu regime inDecember 1989, Romania's foreign trade reboundedbeginning in 1993 and has become a major driving force behind the country's growth performance. While the GDP grew, on average, at a negative rate o f 0.4 percent over 1990-2001, the volume o f exports registered an annual average rate of growth of 8.5 percent and an annual average growth of imports o f 6.8 percent (WDI 2003). The value o f EU-oriented exports increased fivefoldbetween 1993 and2002, while the value o f total exports almost tripled. 2.9 Since the collapse o f communism, the direction of Romania's foreign trade has changed drastically. While some o f these changes have merely beenthe result o f the contraction inimport demand in former centrally planned economies, others have resulted from successful efforts to reorient trade in line with economic incentives and comparative advantage. Trade with the EU has expanded rapidly, and the EU has rapidly emerged as Romania's dominant trading partner. While there has been a significant reorientation o f trade toward the EU, especially on the side o f exports, this does not appear to be the result o f a trade diversion triggered by the European Association Agreement. A unique feature o f this trading relationship has been the uninterrupted growth inRomania's exports to the EU since 1992, offsetting the initial decline intrade with the R O W (Rest o f the World). As a consequence, the picture that emerges from eyeballing statistics for total trade is a different one from that derived simply from trade with the EU. 27 2.10 A cursory examination ofRomania's foreign trade performance since the implementation o f the first stabilization-cum-transformation program in 1991 suggests the existence of three phases: the phase o f expansion in 1992-95; the stagnation phase in 1996-99; and the expansion phasethat began in2000 (Figure 2.1). Imports followed a similar pattern, although they fell more steeply than exports in 1999, and the value of imports continuedto grow in 1996 while exports stagnated. Services accounted on average for 14 percent of total exports and 15 percent of total imports of goods and services. Both trade and services balances were in the red, with exports paying for around 80 percent o f imports--albeit the ratio improvedin2002, reaching 88 percent, up from 81percent in2001. Figure 2.1 Foreign Trade inGoods and Services, 1992-2002 (in millions of US dollars) 100% I 0 18,000 16,000 80% 14,000 I 12,000 60% 1 - 110,000 40% 8,000 6,000 - 20% 4,000 +- lmportsof ~ 0Yo 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 I L I Source. Derived from data of the National Bank of Romania reported in OECD Economic Assessment. Romania, June 2002. 2.11 The three-phase breakdown of Romanian's foreign trade performance, however, does not reflect the developments in exports to its main preferential partner-the EU. According to the EUdata, except for 1996 (4 percent) and 1999 (6 percent), EU-orientedexports grew at double- digit rates, although there was a deceleration in growth in 1996 and 1997. Simultaneously, exports to other preferential partners (mainly CEFTA) even duringthe 1996-99 stagnationphase, grew at an annual average rate o f 15 percent, which would hardly qualify as stagnant (see Figure 2.2). The combination of weaker growth in EU-oriented exports and falling exports to ROW(quite precipitously over 1998-99) contributed to stagnation intotal exports in 1996-99. On the other hand, the combination of unusually strong export growth to the EU and recovery in ROW-oriented exports has been responsible for the current expansionphase. Indeed, the share o f ROW inRomanianexports, albeit not inimports, has increased slightly. 2.12 Gains in competitiveness rather than preferential free trade agreements have been responsible for export expansion, simply because the preferential margins enjoyed by Romanian firms have been low and not exclusive. Apart from textiles subject to a special import regime in the EU (until 2005), EU MFN applied tariffs on manufactures have been very low, with an average of about 4 percent. This has not provided much "breathing space" vis-&vis MFN suppliers from highly competitive countries such as China, Korea, Japan or the United States. Furthermore, Romanian exporters, including exporters of textiles, have had to compete with a large group of producers from countries facing similar conditions in gaining access to EU markets. These countries include not only European transition economies but also a number of developing countries with which the EU has special preferential arrangements. In consequence, 28 the likelihood that EU importers would shift their purchases from non-preferential suppliers to Romanianfirms (i.e., the trade diversion effect) has been low. Figure 2.2 Exports to the EU,Other Preferential and MFNMarkets (inmillions ofUS dollars) 9,000 8,000 7,000 exports 6,000 5,000 ~-e - EU 4,000 3,000 2,000 MFN 1,000 exports 0 - I 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Source: Derived from the UNCOMTRADEdatabaseas reportedby Romania. c.FEATURESTHE OF CURRENTWAVE OFSTRONG EXPORT PERFORMANCE 2.13 The export performance that Romaniahas experienced since 2000 raising questions about its sources and, ultimately, its sustainability. Considering the relatively slow progress in second- generation reforms and the perceived weaknesses in Romania's investment climate, this has surprised most observers. Evenmore surprising has been the fact that the surge inexports has not been confined to the rather flat EU markets over 2000-02, but that, as has been mentioned, exports to non-preferential partners have grown at a similar pace. C.1. OutstandingDynamics ofEU-Oriented Exports: The Earlier Phase 2.14 The current expansionary phase of Romania's export performance inEUmarkets i s not a statistical fluke, as some may argue (see, for instance Vosganian 2003). Nor have EU-oriented exports been its sole lever. Euro appreciation vis-a-vis the U S dollar was not a statistical factor in the high export performance in 2002 when the US dollar began to depreciate against the euro. Indeed, Romania's share in EU external imports-a measure indifferent to currency fluctuations-increased 33 percent in2002 alone on top o f the 20 percent increase in2001 over 2000. Furthermore, exports to other markets have strongly rebounded, halting the process o f growing reliance on EUmarkets. 2.15 The acceleration in export growth in 2000-02 was outstanding. Romania recorded the largest increase in the share in EU external imports among the CEEC-10, with its share rising from 0.64 percent in 1999 to 1.05 percent in 2002, or 64 percent (Table 2.1). The share o f the Czech Republic, the second best performer among the CEEC-10, increased 43 percent from 1.8 percent to 2.6 percent over the same period. 2.16 This improvement has taken place against a superior past performance that followed the launching o f the stabilization-cum-transformation program inApril 1991. Romania and with the 29 Czech Republic are the only countries among the CEEC-10 that had an EU-oriented export growth that exceeded the growth in EU external import demand each year over 1994-2002. All o f the other countries experienced a year or more o f contraction intheir respective shares of EU external imports. Romania's share in total CEEC-10 EU-oriented exports increased each year, except in 1996 and 1998. Table 2.1 Romania's Performance in EUMarkets against CEEC-936;Change in the Share in EUExternalImports and the Share in CEEC-10EU-Oriented ExDorts. 1993-2002 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Index2002, 1999=100 CEEC-10 100 117 133 134 141 159 162 162 179 212 131 Bulgaria 100 127 152 133 142 144 131 140 156 142 108 Romania 100 136 164 163 177 194 199 205 245 326 164 ShareofRomania in EUextemal 0.32 0.44 0.53 0.53 0.57 0.63 0.64 0.66 0.79 1.05 164 demand (%) ShareofRomaniain EUimports 6.34 7.34 7.67 7.64 7.86 7.65 7.67 7.89 8.48 9.43 123 from CEEC-10 Source: Own calculationson the basis of EUdataas reportedto the UNCOMTRADE. C.2. Levers of Export Growth 2.17 EU-oriented exports o f manufactures were the main driver o f Romanian exports throughout 1993-2002, as is still the case during the current ex.pansion-albeit with three important qualifications. First, manufactures have also become the driving force behindthe surge inROW-orientedexports. The value of ROW-orientedexports of manufactures was 70percent higher in 2002 than in 1999, recording the same increase as exports o f manufactures to EU markets. 2.18 Second, although textiles and clothing (TC) and footwear have continued their strong export performance, other manufactures-mainly electrical machinery-have made significant contributions to the current expansion. These manufacturers have been consistent top performers, with their aggregate share intotal exports doubling from 15 percent in 1993 to almost 30 percent in 2002 (see Figure 2.3). Exports of clothing, accounting now for one-third of EU-oriented exports, and footwear, accounting for around 10 percent o f total EU-destined exports, increased fivefold and sevenfold, respectively, between 1993 and 2002, in terms o f value. However, in marked contrast to 1993-98, the growth in exports o f clothing and footwear was in line with the growth intotal exports. Their aggregate share has not seen muchchange since 1999. 2.19 Third, as an illustration of the progress accomplished in industrial restructuring, other sectors have emergedduring this phase as strong performers. Electrical machinery stands out. Its exports more than tripled in terms o f value between 1999 and 2002. It is worth noting that electrical machinery emerged as a top performer only in 2000, when its share in total exports rose to 8 percent from 4 percent in 1998. The value o f these exports exceeded US$1 billion in 2001 and rose by another 50 percent or US$500 million in 2002, accounting for 10 percent of 36 The CEEC-9 includes: Bulgaria, Czech Republic, Estonia, Hungary, Lithuania, Latvia, Poland, Slovenia, and Slovak Republic. 30 total Romanian exports. Exports o f other manufactures(exc1udingclothing and footwear) after a prolonged period o f stagnation and contraction over 1995-99, recovered strongly increasing to US$5.8 billion up from US$3.6 billion in 1999. In short, there has been a significant increase in the diversity of Romania's export offer o fmanufactured goods. Figure 2.3 Exports over 1993-2002 (in millions of US dollars and inpercent) 35 -Share of electrical 6,000 30 machinery in total 5,000 25 and footwear in total exports 4,000 20 Electrical Machinery 3,000 15 2,000 10 I-Clothing and footwear 1,000 5 I-Other manufactures 0 0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Source: Own calculationsbasedon UNCOMTRADE Statisticsas reportedby Romania. 2.20 The degree o f processing embodied in Romanian exports has also been on the increase. Table 2.2, which shows dynamics o f Romanian exports interms o f end-use product categories, indicates an increase in level o f processing, as captured by changes in the weight of foods and feeds together with industrial raw materials.37 In contrast to the 1993-99 period, the current expansion phase witnessed a falling trend in significance of exports o f traditional production inputs,with their aggregateshare falling to 11percent in2002, down from 16percent in 1999. Table 2.2 Dynamics of Exports by End-use Product Categories, 1993 and 2002 In million of US Index, Index, Composition Share of EUin dollars 1995 2002 in YO Y O 1993 2002 1993=100 1999=100 1993 2002 1993 2002 Food& Feed(0+1+2+4-27-28) 48 1 880 163 99 9.8 6.3 27.4 40.1 Industrial Raw Materials 152 589 180 128 3.1 4.2 48.8 47.6 (27+28+68) Machinery, excl auto (7-78) 615 2,598 140 199 12.6 18.7 27.2 66.1 Automobiles& Parts(78) 220 344 78 275 4.5 2.5 7.2 69.1 Consumer Goods(5+6+8+9- 3,423 9,464 170 165 70.0 68.2 47.8 71.3 68) Non-Oil Goods (0 to 9 less 3) 4,404 12,775 165 158 90.0 92.1 41.6 69.3 Fuels (3) 488 1,101 129 266 10.0 7.9 39.1 43.3 All Goods (0 to 9) 4,892 13,876 162 163 100.0 100.0 41.4 67.3 Source: Own calculationsbasedon UNCOMTRADE Statisticsas reportedby Romania. 37 These are regarded as traditional production inputs, that is, they are not processed intheir present form (Feenstra 1998). Interestingly, they accounted for 43 percent of total exports in 1989, but only 12.8 percent o f total exports in 1993, and subsequently they continued to fall. 31 2.21 Simultaneously, there has been a slight increase in the geographical diversity of these exports, combined with differences inROW-oriented and EU-oriented export baskets. The EU's share in purchases o f electrical machinery fell to 76 percent in 2002, down from 92 percent in 1999. The share o f the EUinRomanian exports o f other manufactures (whose value increased 53 percent in 1999-2002) remained relatively low at 57 percent, (which i s well below the EUshare, o f 67 percent and 73 percent, respectively in total or manufactured exports). The EU has remained the dominant market for clothing and footwear taking up 94 percent o f these exports, while ROW has remained an important consumer o f Romanian traditional production input. Foods and feeds together with industrial raw materials were the only product categories that had a consistently strong growth in ROW-directed exports. Their aggregate share in these exports rose from 15 percent in 1993 to 24 percent in 1996, then slightly contracted in 1997-98,`and rebounded in 1999. The share o f these inputs in EU-orientedexports fell from 7 percent in 1993 to an average o f around 5 percent in 1995-2001 and 4 percent in 2002. ROW markets took aroundthree-fourths o fthese exports until 1997 and60 percent thereafter. 2.22 Inall, the differences insources of export growth between the two expansionary phases are significant indicators o f progress in industrial restructuring. Although the increase in the value o f total exports was the same inboth phases, the end-use products that have contributed to the expansion, as measured by their above-average performance, have been different, except for consumer goods, which have displayed a strong performance in both phases. While during the first phase industrial raw materials and foods and feeds registered rapid growth, machinery and automobiles and parts, together with consumer goods, have been the drivers o f the current expansion. C.3. AccelerationinImportsHaveKeptupwith ExportGrowth 2.23 The surge inexports discussed above has been accompanied by an acceleration inimport growth. Imports o f goods have tended to grow at a similar pace as their total exports through 1994-2002. Export growth outstripped import growth in four years: 1994, 1997, 1999, and 2002. On average, total exports amounted to around 80 percent total imports o f goods in 1993-2002. Since the current phase began with the contraction o f imports (-12) interms o f value and with a slight increase inthe value of total exports (1percent), the import dynamics contain a "catch-up" component exaggerating their growth. With a subsequent expansion in exports, imports rebounded more strongly, registering 4 percentage points and 9 percentage points higher in annual increases in 2000 and 2001 than the increases registered in exports (22 percent and 10percent, respectively). The value o f exports increased 22 percent and the value o f imports 15 percent in 2002. In consequence, import coverage by export earnings improved from 73 percent to 78 percent, although (Table 2.3). 2.24 Romania has been a net importer in almost all product categories, classified by end-use duringboth phases. Duringthe first phase o f trade expansion, exports o f automobiles and parts as well as exports o f consumer goods exceeded their imports, but since 1995 the balance has been reversed for automobiles and parts and for consumer goods in 1998. During the second phase, Romania was a net exporter o f industrial raw materials, although the export surplus was on a decline since 2000, and the increase in the value o f their imports was much lower than during the first phase. This improvement has been the result o f exports o f steel to non-EU 32 markets. Exports o f consumer goods remained very close to the value o f their imports, although the latter grew more rapidly in2001-02. Table 2.3 Developments in Imports and Net Exports during Two Expansionary Phases (in millionsofUS dollarsandin percent) Importsin millionof US Change invalue of imports Exportsas percentof dollars imports Average Average Index, Index, Growt Averag Averag 1995 2002 h e e 1993-95 1999-02 2002 1993=100 1999=100 2002 1993-95 1999-02 2000 2001 2002 Food& Feed 1,007 1,184 1,360 98 149 -1.1% 64 75 76 61 65 (0+1+2+4-27-28) Industrial Raw 288 443 463 158 141 -4.0% 75 136 155 123 127 Matl. (27+28+68) Machinery, excl. 1,737 3,406 4,146 169 165 14.8% 43 56 53 56 63 auto (7-78) Automobiles & 188 564 867 235 385 20.2% 119 42 39 33 40 Parts (78) Consumer Goods 2,896 6,962 9,023 200 168 22.2% 140 97 100 95 93 (5+6+8+9-68) Non-Oil Goods (0 6,I 15 12,560 15,859 167 170 16.8% 94 82 84 79 81 to 9 less 3) Fuels (3) 1,854 1,656 2,002 131 190 1.3% 31 44 47 36 55 All Goods (0 to 9) 7,970 14,216 17,862 158 172 14.9% 80 78 79 73 78 Source: Own calculations based on UNCOMTRADE Statistics as reported by Romania. 2.25 The poor performance of the agricultural sector contrasts sharplywith the industry sector. Despite Romania's favorable climate and soil conditions, the country has remained a net importer o f foods and feeds. While the average for exports as a percent o f imports in 1993-95 was lower than the average in 1999-02, this i s due onlyto the fall inimports in 1999 (21 percent) combined with the strong growth in exports (19 percent) raising the ratio to 97. With imports expanding and exports stagnating, there hasbeen a subsequent increaseinnet imports. C.4. Conclusion 2.26 Romania's export expansion has not involved exporting more of the same products. While many products that were competing in international markets have continued to show export growth, many new competitive products have emerged during the current expansion. A marked shift towards more processed goods and away from low processed industrial raw materials can be seen. Romania's export offer has become more diversified, providingproof o f significant industrial restructuring. Another distinctive feature of the current expansion i s the recovery and strong growth inexports to ROW, which i s mainlydue to the growth o f exports of machinery (excluding automobiles). The growth in ROW-oriented exports was sufficiently strong to halt the trend of growing reliance on EU markets in overall trade. To evaluate the sustainability o f the expansion of exports in Romania, it i s important to study the patterns o f trade integration into the EU. This is taken up inthe next section. 33 D.PATTERNSOFINTEGRATIONINTOTHEEU:IMPLICATIONSFORTHE SUSTAINABILITYOF EXPORT PERFORMANCE 2.27 Leaving aside domestic capacity constraints and barring macroeconomic shocks, the patterns o f trade integration into the EU, accounting for the bulk o f Romanian exports, offer clues as to their sustainability. H o w potentially stable are the established commercial links? Do Romanian firms participate inthe most rapidly expanding components o f international trade? 2.28 In contrast to the first surge in EU-directed exports following the initial stabilization- cum-transformation programs o f the other CEEC-10 countries, the second expansion was preceded by a four or five year period of FDIinflows, running at around 5 percent o f the GDP. Hungary experienced this second export push in EUmarkets in 1996-99, and other CEEC-10 countries, such as Czech Republic, Poland and the Slovak Republic experienced it, a few years later. Radical changes in the composition o f exports and the shift towards capital-intensive and skilled-labor-intensive products have been defining features o f this second expansion wave. In contrast to the first expansion, which was dominated by locally owned firms, foreign owned firms havebeenthe main levers o f export expansion inthe second export surge. 2.29 H o w do these stylized facts apply to Romania? We begin with a discussion o f links between FDI and trade, followed by an analysis o f the factor intensities o f Romanian EU- oriented exports andthe development o f trade inparts, networks andtraditional value chains. D.1. FDIInflowsandTrade 2.30 Inadequate progress in the second-generation reforms can provide an explanation o f the variation in FDI inflows. A number o f empirical studies focusing on transition economies have corroborated this finding. Garibaldi et al. (2002) has shown that the quality o f institutions explains the variation in FDI flows to transition economies. In a similar vein, Broadman et al. (2003) plotting the data on FDIper capita and EBRD's governance and restructuring indices for all Balkan countries also finds a very strong positive association between these two variables. The explanation is that countries with a weaker business climate have been less successful in attracting FDI. Ineffective protection of property rights and weaknesses in contract enforcement discourage foreign investors. 2.31 Romania has not scored well in various international assessments, which sheds light on the progress made in second-generation, structural reforms, the quality o f governance, and corruption. For example, the value of Romania's Corruption Perception Index, as annually assessedby Transparency Intemational, has been since 1996 consistently well below the average for the CEEC-10.38 2.32 Nor has Romania been successful in attracting large FDI inflows, which confirms the findings about the importance of the business climate. On per capita FDI and also in relation to 38 In2003 its value was 2.8 (on a scalebetween0 and lo),or 35 percent below the average for the CEEC-10 o f 4.3. Similar deviations from the average were for such governance quality indicators as government effectiveness and regulatory quality. For more detailed information, see Kaminski and Ng (2004). 34 total FDI, it scores lowest among the CEEC-10, despite a very significant improvement in 1997- 2002 (Table 2.4). Table 2.4 FDI Inflows in Comparative Perspective in 1991-96, 1997-2002 and 2002 (estimate) Average FDI Average FDIper capita Cumulative FDIover (in millionsof US dollars) (inUS dollars) 1990-2002 1991-96 1997-2002 2002 1991-96 1997-2002 2002 Total (mln. per capita 2001 GNP ofus $ inUS$ (percent) Bulgaria 85 782 647 10 79 48 4,927 587 27 Estonia 135 422 296 90 249 197 3,05 1 2,034 55 Czech Rep.c 1,089 6,242 9,886 106 520 1,059 39,227 3,808 54 Hungary 2,156 1,890 908 211 179 106 24,484 2,400 40 Latvia 159 323 349 64 142 140 2,926 1,170 38 Lithuania 56 596 744 15 149 201 3,587 969 30 Poland 2,119 6,127 4,371 55 148 113 46,483 1,204 28 Romania 206 1,323 1,210 9 56 51 9,072 400 20 Slovenia 111 612 1,950 55 282 989 4,O 17 2,009 22 Slovak Rep. 175 1,834 4,260 32 325 1,078 10,322 1,911 42 TotaVave. 6,254 20,152 24,621 59 175 257 148,096 1,406 34 Source: Various issues ofEconomic Surveyfor Europe (UNEconomic Commission for Europe), World Development Indicators 2003 (World Bank, 2003) and IMFBalance-of-Payments database. 2.33 Yet Romania has displayed a consistently successful export performance in EU markets since the collapse o f communism. There has been no other case among the CEEC-10 o f a second export boom without huge FDIinflows into industry and business services. There is no reason to assume that Romanian industries could restructure and be successful inEUmarkets without FDI. Indeed, changes in Romania's export offer and its dynamics appear to have been triggered by foreign firms either directly or indirectly. Among the top 30 fastest-growing four-digit SITC exports to the EU over 1995-2002, capital and transportation equipment products (SITC. 7) accounted for three-fourths. As can be seen below, the share o f capital-intensive products inthe growth o f sunrise exports, (that is, exports of products with strong import demand) in the EU tripled between 1999 and 2002, from 23 percent to 70 percent. This points to the expansion of processing activities inthe automotive industries and in electrical equipment production, which, as the experience of other CEEC-10 countries shows, are mainly conducted by foreign owned firms (Kaminski, Smarzynska, 2001). 2.34 Hence, the question becomes that o f why so little FDIhas produced so much in terms o f export performance. The answer has several components. In the first place, there seems to be a view that Romania's institutions are more business-friendly than is generally assumed.39 In addition, the cost of doing business also does not seem to be higher than that in other CEEC-10 countries (see chapter on govemance). This suggests that the business climate alone would not choke o f f supply response to export opportunities. Nor would it prevent domestic businesses from learning from foreign firms, provided that they interact. 39 Murrell(2003a), for example, argues that these institutions are roughly of the same quality as those in other CEEC-9 countries. To illustrate this point in a different context, Murrell (2003) quotes the extensive range o f publications pointing to the adequate enforcement o f contract and property rights. 35 2.35 FDIinflows have some important characteristics that maymagnifytheir actual impact on the economy. As in many CEEC-10 countries, the bulk o f the FDIhas come from firms located inthe EUand the UnitedStates, andmost ofthis inflow has gone to the industrial sector. There were few, if any, investments that would indicate capital flight or money laundering activity. Services and industries-two sectors crucial to exports-have absorbed 70 percent o f the total capital invested inRomania, with the industrial sector accounting for 54 percent o f the total FDI inward stock. Investors from highly developed countries, from the EU and the United States, accounted for almost three-thirds o f the inward FDIstock as o f April 2003.40 2.36 What seems to set Romania apart i s that the FDI inflows have been dispersed across a large number o f relatively small firms with an average capital o fUS$lOO,OO. The average size o f a firm in terms o f foreign capital invested as well as the number o f varies by sector o f the economy. Foreign firms operating in the industrial sector (around 17,500 o f them or 19 percent o f the total o f 93,016 foreign-owned companies, as o f April 2003 tend to be much larger than firms opening in the other sectors. However, given their concentration in the clothing and footwear industries, their average size in terms o f capital investment does appear particularly low. 2.37 A very large number o f foreign firms, combined with a relatively business-friendly environment, may explain Romania's uniqueness interms o f significant knowledge spillovers to domestic firms, as an econometric study o f CEEC-84' countries (excluding Latvia and Lithuania) has shown (Damijan et al. 2003). Indeed, both the sheer numbers and the small average size o f the Germanand Italian companies are particularly striking inindicating their significant presence in foreign trade related activities. As various publications suggest, 13,000 Italian firms, either fully or partly owned, have probably contributed to making Italy Romania's single most important trading partner, accounting for almost one-fourth o f Romania's total trade turnover.42 Judging by the composition o f this exchange, with clothing and footwear accounting for around 60 percent o f Romanian exports andtextiles and footwear parts accounting for around 40 percent of Romanian imports from Italy, some o f these firms have been heavily involved in organizing processing activities. In2002, 90 percent o f footwear parts and 64 percent o f yarns (two major inputs o f Romanian exports to the EU) imported into Romania from the EU originated in Italy. The presence o f many small Italian firms has probably been responsible for the fact that the textile and leather industries have been characterized by much higher foreign penetration than in the case o f other CEEC-10 countries (Hunya 2002). Furthermore, a relatively large presence o f U.S. and East Asian companies in the downstream sectors may also be a contributing factor to the stronger overall effects o f lower FDI. Javorcik, Saggi and Spatareanu (2003) note a positive association between their presence and the productivity o f Romanian firms in the supplying 40 See NTOR 2003, p. 8. The National Trade Registry (NTOR) provides information about statutory or subscribed capital. Incontrast to balance o f payments data, Trade Registry information includes re-invested profits by foreign companies. The largest investor was the Netherlands, accounting for 18 percent o f the total stock o f US$ 9.4 billion, followed by Germany with 10percent and France and the UnitedStates, both with a share o f above 7 percent. 4' The CEEC-8 includes: Czech Republic, Estonia, Hungary, Lithuania, Latvia, Poland, Slovenia, and Slovak Republic. 42 For very interesting insights on Italian firms inRomania, see Cristescu-Voica (2003). 36 industries. No similar link has been found for European investors operating in the downstream sectors.43 2.38 Last but not least, the share o f FDIintotal capital formation, together with the length o f the period, offers some insights into the relative weight in the economy. The average share o f FDIin Gross Domestic Investment o f around 20 percent in the 1997-2001 period (calculated from data in IMF 2002b) suggests the significant presence o f foreign firms.44 With around one- fifth o f domestic investment camed out by foreign firms, the associated influx o f management skills and technology has already had a beneficial effect on the entire economy. With foreign firms accounting in 2000 for 44 percent o f total Romanian manufacturing exports (Hunya 2002),45 the increase in the FDI stock in industry from around US$3 billion in 2000 to US$5 billion in 2002 has also unavoidably raised the presence o f foreign affiliates in Romania's exports. 2.39 Thus, the answer to the question o f "so little FDI producing such strong trade effects" appears to lie in a large number o f foreign-owned firms operating in relatively low tech areas and, because o f this, having strong ties to domestically owned firms. Under these circumstances, the impact o f FDIhas beenmuch stronger than the values o f FDIinflows might indicate. D.2. Factor Intensities ofEU-oriented Exports: The GapbetweenEndowments andFactor Intensities 2.40 The available statistical data suggest that the educational level (with 82 percent participation insecondary education) is significantly higher thanincountries at a similar level o f economic development (see WDI 2003). This is the case despite the fact that Romania's expenditure on education as a percent o f GDP in2000 o f 2.3 percent was the lowest among the CEEC-10. the hourly labor costs in industry and services in 2000 were also among the lowest. Only in Bulgaria was the hourly rate lower (EC 2003a). Furthermore, with employment in agriculture accounting for 40 percent o f total employment and with its favorable climate and soil, Romania should be a strong performer inagricultural exports. 2.41 Hence, considering Romania's large pool o f low-cost skilled labor and its moderate climatic conditions which favor agriculture, skilled-labor-intensive products, together with natural-resource-intensive products, would be expected to dominate its EU-oriented export basket, at least after some adjustment period. 43 The logic behindthis finding i s straightforward. Asian and US. firms, with strong home links, have to look for suppliers within the Pan-European trade area to meet the rule-of-origin conditions and to not pay duties on imports. Onthe other hand, European firms usually have their supplylinks within the EU. 44According to another estimate, the average FDIinflows as a share o f gross fixed capital formation over 1996-2000 amounted to 15.4 percent (UNCTAD 2002). 45 According to the UN, foreign affiliates accounted for 21 percent o f total Romanian exports. This is rather a small share by CEEC-10 standards. Consider that their share intotal Hungarianexports was 80 percent in 1999, 60 percent in Estonian exports, and 56 percent in Polish exports both in 2000 (UN 2003). It seems, however, that Hunya's estimate is more credible, as it was based on a detailed analysis of database covering all manufacturing enterprises that file an income statement in 1998-2000. He observes: "As o f 2000, Romanianpenetration indicators are similar to those inthe Czech Republic one or two years earlier." (Hunya 2002). 37 2.42 Developments in Romanian EU-oriented exports do not corroborate these expectations. Evenallowing for substitutability between capital-intensive and skilled-labor-intensive products, there are no indications that the closing o f the gap between the endowment in a skilled labor force and the EU-directed export basket has begun (Table 2.5). Romania has retained its specialization in unskilled-labor-intensive products, in which it has a strong and recently growing comparative advantage in EU markets. This has been the only product group with a positive strong net export performance. The value of these exports has consistently been almost twice as high as the value of the imports o f unskilled-labor-intensive products from the EU. While exports o f skilled-labor-intensive products experienced the most rapid growth during the first expansion phase, exports o f unskilled-labor-intensive products dominated the stagnation phase, and those of capital-intensive products registered the largest increase in the third expansionphase followed by unskilled-labor-intensive products. Table 2.5 Factor Intensity of Romania's Trade with the EuropeanUnion, 1993-2002 Factor Intensity Product 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 est. Romania's Exports to EU: (%million) NaturalResource Based 338 641 986 837 985 1,004 1,154 1,278 1,428 1,330 Unskilled Labor 1,295 1,670 2,194 2,476 2,824 3,352 3,630 3,887 4,762 6,497 Capital Intensive 200 366 578 644 650 767 842 1,307 1,429 1,534 SkilledLabor 214 406 686 680 685 763 611 653 807 981 All Above Products 2,045 3,082 4,443 4,637 5,143 5,885 6,238 7,125 8,425 10,343 Composition of Romania's Exports to EU: (YO) Natural Resource Based 17 21 22 18 19 17 19 18 17 13 Unskilled Labor 63 54 49 53 55 57 58 55 57 63 Capital Intensive 10 12 13 14 13 13 14 18 17 15 SkilledLabor 10 13 15 15 13 13 10 9 10 9 Romania's Export Specialization Index inEU Natural Resource Based 0.51 0.65 0.70 0.57 0.62 0.64 0.71 0.61 0.58 0.50 Unskilled Labor 3.71 3.31 3.18 3.40 3.44 3.57 3.71 3.81 3.79 3.94 Capital Intensive 0.29 0.34 0.36 0.38 0.34 0.33 0.33 0.46 0.43 0.37 Slulled Labor 0.63 0.79 0.93 0.90 0.81 0.73 0.55 0.57 0.57 0.53 Share inEU'sExternal Imports: (YO) Natural Resource Based 0.16 0.28 0.37 0.31 0.36 0.41 0.47 0.42 0.48 0.55 Unskilled Labor 1.19 1.45 1.70 1.82 2.02 2.28 2.44 2.58 3.10 4.34 Capital Intensive 0.09 0.15 0.19 0.20 0.20 0.21 0.22 0.3 1 0.35 0.41 Skilled Labor 0.20 0.35 0.50 0.48 0.48 0.47 0.36 0.38 0.47 0.58 All Above Products 0.32 0.44 0.53 0.54 0.59 0.64 0.66 0.68 0.82 1.10 Romania's Net Exports to EU: (%million) NaturalResource Based -362 61 7 -311 -59 -103 96 71 -90 -457 UnskilledLabor 654 826 943 1,076 1,215 1,425 1,657 1,717 2,238 3,191 Capital Intensive -745 -843 -1,118 -1,415 -1,376 -1,742 -1,587 -1,792 -1,777 -2,267 Skilled Labor -283 -66 -77 -211 -147 -455 -423 -671 -1,037 -1,515 All Above Products -735 -23 -245 -860 -367 -875 -257 -675 -665 -1.048 Source: Basedon EUas reportedfrom UNCOMTRADE Statistics. 38 2.43 Furthermore, Romania's export profile has not moved toward natural resource intensive products.46The share o f these exports inEU-oriented exports fell significantly during the current expansion phase, mainly because o f the poor export performance o f the agricultural sector. The value o f total agricultural exports (USS880 million) in 2002 was lower than at its peak level o f almost US$1 billion in 1996 as well as its level in 1999 (US$ 885 million). This corroborates one o f the findings in Chapter 4 o f this study, to the effect that the task o f creating an agricultural sector capable o f producinginternationally competitive products has yet to be addressed. 2.44 Inall, the Romanian EU-oriented export basket still appears to diverge from Romania's relative endowments in production factors in the following two dimensions-the under- representation o f skilled labor products and the lack o f natural resource intensive agricultural products. This divergence, which has existed over a relatively long time, is an indication that government policies have prevented the emergence o f competitive markets to allocate resources to sectors with a potential comparative advantage. The agricultural sector has not succeeded in exploiting its favorable climatic conditions. As is argued in Chapter 4 o f this report government policies favored large-scale farming and prevented the emergence of an internationally competitive agricultural sector. Furthermore, the skilled labor force has participated only marginally participated in EU-oriented export expansion. Assuming that skilled labor was significantly under-represented in Romanian exports in 1993, the "endowment" gap was not closed. This sets Romania apart from many CEEC-9 economies, where this gap was closed mainlythrough FDI.This does not appear to have happenedas yet inRomania. D.3. Exports to EUSunriseMarkets: I s the Change inFactor IntensitiesUnder Way? 2.45 Sunrise exports, namely, exports o f products with strong import demand growth in the EU, suggest a possible change in factor intensities in the near future.47 Telecommunications equipment (SITC. 7249) and ships andboats (7353) have been star performers. The value o f EU- destined exports rose from US$1 million in 1995 to US$34 million in 1999 and US$336 million in 2000. Subsequently, these exports fell to an estimated US$166 million in 2002. Ships and boats saw the value o f their EU exports increasing from US$14 million to US$lOO million in 1999 andUSSIOS millionin2002. 2.46 The factor content o f sunrise exports has displayed large volatility over 1995-2002, although with a distinct pattern. The share o f natural-resource-intensive products as well as that o f skilled-labor-intensive products contracted over this period, while the share o f capital intensive products increased. Over the last three years (i.e., during the current expansionary phase) the share o f capital intensive products increased dramatically from 23 percent in 1999 to 70 percent in 2000, thanks to exports o f telecommunications equipment (SITC 7249), and then fell to an estimated 53 percent in2002. 46 Natural resource-based products consist mainly agricultural products and materials and accounts for around 85 percent (agricultural materials-10 percent) of these exports to the EU. After a strong growth during the first phase, their growth was flat over 1996-2002 at an average annual rate of 5 percent. 47Sunrise exports are those that meet the following two criteria: (i) growth inEU-external imports exceeded 10 the percent in 1999-2001 and the value of EUimports exceeded US$50 million; (ii) the value o f the Romanian exports concerned exceeded US$lO,OOO in 1999, with average annual growth rates exceeding 3 percent in 1999-2002. 39 2.47 Romania's performance in the EU dynamic markets, despite the emerging specialization inmore processed, capital-intensive products, does not offer empirical support to the claim that the current export expansion, barring macroeconomic shocks, may be sustainable. D.4 ParticipationinInternationalNetworksof ProductionandDistribution 2.48 Falling transportation and communication costs have created opportunities for outsourcing just-in-time production and supply chain management, has been altering the competitive landscape o f many countries by relocating business activities and providing a new source o f entry into international markets. Value chains have become increasing split with the individual production stages being moved to countries with corresponding comparative advantages. This situation has had far-reaching implications for the global division o f labor and has led to the fragmentation o f the production process within vertically integrated manufacturing industries. Participation in the supply chains o f international networks o f production and distribution brings outside managerial and technological expertise to a local company, and offers a "cheap way" o f marketing products internationally, as the firms do not incur marketing cost which are usually quite significant for newcomers. 0.4.1. Trade in Parts: Impressive Expansion 2.49 Trade inparts, the result o f the fragmentation of the production process thanks to falling costs for service links activities, has beenthe most rapidly expanding component o f international trade. It has been the main lever o f trade expansion in the East Asian economies over the last decade (Ngand Yeats 2003). 2.50 Following Ng and Yeats (1999), we identify 60 SITC. Rev. 2 items capturing trade in parts. Two observations can be derived from data reported in Table 2.6. First, parts have contributed significantly to Romania's export expansion. While the share o f imports o f parts in total imports o f manufactured goods (excluding chemicals) has remained relatively stable, the share o f parts inexports more than doubled between 1998 and 2000 andkept up with the pace o f rapidly expanding overall exports. 2.51 Second, the expansion inexports o f parts was not limited to the EU, although since 2000 the EU has been taking more than two-thirds o f these exports. The share o f parts in exports to markets than the EUincreased from 11percent in 1999to 16 percent in2002. 2.52 Several parts and components have emerged as significant exports, which suggests a wide export specialization. In 1993 only 4 out o f 60 items had comparative advantage (Le., ESI above unity) inEUmarkets. Their number increasedto 15 in 1996 and to 17 in2002. 2.53 In sum, the above analysis indicates the growing specialization and competitiveness in EUmarkets of Romanian exporters o f parts not covered by "network" trade. It also suggests a significant industrial restructuring that has not been captured by the network analysis below. 40 Table 2.6 Trade in Parts and Shareof EU(in millionsof US dollars andin percent) World 1993 1996 1997 1998 1999 2000 2001 2002 Total exports of parts 182 298 382 355 495 868 1,008 1,238 Total imports of parts 401 833 819 1,034 845 1,137 1,305 1,427 Exportsof partsas percentageofimportsofparts 45% 36% 47% 34% 59% 76% 77% 87% Shareofparts inmanufacturedexports(chemicals 5.5% 5.5% 6.4% 5.7% 7.9% 1.8% 11.7% 11.7% excl.) Shareof parts inmanufacturesimports 14.0% 13.3% 12.4% 13.8% 12.2% 3.2% 12.8% 11.7% Shareof the EUin Total exports ofparts 15% 55% 45% 61% 66% 76% 77% 68% Total importso fparts 63% 51% 61% 58% 64% 67% 61% 62% Memorandum: Share ofparts inEU-oriented 1.8% 4.3% 4.3% 4.7% 6.9% 11.7% 11.5% 10.4% manufacturedexports (chemicals excl.) Share of parts inROW-orientedmanufactured 8.7% 8.3% 10.9% 8.3% 11.1% 12.2% 12.5% 15.6% exports excludingchemicals Source:UNCOMTFLA.DE Statistics. 0.4.2. "Network" Trade: Not Yet aLever of Trade Growth 2.54 Have Romanian firms become part of this new division o f labor based on production fragmentation and sharing? This section seeks to provide an answer by examining the participation o f Romanian producers in global production networks operating in EU markets. The focus here is on developments in trade in three major networks-automotive, electronics (referred to as the Information Revolution [IR] network), and furniture (Table 2.7). The automotive and the electronics networks usually involve firms with foreign participation, either as joint ventures or subsidiaries, although there are instances of outsourcing as well. The furniture network operates mainly through marketing organizations that provide suppliers with precise specifications concerning the final product and the parts and components to be used. In contrast to the automotive and IR networks, the fimiture network is less capital-intensive and muchmore unskilled-labor-intensive, although this does not necessarily apply to many parts and components used inthe assembly o f the final product. For these reasons, furniture has been the first network to emerge ina number o f transition economies (Kaminski andNg 2001). 2.55 Romania fits the pattern for the predominance o f the firniture network. In 1993 the furniture network was the only one to generate significant export earnings. It accounted for 95 percent o f all exports by networks in 1993-94, and with 65 percent share in2002 continues to dominate trade inEU-centeredproduction anddistribution networks. 2.56 An examination o f the data tabulated in Table 2.7 leads to the following observations. First, network trade has yet to become the main lever o f Romania's trade, as has been the case with its integration into the EU production centers o f the Czech Republic, Estonia, Hungary, Slovakia and, to a lesser extent, Poland. The closer location to EUmarkets o f these countries and their much higher FDI inflows, explain their success in this area. Except for furniture (both final products and parts), the shares o f these countries in EU extemal imports were well below the average presence o f Romanian products in EU markets, and the values o f ESI were well below unity. Table 2.7 shows that the aggregate share o f network exports was falling until 1999; subsequently, it increased from 8 percent to 11percent in 2000, and to 10 percent in 2001, and then fell back to 8 percent in2002. 41 Table 2.7 Developmentsin Trade with EUin Three Major Networks, 1993-2002 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002(est) Exportvalues to EU($ '000) Automotive network Final 9,603 11,162 15,854 16,722 12,363 10,417 7,393 5,147 4,656 3,862 Parts and components 4,763 6,343 29,703 52,389 44,261 52,505 59,401 41,559 92,906 134,914 Total 14,366 17,505 45,557 69,111 56,624 62,922 66,793 46,706 97,562 138,776 Share in network Exports 4% 5% 10% 15% 13% 13% 13% 6% 12% 16% Exportsas YOof imports Final 7% 9% 11% 11% 7% 3% 3% 1% 1% 1% Partsand components 18% 22% 63% 82% 78% 85% 112% 45% 74% 61% As % of EUexternal imports Final motor vehicles 0.04 0.05 0.06 0.06 0.04 0.03 0.02 0.01 0.01 0.01 Parts and components 0.04 0.04 0.16 0.27 0.22 0.23 0.24 0.16 0.33 0.48 Specializationindices Final motor vehicles 0.14 0.11 0.11 0.11 0.07 0.04 0.03 0.02 0.02 0.01 Parts and components 0.12 0.10 0.30 0.52 0.38 0.37 0.37 0.24 0.42 0.45 Export valuesto EU ($ '000) Information Revolutionnetwork Final 5,408 4,769 4,105 4,379 2,852 4,119 3,738 80,066 50,362 46,070 Partsand components 1,215 871 1,797 2,553 2,586 18,184 31,600 263,035 251,028 129,403 Total 6,623 5,640 5,902 6,932 5,438 22,304 35,338 343,101 301,390 175,474 Exportsas YOof imports Final 39% 32% 14% 13% 3% 2% 3% 55% 37% 30% Partsand components 2% 1% 1% 2% 1% 8% 11% 41% 66% 54% Share in Network 2% 2% 1% 1% 1% 5% 7% 44% 37% 20% Exports As YOof EUexternalimports Final .' 0.00 0.03 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.01 Parts and components 0.00 0.00 0.00 0.01 0.02 0.02 0.21 0.19 0.21 0.(13 Specializationindices Final 0.13 0.07 0.05 0.06 0.04 0.04 0.03 0.55 0.28 0.18 Parts 0.02 0.01 0.01 0.01 0.01 0.06 0.10 0.59 0.59 0.29 Exportvaluesto EU($ '000) Furniturenetwork Final fumiture 290,579 324,043 377,859 364,693 339,890 342,225 342,754 349,169 368,705 490,934 Partsand components 11,079 12,409 17,065 26,017 35,790 44,354 50,274 47,950 55,091 80,407 Total 301,657 336,452 394,924 390,710 375,680 386,579 393,028 397,119 423,796 571,341 Exportsas YOof imports Final 2,547% 1,571% 1,040% 910% 969% 946% 1,259% 1,048% 888% 974% Partsandcomponents 344% 242% 169% 244% 394% 439% 344% 338% 329% 428% Share inNetwork 93% 94% 88% 84% 86% 82% 79% 50% 52% 65% Exports As Yoof EUexternalimports Final fumiture 5.27 5.25 5.10 4.61 4.35 3.94 3.64 3.64 3.75 4.75 Partsand components 0.62 0.57 0.60 0.82 1.08 1.13 1.16 1.08 1.11 1.42 Specializationindices Final fumiture 16.32 11.92 9.64 8.74 7.60 6.28 5.67 5.51 4.74 4.52 Partsandcomponents 1.91 1.30 1.14 1.56 1.90 1.80 1.81 1.63 1.40 1.35 Memorandum: Total network exports 322,647 359,597 446,383 466,753 437,742 471,805 495,159 786,926 822,748 885,590 ShareinexportstoEU 15% 11% 10% 10% 8% 8% 8% 11% 10% 8% Source: Own calculations basedon EUdata as reported to the UNCOMTRADE database. 2.57 Second, inall networks there are clear signs o f two-way trade, with growth inthe exports o f parts and components outstripping that o f final products. While (except for the fumiture network) Romania i s a net importer of both final products and parts and components, significant increases in exports have accompanied imports. These are usually lucrative activities, providing 42 stability for commercial relations and creating opportunities to supply more than one single producer o f a final product. 0.4.3. Participation in Traditional Value Chains: Textiles/Clothing and Footwear 2.58 Clothing and, to a lesser extent, footwear have been the quintessential engines o f growth for many CEEC-10 countries during the initial stages o f transition. They have accounted for a significant share o f value added and manufacturing employment, with significant implications for poverty reduction. With labor costs going up, many o f the outward processing operations in the clothing sector moved to other countries in Central and Eastern Europe through the 1990s. However, these products still dominate the exports o f some transition economies, including Romania. Textiles and clothing together with footwear play a very large role inRomania's EU- oriented exports, accounting for almost half o f the exports to the EU over the last decade (47percent in 2002) and one-third o f the total exports. Both sectors appear to be firmly entrenched inEUclothing and footwear value chains. 2.59 An important question is the extent to which Romania remains solely an assembly shop for EUfirms that are taking advantage o f an available cheap-often female-labor force. Or has the importance o f such inputs for exports as footwear pxts and textiles declined over time, indicating progress in the development o f backward linkages? Have Romanian clothing producers moved from simple cut-make-trim operations where buyers supply fabrics, to FOB operations with the clothing firm responsible for obtaining the fabrics? Without a survey o f firms inthe sectors, or access to input-output tables, itis impossibleto give anunambiguous answer to these queries. However, an examination o f trade data offers some clues. The results are presented inTable 2.8. 2.60 Overall, the importance o f imported inputs for final exports declined during the current expansion phase, which appears to indicate the development o f backward linkages and the increased sophistication o f domestic producers o f footwear parts and textiles. This may also point to the declining weight o f simple cut-make-trim operations in the garment sector. The common feature o f developments in both sectors has been that the expansion in exports has not outstripped the growth inimport o f inputs inthe 1993-2002 period. Infact, the import o f parts as a percent o f exports o f final products andparts-a crude measure o f the importance o f imported materials for final product exports-has been declining since 1999 for textiles/clothing and since 1998 for footwear.48 This measure i s significantly higher for clothing than for footwear. 2.61 Another indication o f a possible increase in local outsourcing and o f further insertion o f firms operating inRomania into the global footwear and clothing markets is the expansion inthe exports o f parts. Although their shares in the total exports o f the respective sectors declined, the value o f these exports has significantly increased. Exports o f textiles in 2002 increased 87 percent over 1999, and the value o f exports o f footwear parts was 62 percent higher over the 48 Similar trends are discernible in terms o f imports o f parts as a percent o f final exports. T h s index fell from 40 percent in 1996-97 to 30 percent in2002, displaying a consistent downward trend. The index for textilelclothing declined from the peak level o f 72 percent. 43 same period. Almost all textiles andthree-fourths o f footwear parts have been shipped to the EU, apparently for further processing. Table 2.8 Trade in Textiles/Clothing (TC) and Footwear and Parts (FP) in 1993-2004 (in millions of US dollars and in percent) Textiles and Clothing 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Exports o f textiles and clothing 770 1,134 1,541 1,729 1,944 2,173 2,212 2,526 3,023 3,563 ~ x & ~ r oftextiles as % o f exports o f t s 17% 12% 12% 10% 10% 9% 8% 8% 8% 9% textiles and clothing Total imports o f textiles 481 625 939 1,055 1,258 1,475 1,577 1,720 2,016 2,374 Imports o f textiles as YOo f exports o f 62% 55% 61% 61% 65% 68% 71% 68% 67% 67% textiles and clothing Memorandum: share o f textiles and clothing in Total exports 15.7% 18.4% 19.5% 21.4% 23.1% 26.2% 26.0% 24.4% 26.5% 25.7% EU-oriented exports 31.5% 33.4% 31.7% 33.8% 36.5% 36.9% 36.5% 34.8% 35.7% 35.1% Footwear 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Exports o f parts and footwear 155 307 423 500 540 603 678 785 976 1,158 Exports o f parts as % o f exports o f 48% 68% 67% 76% 66% 62% 58% 55% 52% 53% footwear Total imports o f parts 25 45 78 103 125 144 154 178 210 231 Imports of parts as % o f exports o f parts 16% 15% 18% 21% 23% 24% 23% 23% 22% 20% and footwear Memorandum: share o f footwear parts and footwear in Total exoorts 3.2% 5.0% 5.3% 6.2% 6.4% 7.3% 8.0% 7.6% 8.6% 8.3% EU-oriented exports 6.2% 9.4% 9.2% 10.2% 10.9% 11.0% 11.7% 11.6% 12.3% 12.0% Source: Own calculations based on UNCOMTRADE Statistics as reDortedbv Romania. 2.62 Italian firms have been mainly responsible for organizing the insertion o f Romanian firms into global networks o f production and distribution o f footwear and clothing. This is particularly visible in the case o f footwear, with Italy taking around 90 percent o f Romanian exports o f footwear parts. The same share for textiles is around one-fifth o f their total exports. 2.63 The large share o f Italy in Romanian trade in clothing and footwear and their inputs provides extra credence to our earlier observation on the role played by Italian-owned firms in organizing local networks of production. Not only do these firms provide the necessary links with Italianretailers and producers, but they are themselves an integralpart o fthe global garment and textile value chain. The implication is that these are not footloose investors. They have a stake inmaintaining competitiveness in foreign markets in operations conducted from Romania, which bodes well for the survival o f the Romanian textile sector following the dismantling o f the EUquotaregime on clothing imports in2005 (see below). 2.64 In all, these two sectors are not in fact enclaves but appear to be immersed in the domestic economy. Both sectors have been very competitive in EU markets, which indicates significant progress inindustrial restructuring. They have become part of the global value chains, organized mainlyby Italian firms. 0.4.4. Concluding Comment 2.65 The picture that emerges from this analysis can be summarized as follows. Incomparison to most other CEEC-9 countries, network trade accounts for a relatively small portion o f Romanian exports. Except for furniture, trade in the networks appears to be moving from one 44 transaction to another. However, the expansion inexports o f parts and components may indicate that many Romanian firms have established a stable position in EU supply chains within the three networks. 2.66 The rapid expansionintrade inparts andthe successfd insertion o f Romanianproducers' global value chains o f clothing and footwear appear to indicate that (i) the current expansion is based on a solid foundation, and (ii)there is an ongoing shift toward finding new niches in external markets. Both o f these indications may bode well for Romania's capacity to withstand the shock ofthe change inEUimport regime for textiles and clothing. D.5. The Challengeofthe ImpedingChangein the TradeRegimefor Textiles andClothing 2.67 The dismantling, on January 1, 2005, o f EU quantitative restrictions on textiles and clothing imports under the WTO Agreement on Textiles and Clothing (TC) will bring about a significant change inthe regime governing the international trade inthese products. Trade inTC, at present subject to quotas restricting competition, will become like trade in other industrial products. With the dismantling o f quotas, MFN tariff rates will be the main border measure protecting domestic and preferential suppliers. The average EU MFNtariff rate o f 9 percent on TC does not offer sufficient preferential margins that might protect Romanian procedures, or, for that matter, EU producers and other preferential suppliers, from the possible surge in exports from the world's most cost-efficient suppliers. 2.68 The challenge facing Romanian suppliers is that they will have to compete with the most efficient suppliers on an equal footing. China in particular is regarded as the chief competitor, with its huge production capacities and its pool o f a cheap labor force. Its performance in EU markets has confirmed these worries. The EU has progressively eliminated quotas on some product categories since China's accession to the WTO in 2001. In all o f those liberalized product categories, unit prices have fallen and China's share has exploded (EC 2003b). The adjustments created by Chinese exports for EUdomestic and external suppliers are not yet over, as almost halfo f the Chinese TC exports are still subject to quotas. 2.69 Romania has assets vis-a-vis many TC exporters from nowEuropean developing countries. First, geographical proximity may give Romanian firms a comparative advantage vis- a-vis Asian or African competitors in some production lines. Clothing is a buyer-driven sector in which the ability to respond quickly to new fashion trends is crucial to a firm's survival inglobal markets. So, also, i s the ability to quickly supply an international retailer. Thus, geographical proximity is an important factor. Second, because o f the large presence o f foreign-owned firms with unique commercial links to EU producers and distributors, Romania may be better positioned than many other countries to withstand the augmented competitive pressures. Italian or German firms operating inRomania are probably less inclined to choose the exit option inthe face o f increasing pressures. Instead, they may look for innovative ways to respond to new challenges. 2.70 Yet the challenge to the sustainability o f the export performance o f clothing in EU markets looms large on the horizon. In view o f the importance o f this sector to the national 45 welfare, a major study is needed o f the TC sector in Romania, and o f ways o f designing measures that will improve the domestic business climate.49 D.6.Implicationsfor Sustainability 2.71 The evidence fi-om the developments in Romanian exports suggests that while a 20 percent per year expansion in exports may not be easy to maintain, it appears to be based on healthy fundamentals. The recent export expansion was not the result o f a temporary explosion in import demand for a single commodity unique to Romania.. Nor was it solely, in its EU- dimension, restricted to the most recent phase. Onthe contrary, it encompassed a larger basket o f manufactures and included markets other thanthose o f the EU. 2.72 Second, the recent expansion is all the more impressive inthat it has occurred against the background o f falling import demand inthe EU. The implication is that Romanian exporters are not marginal suppliers-who are the first to be penalizedby the contraction duringa recession- but theyhave firmly established commercial links to EUmarkets. 2.73 Third, expansion and diversification ingeographical patterns o f trade have accompanied the diversification in export offer and the progress in integrating Romanian firms into EU-based networks o f production and distribution. Clothing and footwear are no longer the sole levers o f Romanian exports to the EU.Other products have also been on the rise. 2.74 Fourth, the developments in the Romanian trade in textiles and clothing and footwear parts and footwear suggest that these two most important export-oriented sectors ofthe economy, which generate almost half o f the EU-oriented exports, are not an enclave but have developed backward links into the domestic economy. 2.75 Whether exports will be sustainable depends to a large extent on developments in the relationship between labor costs and productivity, the emergence o f production facilities using the available skilled labor, the capacity o f Romanian firms to withstand new competitive pressures caused by the dismantling o f MFA quotas, and the restructuring o f the agricultural sector. Sustainability depends critically on the relation between labor roductivity and wage growth in the low-skilled labor-intensive sectors o f the economy?' Among the factors responsible for the improved competitiveness o f Romanian firms in EU markets have been the progress in achieving macrostability and the stability o f the real effective exchange rate. Although the latter has appreciated slightly in CPI terms, it has been stable in terms o f U L C (unitlabor cost) (IMF2002a). Mostimportant isthe fact that productivity gains "offset realwage increases.'fs1 49The EuropeanCommission has launched a major study to examine indetail the impact o f quota elimination inthe EU.The final report will be available in2004. While the IMF (2002a) seems quite optimistic about the sustainability of exports, pointing not only to increases in labor productivity but also to the strengthening o f the profitability of Romanian domestic producers, the OECD country report (OECD 2003) argues that the business environment discourages both domestic and foreign investment. 51Ibidem, p. 10. 46 2.76 While developments in the factor intensity o f Romania's trade suggest that the cost o f labor relative to capital is not too high, this cannot be taken for granted. Increases inwage rates that exceed growth in labor productivity may erode the competitive advantage o f a significant portion o f exports. The opening up o f EU markets to the most efficient TC producers in 2005 will present an extra challenge. Cheap labor, combined with geographical proximity, has been critical to Romania's success. However, the gap betweenthe endowment o f a highly skilled labor force and an abundance o f resources favoring agricultural production and the factor intensity o f the EU-oriented export basket has not been closed. 2.77 Another unexplored area o f opportunity is agriculture. Romania's moderate climate and availability o f land point to the potential for specialization in agricultural production, while the availability o f an inexpensive skilled labor force suggests a specialization in skilled-labor- intensive products. On both counts, the export performance has not bome out these expectations: agricultural net exports have remained stagnant and there has been no perceptible shift toward skilled-labor-intensive exports. 2.78 Although export expansion has been a continuing feature o f Romania's transition from central planning (despite the uneven progress in the implementation o f second-generation reforms), the divergence between the country's export basket and its relative endowments could either smother or stimulate growth in the future. The outcome depends on government policies. Such a divergence over a relatively longperiod o f time i s an indication that government policies have prevented the emergence o f competitive markets to allocate resources to industrial sectors with a potential comparative advantage. Nor has the agricultural sector succeeded in exploiting its favorable climatic conditions, as government policies favoring large-scale farming appear to have prevented this. In a similar vein, skilled labor has yet to become a factor in Romania's export performance, as the FDIcrucial to closing the gap has failed to take advantage o f the low cost skilled labor available in Romania. Once the right policy environment is in place, it might contribute to another wave o f export expansion inthe near future. 2.79 Paradoxically, this combination o f unfulfilled potential in agriculture and industrial restructuring may offer unique opportunities that would provide hrther impetus to foreign trade expansion. However, this would require the implementation o f measures that would foster agricultural restructuring, removing the policy and institutional barriers to the allocation o f capital to the most competitive activities. These measures would attract FDIinflows, which are discussed in more detail below. As the experience o f CEEC economies demonstrates, FDI in activities that close the gap between the endowment o f skilled labor and the factor intensity of production has driven export expansioninCEECs that undertook earlier structural reforms. E.NETWORKPREFERENTIALTRADE OF AGREEMENTS: POTENTIAL FORDISTORTIONS 2.80 Accession to the EU, which has shaped Romania's trade policy and institutions, has been mainly responsible for the very rapid liberalization in access to Romanian markets, with most imports (86 percent in 2002) entering these markets on a duty fiee basis. On February 1, 1993, Romania signed an Association Agreement with the EU and opened EU accession negotiations inDecember 1999.The EU"associate status" has entailed bilateral trade liberalization not only vis-a-vis the EUbut also in trade with other countries enjoying preferential status with the EU. Romania has become a member o f CEFTA and an associate member o f EFTA; and it has had 47 FTA with Israel and Turkey. More recently, under commitments made inthe "Balkan" Stability Pact, Romania has concluded negotiations and signed bilateral FTAs with Albania, Bosnia and Herzegovina, the FYR o f Macedonia, and Serbia and Montenegro. All industrial products originating in Romania's Pan-European preferential partners,52 i.e., 28 countries including the EU, the EFTA, the CEEC-9, and Turkey that are parties to the Pan-European Cumulation Agreement53 are not subject to duties as o f January 1, 2003. Agricultural products originating in these countries are also subject to preferential treatment, although not as deep and extensive as in the case o f industrial products. In addition, all products, agricultural and industrial alike, produced in Moldova have duty-free to Romanian markets. These countries together supply Romania with almost 80 percent o f its industrial imports and 84 percent o f manufactured imports, which otherwise(given the cascading Romanian MFN tariff schedule, and the specialization in highly processed goods as revealed in their Romanian-oriented exports)would be subject to very steep MFNtariff rates. 2.81 The EU accession-driven liberalization policy has resulted in duty free access to Romanian markets for industrial and most agricultural products for producers from 15 EU member countries, three EFTA countries (Iceland, Switzerland, and Norway), Israel and Turkey, and six CEFTA countries.54 Furthermore, commitments made in the "Balkan" Stability Pact Memorandum o f Understanding on Trade Liberalization and Facilitation signed in June 2001 have led to the negotiations (and completion) o f bilateral FTAs with Albania, Bosnia and Herzegovina, FY Republic o f Macedonia and Serbia andMonteneg1-0.~~ 2.82 With the EU enlargement in 2004, Romanian exporters and importers will not face a completely different regional trading environment. To be sure, CEFTA will lose five o f its eight members: Romania will remain in CEFTA with Bulgaria and Croatia. It will gain duty free access for its industrial exports and improved "double-zero" access for its agricultural products to markets inCyprus, Estonia, Latvia, Lithuania and Malta. 2.83 Multilateral liberalization within the WTO framework has not been accompanied by vigorously pursued regional liberalization, however. While Romania has participated in WTO agreements, extending duty free treatment to selected products (e.g., Information Technology Agreement) or countries (Generalized System o f Preferences for the least developed countries), it has kept relatively high MFN tariff rates, especially on agricultural products. "Developing ~ 52The CEEC-9 includes Bulgaria, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, the Slovak Republic, and Slovenia. The CEEC-10 will denote the inclusiono f Romania. 53The Pan-European Cumulation Agreement, which went into effect on January 1, 1997, had set the stage for the formation o f a single Europeantrading bloc. Inaddition to extending the rules o f origin to all Europeanassociates o f the EU (i.e., the CEEC-IO) it has also standardized trade components o f European Association Agreements as well as newly signed agreements interms o f the date o f removal o f tariffs and non-tariff barriers by the end o f 2001. In consequence, a single pan-European trading zone was turned in2003 into a free trade zone for industrial products encompassing 29 countries. 54 Except for Croatia with which duty reductions are subject to a separate schedule; duties on 100 percent o f industrial products from other CEFTA countries were removed as o f January 1, 2003. 55Romania has been part of the Trade Initiative o f the Stability Pact, which over the last three years has focused on providing support to the seven Southeastern European countries-Albania, Bosnia and Herzegovina, Bulgaria, Croatia, the FYR o fMacedonia, Romania, and Serbia and Montenegro. 48 country" status under the WTO has shielded Romania from external pressures to make stronger commitments to liberalizing access to domestic markets. E.l. Applied MFNTariff RatesinComparative Perspective 2.84 Romania has the highest MFNappliedtariff rates among CEEC-10 countries. The simple average tariff rate fell in 2002, yet it was almost 50 percent more than the simple average of 13 percent inPoland, the second most protected economy among the CEEC-10 (Table 2.9). Simple average tariff rates for bothagricultural and industrial products are also significantly higher than inBulgaria, which is scheduled to accedeto the EUjointly withRomania. Table 2.9 Average Applied TariffRates in the CEEC-10and the EUand Implied Preferential Margins vis-his Tariff Rates in the EU AverageMFNApplied Rate(%) a` Impliedpreferentialmarginsover the EUb' Country Industrial Goods Agriculture Goods All Goods Agriculture IndustrialGoods Romania (2001) 19.8 33.9 16.2 15.4 24.5 12.0 Romania (2002) 18.6 27.3 15.9 14.2 17.9 11.7 Bulgaria (2002) 9.6 16.4 9.6 5.2 7.0 5.4 Czech Republic 5.0 11.7 4.2 0.6 2.3 0.0 Estonia 1.6 11.5 0.1 -2.8 2.1 -4.1 Hungary 9.5 25.7 7.0 5.1 16.3 2.8 Latvia 4.1 11.0 2.9 -0.3 1.6 -1.3 Lithuania 3.4 9.4 2.9 -1.0 0.0 -1.3 Poland 12.6 47.4 10.2 8.2 38.0 6.0 Slovak Republic 5.1 8.1 4.2 0.7 -1.3 0.0 Slovenia 9.6 8.1 9.4 5.2 -1.3 5.2 Memorandum European Union 4.4 9.4 4.2 0.0 0.0 0.0 Notes: dThe classificationsof all goods, agricultureand industrial goods are based on HS 01-97, HS 01-24 and HS 25- 97, respectively.b/ The differencebetweena country's MFNappliedtariffrateandthat applied by the EUindicatesthe level of extrareversediscriminationfaced by the MFN supplier in comparison with that faced in EUmarkets. c/ Based on the WTO CD Rom 2001 file, Romania's MFN applied tariffs are even higher: 40.0 percent for all goods, 117.7 percent for agriculture, and 17.5 percent for industrial goods. Data for 2002 were obtained from the Romanian govemment. Source: WTO Members'Tariff Profiles(TN/MA/S/4/Rev. I), andWorld Bank staff estimates. 1999, 2.85 Nevertheless, the combination o f high MFN tariff rates and low preferential rates does not appear to have led to an increase intrade with FTA partners at the expense o f the ROW. On the contrary, both preferential and non-preferential trade has dramatically increased since the collapse o f central planning. The price o f preferential access for Romanian producers to most o f Romania's "natural" trading partners' markets has been the strong competition that they have faced from imports from FTA partners, includingthe EU. Both privileged access and exposure to competition appear to have had a positive impact on the ability o f Romanian producers to successfully compete ininternationalmarkets. 2.86 Thus, if anything, the trade-creating effect o f preferential trade agreements appears to have been well above the trade diversion inherent in the combination of high MFN tariff rates and regional FTAs. Setting aside the "natural trading partner arguments" (geographical 49 proximity and the economic weight o f a country or regional bloc), the apparently limitedimpact o f highreverse discrimination on overall trade flows also stems from the highdispersion intariff rates, with MFNtariff peaks affecting small amounts o f trade and with tariff exemptions granted on discretionary basis. E.2. Preferential Partners: Presence and Conditions inAccess to Romanian Markets 2.87 Current preferential partners accounted jointly for 73 percent o f total Romanian imports in 2002:69 percent of these total imports, or 94 percent of total preferential imports, entered Romania duty free. The remaining 6 percent were subject to duties because they were agricultural products or because duties on industrial products imported from certain countries (Israel and Turkey) were still in effect. While simple average tariff rates on imports from each prospective free trade partner were positive, except for Lithuanianimports, their levels were very low, with a maximum of 3.3 percent for Israel (for details, see Table 2 in Kaminski, Unguru 2004). 2.88 The averages weighted by imports from a respective partner (a more appropriate measures) are mostly zero except for total imports from Hungary (3.2 percent), EFTA (3.1 percent), Israel (3 percent), andEU(1.5 percent) owing to signifkant importsof agricultural products, which still enjoy substantial levels o fprotection inRomanian markets. 2.89 The share of other countries acceding to the EU in 2004(Cyprus and Ma1ta)in Romanian total imports in 2002 was paltry amountingto 0.09 percent, whereas the share of Stability Pact countries was slightly higher at 0.5 percent of total imports. Both simple and weighted average duties on imports from Stability Pact countries and from 2004 EUentrants that do not have FTAs with Romania (that is Cyprus, Estonia, Latvia, Lithuania, and Malta) are very high, at double digit levels. However, while an increase intheir exports to Romania may be expected once they have duty free access, this will not have a significant impact on Romania's overall imports. It should be noted also that under the current tariff regime some portion of their imports enter Romania on a duty free basis either because o f limitedbilateral liberalization (e.g., 0.4 percent o f Croatian imports) or because MFN duty rates are zero. For example, the latter applies in particular to Latvia, with 29 percent of imports subject to zero MFNapplied rates. E.3. Tariff Barriers in2002: The ExtentofReverseDiscrimination 2.90 Shortcomings o f regional liberalization, unless it is accompanied by the simultaneous liberalization o f MFN tariff rates, are threefold: first, it usually suppresses FDI inflows from firms located in non-preferential countries; second, it raises the prices o f imports; third, it exacerbates the potential for distortions inproduction and consumption patterns owing to the fact that variation inthe tariffstructure seriously distorts price signals. 2.91 The Romanian MFN tariff structure contributes to distortions in production and consumption patterns, which relate to high levels in MFN tariff rates and differences between MFNapplied rates andpreferential rates on imports from FTA countries (Table 2.10). The latter are significantly higher on industrial products than on agricultural products. 50 Table 2.10 Characteristics of Tariff Structures Levied on Agricultural and Industrial Products Imported from Non-preferential MFN Partners and FTA Partners in 2002 AgriculturalProducts (HS 0 through 24) Average All FromMFN FromFTA Partners Simple average 20.2 27.9 17.2 Weighted average 23.8 34.8 28.7 Coefficient o f variation 1.04 0.72 1.12 Maximum tariff rates 225.5 225.5 225.5 IndustrialProducts(HS Chapters25 through 97) Simple average 5.2 15.9 0.1 Weighted average 1.2 4.4 0.02 Coefficient o f variation 1.82 0.63 0.22 Maximum tariff rate 225.5 225.5 77.5 Note: Calculatedon 8-digit HarmonizedSystemitems for the 2002 tariff schedule. Source: World Banks staff estimates based on 2002 tariff schedule. 2.92 Regional preferences applied to agricultural products seem to be less distorting in terms of production and consumption patterns than a simple examination o f respective tariff schedules would suggest. While almost 40 percent o f tariff items on imports from FTA partners were not subject to duties as compared with 6 percent from nnn-preferential partners, the value o f the MFNweighted average tariffrate is only 6 percentage points or 22 percenthigher than the same average on agricultural imports from the EU.56This means that preferential concessions have not been granted for products deemed sensitive and, by the same token, they contribute less to trade diversion.57 2.93 Tariffs and preferential conditions for access to Romanian markets for industrial rather than agricultural products are a much more important source for creating distortions in the economy and opportunities for rent seeking. Romania's tariff regime for industrial products has the following features. First, although MFN zero tariff rates account for only 11.5 percent o f eight-digit H S tariff items, most o f the industrial products imported into Romania were duty free in 2002. While "developing country'' status has allowed Romania to keep MFNtariff rates at highlevels, this has beenpartly offset by zero tariff rates on a number of raw materials andby Romania's participation in WTO-sponsored arrangement^.^^ As a consequence, 71 percent o f MFNimports and63 percent ofGSPPl6 imports were subject to MFNzero tariffrates, whereas practically all imports o f industrial products from preferential partners entered duty free in 2003 (Table 2.11). 56Imports from CEFTA countries (excluding Croatia) are subject to slightly lower tariffs than those from the EU, but the differences are not large. The simple average tariff rate on CEFTA agricultural imports o f 16.1percent was one percentage point lower than that on imports from the EU (16.1 percent). All agricultural exports originating in Moldova are undera duty free regime. 57Butthis should not suggest that high levels of agricultural protection impose costs on the economy. As is argued inChapter 4 ofthis study, they have had serious negative consequences for domestic efficiency insectors producing agro-food products. 58Romania i s a party to the conventions on Preferential Trade among Developing Countries ("The 16") and the Generalized System o f Trade Preferences (GSP). It is also a party to the WTO Information Technology Agreement (ITA), which has ledto zeroing o fMFNtariff rates onproducts coveredbythe ITA. 51 Table2.11Duty-freeImportsof IndustrialProductsunderMFNandPreferential Agreements in2002 - Duty-free Imports(in Shareof duty-free Share intotal Duty regime Imports(in importsunder millionsof US Ofus respectiveduty industrial dollars) dollars) regime imports ~~MFNimports 2,610 3,702 71% 17% GSPF16 423 671 63% 3% FTApreferentialpartners 12,109 12,135 100% 80% Total industrial imports 15,142 16,661 91% 100% Source: Derivedfromdataprovidedby the RomanianCustomsAdministration 2.94 Second, in spite o f the dominance o f duty free imports in industrial imports from these three groups o f partners, preferential margins lead to acute reverse discrimination. The simple MFNaverage tariff rate implies that an "average" MFNsupplier would have to offer prices that were 16 percent lower than those o f preferential firms in order to be competitive in Romanian markets. But the average is usually misleading. This may quite effectively cut access for Romanian import users to the most cost-effective sources o f supplies for a wide range o f products, where, for example, Japanese, Korean or US.firms may have a competitive advantage and may offer higher quality products. 2.95 It appears that this indeed may be the case, as Romanian importers o f industrial products display interest in MFN suppliers only insofar as preferential margins are relatively low (see Table 2.12). Once one starts to move up the MFN tariff ladder the respective shares dramatically change in favor o f FTA partners, whose share intotal imports reaches a maximum o f 93 percent o f these imports, where MFN suppliers face tariff rates o f 15 percent or higher. Since differences in specialization profiles o f MFN and FTA producers may provide some explanation as to the concentration o f MFNimports inproducts subject to zero MFNtariff rates, this study will exclude them for comparative purpose^.^' 2.96 The exclusion o f MFNimports o f energy and industrial raw materials produces a striking (though not a counterintuitive) picture. MFN imports subject to rates exceeding 35 percent account for a paltry 0.3 percent o f all non-zero MFNimports. The share o f MFNimports intotal industrial imports o f items subject to higher MFNtariff rates rapidly falls from 14percent for 10- 14percent rates to 6 percent for items subject to rates between 30 and 35 percent. MFNimports account for 5 percent o f total imports o f items subject to ratio between 38.4 percent and 44.2 percent and to the 64 percent rate, but disappear at higher rates (see Table 2.12). 2.97 Nevertheless, it is surprising that these imports are taking place at all. There are three possible explanations for this situation. One explanation is tariff exemptions, which-judging by customs revenues from tariffs-appear to be widespread. Furthermore, goods imported by firms operating in Romanian Free Trade Zones and economically depressed areas are exempt from duties. So are imports o f capital equipment by small and mediumenterprises or by larger firms if their value exceeds US$ 1 million. Imports o f goods to be exported within 45 days are also 59Imports o f fuels alone accounted for 28 percent o f non-preferential imports, as opposed to 1.4 percent o f imports from preferential partners. 52 exempt. As mentioned earlier, this may have weakened the trade diversion effect usually triggered by highpreferentialmargins. Table 2.12 MFNImports at Various MFNApplied Tariff RateBracketsand Their Share in ImDortsof These Tariff Itemsin 2002 (in millions of US dollars and inDercent) All MFNappliedrate 225 155 77.5- 86.5 (excluding 64 38'4- 30-35 21-25 15-20 10-14 44.2 o% tariff Total imports) MFNimports (percent o% o f total) 0% 0% 5% 5% 6% 7% 8% 14% 9% 24% Imports (inmillions of 15,21 US$) 6.8 2.6 13.8 0.2 60.4 1,350 1,846 4,796 1,928 11,411 As percent ofnon-zero MFNimports 0.0% 0.0% 0.0% 0.0% 0.3% 8.0% 11.3% 35.7% 25.5% 100.0% NIA Source: Derivedfrom dataprovidedby the RomanianCustomsAdministration. 2.98 Another possible explanation may relate to the unavailability o f competitive products in the pan-European industrial free trade area. Inthis case, there i s no choice but to pay high duty rates. 2.99 Last but not least, MFN suppliers may lower export prices to stay competitive in Romanian markets. Chang and Winters (2002) provide empirical evidence showing that firms lowered the export prices that they offered to consumers from MERCOSUR countries, in order to compete with firms from this regional trading bloc. This may explain imports o f items subject to 10-20 tariff rates, but it seems unlikelythat MFNproducers would be willing to slash export prices to internalize the higher double-digit preferentialmargins o f their competitors. F.IMPLICATIONSOFTHEALIGNMENTOFMFNAPPLIEDTARIFFRATESONINDUSTRIAL PRODUCTSWITHTHE EUCET 2.100 The exclusion o f MFNimports o f energy and industrial raw materials produces a striking (though not a counterintuitive) picture. MFN imports subject to rates exceeding 35 percent account for a paltry 0.3 percent o f all non-zero MFNimports. The share o f MFNimports intotal industrial imports o f items subject to higher MFNtariff rates rapidly falls from 14percent for 10- 14 percent rates to 6 percent for items subject to rates between 30 and 35 percent. MFNimports account for 5 percent o f total imports o f items subject to ratio between 38.4 percent and 44.2 percent andto the 64 percent rate, but disappear at higher rates (see Table 2.12). 2.101 N o matter which o f the above reasons explains the decreasing levels o f MFN imports with the increase in rates, the bottom line is that Romanian MFN applied tariff rates are unnecessarily high. High tariff protection is always a bad trade policy. In the absence o f extensive regional trade arrangements, tariff protectionism can be defended (or explained) in terms o f infant protection argument, government revenue needs, or political economy reasons related to political pressures from powerful interest groups. If, however, the same sectors are open to duty free competition from firms from such highly developed economies as the EU or EFTA, not to mention the internationally competitive imports from other EU entrants, none o f the above applies unless a protected product i s not manufactured in any o f the FTA countries. 53 This is hardly a plausible occurrence. If this is not the case, then maintaining high MFNtariffs amounts to protecting FTA suppliers from MFNcompetition. 2.102 Since Romania i s on the accession path to the EU, MFNtariff rates on industrial products should be harmonized with those in the EU CET (Common External Tariff). Their alignment would send a signal to the investment community indicating the government's commitment to addressing weaknesses in Romania's domestic business climate. The cost o f such a move is negligible. This is clearly a win-win case. F.1.A "Shadow CustomsUnion"with theEURestrictedto IndustrialProducts 2.103 In view o f Romania's involvement inthe EU accession process, the best option-that o f free trade-is not acceptable, as it might trigger demands for compensation from Romania's MFNtradingpartners onceRomaniabecomes amember ofthe EU. 2.104 But since Romania's MFNtariff rates on industrial products are significantly higher than those inthe EUCET, the harmonization o f tariff rates applied on industrial imports with those in the EUwould lower the extent o f reverse discrimination considerably. The simple average MFN rate i s almost 4 times higher, and the maximum tariff rate is almost 15 times higher.60Romanian MFNappliedtariffrates are lower thanthose inthe EUfor 165 six-digit HStarifflines out ofthe total o f 5,492 lines. The maximum difference for these items is fairly small, amounting to only 10 percent. On all other items, Romanian tariff rates are much higher than rates in the EU. For example, the difference between Romanian and EU applied tariff rates exceeds or is equal to 10percent for 2,778 lines, or 51percent o f all industrial six-digit tariff lines., 2.105 Hence, the alignment of applied MFN tariffs on industrial products to those o f the EU CET would represent a significant improvement in the contestability o f Romanian domestic markets. As a result o f this, competitive pressures on either preferential suppliers or domestic producers, or both, would significantly increase. 2.106 However, concerns might be raised about the compatibility o f the alignment with WTO rules, the negative fiscal impact, and the negative impact on sensitive sectors. None o f these concerns appears to be well grounded. F.1.1. Implications: WTOI s Not an Obstacle 2.107 As regards WTO rules, three comments are worth noting (for details, see Box 2.1). First, these rules impose discipline on the increase o f applied tariff rates beyond their bound levels. Member countries may be penalized for erecting barriers to market access, but not for their removal. 6oThe accuracy o f these estimates suffers fiom a drawback related to the absence o f full concordance between Romanian and EU tariff lines. This is mainly the result of the use by the EU of specific and compound tariff rates, which have not been includedinthe analysis. 54 Box 2.1 Harmonization of Tariffs and WTO Rules From the point o f view ofthe WTO rules, accession is simply the shift from a free trade area (FTA already fully operational under Europe Agreements) to a customs union (CU). The only new (in relation to FTA already in place) relevant article that will apply when Romania either joins the EU or signs a C U agreement is Article XXIV:6. Article XXIV:6 requires WTO membersjoining a C U and consequentlyseeking to increasebound rates oj duty to enter into negotiations-under Article XXVIII (Modification o f Schedules)--on compensatory adjustment, netting out reductions in duties on the same tariff lines made by other members o f the customs union. If such reductions are regarded as insufficient compensation, the CU will either reduce duties on other tariff lines, or provide compensation. Ifno agreement is reached on compensatory adjustment within a reasonable period o f time, the C Umay modify or withdraw the concessions and affected WTO members are free to retaliate. A reasonable period o f time, as agreed during the Uruguay Round, is to be no more than 10years. This has the following implication: the legal situation vis-a-vis the WTO will remain the same as long as Romania will not insist upon accession that the EUraise its bound rates to higher Romanian ones. Other issues are unlikely to appear as legal impediments. Reference is occasionally made to Article XXIV, 5 (a), which requires that duties and other regulations of commerce after the establishment be either less restrictive or n o more restrictive than prior to the establishment o f a CU. If the applied rates are after accession on the whole higher than those applied before accession, the enlarged EUwould be in breach of its WTO obligations and any WTO member will be entitled to ask for compensations. This is a misinterpretation. First, as noted above, the trigger for compensation adjustment is the increase in bound rates. Second, an assessment under this Article is based on an overall assessment of weighted average tariff rates and customs duties collected. The WTO Secretariat is responsible for conducting an assessment, which has little practical relevance. It should be noted that, since 1948, out o f more than 60 FTAs and preferential trade agreements that have been reviewed under Article XXIV provisions, only 4 agreements (the list includes neither EU CU nor NAFTA) were declared hlly compatible with Article XXIV requirements. However, a working group has not censured a single agreement as being incompatible with GATT-WTO rules. 2.108 Second, lowering tariff rates on a unilateral basis does not reduce the govemment's leverage in tariff negotiations, as such negotiations concern the bound levels, not the actually applied levels. Moreover, this is o f little relevance-the Doha round o f multilateral trade negotiationsmay go beyond the date o fRomania's scheduled accession to the EU. 2.109 Last but not least, WTO rules pertain to bound tariff rates not actually applied. In other words, problems with WTO members may arise only if Romania persuades the European Commission to raise the WTO bounds to its own levels. Otherwise, Romanian accession to the EU, or more precisely its membership in the EU Customs Union, will not change the overall level o fprotection inEUindustrial markets. F.1.2.Implications: Tariff Revenues and MFNImports 2.110 With the expanding share o f duty free imports, customs revenues have been on the decline. Inrelation to the value o f imports, they fell from around 6 percent in 1998 to 2 percent in 2002 and may have declined further in 2003, as all tariffs on industrial imports from Pan- European Cumulation Agreement countries have been zeroed. Furthermore, tariff exemptions under various schemes have reduced MFNdutiable imports. The latter seem to be accountable for the difference between the values o f a duty collection rate o f 2.0 percent and a weighted average tariffrate of 2.8 percent. 55 2.111 With reductions in preferential tariff rates on industrial products, duty revenues from agricultural imports have been accounting for a dominant share o f customs revenues from duties. Assuming that a weighted average tariff rate on agricultural products o f 24 percent roughly equals the implicit duty collection rate on these imports, duty revenues generated by agricultural imports amounted to 82 percent o f the total in 2002, with only 18 percent coming from MFN industrial imports. The projected value o f the duty collection rate o f 1.7 percent in 2003 i s very close to the weighted average rate on agricultural imports multiplied by their share in total imports, which yields 1.5 percent. The projected collection rates appear to be based on the assumption that duty revenues from agricultural imports will be even more dominant than inthe past. 2.112 In consequence, the impact on tariff revenues is unlikely to be significant. But the alignment will nevertheless affect the duty income from MFN imports, as tariff rates would substantially fall. The fall may be offset by increased MFNimports as discrimination levels drop. How large will the change in duty revenues from industrial imports be with the adoption o f EU CET on industrial products? To answer this question, we use a very simple standard partial equilibrium trade model to simulate trade creation and assess tariff revenue loss. It is based on the assumption that the percentage change in imports equals the elasticity o f import demand multiplied by the change in price (or here, in tariff rate) for this product.61 It is assumed that preferential imports will not change, since tariff rates for them remain unchanged. 2.113 The results o f a partial equilibrium simulation presented in Table 2.13 confirm our expectations. Ifwe assume the full alignment with the EU applied tariff rates, the value o f MFN imports would increase 6.2 percent. This would not have a significant impact on the geographic pattern o f industrial imports. The share o f MFN suppliers would increase 1.4 percentage points (or 5 percent) from the current o f 22.0 percent to 23.4 percent o f the total. This is clearly not a dramatic change. 2.114 Neither is the change incustoms revenues from duties on industrial imports, which came exclusively in 2003 from MFN imports. Although the implicit duty collection rate would fall rather dramatically (70 percent, from 4.8 percent to 1.5 percent), there are two important caveats. First, the actual collection rate, because o f the previously mentioned exemptions, was significantly lower than the rate assuming full collection o f due duties (implicit collection rate)- 4.8 percent as compared with 1.9 percent or almost 70 percent lower. Second, because o f the large share o f preferential imports, the concentration o f MFN imports in products subject to relatively low applied tariff rates and tariff exemptions, the duties from industrial imports would account for a relatively low portion o f total customs revenue from duties. In consequence, a 24percent fall in duty revenues from importers o f industrial products would produce only a 4percent decline in total revenue from duties. Considering that customs revenue from duties accounted for a mere 2.01 percent o f the total budget revenue in 2004, this would be a barely perceptible change. 6'For the discussion o f issues involved inthe use o f partial equilibriummodels, see, for example, Laird and Yeats (1989 and 1990) and Stem (1976). Elasticities used inour calculations were taken from Stem (1975). 56 Table2.13 Simulationof ChangeinMFNImportsandTariff Revenueunder the Scenarioof the FullHarmonizationof IndustrialMFNRateswith the EUCET Full alignment Change MFNindustrial imports 3,658 3,886 6.2% Income from duties on industrial products 69 56 -19% Impliedduty income 176 56 -68% Implicit collection rate 4.8% 1.4% -70% Actual collection rate 1.9% 1.4% -24% Total duty income 357 344 -3.7% Memorandum: Share ofMFNindustrial imports intotal industrial imports 22.0% 23.4% 6.2% Source: Own calculationsbased on data providedby the Customs Administration. 2.115 The results obtained are not counterintuitive. Considering the large proportion o f MFN imports entering Romania on a duty free basis and the trade patterns o f CEEC-9 with much lower MFNtariff rates, one would not expect a dramatic shift in imports toward non-preferential partners with the drastic cut in tariff margins currently enjoyed by preferential suppliers. The calculations that are aimed to provide a numerical answer to this question are based on three simplifying assumptions. First, we assume that markets are not ccmected (ie., the change in demand in one has no impact on another). Inthe real world, markets are connected and higher imports may ceteris paribus lead to the depreciation o f local currency and thereby limit the increase inimports. 2.1 16 Second, since preferential suppliers enjoy duty free access, we assume that only MFN imports are affected by lower tariffs. This i s not necessarily the case, as preferential exporters faced with stiffer competition from MFNproducers may seek to maintain their market shares by lowering their offer prices. Onthe other hand, MFNexports may be higher than predicted, as the model used here does not capture the possibility o f the entry o f MFN suppliers that were previously crowded out by high preferential margins. Third, calculations may actually underestimate the increase in MFN imports, as they do not take into account the possibility of imports o f products that were not imported from MFNsuppliers in2002. For products subject to highMFNapplied tariffrates, there were 135 HS eight-digit items that were not imported at all. Furthermore, there were 20 products subject to non-zero preferential rates, 18 o f which were imported exclusively from preferential countries (for the list, see Kaminski and Unguru, 2004, Table 3.13). It is quite likely that, once MFN tariffs fall, for example, from 62 percent to 6 percent, competitive offers from MFNproducers will appear. F.1.3. Implicationsfor Domestic Firms: Sensitive Sectors 2.117 Lowering MFN applied tariff rates on industrial products to EU levels may have a negative impact on Romanian producers o f a product, subject simultaneously to an MFNtariff peak and a zero-duty rate for preferential partners that i s not competitively manufactured in any o f 29 countries with which Romania has FTA within the pan-European framework. The indication o f such a situation would be the absence o f imports o f products subject to MFNtariff peaks (30 percent and above) from preferential partners. Out o f 996 tariff lines, this applied in 2002 to 103 eight-digit H S items-for these products there were neither preferential nor MFN 57 imports. With the lowering o f tariffs to EU levelsY6*producers o f these products may face competitive pressures from MFN imports. The average weighted tariff rate for these imports from MFN countries would fall from 31 percent to around 10 percent after the alignment with the EUCET on industrial imports. 2.118 However, only a few producers may face the challenge o f adjusting to higher competition from imports. With few exceptions, products subject to MFN tariff peaks from preferential partners entered Romanian markets duty free in 2002. Since these remaining tariffs were zeroed in 2003, the producers of these products now face duty free competition from preferential partners. For the remaining products, the producers had to compete with preferential duty free imports. Preferential imports o f these products amounted to US$ 1.4 billion and MFNimports to US$89 million in2002. 2.119 One may thus conclude that the majority o f Romanianproducers o f products protected by MFNpeaks do not face any significant adjustment problem associatedwiththe harmonization of MFNtariff rates on industrial products. Most of these producers have already been exposed to competition coming from pan-European firms. And most o f them have either gone out o f business (not being able to withstand the competition) or have successfully adapted to the new competitive conditions. F.2. Summary: Costs andBenefitsofIndustrialTariff Harmonizationwith EUCET 2.120 The harmonization o f industrial MFN tariffs entails several benefits and few, if any, costs. The major benefits stem from the rationalization o f the tariff structure and, by the same token, the minimization o f the distortions inevitably produced by diversity in tariffs. EU tariffs on industrial products are significantly lower and less diversified and tariff peaks are lower. Romanian users of imports would benefit from lower tariffs, and their bargaining position vis-h- vis preferential suppliers would increase once preferential margins were reduced. Furthermore, with the fall intariffs, the transaction costs associated with exports and the corresponding tariff rebates on imported inputs would also decrease considerably, as lower amounts of capital would be tied up. In a similar vein, importers o f MFNproducts would be less inclined to actively seek duty exemptions onMFNimports. 2.121 The costs are insignificant. Although the budget will loose customs revenue from duties, the amounts involved are relatively small as most o f the imports that would potentially generate tariff revenues come from preferential rather than MFN suppliers. In spite o f the significant differences inthe respective structures o f tariff protection-as captured inter alia by the value o f a correlation coefficient o f 0.51-the cost o f adjustment will be low, simply because most producers have already been exposed to competition. However, this should not suggest that a more detailed analysis o f the employment impact o f some tariff cuts is not warranted. 62 This would implythat the simple average tariff rate would be reduced from 33 percent to 9 percent. The average level o f reverse discrimination against MFNsuppliers interms o f that situation inEUmarkets would then fall from the current average o f 24 percent to zero. 58 G.CONCLUSIONSAND POLICYRECOMMENDATIONS 2.122 Empirical analysis supports the conclusion that Romania's foreign trade performance appears to be based on solid ground, and provides reasons for optimism as to its sustainability. This conclusion can be summarized inthe following ways. First, the expansion inexports, driven bythe import demand for Romanianproducts inEUmarkets, has continued unabatedsince 1993. Regarding a recent expansion in exports to both preferential and non-preferential partners, its sources have nothing to do with a simple redirection o f trade following the collapse o f central planning. Its sources are the new, internationally competitive, industrialcapacities. 2.123 Second, producers located in Romania do not appear to be marginal suppliers, excessively vulnerable to swings in import demand in highly developed markets. The proof o f this would seem to be that the dramatic increase in the Romanian presence in EU markets occurred in2000-02, against a background o f falling import demand inthe EU. 2.124 Third, Romania is not a "single product" exporter. Although textiles/clothing towers over other products in Romania's export offer, its dominance has recently been undermined. First there was the shift inRomania's exports toward sunrise products. Inaddition, the share insunrise exports o f capital intensive products (which have been the main drivers o f this growth), was four times higher than intotal EU-oriented exports. Finally, there has been a healthy shift within the "network trade" toward parts and components. While these exports might be less "photogenic" than exports of final products, they provide for the stability associated with being part o f the well-established supply chains of large firms. 2.125 Fourth, the country's export offer has become not only more diversified but also more attractive to non-preferential markets. Capital equipment has driven export expansion, not only to the EUbut also to other markets. The combination o fproduct and geographical diversification seems to augur well for the sustainability of exports. 2.126 Fifth, although the FDIinflows in relation to GDP were still well below the average for the CEEC-10 in 1997-2001, they accounted for one-fifth o f Gross Domestic Investment in this period. The earlier experience o f Hungary or Poland shows that these new firms with foreign participation may soon become major forces behind export growth. In addition, the large presence o f Italian firms (almost 13,000 in 2003) with a total capital investment o f about US$600 million, appears to indicate that trade with Italy, which accounts for almost one-fourth o f Romania's total trade turnover, i s on a firm footing. These firms are largely responsible for exporting clothing and footwear that account for around 60 percent o f Romanian exports, and for importing textile and footwear parts that account for around 40 percent o f Romanian imports from Italy. The implicationi s that these firms are probably not footloose investors and they may have an interest inmaintaining their competitiveness in foreign markets in operations conducted from Romania. 2.127 While these factors point to the potential for sustaining Romania's impressive export performance in the future, the government's policies in staying the course o f macroeconomic stability and inundertakingstrong measuresto improve the business climate are critical. A better business climate i s crucialto attracting capital, both domestic and foreign, and i s indispensable to address certain disturbing developments in Romania's export performance. The first o f these 59 developments is that low FDIinflows appear to beresponsible for Romania's idiosyncrasy inthe CEEC-9 context, namely, the very limited (and not expanding) presence o f skilled-labor- intensive products in Romanian exports. In fact, the factor intensity o f the EU-oriented export basket, after a significant change in 1994-95, reverted to its original composition as it was in 1993, with one caveat. The share o f capital-intensive products increased at the expense o f natural-resource-intensive products. While the line between capital-intensive and skilled-labor- intensive products i s blurry, the continued dominance o f unskilled-labor-intensive products defies the experience o f most other CEEC-9 countries. Inmost o f these countries, the initial gap between their endowment in relatively high skilled labor and the factor intensities o f their exports has been closed fairly rapidly, thanks to foreign investors. 2.128 The second development is that low FDI inflows are also responsible for a relatively underdeveloped trade situation in its most dynamic sector in the contemporary global economy that i s associated with the internationalization and fragmentation o f production. Trade triggered by fragmentation is referred to as "intra-product" or "network" trade. Incomparison with most other CEEC-9 countries (apart from Bulgaria, Latvia and Lithuania), network trade accounts for a relatively small portion o f Romanian exports. Except for furniture, trade in the networks appears to be highly volatile and to move from one transaction to another. This may indicate that Romanian firms have yet to establish a stable position in the EU supply chains. Intra-product trade, based on "just-in-time" supplies, requires a very strong trade-facilitating environment. 2.129 The divergence between the export basket and the relative endowments, together with underdeveloped intra-product trade, points to weaknesses inthe business climate. Both situations suggest that government policies have prevented the emergence o f competitive markets to allocate resources to industrial sectors with a potential comparative advantage. While potential production costs and proximity to the most important markets are the necessary conditions to attract FDI, they are not sufficient to attract MNCs that establish "outside" production blocs. Operations o f value-chains are particularly vulnerable to potential delays and disruptions in various stages o f the supply chain. Ultimately, these operations determine the location decisions o f MNCs. Burdensome customs procedures, transportation delays, or telecommunications problems usually prevent the emergence o fborder-spanningproduction networks. 2.130 Measures that are taken to improve the business environment should be extended to agriculture. Inview of its soil and climate conditions, Romania should be a net exporter o f foods. Instead, it has been a net importer, which clearly suggests that external barriers to trade in agricultural products are not responsible but that domestic policies are to blame. The creation o f an agriculture-friendly environment would go a long way toward boosting domestic output, and would contribute to economic growth and to poverty reduction. 2.13 1 Through the European Association Agreement and the EU Eastem Enlargement Project, Romania has become part o f the process o f policy and institutional changes that facilitate integration into the global economy. The EU associate status has entailed bilateral trade liberalization not only vis-a-vis the EU but also in trade with other countries-most o f them geographically close. As a result o f actively pursued regional liberalization within the pan- European framework, Romanian producers o f industrial products have duty free access to most 60 o f their partners' markets and are exposed to intense competition from imports-motivating investment decisions. 2.132 Romania has, however, failed miserably on one count: multilateral liberalization has not accompanied regional liberalization. Incontrast to exports, where the improved market access to regional partners has been significantly offset by much lower MFNrates than inRomania andby competition from EUproducers, the scope for reverse discrimination in Romanian markets has been several times higher. 2.133 A detailed examination o f Romania's imports at all ranges o f applied MFN and preferential tariff rates shows that maintaining the applied MFNtariff rates on industrial products at levels exceeding those in the EU's CET is a mistake, for three major reasons. First, since producers from 30 countries, including the most advanced economies, presently have duty free access to Romanian markets, high MFN tariff rates serve no protectionist purpose. With the zeroing o f all duties on industrial imports from preferential sources in 2003, they are simply redundant. 2.134 Second, maintaininghighMFNtariff rates on industrial imports, which are now duty free under Pan-European agreements, exacerbates the reliance on less cost-effective sources o f supply. Importers are inclined to import from less cost-effective sources even inthe presence o f duty-drawback or tariff rebate schemes for MFNimports. 2.135 Third, aligning MFNapplied tariff rates with the EU CET will reduce the administrative burden associated with running duty-drawback or tariff rebate schemes, which are now prohibited under the Pan-European Cumulation Agreement for exports within the Pan-European market for industrial products. These schemes, as well as special economic zones, are administratively burdensome and also increase the potential for corruption. All such arrangements raise the transaction costs o f imports and thereby raise the domestic prices of imported goods as well as domestically produced goods. Lowering tariff rates would increase the level o f competition in domestic markets, would reduce the administrative burden and thus would benefit both consumers and users o f imports. 3. RESTRUCTURINGTHE ENTERPRISESECTOR 3.1 carrying out privatization to strategic investor^.^^ However, inno other area, has the legacy o fthe Inthe last few years, Romania has made some progress in enterprise reform including socialist years weighted so heavily in Romania as in enterprise reform and private sector development. The slow and inefficient enterprise reform o f the past has left Romania with a larger number o f enterprises to be privatized or liquidated than in all o f the other CEECs combined. Public enterprises are the core o f the un-restructured part o f the economy. As discussed inthe previous chapters, there are also dynamic private businesses, which are leading Romania's solid trade performance in the EU and world markets. Their scope of action, however, is limited to a few activities operating around the un-restructured enterprise sector. Furthermore, the expansion o f these internationally competitive private firms i s hampered by the sizable un-restructured public enterprises and the soft budget constraints that keep unviable enterprises from exiting. This chapter is divided into three sections. Section A described early transition reform efforts, as compared to those undertaken by other countries. Section B documents their effects on economic growth. Section C suggests some dimensions o f the remaining reform agenda. Figure3.1: Many new enterprisesenter, few exit (1993-2002 average) S -6 E Q f nentry mexit Source: Bartelsman et a1(2004). ~ 63This chapter was prepared by Simeon Djankov (CICMA), the World Bank Group. Juan Carlos Ginarte prepared the section on nonpayment and accumulation o f arrears, and Stella Ilieva prepared the analysis of total factor productivity; bothEconomists ECSPE, the World Bank. 62 A. ECONOMIC RESTRUCTURINGREFORMS: ROMANIA A REGIONAL PERSPECTIVE IN 3.2 The main thesis o f this chapter i s that the Romanian enterprise sector is characterized by a vibrant and expanding fringe o f successful private businesses, which operates around a slowly- shrinking core of non-competitive state owned enterprises. The competitive sector finds opportunities in some domestic markets: services, construction, and light manufacturing; and has accounted for almost all o f the recent boom in exports. Many new firms have started operations inthe last few years (Figure 2.1) and these entrepreneurs make good use of existing institutions. However, their expansion is hampered by the continued existence o f a non-competitive core o f enterprises, which crowd out opportunities for extemal financing and sales. This effect is apparent inthe difference inthe share o f new entries and exits as a share o f economic activity. In the typical market economy, these numbers go hand inhand. InRomania, exit rarely happens. 3.3 The result is an unstable business environment in dealings between private and state- owned companies. Market economy institutions are sometimes obviated by non-competitive practices intolerance o f arrears to utilities and the tax office. This sends a signal to entrepreneurs that perhaps they, too, can escape payments to suppliers in the state sector. Incontrast, business between private companies is performed on market principles. The significance o f this dual system is diminished over time, but the slow progress retards the further development o f a business-friendly environment. 3.4 The first ten years o f enterprise reform in Romania were characterized by slowly unfolding privatization, with a large number o f enterprises - between 1,500 and 2,000 - remaining fully or partially in state hands. Recently, the government has carried out a number o f successful privatizations to strategic investors, including Siderurgica Hunedoara, Tractorul, RomanBrasov, andARO Campulung. Still, no other economy inCentral and Eastern Europe has such a large remainingprivatization agenda. 3.5 The prevalent methods o f privatization in the 1990s-insider privatization and mass privatization-resulted in weak corporate governance. Only recently have sales to strategic outsiders and hence boosted economic growth. Lackadaisical privatization was accompanied by soft budget constraints, resulting inlarge arrears to the government, the utilities, and the financial sector. Many unviable enterprises have been kept in operation. Even profitable enterprises benefited from accumulating arrears, reducing their incentives to restructure. Finally, domestic competition and labor reallocation opportunities has been hamperedby the slow exit o f unviable firms, explained both by the presence o f soft budgets, the dysfunctional insolvency system, and labor market rigidities. A.l. Privatization Unfolds Slowly, Far from Finished 3.6 The 1990s were a period o ftransformation o f ownership in Central and Eastern European Countries (CEECs). In Hungary, companies were sold to outsiders, mainly foreigners, based on the strategy that foreign technology is needed to boost productivity. Inthe Czech Republic, mass privatization was implemented in 1992-1994, and in the following five years a rapid concentration o f shares in the hands o f strategic investors took place. By 1998, an estimated 85 percent o f companies had a strategic owner. Foreign investors flowed in, buying both the remaining large state owned companies and many already privatized enterprises. By the end o f 63 2003, a third o fthe enterprise sector has significant foreign ownership. The Slovak Republic also started with mass privatization, but cancelled it in 1994 and instead reverted to direct sales, mainly to domestic investors. However, large enterprises, primarily in the heavy machinery, automotive, and financial sectors, went directly to strategic foreign investors. Since 1999, the Slovak Republic has been the favorite destination of foreign investors in the region, with about 40 percent o f value addednow producedinforeign-owned companies. Poland initially privatized all small companies and allowed for new entry, while weighing options in the privatization o f large enterprises. In 1995, mass privatization started in earnest while banks and other financial institutions attracted foreign investors. Finally, Bulgaria toyed with mass privatization but, like the Slovak Republic and Poland, changed course mid-way and engaged in sales-for-cash that brought in much needed foreign currency. By the end o f 2003, more than 80percent o f productive assets in these economies were in private hands (Figure 3.2). More importantly, nearly all had strategic owners. Figure 3.2 Private Sector ShareinGDP, end- 2002 85 80 75 70 65 60 55 I 50 Bulgaria Czech Estonia Hungary bland Romania Slovak Slovenia Republic Republic Source:EBRDand World Bank. 3.7 In the meantime, Romania experimented with various alternatives. Those took considerable time to develop, and after 10 years o f transition Romania had privatized only 40 percent of its large enterprises, and about two-thirds o f its medium-size enterprises (Table 3.1). By end-2003, there were still approximately 1,300 state-owned enterprises in Romania, including in manufacturing and services; and another 600 enterprises that were de facto in state control. This is a larger number than inthe rest of CEECscombined. Table 3.1 PrivatizationinRomania, 1993-2002 Companies for Companies privatized privatization (share of total) Large 708 288 (40) Medium 2549 1582 (62) Total 3257 1870 (57) Source: The Romanian Authority for Privatization and Management of State Ownership 64 3.8 The goal of the isolationprogram was to bringcompanies back to health or, ifthis proved difficult, to privatize or liquidate them. However, managers remained in place, and were reluctant to take drastic measures since they were either elected by the workers or their appointment was approved by the union. Instead, they focused their energies into vying for access to Romania's own "Structural Funds", and giving explanations o f their inability to pay various obligations to the govemment. 3.9 Not surprisingly, little restructuring took place, with most o f the funds used instead to maintain employment. At the time o f closing the program in February 1997, only four firms graduated into profitability, two were privatized, and two were liquidated, out o f the 147companies that entered the isolation program. Hadthe government usedthe resources o f the Structural Funds to offer generous separation packages to workers, the amount used during the life o f the isolation program would have been sufficient to give 30 monthly wages to every employee. A generous package that would have been difficult to reject. The program instead served to delay privatization and when it became clear that there was no other route, many enterprises were already too troubled to be worth resuscitating. 3.10 The State Ownership Fund. The protracted approach to enterprise restructuring was set from the start. In 1991, the government carried out a legal conversion state-owned enterprises (SOEs) were divided into two groups: remaining SOEs and commercial companies. The former group was smaller in number but included all strategic companies, about 400 in total, which accounted for nearly half o f productive assets.64 A number o f those were entered the isolation program, others were deemed too strategic to do so. Incontrast, the commercial companies were slated for privatization, albeit o f an unusual kind. Their shares were placed in a newly established State Ownership Fund (SOF) and one o f five Private Ownership Funds (POFs), in a ratio o f 70:30 percent. The management o f the SOF was identical to that o f State Property Funds inevery other transition economy. However, unlikeprivatization funds inany other country, the Romanian POFs remained state-managed, their boards of directors appointed by government and approved by Parliament. Their nominal owners, about 18 million Romanian citizens who received shares through mass privatization, received no control. 3.11 Thus the mass privatization program inRomania was unique inits design. Inevery other country that distributed vouchers to citizens-Bulgaria, the Czech Republic, Poland, and the Slovak Republic-the privatization funds were privately owned and operated. This created incentives for the new owners to establish profitable operations or liquidate the companies in distress and sell their assets. Also, it resulted inrapid concentration o f ownership inthe hands o f strategic owners, either a privatization fund or another legal entity that bought the controlling package. For example, by the end o f 1997, or within 3-4 years after mass privatization, over 70percent o f mass-privatized companies in the Czech Republic had a majority owner.65 In Bulgaria, by 1999 nearly 85 percent o f mass-privatized companies has a majority owner.66 In both cases, the concentration took place in secondary-market trading as owners were interested inconsolidation of stakes andimproved governance. Insharp contrast, by end-1998, the largest 64RomanianDevelopment Agency (1997). 65Claessens and Djankov (1999). 66Djankov (2002). 65 private owner in mass-privatized companies in Romania had on average a 24 percent stake.67 This made it difficult to make strategic decisions and commit to long-term restructuring. 3.12 The other major privatization method employed in Romania was management employee buy outs (MEBOs). These started before the mass-privatization program (its heyday in 1995- 1996) and developed into 1997, when the practice largely ceased. By the end o f 1998, about a third o f all industrial firms inthe SOF portfolio has insider owners. Two features o f the process made it unique. a First, the ownership share that went to insiders was very large-65 percent on average, with a median o f 71 percent.68To the extent that there was any outside ownership, mostly through residual mass-privatization shares, it could play no effect incorporate governance; and e Secondly, usually the entire stake was transferred to employees, so the method should really be deemed employee buy-outs. 3.13 MEBOs also took place in Russia and the countries o f former Yugoslavia, especially in Croatia and Slovenia. In Russia, managers o f state-owned enterprises provided the backbone o f the resurgent Communist party and Yeltsin offered them ownership stakes in exchange for support inthe 1996 elections.69 Employees also gained some shares, but had no power over the decisions o f the companies. In effect, these were management buy-outs. By 2001-2001, about three-quarters o f these companies had changed ownership (and management), taken over by large conglomerates or being sold to outside (primarily domestic) investors. Similar developments, with much smaller magnitudes, took place in Hungary and later in Poland. In Croatia and Slovenia, the worker councils traditionally had a large role incompany management and successfully retained power in the transition period. As a result, restructuring o f these companies has been slow and only since 2002 have they experienced any significant interest from foreign investors. Today, many Croatian and Slovenian companies face similar issues as do Romanian companies. However, a major difference is the higher level o f competition in the domestic market, as former Yugoslavia hadhistorically beenmore open to imports from Europe. 3.14 Privatization to strategic investors. Since 1998, sales for cash gained momentum in Romania. Some of these companies were too troubled to be o f interest to insider owners, other profitable companies were offered to foreign strategic investors following the Hungarian experience o f direct sales. By 2002, about two-thirds o f GDP was produced in private companies. However, 33 percent was still in state hands, 24 percent was insider-controlled, 19percent had strategic outsider investors, 13 percent was controlled by shares obtained in the mass-privatization, with the remainder in the hands o f privatization finds and dispersed owners (Figure 3.3). 61EarleandTelegdy (2002). 68EarleandTelegdy (2002). 69Shleifer andTreisman(2001). 66 Figure3.3 OwnershipTypes inRomanianIndustry,end 2002 30 25 20 15 10 5 0 State Insiders Outsiders Mass Others Source: World Bank staffestimates. A.2. PrivatizationModesAffect Outcomes 3.15 Transition economies, in Central and Eastern Europe and further east, have privatized over a quarter million enterprises in the last decade. Various privatization methods have been tried, successes varied, andprivatization strategies were revised as the evidence accumulated. By now, there i s sufficient empirical evidence and analysesthat show that: 0 Privatization in almost any form resulted in better productivity and more employment opportunities than state ownership; and 0 There were differenteffects o f different types o f newprivate owners.7o 3.16 Privatization to strategic outside investors is associated with nearly six times higher productivity growth than privatization to insiders (Figure 3.4). In Central and Eastern Europe, privatization to employees made enterprises worse off than keeping commercialized state ownership. Privatization to managers-insiders also worsen performance. Foreign ownership was particularly beneficial, and this result is supported by the experience inevery country. 3.17 Figure 3.4, one can associate the apparent success of various transition economies with the privatization mode. Estonia and Hungary went for outsider privatization, mostly to foreigners. So did Bulgaria, Latvia and the Slovak Republic, after initial delays and hesitation. The Czech Republic, Lithuania, and Poland had a mixed strategy, but in the end ownership quickly concentrated inthe hands of outside investors. Unique among its peers, Romania ended up with privatization that often made things worse, not better. Romania today is at par with Georgia and Ukraine, far behind the rest o f CEECs. Only recently has Romania carried out privatizationto outside investors and subsequent increase inforeign investment. 'O Djankov and Murre11(2002). 67 Figure 3.4 The Benefits of Outsider Ownership in Central and Eastern Europe (in percentage average annual productivity growth) 8 6 4 2 0 I Note: Insidersdoes not simply add the analyses for employees and managers as some studies do not distinguishamongthe two groups. Similarly, Outsidersdoes not simply add the various groups of outsider investors. Insome cases, these are reported and analyzedtogether. The two categories-Insiders andOutsiders-should only be comparedto each other. Source: Djankov andMurre11(2002). A.3. HardBudgetConstraints,Yet to Come 3.18 Most CEECs have had to deal with loss-making financially unviable enterprises. However, inmost cases distressed assets were isolated under state fund management and sold off or liquidated. Unlike most other CEECs, nonpayment and arrears in the state-enterprise transactions in Romania is common-both financially unviable and viable companies rely on nonpayment to finance their operations. In2002, arrears amounted to nearly 4.5 percent o f GDP. For example, tax arrears amounted to a third of total assets of SOEs; excluding payments to utilities. Wages also grew at a very fast pace, at nearly 9 percent in real terms. Energy utilities also continued to tolerate arrears, allowing unviable firms to survive and preventingmarket exit. In contrast, this practice is nearly absent in private-to-private transactions in Romania. The challenge i s to extend hardbudget constraints observed inprivate-to-private transaction interface to the state-private transaction interface. 3.19 Other transition economies have faced similar issues but the arrears phenomenon is virtually unknown there. In all cases, distressed assets were isolated under state fund management and sold off or liquidated. With few exceptions, as in Polish mines and shipyards, has this issue resulted in significant budgetary allocations. And as many utilities were privatized early on, non-payment usually resulted indelivery stoppage. Unviable enterprises were forced to cease operations. Also unique in Romania, as the banking sector remained in state hands longer 68 than in any other Central and Eastern European country,71 enterprises also continued to receive bank credit to fill the shortfall. In short, through social security and tax arrears, through the banking sector and the utilities, Romania has delayed the sale or liquidation o f many unviable companies. 3.20 More worrisome is the fact that many profitable private companies have been allowed to accumulate arrears too.72In2002, profitable tax arrears o f private companies as a percentage o f their total assets averaged over 10 percent, a share that is similar to the tax arrears share in companies inminority state ownership, at 11.5 percent. 3.21 When both unviable and viable companies face soft budget constraints, there is little incentive for productivity improvements. This behavior o f government agencies i s particularly puzzling, as the main rationale for the slowness o f privatization was declared to be the financial betterment o f companies. This would take place in the context o f hardened budgets. Indeed, the evidence from other transition economies show that hardened budgets are the second most important determinant o f enterprise restructuring (Figure 3.5). Absent that, and inthe absence o f outsider privatization until recently, without enterprise restructuring, Romania's competitiveness has been limited. Figure 3.5 Hardened Budgets Increase Productivity Growth in Transition Economies (in percentage average annual productivity growth) _ _ ~~ outside hard budgets domestic import conpetition privatization competition Source:Adapted from Djankov andMurre11(2002). 3.22 Strengthening financial discipline o f the enterprise sector is also fundamental to maintaining macroeconomic stability. If enterprises operate profitably without dependence on either direct or implicit government subsidy, their financial difficulties will not significantly affect government budgets. Financial imbalances in the economy will not meaningfully pose a threat to the stability o f the fiscal accounts, and the central bank will be able to maintain control "CoricelliandDjankov(2002). 72Fennema and Schaffer (2003). 69 over monetary policy. To the extent that much of the public enterprise sector is unprofitable and un-restructured, however, companies will depend on budgetary assistance and quasi-fiscal subsidies to continue operating. These are a risk to sustaining macroeconomic stability. In addition, subsidization distorts resource allocation, crowds out profitable firms from the credit markets, and limits growth prospects. Large loss-making entities that are subsidized by the government will b e able to block the entry of efficient competitors, including foreign companies. Under these circumstances, resources that could have been used to invest in new projects, and which could have led to faster job creation, are instead used to finance wasteful activities. 3.23 Furthermore, without restructuring public enterprises these will remain a source o f inflationary pressure and low competitiveness since historically they have been a source of rapid wage increases. The wages of state enterprises are used as reference for wage adjustment throughout the economy, includingprivate enterprises andthe state administration, and increases have often not been justified by productivity Rapid real wage growth was partly responsible for the strong real appreciation and consequent sharp exchange rate devaluation o f 1998 and 1999 needed to restore current account ~ustainability.~~ 3.24 Table 3.2 presents estimates o f the amount of budgetary subsidization of the enterprise sector. Direct subsidies have declined from 2.6 percent o f GDP in2001 to 2.2 percent o f GDP in 2002.`i5 This is primarily due to a decline in agricultural subsidies as a result of improved harvests since 2000 and increased opportunity for export. Energy and transport subsidies have remained largely unchanged in2001 and2002. 3.25 However, explicit budgetary subsidies are not the only public resources transferred to enterprises. In Romania, there are a variety o f implicit subsidies provided to both state-owned and private companies. The transfers include budgetary credits, accumulation of tax arrears, transfers from off-budget funds, and government payment of guaranteed loans. In total, these implicit non-energy subsidies have increased from 4.1 percent o f GDP in 2001 to 5.5 percent o f GDP in2002. Inaddition, there are significant implicit subsidies from the energy sector to non- energy producers and households, o f which about two-thirds accrue to firms.76 In 2002, these subsidies were 1.1percent o f GDP in2002. l3According to its Pre-Accession Economic Program, the GoR aims to limit wage growth inthe budgetary sector to the sum o f CPI inflation and growth of GDP. For the public enterprises, wages increases are to be limitedto the sum o f CPI inflation and productivity growth. l4 The Russian financial crisis o f 1998 also triggered a decline o f confidence inRomania and necessitated extemal adjustment. 75 Until the late 1990s, state-owned enterprises in Romania depended heavily on directed bank credits to finance losses and inter-enterprise arrears, which often were driven by rapid wage increases. As the loans were seldom repaid, the financial imbalances were effectively "monetized." As a result, monetary policy could not be reconciled with the strategy o f using the exchange rate as a nominal anchor to control inflation. After 1997, soft-budget constraints were no longer financed by the central bank. Firms also began to impose hard-budget constraints on each other. Today, because the enterprises remain largely un-restructured, continuing losses have been financed with direct budget subsidy, tax arrears, unserviceable foreign loans that repaid by the government, and cheap sources o f energy. l6 Energy sector subsidies have been estimated by the IMF. See the September 2003 Staff Report for the Fourth Review Under the Stand-By Arrangement and Request for Waivers o f Performance Criteria. These estimates exclude the district heating companies that were not spun away from Termoelectrica. The estimate o f the allocation 70 Table 3.2 Subsidization of the Enterprise Sector 2001 2002 2001 2002 (billions o f lei) (%of GDP) Direct Subsidies 11 30923 32772 2.6 2.2 Energy 9633 12372 0.8 0.8 olw local government 3841 4564 0.3 0.3 Transportation 7765 8620 0.7 0.6 olw local government 2808 2875 0.2 0.2 Agriculture 11 6280 2581 0.5 0.2 Other 7245 9199 0.6 0.6 Implicit Subsidies 47476 83864 4.1 5.5 Budgetary credits, net 978 1165 0.1 0.1 Tax debt 35417 65033 3.0 4.3 Change inend o f year stocks 26901 63423 2.3 4.2 Cancelled interest and penalties 8516 1610 0.7 0.1 Energy fimd 3068 4175 0.3 0.3 Loan guarantees paid 8013 13492 0.7 0.9 olw energy sector 7435 6050 0.6 0.4 Quasi-FiscalEnergy Subsidies21 39027 25554 3.3 1.7 L o w tariffs 31763 22304 2.7 1.5 Non-payments 7264 3250 0.6 0.2 Memorandum: Total tax arrears (eop stock) 126587 190010 10.8 12.6 olw arrears o f 76 large loss makers 46944 68274 4.0 4.5 rescheduled arrears 6259 5019 0.5 0.3 Notes: 1/ Includes items classified in the state budget as transfers to mining and agriculture. 211 Based on IMF estimatedofthe allocationof subsidiesbetweenhouseholdsand firms in 2001. Source: Ministryof Finance; Ministry ofIndustryand Resources;WE3 staff estimates. 3.26 The two principal types o f implicit non-energy subsidies are tax arrears and loan guarantees, both strong indicators o f widespread weaknesses in corporate financial discipline. In 2002, subsidization via tax arrears, including cancelled interest and penalties, was 4.3 percent o f GDP. This includes debt to the state budget, social insurance funds, and local government. It includes 4.2 percent o f GDP in new arrears, which itself includes an estimated 2.0 percent o f GDP innew interest andpenalties on previously overdue obligations, and 0.1percent of GDP in cancelled interest and penalties." The rate o f implicit subsidization from tax debt is 1.4 percentage points o f GDP higher in 2002 than in 2001. However, these numbers include cancelled arrears. New arrears actually increased 1.9 percentage points o f GDP. 3.27 Activated loan guarantees increased to 0.9 percent o f GDP in 2002 from 0.7 percent of GDP the previous year. This form of subsidization, in fact, has become increasingly important. In 1995, outlays were just US$ 1.6 million. Between 1995 and 2000, they grew at an average annual rate o f US$ 40 million, while in each o f the last two consecutive years the growth has of subsidies between households and producers is based on the 2003 IMF Policy Discussion Paper (PDPl03102) by Stephane Cosse. "ThisisaWorldBankstaffestimateofthewholeeconomybasedondatafrom76largeloss-making enterprises being monitoredbythe Ministryo f Public Finance. 71 nearly doubled. In 2002, outlays for loan guarantees reached US$ 408 million, nearly half o f which pertained to energycompanies.78 3.28 Romania has tried to reduce the financial pressure on the state enterprises by rescheduling and canceling some arrears. Rescheduled arrears totaled 0.5 percent o f GDP in 2001 and 0.3 percent o f GDP in2002. Cancellation o f interest and penalties were 0.7 percent and 0.1 percent o f GDP, respectively. As demonstrated by the rise in new tax arrears, these operations ultimately do not restore financial stability. They can only encourage financial indiscipline and retard the restructuring needed to profitable business operation. As such, they pose a threat to macroeconomic stability andpromote wasteful use o f scarce resources. 3.29 This type o f subsidization is widespread inRomania. Both public andprivate enterprises use tax arrears to finance themselves. Neither loss-making nor profitable companies are excluded, though the degree o f use o f tax arrears i s inversely related with profitability. As shown inTable 3.3, majority and minority state-owned companies have had the largest tax debts as a percentage o f total assets during the last two years, averaging 30 percent and 7 percent, respectively. Of these companies which had losses inboth 2001 and 2002, the arrears averaged, respectively, 37 and 11percent o f assets. For private companies, persistent loss-makers averaged more 30 percent o f assets in tax debt, while profitable entities owed the equivalent o f 2 percent o f assets. Furthermore, the distribution by type o f arrear tends to be about the same irrespective o f the level of profitability, with debts to the state budget, social insurance funds, and local and special fund on average accounting for, respectively, 51 percent, 37 percent, and 12 percent o f the total amount owed. 3.30 Eliminating this financing option is key to strengthening the financial discipline of the enterprise sector. Infact, firms largely impose hard-budget constraints on each other: the scale of overdue trade credit among enterprises inRomania is inthe range found inwestern Europe.79In terms o f overdue debt service to banks, there i s indeed a large stock of outstanding liabilities, though mostly specific to state-owned companies. However, this appears to be a stock problem rather than a flow problem, as the outstanding debts were mainly accumulated prior to 2002 and banks now are generally imposing hard-budget constraints on enterprises better than before. 78Under the current IMFprogram, total contracting or guaranteeing o f external loans is capped at US$600million, or 1.3 percent o f GDP, for loans with one- to three-year maturities, and US$3.6 billion for loans with a maturity greater than one year. 79See Julian Fennema and Mark Schaffer, "Financial and Economic Performance o f Romanian Firms 2001-2002", a 2003 background report prepared for the Romania Country Economic Memorandum on the basis of the financial reporting o f approximately 49 thousand companies. 72 Table 3.3 Tax Arrears (% of total assets) Large Losses in Losses in at Profitablein both 2001 and 2002 Least One Year both years Total 2001 2002 2001 2002 2001 2002 2001 2002 Regie Autonomes .. 3.7 3.6 1.1 1.1 1.8 1.8 State-owned 69.4 74.6 30.7 6.1 1.1 1.3 6.4 4.3 Majority state 32.6 40.7 25.1 28.8 14.4 17.6 26.7 33.1 Minority state 11.7 10.5 10.3 11.4 1.6 1.0 7.3 7.1 Private 31.1 36.1 7.9 10.4 2.3 2.1 5.7 6.3 Cooperative 44.5 76.9 18.9 18.7 5.7 4.6 9.2 8.8 Total 27.9 32.4 9.5 10.7 2.4 2.2 7.1 7.4 otw state budget 13.7 14.7 5.2 6.1 1.1 1.0 3.6 3.8 social security 11.2 14.2 3.0 2.7 1.0 0.9 2.6 2.6 local gov. & spec. funds 3.1 3.5 1.3 1.9 0.2 0.2 0.9 1.o Source: Julian Fennema and Mark Schaffer, "Financial and Economic Performance of Romanian Firms 2001-2002." A.4. Domestic Competition Slow to Come 3.31 There are examples o f countries which have been unwilling to have rapid privatization and still have reaped the benefits o f markets. China comes to mind, as well as Poland and the Slovak Republic in the early transition period. Inboth cases, the main policy was to ensure the entry o f new firms and the use o f unviable assets in these new operations, through liquidation, leases, or asset stripping. It is more difficult to point to examples where restructuring has taken place in the absence on both strategic privatization and hardened budget constraints. One can perhaps characterize the dual nature o f Chinese restructuring inthis way, with some sectors and regions being subject to both private (and foreign) ownership and strong competition from new entrants, while other sectors remaining protected and the recipients o f state transfers. With the exception of Belarus and several central Asian countries, no one has tried to restructure without the benefits o f both hardened budget constraints and increased domestic competition. This i s what Romania has tried for some time. 3.32 As shown in Figure 3.6, increased domestic competition is associated, in the context o f transition, with faster enterprise restructuring than import competition. The main reason is that new domestic entrants can easily find their comparative advantage and fill these market niches. Incontrast, foreign companies typically haveglobal or regional strategies andtry to complement their operations abroad. Rarely do they know where the gaps inthe local market are. 3.33 Increasing domestic competition is the easiest o f all transition reforms. First, allow for the registration o f many new firms, by making the process free o f administrative burdens. Second, push unviable incumbent companies into liquidation, so that their assets can be used in new ventures. The latter feature i s more important in countries with underdeveloped financial markets, like Romania, where buying new assets on credit is nearly impossible and instead entrepreneurs need to recycle second-hand assets. 3.34 Starting a new company in Romania is not too difficult, although a lot can be improved (Figure 3.6). In particular, the commercial registry is one o f the least efficient in the region, 73 taking on average three weeks to issue a company identificationnumber. The use of notaries in the registrationprocess also makes it unnecessarily complex.80 Figure 3.6 Room for Improvement in Entry Regulation, 2004 (number of days to legally start operations of a new business) 35 , I , ... ...^..._I _I^.___ . . ._ - . I __. - ,. 30 25 20 15 10 5 France Turkey Latvia Slovakia Romania Bulgaria I Source: DoingBusinesswebsite, \~ww.worldbank.oreiI>oinebusinrss 3.35 Moreover, exit rarely happens and by the time it does, assets are often obsolete. Romania's bankruptcy law has the weakest protection of creditors of any transition economy (Figure 3.7). Once a company i s in distress, its management dictates the terms o f the insolvency process. The result is an unwilling use of formal insolvency by creditors, coupled with the general stance of government to allow arrears to accumulate. Assets stay unused for years. Figure 3.7 Creditor Rights in Insolvency Index (varies from 0 to 4, high i s better protection) I Bulgaria Czech Hungary Poland Romania Slovakia Slovenia Republic Source:DoingBusinesswebsite, www.worldbank.orpiDoingbusiness so World Bank (2004). 74 3.36 Not only do new companies have trouble getting such assets, but their managers also have trouble distinguishing themselves from other companies. Private banks are unwilling to lendlong-term credit, as recovery is difficult. The government has attempted to address this issue with the establishment o f a public credit registry in the year 2000. The idea behind it is to provide would-be creditors with sufficient knowledge o f the credit histories o f companies and their owners so they can distinguishbetween good and badrisk. As work inprogress, the registry has already had some impact, with muchto be done. B. GROWTH THE ENTERPRISE IN SECTOR B.1. Sales Growth of Enterprises 3.37 The issues summarized here resulted in the shrinking o f the Romanian economy in the 1990s, with growth taking place only inthe last four years. Table 3.4 compares the sales growth performance o f the Romanian enterprise sector with that in other six transition economies. Romania has the sharpest declines inproduction in 1992-95 and 1996-1999. Bulgaria is the only other country that records declines inboth periods. Since the year 2000, Romania has rebounded strongly, with average annual growth of over 5 percent. However, similar growth has been recorded inBulgaria, and the Slovak Republic, which suggests that transition laggards are inpart catching up with early reformers. Table 3.4 Sales Growth in Enterprises, (%, average over the period) Bulgaria Czech Rep Hungary Poland Romania Slovakia Slovenia 1992-1995 -11.1 1.4 0.8 1.1 -13.7 -6.1 3.6 1996-1999 -5.2 1.1 3.6 6.9 -5.8 6.4 3.8 2000-2002 5.1 3.4 4.6 2.4 5.2 4.9 2.7 Source: Private Sector Vice Presidency, World BankGroup. 3.38 Another heartening feature o f recent growth in Romanian enterprises is that it is accompanied by some adjustments in the labor force. This has resulted in even higher labor productivity growth, at over 7 percent on average since the year 2000. Only the Slovak Republic rivals this growth. 3.39 Still, the missed opportunities in the early transition years put Romania at real GDP equivalent to 87 percent o f its 1989 level. This is the lowest level o f any central and eastern European economy, bar Bulgaria (Figure 3.9). 3.40 In the 1990s total factor productivity (TFP) growth has accounted for nearly all of economic expansion.8' Between 1991 and 2002 the contribution from labor was negative as a result o f labor shedding. Due to periods o f contracting investment followed by weak capital accumulation, on average there was no contribution from capital. Growth was entirely due to more efficient use of labor and capital, with average TFP growth o f 2 percent over the whole *' TFP growth i s measured as a residual source o f growth once labor and capital contributions have been accounted for. Generally speaking, the size o f the physical capital stock is especially difficult to quantify. In the case o f Romania there is also a large informal economy which may lead to inaccurate estimates o f output and labor and capital input. TFP estimates, therefore, canbe highly sensitive to measurement error. 75 period. This is less than half o f the TFP growth experienced in central Europe: the Czech Republic, Hungary, Poland, and the Slovak Republic recorded TFP growth o f close to 4.5 percent over the same period.82 Again, this growth was largely explained by shedding excess employment but since the mid-1990s new fixed investment started coming inas well. Figure 3.9 2002 GDP Relative to the 1989 GDP Level Bulgaria F " i a Hma Cz& Slovak km~ary Slovma Paland Repblic W i c Source: EBRD(2004). B.2. Total Factor Productivity 3.41 TFP growth in the Romanian enterprise sector has risen to 3.3 percent in more recent times-1 999 to 2002-but still trails all transition economies but Croatia and Slovenia (Table 3S). Whether productivity growth can accelerate depends on faster investment growth, particularly FDI. In the long-run, efficiency can only increase from the adoption o f new technologies, including managerial techniques and advanced capital goods. To achieve this, the incentive structure o f the economy needs to be changed to encourage greater domestic and foreign investment. This will also require reforms that encourage more domestic savings and further financial restructuring and a business climate that attracts more long-term foreign capital to improve the intermediation between savers and investors. This will lead to a reallocation of resources to the most productive sectors o f the economy. The example o f other transition economies-first Hungary in 1991-1995, then the Czech Republic in 1994-1998, and recently the Slovak Republic-shows that investment can increase significantly in a very short period, if the business environment improves. 82Djankov andMurre11(2002). 76 Table 3.5 Accounting for Economic Growth Average annual growth rates (percentages) Contributions to growth (percentages) GDP Labor CaDital Labor CaDital Total factor Droductivitv 1999-2002 4.1 -0.5 3.5 -0.3 1.2 3.3 1991-1998 -0.5 -1.8 -2.4 -1.2 -0.8 1.5 1991-2002 0.7 -1.5 -0.8 -1.0 -0.3 2.0 Source:World Bank staffestimates. c.THEREFORMAGENDA C.l. Privatizing to Outside Strategic Investors 3.42 Privatizing the enterprise sector and imposing hard budget constraints is central to Romania's economic restructuring. The best mode o f privatization is to outside strategic investors. Recently, Romania started to move in this direction. Pace o f reforms needs to be accelerated since the protracted enterprise reforms and failed privatization methods of the pas have left a legacy o f a large reform agenda going forward.. This process needs to be carried out decisively and ina transparent framework. C.2. LicensingRegimes 3.43 In mid-2003, the government introduced the silent consent rule that had immediate impact on 480 licenses. The new rule allows companies to undertake activities if the relevant government authority ha snot responded within 30 days. Also, new legislation i s to be accompanied by a regulatory impact assessment, inline with OECD standards. Finally, statutory limits on the processing o f licensing requests have been mandated. All these changes will undoubtedly reduce the administrative burdenon business. Much remains to be done. Businesses continue to complain from burdensome regulation (Figure 3.10). Figure 3.10 A Quarter of Businesses Identify Licenses as a Major Problem (% of businessesthat say business licensing i s a moderate to major obstacle) 35 30 25 20 15 x) 5 Poland Estonia Slovenia m a r y Romania Slovdc Czech Bulgaria Republic Republic Source:BEEPS2002 survey. 77 C.3. BusinessEntry 3.44 Regulatory reform inthis area i s not difficult and yet inRomaniathis regulatory reform is wanting. Several other EU accession countries have recently revised their regulation to allow quick entry, Hungary in 2002, Slovakia in 2004. In 2003, Turkey reduced the number o f procedures necessary to start business operations from 13 to 6, and the number o f days from 38 to 7. Reform is also taking place in the registration processes o f many European Union countries, including Finland, Greece, Italy, Netherlands, and Portugal. Italy, for example, has removed business registration from its courts and placed it in administrative units. In all cases, these reforms have lowered the current and expected time needed to register new busine~ses.~~ A recent study for the European Commission found that "Considerable progress has been made in improving the process o f administering business start-ups." In particular, many EU Member States have: 0 Improved the administrative efficiency o f current procedures by introducing statutory response times ("silence is consent'' rules). This reform has been undertaken in Romania although business surveys do not uniformly register its effects. 0 Introduced Intemet sites to provide information, improvements in the availability o f information for entrepreneurs; the mapping o f all procedures and licenses; the provision o f information electronically; and consolidation o f information into a single source. None o f these steps are taken inRomania. 0 Reduced the scale and complexity o f the documents required to establish new businesses by using single registration numbers, and notification and self- certification rather than authorization. This has been stalled on numerous occasions inRomania, with no results insight. 3.45 Ifreform is revitalized and Romania implements these reforms, it should not stop here. Registration should be made into a purely administrative process and available as an electronic service. The current organization o f Romania's registration system is low in efficiency. Its location inthe courts and involvement byjudges in individual registration decisions are contrary to good European practices. This also damages economic growth by reducing the capacities o f the commercial courts to deal with legal disputes. A better option is an expert registration agency. The national registry database should be publicly accessible, updated continually, and eventually accessible by Intemet. 84 83 European Commission, Enterprise Directorate General (January 2002) Benchmarking the Administration of Business Start-ups, Final Report. Prepared by the Centre for Strategy and Evaluation Services (CSES), Brussels, p. 25. 84 For example, the main recommendations relating to the European Union's First Directive o n company law consists partly of the need to accelerate the filing and disclosure of company documents and particulars by the use o f modem technology. 78 3.46 Regulatory reforms are indeed a first step towards a more competitive business environment. However, reforms need to be sustainedto provide a business environment o f entry andexit. C.4. Labor Regulations 3.47 Romania has revised its socialist-era Labor Code, making it in many areas more restrictive than the original. Especially burdensome is the area of conditions of employment, where firms inRomania are at a significant disadvantagerelative to their peers ineasternEurope (Figure3.11). Figure 3.11 Employment Regulation Index (varies between 0 and 100, highervalues for more rigid regulation) Czech Slovak Estonia Hungary Bulgaria Romania bland Slovenia Republic Republic Source: DoingBusinessproject, www.worldbank.ordDoingbusiness. 3.48 Businesses report that rigid employment laws are a significant obstacle to their performance. In cross country comparisons, Romania has some of the most rigid conditions of employment, with tight restrictions on hours worked, night and holiday work, and a relatively high minimum wage. Several details in the labor regulations create perverse incentives for workers. For example, one business interviewed reported that labor regulations create problems inmanagingabsenteeism. 3.49 Another example i s the regulation requiring periodic automatic increases in wages by years worked, so that salaries are in effect based on seniority, not productivity. This is a significant problem in labor-intensive industries like textiles, where a very low turnover of workers and salaries continually creep up. 3.50 Rigid labor regulations encourage informality. A common practice in order to avoid social security payments and taxes i s to pay official wages at the minimum wage threshold, and 79 to pay additional cash wages directly to employees. This practice was prevalent inthe late 1990s, when an estimated 50percent o fthe workforce inprivate business was paidinthis manner. C.5. Bankruptcy 3.51 Lenders also face problems with bankruptcy laws. In a good case scenario it takes four years and costs around 8 percent o f the value o f the estate to close a business through bankruptcy proceedings, Interviews with insolvency practitioners suggest that it takes five years to resolve bankruptcy when the main creditor is a private bank or company. This makes Romania the second least efficient country among all central and eastern European economies, trailing only the Czech Republic (where it takes over eight years). As a result most banks report that they foreclose before bankruptcy-all stated that bankruptcy was rarely used to enforce on default. About 90 percent o f the bankruptcy cases that reach the courts-an estimated 4,000 cases a year-are brought forward by the state as a means to collect on overdue taxes. 3.52 Revisions to the bankruptcy law were proposed by the government in early 2003, but have yet to be adopted by Parliament in full. Government Ordinance 38/2003 introduced some minor revisions, most importantly an article that stays management during the insolvency process. This is a setback from a creditor rights perspective as the incumbent managers have more opportunity to strip assets incases where liquidation is the most likely outcome. 3.53 A draft law has been proposed to parliament and will have significant beneficial effects on improving creditor rights in Romania if passed. Among the most salient features are a reduction o f the period o f appeals-from 30 to 10 days; the introduction o f subpoena for creditors published in newspapers, which will reduce the time o f notification; removing the requirement for the presence o f a court clerk during the creditors' meeting; and a provision for out-of-court restructuring. All these features would undoubtedly make lending inRomania a less risky proposition. 3.54 Two other areas o f judicial reform merit consideration. First, the application fee for entering bankruptcy is the highest o f any country inEurope, at 5,000 euro. It i s unclear what the rationale for such a fee is. Second, under the current structure o f the Romanianjudiciary, one can become a bankruptcyjudge fresh out o f law school, at 23 years o f age and with no experience in either insolvency practice or management and accounting. Incontrast, Italy and France, the two countries used as the origin o f the current judicial system, have both implemented reforms that ensures that judge shave some experience in handling commercial matters prior to taking the bench. Such reform would take time to develop and is likelyto have a large positive effect on the quality o fjudgments. C.6. EfficiencyandPredictabilityof CommercialContracts 3.55 Enforcement o f contracts in Romania faces more serious challenges than it may first appear. As described in Section 1, judiciary reform is highin the agenda in Romania's process towards EU accession. In late 2003, important initial steps have been taken including adopting Judicial Reform Strategy 2003-2007 and adopting constitutional amendments overhauling the 80 legal framework for judicial reform. Regulatory and organizational changes have to be fleshed out and more importantly actions need to be taken to develop a functional, transparent, independent judiciary system. These changes will improve enforcement o f contracts. There is evidence that deficiencies in the judiciary are ~ignificant.'~ A recent World Bank study reports that these deficiencies have for the most part discourage business to resolve disputes through litigation.86Only very large enterprises seem to be more litigious compared to medium and small business. In addition, the large and dominant presence o f public enterprises combined with the practice o f barter settlement further reinforces these findings since firms have these alternative mechanisms to settle disputes. Hence, the low number o f days o f 240 days to enforce a simple debt contract inRomania, compared with an average o f 252 inCEEC reflects deficiencies rather than strengths since only large companies use the courts to settle disputes andthejudiciary lacks transparency (Chapter 3). Furthermore, in2003 the Emergency Ordinance 58/2003 reintroduced two levels o f appeals for all types o f commercial disputes and hence represents a step backwards inthe efficiency of commercial cases. Data of January 2004, indicate that debt collection cases now can take on average one year to resolve. In addition deficiencies in the enforcement o f contracts protecting property rights on land in agriculture is a major stumbling block for agricultural transformation discussed inthe next section (Section 4).87 3.56 Clearly judiciary reform is central. Recognizing these deficiencies, the government has put judiciary reform high on the agenda in Romania's process towards EU accession. In late 2003, important initial steps were taken, including the adoption o f Judicial Reform Strategy 2003-2007 and constitutional amendments overhauling the legal framework for judicial refom. Regulatory and organizational changes have to be fleshed out and more importantly actions need to be taken to develop a functional, transparent, independent judiciary system. Implementing these reforms are essential to improve enforcement o f contracts, and thus for growth and competitiveness prospects. In addition, further streamlining o f contract enforcement is also needed. Four successful reforms inother countries are o f particular relevance for Romania: 0 Establishing information systems on caseload andjudicial statistics, as in the Slovak Republic can help identify the courts primary users andthe biggest bottlenecks. 0 Taking out o f the courts transactions that are not disputes-in particular the registering o f new business entities-will free up resources for commercial litigation. Because such reform will require new legislation, in the meantime government can reorganize the workflow in the courts so that clerks, not judges, are responsible for company registration. 0 Simplifying procedures. The priority is to reduce opportunities for defendant appeal and delay. Reducing reliance on written procedures and legal justification are also necessary. More extensive reforms include introducing summary proceedings to alleviate court congestion (as has recently been introduced in Mexico). Default 85 Romania: A Public Expenditure Review, 2002, The World Bank and Regular Reports on Romania by the EC, 2002and2003 86 Building Market Institutions in South Eastern Europe: comparative prospects for investment and private sector development, forthcoming in2004, The World Bank. ''See Chapter 4 inVolume 2 of this report. 81 judgments-automatic if the defendant does not appear in court-will also significantly cut delays. 3.57 Reforming the enforcement o f judgments. Currently, judicial enforcers in Romania are highly regulated, and the service is an expensive part of the commercial dispute resolution process. The minimumfee i s ROL 3 million, and it rises steeply with the amount of the case. 4. IMPLEMENTINGAGRICULTURAL TRANSFORMATION 4.1 Inthe early 1990s, after the first democratic elections, Romaniaembarkedonaprocess of general economic reform, which also included specific and ambitious reforms inthe agriculture and food sector. The reform program envisaged the transformation o f agriculture to a sector based on the principles o f private ownershipo f land and other agricultural property. The aim was to create a market-oriented and internationally competitive agriculture. Progress was rather modest until 1997. Since 1997 Romania has made significant, if sometimes erratic progress in implementing and maintaining the sectoral reform program. In 2000, the substance o f reform began to shift towards measures aimed at EU accession. The overall process, however, has been slower and the results are less consistent than in other Central European countries in the final phases o f the EU accession process. Further substantial reform and changes in agricultural policies are necessary before Romania can proceedwith accession to the EuropeanUnion. A. AGRICULTUREAND FOOD THE ECONOMY SECTORALPERFORMANCE IN AND 4.2 Romania i s a rural, agrarian country with 45 percent o f the population living inrural areas Figure4.1 Romania:ShareofAgriculture and 36 percent o f the labor force employed in inGDPand Labor Force agriculture. Agricultural development has followed an unusual trend since 1990: while the 50 I 1 share o f agriculture in GDP declined from 29 percent in 1990 to 18 percent in 2000, in line with the traditional trends in market economies, the share o f agricultural labor paradoxically increased from 30 percent to 40 percent o f total employment (Figure 4.1). The 4.8 million people I I working in Romanian agriculture today 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 constitute an astonishing 72 percent o f 1AgShar -EmpShar/ agricultural labor inall the 15 EUcountries. Source:NSI andMinistry of Agriculture. 4.3 Romanian agriculture did not suffer a dramatic collapse after 1989: agricultural production went through moderate ups and downs during the transition and in 2001 it was basically at the same level as in 1989-90. Agricultural growth since 1990 has largely paralleled the total growth indomestic product andthe growth o f industrial product. 4.4 The product mix has remained fairly stable since 1990 at 60 percent crops and40 percent livestock. Over the long term the Ministry o f Agriculture aims to achieve a more balanced mix, raising the share o f livestock to 45-50 percent of agricultural product, as in the late 1980s. This chapter was prepared by Csaba Csaki, Holger Kray, and Zvi Lerman. Henry Gordon provided useful contributions to the section on rural development. 83 Among crop enterprises, cereals and legumes consistently account for more than 60 percent o f the cropped areas, while technical crops represent over 10 percent. The share o f land cropped to cereals and legumes increased by 5 percentage points between 1990 and 2000 (from 62 to 67 percent). The area under technical crops increased by 3 percentage points (from about 10percent in 1990 to 13 percent in 2000). These increases have been made possible by the decrease in the second largest user o f sown land-fodder crops, whose share dropped from 21 percent in 1990to 13 percent in2000. 4.5 Agricultural productivity in Romania leaves much to be desired. Figure 4.2 Romania's Yields Relativet&15 The low productivity o f labor is underscored by the fact that 44 percent o f Romania's labor force employed in agriculture contributes only 13 percent 7o percent of -15 o f total GDP (2001 data). The low I I productivity o f land is manifested in the generally low yields, which achieve at most 50 percent o f the corresponding EIJ-15 yields (Figure 4.2). In livestock enterprises, milk yields hover around 3,000 liters per cow per year compared with 5,800 in EU-15. Based on the aggregate value Cereals Sugar Rape see6unflower Milk Wine o f production, Romanian agriculture ource: EUand OECD Statistics. generates about 500-600 per hectare, compared with 2,000 per-hectare in EU-15, and about 1,500 per worker, compared with 21,000 per worker in EU-15 (1999 data). A weighted ranking o f EU-15 and 9 new EU members by the yields o f milk, cereals, and sugar puts Romania at the very bottom o f the scale (together with Bulgaria). 4.6 The poor performance o f Romanian agriculture cannot be blamed entirely on the Figure 4.3 Change of Output, Labor, and availability o f farm inputs. While Productivity 1990-2000: Food Industry and application o f fertilizers and herbicides ManufacturingIndustries literally collapsed between 1989 and 2001, the number o f tractors, mechanical ploughs, and cultivators today is greater than it was in 1989 and certainly in the crisis years 1990-91. 4.7 Contrary to stagnating output and low productivity in agriculture, the food industry has achieved significant growth in -100 1 recent years. The volume o f production in Food Manufacturing 2000 reached about 130percent o f the 1989 Source: NSI. level, while the manufacturing industries had shrunk by 2000 to 60-70 percent o f their level in 84 1989 (Figure 4.3). The number o f employees in food industry decreased by nearly 40 percent since 1990, roughly in line with the decrease in the national economy and somewhat more than inthe manufacturing industries (where the number of employees dropped by 55 percent). The combination o f growing output and decreasing labor force naturally has had a strong positive impact on productivity in the food industry, which more than doubled between 1990 and 2000, increasing at a mean annual rate o f 7.8 percent. As a sector, the food industry accounts for about 10 percent o f total national production and 7 percent o f gross value added (averages for 1990- 2000)-somewhat less than the share o f primary agriculture. B. STATUS OFSECTORALREFORMS 4.8 Reform o f the Romanian agriculture and food sector and the evolution o f the agricultural policy framework have passed through several phases since the beginning o f the transition. During the early years (up to 1997) important sectoral reforms were initiated, but their implementation was slow. The most important landmark o f sectoral reforms during this period was the 1991 Land Law, which resulted inthe restitution o f land ownership and land use rights to private holders and effectively dismantled the cooperative farming system. Policies in that period aimed to minimize the difficulties created by the changing economic environment both inside a ~ outside the country; relatively little attention was paid to the need tc. ;dorm the basic d institutions and enterprise structures and to adjust the policies themselves to the conception. The support program in agriculture was heavily biased toward large agricultural units and almost exclusively benefited state-owned farms. Small household farms could have access to subsidies only via integrators (state marketing intermediaries), and these purchased their output at fixed "state-guaranteed" prices, which could be attractive or not depending on the specific point in time and the specific location. As a result, input andoutput markets remained highlydistorted. 4.9 Driven by economic crisis and a growing consensus that more sustainable reforms had to be put inplace, Romania adopted a radical reform package in 1997, which marked a fundamental adjustment o f economic policies. The implementation o f the 1997 Stabilization Plan accelerated the overall economic reforms, as well as the specific reforms in agriculture. Agricultural price controls were eliminated, tariffs were lowered, directed credits were discontinued, and the privatization process was jump-started. The emphasis on hard budget constraints and increased financial discipline aimed to create the conditions for the development of a sound and sustainable sector. Agricultural reforms resulted in a new form o f support-the input voucher system- which was designed to redirect subsidies away from the state farms and to increase farm output through better access to inputs. Reforms also included the privatization o f some state input suppliers and Banca Agricola, and more recently, the privatization o f state farms. 4.10 The uneven course o f agricultural policy development, however, still remained inrecent years. The voucher scheme came to be seen more and more as a social welfare scheme and was blamed for the unsatisfactory increases in agricultural production. This led to the adoption o f a new policy framework in 2000, which aims to boost production by again redirecting support to larger farms and increased support to agriculture significantly. Nevertheless, under this overall new direction, the policy framework has remained fluid: the system is characterized by frequent changes, annualbargaining, and unpredictability. 85 B.l. Distortive Agricultural Support Policieswith HighLevels of Subsidization 4.11 Romania's agricultural policy has continually relied on the application o f direct and indirect payments as a measure for supporting the income o f primary agricultural producers. However, even though the hndamentals have remained relatively unchanged, the agricultural support policies have undergone repeated changes and adjustments since 1997. The support system in effect was re-written each year, with additional changes sometimes introduced in the course o f a given year, as in2002 and2003. 4.12 Romania's agricultural support policy explicitly favors large-scale farming operations. Government ordinance adopted in June 2001 stipulated that only relatively large farms, e.g., with at least 110 hectares o f crop land or 15 dairy cows (legally defined as "commercial" farms), would be entitledto the full range o f agricultural subsidies. Farms that did not meet the stipulated size thresholds (legally defined as "family" farms) were not entitled to any subsidies (other than fkee extension services). Intense public pressure and unfavorable reaction from the European Union led the Parliament to amend this ordinance inApril 2002 and to add a provision that allowed (unspecified) support payments to "family" farms willing to increase their market sales. Although the law opened a door for eliminating the large-farm bias in subsidies, no comprehensive implementation regulations have been issued so far and the support to small farms remains very limited (see some details below). Moreover, the Parliament did not intervene to eliminate the complex documentation and formal registration requirements introduced as part o fthe eligibility criteria, although these requirements cannot be met by the fast majority o f Romanian peasant farmers. It should be stressed in this context that the farm size thresholds introduced in Romanian legislation are much higher than those foreseen by CAP eligibility criteria and that the Romanian definition o f "farm operation" for the purposes o f subsidy payments i s not harmonized with the EU definition. Until the end o f 2002, less than 15,000 farms were registered by the Ministry o f Agriculture in the framework o f the new legal requirements, o f which 12,918 are commercial farms and 1,721 are family farms. The registered farms cultivate 3.2 million hectares o f arable land, which is about one-third o f all arable land in Romania (9.4 million hectares). 4.13 The main support instruments applied in the last few years (although with changing emphasis) can be divided into direct income support measures and indirect support programs: Direct income support measures Price support with price supplements for a number o f major commodities. Financial support for inputs in the form o f free fertilizer, reduced prices for certified seeds, and price supplements for the use o f other inputs such as fuel (this program replacedthe former voucher system). Investment grants for up to 55 to 70 percent o f the cost o f agricultural machinery. Other direct payments or compensationto farmers. Credit subsidies and investment support. Indirect support programs Support for general agricultural services, such as animal and crop breeding, water users associations, land reclamation, etc. 86 0 Export subsidies. 4.14 The 2003 budget o f the Ministry o f Agriculture projects a total support volume o f 14billion Romanian lei (ROL) (Table 4.1). Most o f the sectoral support i s earmarked for the crop Table 4.1 Structure of Agricultural sector (72 percent) with another 25 percent going Support Program in2003 to the livestock sector (the remaining 3 percent go (million) RoL Percent to export measures, forestry, and agricultural Direct Support research). The crop sector thus enjoys a Measures 13675.50 97 disproportionately high support-more than its Price Support 5101.00 36 share in agricultural product. The relatively small Input 7140.60 51 allocation to the livestock sector is a tangible OtherDirectPayments 685.00 5 reflection o f the large-farm bias, as most o f Creditand support 749.90 5 livestock production is in small household farms, Indirect Support not in large legal entities. Of the various support ~easures 328.6 3 programs in Table 4.1, commodity price support 103.60 1 supplements and financial support for inputs are Export Subsidies 225.00 2 the two main categories, accounting for 87 percent Total 14005.10 100 o f all support programs. The composition of the Source: Ministryof Agriculture. support programs mainly serves as a conduit for income transfer and social purposes rather than a tool for fostering competitiveness and productivity enhancement. Figure 4.4 Measures of Agricultural Support, Figure 4.5 Producer Support Estimates for 1991-2002 Selected Countries, 2001 Measures of Agricultural Support 1991-2002 -g 40 -3 ' 35 a % m ebudget %GDP *E C 30 a .-2 = 25 8 20 6 15 5 10 4 n 5 $ 2 0 n ~~ Source: Ministry ofAgriculture, PSEfrom OECD. Source: OECD. 4.15 The 2003 budget and subsidy program envisages a further significant increase in support to agriculture. Expressed in U S dollars, the 2003 program is close to US$450 million, up from about $350 million in 2002. This support program represents about 4 percent o f the total state budget (down from 10percent in 1992-93) andless than 1percent of GDP (down from 3 percent in 1992-see Figure 4.4). Like the modes o f intervention, the level of overall support to agriculture has fluctuated widely during the transition period. The level o f support as expressed by the Producer Support Estimate (PSE) was relatively modest during most o f the 1990s, but it 87 definitely seems to be higher since 1998 than previously (Figure 4.5). The liberal policies introduced in 1997 (when the PSE dropped to a 10-year low o f 3 percent) were short-lived and support to agriculture has increased significantly since then. Thus, the average PSE during 1991- 96 was 13 percent, whereas after 1998 it rose to 20 percent-25 percent. The level o f support varies significantly by commodity. The sugar, milk andpoultry sectors receive the highest levels o f support at 64 percent, 46 percent, and 34 percent respectively. On the other hand, oilseeds hardly get any support, while the pork sector is actually taxed. 4.16 An international comparison shows that Romania's PSE is close to that o f USA and is significantly lower than the OECD and EU averages (Figure 4.6). At the same time, it is significantly higher than for most o f the other CEFTA and EU accession countries (with the exception o f Croatia and Slovenia). The results o f this comparison are somewhat surprising because the total value o f Romania's agricultural support program i s not excessive when compared to other CEFTA countries. One has to remember, however, the significantly lower productivity and output levels inRomanian agriculture, which result inmuch higher PSEs. B.2. Light and InconsistentTaxation inthe Agricultural Sector 4.17 There is no tax on agricultural land and no tax on small farm income inRomania. Farms registered as legal entities, however, pay 25 percent tax on profits. The sweeping tax exemptions for agriculture were htroducedin 1990 as a compensation for the farmers' economic difficulties and the government's limited ability to pay direct subsidies. The failure to impose land tax was moreover associated with the uncertainties o f land ownership during the "reconstitution" phase inthe early 1990s. At that time, the tax-exemption decisionhadits merits andit played a role in improving the financial situation o f farmers. 4.18 There have been some attempts over time to subject agriculture to general taxation principles. Tax on farming was introduced in 1994, but because o f difficulties with income reporting by small farmers the tax was assessed on the basis o f the area owned or used and was in effect an agricultural land tax. Only the income from especially lucrative farming activities (flowers, vegetables, mushrooms, seedling nurseries) had to be actually reported. Income from livestock remained exempt from tax. The tax rate was initially 15 percent of normative income and it was lowered to 10 percent during 1994-1996. The normative tax assessment and collection never really took off, and taxation o f farming activities was suspended in 1997, after the formation o f a new coalition government. When general income tax was introduced in 2000, income derived from fanning (except flowers, vegetables, and mushrooms) was exempted and has remained so ever since. 4.19 In2001, the government imposed a landtax on holdings larger than 10hectares, because the winningparty inthe 2000 elections hadpromised not to tax poor small farmers. However, the parliament rehsed to allow any taxation o f agricultural land and annulled the government decision. After the failed attempt to tax farmers and perhaps in view o f the next election campaign, the government has announced suspension o f the tax until2005. 88 B.3. FluctuatingTrade Policies 4.20 Romania's agricultural trade policy has evolved through various stages since the early 1990s. The major determinants o f agricultural trade policy developments have been the WTO agreement, commitments made within the framework o f the A S A L (Agricultural Sectoral Adjustment Loan) agreement with the World Bank, evolving relations with the EU, and Romania's CEFTA membership. 4.21 WTO. Under the WTO agreement Romania was granted "developing country" status. This status is quite unique among Central and Eastem European countries and it permitted negotiating very high tariffs, which became the main instruments for protecting ago-food producers. The high tariff ceilings, however, are very rarely This tariff protection created a wall around Romania and has had negative consequences for domestic efficiency in ago-food products. Developing country status also allowed smaller reduction obligations for average aggregated agricultural tariffs and a higher level o f domestic support allowances. If Romania were forced to give up its developing country status before EU accession, the level o f domestic support might become an issue, although at current levels it is not excessive when compared to other countries inthe region. 4.22 EU Agreements. Romania's Association Agreement with the EU became effective in 1995. This agreement provided preferential quotas and tariff reductions for most livestock products, wheat, and vegetables. A special agreement was adopted for wine. In 2000, Romania reached a Double-Zero Agreement with the EU on agricultural trade liberalization. A number o f products (poultry, meat, cheese, wine, fruits, grain) received full or significant tariff reductions and a schedule for increasing trade quotas with the EUwas approved. Bothparties agreed on the elimination o f export subsidies for commodities within the framework o f the Double-Zero- Agreement inbilateral trade. 4.23 CEFTA. In 1997 Romania became a member o f CEFTA and adopted preferential treatment o f a number o f agricultural products in CEFTA-related trade. CEFTA countries are Romania's major trading partners in agricultural products. The limited competitiveness o f Romanian food and agricultural products has become most evident inthese trade relations, which hasresulted inthe application o f a safeguarding clause and increasing import tariffs temporarily for a number o f products during the last few years, especially against Hungary. With the EU enlargement in 2004, CEFTA preferences will be transferred to the Europe Agreement and Romania will continue to receive the benefits it previously had with the five former CEFTA members (now EUmembers).. The acceding countries will then benefit from all EUconcessions to Romania. 4.24 Romania's agricultural trade policy can be characterized by change and considerable volatility throughout the transition period. Temporary exceptions and reductions benefiting particular exporters and importers were a common practice during this time. Inrecent years, the volatility took the form o f frequent use o f safeguards and temporary interventions on behalf o f domestic producers. 89 B.4. Incomplete Transition inthe FarmingSector 4.25 Landreform and the dismantling o fthe inefficient large-scale farming structures has been one o f the most difficult and complex tasks o f transition. Romania has made sweeping changes inland tenure and in farm structure, but the task of creating a farming system able to produce internationally competitive products has yet to be completed. B.4.1. Land Reform Virtually Completed 4.26 The agrarian transition had to cope with the dual task o f reversing the post-1945 expropriations and decollectivizing the cooperatives which had been the dominant farm structure since the 1950s. The first post-communist land law passed in 1991 (Law 18/1991) laid the foundations for a dual land-reform mechanism: (a) the restitution o f ownership rights to former owners whose land had been expropriated by the state after March 6, 1945 and (b) the return o f use rights to individuals whose land had been managed for years and decades by cooperatives. This dual mechanism is referred to as reconstitution of land ownership rights, and it is complemented by an additional "constitution" mechanism that prescribes distribution o f land to landless farm employees and landless young families willing to start farming. The Romanian land reform i s thus distinguished by a combination o f land restitution and land distribution elements, which sets it apart from the strictly restitution-based and thus less equitableprocedures adopted inother Central European and Baltic countries (with Hungary a notable exception). 4.27 Romanian land reform had to deal with reconstitution o f 10.5 million hectares out o f a total o f 14.8 million hectares in use; the process involved 4.7 million claimants, mostly in rural areas but many also among urbanresidents. This was a huge task, and completion rates o f about 97.5 percent as o f early 2004 are a important achievement for the country. In the process o f reconstitution, cooperative members got their land back in the form o f physical plots, whereas ownership rights in state land were initially restituted by allocation o f shares in the former state farms. The situation o f the "shareholding" restituents was gradually equated to the situation o f former cooperative members, and in 1999 it was finally resolved that all land owned by individuals within the holdings o f a former state farm should be restituted in physical form and the owners should be free to dispose o f it at their discretion. 4.28 The changes in land tenure between 1991 and 2001 are presented in Table 4.2. Most o f the land today is in private ownership o f individuals. One-quarter o f agricultural land is owned by public authorities (the state and local government) and most o f it can be ultimately privatized (the "private domain" component). Currently, however, only 1 million hectares o f "private- domain" state land is under active consideration for transfer to private use (but not private ownership!) through leasing and concession arrangements with the State DomainAgency. 90 Table 4.2 Landby Ownership Type 1991-2003 1991 2000 2003 thousandha thousandha percent Agricultural surface 14,758.4 14832.2 14836.5 Private ownership 1,876.3 14281.5 14307 Judicial persons 4,301.1 3,717.7 Physicalpersons 1,876.3 9,980.4 10,584.3 Public property 12,882.4 550.6 529.5 Source: Ministry of Agriculture B.4.2. Shqt to Predominantly Individual Farming 4.29 The landrestitution and distribution laws were complemented with laws focusing on new organizational forms inagriculture. After more than a decade of transition, Romanian agriculture i s characterized by farms of four organizational forms. 0 The individual or household sector consists of 4.2 million farms o f less than 2.5 hectares on average, which inaggregate control nearly 70 percent o f agricultural land (up from 15 percent in 1989). This is the major component of Romanis's private agriculture. 0 Private corporate farms, i.e., legal entities that include so-called formal or legal associations created since 1991 as successors o f the fomier cooperatives and also private agricultural companies created more recently by various mechanisms. The legal associations and other companies jointly control around 10 percent of agricultural land (down from 67 percent controlled by cooperatives in 1989). There were somewhat less than 4,500 legal associations in 2001 averaging around 400 hectares each; no detailed statistical data are available on the subgroup o f private agricultural companies. 0 Family associations or informal associations. These are spontaneously created voluntary associations o f individual farmers that are not registered as legal entities. There are about 6,500 such family associationsinRomania and they are much smaller than the legal associations, averaging about 120 hectares and cultivating in total 5 percent o f agricultural land. Family associationsrepresent a kindo f "hybrid farming", as many individuals (about 30 percent according to some estimates) choose to entrust part of their land to the family association while continuing to farm the remainder within the family. 0 Inadditionto thesethree components ofprivate agriculture, Romaniastill has asector of state-controlled agricultural companies-the successors o f former state f m s - which hold about 15 percent o f land, not all o f it under active cultivation. The statistics on the state sector i s ambiguous. 91 4.30 The dramatic changes inthe land use pattern during the decade of Figure4.7 Change inStructureofAgricultural transition are illustrated in Figure 4.7. LandUse 1989-2001 The changes in land tenure are reflected inthe structure of agricultural production as it i s observed today. Individual farms loo%- control the bulk o f agricultural land and accordingly produce 87 percent of agricultural output. Private corporate farms (legal associations and other private companies) contribute 10percent while the state sector produces only 3 percent o f agricultural output. This structure is the end result of an " I " evolutionary process that began in the 1989 2001 early 1990s with the onset of landreform wrce: Ministryo fAgriculture,NSI. inRomania. The share of the state sector in agricultural output is steadily shrinking, while the share of the private sector is increasing- mainly due to the growth of individualfarming. B.4.3. LaggingPrivatizationof Former StateFarms 4.31 Back in 1989 Romania had 411 state farms that controlled 2 million hectares of agricultural land-1 4 percent oftotal agricultural land and 50 percent o f state-owned agricultural land. The strategy for transition to a market-oriented agriculture naturally includedprivatization of state farms. As a preliminary step to privatization, the state farms were technically transformed in 1990-91 into joint-stock companies. The assets of the former state farms became the property o f the new commercial companies and the state continued in the role of the sole shareholder (at least initially). The actual management of the former state farms remained inthe hands of state-controlled holding companies-the State Ownership Fund (SOF) and five Private Ownership Funds (POF). The formal reorganization, however necessary from legal considerations, did not convert the former state farms into privately owned, profit-orientedunits. 4.32 In December 1999, after years of deliberations and debates, the government finally decided on a policy o f selling the state-farm assets to private investors while leasing the state- owned land to the new operators and other interested users. According to this policy, the state land had to be privately farmed, based on a 49-year concession or a long-term lease contract betweenthe administrator of state land and a private investor. This transfer of land use rights may or may not have included the sale o f shares in the state farms assets, which may be privatized separately through an auction or another mechanism. 4.33 As part of the new policy, the responsibility for state-farm privatization was transferred from the largely unsuccessful SOF to the State Domain Agency (SDA) in the Ministry o f Agriculture. So far SDA has sold to private investors about 280 former state farms (Table 4.3). Another 80 are considered good prospects for privatizationinthe near future. The remaining 382 92 former state farms, or more than 50 percent o f the original SDA portfolio, are either in the process of liquidation through bankruptcy or are candidates for bankruptcy. Table4.3 Status of State FarmPrivatization(April 2003) Numberof farms Percent Startingnumber o f state farms inSDA portfolio, Jan. 2000 739 100 Inliquidatiodreorganizationhankruptcy(Law 64/1995) 382 51 Privatized 279 38 Remainto beprivatized 78 11 Source: SDA of Ministryof Agriculture. B.5. AgroprocessingIndustryUnpreparedfor EUEntry 4.34 A well-functioning and competitive food industry is an essential pre-condition for EU accession. The privatization o f Romanian agroprocessing has been almost hlly completed. Overall, less than 2 percent o f enterprises remain in state ownership, although there are obvious fluctuations between subsectors. While milling, baking, and meat processing are totally in private hands, the state still owns and operates 40 percent o f sugar plants and 12 percent o f fruit and vegetable processors. Progress with privatization may have been responsible for the growth inoutput andproductivityinagroprocessingsince 1990. The options for privateownership have also attracted increasing flows of foreign direct investment, but the level o f FDI in the food sector (and primary agriculture) remains low. Moreover, foreign investors generally prefer food retailing and restaurants, while FDI in food processing i s limitedto breweries and dairies. As a result, the small private owners (which constitute two-thirds o f the sector) generally lack financial resources for technological improvements, while the oversized and technologically obsolete state-owned plants work at low capacity and efficiency. Privatization o f the remaining state-owned processors (especially in the important sugar and fruit and vegetable subsectors) combined with aggressivepromotionofFDIshould become a highpriority. 4.35 The Romanian food industry is generally not prepared for the new competitive environment associated with EU accession. This lack of preparedness i s highlighted by NIS surveydata regarding I S 0 9000 systemimplementation. More than 40 percent o frespondents in the food industry in 2002 indicated that they were not interested in I S 0 9000 certification, and this percentagehad remained constant since 1999. Only 16 percent o f companies had I S 0 9000 certificates in 2002 (encouragingly up from 8 percent in 1999) and another 13 percent where in the implementation and design stages (up from 10 percent in 1999). To ensure competitiveness with EUindustries, the Romanian agrobusinessesmust quickly change their attitude toward I S 0 9000 certificationandmake concerted efforts to ensure system implementation. B.6. CommodityandFactor MarketsNotAdaptedto the Needs of PrivatizedAgriculture 4.36 During the last decade, the structure of agricultural production inRomania has changed dramatically from large-scale collective production to predominantly small-scale individual production. The changes in the type of farming have not been accompanied by an appropriatechange of marketingand supply structures. Largeprocessors are understandably reluctant to buy their raw materials in small batches from a multitude o f small producers: this 93 arrangement involves high transaction costs and does not guarantee uniform quality standards. Small producers need to rely on assembly markets and integrators for selling their farm products-and such institutions have not yet developed in Romania. As a result, direct sales in town markets are the main outlet for the basic commodities o f small producers, such as vegetables, milk, and meat. This naturally involves huge time inefficiencies and leads to high spoilage rates for these commodities. Technical crops, mainly sunflower and sugar beet, are generally produced and sold under advance marketing contracts between processors and producers; the demand for these crops is so high that even small farmers fit within the contract marketing scheme. Cereals, on the other hand, are not produced under advance marketing contracts. Yet the large number o f recently established small mills and bakeries is a positive force enabling small producers to sell their grain to the milling and bakingindustry. 4.37 One o f the main characteristics o f the CAP is the fact that price and market support is mostly done at the wholesale level and not at the farm gate. It constitutes an indirect support to farmers through wholesale prices and not a direct intervention by buying directly from farmers. Under current market conditions, small Romanian farmers relying on direct sales in the marketplace will not be able to benefit from such interventions. They need a network o f wholesalers or, in their absence, they need to organize in marketing cooperatives that allow wholesale-level support. Farmer marketing cooperatives will help small producers overcome the barriers imposed by smallness, and improve their bargaining position vis-&vis the processors. Moreover, marketing cooperatives will be able to provide sorting and packing services that are essential for meeting uniform quality standards; they will also be able to invest in storage and cooling facilities (e.g., for milk) and ensure efficient collection and shipping. 4.38 At present, private input supply channels seem to be better adapted to small farmers than output marketing channels. Networks o f input dealers have emerged inresponse to the extremely fragmented farm structure. These emergent dealer-operators are trying to supply a full range o f inputs to farmers, with the same operator selling pesticides, seeds, growth stimulators, veterinary drugs, and fertilizers (these last only in small quantities). Although no firm figures are available, the number o f dealers may exceed several hundred, and has generally been increasing. Foreign companies are entering the farm supply arena, especially in the pesticide and seed markets. Despite the generally competitive nature o f the input marketing, domestic production o f inputs, especially seeds and pesticide, is done by only a small number o f producers. The domestic market thus still has oligopolistic elements, and establishment o f input supply cooperatives by associations o f farmers will certainly improve their bargaining position and enable them to benefit from volume discounts and other scale economies, as well better access to importers, who private an alternative to domestic producers. B.6.1.Emerging LandMarkets Constrainedby ComplexBureaucratic Requirements 4.39 Buying and selling o f land became universally possible only with the adoption o f the Land Circulation Law in 1998. Legal entities are allowed to buy and sell land, whereas foreign nationals may only lease land. The state has released its land reserves only for leasing, and no privatization o f state land is contemplated. During the four years 1999-2002, private landowners sold 312,000 hectares o f agricultural land, which is a very respectable 2.7 percent o f private 94 agricultural land in rural areas; more than 4 percent o f rural landowners entered into land transactions duringthis period. 4.40 To this very day Romania has a dual land registration system inherited from before World War 11: the "land book" system and the "inscription-transcription" system. Inpreparation for the transition to the land book system envisaged by the 1996 Law on Cadastre, it is now required to have each plot surveyed by certified surveyors inthe process o f titling and ownership transfer. Notaries play a central role in transfer o f ownership, as they are responsible for validating the title and only a notarized sale contract can serve as a basis for recording ownership transfers. Notary fees and survey fees are thus two unavoidable components o f the cost o f property transfer. The cost o f surveying by private companies may runup to 3 million ROL per hectare. The total titling and registration costs may reach 4-6 million ROL per hectare, which is sometimes higher than the value o f the land. The high transaction costs constitute a serious obstacle to completion o f cadastral registration and development o f land market transactions. B.7. Urban-Rural Gap: Significant RuralSocialProblems 4.41 Romania's rural population of 10.1millionresides in 13,000 villages, which are clustered administratively into 2,700 communes-the traditional seat o f local government. Rural incomes per capita in 2001 were 27 percent below urban incomes, and this gap had increased from only 5 percent in 1997. Poverty is thus a significant problem in rural communities. In 2002, the relative risk o f poverty in rural areas was more than double the poverty risk in urban areas: 42 percent for rural residents compared with 18 percent for urban residents. The rural poor account for two-thirds of the total poor inRomania. 4.42 Agriculture-primarily self-employed agriculture-is the main source o f employment in rural areas: nearly 70 percent o f the rural employed report agricultural activities (compared with about 40 percent national average). As a result, rural households rely to a much greater extent on in-kindincome: 46 percent o f total income inrural households is derived from in-kindsources compared with only 12 percent for urban families. Rural residents also achieve significantly better nutritional levels. The caloric intake is 2,600 kcallday for rural residents compared with 2,380 kcal/day for urban people (1998 data); only 14 percent o f the rural population consume less than the minimum caloric requirement compared with 26 percent for urban residents. Yet farming incomes are generally low and the occupational structure o f the poor is biased toward self-employment in agriculture. Thus, 25 percent o f the poor are self-employed farmers, compared with only 15 percent in the total population. Many o f those self-employed in agriculture are semi-subsistence farm households, and many households inthis group are headed by pensioners. While the share o f farm households inextreme poverty increased inrecent years, pensioners have been shieldedbyrelatively well-targeted social security programs. 4.43 Because o f the reliance on self-employment in agriculture, the unemployment rate in rural areas is much lower than among the urban population (about 3 percent compared with 10percent in urban areas in 2001). Moreover, rural unemployment dropped from 5 percent in 1996 to 3 percent in2001, while urban unemployment hovered around 10 percent with a certain tendency to increase. Long-term unemployment is primarily an urban issue, whereas in rural 95 areas the major problem i s underemployment combined with low productivity in existing farmingoccupations. 4.44 The rural population is disadvantaged by both human and social capital indicators. Interms o f educational attainment, only 1percent ofthe ruralpopulationhave higher education compared with 9 percent in urban areas; more than 7 percent o f the rural population have not completed any formal schooling compared with less than 2 percent inurban areas. Social capital indicators also show locational disparities: rural inhabitants have less access than urban residents to contacts who can help with problems relating to employment, the health and legal systems, the police, and the banks. Thus, only 25 percent o f rural residents "know someone who can help" solve problems with these institutions comparedwith 35 percent for urban residents. 4.45 There is a large discrepancy between urban and rural Romania in terms o f both physical and social infrastructure (Table 4.4). This discrepancy and the state o f rural areas ingeneral are a significant challenge for the country and a critical constraint for integrating into an enlarged EuropeanUnion. Table 4.4 Urban-Rural Disparities in Household Amenities and Community Services (percentage of Households) Urban Rural Consumer assets no gas stove 4 27 no refrigerator 11 34 no washing machine 28 67 Household amenities less than 5 sq. meters per Person 4 10 no bathroom 13 84 no flushtoilet 8 68 no hot water system 15 86 Community services no post office inthe area 0 16 no health service inthe area 0 15 no pharmacy inthe area 0 40 no culturalientertainment Center 2 25 no children playground 87 poor roads 98 20 Source: World Bank estimations based on HouseholdBudgetSurvey(ABF 2002) and Living ConditionsSurvey (ACOVI2002). 4.46 Rural infrastructure, including rural roads and water supply, has a potentially important impact on rural development in general and on local non-farm investment incentives in particular. Water and sanitation systems additionally affect family health in rural communities. Yet only half the communes (1,355 out o f total 2;868) are connected to water supply systems, and more seriously these communes cover only one-quarter o f the rural population (2000 data). According to the 1992 population census, only 12 percent o f the rural population had access to water supply from public systems or private wells. Because o f budget constraints, both rural water supply systems andrural roads suffer from inadequate maintenance. 4.47 Health conditions in rural areas are poor. Infant mortality is 22 per 1000 live births compared with 15 in urban areas. There is one doctor per 1,500 rural residents compared with one doctor per 350 residents in urban areas. Most o f the rural communes provide only the most basic medical services, and for special health needs rural residents have to travel to the nearest town. The quality o f the medical services in rural areas i s relatively low, mainly because o f inadequate infrastructure and equipment. Romania lacks any substantial information and awareness campaigns on environmental health. Resources to implement such campaigns are scarce, andthe organization and skills to carry them out are limited. c.SELECTINGANAPPROPRIATECOURSEFORAGRICULTURALPOLICY C.1. Shiftin Policiesfor IncreasedCompetitivenessandfor the Introductionof CAP 4.48 Untilrecently, the completion o f the transition to a market-consistent agricultural policy environment and the provision o f certain safeguards for farming incomes have been inthe focus o f the government's attention. Different forms o f market interventions have been tried, and the harmonization with the CAP has become an additional, but still not a major, objective. The current agricultural support system mixes heavy government intervention with the principles o f a market-based agriculture. However, it does not fully correspond with the level o f support and instruments o f the CAP. Although there is pressure for immediate increases o f support and protection from the farmers' side and the obvious need to converge on EU levels before accession, the level o f support and border measures should remain unchanged in order to maintain pressure to complete reforms and to remain within budgetary constraints and WTO commitments. The recommendation is to align the level o f support only after the accession. 4.49 In moving toward EU accession, first of all the emphasis on increasing production needs to be removed.The focus on increased production has shaped past agricultural policy by promoting intervention in favor o f specific commodities and introducing distortions in agricultural markets. Given the open trade environment o f our time in general, and the trade goals o f EUaccession inparticular, Romanian agricultural policy makers have to make a mental adjustment away from increasing production and toward competitiveness. Romania suffers from generally low yields inboth crop and livestock production, combined with the lowest agricultural labor productivity in the region by far. It is on these issues o f agricultural reform that policy makers must concentrate instead o f the traditional emphasis on increased production. Competitiveness requires higher efficiency in agriculture. This can only be achieved by policies that facilitate the structural reorganization o f agriculture by allowing inefficient farms to close down and removing obstacles to the expansion o f new and more efficient farming units. 4.50 The recommended agricultural policy framework and support system would need to focus on facilitating the required structural changes and improvement in competitiveness by pursuing the following objectives: Quick completion o f the most important pending transition tasks, such as farm consolidation, landprivatization, creation o f functioning land markets. e Facilitation o f the consolidation o f the small-scale farming sector through the creation o f rural non-agriculturaljob opportunities resulting inthe reduction o f the agricultural work force and improvements o f labor productivity. 97 0 More effective use o f budgetary support to agriculture through refocusing o f support programs on efficiency enhancement rather than price support and export subsidies. 0 Integration o f the various instruments o f government intervention inthe sector into a more consistent and predictable framework, including provision o f a reliable orientation for farmers untilthe introduction o f the CAP. 0 Clear separation o f rural social measures (e.g., measures to reduce social tensions and provide social protection inthe rural areas) from the major instruments o f agricultural policy aimed at improving efficiency and competitiveness. 4.51 Efforts should also be concentrated on the adaptation of the same policy instruments as in the EU, but not yet necessarily applying them, or applying them at the same level. The current CAP should be used as a benchmark in this process. The so-called SIPS framework (Simplified Implementation o f Payments System), which provides opportunities for a simplified implementation of the CAP for new member countries, should also be taken into account. In particular, the following adjustment would needto be considered: 0 The market support system needs to be adjusted to a transparent CAP-conforming framework incorporating the potential use o f quotas for some products. Changes should be made inthe current intervention system, shifting from ex-ante mechanisms (when the intervention is based on short-term forecasts on production and consumption) to an ex-post CAP-type mechanism, when market intervention decisions by the private sector are based on the observed evolution o f EU market prices. Inaddition, to be consistent with the CAP principles, price support would need to be shifted to the wholesale level instead o f the farm-gate level for most products. 0 The direct payments and other current structural support measures have to be adapted to the EU format. Direct payments for irrigation inputs and for the preservation o f selected genetically valuable livestock are not recognized in the EU and should be eliminated. On the other hand, new support measures as envisaged under the EU agro-environmental programs will have to be introduced. Thus, the EU provides direct payments for the rearing o f endangered livestock breeds in the context o f its rural development measures. The generalized area payment scheme outside o f the less-favored areas could be adapted and fitted into the national envelopes proposed for the dairy andbeef sectors under Agenda 2000. 0 A more integratedregional and rural policy and related support programs have to be developed to be able to utilize the EUstructural fund instruments. 0 The current credit guarantee scheme should be made fully consistent with EU standards and most credit subsidies will have' to be discontinued in their present format. None o f the current support measures aiming to improve the access to financing o f capital andnon-capital expenditures is admissible under the CAP. Credit subsidies may temporarily be replaced with investment grants from national 98 budgetary sources (or financed by the EU, ifpossible). Support for modernization and restructuring also needs to be revised in line with EU guidelines. The EU currently allows some grants for structural development measures, but these must meet one or more o f the following objectives: reduce production costs; improve product quality; preserve and improve the environment; meet hygiene and animal welfare conditions; encourage diversification in agricultural activities. N o support for modernization and restructuringto agroprocessors i s admissible inthe EU. o The taxation system and the providedtax concessions need to beharmonized with the EUpractices and requirements. N o tax preferences for farmers are allowed in the EUand these will have to be eliminated entirely before accession. C.2. SimulatingFuturePolicyOptionsfor Romania'sAgriculturalSector 4.52 The impact o f various agricultural policy scenarios on farmer incomes, on consumer real incomes, on poverty levels, and on the national budget has been assessed by a simulation in a partial equilibrium modeling fi-amework (see Annex 2). Nine different policy scenarios have been used, seven o f which represent evolving variants o f the CAP, one represents the theoretically ideal situation when all distortions are removed, and one is the benchmdc scenario corresponding to the current trade and market policy. The list o f scenarios is given in B o x 4.1 and the detailed simulation results are provided inAnnex 3.*' Box 4.1 Scenarios for Simulating Future Policy Options Scenario'4: Romania maintains its current trade and market policy Scenario BI: Partial adoption of current CAP. Adoption o f Agenda 2000 price policies, but without direct payments. ScenarioBz: Complete adoption of the agreement on the CAP as applied to first wave accession countries (Copenhagen Agreement)-m levels. Simulating the final level of direct payments (foreseen for 2013), Le., 100percent o f those applied to EUfarmers. ScenarioBs: Complete adoption of the agreement on the CAP as applied to first wave accession countries- -levels.Similar initial to Scenario BZ,but direct payments amounting to 25 percent o f those applied to EUfarmers (phasing-in or starting level). ScenarioBq: Complete adoption of the provision on a "Simplified Implementation of Direct Payments." Similar to B3but opting for SIPS, i.e., granting direct payments in the form o f a de-coupled area payment applied to the whole utilized agricultural area (no livestock premia!) Scenario CI: Complete adoption of the future CAP (Council Compromise), including the full decoupling o f CAP-type direct payments (i.e., no compensatory payments granted to individual production activities). Scenario Cz: Partial adoption of the future CAP (Council Compromise). Similar to C,, but including 100 percent direct payments. Scenario D: Implementation of the Romanian negotiation position. Simulation of the Romanian position for the negotiations onChapter 7 (Agriculture) withthe EuropeanCommission. Scenario E: Complete removal of current divergences ('reference scenario'). Close to a free trade scenario, it provides anorder ofmagnitude ofthe effects ofthe current trade regime inRomania. *' All scenarios were simulated for ten major agricultural commodities: wheat, barley, maize, sunflower, potatoes, milk, beef, pork, poultry, and eggs. These commodities account for almost 70 percent o f gross agricultural production. 99 C.3. TheWelfare Impactof Introducingthe EUCommonAgriculturalPolicy 4.53 The overall effect of Romania's Figure 4.8 Change in Farmers' Income under integration into the EU will be positive: a - Variousscenarios preliminary estimate of the expected budgetary transfers from the EU Structural and Cohesion Funds (SCF) indicates that the 81 No directpaymen benefits of EU membership greatly outweigh the costs. Net transfers from the EU budget 82 100%directpayments , (direct compensation payments to farmers and SCF transfers net o f Romania's expected contribution to the EU budget) can be C1 Compromise no dir pay estimated at between 36 trillion ROL and C2 Compromise 100% 81 trillion ROL (3 percent to 7 percent of D Romanianposition Romania's 2001 GDP). E NOn-inteNenti0 I -40 -20 0 20 40 60 80 100 120 4.54 However, for the individual groups of b u r c e : World Bank staff estimates. producers and consumers of agricultural products and for the agricultural administration in general the impacts o f CAP adoption are mixed. Our simulations indicate that the implementationof CAP-type agricultural policies would significantly benefit the farmers, while the population as a whole would be worse off due to higher food prices and the number of the poor would actually increase. Compared to current policies, an implementation of CAP-type policies would lead to a doubling of farmers' income (equivalent to a national farmers' income gain of 131 trillion ROL), but an overall real income loss of about 3 percent for consumers (equivalent to a national consumers' real income loss of 244 trillion ROL). Combined with the positive effect on the budget, these oppositely directed effects produce a net welfare loss of about 2 percent of GDP. The negative impact on Romanian consumers i s further emphasized by the estimated growth o f poverty under the CAP-type price regime: in the long term, the number of poor people would increase from 2,442,500 (10.9 percent o f the population) to 2,845,600 (12.7 percent). Inthe short run, this effect would even be even more severe as poverty might increase to 13.7 percent. The actual magnitude o f these effects depends on the specific scenario, but paradoxically they are especially pronounced underthe assumptionof full adoption ofthe Romaniannegotiating position. 4.55 The exact increase in farmers' incomes (approximated by aggregating commodity- specific value added from the analyzed activities) depends on the amount of direct payments granted to Romanian farmers. The highest increase in farmers' income would be achieved by a full adoption of the Romanian negotiation position (Figure 4.8). The other CAP-type scenarios would also induce significant increases in farmers' income as long as they involve direct payments. Introduction o f CAP-type price policies without direct payments (Scenarios B1, C1) would actually lead to a reduction in farmers' income (by 2 percent or 16 percent in the short run, much more in the long run-see Figure 4.8); this effect i s determined by the livestock production activities which suffer from higher input costs (grainprice policies!) combined with a more restrictive price policy andquota application." 90 The more pronounced effect for the Council Compromise scenario C1 appears to be due to the fact that our simulation only takes activity-based (commodity-specific) direct payments into account. Under a full decoupling o f 100 4.56 The impacts of different agricultural policy regimes vary across agricultural activities and regions. As a result of the changes in relative protection, different farm types and different regions will be affected in very different ways, with some of the farmers gaining and others losing out from the reforms. Broadly, crops would expand, but livestock production might contract depending on the specific reforms. The new policies create a clear incentive to modify the land production mix in favor o f directly supported crops. As a result, the output o f all major crops is projected to expand with the extent of the increase depending on the individual scenarios. Milk, beef, and eggs production also might expand; however, the change in the production of other livestock activities depends entirely on the individual scenarios and would not necessarily increase. 4.57 The implementation of CAP-type agricultural policies implies an increase in agricultural prices in Romania, leading to an increase in food prices. On average, CAP-type changes in agricultural policies would induce an increase in food expenditures o f 3 percent to 7 percent compared to the current base-period conditions (Figure 4.9). The highest increases in food expenditure would occur under the Romanian negotiating position and under the agricultural policy currently implementedinthe first wave o f accessioncountries. Expenditures are projected to increase for all food product groups, with the highest increases projected for milk, meat, and eggs. 4.58 Since Romanian households on average spend more Figure 4.9 Impact o f CAP-type policies versus intervention ?han half their disposable income on food, policy reforms that raise food prices would produce a decrease in consumers' real income. As a consequence o f higher food expenditure, the introduction of CAP-type 2 agricultural policies would reduce - -CAP-type real household incomes by 2 Non-interventlo2 percent to 4 percent (Figure 4.9). 0 The model estimates that the real -1 income loss would be highest under the conditions of the current -3 ' Romanian negotiation position and Food expenditurdousehold income Poverty rate the Agenda 2000-type policies. I The real income loss from an Source: World Bank staff estimates. introduction of the current Council Compromise is more moderate but still significant. The population at large stands to loose from the introductionof CAP-type policies. Households that do not produce food on their own plots and spenda greater share oftheir cash expenditureson food will be hit hardest. direct payments, these eventually would be paid to the farm and not to the (analyzed) commodities. Our commodity- coupled results therefore underestimate the positiveimpactr and overestimate the negative impacts on total farm income. 101 4.59 CAP-type policies would have a negative impact on the poorest. Inthe short run, the introduction o f CAP-type policies would increase poverty from 10.9 percent in the current situation to 13.7 percent under the Romanian negotiating position, which corresponds to an absolute increase of about 626,500 people. Once consumers adjust, the medium- to long-term effects would still be significant with poverty levels at 12.7 percent o f the population (Figure4.9). The government should introduce policies to counter the adverse effects o f CAP adoption on the poorest people. 4.60 Net transfers from the EU budget would be between 36 trillion ROL and 219 trillion ROL, depending on the imposition of EUrestrictions. Without any restrictions on transfers from SCF, net transfers from the EUbudget would be between 173 trillion ROL and 219 trillion ROL (5.5 to 7.0 billion), corresponding to about 15 percent to 19 percent of Romania's 2001 GDP. However, Agenda 2000 currently limits the maximum annual receipts from SCF to 4 percent o f national GDP o f any Member State. Therefore, the actual SCF transfers would not exceed 46 trillion ROL and net transfers from the EUbudget would range between 36 trillion ROL and 81 trillion ROL (3 percent to 7 percent o f Romania's 2001 GDP). 4.61 Projects receiving EU structural funding require co-financing by the recipient country, and the amount estimated as the EU share in project funding will have to be accompanied by funds from the Romanian budget. Assuming that Romania qualifies for assistance under Objective 1, and assuming a EU participation rate o f 75 percent, the maximumEU contribution o f 46 trillion ROLwould have to be matchedby Romanian co-financing of about 15 trillion ROL from the budget. C.4. The Impact of Romania's Current Agricultural Policy 4.62 The simulation o fthe introduction o f CAP-type policies has been discussed relative to the benchmark o f Romania's current policies. Analysts and policy makers may be interested in evaluating the effects of these current policies against a completely non-distorted, free-trade environment. A scenario has been run to assess the impact o f current policies relative to the theoretically ideal situation involving removal of all government policy interventions and assuming no EUtransfers (Scenario E). Some o f the comparative results are shown inFigure 4.9. 4.63 The simulations show that the removal o f all policy distortions in agriculture would slightly reduce farm incomes and significantly improve the real consumer incomes due to a reduction infood prices. 4.64 The removal o f farmers' support would result ina reductiono f 0.2 percent (short-term) or 1.4 percent (medium- to long-term) in the aggregate value added for the sector as a whole (seeFigure 4.8 above). This i s less than the estimated reduction in gross agricultural output (5.0 percent or 9.3 percent, respectively), because the scenario also assumes liberalization o f input prices (removal o f protection on fertilizers and energy), mitigatingthe negative impact on farmers' income. 102 4.65 A significant positive effect o f the removal o f all current distortions is observed on the consumer side. Here, such a policy reform would lead to an immediate increase o f 1.O percent in household incomes. The positive effect would persist in the long run, with household incomes risingby 0.7 percent (Figure 4.9). 4.66 The removal o f all current market distortions would reduce the number o fpoor by 66,822 to 100,335 people. Assuming no demand response, the share o f poverty would decrease from 10.9 percent under the current conditions to 10.5 percent, corresponding to the aforementioned headcount. The results again emphasize that the poor are negatively affected by distortive agricultural policies, and they would benefit from a removal o fmarket distortions. D. CONSOLIDATIONINTHE FARMING SECTORAND REDUCINGEXCESS AGRICULTURALLABOR 4.67 Creating a more efficient and competitive farming sector i s probably the most important task inpreparing for EU accession. This i s extremely complicated and requires several specific actions. D.l. Completion of FarmRestructuring and FarmPrivatization 4.68 Romania has made significant progress toward the establishment o f a genuine private agriculture characterized by a mix of organizational types spanning the entire spectrum o f individual and corporate farms. Yet some important tasks remain to be completed. he first and foremost among them concems the responsibilities of the State Domain Agency, i.e., completing the privatization o f state farms and ensuring that the bulk o f state-owned land is privatized by auction or, failing that, expeditiously entrusted to private operators. More than 50 percent of the original number o f state farms inthe SDA portfolio arejudged to be insolvent and are slanted for liquidation rather than privatization (see Table 4.3). The liquidation procedures must be completed without further delays. Large areas o f agricultural land will thus be released from their currently locked status o f pre-emptive concession and become available for free leasing to private operators. Every possible priority must be given to disposing of these large reserves of state land and returning them to productive cultivation by private farmers. 4.69 The bankrupt state farms have been allowed to exist and accumulate losses for more than a decade due to the traditional regime o f soft budgets that encouraged financial irresponsibility. A preliminary analysis o f the financial data for some 10,000 legal entities provided by NIS suggests that more than half o f these corporate farms report losses during the last 4-5 years. Not all of these farms have reached a state o f true insolvency, but the frequency o f unprofitable farms i s alarming. The government should avoid the potential financial disaster that may evolve from this situation and take immediate measures to impose hard-budget constraints and strict financial discipline on all farms. Bankruptcy procedures, however painhl, should be enforced for all farms that are seen to be chronically unprofitable. 4.70 One o f the original goals o f transition involved the elimination o f the large-farm bias built into the Romanian agricultural mentality since 1948. Unfortunately, recent policies only strengthen the feeling that large-farm bias persists in Romania; the most recent manifestation o f this bias i s the curious classification into "commercial" and "family" farms, which allows 103 subsidies to be paid mainly on the basis o f pure size criteria. Commercial farming should definitely be supported and encouraged, but this should be done based on measures o f commercial activity, not size. The objective should be to help small farmers increase their level o f commercialization, andthis cannot be achieved by cutting them off from subsidies because o f their smallness. Successful development of the farm sector requires abolishing the traditional preferential treatment of large farms and establishing a level playing field for farms o f all organizational types and sizes. D.2. Consolidation of Small-scale FarmingSector and ReducingLabor inAgriculture 4.71 Romanian policy makers are preoccupied with the fact that their country has 4.2 million farming units, o f which 60 percent are smaller than 2 hectares. For comparison, the 15 countries o f the European Union in aggregate have 6.8 million farming units and 60 percent o f the EU-15 f m s extend in size up to 5 hectares. The average farm size in Romania is 2.9 hectares (including all individual and corporate farms), while EU-15 farms average 19 hectares. These phenomenahave led to heated discussions o f fragmentation and consolidation. 4.72 Consolidation o f small f m s should be encouraged because, based on empirical evidence from farm surveys inRomania and other countries, more land is at least a partial guarantee for a better standard o f living in rural households. Consolidation will occur naturally if farming is a profitable business: farmers will seek ways to increase their holdings in a supportive environment, where they can eam enough money from agriculture. The government should accordingly implement policies that will enable farmers to improve their financial performance. These policies include e Development o f marketing and supply channels for agriculture, including encouragement o f service cooperatives for small farmers; 0 Provision o f market information, extension, and farm management training for small farmers (all these are public goods, and thus the responsibility o fthe government); 0 Attention to the establishment o f rural credit systems accessible to farms o f all profiles; and 0 Targeted grants and subsidies intended specifically for development of commercial activities and introduction o f highvalue added commodities with a significant export potential. 4.73 One conclusion regarding farm consolidation emerges very clearly from the international experience: formal government-initiated land-swapping programs (either directly between small landowners or through a state-owned pool) are never very successful. Consolidation must be left to market forces, with the government standing ready to provide budget support to successful and sufficiently motivated farmers willing to enlarge their farms ina market-driven environment. 104 4.74 Consolidation o f fragmented farms naturally requires that some people give up the use (ifnot the ownership) oftheir landinfavor or moreefficient andmoreproductive farmers. There are at least three market mechanisms that may effectively act inthis direction: 0 development o f markets for land transactions; 0 provision o f adequate pensions to elderly rural landowners; and 0 development o fnon-agricultural employment opportunities inrural areas. 4.75 Land markets provide a mechanism for retiring elderly and non-productive landowners. The development o f land market transactions, including land leasing, i s particularly important for Romania where restitution and reconstitution transferred ownership and use rights to 4.2 million rural beneficiaries, o f which 40 percent are older than 60 and nearly 30 percent are pensioners. According to certain estimates, pensioners own 65 percent o f land in rural areas. These elderly people are obviously not the most productive among farmers and they are probably on the verge o f becoming passive landowners. Another 500,000 restituents are urban residents with alternative careers who have no interest in resuming active farming. In this setting, land markets have an important role in providing a medium for the flow o f land resources from less active to more active and more efficient producers and thus generating potential productivity improvements through consolidation. To exploit landmarkets as a tool of land consolidation, the government should adopt policies that simplify registration and titling procedures, minimize transaction costs, and ensure security for both lessors and lessees through effective contract enforcement. 4.76 Targeted income support can reduce the reliance of elderly and poor households on farming and release land for consolidation. Households o f pensioners derive only 30 percent o f their income from pensions and fully 58 percent from the household farni (50 percent inkind and 8 percent in cash from product sales). The only way to encourage elderly people to stop subsistence farming and release their land resources to more productive users i s by targeted income support, andby strengthening contracts to enable deriving income from land leasing. 4.77 Non-agricultural rural employment will release farm labor and facilitate consolidation. Ina country with 4.2 million farm units on 10million hectares o f arable land, any significant consolidation o f farm sizes would force large numbers o f workers out o f agriculture and may lead to a massive migration o f rural unemployment to urban areas. Development of non-agricultural rural employment opportunities in services, construction, and manufacturing is essential in this situation, especially for people o f pre-retirement age. This can be achieved through policies that encourage establishment o f various small- and medium-sized enterprises in rural areas. However, this is a very long process, and the government should initiate appropriate policies as soon as possible to ensure successful development o f non-agricultural jobs a few years down the road. 4.78 Factors other than land consolidation have a strong positive impact on rural incomes. Attention to education, rural infrastructure, and development o f market channels has a beneficial effect both on farming incomes and on non-agricultural jobs. Empirical evidence suggests that land is not always the most important factor for increasing the standard o f living o f rural families. Access to machinery, human capital (Le., education), and willingness to engage in 105 commercial sales have an even greater impact on household well-being than land. Emphasis on physical and human capital and development o f market channels for commercial sales are probably the priority policy directions that the government should explore, in addition to land market development, in its efforts to counteract the impacts o f fragmentation. These measures may definitely produce additional synergies, as they will enhance the motivation and capacity for the development of non-agricultural activities inrural areas. E. ACCELERATINGTHE NECESSARY REFORMSINAGROPROCESSING, INPUT SUPPLY, AND MARKETING 4.79 Two main tasks face policy makers in the agroprocessing sector. First, second-wave privatization should be launched without further delay, especially in state-dominated subsectors, such as sugar and h i t and vegetables. Second, every effort must be made at the government level to promote and attract FDIinto agroprocessing (and primary agriculture). The government should study the experience o f other European countries-and especially Ireland-in encouraging foreign investment. No special measures are needed for further encouragement o f FDI in food retail and restaurant business, but it is imperative to attract foreign capital for modernizationandupgradingofprivatizedagroprocessing. 4.80 In parallel with encouragement of foreign investors, the government should develop policies for the encouragement o f domestic investment in small and medium size processing enterprises inrural areas. Foodprocessing is an ideal complement to the agricultural activities of the rural population and it can be set up in villages with a minimum o f effort and investment. In addition to augmenting the income o f entrepreneur families, this activity will generate localjobs and thus contribute to agricultural labor reduction and non-agricultural rural development in a broad sense. These policies should be part o f a general forward-looking rural development strategy that must not remain limited to the traditional course o f providing subsidies for the purchase o f agricultural machinery and equipment. 4.81 The marketing system appears to be incompatible with the needs of the new privatizedagriculturein Romania.Farmers should not be forced to rely on the inefficient and time-consuming method o f conducting their sales face-to-face with the end consumer inthe town market. The government should facilitate the entry o f wholesalers and other trade intermediaries into the food chain. Marketing farmer cooperatives should be established to overcome the problems o f smallness and in particular to alleviate the concerns that processors and supermarkets have indealing with small producers. 4.82 Farmer cooperatives will also improve the bargaining position o f small producers in the semi-oligopolistic input supply system that prevails in Romania, despite the encouraging emergence o f many relatively small diversified suppliers. As part o f its FDI promoting policy, the government should also focus on the entry o f foreign input suppliers into the Romanian market with the aim o f increasing competition. 4.83 Romania must expeditiouslycomplete its efforts to meet the quality and standards requirements of the EuropeanCommunity relatingto food industry. Considerable progress has been achieved in harmonizing the local standards and regulations with the EU, yet much 106 remains to be done. This i s not so much the task o f the Ministry o f Agriculture in Bucharest as the responsibility o f the regional field offices, which much spearhead an extensive educational program aimed at producers and processors in all parts o f the country. It seems that the Romanian producers are still not fully aware o f the crucial importance o f meeting the EUquality standards for their future access to Europeanmarkets. The Ministry o f Agriculture should deploy its regional forces so as to overcome these behavioralbarriers and persuade all actors inthe food industryto take the necessary steps toward gaining EUapproval. F.CREATINGEUCONFORMINGINSTITUTIONSANDREGULATORYSYSTEMS 4.84 Romania will have to implement the full acquis communautaire by the time o f its accession to the EU. This involves two main prerequisites: (a) creation o f a competitive market with private and public institutions capable o f meeting common market requirements; and (b) establishment of EU-compatible institutionscapable o f administering the CAP. 4.85 The implementation o f CAP-conforming institutions and the training o f qualified staff has proven to be one o f the main challenges on Romania's road to accession. Romania has lagged in this area and remains well behind the level o f its neighboring second-wave accession country, Bulgaria. Acknowledging the need for accelerated progress, t3e Romanian Government inJuly 2002 took several initial steps: it passed a revisedNational Program for the Adoption of the Acquis (NPAA); and it opened the agricultural chapter for negotiations in October 2002. These are indeed steps toward acceleration o f institutional development in the agricultural and rural spheres, but to become reality they require sustained political will and timely budgetary support.9' 4.86 While significant progress has been made in some areas (e.g., veterinary and phytosanitary acquis, including food safety chapters), concern remains about Romania's ability to comply with the complex administration requirements o f the CAP. Further institutional development should focus on the following issues. e Efforts to set up institutions and horizontal structures related to the administration of Common Market Organizations (CMOs) need substantial acceleration. The legal basis for the administration o f many CMO-related instruments, Law 73/2002 on the Organization and Operation o f Agricultural and Food Markets, still needs substantial adjustment. Specifically, the implementation rules should be clarified and concrete market mechanisms for state intervention should be assigned to each production subsector. Insufficient progress has been achieved in developing administrative structures for the management o f CMOs for specialized crops and for animal products. According to an EU assessment, satisfactory progress has been achieved only inthe sugar andwine subsectors. 9'According to a Ministry of Agriculture estimate (November 2002) a budgetary effort o f around 220 million is needed for the implementation o f the acquis on Common Market Organizations, o f which 87 percent is the cost o f acquis implementation and 13 percent is operational costs. Of the total amount, 54 percent is meant for institutional capacity building, 36 percent for acquis adoption and stafftraining, and 10percent for investments. 107 0 Institutional and legislative arrangements regarding the administration of EUFunds still remainto be implemented. Insufficient progress has beenmade inthe implementation of the acquis regarding EAGGF(Guarantee and Guidance Fund). The attempt to establish a Romanian EAGGF (composed o f 50 percent customs duties for agricultural imports, 20 percent excise taxes for alcohol and cigarettes, and 10 percent VAT for foodstuffs) has failed and the fund is not expected to be functional until2007. 0 The establishment of a Paying Agency i s a prerequisite for the implementation of CAP-like arrangements. The Agricultural Intervention Agency's main purpose would be to manage operation o f the agricultural support regimes and their transition towards the CAP. The Payment Agency would disburse direct payments and other subsidies to farmers, and according to plans, the paying agency will start activities in 2004 and will be fully operational in 2007. With the accreditation o f the SAPARD Agency for the implementation o f three approved SAPARD measures, Romania has taken the first step towards the establishment o f mechanisms for the absorption o f rural development funds (and ultimately EU structural funds). However, SAPARD implementation is still at a very early stage, and only a small portion o f the available funds have actually been spent. Two frequently mentioned problems are complex application forms and lack o f assistance to farmers and other potential beneficiaries in preparing project proposals. 0 The introduction of the Integrated Administration and Control System (IACS) is a prerequisite for the functioning of the CAP. The development o f the IACS centralized database i s particularly challenging because the current system inRomania is entirely different from the EU-style IACS, and because the data required for the IACS system are not readily available. Little progress has been made to date on this front. The design, building and implementation o f the IACS system needs considerable investment (estimated at over 18 million). The absence o f IACS accreditation upon EU accession would lead to the loss o f EU support funds. 0 Promotion and facilitation of voluntary professional associations. Agricultural growth will be enhanced if a broad range o f farmers and processors associate to provide services and seek representation inthe market and at the policy table. The current pattern i s for such organizations to represent narrow vested interests, to the detriment o f public policy. The current legislation allows farmers and processors to establish farm organizations and to have their interests represented inCMO management. Seveninter-branchorganizations have emerged(e.g., inthe sugar industry), and nine additional ones are being formed. Preparation o f an administrative structures for C M O operation is also slowly taking shape. 0 Modernization of the sanitary-veterinary laboratories requires attention. Romania has joined the Animal Disease Notification System (ADNS), a key step in designing the animal health infrastructure necessary to implement the acquis 108 regarding livestock. Efforts are being made to improve the infrastructure and accredit laboratories. a Enforcement capacities in the veterinary and phytosanitary field have to be strengthened.According to the EC Regular Report for 2002, significant progress has been made in adjusting the legislation to the E C veterinary and phytosanitary acquis, but an appropriate legal basis is still missing for effective operation in these areas. By 2004, a suitable legal basis in the phytosanitary area was created but an appropriate legal basis for veterinary controls is yet to be developed and adopted. Additional human and financial resources must be allocated to strengthen the currently limited implementation and enforcement capacities o f the veterinary andphytosanitary administration. a Adoption of EU food-safety standards requires a radical reorganization of administrative bodies. In order to increase their effectiveness, a reorganization o f supervising and surveillance bodies inthe field o f food safety (i.e., the national food laboratory and the plant hygiene laboratory) i s a further crucial area for meeting the acquis. This relates to both the complex numerousness o f super- ordinated ministries and agencies asslged to regulating the food security field, as well as to deficiencies in intemal organization and sharing o f capacities and information among the responsible laboratories. No system has yet become operational to trace non-compliant products domestically, although a system for registrationo fproducers is operational. Inorder to meet acquis requirements, this monitoring must equally cover domestic production. a A system for monitoring compliancewith pesticide residue regulations must be put in place. A system for registration o f pesticide manufacturers has been operational since February 2002. A national monitoring program has been established and residue controls have been initiated for imported fruits and vegetables. Residue monitoring should now be launched for domestic production inline with the acquis. G.OVERCOMING OBSTACLESTO RURAL DEVELOPMENT-REDUCING RURAL POVERTY 4.87 Reduction of rural poverty requires attention to a number o f rural development issues. First, non-farm employment will become increasingly important in absorbing the labor that will be shed from the farm economy as farms consolidate and become more efficient in response to competitive pressures. International experience shows that creation o f a vibrant rural non-farm economy depends upon a number o f factors that are relevant to Romania: a maintenance o f a market-friendly policy environment for small- and medium-size businesses (with specific technical or financial supports for rural SMEs incertain cases); a public investments in physical infrastructure, including roads, water, communications, and energy; and 109 e investment in human development in terms o f basic education and training, to better prepare the rural population for off-farm work. 4.88 Second, increasing productivity on small- and medium-sized f m s will require sustained attention. The areas o f support for these farms should include: e measures to stimulate agro-processing and agri-business, including a conducive business environment, and access to term finance via private lenders on commercial terms; a improved agricultural extension for small- andmedium-sized farmers; e upgraded marketing infrastructure, including simple standard contracts and enforceable contracting mechanisms; e support for development o fmarketing associations; and e an improved environment for land transactions, including less complex administrative procedures and adequate contract enforcement. 4.89 Third, maintenance o fthe social security o frural pensioners requires continued attention, particularly given budgetary constraints. As it now stands, land holdings and the income received from them, in the form o f food and rental income, are crucial to maintaining the precarious standard o f living for most pensioners. Since increasingpension payments is not an option due to fiscal constraints, a number o f indirect measures can be proposedto assist the vulnerable elderly: e encourage elderly landowners to engage in land rental transactions by preparing standardrental contracts andproviding assistance inexplaining and enforcing the legal options; e increase the demand for landrentals from elderly people by increasingthe number o f productive farmers through improved credit schemes that could be used for the purchase o f farm equipment; a expand the range o f annuities attractive to elderly investors, so that rural residents have an alternative investment option ifthey make the irrevocable decision to sell their land; and e reduce the isolation o f the rural elderly through selective investments intransport and communications in remote villages to help them maintain access to social services and to the outside community. 5. ENHANCINGLABOR MARKETADJUSTMENT 5.1 Romania's development prospects and its efforts to catch up with other countries in the region in integrating with the EU are determined by the ability to reallocate resources from low to higher productivity activities across the economy. Success depends to a large extent on the functioning o f three factor markets: labor, financial, and land markets. This chapter discusses policies to enhance labor market adj~stment.~~ It first describes the recent labor market trends in participation, employment and non-employment. It firther analyzes sectoral employment imbalances and job creation and destruction. The chapter then turns to discussing wages and productivity. Following an assessment o f the effectiveness o f the labor market policies, the chapter concludes byproposing a set o fmeasures to increase labor market flexibility. 5.2 The chapter acknowledges there has been efforts inrecent years to modernize some o f the institutional framework o f labor market. At the same time, the research demonstrates that the constraints imposed by the slow pace o f reforms inthe enterprise sector throughout the nineties, compounded by shortsighted sectoral policies, to a large extent have undermined labor market flexibility. These policies have destroyed a large number o f jobs and have created few sustainable ones, have pushedworkers out of the labor force and into low productivity activities, and have facilitated and supported the preservation o f a distorted sectoral distribution o f employment. Wage policy rigidity, especially in the SOE sector, has hampered reallocation, and has increased labor mismatch and labor market segmentation. 5.3 The analysis identifies three perverse labor market trends o f high magnitude which are largely specific to Romania, namely: (i)the migration o f active workers into subsistence agriculture and other low productivity/low earnings activities; (ii) the large number of drop-outs from the labor force, mainly through retirement and early retirement and through labor market discouragement; and, (iii) flows out o f employment and into long-term unemployment. These trends have been driven primarily by the collapse o f output andby short-term constraints which require reversal, and should be the focus o f labor market policies in the near future. These large scale labor market trends have not only affected labor market flexibility, but have also distorted the balance between the occupied and the working age population and have affected the competitiveness ofthe Romanian economy. 5.4 The chapter shows that these trends are not sustainable and are bound to be reversed. Signs o f change can be seen inthe chapter. The number o f workers inagriculture appears to have started to contract. Employment in the private sector and SMEs is growing robustly. Wage distribution is widening and the correlation with productivity i s improving. Nevertheless, addressing the existing distortions requires, apart from significant labor reallocation and labor policy reforms, a broader spectrum o f policy corrections, including changes in education, agriculture, the enterprise sector, and the taxation system. 92 This chapter was preparedby CatalinPauna (ECSPE). 111 5.5 The liberalization o f trade with the EUinindustrial goods has placed important pressures on Romanian companies to cut costs and restructure internally. However, the solid exports performance is limited to a relatively small number o f activities and is dominated by unskilled, low technological products (see Chapter 2). This contrasts sharply with the poor performance o f the largely unrestructured inward-oriented industries, where large scale productivity-driven labor adjustments are yet to occur. 5.6 Inthis context, the chapter is aimed at addressing two inter-related policy issues. First, it evaluates the extent to which labor policies support labor market adjustment. Second, it identifies the challenges to labor market policy arising from integration with the EUand from meeting the European Employment Strategy objectives. The central question that the chapter attempts to address is the extent to which Romania has succeeded inreallocating labor toward more efficient uses. 5.7 The study argues that the current labor market framework fails to provide the necessary relief by obstructing the reallocation o f labor from the declining enterprises and sectors to the growing ones. In particular, the newly adopted Labor Code discourages labor mobility and complicates further progress in structural reforms by imposing large transactions costs on companies. This risks affecting the speed and the direction o f the reforms and slowing down the process o f EU convergence. The Code introduces new impediments to enterprise restructuring and, thereby, adversely affects the overall competitiveness o f the Romanian economy. The chapter concludes that from a labor market perspective additional effort is needed to reduce the existing legal and institutional rigidities. A. PARTICIPATION, EMPLOYMENT, NON-EMPLOYMENT AND 5.8 . Following the collapse in output at the beginning of the transition, labor force participation and employment rates are decreasing, while non-employment continues to expand. Romania began the transition with high labor participation rates in comparison with countries with similar GDP levels.93 These higher rates were due to the extensive use o f labor resources, and an especially high female participation, to compensate for low productivity and labor mobility. The country inherited no open unemployment but an abundance o f hidden unemployment, a distorted sectoral allocation o f labor, and a compressed wage distribution which bore little, ifany, relationship to productivity. 5.9 The collapse inoutput at the beginning o f the 1990sprompted a gradual reduction inboth participation and employment. The main policy response to the sharp output contraction was, however, the adjustment o f real wages, with a view to limiting job destruction. This was complemented by a series o f early retirement programs. While this policy approach succeeded in containing the increase in open unemployment, it concomitantly pushed workers out o f the labor force and into low productivityjobs, primarily in agriculture. This route was eventually abandoned inthe late 1990sandsomeofthe initial imbalances areintheprocessofbeingcorrected. 93Barr, N.(1994), Labor Markets and Social Policy in Central and Eastern Europe, OxfordUniversity Press. 112 5.10 Table 5.1 shows the changes in output relative to the changes in employment and real wages94between 1989 and 199495,and between 200196 and 1989. In many respects, the table synthesizes the reform strategies pursued by Romania and other transitional economies in adjustingto market conditions. Ifthe initial output collapse was accepted as inevitable, owing to increased external competition and the abolition of CMEA, the table shows that while some countries responded with sharp adjustments in employment wherejob cuts exceeded output fall, Romania resisted making painhl reductions in employment to match the contraction in GDP. Although by 1994 the Romanian economy had contracted more than 20 percent relative to 1989, its employment had shrunk by only 9 percent, with the adjustments mostly taking place in real wages, which had fallen by around 37 percent, and inhours ofwork.97 Table 5.1 Output,EmploymentandWage Adjustment (percent change relative to initial levels) Country AGDP AEmployment AWage 1994/1989 2001/1989 1994/1989 2001/1989 1994/1989 2001/1989 Bulgaria -23 -17 -24 -31 -48 -57 Czech Republic* -10 6 -10 -12 -24 1 Hungary -16 12 -26 -23 7 14 Poland -18 15 -14 -16 -12. 32 Romania -21 -14 -9 -22 -37 -37 Slovak Republic -21 9 -16 -15 -34 -25 * Forthe CzechRepublic GDP changesare computedbetween 1994and 1990, and2001 and 1990. Source: Staff computations usingOECD, ILO andNIS data. 5.11 The delay in the employment adjustment failed, however, to yield long-term positive results. Comparisons o f the changes between 2001 and 1989 prove the unsustainability o f the employment preservation policies promoted inthe early phases o f the transition. By 2001, in all CEECs the adjustment in employment was significantly larger than that in output. InRomania, between 1989 and 2001, employment contracted by 22 percent while GDP shrank by only 14 percent. Furthermore, the decline in employment was not matched by a proportional rise in registered unemployment, as many workers chose to respond to the adverse shock by leaving the labor force. 5.12 Table 5.1 also shows the divergence inthe dynamics o f real wages between the countries that by 2001 had recovered and surpassed the GDP level o f 1989, and slower reformers. In the case o f the former, output recovery was accompanied by an increase in real wages relative to 1989, reflecting substantial productivity gains. By contrast, wages inRomania, Bulgaria and the Slovak Republic were at levels visibly inferior to those of 1989. In Romania, the real consumption wage9*was, in2001,37 percent lower than in 1989. 94Consumption wages, CPI deflated. 95Considered the mid-point inthe analysis for the purposes o f the work. 96The last year for which comparative statistics for the selected transitional countries are available. 97The average annual number of hours effectively worked in industry was 1588.2 in 1994, down from 1759.2 in 1990(NIS data). 98CPI-deflated real wage. 113 5.13 The substantial increase in non-participation, especially in the early years o f the transition, seriously affected the balance between the contributors to and the beneficiaries o f public transfers, principally between wage eamers and pensioners. The large-scale introduction o f early retirement programs, with the false hope o f vacatingjobs for new entrants, has proved a main factor inthe deterioration o fthe balance of the pension scheme andthe subsequent perverse employment dynamics. It has placed significant pressure on the public budget, and has generated a fiscal gap that is difficult to correct in the medium term, owing to the rigidity it induces in public expenditure. Between 1995 and 2001, the PAYG pension system posted annual deficits estimated at 0.8-1.6% o f GDP.99 5.14 As Table 5.2 reveals, both participation and employment, although still declining, remain above the average for CEE and countries with a comparable GDP per capita, and slightly below the EU average. Moreover, Romania exhibits higher participation rates than some current EU members, such as Italy or Greece. The participation o f women remains lower than that o f men, with the gap, on average, inferior to that inthe EU. Table 5.2 Employment Rate, 15-64 Years, 1997-2002 (inpercent) 1997 1998 1999 2000 2001 Country Total Women Total Women Total Women -Total Women Total Women Czech 68.7 59.9 67.5 58.7 65.9 57.4 65.2 56.9 65.3 57.0 Republic Hungary 52.7 45.5 53.8 47.3 55.7 49.0 56.4 49.7 56.6 49.8 Poland 58.8 51.8 58.9 52.2 57.5 51.6 55.5 48.9 53.5 47.8 Romania 65.8 53.3 64.3 52.5 63.5 52.2 63.2 52.0 62.6 51.9 urban 59.8 56.4 55.6 55.1 Slovak Rep. 61.1 54.0 60.5 53.5 58.1 52.1 56.8 51.5 56.9 51.8 Greece 54.8 39.1 55.6 40.3 55.4 40.7 55.9 41.3 55.6 41.2 Italy 51.6 36.4 52.2 37.3 52.9 38.3 53.9 39.6 54.9 41.1 EU 60.9 50.9 61.7 51.8 62.6 52.9 63.8 54.1 64.1 54.9 Source:OECD, INS, and World Bank staff estimates. 5.15 Focusing on the aggregate figures for labor participation and employment is deceptive in the case o f Romania, owing to the high share o f employment in subsistence agriculture and the existence o f borderline employment categories. With the exception o f Bulgaria and Poland, the share o f employment in agriculture is much lower in the CEEs and the EU than in Romania. As evidence we have computed participation and employment rates for the urban labor markets, as involvement inagricultural activities occurs primarilyinthe rural areas. As the table reveals, urban employment rates are lower, on average, by 5-6 percent, more inline with the CEE average. 5.16 Furthermore, the existence o f large borderline employment categories in Romania substantially influences key indicators o f labor market performance, such as the employment to population ratio. To illustrate this aspect, this section provides altemative measures for the employment to population ratio, by excluding some labor categories which, under different interpretations, can also be classified as out-of-the-labor force, such as the unpaid family helpers, some involuntary part-timers or people in "technical" unemployment (Table 5.3). With these adjustments, the employment ratios decline substantially. The importance o f these categories 99Romania Public Expenditure and Institutional Review (2002). 114 goes beyond their impact on the aggregate statistics. They suggest that many are ready to work but that the limited job opportunities prevent them from doing so. Identifying the means o f bringing these groups into standard jobs would boost employment and result in important benefits interms ofhigher output, improved budgetrevenues, and reducedpoverty. Table 5.3 Alternative Measures of the Employment Population Ratio, 1994-2002 (percent o fpopulation aged 15-72) Employmentratio 1994 1996 2000 2001 2002 Standard 62.4 57.5 58.3 58.7 53.4 Excluding subsistence farmers (both own account workers and unpaid family n.a. 47.8 46.3 46.3 n.a.2 helpers) Excluding other own account workers and unpaid family helpers inagriculture 43.4 43.4 40.4 40.1 37.4 Excluding halfo f involuntary part-timers Excluding laid-off persons with a formal 43.3 43.2 40.2 39.9 37.3 attachment to the job 43.0 42.9 40.0 39.8 37.2 Notes: Subsistencefarmers include own account workers and unpaidfamily helperswho work intheir own agricultural household for their own consumption only. Involuntary part-timers are persons who declared having a part-time program, were activelylooking for work andavailable, andwhose actualtotalhoursof work at alljobs were fewer than 30. Laid-off persons with a formal attachment to the job are persons who worked zero hours due to technical unemploymentor unpaidleaveinitiatedby the employer. 2002 data is preliminary. Source: CEU Labor Project, "MeasuringNonstandardFormsof EmploymentandUnemploymentin Romania, Working Paper, 2003. 5.17 Employment rates also differ substantially between age groups. Compared with the average, these rates are particularly low for youth (age 15-24), at between 30-40 percent, and older persons (age 50-64), at 40-60 percent, suggesting limited labor opportunities for these labor categories and probably the segmentation o f the labor market in favor o f the prime-age working population. Indeed, unemployment i s highamong young persons, and a large share o f older persons choose to withdraw from the labor force, through early retirement or simply out o f discouragement.'00 5.18 One source o f concem is that Romania started the transition with a distorted employment composition, and the initial imbalance has been exacerbated through the transition, leaving the economy with a remaining heavy burden. One o f the most conspicuous features o f the Romanian pre-reform labor market, even when compared to that o f its neighbors, was the different structure o f employment across broad industrial sectors, involving an especially large number o f workers in (heavy) industry andagriculture, and a very low employment inservices. 5.19 Table 5.4 shows the employment structure o f Romania with that o f other CEECs and Westem European comparator"' economies at the beginning o f the transition. Incomparison with looDiscouraged workers are persons not working, available for work and not searching for a job for the following reasons: believe there are no available jobs or do not know where to seek work; believe that they do not have suitable skills; believe they will not findajob because o fage; have sought for ajob, but didnot find one. lo'For purposes o f comparability we divided Westem European countries into a Northem group (Denmark, Germany, the Netherlands, the United Kingdom) and Southem group (Greece, Italy, Portugal and Spain), to allow for the 115 the market economies, there was excessive employment in agriculture, mining and the manufacturing industries. Services were underdeveloped, most obviously in finance (which covers business and professional services in addition to banking, insurance, etc.), but also in trade and community services (which include health and education, as well as public administration). Construction and transport were much in line with a market economy profile. As Table 5.4 shows, Romania had the largest share o f agricultural employment (27.9 percent), one o f the largest manufacturing sectors (33 percent), and the smallest community services (15.3 percent) and trade (5.9 percent) sectors. Table5.4 Structureof EmploymentbyMainSector in 1989(percentoftotal) Sector Bulgaria EU EU Republic Czech Hungary Poland Romania Slovakia South North Agriculture 19.0 11.7 16.6 26.8 27.9 13.8 10.7 4.1 Mining 2.6 3.6 2.0 3.4 2.3 1.0 0.4 1.o Manufacturing 34.9 34.0 28.6 24.5 33.0 32.1 22.0 26.3 Electricity, gas, water 0.8 1.4 2.6 1.1 1.2 1.6 0.9 1.1 Construction 7.8 7.3 7.0 7.8 7.0 11.6 8.1 6.4 Trade 9.2 11.5 11.3 8.9 5.9 11.1 19.3 17.4 Transportation 6.8 6.5 7.7 7.2 6.9 6.4 6.0 6.0 Finance 0.6 0.5 0.8 1.o 0.3 0.4 6.1 8.6 Community services 18.4 23.5 23.4 19.3 15.3 22.0 26.5 28.7 DI-South 24.2 17.2 16.5 23.0 31.3 18.4 10.0 DI-North 27.3 19.6 19.6 27.7 33.4 21.6 10.0 Note: DI is a coefficient of departure, defined as the overall excess employment in the sectors where employment in the Eastern European country exceeds mean employment in the comparator countries. EU North includes Denmark, Germany, the Netherlands, and the United Kingdom; EUSouth includes Greece, Italy, Portugal, and Spain. Source:OECD-Labor Force Statistics (1998), ILO, country statistics and staffcomputations. 5.20 It is evident from the table that the extent o f structural imbalance differs considerably among countries. To capture this irregularity, the final rows of the table shows an "index o f departure" which measures the proportion o f the workforce in each country that would need to change sector to enable the country to attain the same structure o f employment as that o f a comparable Western European economy in 1989. As an approximation, the index o f departure measures the inherited "distortion" o f former centrally planned economies at the beginning o f the reform, in the sense o f departures from the average Western European economy. The larger the index, the higher the distortion is, and most likely the required adjustment and restructuring costs willbe. 5.21 Among the CEECs, Romania inherited the most distorted employment structure, even in relation to Southem Europe. As compared to the structure of employment in Southem Europe, the extent o f reallocation required inRomania, at 31.3 percent o f the workforce, was, at the outset, the largest. Inother words, inorder to attain a Southern European employment structure inRomania in 1989, on average 31.3 workers out o f 100inRomania would have had to change their broad sector, and presumably occupation, skills and even work area. The discrepancy is even larger if we considerable differences inemployment inagriculture, the average for the Northbeing 4.1 percent, while for the Southit is 10.7 percent. These differences may be attributable inpart to permanent features of the economic landscape such as climate andpopulationdensities, and inpart to historical and institutional considerations. 116 compare Romania with Northern Europe (33.4 percent), with most o f the difference coming from the large employment share inagriculture, and, to some extent, inmanufacturing. 5.22 Inthe context of the labor markets inRomania, restructuring therefore becomes anissue of industrial composition. Employment inthe private sector can grow simply because of changes inthe ownership o f enterprises, with no change inanyworker's place o f employment. The privatizationo f state manufacturing enterprises and the introduction o f new technology may raise efficiency, but will not remove the need to shed labor, andthus are not an altemative to labor mobility. A shift in industrial compositionwill ingeneral be associatedwith a decline inthe state sector anda growth in private sector employment, but in the context o f labor reallocation the fundamental issue is the change inthe composition o femployment byindustrial sector. 5.23 Fundamentally, labor reallocation is necessary because the Romanian companies have to adapt to a new environment. The gradual integrationo f the country into the EUeconomic structures and the liberalization o f trade inindustrial goods with the EU in2002 have resulted inincreased extemal competition, forcing enterprises to restructure and inevitably to shed workers. At the same time, sectors that received little attention inthe past, mostly in services, are expected to grow, and their demand for labor should pick up. Accelerating the removal o f barriers to the movement o f labor toward services, away from the declining industrial sectors and agriculture-experienced not only by the CEECs, but also bythe later entrants into the EUsuch as Greece, Portugal andSpain- could speedup real convergence and, consequently, couldbe felt inreal incomes andproductivity. Table5.5 StructureofEmDlovmentbvMainSector in2001(Dercent of total) ~~ Sector Bulgaria Slovak EU EU Republic Czech Hungary Poland Romania Republic Souih North Agriculture 26.3 4.7 6.2 19.1 42.3 6.1 7.3 2.2 Mining 1.2 1.4 0.3 1.9 1.4 1.0 0.3 0.4 Manufacturing 20.1 27.7 24.8 19.9 18.9 26.1 20.5 19.6 Electricity, gas, water 2.0 1.9 2.1 1.9 1.9 2.5 0.7 0.7 Construction 4.3 9.1 7.1 6.7 4.0 8.0 9.5 7.5 Trade 15.3 16.1 17.9 15.9 10.1 15.4 21.0 18.4 Transportation 7.3 7.7 8.1 6.0 4.9 7.6 5.6 6.3 Finance 5.6 7.6 7.7 6.8 1.9 6.7 9.6 13.7 Community services 17.8 23.9 25.9 21.7 14.6 26.5 25.4 31.3 DI-South 22.9 11.6 8.7 15.0 37.3 11.2 10.6 DI-North 27.7 15.8 12.4 19.9 42.3 14.6 10.7 Source: Staff computations using ILO and OECD statistics. 5.24 Turning to the issue o fthe amount o f restructuringachieved, an indicative measure can be estimated by replicating the previous table for the most recent year for which comparative statistics are available, namely, 2001, and contrast the structural imbalances with those that existed at the outset. These estimates are shown in Table 5.5. The significant feature here is that the imbalances appear as great now as they were in 1989, while the proportion o f the workforce employed in manufacturing has fallen everywhere, including by a remarkable 14 percent in Romania, the proportion employed in trade and finance has increased, and employment in agriculture has also risen. Most remarkably, the departure index has actually slipped back. This suggests that the gap with Westem Europe appears to have widened, at 37 percent relative to the South, and as much as 42 percent relativeto the North. 117 5.25 The apparent lack o f progress in restructuring despite substantial job losses reflects that progress in other countries i s faster than in Romania. Market economies are themselves adjusting faster to the pace o f market changes. Transition economies need a faster pace o f employment reallocationsimply to avoid falling further behind. 5.26 The depth o f employment adjustment has varied significantly across economic sectors. Probably a part o f the observed change insectoral employment is not attributable to the reallocation o f labor across sectors, but simply reflects the uneven incidence o f macroeconomic recession and sectoral shocks. A sector with excess employment at the outset maybe sheddinglabor, but ifits rate o f employment decline i s slower thanaverage, its share o f employment will be increasing. 5.27 InRomania, for example, the share of employment inutilities such as electricity andwater has risen even though the initial shares inthese sectors were already too high, presumably because they are less vulnerable to recession. At the same time, the share o f employment inmanufacturing severely declined, from 33 percent to 19 percent. On the other hand, one expects the highly unionized industries, such as mining, to be more difficult to reform, and to continue to maintain some o f their privileges, including subsidies, and hence to have higher levels o f employment for a longer time. The slow growth o f the services, where some sectors, such as hotels and restaurants, have actually contracted, suggests the limited progress achieved in creating the conditions for expansion, andhints at apotentially large informal sector. 5.28 Another important reason for the slow sectoral reallocation o f labor is related to the role o f the agricultural sector in Romania as an "employer o f last resort." In some transitional countries, particularly those in which employment in agriculture was initially high, the collapse o f regular, urban employment in industry has led to a reversion to small-scale farming. There is a vicious tendency for those countries that had the greatest surplus o f employment in agriculture in 1989 to experience the slowest declines, or even, as in the case o f Romania, an actual increase in agricultural employment. 5.29 Romania is in fact the only country among the EU candidates in which the workforce in agriculture increased. A large agricultural sector interms o f employment suggests the existence o f a two tier system, consisting o f a large number of people involved in low productivity jobs, small farming, and subsistence agriculture coexisting with large, potentially profitable but unreformed farms (see Chapter 4). The case o fPoland shows, a country with a large inherited agricultural sector (27 percent o f employment in 1989), where employment in agriculture declined by almost 1.8 millionpeople between 1989 and2001 (reaching a share o f 19percent). 5.30 The task o f labor reallocation across sectors in Romania must therefore fall on the extemal labor market. This can be achieved only through the physical mobility o f workers leaving one enterprise, likely in a declining sector, and finding work in another enterprise. It is thus the dimension o f restructuring in which any deficiencies inthe labor market-such as impediments to labor mobility, poorly developed employment exchange facilities, excessive specific training on the previousjob which is o f limited use in altemative employment, or industry rents -- are likely to have the most severe effects. 118 B. JOBDESTRUCTION,JOBCREATION,AND JOBLESSNESS 5.31 As Romania becomes increasingly integrated into the EU, trade barriers are removed, trade flows are liberalized, and competition increases. People are pushed into changing their jobs, companies, sectors and even occupations. Romania shows a highdegree o fjob destruction occurring not only inRomania, but throughout the region. It is therefore crucial that a successful reform must entail enhanced labor flexibility that facilitatesjob creation so labor cohorts released by over-manned companies are absorbed quickly by the growing sectors and firms in order to avoid highlevels and long spells o f costly non-employment. Table5.6. ChangeinEmployment,2001/1989 (millions) Mining -0.079 -0.129 -0.087 -0.304 -0.109 -0.003 -0.021 Manufacturing -0.904 -0.524 -0.452 -1.343 -1.588 -0.247 0.085 Electricity, gas, water 0.023 0.010 -0.05 1 0.087 0.066 0.012 0.010 Construction -0.206 0.038 -0.072 -0.363 -0.337 -0.120 0.959 Trade 0.055 0.147 0.136 0.745 0.434 0.049 1.716 Transportation -0.076 0.013 -0.069 -0.370 -0.238 0.001 0.277 Finance 0.140 0.335 0.261 0.795 0.165 0.134 1.300 Community -0.265 -0.110 -0.154 -0.205 -0.113 0.014 1.313 Total Change -1.35 -0.63 -1.07 -2.80 -0.25 -0.37 4.56 Job Creation: 0.22 0.54 0.40 1.6 2.1 0.2 5.7 % change 5.1 10.1 s.1 9.6 19.5 8.4 27.9 % oflabor force 4.0 7.5 5.7 6.0 13.6 5.3 Job Destruction: -1.57 -1.17 -1.47 -4.4 -2.4 -0.6 -1.1 % change -36.6 -21.7 -21.o -29.7 -26.0 -21.8 -23.4 -5.4 % of labor force -28.5 -16.3 -16.4 -15.6 -15.8 Memo items: Labor force (2001) 5.5 7.2 7.0 26.8 15.4 3.8 Population(2001)' 8.0 10.2 10.1 38.6 22.3 5.4 _.. Source: World Bank staff estimatesbasedon ILO, OECD and NIS statistics. 5.32 As Table 5.6 shows, inRomania and inthe countries covered by the analysis, millions o f jobs have been destroyed over the transition period. In Romania, during the period of investigation, job destruction exceeded 2 millionjobs, a figure surpassed only by Poland, with 4.4 million, although the labor force there is significantly larger. Job destruction relative to the labor force is, however, 15.6 percent, which is at the low end o f the CEECs. In the declining industrial sectors, especially in manufacturing but also in construction and transportation, more than one in three workers employed in 1989had lost their jobs bythe end o f 2001. 5.33 Job creation on the other handhas been low. It appears highat first glance only because o f the expansion o f employment in agriculture. However, the job creation to labor force ratio excluding agriculture, is about 4 percent, which, together with Bulgaria, is the lowest ratio o fjob creation among the CEECs. 5.34 Not surprisingly, the probability o f an unemployed person's staying inthe pool for more than one year are higher than 50 percent, and the likelihood o f that person's finding ajob is one in three (see Table 5.7). Around 15 percent o f the unemployed withdraw from the labor force. 119 For a recent graduate, the chances o f entering directly into unemployment are almost half those o f finding a job. An employed person is far more likely (the probability ratio is 4.5:l) to move out of the labor force than into unemployment, most likely through discouragement. The figures also suggest a surprisingly high number (one in five) o f homemakers that choose to enter (or reenter) the labor market. This indicates a strong added worker effect, presumably to compensate for loss o f income due to the disappearance o f the mainjob(s) inthe household. While this might capture some seasonal effects as well, its magnitude requires more detailed research. Table 5.7 Transition Probability Matrix, 4 4 2000 4 4 1999 (percent) - Employed Unemployed Out of labor force Total Employed 91.2 1.6 7.2 100 Unemployed 32.3 52.2 15.5 100 Out oflabor force, o f which: 12 2.5 85.5 100 Student 7.6 3.4 Homemaker 19 2.2 Note: The first column refers to the state in 44 1999.The first row refers to the state in 44 2000. Source: World Bank staffestimatesbased on LFS data. 5.35 As a result, a particularly largeportion ofthe population inRomaniais represented bythe workers displaced inthe restructuring o f the industrial sector and by new entrants and re-entrants into the labor market, some directly into unemployment, who became discouraged and stopped searching. As Table 5.8 reveals, the discouraged population is high.The vast majority, around 90 percent, have become discouraged because there are no jobs available. Particularly worrying is the fact that more than halfo f the discouraged population is below 35 years old. Table 5.8 DiscouragedPeople, by Reason and Age Category, 1996-2001 1996 1997 1998 1999 2000 2001 Discouraged persons (total) 279,358 249,875 263,988 304,908 305,163 340,720 Reasons for discouragement believe there are no availablejobs or do not know where to seek work 46% 39% 41% 41% 33% 27% believe do not have suitable skills 3% 2% 2% 2% 1% 1% believe will not findajob because o f age 4% 5% 5% 5% 7% 8% sought for ajob before, but didnot find one 47% 55% 52% 52% 59% 64% Distribution of the discouraged by age 15-24years 50% 43% 41% 50% 33% 30% 25-34 years 21% 27% 27% 21% 25% 25% 35-49 years 22% 23% 26% 23% 31% 33% 50-64 years 7% 7% 6% 6% 11% 12% Source: World Bank staff estimates basedon NSI. 5.36 The characteristics and behavior of the discouraged, such as their labor supply elasticity, are largely unknown. Identifyingmeasures and incentives to attract this large group into standard employment should help ease their current situation considerably. Treating the discouraged as unemployed would also significantly influence the official statistics. For example, the alternative 120 measure o f the unemployment rate"* in 2002 would become 15.3 percent, up from the official 11.1percent. C. EMPLOYMENT DISTRIBUTION CHARACTERISTICS AND 5.37 Aggregate employment appears to be on a gradual recovery path, but sectoral imbalances persists suggesting that substantial labor reallocation might follow. Throughout the transition, the employment dynamics in Romania have been correlated with the fluctuations in output and characterized by periods o f growth alternating with intervals of contraction (Figure 5.1). Benefiting from a protractedupturnineconomic activity, employment inRomania, as inCEECs, is currently on a slow recovery trend, following the initial collapse. While inthe case o f the EU frontrunners employment recovery has been sustained since 1994, in Romania (and also in Bulgaria) an employment upturn has recently begun. Inthis context, the main question concerns the sustainability o f the current dynamics. Figure5.1 EmploymentDynamics, 1989-2001, SelectedCEECs (1990=100) Emplqymerrtdynamics, 1989-2001[selected EUcandidates, 1990=100) 160.00% 140.00% 120.00% 10000% 80.00% 60.0036 40.00% 20.00% 0.00% 5.38 One of the most conspicuous reflections o f the structural transformations that took place in the Romanian economy in recent years is the significant increase in employment in the private sector (Figure 5.2). Between 1996 and 2001, employment inenterprises having a private entity as majority shareholder has expanded from 50 to 70 percent o f total employment. At the same time, lo*Defined as total unemployed plus discouraged workers, as a percent o f total labor force plus discouraged workers. 121 employment inthe broadpublic sector has decreasedcontinuously. In2001 it contributed around 25 percent to total employment. Figure5.2 EmploymentDistributionby EnterpriseOwnership(percent oftotal) Employmentdistribution by ownership (%of total) 80.0% 70.0% 60.0% 50.0% , 40.0% +Fubiic ' 30.0% +Rivate I 20.0% k t 1 10.0% ~ --w- Other (including 0.0% cooperative) ~ I I 1996 1997 1998 1999 2000 2001 ~ Source:NIS. 5.39 As privatization advances, with most of the companies in the energy sector-a large employer, earmarked for privatization-employment in the private sector is likely to expand. The share o f employment inthe S M E sector has also increasedconstantly since the inception of the transition, reaching almost 50 percent at the end o f 2001. The figure exceeds the 40 percent threshold, considered the minimum share below which economies are unlikely to take off in terms of growth.lo3 However, some of the growth in the S M E sector was attributable to the favorable fiscal treatment received at the expense of the larger firms. The removal of most of the fiscal facilities for the SMEs in 2001, through the introduction o f a new non-discriminatory profit tax and VAT legislation, might explain the slowdown inemployment growth inthe SMEs that year. 5.40 By sector, employment in SMEs is concentrated primarily in retail services, trade, construction, and hotels and restaurants, and has a low participationinindustry. Surprisingly, the number of employees in services suffered a decline between2000 and 2001. The expansion of the large companies, such as the entry into the retail trade market of the international chains, appears to have contributed to this effect in spite of the improved access of the SMEs to financing sources. Transition -the First Ten Years -Analysis and Lessonsfor Eastern Europe and the Former Soviet Union, World Bank, 2002. 122 5.41 Two features o f the employment structure cause some concern. First, the share o f employment for those with primary education only remains high in Romania, even by regional standards (Table 5.9). Together with Poland, Romania i s the country with the highest share o f employment for those with primary education only, at 14.8 percent o f the total. Bulgaria, the Czech Republic, Hungary and the Slovak Republic all have a significantly lower level o fprimary education employment. At the same time, the contribution o f a highly educated labor force to total employment is comparable with that in CEECs, with only Hungary and Poland exhibiting higher shares. Table5.9 StructureofEmploymentby EducationLevel Tertiary Secondary Primary Bulgaria 11.6% 86.3% 2.1% CzechRepublic 12.2% 87.7% 0.1% Hungary 16.4% 82.7% 0.9% Poland 14.7% 67.8% 17.5% Romania 12.9% 72.2% 14.8% Slovak Republic 12.3% 81.O% 6.7% ~~ ~~ ~ Source: ILO. BulgariaandCzechRepublic in 2001, the rest in 2000. 5.42 Improving the average level of education for the labor force requires further reforms in the education system, including more emphasis on quality education oriented towards the poor and low-skilled, and the adaptation o f curricula to the needs of the labor market. Limited spending on education, at around 3.4 percent o f GDP, presents risks o f the de-skilling o f the human capital and has a negative impact on productivity and labor market flexibility. Inthe long run,thistranslates into anirrecoverable output loss. 5.43 The second cause for concern is the fact that the share o f part-time non-agricultural employment i s low. As Table 5.10 shows, although part-time employment inRomania is highby CEE standards, the vast majority o f this employment is concentrated in agriculture. Taking agriculture out, the share o f part-time employment is lower than anywhere else in Eastern Europe. Table5.10 Part-timeEmploymentin Selected Countries (percent o ftotal employment) Country 1996 1997 1998 1999 2000 2001 CzechRepublic 6.1 5.9 5.7 5.6 5.3 4.8 Hungary 5.5 5.8 3.5 3.5 3.2 2.8 Poland 10.6 10.5 10.4 10.9 10.8 10.3 Romania, of which: 13.8 14.7 15.4 15.7 16.3 16.4 non-agriculture 1.9 1.8 1.8 1.7 2.0 2.0 Slovak Republic 2.7 2.3 2.3 2.1 2.1 2.3 Italy 6.5 6.9 7.4 8.0 8.5 8.5 Netherlands 36.5 37.1 37.9 38.3 39.5 UnitedKingdom 24.6 24.9 24.9 24.8 24.9 24.8 Note: Employmentby full-time/parttime distinctionbasedonnationaldefinitions. Source: World Bank staffestimates basedon OECD and NIS. 123 5.44 The future dynamics of part-time employment inRomania will dependon the net effect of two factors. First, the adoption of a new Labor Code at the beginning of 2003 allows for the establishment o f standardpart-timeand temporary employment agency (TWA) types of contracts, which did not exist until recently. This should encourage part-time employment, as out-of-the- labor-force individuals for whom full time employment i s not an option might be tempted inthis way to reenter the labor force. Second, the elimination of the civil c~ntracts,'~~coupled with the high statutory taxes on labor, is likely to push workers to enter the informal sector, rather than registering as part-time workers (and paying the appropriate social insurance contributions). If the latter were to prevail in the short run, in the medium term the recent legislative changes should benefit the labor market, judging from the experience o f other European countries, and part-time employment would be expectedto increase. D. WAGESANDPRODUCTIVITY 5.45 Wage distribution has widened and the linkage between wages and productivity has improved, but occasional slippages occur and rigidities (and rents) remain. Wages have yet to become the main mechanism for the allocation of labor resources. In terms of wages, the big losers appear to be agriculture, (where in 2001 wages were 71 percent of the average, down from 106 percent in 1990) and construction (at 85 percent in2001, relative to 111 percent in 1990). The big gainers are the sectors in services, and primarily the financial services, where salaries in 2001 were more than two times higher than the average, up from below the average in 1990, and transport and storage. Overall, wages inindustry have fared better than the average, thanks to male wages, which were 21 percent higher than the average. In general, male wages are higher than female wages, withthe exception of agriculture andconstruction. Figure 5.3 Hourly Labor Cost inIndustry and Servicesin2000 (in euros Der hour) HOURLY LABOURCOST IN INDUSTRYAND SERWCESIN 2000 (in euros per hour) 10 1 s+ ............................................................................................................................................... " I 4 2 0 Bulgaria Romania Slovakia Hungary Czech Poland Republic Source: Vienna Institute for Intemational Economic Studies (WIIW), research report no. Casual evidence suggests that in the past most of the non-agriculture part-time contracts took the form of civil conventions, where holders didnot pay social insurance andrelated contributions. 124 5.46 Despite the existing rigidities, unit labor costs remain attractive relative to the EU area and CEECs. Although unit labor costs have increased (ineuro terms), Romania remains, together with Bulgaria, the country with the cheapest labor among the EU candidates. Romania's labor cost, measured in euro-equivalent, increased by 48 percent between 1995 and 2002. However, as illustrated in Figure 5.3, at around 1.5 euros per hour in industry and services, it remains well below that inthe acceding countries and over five times lower than inPortugal, the EUmember state with the lowest labor cost (at around 8 euros per hour). Following in the steps o f the countries which joined the EU this year, Romania is likely to face further real exchange appreciation as the inflation differential with the Eurozone will not lessen fully (owing to price catch-up) and the labor costs, expressed ineuro-terms, is likely to increase. 5.47 If the wage increases match the productivity gain differentials, external competitiveness would not deteriorate. Hence, inthe short runthe departure o f real wage growth from productivity advances, which we are currently witnessing, constitutes more o f a threat for the external balances, and we can already see a widening of the current account gap. However, recurrent unsustainable wage expansions represent not only a source o f inflationary pressures and external balances deterioration, but can constitute a bad signal for the attractiveness o f the business environment, with anadverse impact onFDIinflows. 5.48 A substantial burden is placed on labor by high social insurance contributions (see Table 5.11). At 49.5 percent o f the gross wage, these contributions are the highest among the selected CEECs, representing a disincentive for job creation andpushing labor into the informal sector. The bulk o f this burden is made up o f the contributions to the pension system, at 31.5 percent. At the same time, the effective collection o f the contributions is among the lowest in Europe, with only Albania displaying a poorer performance. Table 5.11 Non-wage Components of Labor Costs in Selected Selected CEECs '' Social security contributions Healthcare contributions Contributionsto unemployment fund Total Employer's Employee's Total Employer's Employee's Total Employer's Employee's Total Bulgaria" 0.37 0.02 0.39 0.00 0.00 0.00 0.04 0.01 0.05 0.44 Czech Republic 0.20 0.07 0.26 0.09 0.05 0.14 0.03 0.00 0.04 0.43 HU%ary 0.22 0.05 0.27 0.11 0.03 0.14 0.00 0.00 0.00 0.41 Poland 0.10 0.10 0.20 0.07 0.09 0.16 0.04 0.00 0.04 0.39 Romania3' 0.22 0.095 0.315 0.07 0.065 0.135 0.03 0.01 0.04 0.495 EU 0.24 0.13 0.37 1/ Contributionsare based on the gross salary receivedby the worker. Datarefer to July 1999. 2/ Datarefer to the third categoryof labor. 31Romania-data refer to 2003. Source: IMF, CountryReports(various issues), RomanianMinistry of Public Finance. 5.49 Despite several successive reductions inrecent years, taxes on labor remain well above the EU average. Owing to a chronic imbalance between the number o f contributors to the PAYG pension system and the number o f pensioners, and the inadequate collection o f the social security contributions, the high rates will not solve the problem of the social assistance budget, which has been in an endemic deficit inthe last years. This deficit was estimated at around 1percent o f GDP 125 in 2002 alone, and declined slightly in 2003. Lowering labor taxation should make a major contributionto bringinginformal employment into the formal sector. 5.50 The informal sector i s sizable in Romania. Estimated at over 24 percent o f the labor force, informal employment i s higher only in Bulgaria, among the comparator countries. The large increase in p ~ v e r t y " during the transition, following the sharp decline in output and ~ incomes and coupled with limited opportunities for finding employment inthe formal sector, has contributed to a significant expansion in informal economic activities, not only inRomania but throughout the CEECs. 5.51 The evidence suggests that the relationship between poverty and the underground economic sector is strong.'06 This indicates that informal economic activities are used as a survival strategy for the poor to cushion negative income shocks during the transition. Informal economic activities thus decrease poverty and contribute to reducing inequality in Romania. However, a systematic diagnostic o f the size, determinants and effects o f a seemingly significant underground economic sector inRomaniahas yet to be undertaken. Figure 5.4 RealNet Growth and Productivity - 14.0% r------------- +Productivity in Industry,y/y --- I -t-Realnetwagegrowth,yly. 1 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% -2.0% Source:World Bank staff computationsusingNIS data. 5.52 To contain unjustified wage growth in the SOE sector, wage bill ceilings have been imposed as part o f successive Stand-By agreement^"^ with the IMF. In addition, the government has promoted a series o f legislative acts that attempt to link wages to performance, such as by connecting management salaries inenterprises with key performance indicators,lo'by including the reduction in the level o f payment arrears and claims, and by observing wage bill ceilings. Inaddition, company performance monitoring has been strengthened. ' O sSee Romania Poverty Assessment,World Bank, 2003. Io6Kim,B.-Y. (2001), "Poverty and Informal Economy Participation in Transition Countries: Evidence fiom Romania", EssexUniversity Working Papers, UK. lo' The most recent Stand-by Agreement was successfully completed inOctober 2003. log For example, Emergency Ordinance 7912001. 126 5.53 Despite these attempts, the policies pursued to date have achieved limited success and overspending has often occurred inthe monitored SOEs, suggesting that the problems are rather of a corporate governance nature. The EU Joint Assessment o f Employment Policy Priorities (2002) recommends that Romania "develops together with the social partners a wage setting system that can be supportive o f economic and labor market reforms and able to guide skill formation, mobility and reallocation o f labor and ensure that wage developments are employment friendly." 5.54 Inrecent months real wage growth has exceeded productivity gains, as Figure5.4 shows. This comes after a period o f sustained productivity growth. The sharp increase in real wages occurred at the beginning o f 2003 and appears to continue in 2004. This trend is primarily attributable to a rise in the minimumwage o f around 43 percent in January 2003. The increase has brought the minimum wage to around 40 percent o f the gross average wage, in line with figures for other CEECs. A further substantial hike in the minimumwage in the future, pursued by the trade unions, would complicate the situation even more and would reduce employment opportunities for those at the lower end o fthe wage distribution. 5.55 Evidence sugge~ts''~that around 9 percent o f the employees receive the minimum wage"' and a further 24 percent are just above this level, including many young and low skilled workers, most o f them located in retail trade, construction, manufacturing, transport and public administration. In Romania the private sector ays less than the SOEs. In the private sector, 17 percent of workers receive the minimumwage" and a further 32 percent are paidjust above the minimumwage. By contrast, less than 1 percent o f workers in the SOEs are at the minimum wage (Table 5.12). Table 5.12 EmployeeDistributionby MonthlyWage and Ownership,October 2002 Less US52.6 US90.2 USU50.4 US$210.5 US%300.8US451.2 Over than US$52.6* - - - US%52.6 US90.2 US%150.4 US$210.5 US%300.8 US%451.2 US$601.6 US601.6 State 0.05 0.07 6.01 26.38 27.42 21.35 13.55 3.33 1.84 Majoritystate 0.13 0.63 9.86 22.99 20.94 22.71 13.09 4.70 4.96 Majority private 0.56 1.50 20.46 33.25 20.76 12.52 6.18 2.31 2.46 Private 0.37 17.24 32.29 27.64 11.51 6.03 2.78 0.97 1.18 Foreign 0.14 6.28 31.87 27.48 13.24 7.83 4.75 2.66 5.75 Note: * Minimumgross monthly wage inOctober2002. Source: World Bank staffestimatesusingNIS data. 5.56 Furthermore, considerable segmentation persists in the labor market, reflecting the protection afforded by the companies enjoying soft budget constraints. Earning functions analyses' l2suggest that wages o f otherwise identical individuals from a labor market perspective differ significantly across industries, occupations, firm sizes and locations, after controlling for I O 9Staff computations using NIS data o f October 2002. 'lo US$ 52.6 per month in October 2002, at the October average exchange rate (ROL/US$). Anecdotal evidence also suggests that many private employers understate the salaries o f their employees to avoid paying highpayroll taxes and social contributions. This might explain some o f the figures. 'I2 Using data fiom the Integrated Household Survey we highlight the main determinants of wages. The detailed results are presented inAnnex 4. 127 other relevant variables such as general human capital characteristics or for compensating differentials. Jobs and industries do pay rents, and they range from 41 percent in mining to 18 percent inthe energy sector above manufacturing wages, ceteris paribus. At the other end, wages in education are on average 26 percent, and inhealth 15 percent, lower than inmanufacturing. Unskilled workers are paid on average between 20 and 26 percent less than skilled workers, with the difference lower for females. 5.57 Exposed to competition, private companies pay less than state firms, by 6 percent in the case o f males while in the case o f females the difference i s statistically insignificant. The regression fbrther shows that Roma workers are not paid less than others. In general, wages increase with the size o f the locality, suggesting betterjob opportunities inlarger cities. Wages in the Northeastern region, Moldova, are the lowest. The returns to education, at around 4 to 5 percent for an additional year o f schooling, are relatively similar for men and women and lower than inmature market economies, althoughthey show a gradual increase.113 Table 5.13 Estimated Male-Female Wage Differentials in 2000 Males equation Females equation coefficients coefficients Percent Percent Wage difference 0.232 100 0.232 100 Controlling for differences in characteristics: Experience 0.040 17.0 0.040 17.2 Education -0.008 -3.3 -0.007 -2.9 Marital status 0.008 3.3 -0.002 -0.7 Industry 0.047 20.1 0.031 13.4 Occupation 0.029 12.6 0.026 11.0 Legal form o fthe company 0.007 2.8 0.003 1.3 Nationality 0.000 0.1 -0.001 -0.2 Population -0.001 -0.5 -0.001 -0.6 Region 0.002 0.9 0.001 0.4 ln(D+l) 0.109 46.9 0.182 61.2 D* 0.115 0.200 D*=coefficientofdiscrimination(Oaxaca). Source: World Bank staff estimates using Integrated Household Surveydata. 5.58 Concerning male-female wage differentials and labor market discrimination, the analysis shows that the wage gap between men and women is around 23 percent114 (Table 5.13). The average value o f the Oaxaca'15 discrimination coefficient is around 15 percent, lower than in most o f the market economies, where it ranges between 20 and 30 percent. Under the specification presented in Annex 4, differences in education, experience, marital status, legal form o f the company, and industrial and occupational segregation explain around half o f the `13Returns on educationwere around 3 to 3.3 percent for an extra year o f school in 1995. `14 * l5Oaxaca, Widening from around 15 percent in 1995. R., 1973, "Male-Female Wage Differentials inUrban Labor Markets," International Economic Review, vol. 14, no. 1, pp. 693-709. 128 wage differentials between males and females (see Table 5.13). The rest can be attributable to discrimination inthe labor market. 5.59 The investigation points to the existence of considerable occupational segregation, with women concentrated mostly in low paying light manufacturing industries, such as textiles or clothing, while males dominate numerically in the heavy, and better paying, sectors such as miningandheavy industry. E.LABOR MARKET POLICIES 5.60 Employment protection legislation and the passive labor market policies promoted by the Romanian government did not substantially hinder labor market flexibility, as is evidenced by the highjob destruction and creation before the adoption o f the new Labor Code (Box 5.1). This might change with the introduction ofthe new Code, which many employers fear would increase hiringand firing costs andadversely affect job creation andcompany competitiveness. 5.61 The Code increases transactions costs for companies and thus risks having a negative impact on the perception o f the investors, both domestic and foreign, about the business environment. Romania has the potential to draw levels o f FDI that are well superior to the current figures, and this requires the establishment o f an attractive business climate, including the existence o fprogressive labor legislation. 5.62 The World Bank PAL program, currently under preparation, supports the government's amending o f the Labor Code in order to enhance growth and job creation, to stimulate investment, and to improve the competitiveness o f the Romanian economy. 5.63 The Ministry o f Labor, Social Solidarity andFamily (MLSSF) and its agencies, primarily the National Employment Agency, are responsible for drafting and implementing labor policies. In recent years the institutional capacity of these agencies has been consolidated and their responsibilities have received further clarification. The Employment Agency administers, apart from the Unemployment Fund, the Active Labor Market Policies (ALMPs). Guided by the Strategy for Employment for 2002-04, 116 the Agency employs a spectrum of active measures, including public works, employment andbusiness start-up subsidies, training, job counseling and brokerage, etc. 5.64 The framework for labor market policies has been strengthened in recent years. Overall expenditure on labor policies is comparable with that in other CEECs, but is lower than that in the EU-15. Expenditure on ALMPs, increased starting with 2001 and stands at around 15 percent o f the total117 (Table 5.14). `I6Adopted inOctober 2001. I" Recent information provided by MLSSF shows that expenditure on ALMPs has increased to 20.8 percent in 2003, and expected to reach23.6 percent ofthe total in2004. 129 Box 5.1 The Labor Code A new Labor Code was approved inFebruary 2003, and came into force onMarch 1 ofthe same year. The Labor Code defines the legal framework for the labor market and regulates aspects relatedto collective and individual contracts, working and leisure time, remuneration, health and working conditions, training, social dialogue, and conflict resolution. Although the Labor Code has only recently beenenacted, it has already triggered numerous complaints from employers, both domestic and foreign. The Code is widely criticized for introducing significant rigidities inthe labor market, with adverse consequencesonjob creation and labor costs. Among the positive changes proposedby the Code, the elimination o f labor booklets should simplifyrecording procedures and limit the opportunities for abuse and corruption. The Code also stipulates that the payment o f the social assistance, unemployment and healthinsurance contributions is mandatory, and failure to pay on time constitutes a criminal offense, punishable by fine or even imprisonment. The effectiveness o fthese stipulations will, however, crucially hinge on their vigorous enforcement. The introduction o f part-time employment and temporary work, which boosted employment inseveral EUcountries (e.g. the UnitedKingdomand the Netherlands) is another positive aspect o fthe Code. At the same time, some existing stipulations inthe Labor Code require revisionifthey are to enhancethe needed labor market flexibility and the competitiveness o fthe Romanian economy. These include: a) The relaxation o frestrictions onthe use ofterm contracts. The current regulations o f fixed-term contracts are restrictive by international standards. They canbe m i t e only on an exceptional basis (Arts. 80 and 8l), with no possibility o frenewals (Art. 84 (1)). Admittedly, these stringent provisions are partly offset by the introduction of the institution o f a temporary work agency, but the Code does not allow for the existence o f fixed-term contracts, inthe standard sense. Inaddition, the probationperiod stipulatedby the Code is too short to be a substitute for fixed-tenn, as a means to assess one's aptitudes to occupy a position, especially a highly skilled one. b) The introduction ofemployer rights for labor retrenchment for economic reasons. The flexibility inthe area o f labor and labor relations includes a possibility to employ, adjust, distribute and dismiss work force based on the job requirements, to ensure the ability to adjust to the external market criteria, primarily the demands of customers and changes inthe economic environment, to maintain competitiveness by means o f cost control and to guarantee steady increasesinthe quality o fproducts and services. Economic reasons (e.g. increasing productivity and profitability) should be added to a list o f a valid reasons for dismissal. Currently, Art 65 (1) mentions only economic hardship, which is not sufficient. Firmsmay need to lay labor off even ifthey do not suffer from economic hardship but want to improve their market competitiveness. C) The elimination o fthe application ofbranchcontract agreements to all contracts inthe sector. Mandatory extension o f industry-wide collective labor agreements to non-participating employers should be discouraged (Art. 241 [2]). Or, there should at least be an "opt-out" option for employers, for whom complying with the industrylevel agreement is too costly. Evidence shows that these are usually small firms who are, as a rule, not well represented at the bargaining table. Anindustrylevelagreement negotiated by large employers, who can afford higher wages and benefits, may pushsmall employers out of business. Inaddition, critics notedseveral other issues ofimportancefor the labormarketwhich needfbrther clarificationand refinement, such as the provision for the establishment o f the wage guarantee funds, the elimination o f the use o f civil contracts by converting them into regular contracts, the adoption o f a 48 hour ceiling on overtime, and the requirement for employers to provide training for employees. 130 Table 5.14 Public Expenditure on Labor Market Programs in Selected Countries, 1991- 2002 (percent o f GDP) 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 CzechRep Tota1,ofwhich: 0.4 0.5 0.3 0.3 0.3 0.3 0.3 0.4 0.5 0.5 active(%) 43.6 62.2 53.3 48.4 51.9 46.2 34.4 36.1 36.7 43.1 Hungary Total, of which: 2.8 2.8 1.8 1.3 1.1 1.1 1.0 1.0 0.9 active ("3) 21.5 23.6 33.5 31.8 33.0 41.1 38.6 41.2 45.3 Poland Total, of which: 2.5 2.5 2.2 2.2 2.2 1.1 0.6 0.6 0.8 active ("3) 0.0 23.7 21.9 18.6 22.4 0.0 0.0 0.0 0.0 Romania Total, ofwhich: 0.9 0.7 1.2 1.4 1.5 1.0 0.7 0.7 active (YO) 15.7 11.7 2.3 2.5 2.7 2.5 12.9 14.8 Netherlands Tota1,ofwhich: 3.7 4.0 4.3 4.5 4.2 4.8 4.5 4.1 3.9 3.6 active(%) 34.9 36.7 35.0 32.5 32.2 29.0 32.7 38.5 41.6 43.1 Spain Tota1,ofwhich: 2.6 4.0 5.0 3.5 2.9 2.5 2.3 2.1 2.1 2.2 active(%) 27.4 18.0 10.9 14.9 18.7 18.0 21.2 28.0 33.3 37.7 Sweden Total, ofwhich: 4.0 5.6 5.7 5.5 6.8 4.1 3.9 3.5 2.7 active(%) 59.9 53.0 51.9 54.1 51.0 49.2 50.4 51.9 50.7 United Tota1,ofwhich: 1.91 2.18 2.14 1.91 1.69 1.43 1.15 0.96 0.92 Kingdom active(%) 28.8 26.6 26.2 27.7 26.6 28.7 33.0 34.4 39.1 Source: World Bank staff estimatesbasedon OECD andNIS. 5.65 Public works programs did not have a significant impact on employment and earnings, although they obviously reduced the duration o f unemployment. As expected, the measures had different effects on various sub-groups. For example, the public works program did have a positive impact on older participants, but not on other categories. 5.66 The evidence suggests that the Employment Agency should improve the targeting o f its programs, which i s defined too broadly, as well as revising some o f the instruments used. Linkingoutcomes with program expenditure should improve allocation efficiency and should be considered by the MLSSF as a useful policy choice. Defining clear performance indicators, measuring the extent to which policies achieve their desired objectives, and, in the longer term, introducing budget programming, should enhance the capacity o f the MLSSF to design and implement coherent labor market programs. F.CONCLUSIONSANDRECOMMENDATIONS 5.67 The process o f EU accession is becoming the main driver for the productivity-driven labor reallocation and labor market reforms in Romania. While in the short run the unfinished enterprise sector restructuring agenda constitutes the main determinant o f the labor market dynamics, in the medium term the speed and depth o f the integration with the EU will set the trends and the magnitude o f labor market adjustments. 5.68 The liberalization o f trade in industrial goods with the EU, starting from January 2002, and the implementation o f the core o f the EU legislation and regulations under the acquis communautaire are gradually opening the markets to competition and increasing the pressure on the Romanian enterprise sector to restructure internally and diversify its product range and quality. This trend is visible primarily inthe tradable sector and has already resulted inimportant productivity gains. Enterprises are responding to increasedcompetition with cost cuts, which are likely to entail further labor retrenchment. Inthe sectors less exposed to competition inthe past, and especially in the network industries such as the energy sector, efficiency-driven reforms, 131 through privatization, restructuring and liquidations, are likely to result in large job losses. Consequently, one should expect additional inflows into unemployment and a further increase in the unemployment rate. Determined by the skill mismatchbetween the labor supply and demand, this unemployment i s o f a structural, long-term nature. 5.69 Against this background o f significant economic restructuring, which, given the complexity o f the processes entailed, i s expected to continue well after accession, this research has identified several priorities for labor market reforms. These come under several headings, as listed below. 5.70 Revisit the legal and institutional framework for the labor markets and address the rigidities imposed by the new Labor Code. The new Labor Code includes aspects that threatens to reduce labor mobility. By discouraging labor mobility, the Code complicates further progress in structural reforms and introduces unnecessary rigidities in the labor market, which could affect the speed and the direction o f the reforms and could slow down the process o f EU convergence. Inparticular, the Code will create new impediments to enterprise restructuring and will thereby adversely affect the competitiveness o f the Romanian economy. The Code increases the risk that Romania will repeat the experience o f many EU countries that paid with high and persistent natural unemployment rates for their reluctance to increase labor market flexibility. 5.71 Substantial changes to the current draft are required if the Code is to support an efficient labor market which will eventually benefit all citizens. Particular attention should be paid to the most critical areas, which are the justification and procedural costs o f dismissals, the limited use of fixed-term contracts, working time' flexibility, and the role o f labor unions in corporate governance. Strengthening the institutional capacity o f the agencies involved in the design and implementation o f labor market policies should continue as the country prepares for substantial European structural funds inflows after accession. This is especially needed, given the increased delegation o fresponsibilities and resources from the central agencies to the local labor offices. 5.72 Encourage labor participation by bringing the working-age population back into the labor market. Identifying ways to increase employment rates, especially for women, constitutes a key condition for achieving sustainable economic growth and real convergence in incomes with the EU. Raising employment rates will help restore fiscal sustainability by broadening the taxation base and increasing the resources allocated for public programs, such as the PAYGpension system. Inthis context there i s a need for tailor-made programs focusing on increasing the employability o f the labor categories adversely affected by this transition. Among these categories one can single out discouraged workers, early retirees, unpaid family workers and the long-term unemployed. A combination is needed of targeted measures that will enhance the flexibility o f employment by encouraging temporary and part-time job creation, and the use of TWAs and active labor market policies focusing on specific disadvantagedlabor groups. 5.73 Introducing incentives for those who wish to continue to work beyond the retirement age would not only result in higher levels of output and higher tax revenues, but would also have a positive impact on poverty reduction, as these groups are vulnerable to or are already in poverty.'18 Relaxing the provisions o f the Labor Code to address this obstacle, principally by "* See the WorldBank, 2003, RomaniaPoverty Assessment. 132 easing hiring and dismissal conditions, is likely to result in important added-worker inflows which would boost participation. 5.74 Encourage employment mobility to support a more rapid reallocation of labor across sectors. Labor reallocation i s fundamental for a successful transition in Romania. Throughout the transition, employment trends have been determined by a combination o f conjectural and permanent factors, with the weight unbalanced towards the former. The sectoral employment structure differs significantly between Romania and the EU economies in spite o f important changes that have taken place inthe last several years. Employment in agriculture and inthe heavy and network industries remains unsustainably high.At the same time, employment inservices ishalfofthat inthe EU. 5.75 Some o f the current labor market trends are likely to be reversed through the liberalization o f the markets and the removal o f the existing constraints, such as the high transaction costs that restrict entry into some sectors and jobs. Employment reallocation is therefore a task that must fall upon the "external" labor market. It can be achieved only by the physical mobility o f workers who leave one enterprise, probably in a declining sector, and find work in another. Reallocation in the context o f Romania will have to be multidimensional, with workers changing not only their companies and sectors, but also their occupations and probably their job locations. It i s thus the dimension o f restructuring inwhich any deficiencies inthe labor market-such as impediments to labor mobility, poorly developed employment exchange facilities, excessive specific training on the previous job which is not transferable to alternative employment, or industryrents-are likely to have the most severe effects. 5.76 Active Labor Market Policies (ALMPs) can only be useful if these complement flexible labor market rules. ALMPs are expensive and their effects are not sustainable unless there are flexible market rules. Recent legislative changes, and primarilythe adoption in2002 o f the Law on Employment Promotion, have led to the establishment o f a framework for ALM programs. Public works schemes, employment andbusiness start-up subsidies, andjob brokerage have been introduced, and access for the unemployed has been widened to cover non-recipients of benefits. This has resulted ina larger participation o f the unemployed inthe programs, and has stimulated job creation. Their costs, benefits and sustainability, however, is not known. These need to be monitored closely and their net benefits assessedbefore expanding them further. To facilitate employment o f youth and low-skilled and displaced workers, as well as the long term unemployed, policies should continue to reduce the costs o f hiringthese categories. 5.77 Continue the reforms of the tax and benefits systems, in terms of policy and administration, to further reduce labor costs. Hightaxes represent important barriers to job creation, affect the competitiveness o f Romanian enterprises and push firms into the informal sector. The simplification o f the tax system and the recent adoption o f a new Fiscal Code should stimulate job creation and give a boost to the SMEs, with a positive impact on employment and productivity in this sector. The unification o f the collection and administration o f social assistance, health and unemployment contributions under a single agency coordinated by the Ministry o f Finance is expected to improve collection rates. However, the statutory rates o f labor taxation remain high, despite the recent reductions in social assistance contributions, and effective collection as a percentage o f GDP remains low. 133 5.78 Develop a wage setting system where wage increases correlate with productivity growth at the firm level, so that it can be supportive of employment-friendly economic and labor market reforms and able to guide skill formation, mobility and the reallocation of labor. Romania should adhere more closely to discipline in incomes policy and hence better control on wages inthe public sector is needed. Wages in the private sector should be set at the firm level in line with productivity. Benefits o f growth can not be more broadly shared unless wage income is set according to performance and productivity gains at the firm level. 5.79 To further lower labor costs, the government should contemplate more moderate increases in the minimumwage in the future, eventually by linking it to the expected inflation. Since many low-skilled and young persons are clustered around the minimumwage, they will be among the first to benefit from increased employment opportunities ifthe minimumwage stays at a moderate level relative to the average wage. 5.80 Accelerate the restructuring of agriculture and design programs to assist the redeployment of labor released from this sector. Agriculture necessitates special attention in Romania because o f its high contribution to employment. While its share in GDP is around 13- 15 percent, agriculture continues to be the largest employer in Romania, with more than 30 percent o f the labor force."' These numbers are unsustainable, as the experience o f the EU cohesion countries suggests, and we should expect a sharp decline in agricultural employment in the coming years. Encouraging alternative employment opportunities and supporting income- generating activities in the rural areas, away from agriculture, should reduce the restructuring costs for the sector and facilitate integrationwith the CAP. 5.81 The high average age o f the population working inagriculture and its reduced skill levels limitthe transferability o fworkers to alternativejobs. We canthus expect the productivity-driven reforms o f the agricultural sector and the adoption o f the CAP regulations to trigger a strong push effect, sending labor from agriculture out o f the labor force. Mitigating the anticipated adverse consequences o f this phenomenon would require paying attention to the social assistanceneeds of those with a highchance o f being displaced from agriculture. The success in reducing extreme poverty exhibited by the Minimum Income Guarantee program introduced in 2002, set a good example o f effective instruments o fthis type o f assistance. 5.82 Increase investment in human capital and address the skill mismatch gap. Investment in education inRomania, at 3.4 percent o f GDP, i s low by regional standards. As the country is undergoing significant economic restructuring, and as economic growth takes off, the demand for labor could become the reality check for the relevance o f education. Present unemployment features such as high and long-term unemployment among new graduates and low level educated youth indicate a mismatch between the skills that the education system provides and the labor market demand. Policies designed to assist recent graduates and inexperienced workers have contributed substantially to the improvement of the employment chances o f these categories. Notwithstanding these successes, more is needed to reduce labor mismatch. Educational institutions (VET, post-secondary and higher education) should conduct The 2002 census data indicate a recent decline inthe number of people working inagriculture. 134 tracer studies to identify both match and mismatch situations and reorient the profiles, study streams and curricula accordingly. 5.83 The education system i s undergoing a comprehensive reform that has already produced significant changes, especially in compulsory education. Reform measures were piloted in the vocational education and training (VET) system, with support from the EU. These measures need to be scaled up to the level o f the entire system. Improvements are still needed, mostly with regard to the relevance o f secondary and tertiary education. Universities were also included in a modernization process, which entailed improvements in management and financing, accreditation, and curricula. These reforms were carried out more inan anticipatory manner than inresponse to economic andsocial pressures. 5.84 Insum, EUaccession is becoming the main driver of labor market reforms inRomania. Inthe mediumterm, labor market reforms are needed to accelerate and deepen integrating with the EUinorder to alter the trends andmagnitude o f labor market adjustment. The urgency of the above reforms is highlighted by the worrisome trends o f an expanding share o f the labor force in agriculture. Without enhancing labor market adjustment, Romania's economic potential can remain largely untapped. 6. FROMQUASI-FISCAL FINANCINGTO EFFICIENT FINANCIALINTERMEDIATION 6.1 Inadditionto labormarkets, financialsector developmentisthefactor marketthatplaysa critical role in economic restructuring. 120 A sound and efficient financial sector is essential if Romania's efforts to restructure its economy from low to higher productivity activities are to succeed.'21 Without a financial sector that efficiently mobilizes and directs finances toward the most productive investment opportunities, the ,accumulation o f savings i s limited and these resources are not channeled to the most productive investments.'22 The 2003 Joint Bank-Fund Financial Sector Assessment Program for Romania'23includes a comprehensive analysis of, and policy recommendations for, the challenges o f financial sector development. This study is based on this work and builds on it from the perspectives o f stability, growth, and integration with the EU. A. INTRODUCTION 6.2 Romania has made important progress in consolidating the financial sector since 1999, whenbank restructuringtook hold. The regulatory and supervisory environments have improved, which will provide the foundation for a stronger financial sector. However, the basic financial infrastructure needs to be further developed to ensure an efficient and secure financial market that will help stabilize the economy and will set it on a higher growth path by mobilizing and directing finance toward more productive economic Inview of these facts, this chapter provides a rationale for a public policy that is aimed at improving the fundamentals o f financial markets in Romania, presents a summary o f the structure and performance o f these markets, and suggests policy recommendations to achieve these objectives. B. THESTRUCTUREOFTHE FINANCIALSYSTEM 6.3 Romania's financial system has undergone a major transformation since the start o f the transition. Like to most o f its peers, Romania had to fundamentally reform the mechanism o f the financial sector, as the monobank structure needed to be replaced with a banking system, and equity and corporate debt markets had to be created from scratch. In the past, poor macroeconomic management and the failure to establish financial discipline inboth the financial and the corporate sectors made the development of a financial sector a difficult undertaking in Romania. The financial sector i s dominated by the bankingsector (Table 6.1). 120 This chapter was prepared by Stella Ilieva (ECSPE), Economist, World Bank. The chapter builds on the Joint Bank-Fund Financial Sector Assessment Program, World Bank and InternationalMonetary Fund, 2003. '" Financefor Growth: Policy Choicesin a Volatile World,World Bank. 122Financial infrastructure refers to legal and regulatory structures, supervisory rules and practices, accounting and auditing rules and practices, credit bureaus and registries, payments systems, and exchange systems. 123Joint Bank-Fund Financial Sector Assessment Program, 2003. 124Finance and the Sources of Growth, Beck, Levine, and Loayza, 1999. 136 6.4 Bank restructuring got off to a comparatively late start inRomania and still remains high on the agenda, whereas it has been largely completed in other CEECs. As shown in Table 6.2, the state has largely withdrawn from the banking sector in most of the CEECs, with its asset shares at the single digit levels. State ownership inRomania, however, continues to be high-at around 40 percent.'25 End-2002, state ownership of the banking sector was close to the median share of its peers seven years ago. Foreign ownership of banks in Romania i s also substantially lower, compared to more than 80 percent inthe other CEECs. Table 6.1 Structure of the Financial System, 2002 As of end-2002 Assets Number of Inpercent Institutions I n ROL Billions of GDP Commercial banks 39 469,712 31.1 Credit cooperatives 1network 2,687 0.18 Credit unions 3,895 5,079 0.34 Financial investment companies (SIFs) 5 21,969 1.45 Financialinvestment service companies 77 723 0.05 Investment funds 26 1,288 0.09 Of which: 23 999 0.07 Open-ended investment funds 3 289 0.02 Venture capital funds Insurance companies li 49 22,841 1.51 l i The pensionscheme authoritiesand Fund estimatesis a state-organized "pay as you go" system and there are no private pensionfunds. Source: NBR. Table 6.2 Asset Share of State-Owned Banks inCEECs, 1996-2002 1996 1997 1998 1999 2000 2001 2002 Lithuania 54.0 48.8 44.4 41.9 38.9 12.2 0.0 Estonia 6.6 0.0 7.8 7.9 0.0 0.0 0.0 Slovak Rep. 54.2 48.7 50.0 50.7 49.1 4.9 2.9 Latvia 6.9 6.8 8.5 2.6 2.9 3.2 4.0 Czech Rep. 16.6 17.5 18.6 23.1 28.2 3.8 4.6 Hungary 15.3 3.5 9.8 7.8 7.7 9.1 10.8 Bulgaria 82.2 66.0 56.4 50.5 19.8 19.9 14.1 Poland 69.8 51.6 48.0 24.9 23.9 24.4 26.6 Romania 75.9 73.0 71.0 46.8 46.1 40.4 40.4 Slovenia 40.7 40.1 41.3 42.2 42.5 48.9 48.6 Median fexc. Romania) 40.7 40.I 41.3 24.9 23.9 9.I 4.6 Source: EBRD Transition Report, 2003, and NBR for Romanian data. 6.5 Indeed, Romania made important progress in changing the ownership structure of its banking system in 1999 and 2001. State ownership in banks declined from over 70 percent o f banking system assets in the mid-1990s to 47 percent in 1999. Bank restructuring became 12'InMay2004, the govemment privatized 33 percentofthe shares ofBCR. 137 inevitable after the crisis in 1998-99 which was triggered by the dire condition o f two large state- owned banks-Bancorex and Banca Agricola. Several years o f directed lending to nonviable SOEs and to agriculture, and massive mismanagement o f funds, led to the accumulation o f large amounts o fbad loans inbothbanks. In1999, Bancorex (accounting for one-fourth o f the banking system's assets) was liquidated after several failed attempts at its restructuring, including through recapitalization.126 Its bad assets were transferred to the Asset Recovery Agency (AVAB), and the rest were transferred to Banca Comerciala Romana (BCR). In contrast, Banca Agricola was recapitalized, its bad assets were removed, and it was prepared for privatization. The same year saw the privatization o f two state-owned banks (the Romanian Bank for Development and Banc Post), which were sold to foreign investors, and in2001 BancaAgricola was finally privatized. 6.6 Presently the state continues to be one o f the major players inthe banking sector. Out o f the 39 banks operating in Romania in 2002, three are state ~wned-BCR'~~,the Savings Bank (CEC), and the Export-Import Bank o f Romania (EXIMBANK). They account for nearly 40 percent o f bank assets and deposits, and 32 percent o f loans (Table 6.3). BCR is by far the most important bank in the market, accounting for about 31 percent o f the banking system's assets, while the Savings Bank enjoys a full state guarantee on deposits. Domestic banks also include a small number o f privately owned banks, which constitute a very small share o f the market (4 percent of assets and 3 percent o f deposits). The remaining 32 banks account for 56 percent o f the system's assets. These banks include subsidiaries o f large multi-national banks, subsidiaries o f other smaller foreign banks, or branches o f foreign banks. All these banks hold 57 percent o f the deposits and extend 64 percent o fthe loans inthe country. 6.7 While the sale o f 33 percent o f BCR inM a y 2004 i s a step fonvard, the continuing strong presence o f state-owned banks places Romania at a disadvantage compared to other EU accession countries, where the benefits o f private sector ownership are already visible. These benefits include deeper financial intermediation, diversification and innovation o f the financial services provided, and increased competition, which leads to better and less costly services for supporting a stronger corporate sector. 6.8 Non-banking financial institutions are underdeveloped and consist mainly o f insurance companies and financial investment companies (SIFs). Capital market capitalization accounted for 11percent at end-2002.'2* Annex 5 provides a brief summary o f the non-banking financial institutional sector inRomania extracted from the Joint Bank-Fund Financial Sector Assessment Program o f 2003. 126 Estimated at an equivalent of US$600million ingovernment bonds (2 percent of GDP). 12' InM a y 2004, the govemment privatized 33 percent of the shares of BCR. 128 The insurance sector is very small, despite the large number of insurance companies. Its assets are estimated at only 1.5 percent o f GDP. There are also five financial investment companies of a similar size. Credit unions and credit cooperatives together hold assets of about US$230 million. 138 Table 6.3 Composition of the Romanian Banking: Sector, 1996-2003 1996 1997 1998 1999 2000 2001 2002 H1-2003 Number of commercial banks 40 43 45 41 41 41 39 38 Romanian incorporatedbanks 31 33 36 34 33 33 31 30 State-owned 7 7 7 4 4 3 3 3 Private 24 26 29 30 29 30 28 27 ofwhich, jointventure with foreign 10 13 16 19 21 24 24 23 Branches.of foreign banks 9 1 0 9 7 8 8 8 8 Share of total banking sector assets (YO) All commercial banks 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Romanian incorporated banks 96.2 92.7 94.3 92.9 92.2 92.1 92.6 92.7 State-owned 75.9 73.0 71.0 46.8 46.1 41.8 40.4 40.4 Private 20.3 19.7 23.3 46.1 46.1 50.3 52.2 52.3 of which, joint venture with foreign 7.5 10.5 14.3 40.4 43.0 47.3 49.0 48.6 Branches bfforeign banks 3.8 7.3 5.7 7.1 7.8 7.9 7.4 7.3 Share of total banking sector loans YO All commercial banks 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Romanian incorporated banks 97.7 93.7 92.8 90.0 88.5 88.9 89.5 90.3 State-owned 79.9 73.0 72.1 38.8 30.4 31.8 31.4 31.7 Private 17.8 20.7 20.7 51.2 58.1 57.1 58.1 58.6 of which, joint venture with foreign 5.9 8.7 11.6 45.1 54.7 53.5 54.5 54.5 Branches bfforeign banks 2.3 6.3 7.2 10.0 11.5 11.1 10.5 9.7 Source:National Bank of Romania. c.THEROLEOF THE BANKINGSECTOR 6.9 The level and quality o f financial intermediation in Romania has improved substantially over the last four years. At present the banks are liquid, well capitalized, and relatively profitable, and the systemic risk appears low. However, compared to other CEECs, financial intermediation and monetization in Romania remain shallow, reflecting a difficult macroeconomic environment, delayed structural reforms, and an uncertain and insecure legal framework (Figures 6.1 and 6.2). The slow pace o f the restructuring o f the corporate and financial sectors inRomania has encouraged the persistence o f non-payment inthe economy, and this has significantly constrained the deepening o f the financial sector in the country. Thus, the banking sector inRomania has not been provided with the incentives to effectively mobilize the resources to finance investment and growth. 6.10 Monetization as measured interms o f the ratio o f broad money to GDP has been slowly recovering to pre-crisis levels and accounted for about one-fourth of GDP in 2002, which is two to three times lower than in the Czech Republic, Hungary and Poland. High levels o f inflation and negative real interest rates, coupled with a low level o f confidence inthe bankingsystem and a dynamic informal sector, have discouraged savings in Romania. Nevertheless, with the improvement o f the macroeconomic environment, deposits have been gradually increasing since 2000. 139 Figure 6.1 Broad Money to GDP, 2002 Figure 6.2 Private Sector Credit, 2002 ercent) (percent) Fig.1: Broad Money, 2002 (percent of G D P ) Fig.2 Private S e c t o r c r e d i t , 2002 (percent of GDP) Czech Rep 72 4 Croatia Croatia wary Slovak Rep Latvia Hungary Estonia Estonia 3lovak Republic Bulgaria CzechRepublic Poland Bulgaria Latvia Poland Lithuania Lthuania Romania Romania 0 20 40 60 80 Source: EBRD TransitionReport 2003 0 10 20 30 40 50 Source EBRD TransitionReport.2003 6.11 Domestic credit has been low inRomania compared to othc-r countries inthe region, and this was the case even before the bankingcrisis that Romania experienced in 1999. The distorted macroeconomic environment, particularly the high interest rates and volatile exchange rates, presented to obstacles and perverse incentives for the activities o f the commercial banks and also shortened the horizon o f credits. In addition, bank access to long-term funding was limited, and thus banks were not able to provide medium term and long-term credit to finance long-term investments. The prevalence o f state ownership in the banking system led to weak intemal governance and poor lending decisions. This caused a dramatic deterioration in the banks portfolios, which was further exacerbated by the weak legal and institutional capacity for debt collection and contract enforcement. Sound banking under these conditions meant mainly investing inT-bills: the rates o f return on this safe asset were higher than on riskier altematives. 6.12 Apart from the legacy o f the banking crisis o f 1999, the low intermediation and investment levels may be explained by: (i) the liquidity preferences o f banks; (ii) high cost the of bank lending; (iii) the lack o f growth opportunities; and, (iv) the structural deficiencies o f the credit market. 6.13 Banks prefer to keep high liquid positions rather than involving themselves in risky lending, as most of the companies are incurring losses and also maintain high stocks o f arrears and lack growth opportunities. Moreover, the interest rates offered by the NBR on non-reserve deposits have often been a better alternative than loans. Compared to other countries in the region, inter-bank activity in Romania is low. The limited inter-bank market in Romania contributes to banks holding additional liquid assets in order to self-insure against a higher than usual demand for liquidity. 6.14 Similarly, the combined effects o f non-payment and accumulation o f arrears by enterprises and high cost o f capital have depressed the demand for bank credit. Non-payment is common not only among companies that are facing soft budget constraints, but also among 140 healthy enterprises that chose to delay tax payments because they are encouraged by a long history o f tax arrears and forgiveness and by the comparatively low penalties for outstanding tax debts. 6.15 As shown inTable 6.4, bank credit represents only 8 percent o f the total financing o f the corporate sector, and firms are predominantly financed by equity'29 and arrears to the state.'30 The latter constitute more than on-third o f the liabilities, with tax arrears accounting for about half o f the total. The amount o f overdue bank debt is small relative to total debt, and both the percentage o f firms with overdue credit and the share o f overdue credit in total debt have fallen since 1999. There is also evidence that private domestic and foreign-owned companies tend to rely more on bank debt as opposed to state-owned companies, which have higher tax and social security arrears. Trade credits provide for about 16 percent o f firms' financing, with more than one-third o fthe trade credit being overdue. The bondmarket is virtually non-existent. Table 6.4 Sources of Funds for Romanian Firms in 2001 (Percent o f total financing) stocks 1999 2000 2001 Number of firms 1,155 932 726 Internal Financing 7.4 10.0 12.8 Retained earnings 6.8 8.9 11.9 Reserves 0.6 1.1 0.9 ExternalFinancing 58.8 51.8 52.5 Share capital 35.6 29.2 28.3 Bonds 0 0 0 Bank debt 10.1 7.6 8.3 Trade credit 13.1 15.0 15.9 Arrears 33.6 38.2 34.6 Taxes 19.2 22.6 15.2 Social security 6 5.7 7.2 Sources: Ministryof Finance and staff estimates, Financial Sector Assessment Program, 2003. 6.16 Banks are reluctant to lend to enterprises, either because they are perceived to be especially risky or because they are unprofitable and lack growth opportunities. Only a few enterprises are profitable, and are sufficiently liquid to commit an acceptable amount o f their own equity to a new investment project. Moreover, only a few enterprises are able to establish their creditworthiness, given the quality o f their financial statements and the lack o f audited reports. Despite the fact that since 1999 the corporate sector has shown improvements in profitability, about one-third o f the firms are still unprofitable and do not generate the cash flow to service their debt. Large state-owned manufacturing firms with high tax and inter-enterprise arrears have less than halfo f the bank credit that is given to private firms with hightax arrears. 6.17 Bank lendingdecisions are influencedto a large extent bythe reliability o fthe borrowers' financial statements. Moreover, banks find it difficult to monitor managers and to exert corporate control after the financing has been made. Poor financial reporting, accounting and auditing lZ9Whichincludes mainly share capital and revaluations as, equity issues are limited. I 3 OThe firms inTable 6.4 are BSE and RASDAQtraded tiinns with more than 50 employees. 141 norms and practices prevent banks from effectively allocating resources to productive investments and thus influencing capital accumulation and long-term growth. The recent introduction of international accounting standards (IAS) is important for improving the quality of firms' financial statements and building confidence among market players. Their full implementation i s central to strengtheningRomania's basic financial infrastructure. The adoption of the inflation accounting principles (IAS-29) i s especially important because most Romanian companies have not revalued their assets since 1994 despite the high inflation during the period. As aresult, most ofthe current balancesheets underestimatethe value of assets. 6.18 Another important factor in the development o f efficient financial intermediation i s the availability and transparency of information about borrowers. A reliable and widely accessible credit registry would not only allow banks and other intermediaries to better assess potential borrowers, but it would also contribute to greater competition betweenlenders, and, last but not least, might have a disciplining effect on borrowers.13' Currently, the Credit Information Bureau at the NBR includes only those financial institutions regulated by the NBR and does not take account of small commercial or most consumer loans. It is important to make the credit history of entrepreneurs, whichi s generally usedto grant loans to SMEs, readily accessible to creditors. 6.19 Romanian banks have limited operational capacity for providing credit to profitable entrepreneurs, new businesses, and SMEs. There is a need to improve credit skills, financial statements, and develop credit history information inthe credit bureau. The technologies used by the banks need to be modernized. Improvements are being carried out mainly by foreign banks. 6.20 Credit growth, especially credit to the Figure6.3 DomesticCredit to Non- non-government sector, has accelerated with Government 1996 -2003 (percent of GDP) improved macroeconomic performance (Figure 6.3). Domestic credit grew from 14 30 percent of GDP in2000 to close to 15 percent 25 in 2003. As shown in Figure 6.3, credit to 20 households, mainly in the form of consumer credit, was the fastest growing component of 15 credit to the non-government sector. It more 10 than doubled in 2000-02 and in 2003 was 5 close to 4 percent of GDP, or one-fourth o f credit to the non-government sector. Credit to 0 private enterprises experienced a growth o f close to 2.5 percent of GDP between 2000 and2003. Claims on the SOE's have steadied Source: NBR at around 1.5 percent of GDP since 2000. However, these developments are taking place in an environment in which the fundamentals for lending are o f still developing, thus making reforms to strengthenbasic financial infrastructure of critical importance. Beck, T, 2000, Impedimentsto theDevelopmentandEflciency of FinancialIntermediation in Brazil. 142 6.21 While the expansion o f lending over the last couple o f years is not uncommon among most o f the CEECs, and is even stronger in these CEECs than in Romania.132credit growth in Romania has been associated with a lower degree o f restructuring o f the financial sector.'33 poorer financial discipline and a weaker corporate sector. Thus, while the fact that banks are providing more o f their core services to the private sector i s a welcome development, there are some macro and financial stability implications that have to be considered. 6.22 From a macroeconomic point o f view, the expansion o f consumer credit has already negatively affected the external current account deficit, which widened to a projected 4.8 percent o f GDP in 2003. Moreover, as most o f the domestic banks have limited access to international funding, the demand for credit byhouseholds and firms has to be offset by a comparable increase inthe savings rate, yet the dynamics of household deposits inRomania is far below the growth rate o f credits. Another significant risk could materialize inthe case o f lei devaluation on the part o fun-hedged borrowers, as they may have difficulties inservicing their debt. 6.23 Inview ofthe recent developments infinancial intermediation, banksupervisionneeds to be significantly strengthened and underpinned by an effective regulatory framework to limit the incentives for undue risk-taking by banks. This entails ensuring timely and accurate information from both the financial sector and the corporate sector, and improving contract enforcement. The importance o f a good regulatory and supervisory infrastructure for lending is evident especially with a dynamic bank lending environment anda comparatively weak corporate sector. D.THESOUNDNESSOFTHEBANKING SECTOR 6.24 The dominance of the state in the banking system, arid the lack of effective supervision and regulation, led to weak internal governance and poor lending decisions prior to 1999, which caused a dramatic deterioration in banks' portfolios. Non-performing loans soared to more than 70 percent o f gross loans. Sound banking under these conditions meant mainly investing in T- bills as the rates o freturn on this safe asset were higher thanriskier alternatives. 6.25 With start o f the restructuring o f state-owned banks and the introduction of stricter rules and regulations for bank supervision, the soundness o f the banking sector in Romania has improved significantly. The closure o f Bancorex in 1999 and the transfer o f the bad assets inthe system to AVAE3'34improved the overall health o f the banking system (Table 6.5). Since then, banks appear to be well capitalized, liquid, profitable and they have limited market and credit exposure. Capital adequacy ratios improved from 18 percent in 1999 to 23 percent inJune, 2003. Non-performing loans declined from 53 percent in 1999 to close to 2 percent in 2002 and are now on a par with those reported by other CEECs. Return on assets has more than doubled over the sameperiod, while liquidityratios havebeenmuchhigher thanrequired. 13*For more details, see Early Birds, Late Risers and Sleeping Beauties: Bank Credit Growth to the Private Sector in Central and Eastern Europe and the Balkans, IMFWorking Paper, Carlo Cottarelli, Giovanni Dell'Ariccia, and Ivana Vladkova-Hollar, November 2003. 133The EBRDIndex o f Bank Reform for Romania is the lowest among the CEECs. 134 Close to US$2.3 billion in bad assets from Bancorex and Banca Agricola were transferred at that time to the Asset Recovery Agency. 143 6.26 However, with the recent expansion o f credit a slight worsening o f prudential indicators has taken place, which raises concerns about the sustainability o f rapid credit growth in a system that is still dominated by large state-owned banks that lacks comprehensive credit information andreliable financial statements, andthat follows inefficient debt enforcement practices. 6.27 The capital adequacy ratios show that banks in Romania are well capitalized: reported capital adequacy ratios have been consistently higher than the required 12 percent,'35 although they have been declining lately. The challenge is now to implement the new accounting rules by hlly implementing U S , particularly IAS-29. Similarly, it is central to implement the requirements, introduced in 2004, to report on a consolidated basis, as a number o f banks have subsidiaries outside the banking sector. Table 6.5 Financial Soundness Indicators, 1998-2003 1998 1999 2000 2001 2002 2003' Capital Adequacy Ratio' 10.25 17.90 23.79 28.80 25.04 22.82 Tier 1I RiskWeighted Assets2 na 15.82 18.90 26.21 22.93 21.15 NPLsl Gross Loans' 71.7 52.6 5.2 3.3 2.3 9.1 ReturnonAverage Assets 0.06 1.47 1.49 3.10 2.64 2.43 InterestMargidGross Income na na -0.21 5.17 7.77 9.09 Liquid AssetslTotal Assets' na na na 77.10 78.59 70.88 Liquidity Ratio (actualhequiredliquidity) 1.30 1.37 3.48 LoansI Deposits 56.93 43.92 43.91 44.69 48.03 58.22 Notes:1. Dataas ofJune 2003, with the exception ofNPLsiGrossLoans(as ofNovember2003), andliquid assetsitotalassets (as ofMay 2003). 2. Foreignbanksbranches are not included. Source: NRR. 6.28 Another factor behind the current liquid position o f the industry relates to the monetary policy o f the NBR, which is aimed at absorbing liquidity associated with capital inflows by issuing deposits. Government securities and NBR deposits have been attractive investments for banks since the beginning o f transition. Banks also maintain high liquidity because o f the underdeveloped inter-bank market as a result, to some extent, o f the perceived risks in the system. The planned implementation in 2004 o f the new electronic inter-bank Real Time Gross Settlement System (RTGS) would improve the infrastructure for an efficient inter-bank market andwould allow banksto better manage their liquidity. 6.29 Liquiditydiffers across the different bank groups. The subsidiaries o f large international banks have a large liquid position, which to some extent is a result o f recent changes in their lending strategy. They were actually the only banks that decreased their lending to the private sector in 2002. Small foreign banks are less liquid, while the liquidity o f state-owned banks, although comparatively high, has beendeclining over time. 6.30 The quality o f loan portfolios in Romania has improved substantially since 1999 and compares favorably with good CEEC performers (Table 6.6), although some worsening has been seen lately. Non-performing loans have declined from more than 50 percent before 1999 to about 2-3 percent o f the loan portfolio. However, as a result o f the rapid growth o f lending and the 135The minimumcapitaladequacy ratio was raisedfrom 8 to 12percent in 1999. 144 introduction o f stricter rules'36 for loan classification and provisioning, the share o f non- performing loans has quadrupled in a year's time and at end-November 2003 stood at 9 percent o f total loans. Table 6.6 Non-PerformingLoans, 1996-2002 (percent) 1996 1997 1998 1999 2000 2001 2002 Bulgaria 15.2 13 11.8 17.5 10.9 7.9 10.4 CzechRep. 21.8 19.9 20.3 21.5 19.3 13.7 9.4 Estonia 2 2.1 4 2.9 1.3 1.2 0.8 Hungary na 6.6 7.9 4.4 3.1 2.9 4.6 Latvia 20 10 6.8 6.8 5 3.1 2.1 Lithuania 32.2 28.3 12.5 11.9 10.8 7.4 5.8 Poland 14.7 11.5 11.8 14.5 16.8 20.1 24.6 Romania 48 56.5 58.5 35.4 3.8 3.4 2.3 Slovak Rep. 31.8 33.4 44.3 32.9 26.2 24.3 11.2 Slovenia 10.1 10 9.5 9.3 9.3 10 na Source: EBRD TransitionReport, 2003. 6.3 1 Despite a good start, the Credit Information Bureau at the NBR could still be improved in terms o f the coverage o f institutions and loans. Currently, it includes only financial institutions regulated by the NBR and does not take account o f small commercial loans or most consumer loans. The credit history o f entrepreneurs, which i s generally used for the granting o f loans to SMEs, is not accessible. 6.32 The regulatory framework guiding the activities o f NBFIs is still incomplete, and the supervision o f NBFI activities is weaker. As a consequence, the non-bank financial market is small and underdeveloped. Current regulations do not define the funding options for non-bank lenders or the rights o f these institutions in case of debtor's default. Their tax regime is less favorable than that or the banks: borrowers are not eligible to deduct interest payments made to NBFIs andamortization o fthe VAT over the lifetime o fthe lease is not permitted. 6.33 Stress tests performed for the Joint World Bank-IMF Financial Sector Assessment Program suggest that currently banks in Romania do not face significant market and credit risk shocks, partly owing to the low level o f intermediation. Systemic risk appears to be low as well, which, however, reflects the low level o f the inter-bank market. Nevertheless, the credit risk arisingfrom loans to the corporate sector is important, as the corporate sector is vulnerable to an exchange rate depreciation due to its large share o f borrowing in foreign currency. A hypothetical sharp outflow o f capital is estimated to result in a 20 percentage point depreciation o f the lei and a 10 percentage point rise inlei interest rates. Another source o f risk to the banking sector that could be catalyzed when competition among banks grows is the impact on retum on equity (ROE) o f a sharp narrowing o f lending margins. The enhanced supervisory and regulatory rules, especially in terms o f interest rate risk management, stress testing, accounting for country risks, and transfer and operational risks, would provide incentives for banks to remain sound despite marketvulnerabilities. `36 The old loan classificationrules disregarded the borrower's financial condition and thus led to an underestimationof the non- performingloans inthe banks' loan portfolio. 145 E.REGULATORY AND SUPERVISORYISSUES 6.34 Important progress has been made in improving the regulatory and supervisory environment inorder to quickly adapt it to the demands o f Romania's economic restructuring. A large number o f new laws and substantial amendments to the existing financial legislation have been adopted' since 1998 in an effort to harmonize the legislation with international and EU requirements. A broad range o f measures has targeted all stages of the prudential supervision o f banks, from licensing to exit proceedings. Appropriate licensing and sanctioning mechanisms were set in place, a modem early warning bank-rating system is being implemented, and the frequency and coverage o f on- and off-site supervision has increased. At the same time, some o f the instruments necessary to support supervision have been created or streamlined, namely, the Credit Risk Bureau andthe Payment IncidenceBureau. 6.35 Nevertheless, important challenges remain for the development o f a strong financial sector infrastructure that ensures that the right incentives for financial market participants are in place. These challenges include improvements in the following areas: (i) assessment o f bank capital; (ii) access and coverage o f credit registry; iv) capacity to assess and manage risk. 6.36 Assessment of bank capital. While the capital adequacy regulation is largely compliant with the Base1Core Principle, legislation inthe area o f supervision on a consolidated basis has needs to be hlly implemented. Similarly, full implementation o f I A S will improve valuation o f capital, and national accounting standards and practices. Without supervision on a consolidated basis, capital figures, compliance monitoring with prudential standards, and asset quality assessment can become meaningless inlight o f the ease with which weak assets can be hidden in subsidiaries, and double gearing can take place. To address this concern, the government adopted the Law 485/2003 which introduces principles o f consolidated supervision in 2004. In addition the government is developing regulations and norms to further strengthen the legal framework for supervision on a consolidated basis. In addition, the supervision capacity is being strengthened to exercise supervision on a consolidated basis, includingthrough a mechanism for consultations with non-bank financial supervisors or supervisory bodies in other countries.137 In 2002, the NBR signed a Memorandum o f Understanding with the National Securities Commission and the Insurance Supervision with a view to improving the supervision o f the financial sector as a whole. 6.37 In spite of a well-functioning collateral registry, the ability o f banks to mitigate risks through collateral remains limited. Ordinance 61, introduce in 2002 and effective in 2003, permitted the authorities to seize any collateral pledged to a bank in order to collect tax arrears, which presented a significant impediment to secured lending by banks, in part owing to inefficiencies inthe courts and difficulties in selling reclaimed assets on secondary markets. The government however has now repealed Ordinance 61 and this is an important set forward. Further efforts, however, are needed to strengthen the regulatory and institutional framework, as discussed inChapter 2, the bankruptcy framework andprocesses need to improve. 13'The government has already signed agreements with the authorities o f Germany, Italy, Greece, Turkey, Cyprus, and Moldova. 146 Table6.7 CoverageofPublicandPrivateCredit RegistriesinEurope (number of borrowerswith creditrecordsper 1,000 capita) PublicCredit Private Credit Public Credit Private Credit Country Registry Bureau Countrv Reaistrv Bureau Austria 9 308 -Bulgaria 5 Belgium 68 42 Croatia Denmark 58 Czech Republic 10 136 Finland 96 Hungary 15 France 12 Latvia Germany 5 693 Lithuania 7 Greece 86 Poland 543 Ireland 730 Romania 1 Italy 55 416 RussianFederation Netherlands 530 Slovak Republic 2 Portugal 496 24 Slovenia 14 Spain 305 48 Turkey 7 204 Sweden 489 United Kingdom 652 Source: Doing Business in 2004. 6.38 Access and coverage of credit registry. To increase the transparency o f the financial market and reduce financial risk, the credit register should be expanded and made easily available. Currently, the credit history o f loan applicants is neither complete nor widely accessible. The public Credit Information Bureau includes only financial institutions regulated by the NBR and does not take account o f small commercial loans or consumer loans. Inaddition, there is no available credit history o f individuals, which i s generally required for lending to new and small firms. To accelerate the expansion o f the credit bureau, NBR may encourage a private bureau to operate and distribute credit information to banks. Such private credit bureaus exist in manyEUcountries andhave also beenestablished insome EUacceding countries (Table 6.8.) A major obstacle to the establishment o f a private bureau inRomania is Banking Law 58/1998 and law no. 677/2001 on protecting personal data, which restrict the sharing o f credit information among banks. The National Bank o f Romania has encouraged the large commercial banks to establish a bureau to share positive information on the creditworthiness o f borrowers and to be open to non-bank financial institutions as well. An outstanding issue is the sharing o f information on existing loans. Under the current law on personal data, such information cannot be shared without the prior consent o f individuals. In practice, this means that historical and outstanding loans may not be included in the credit bureau unless a dedicated law i s passed or the current personaldata protection law i s amended. 6.39 Capacity to assess and manage risk. To safeguard the health o f the banking sector, the bank supervision needs to develop a methodology to stress test banks' portfolios for various risks--credit, foreign exchange, interest rate, and systemic inter-bank risks, and the combination of shocks in these areas. Banks need a comprehensive risk management process that will identify, measure, monitor, and control all material risks and hold capital against these risks. It is especially important that banksmanage interest rate risk so as to mitigate risks arising from loans extended in foreign currency to non-foreign currency eamers. With the adoption o f the Law 485/2003, in 2003, more power is given to off-site supervision to receive information on risk management. This is a very important step forward considering that until 2003, some o f the information was provided bybanks only duringon-site examination, while there were limitations to acquiring this information for off-site surveillance purposes. 147 G.CONCLUSIONS 6.40 The Joint Bank-Fund Financial Sector Assessment Program for Romania includes a comprehensive analysis and policy recommendations to address the challenges o f financial sector development. This study is based on this work and builds on it from the perspective o f stability, growth and integration with the EU. Inthis context, this study highlightsthe following recommendations to support the development o f a sound banking sector. 0 Privatizing state-owned banks. The continuing strong presence o f state-owned banks, representing about 40 percent o f the bank assets and deposits and 32 percent o f loans, places Romania at a disadvantage compared to other EU accession countries, where the benefits o f private sector ownership are already visible: deeper financial intermediation, diversification and innovation o f the financial services provided, and increased competition leading to better and less costly services to support a stronger corporate sector. 0 Improving banking regulations and supervision. In 2003 and first-half o f 2004, important progress has been made in improving the regulatory and supervisxy environment that are necessary to quickly adapt to the demands o f Romania's , economic restructuring. Key challenges going forward include to fully implement the recent regulatory changes, further strengthen supervision capacity on a consolidated basis, to focus on adopting and enforcing risk assessment and management rules, and to improve the valuation of capital adequacy. Progress in these areas is a cornerstone o f deepening financial intermediation. 0 Eliminating the practice of nonpayment, arrears and barter settlement. Reliance of firms on nonpayment as an important means o f financing is crowding out regular banking credit and contributes to the country's low level o f financial intermediation. In this context, the government is commended for having abolished Emergency Ordinance 61/2002, which permitted the authorities to seize any collateral pledged to a bank in order to collect tax arrears. There is a need, however, to abolish the legal framework that allows compensation without cash o f the bilateral obligations stipulated in the Government Ordinance No. 77/1999. Romania should rely on revenue collection and administration reform, particularly prompt cash payments o f tax obligations, as an appropriate and effective approach to eliminating the practice o f nonpayment. 0 Introducing and enforcing international accounting and auditing standards and practices. The government plans full implementation o f international accounting standards as a key step toward improving both the level and the quality o f financial intermediation. These efforts will improve financial reporting, accounting and auditing norms and practices enabling banks to move to a more effective allocating resources to productive investments and thereby improve capital accumulation and long-term growth. The adoption o f the inflation accounting principles (IAS-29) is especially important because most Romanian 148 companies have not revaluedtheir assets since 1994, despite highinflation during the period. As a result, most o f the current balance sheets may underestimate the value o f assets. e Improving credit risk information by expanding the coverage of and access to the Credit Information Bureau. Despite a good start, the Credit Information Bureau at the NBR could still be improved in terms o f the coverage o f institutions and loans. Currently, it includes only financial institutions regulated by the NBR and does not take account o f small commercial or most consumer loans. The credit history o f entrepreneurs, which is generally used to grant loans to SMEs; is not accessible. As has been mentioned, an efficient and widely accessible credit registry, in addition to enabling banks and other intermediaries to better assess potential borrowers, would encourage higher competition between lenders, and, in particular, might have a disciplining effect on borrowers. 7. THE QUASI-FISCAL CHALLENGE: ENERGYSECTOR REFORM 7.1 As in many transition economies, the energy sector in Romania has played a central quasi-fiscal role.138Cheap energy has simultaneously been used to support loss-making public enterprises and to provide an implicit subsidy to consumers. Inefficiencies inthe sector are not only costly in terms o f direct resources used, but inefficiencies permeate the economy. Broad subsidization, particularly the practice o f nonpayment and arrears, i s related to costly quasi-fiscal financing that takes away limited public resources from priority public investment, such as education (Section 5) and undermines macroeconomic stability (Section 1). Not surprisingly, operating costs inthe sector remain high since the incentives to restructure have been lacking. In sum, reform that will develop an efficient energy sector in Romania is critical to Romania's stability, economic restructuring, competitiveness, and integration with EUand global markets. 7.2 Romania faces a triple inter-related challenge: (i)eliminating quasi-fiscal financing related to energy; (ii) aligning policy with EUdirectives; and (iii) integratingwith the South East Europe (SEE) market.13' Reforms and progress inany o f these three areas support progress inthe other two, given their complementarities. This chapter discusses key reforms ineach o f the these areas. A. THEENERGY SECTOR AND STATE SUPPORT 140 7.3 Energysubsidizationhasbeenprovided implicitlythrough tolerance o f low bill collection rates and low pricing o f electricity, gas, and district heating below their full cost-recovery tariffs.141 As early as mid-2002, natural gas prices were about one-third lower than those in OECD countries and one-half o f those in EU accession countries.142 For electricity prices, the disparity was about 50 percent in relation to the EU and 20 percent with respect to accession countries. Though close to the level o f several accession countries, district heating prices have been about half o f the average price in the EU. Under the current Stand-By Arrangement with the IMF, energy prices have risen since 2001, however reforms are yet to be completed. In electricity, there is wide variance in operating costs with some units operating with some efficiency, yet others are very inefficient. This in turn, poses important challenges to implementing full cost recovery reform. Owing to reforms, collection rates in the power sector 138This chapter was prepared by David Kennedy (ECSIE). Juan Carlos Ginarte prepared the section on the energy sector and state support. `39Annex 6 provides a briefdescription o f the Athens Memorandum signed inNovember 2002 by Albania, Bosnia and Herzegovina, Bulgaria, Croatia Former Yugoslav Republic o f Macedonia, Greece, Kosovo, Romania, Turkey, and Serbia and Montenegro. I4OThis section was preparedby JuanCarlos Ginarte, Economist, ECSPE, The World Bank. 14'The pricing structure o f the oil sector is considered undistorted. 14*A detailed discussion o f the pricing structure and financial performance o f the sector inrecent years is provided by the 2002 IMF Policy Discussion Paper "The Energy Sector Reform and Macroeconomic Adjustment in a Transition Economy: The Case o fRomania." 150 also have improved inrecent years, but monthly rates can vary significantly and on average need to rise substantially more.143 Furthermore, despite improvements in collection, enforcement of payment discipline on customers has been weaker than in the rest o f the industrial sector. For example, in 2002 the average payments period o f electricity companies was over five months, and less than two months for other firms.144 The district heating companies also confront significant collection problems, and they often fail to meet their obligations to gas distributors. As a result, the collection rates o f Distrigaz Sud and Distrigaz Nordhave at times been low and volatile. Table 7.2 ResourceTransfers To and From the Enerm Sector 2001 2002 2001 2002 0. Total Subsidization of energy (1+2+3) 28865 3 1267 2.5 2.1 1. Direct subsidiesto energy 4250 5166 0.4 0.3 Electricity 1052 1344 0.1 0.1 Mining 3198 3822 0.3 0.3 Gas 0 0 0.0 0.0 2. Other resource transfers 371 561 0.0 0.0 3. Hidden subsidies 24244 25540 2.1 1.7 Energy Fund(electr.) 3068 4175 0.3 0.3 Foreign loanguaranteespaid 7435 5050 0.6 0.4 Tax debt 5186 13856 0.4 0.9 Change inend o f year stock 1/ -3330 13550 -0.3 0.9 olw social insurance 2547 2484 0.2 0.2 Cancelled interest and penalties 8516 306 0.7 0.0 Bank payments overdue (domestic) 951 -187 0.1 0.0 Supplier arrears 7604 1647 0.7 0.1 4. Energy sector losses21 56165 37220 4.8 2.5 Low tariffs 45792 32527 3.9 2.2 Non-pap e n t 10373 4694 0.9 0.3 5. Net transfers to the energy sector (0 4)+ -10162 5714 -2.3 -0.4 Memnrandum: Rescheduledarrears 6259 4947 0.5 0.3 Note: 1/ Excludesaccumulatedpenalties. 2/ comprises implicit subsidiesto both households and firms. Excludesdistrict heatingcompanieswhich were not formerly units ofTermoelectrica. Source:Ministry of Industry andResources. 7.4 The resulting operating losses have been financed with arrears and various budget and off-budget funds (Table 7.1). These losses have resulted in under-investment in modem machinery and equipment and have reduced primary production o f oil, coal, and gas. The lack o f investment has left many facilities obsolete and highly inefficient, leading to increased import dependency.145 Because energy is a critical component o f production, energy subsidies have distorted relative prices and reduced efficiency throughout the economy. As prices are kept artificially low, there is over-consumption and waste o f scarce energy resources. Furthermore, 143 For example, in September 2002 the collection rate o f Termoelectrica's heating unit was 98.5 percent, which included bothpayments due that month and overdue bills. According to the Ministryo f Industryand Resources, this was followed by a steady drop to a 22 percent rate in December, then the rate increased to 52 percent in February 2003. The average for the year ended February was 76 percent. `44 See Fennema and Schaffer, 2003. 145 The inefficiency o f electricity productionis partly due to considerable excess capacity. 151 since energy company losses are often financed with tax arrears and unpaid loans, they have added to a rising stock o f public debt. 7.5 Recognizing the importance o f eliminating these losses for macroeconomic stability, the authorities have been adjusting the prices o f electricity, gas, and heating upward since 2001 and have redoubled their efforts to increase bill collection rates.'46 This has helped to reduce total losses from 4.8 percent o f GDP in2001 to 2.5 percent in2002 (Table 7.1).147This improvement reflects a reduction in losses due to low prices o f 1.8 percentage points o f GDP, and a 0.6 percentage point reduction due to improved collection. However, this excludes about 500 small district heating companies that were not previously a part o f Tennoelectrica. Arrears to these companies have been growing in the form o f unmet "subsidies" that the state and local governments are required to pay to close the gap between the administered prices they receive andtheir production costs. At the end o f 2002, these companies had received only 63 percent o f the amount due from both levels o f g~vernment.'~~ 7.6 Giventhe state o f financial indiscipline inthe enterprise sector, the increase inprices may have accounted for much o f the growth in tax arrears, and possibly also for the greater rate o f activation o f loan guarantees. In fact, the 2.1 percentage points o f GDP increase in tax arrears and loan payments in 2001-02, for the economy, mirrors the 2.3 percentage point decline in energy losses. The elimination o f losses in the energy sector, therefore, is by itself unlikely to improve the prospects for long-run macroeconomic stability without the enforcement o f tax payment di~cipline.'~'Smaller losses in the energy sector will give way to larger losses elsewhere and to increasingpublic indebtedness. 7.7 More recently, particularly in 2003, the Government o f Romania has acted to reduce energy sector subsidy. Specifically, power tariffs have been increased to cost recovery levels, andthere have also been significant increases ingas tariffs. Inaddition, payments discipline has improved inthe residential power sector. 7.8 Notwithstanding this progress, further power tariff increases will be required to finance investments going forward, and further gas price increases will be required to meet parity with border prices. District heat prices remain below cost recovery levels, and significant price increases are required in this sector. Regarding payments discipline, further improvement in power collections from state-owned enterprises (SOEs) is required, as are reductions in power '46 There is also an intention to accelerate privatization throughout the energy sector to raise investment and boost efficiency, but this plan is proceeding slowly. See the Ministry o f Industry and Resources draft "Road Map for Energy FieldinRomania" o f March, 2003. 14' Other than Termoelectrica's "extemalized units," district heating i s excluded. However, the losses from low pricing may be estimated by the arrears o f the state and local budgets to the heating companies to cover the difference between the national reference price and production costs. In 2002 these arrears totaled 0.2 percent o f GDP, up from 0.01percent o f GDP in2001. The district heating companies were due 7.3 billion lei in 2002. Of this amount, local governments were responsible for 55 percent required by law, and the state had the obligation to pay the remaining 45 ,percent. Inthe end, the state met over 90 percent o f its obligation, while municipalities paid less than 40 percent o f the amount due. This is down from 99 percent and 92 percent, respectively, in2001. 14'Disinflation mayput further financial stress on loss-makers, as the erosion o f the real value o f arrears will tend to slow. 152 sector commercial losses. Furthermore, improved collections inthe gas sector, and significantly improved collections indistrict heating, will be required for financial viability. B. POWERPRICINGAND PAYMENTSDISCIPLINE ROMANIA IN B.l. Pricing and PaymentsDiscipline 7.9 The methodology used in Romania to set power prices i s designed to allow power companies sufficient revenue to cover operating costs plus depreciation and - abstracting from payments discipline - should fulfill both economic andfinancial pricing criteria. 7.10 Inpractice, application of the methodology resulted in an average price o f 5.6 cents / kWh at the distribution company (end user) level; time series tariff data are presented inTable 7.2. This represents a 40 percent real increase in tariffs relative to 1999 (above 90 percent in nominal terms), when the average price was 4.0 cents / kWh. It is reasonably highrelative to other countries inSEE, where the average price i s 3.5 cents / kWh. 7.11 Interms o f economic and financial viability, current tariffs inRomania are sufficient on average to cover SRMC and to provide positive cash flow. More specifically: e Tariffs for thermal power generation provide for cost recovery, including cost o f debt, and also provide for the recovery o f "general" liabilities camed over from the past. 8 Tariffs for cogeneration (combined heat andpower) plants are cost reflective (Le., there are no cross subsidies betweenpower and heat). 8 Tariffs for hydro power are low compared to thermal power, but they do cover operating costs. They do not provide for rehabilitation, but do provide for recovery o f losses incurredduringthe drought o f 2003. 8 Tariffs for nuclear power cover operating costs plus provision for decommissioning. e Tariffs are sufficient to cover both transmission and distribution margins, though not investments inthese industry components. e Cross-subsidies within the power sector have been eroded, and tariffs now reflect costs by customer categories (e.g., residential / business / industry). 7.12 Going forward, and as investments are required, tariffs should be increased to cover the long run marginal cost (LRMC), in the range 7-9 cents / kWh in SEE; Romanian prices would then be comparable with those inthe accession countries, together with Westem Europe. This, o f course, presumes significant restructuring o f the energy companies with the result that their operating costs are as efficient as those in accession countries and Western Europe. Both tariffs andoperatingcosts need adjustment andrestructuring. 153 Table 7.3 RomanianDistributionCompanyTariffs andPayments,1999-2003 1999 2000 2001 2002 2003 Average revenue (cents / 4 3.9 3.5 4.1 5.3 kWh[excluding tax]) Collections to billings ("A) 89.6 94.8 94.1 93.2 98 Ratio o f cash to noncash 60.5 60.9 60.6 65.1 transaction ("A) Technical and commercial 12.1 12.8 12.8 12.8 losses (%) ~~Source: Ministryo f Economy and Commerce. 7.13 The fact that a major price correction may not be required in the short term does not mean, however, that the industry i s financially viable (Box 7.1). Indeed, though operating cash flow for the sector may be positive, receivables/payables are increasing, and ongoing state support i s required in the form o f transfers from the budget and deferred tax payments; the current situation i s not sustainable fiom a financial point o f view ifthe industryis to operate on a stand alone basis (Le., without state support) in the context o f a well-functioning economy where payments are made infull and on time. Box 7.1 Power Sector FinancialViability andAffordability: Basic Pricing From an economic perspective, and at a minimum, prices should cover Short Run Marginal Cost (SRMC), approximated by average operating cost. Over time, and as investments are undertaken, prices should increase to cover LongRunMarginal Cost (LRMC), that is, SRMC plus capital cost. From a general economic perspective, and as a practical consideration inthe context o f a liberalizingpower market, the pricing rules above should apply for each category o f consumers, so that there is no cross-subsidy, for example, between industrial and residential consumers. Given the cost structure o f the power industry, specifically the fact that the cost o f serving large customers is lower than the cost o f serving residential consumers, prices for the former group should be lower than those for the latter. In the context of regional power market development, prices in countries signatory to the Athens Memorandum should converge over time as the market is liberalized. Where there is regional excess capacity, the regional price should converge to SRMC. Where capacity constraints are binding, the regional price should move to LRMC. Regarding payments discipline, this should be hlly enforced fiom an economic point o f view in order that price signals are effective in supporting beneficial levels o f consumption; where consumers do not pay they will over- consume from an economic point o fview. Froma financial point o f view, to ensure short-term financial viability, power prices should be set so as to provide a positive operating cash flow, that is, to cover operating costs, debt service, and any costs related to new investment. For sustained financial viability, power payments discipline should be fully enforced. For companies with increasing receivables, though this may be bearable inthe short term, it will quickly result incash flowlfinancing problems. Economic and financial principles, while not synonymous, may be reconciled. Where SRMC pricing does not provide a positive cash flow, then the price may be marked up to cover the shortfall. In cases where payments discipline is a problem, prices should not be marked up; rather, and in order to avoid further distortion o f price 154 signals, cash shortfalls should be reduced through the undertaking o f measures to improve cash collections relative to billings, and to reduce non-billed consumption. 7.14 Financial problems for the sector arise from three main areas: e Poor payments discipline inthe power sector 0 Inefficiency inthe power sector 0 Tariffs andcollections inthe heat sector. 7.15 Focusing first on payments discipline, it i s useful to start by defining categories o f non- payment: (i) collections are defined as revenue collected over revenue billed and measure non- payment by billed customers; (ii) collections are cash collected over revenue billed and cash measure non-payment by billed customers together with barter transactions; and (iii) commercial losses relate to non-billed consumption (theft from the system). Payments discipline inRomania is now assessed from these three perspectives (Table 7.2). 7.16 Regarding collections, there has been substantial improvement in recent years, with current average distribution collections at 98.1 percent in 2002 (Table 7.2). Though this level is comparable with other SEE countries, further progress is required in order to meet standards in Western Europe. 7.17 At a minimum, gains madethrough 2003 should be sustained going forward to secure the ongoing financial viability o f distribution companies, particularly since the tariff methodology for the sector does not allow for failure to fully collect. B.2. Rules for allocatingsector revenues between generation, transmission, and distribution. 7.18 One short-term measure for sustaining improved payments discipline would be to change the algorithm for allocating sector revenues between generation, transmission and distribution companies, which currently shares cash shortfalls between these companies. Changing the algorithm so as to require that distribution companies bear more o f the burden o f non-payment would create incentives for distribution companies to maintain and improve payments discipline, and to more actively implement the disconnectionprograms currently inplace. 7.19 While modification o f the algorithm may yield positive results, it is unlikely to alone provide a full solution to the payments discipline problem. Currently, the bulk o f non-payment stems from large SOEs. In order that these entities will start to pay, government action is required, with the need to adopt and implement an industrial policy vis-a-vis restructuring and privatization / liquidation o f large enterprises. 7.20 The evidence suggests that the introduction o f the private sector inpower distribution can help to strengthen payment discipline; this has been successful to varying degrees in Albania, 155 Georgia, Kazakhstan, Moldova and Latin America.150 Ideally, in Romania the private sector should be introduced in the distribution through sale o f assets to strategic investors, though this might be difficult under current market circumstances; this is discussed inmore detail inSection 3. As an alternative, there may be scope for introducing the private sector under incentive-based management contracts, although incentives here would be weaker thanunder asset divestment. 7.21 Equally problematic to non-payments i s the high level o f non-cash payments in the sector, as shown in Table 7.2. These are high by SEE standards, and use on a level with non - cash payments in Montenegro and Republika Srpska (the highest non-cash systems in the region). Barter payments are problematic because they are non-transparent and are typically valued at non-market (inflated) prices. 7.22 Possible solutions to the problem here would be to cap the maximum percentage o f barter payments in the energy sector and to restrict the number o f companies allowed to engage in barter. In any event, if investment finance i s to be secured, collections should be improved and barter reduced. 7.23 Distribution losses - including commercial losses - in Romania at around 12 percent are not highcompared to other countries in SEE, but are highcompared to Western Europe. Though losses do not currently jeopardize cash flows because they are allowed for in the tariff calculation by the regulator, there is a question about how long this practice will continue given the scope for improved performance. Inorder to achieve improved performance, investments in network rehabilitation and metering, together with more active disconnection, are likely to be required. As with collections, experience shows that the introduction o f the private sector in distribution can help to reduce commercial losses. 7.24 Turning to other potential sources o f financial problems, the sector does not currently perform efficiently. Indeed, highcost thermal generation has been responsible inthe past for part o f the financial lossesrelated to the power sector. As noted above, tariffs now cover costs for this part o f the industry. Nevertheless, high cost power generation, with costs as high as 5 centskwh for some units, does represent a potential liability going forward. Two key areas inwhich there is scope for efficiency improvement are in retiring / rehabilitation o f old plants with low thermal efficiency, and - given the current over-manning, particularly in power generation-in labor restructuring. Measures to achieve performance improvement include further industry restructuring (discussed inSection E) and investment (discussed inSection H). 7.25 Power sector financial viability is also strained through effective price distortions in the district heat sector, given the cross-ownership o f district heat cogeneration assets. Effective price distortions in heat are manifest in a 2003 collection level o f only 80 percent for cogeneration plants. 7.26 The end user price for district heating is called the National Reference Price (NRP), and i s set on the basis o f what i s thought to be reasonable in affordability terms. The shortfall between the NRP and the allowable cost is financed by government transfers, formerly 45 percent from central government and 55 percent from local governments, although following For a discussion, see Transition Report 2001, London: EBRD. 156 recent changes, the central government will now pay a larger share. Problems as regards district heating financial viability ensue because the allowable costs are less than the actual costs, and government transfers have not been paid in full, particularly the component due from the municipalities. In addition, end user collections in the district heating sector which currently average around 70 percent are a major problem. 7.27 The key reform required to improve the financial viability o f district heating - and, indirectly, o f the power sector (to the extent that CHP assets remain under the ownership o f the national power generation company) - is to increase the NRP in a phased manner towards full cost recovery, and to take measures for offsetting adverse affordability consequences (discussed in more detail below in the context of gas price increases and affordability). In addition to mitigating affordability concerns, strengthening the social safety net should also facilitate increased collections, although this would also require that a more commercial approach be adopted by district heatingcompanies. B.3. The Power Sector Social Safety Net 7.28 Returning to the message mentioned above that power prices have increased and will have to continue to increase over time as investments are undertaken, it is important from the perspectives o f social acceptability and political feasibility that power remains affordable for all groups inthe population, through support provided by a social safety net. 7.29 The social safety net inthe Romanian power sector consists o f an optional tariff, whereby customers can choose to pay a low marginal rate for a first consumption block, and a higher marginal rate for additional consumption. Alternative optional tariffs are based on constant marginal tariffs somewhere between the two rates under the block tariff structure. Tariff levels under alternative options are set insuch away that the block tariff is attractive to customers who consume only to cover basic needs and no more. For other customers with higher consumption, alternative options to the block tariff are more attractive. 7.30 The social safety net for power can be judged by three criteria: (i) targeting-the extent to which it is focused on the poor as opposed to all consumers; (ii) coverage-the extent to which the poor are supported by the safety net; and (iii)incentives-any distortions to consumption decisions related to the safety net. 15' 7.31 The Romanian power sector's social safety net would target the poor assuming that consumption i s related to income (then the block tariff supports low consumption / low income customers). Under the same assumption, the subsidy covers the poor effectively but less so for the poor consuming in the second block. The block tariff is not ideal from an incentive point o f view, given that the marginal tariff for the second consumption block is relatively high in relation to marginal cost and encourages under-consumption for a class o f customers. For this reason, the current social safety net provides good interimsupport, although inthe mediumterm the target should be to replace this with a targeted subsidy (voucher or cash payment) for low income consumers. Is'See L.Lovei et al. (2002), "Maintaining Utility Services for the Poor," Washington, D.C: World Bank. 157 c.GASSECTORFINANCIAL VIABILITYAND AFFORDABILITY C.l. Gas PricingandAffordability 7.32 As a SEE regional gas market evolves, trade opportunities for Romania will increase. Thus, the opportunity cost o f gas consumption related to potential exports will be the border price. As demand grows andRomanian reserves are imported, the actual cost o f gas consumption in Romania will be the border price. On both counts, in the context of regional market development, gas prices inRomania should reflect border prices. 7.33 Current gas prices inthe region are set by Gazprom and are inthe order o f US$140 - 150 / tcm. Caspian gas imports to SEE via Turkey are likely to commence in 2007, at which point there will be the potential for shipping 7 bcm annually to the region, providing scope for competition with Russian gas and price reduction. Turkey has entered into off-take agreements for Caspian gas at a price believed in the order o f US$lOO-110 / tcm. Depending on the necessary infi-astructure investments and on transit fees, this might allow a slightly reduced gas price at the Romanian border. It is unlikely, however, that the border price would fall below US$120 / tcm inthe foreseeable future. 7.34 The January 2003 ,the end user gas price inRomania was around US$90/tcm for captive customers, including all residential consumers. Working back from this price and deducting margins for distribution, transmission and storage gives a well head gas price averaging around US$40 / tcm in2002, falling to US$15 / tcm inthe first six months o f 2003. Comparing the well head price and the border price, it can be seen that from an economic point o f view, and based on a regional perspective, large price increases for all customer categories (both residential and industrial customers pay a price well below economic cost) are required in the Romanian gas industry; the associated subsidy to gas consumers is worth around 2 percent o f GDP. 7.35 A gas tariff increase would bolster fiscal revenues and provide additional h d s for investment in the upstream gas industry. It would also provide the correct signals for decision making by residential consumers vis-a-vis the use of gas for heating rather than district heating. The current price o f gas is low relative to that o fdistrict heating (US$11/ Gcal for gas compared to US$20 / Gcal for district heating) thus encouraging direct consumption o f gas as opposed to district heating even though this may not be desirable from an economic point o f view. 7.36 Recognizing this, the Government o f Romania has increased gas prices significantly in recent years; most recently, there was a 10 percent price increase in July 2003 and 20 percent increase in September 2003. By end o f 2003, residential gas prices had reached $125 per tcm whilst other distribution customers paid $115 per tcm, and large industrial customers paid $99 per tcm. Though these prices increases are to be welcomed, further increases will be required going forward to ensure gas industry sustainability. The government has made a commitment in the Road Map to increase the domestic gas price to the border price by 2007, and proposes quarterly increases of 5 percent through 2004 in order to help meet this objective. The challenge now is to implementtariff increases according to this schedule. 158 C.2. The DirectAffordabilityImpactof Gas PriceIncreases'52 7.37 Significant gas price increases could potentially have an adverse effect on the affordability o f gas for residential consumers. This would not be desirable either socially or politically. Inspection of data suggests that there could be adverse affordability impacts for low and middle income consumers using gas-fired central heating and gas burners. This is not an argument against tariff increases, but is rather argument that these increase should be implemented in a context inwhich there is a social safety net based on properties outlined inthe above discussion o f the power sector (the subsidy should have good coverage, should be targeted, etc.). Finance for the targeted subsidy could come from increased fiscal revenues ensuing from tariff increases. 7.38 Inthe case o f Romania, a targeted subsidy for residential gas consumers is inplace. It was not possible for the purposes of this chapter to assess the proposed subsidy for next winter. Ifthe proposed subsidyisto be effective, then it shouldbeincreasedinlinewith tariffs to ensure that affordability ratios remain below critical levels. As regards a benchmark for the critical affordability ratio, the World Health Organization (WHO) defines fuel poverty as a situation in which a household spends more than 10percent of its income on heating. C.3. The Gas PriceImpacton DistrictHeating 7.39 Given that much o f district heating generation is gas fired, an increase in the gas price would result in an increase inthe price of district heating with associated affordability impacts. The indirect impact o f gas price increases - through the district heating sector -could be more wide-ranging than the direct impact - o n residential gas consumers - given that 60 percent of the total heat demand is met by district heating, with 2.8 million households supplied by district heating compared to 900,000 households supplied by gas. Inaddition to affordability impacts, an increase in the gas price would have a fiscal impact, since district heating is currently heavily subsidized. 7.40 District heating consumers are protected by the NRP, as discussed above. In addition, a subsidy is paid to end users on a means-tested basis under Emergency Ordinance No. 6/2003 as shown in Table 7.4. Notwithstanding this subsidy fkamework, the household budget data, together with data on non-payments, suggest that affordability remains a problem for low income groups inRomania. 7.41 Ifthe gasprice were to be increasedwhile a constantNRP ismaintained, thenthe impact would be purely fiscal, with increased subsidy from centraVloca1 government. 153 This would represent a transfer from the government to the gas industry which, while it would provide cash flows for investment, would not be desirable given the fiscal constraints. It would also not be desirable from the perspective of providing incentives for good performance in district heating 152This discussion draws on the report "Study on Gas Pricing and on the Fiscal System Applicable to Gas Production," carried out for the World Bank by HIS Energy, together with data from HouseholdBudget Surveys and calculations provided to the Bank by the government o f Romania. 153This assumes that all subsidy is paid. Inpractice, local government contributions are not always forthcoming. In this situation, the tariffincrease would firther strain the cash flow ofdistrict heating companies. 159 and o f providing the correct price signals for consumers, so that they will behave in an energy efficient manner. This is not an argument against raising the gas price rather, it highlights the shortcomings of the current district heating subsidysystem. Table 7.4 Householder DistrictHeating Support, 2003 Income (ROL per Support against person), up to actualbill(ROL per person), up to Monthly Annual Monthly Annual Over 5 months Over 12 months 750,000 9,000,000 1,680,000 700,000 8,400,000 1,250,000 15,000,000 720,000 300,000 3,600,000 1,500,000 18,000,000 480,000 200,000 2,400,000 2,106,000 25,727,000 360,000 150,000 1,800,000 Source: Reportby HIS Energy. 7.42 A better alternative from an economic perspective would be to increase district heating tariffs in line with gas prices, and to increase end user subsidies as necessary to maintain affordability ratios below the WHO threshold. The fiscal impact here could be reduced by rationalizing the existing subsidy system, and by phasing out subsidies paid directly to district heating companies and replacing these with targeted subsidies at the end user level. 7.43 A preliminary assessment carried out bythe World Bank,'54 shows that rationalization of the subsidy system could reduce govemment support to district heatingby around US$50 million per year. This would provide a goodbasis for increased targeted support following end user price increases. It should be stressed that phasing is important, given that the relative price increases for district heating could lead consumers to switch to gas when this is not desirable from an economic point o f view (because the gas price is distorted). 7.44 The work done by the World Bank is a first step towards reform of the district heating subsidy system. The next step would be for the govemment to undertake a fillreview o f district heating subsidy, using household budget data to assess the affordability impacts o f removing the subsidy to district heating companies and costing the targeted subsidy to maintain affordability for all income groups. As part of this step the impact of gas price increases and of offsetting government support could be evaluated. To reiterate, increasing district heating prices together with the end user subsidy would tighten budget constraints for district heating companies and would encourage consumers to behave ina more energy efficient manner. 154Daniel Aizic, "Analysis of the District Heating Sector in Romania 2000-2002," Washington, D.C.: World Bank. 160 C.4. Payments in the Gas Sector 7.45 Collections in the gas sector are currently around 90 percent, a figure that is low by industry best practice standards. As in the power sector, non-payment stems from large state- owned industrial enterprises, including district heating generation companies. As in the power sector, the solution here could be to introduce the private sector in gas distribution. As in the power sector, any solution will require the government's commitment to allow the disconnection o f non-payers, supported by industrial policy for the restructuring o f large SOEs, and will also require measures to improve the financial situation o f the district heating sector. D.REGIONALENERGY MARKET: CHALLENGESAND OPPORTUNITIESFORROMANIA 7.46 Regional energy market development poses three key sets o f challenges for Romania: (i) the need to move to energy industry economic and financial viability without government support; (ii)the need to provide institutional reform through industry commercialization / restructuring and regulatory development; and (iii) need to mobilize finance for investments. the 7.47 These challenges overlap with those deriving from a macroeconomic perspective: industry financial viability would reduce the energy sector's quasi-fiscal deficit and would support increased macroeconomic stability; institutional reform should lead to increased efficiency in the energy industry, which in turn would support the increased competitiveness o f Romanian industry; the securing o f finance and the undertaking o f investments would improve the security o f the energy supply-- a pre-condition for GDP growth. 7.48 As recently as 2000 the energy sector was largely unreformed and contributed around 5 percent o f GDP to the quasi-fiscal deficit through power and heat prices that were below cost recovery, domestic gas prices that were well below the import price, and collections problems in power, heat, and gas. Since that time, the energy-related quasi-fiscal deficit has been reduced owing to increased power, heat and gas prices, in tandem with institutional reform through the passage and implementation o f an Energy Law. Notwithstanding this progress, a number o f challenges remain if a commercially oriented energy sector is to ensue. 7.49 On the institutional reform side, Romania has made good progress. The outstanding challenges are to further commercialize both the gas and power sectors through private sector participation where this is feasible. 7.50 Regarding investments, a major rehabilitation o f power generation assets is required in the medium term. Failure to secure investments could see Romania go from its current status as a net exporter o f power to SEE, to a country dependent on imports and suffering from an insecure supply. 161 E. ENERGY INDUSTRYINSTITUTIONAL REFORM E.1IndustryRestructuring 7.51 The formerly monolithic power utility has been vertically unbundled, first through the establishment o f subsidiaries in 1998, and then by the full separation of the different industry components in 1999-2000. The industry now comprises a transmission company (Transelectrica), a company owning thermal generation assets (Termoelectrica), a hydro generation company (Hidroelectrica), and a distribution company (Electrica). This structure is consistent with requirements under the Athens Memorandum. 7.52 Labor productivity in Romanian power generation is currently low, partly because o f over-manning. Though this may not be a problem now, as wages increase and as competition in the regional market intensifies, low labor productivity could lead to the stranding o f generation assets. One way to avoid this would be to introduce a generation industry structure to promote increased productivity. Experience other countries elsewhere suggests that restructuring generation into a number of competing companies leads to performance impr0~ements.I~~ 7.53 The Government o f Romania has recognized the importance o f restructuring power generation and has unbundled Termoelectrica into a holding company with five subsidiaries: Electrocentrale Rovinari and Electrocentrale Turceni (owning lignite fired power generation assets); Electrocentrale Deva (owning hard coal fired power generation assets); Thermal power branches andElectrocentral Bucuresti (owning Combined Heat and Power [CHP] assets). 7.54 Such a structure would seem to provide a basis for the promotion of competition in generation, given an adequate number o f companies for the installed capacity in Romania, and such a structure would be consistent with promoting competition in generation. The next step should be the commercialization of subsidiaries, so that each subsidiary is able to operate in a regional market context on a stand-alone basis. It i s likely that commercialization o f this nature would require spinning o f f subsidiaries from Termoelectrica and introducing o f private management where this is feasible. 7.55 Inthe gas industry, the formerly integrated gas company was unbundled in 1998 in a manner consistent with requirements under the Athens Memorandum to an exploration/ production/storage company (Romgaz), a transport company (Transgaz), and two distribution companies (Distirgaz Nord and Distirgaz Sud) Inorder to consolidate the reform progress made inthis case, the introduction o f the private sector inthis unbundled structure is likely to leadto improved performance. To this end, a privatization adviser for the sale o f gas distribution companies has been hired, andthe privatizationprocess is proceeding. E.2 PrivateSector Participation 7.56 As noted above, the private sector has a potential role in energy sector performance improvement, particularly as regards the enforcement of payments discipline, but also more lS5 See D.Kennedy, "Liberalization o fthe Russian Power Sector," Energy Policy 31(2003) for a discussion. 162 generally, for example, to unlock efficiency gains in power generation and power gas / distribution. 7.57 However, there i s the a question o f whether the private sector will be interested inbuying / rehabilitating assets in Romania. More generally, the appetite o f the private sector for the purchase o f assets in the power sectors o f transition economies has declined recently for the following reasons: 0 The collapse o f ENRONhas led energy traders to focus on balance sheet strengthening rather than on intemational expansion. 0 Investors have lost money inthe United States andthe UnitedKingdom. 0 U.S. and Europeancompanies lost money inLatin America following currency devaluation. e Political / regulatory riskperceptions have increase following problems with PPAsinIndonesia, the Philippines, andPakistanandfollowingregulatory problems inHungary,Kazakhstan andMoldova. 7.58 Currently, there i s no interest in increased equity intransition economy power sectors on the past o f U.S., English, Spanish and Swedish companies that were formerly interested inthese markets. There is only limited interest from French, German and Italian companies. However, there is firm interest from one investor intwo Romanian power distribution companies, and the government i s indiscussionwith investors interested inpower generation projects. 7.59 The successful closing o f these deals and other privatization projects will require the development o f a strong institutional framework to support investment and to stimulate private sector interest. Even if the private sector were not to enter in Romania, the development o f the institutional framework would help to mobilize finance for public projects, possibly undertaken with private sector participation under incentive-based management contracts. One key aspect o f the institutional framework inthe context ofprivate sector participation is regulation. F.INDEPENDENT REGULATION 7.60 Private sector entry and, more generally, ongoing industry financial viability, will depend on the energy sector's regulatory framework. The development o f an independent regulatory framework can encourage private sector participation and mobilize investment finance by demonstrating the government's commitment to sustained cost recovery tariffs. 7.61 Romania has made significant progress in establishing an independent regulatory framework for the energy sector, establishing an independent power sector regulator - the National Agency for Electricity (ANRE) - in 1998, and a gas sector regulator - the National Regulatory Authority for Gas (ANGRN) - in2001. Bothregulators may be deemed independent under standard definitions: regulators are appointed for fixed terms; circumstances for dismissal are clearly specified in the primary legislation; funding for regulators comes from the energy industry through license fees; and regulators have full tariff-setting power. For both industries, Romania i s in compliance with the requirements for independent regulation under the Athens Memorandum. 163 7.62 Interms o f the regulatory tariff methodology for the power sector, as discussed above, this provides adequate revenues to cover costs, including depreciation, and to allow for losses, but it does not allow for non-payment by billed customers. The methodology provides some security for potential private investors/financiers, together with incentives to improve collections. The evidence from the power sector suggests, however, that a long-term tariff methodology (sometimes called Price Cup or RPI-X regulation) can best provide investor security and incentives for efficient perfom~ance.'~~ 7.63 Under a long-term methodology, tariffs are set on a forward looking basis over a period typically between five and seven years. Adjustments can be made for changes in key variables such as inflation, the exchange rate, demand, and input prices. The benefit o f a long-term methodology is that it gives regulated companies security over medium-term cash flows, something that is not necessarily true o f tariff methodologies which reset tariffs on an annual basis. Recognizing the advantage o f a long-term methodology, the Romanian power sector regulator has made a commitment to introduce a long-term (price cap) methodology from 2005, although there would be benefits inmovingthis date forward. 7.64 Inthe gas sector, prices are set inconformance with the objective stated inthe RoadMap to move to import parity by 2007. The challenge for the regulator is to continue to implement tariffs increases in the face o f potential political resistance. The regulator has started to build a track record in this respect with the two tariff increases in 2003, and should now build on these achievements. Regarding the time frame for tariff setting, the gas regulator has recognized the benefits o f adopting a long-term methodology and will move towards adopting such a methodology inthe medium term. G.MARKET RULES 7.65 There are a number o f important questions in the context o f power market liberalization as regards institutional design. These questions include the market base -- whether the market should be based on bilateral contracts or on a pool model; the role o fbalancing markets: the real time clearing o f demand and supply: the role o f a market operator: investment support mechanisms: and generationprice regulation. These questions will be discussed inturnbelow. 7.66 Evidence from transition economies (Georgia, Kazakhstan, Ukraine) suggests that power markets based on bilateral contracts between generators and large consumers are the most appropriate in a setting in which payments discipline is a problem and institutional capacity constraints may be binding.'57This model i s widespread inEurope and was recently adopted in Englandand Wales to replace the .formerpower pool. It is envisaged that the SEE REMwill be based on a bilateral contracts model. 7.67 A bilateral contracts market typically does not cater for the whole o f the demand (customers in the market aim to avoid over contracting, that is, the purchase under contract o f capacity which they may not use). Inorder to meet residual (i.e., non-contracted) demand, and to "'SeeTransition 15'See Transition Report 2001. Report 2001. 164 allow trade o f contracts, an additional mechanism is required. One such mechanism is a day ahead spot market where generators and customers canbuyand sell power. 7.68 A day ahead spot market requires amarket operator, responsible for clearingdemandand supply, and subsequent financial clearing (ie., overseeing the flow o f funds from purchasers to sellers). A regional spot market is likely to perform better than a set o f subregional markets in SEE, particularly given the small size o f some o f the countries involved. The market operator would also be responsible for overseeing the contract market and ensuring the consistency o f transactions with the commercial code governing the market. 7.69 Since the power demand in Romania is not responsive to price in real time.15* it is not possible to operate a real time market (where price clears demand and supply). To meet the differences between real time and forecast demand, the best that can be achieved is to purchase power efficiently, for example, through the tendering o f contracts by the system operator for spinning / other reserves. InRomania, tendering would ideally take place on a regional basis to allow the full sharing o freserve between countries. 7.70 Since price signals do not function inpower markets as inother markets (i.e., they cannot fully clear demand and supply), and in view o f un-hedged price volatility, free entry in generation is not a sufficient condition to guarantee that the optimal level o f capacity will be provided'59. Given that this i s the case, countries in which power markets have been liberalized rarely rely on free entry to provide an adequate systemreserve margin. 7.71 Examples o f wage inwhich countries have dealt with this problem include the following: (i)transmission system operators inAustralia and Scandinavia have contracted inreserve capacity to meet target reserve margins; (ii) large consumers (e.g., in the United States and Canada) are required to purchase to cover demand plus a reserve margin; and (iii) capacity payments are made, (for example, in Chile and (before reform o f the pool) in England and Wales. 7.72 Inthe case ofRomania, the following additional considerations will be important: (i) the market rules will be untested; (ii)the regulatory rules will be untested; and (iii)the creditworthiness o f participants inthe market will be untested. Under circumstances, and inlight o f significant risks for investors, an investment support mechanism is likely to be required if an adequate reserve margin is to be maintained. 7.73 Inview ofthese considerations, the proposalbythe Government ofRomaniainthe Road Map to require forward contracting would seem reasonable on the basis o f intemational experience and the Romanian context. As regards trading, the government's proposal for a day ahead market together with a simple arrangement for the efficient purchase o f power inreal time by the systemoperator would seem appropriate. Although detailed market design inthe context o f the SEE REMhas not yet been developed, it is likely that market rules will involve bilateral contracts with day ahead trading and simple arrangements for real time power purchase. 15' This will be the case for the foreseeable hture given the installed technology for metering and data communication inRomania and SEE. See R.Turvey, "Ensuring Adequate Generating Capacity," Utilities Policy 11(2003), for a discussion. 165 7.74 Whether the SEE REM will comprise a regional market operator or a set o f national operators is not yet determined. Ideally, a regional market operator would probably result in more efficient plant dispatch. As a second best, coordinated national markets would result in benefits from international trade. The proposed power exchange inBucharest will have a role to play inthe regional market (this to be determined as SEE REMmarket rules and institutions are agreed between participating countries). 7.75 The experience from Canada and the United States shows that reliance on price as the sole means o f rationing capacity can result in sustained high prices, which may be undesirable from an economic viewpoint and may be socially and politically unacceptable. This could be a problem in Romania and in the SEE REMmore generally, given the possibility o f national and regional supply deficits. 7.76 One way to avoid unduly high prices is through contracts between generators and residential consumers (e.g., distribution companies). Inthe case o f Romania, long-term contracts between distribution companies and hydro companies might be considered as a way o fproviding consumer protection. Alternatively, market design might include generator price caps and/or supply obligations (e.g., to tariff consumers). 7.77 On network access, zonal transmission pricing in Romania is consistent with the basic principal that tariffs should reflect locational costs so that dispatch and entry in a liberalized market may be efficient. It is important in a regional context that transmission tariffs support imports where these are the most economically efficient means o f meeting demand. Romania i s working together with other SEE REM countries to develop a cross-border pricing mechanism that will support efficient trade. 7.78 Going forward, the challenge for Romania is to buildon the broad principles which it has endorsed, elaborating detailed implementation proposals, particularly inthe following areas: 0 Defining the inter-relationship between contracts and day ahead markets 0 Developing a (real time demand-supply) balancing mechanism 0 Developing a capacity obligation - or other investment support -mechanism 0 Defining the relationship between the Romanian and SEE REMmarket operators 0 Developing a mechanism for the protection o f residential consumers ina liberalizedpower market context 0 Working with other SEE REMcountries indeveloping a common cross-border transmissiontariff methodology. H.INVESTMENTINTHEPOWERSECTOR 7.79 From a technical point o f view, Romania is currently connected to the regional power network, operating synchronously and in parallel with other SEE REM countries (excepting Turkey) that together comprise the second zone o f the Union for Coordination o f Transmission o f Electricity (UCTE). It is envisaged that inthe medium term, following network rehabilitation investments inBosnia and Herzegovina and Croatia, the second UCTE zone will be connected to 166 the main UCTE European network. At this time, trade between Romania and other countries in SEE and Europe will be technically feasible. 7.80 Inrecent years Romaniahasbeen amajor player as regards power trade inSEE. In2001, for example, o f 14 TWh total power trade in the region, 3.4 TWh o f this trade was due to Romanian exports. Whether Romania continues to be a major power exporter will depend on the performance o f Romanianpower generation relative to other generation inthe region. This poses a major challenge for Romania given the profile o f the current installed generation capacity. 7.81 Installed capacity in Romania is roughly 19,000 MW, comprising 32 percent hydro, 64 percent thermal, and 4 percent nuclear. Regarding the thermal capacity, 32 percent o f this i s over 30 years old, and 50 percent is between 20 and 30 years old. Following the planned decommissioning to 2008, the available capacity will fall to around 10,000 MW, Which will leave Romania with an inadequate reserve margin. Without investment in power generation, Romania would have to forgo economically viable power rehabilitation projects, and would have no alternative but to increasingly import power from the regional market. Investments on the order o f US$7 billion to 2010 would provide Romania with adequate capacity,'60 allowing continued exploitation o f indigenous natural resources (such as lignite from the Oltenia surface mines) and export o fpower to SEE. H.1.TheFrameworkfor Investment 7.82 Large investments will be required if Romania is to compete in the SEE REM, and mobilizing finance for investments represents a major challenge for the country. Small investments may be self-financed by the industry or financed commercially without the need for long-term contracts as security. For larger investments (such as major rehabilitation), given the current financial market conditions, some long-term contracts between generators and consumers/intermediaries are likely to be required for finance to be secured. The reason for this i s the risk aversity of investors (particularly financial investors as opposed to strategic investors) together with the high degree o f uncertainty in SEE over market rules (as regards both design and implementation). 7.83 There is a problem with long-term contracts that are signed before effective competition sets in: these contracts may support investments that would not be selected in a competitive market. In addition, generators with contracts are insulated from competitive pressure. Furthermore, reducing the number o f buyers in the market through locking into long-term contracts can undermine incentives for new entry ingeneration. Under these circumstances, long -term contracts may result inhigherprices for consumers. 7.84 Long-term contracts have proved to be an obstacle to power sector liberalization around the world, (for example, inthe United States and Spain) and, closer to Romania, inHungary and Poland. In the latter two countries, long-term Power Purchase Agreements (PPAs) signed between power generators and national transmission companies would not be sustainable in a market context: prices in PPAs are high relative to forecast market prices. The authorities in I6O This figure comes from the Road M a p and has not been subjected to independent assessment by the Bank. 167 Hungary and Poland are moving to buy out PPAs through a stranded cost resolution mechanism prior to full market liberalization. Stranded cost resolution mechanisms are typically financed by consumers through a power system levy. 7.85 Long-term contracts can slow down the process o f liberalization until the issue o f stranded costs is resolved, and they can also undermine benefits that would ensue from competition. Thus, the need to provide security for investors should be balanced by the objective of promoting sector competition. 7.86 Inorder to strike the appropriate balance, standard policy advice is that all but the most urgent investments should be postponed until after sector liberalization takes place. Inthe case o f Romania, this may not be feasible: investments are required inthe mediumterm before markets are to be liberalized in the context o f the SEE REM. Given that investments are required, and that these would not be delivered on a merchant basis (i-e., without long-term contracts), the objective should be to minimize the scope of the long-term contracts required to support necessary investments. 7.87 One way to reduce the scope o f contracts is to design these contracts to cover capacity but not to contain off-take obligations. For example, a distribution company with a capacity contract would agree to pay the capital cost associated with power generation covered under the contract, and would have an option to purchase power at a specified price or according to a specified formula, but would not have to purchase power from the plant. The advantage o f such a contract over a PPA is that it allows efficient dispatch, whereas under PPAs plants with relatively high marginal costs may continue to generate even though this is not economically desirable. Efficient dispatch would ensue through the trading o f capacity contracts in Romania's proposed forward contract markets and the day ahead market. 7.88 Other dimensions for reducing the scope o f long-term contracts are the related capacity covered by the contract and the contract's duration. For green-field investments, experience elsewhere inSEE (e.g., inBulgaria) suggests that financiers require contracts covering the whole o f capacity (i.e., debt and equity financed) for a period extending beyond loan tenors for debt. In the case o f Romania, medium-term investments are likely to be brown-field, with only some units inthe existing plant requiringrehabilitation. This couldprovide scope for reducing contract coverage. 7.89 Where assets requiring rehabilitation are bundled together with operating assets, cash flows from the latter can provide extra security for investors. In other words, the bundling o f assets would facilitate corporate as opposed to project finance o f investments. Under these circumstances, long-term contracts might only relate to a level o f capacity required to meet debt service, while equity returns would be derived from capacity competing in the market. Dependingon the level o fcash flows from operating assets, there may be scope to further reduce the coverage o f long term contracts so that they cover only a part o f debt service (in any given year, or beyond a certain date). The duration ofcontracts will probably have to be longer than the proposed five-year vesting contracts inthe Road Map ifdebt finance is to be secured. 168 7.90 One possibility would be to further enhance the cash flows o f project companies by bundling thermal assets together with hydro assets. Rents to hydro assets deriving from low marginal costs could provide further support for rehabilitation with reduced coverage o f long - term contracts. The question is whether project companies would be sufficiently creditworthy without hydro assets. Ifthis is the case, then the bundling o f hydro assets with thermal assets would risk reducing future privatization revenues. The alternatives o f keeping hydro assets under public control, or o f privatizing on a stand alone basis, would probably yield greater benefits to taxpayers andpower consumers. 7.91 It is likely that some o f the proposed thermal generation companies will have adequate cash flows to support investments with limited scope long-term contracts. Inparticular, Rovinar and Turceni - two candidates identified by the government for privatization - are both o f adequate scale and comprise a mixture o f units o f different ages -- some newly rehabilitated and some with a number o f years o f design life left. InRovinar, two units are mothballed and there are plans to add a new 300 MW unit, while four units o f Turceni require rehabilitation. The thermal efficiency o f these plants i s comparable to that o f other plants in the region, and both plants are located close to reasonably low cost lignite mines, thus continued dispatch in a regional market would be likely. H.2. ContractingParties 7.92 In principle, generators could have long-term contracts with Transelectrica or with distribution companies. Since distribution companies have stronger incentives to enter into contracts that would be viable in a competitive market (they would have to bear any contract- related stranded costs, whereas a transmission company might pass on stranded costs to consumers) the latter i s preferable. This assumes that distribution companies act in a commercial manner, which would seem to be a reasonable assumption for at least some o f the Romanian distribution companies, particularly as these are privatized. 7.93 Commercial distribution companies might enter into contracts to realize attractive purchase prices, to hedge price volatility, or because o f license obligations. Indeed, the proposal by the Government o fRomania inthe Road Map is that distribution companies will be required to forward contract o f sufficient capacity to cover the expected demand plus a reserve margin. To the extent that distribution companies are not commercial, the danger is that contract-related costs could at some future time become stranded. In order to mitigate this risk, investments in power generation might be demonstrated to be least cost from a regional perspective as part o f project feasibility anddevelopment. I.CONCLUSION 7.94 The analysis in this chapter finds that effective tariffs-reflecting tariffs and payments discipline-are still below cost recovery levels in the Romanian energy sector. In the power sector, power tariffs will have to increase over time to support the financing o f investments. Collections remain a problem in the power sector, owing to failure to pay on the part of large SOEs and a high degree o f non-cash settlement. In the gas sector, the main challenge is to 169 increase tariffs to reflect border prices. Though some progress has been made here, further increases are necessary. 7.95 This chapter has arguedthat while good progress has been made inpower tariff reform, further price adjustment and restructuring will be required as investments are undertaken. Inthe gas sector, while there have been significant price increases, further large increases are required in order to meet import parity. The challenge for the government is make good on its commitment inthe RoadMap to move to import parityinthe medium term. 7.96 Regarding the affordability consequences o f tariff increases, these are unlikely to be problematic in the power sector where expenditure is small relative to income and where an effective social safety net is inplace. Gas price increases, on the other hand, are likely to strain affordability inthe gas and heat sectors. Targeted social safety nets for gas and heat will require additional financing if affordability risks are to be mitigated; part o f the increased finance could be unlocked through the rationalization o f the current blanket subsidy to the district heating sector. 7.97 Payments discipline remains a problem in power and gas, notwithstanding increased collections in recent years. For collections to be increased to levels which would permit energy industry financial viability on a sustained basis, government action is required, through allowing the disconnection o f large SOEs and through developing an industrial policy for their restructuring / privatization. 7.98 On institutional reform, Romania has progressed further than all SEE REMcountries bar Croatia in regulatory development, and further than all countries bar Bulgaria on industry restructuring. Interms o f power market development and gas industry regulation / restructuring, Romania has outperformed all SEE REMcountries. Outstanding challenges are to build on the track record inregulation through continuing to implement price increases as necessary inpower and gas, and to move towards the introduction o f tariff setting on a long-term basis. Regarding industry restructuring, the challenges include the need to commercialize the o f generation subsidiaries and the need to introduce o f the private sector in the unbundled gas structure. On market design, where proposals are consistent with those for the SEE REM,the challenges relate to detailed implementation issues. 7.99 As regards investment, Romania's continued status as a net power exporter to SEE will require large-scale generation rehabilitation. The challenge here is to develop a framework to secure investment that is consistent with the objective o f liberalizing the power market. The chapter argues that capacity contracts could provide a solution here, and that these contracts would be consistent with proposedmarket design inRomania. 7.100 In summary, if Romania is able to meet the challenges in effective energy tariffs and operating efficiency, through applying hard budget constraints, closing unviable thermal generation units, and implementing institutional reforms, together with undertaking the necessary investments, this should bring about a domestic competitively priced energy as well as export opportunities. Such an outcome would result in economic benefits in a sub-regional energy market context and, more generally, ina regional market context. 8. ENVIRONMENTAL STANDARDS:UPGRADINGAND CONTAININGTHE COSTS A, INTRODUCTION 8.1 The experience o f countries that joined the EUon M a y 1o f this year shows that estimates o f investment costs cover a wide range, depending on the their respective initial conditions, and depending also on the choice o f the types o f investments, which also cover a wide range o f possibilities. Similarly, timetables vary across countries, with earlier reformers having more generous timetables than other selected CEECs. The experience to date o f acceding countries also shows that there may be important differences between planned and actual investments. However, the total investment costs do not stop with the accession date, since compliance with environmental standards, while it offers a range o f technology choices, must fknction ina single market. The estimates o f total investment costs used in this study are, therefore, only indicative and can vary within a wide range depending on the investment type and the sequencing o f investments. Table 8.1 Key Environmental-Related Indicators: a regional perspective YOPopulation co2 Under 5 YOPopulation With access to GDP/unitof emissions/ Mortality Country with access to improved energy unitof GDP Rate/1000 sanitation water source (PPP$/Kgoe) Kg/PPP$GDP live births Bulgaria - 100 100 2.8 0.9 16 Cyprus 100 100 6.3 0.4 6 CzechRep. 3.6 0.8 5 Estonia 2.9 1.4 12 Hungary 99 99 4.9 0.5 9 Latvia 4.6 0.4 21 Lithuania 67 67 3.9 0.5 9 Malta 100 100 6.7 0.7 5 Poland 4.0 0.9 9 Romania 58 53 3.4 0.7 21 SlovakRep. 100 100 3.6 0.7 9 Slovenia 100 100 5.0 0.5 5 Turkey 82 90 5.3 0.5 43 Source: World DevelopmentIndicators,2003, the World Bank. Countrieswithout dataon sanitationand improved water generallyhavevery high levelsofprovision. 8.2 This situation presents opportunities and challenges for Romania in the area o f the environment. Of all these countries, including Turkey, Romania has, in environmental terms, the greatest ground to cover to reach the standards required o f member states. Table 8.1 provides some o f the key indicators, which show Romania as having the lowest level o f access to improved water and sanitation, the third poorest energy efficiency measured in terms o f GDP 16' This chapter was preparedby Ani1 Markandya (ECSSD) of the World Bank. 171 produced per KG o f oil used, and the second highest under 5 mortality rate (after Turkey). In terms o f C02 emissions per unit o f GDP, Romania is exceeded only by Bulgaria, the Czech Republic, Estonia and Poland.'62 8.3 Inview o fthe size oftheproblem facing the country, andgiven that Romania's percapita GDP is the lowest among the selected countries except for Turkey, it is not surprising to find the burden o f environmental improvements is considerable. One measure o f this is provided in the early European Commission (EC) study o f the environmental costs of meeting the environmental part o f the acquis communautaire, covering 10 o f the 13 countries in Table 8.1 (excluding Cyprus, Malta and Turkey). Romania's costs o f meeting EU environmental standards amount to 56 percent o f its 2001 GDP, compared to an average o f 29 percent across the ten countrie~.'~~ The total cost for Romania was estimated at around 21 billion. More recent figures actually come up with an even higher number (over 29 billion, see below). While this type o f exercise remains somewhat imprecise, the conclusion that the environmental component o f the acquis presents the country with a formidable challenge is uncontested. 8.4 This chapter focuses on how Romaniawill meet this challenge. Although the government has now closed the environment chapter, there are many issues that remain to be discussed. Prominent among these issues are how the required public funds for meeting the targets will be generated and whether those programs that depend on non-budgetary contributions will infact be able to be funded at the level that is expected inthe agreements with the Commission. There are also concerns about the government's administrative capacity for managing the implementation of the acquis in this area,'64 and about the impacts o f the programs on the affordability o f the services (especially water supply and sanitation) for the population. 8.5 The discussion in this chapter is structured as follows. Section B looks at the potential benefits, o f meeting the environmental targets o f the acquis, drawing on a major E C study o f the benefits that included Romania. Based on this and on the financing gaps referred to above, the section puts forth some recommendations for the sequencing o f the environmental investments. It should be understood that even within the agreements reached with the Commission on the environmental chapter, there is some choice regarding the phasing and sequencing o f investments. This will be driven largely by the available funds but should also be responsive to the potential benefits o f makingthe investments. 8.6 Section C looks at the costs of the environmental acquis inmore detail, by directive and by sector (budgetary versus non-budgetary, and, within the latter, local government, state enterprises and the private sector). The Section compares, to the extent possible, these estimated costs with past expenditure on the environment by each o f these groups. Where funding sources 162 Not all important environmental indicators are included, as comparable data are not available. One other indicator for which Romania also has a poor record is air pollution interms o f suspended particulates. Inmajor cities this is well above the recommended maximumWHO levels. 163Only Bulgaria had higher costs as a percent o f GDP. The study being referred to is EDC (1997), "Compliance Costing for Approximation of EULegislation inthe CEEC,". EuropeanCommission, DGEnvironment, Brussels. 164This was expressed quite forcefully by the Commission itself in "Commission o f the European Communities, 2002: "Regular Report on Romania's Progress Toward Accession," COM (2002) 700 Final," EC, Brussels. 172 have been identified these are noted and from this a `financing gap' is estimated. Such a gap is, o f course, not set in stone. It can be reduced by taking measures that shift spending from one category to another (e.g., from the public to the private sector), or that reduce the required level o f spending intotal (e.g., by reducing demand for services where it i s at present too high). These options are reviewed and discussed qualitatively. Mobilizing additional resources, such as through the proposed environmental charges can also fill the financing gap and the potential for these is also evaluated. fl 8.7 Section D examines some o f key aspects o f the acquis program inmore detail. It looks at the following (i) ways in which the budgetary burden can be reduced: (ii) potential for the mobilizing additional resources internally and externally for the environment: (iii)the administrative capacity constraints that have to be addressed: and (iv) the affordability o f the tariff increases and the constraints that they pose interms o f implementing the program. Ineach case the chapter offers some suggestions and recommendations for action. Finally, Section E provides the conclusions to the chapter. B. BENEFITSOFMEETING ENVIRONMENTAL THE PART OF THE ACQUIS INROMANIA 8.8 In view of the high costs of meeting the environmental directk-es, it is important to identifythe benefits o f compliance. The EC has recently quantified the expected benefits for the CEECS.'~~The results o f this exercise are given in Table 8.2.'66 The bznefits covered in this exercise include health, resource, and ecosystem protection, and the directives covered include all those with an impact on air, water, and solid waste. The results are reported as the net present value of benefits, discounted at a real rate of 4 percent. 8.9 The expected benefits arising from the meeting air standards are generally the highest, followed by those arising from stricter water standards, especially if the upper estimates are taken (the exceptions are Bulgaria and the Czech Republic). However, estimates range widely across categories and ~0untries.I~~The greatest uncertainties are associated with the potential benefits o f EU waste treatment standards, followed by the benefits o f air and water standards. Some o f the variations across countries, on the other hand, result from differences inthe baseline environmental conditions. The Czech Republic has the greatest river length o f all the CEECs (76,000 km). At the same time, not a single river is o f "good" quality (by EUstandards); that is, 10 percent are of "fair" quality, 10 percent are o f "very bad" quality, and the remaining 80 percent are o f either "poor"' (40 percent) or "bad" (40 percent) quality. Compliance with EU water directives will improve this situation considerably: 10percent are expected to be of "good" quality, and all rivers o f "poor," "bad," or "very bad" quality are expected to improve to fair quality after successful implementation. This yields substantial benefits to the Czech Republic, giving it muchhigher figures than those for some o f the other countries. ~~~ The study covers all 13 countries: Bulgaria, Cyprus, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, the Slovak Republic, Slovenia, and Turkey. 166Inactual practice, countries mayuse higher discount rates, ifonly to ensure that theyprioritize those investments that will yield earlier benefits for the populations. 167The high estimates vary by as much as an order o f magnitude as the low estimates, reflecting the substantial uncertainties surrounding environmental benefit estimation ingeneral. 173 Table 8.2. TotalBenefitsover the Period2005-2020 ( Billion) Water Air Waste Total Benefits Per Low High Low High Low High capita Low High EuroOOO Bulgaria 1.58 4.20 1.07 11.00 0.20 6.62 2.85 21.82 3.01 CzechRepublic 15.23 24.05 7.10 35.10 0.93 11.20 23.26 70.35 9.09 Hungary 2.72 10.49 5.74 39.92 1.12 18.50 9.58 68.91 7.77 Poland 13.59 31.96 25.80 149.90 1.60 26.30 40.99 208.16 6.44 Romania 3.96 12.15 7.59 56.95 0.83 26.30 12.38 95.40 4.79 Slovak Republic 3.00 6.61 3.40 21.90 0.29 4.28 6.69 32.79 7.31 Slovenia 1.47 3.44 0.68 4.62 0.24 2.82 2.39 10.88 6.64 Baltics Estonia 0.26 0.99 0.39 2.05 0.09 1.75 0.74 4.79 3.95 Latvia 0.38 1.34 0.49 3.12 0.05 1.07 0.92 5.53 2.69 Lithuania 1.23 2.75 1.56 7.98 0.06 2.00 2.85 12.73 4.21 Total 43.42 97.98 53.82 332.54 5.41 100.84 102.65 531.36 6.06 As percentoftotal 42.00 18.00 52.00 63.00 5.00 19.00 100.00 100.00 Note: Net present value at a 4 percent discount rate. Source: Ecotech, 2001, The Benefits of Compliance with the Environmental Acquis for the CEECs. Brussels: European Commission. 8.10 For Romania, Table 8.2 shows that the benefits o f upgrading air quality are the highest, followed by water and waste. Taking the lower bound estimates, the air benefits have a present value o f7.6 billion against an estimated investment cost o f2.8 billion. The lower bound o f the water benefits, however, is about 4.0 billion against a capital cost o f 7.2 billion, and the waste benefits come out at a minimum o f 825 million against an investment cost o f 8 billion. Overall, the distribution o f benefits between the media does not depend on whether one takes the lower or higher benefits figures. Within the water category, however, the breakdown among drinkingwater, bathingwater, andriver quality improvements does depend onwhether one takes the lower or higher values. With the lower values, bathing water improvements account for about two-thirds of the total, whereas with the higher values, drinking water accounts for about 72 percent o f the total and bathing water for only 22 percent.I6* In other words, there is much uncertainty about the drinking water benefits, because it is difficult to ascertain exactly how muchhouseholds are willing to pay for these improvements. 8.11 The main implications o f the benefit analysis are the following: 0 Greater care is needed in the phasing o f the water and waste investments than in the following ofthe air investments 0 Further analysis is needed inthe areas o f solid waste and drinkingwater. "* The breakdownwithin the water sector is not showninthe table. 174 c.COSTS OFUPGRADINGENVIRONMENTAL STANDARDS C.1. Estimates of Total Costs and Costs by Year 8.12 Table 8.3 provides the latest estimates o f the costs o f the main environmental directives, the time period o f compliance, and an estimate o f the shares o f the investments that will have to come from the different sectors. The data are based on detailed Romanian studies carried out for the directives, where these were available, and on detailed estimates from Poland and other countries where these were not available. The total investment cost comes to nearly 30 billion, which is about 8.7 billion or 29 percent morethanthe 1997 EC estimate. Calculations havealso been made for the additional operating costs, and these come to between 80 and 120 per personper annum, or about 2.2 billion intotal. 8.13 How much o f this investment expenditure can Romania expect to meet from EU funds? The best estimate indicates that from 2004 to 2006, PHARE and ISPA funds, which are dedicated to supporting the environmental directives, will provide about 2.8 billion. This i s based on the EC 2002 communication covering Bulgaria and Romania.'69 which details the total allocation o f PHARE, SAPARD and ISPA funds and the division between SAPARD and other funds obtained by direct communication with the EC. After 2006, the Cohesion Funds will take over, and the estimated amount available to 2015 is about 4.1 billion, based on similar calculations for Bulgaria. Thus, in total, Romania can expect about 6.9 billion, which leaves 22.8 billion to be funded from other sources, principally domestic, over about 18 years (the longest negotiated period i s 2022). It is often not fblly appreciated how muchthe environmental upgradingwill cost the candidate states, and the view that EUfunds will take care ofmost of the additional expenditure i s clearly false'70. 8.14 These expenses are, o f course, spread out over several years, as Table 8.3 indicates, and the expectationis that economic growth will make it easier to implement the program inthe later years. Also the said expenses are provided by the line ministries, hence they are presumed to be inlinewith commitments made duringnegotiations. Since it is particularly interestingto see the annual expenditure implications o f the environmental chapter, some year by year breakdown is provided inTable 8.4, which shows annual expenditure plans o f the government where available. Inother cases, which because of data the limitations cover most of the directives, the annual sequencing has been assumed to be similar to that o f the 2004 entrants (for examples Poland).17' The resulting figures show annual investment starting at 1.6 billion in 2004, increasing to as high as 4.0 billion in 2007, and then declining slowly to 1.4 billion by 2015 and to around 400 million in2022 (not shown intable). "Communication from the Commission to the Council and the European Parliament: Roadmaps for Bulgaria and Romania," COM (2002) 624 final, Brussels. I7OThe amount available to Poland for the environment is about the same as that for Romania (5 billion) in spite o f its population being about 75 percent higher. Bulgaria, on the other hand, will get about 3 billion, with less than halfo fRomania's population. 17` The only directive for which sequencing information was provided was 2001180 (air quality). For the others, the annual shares o f the total are assumed to be the same as for Poland. 175 Table 8.3 Estimates of the Investment Costs of the Environmental Directives Directive Number Cost Mn. Period for Source of Compliance Sectors Data Water Quality 761474 933 2015 Enterprises 11 911271 8.000 2022 State: CentrallLocal 21 91I676 1,200 2014 State andEnterprises 21 2001180 1,402 2011 Mainly Enterprises 21 93112 140 2007 Enterprises 21 Air Quality 94/63 363 2009 Mainly Enterprises 21 96/62 42 2007 CentralState Budget 21 98/70 380 2007 Enterprises 21 99/13 474 2015 Enterprises 21 Waste Management 94/62 243 2010 State and Enterprises 21 99131 8,196 2020 State and Enterprises 21 IPPC 96161 4,147 2015 Mainly Enterprises 11 791409 10 2006 State Budget 21 92/43 10 2006 State Budget 21 Other 96129 19 2007 State Budget 21 97143 2,500 2007 State Budget 21 2003154 903 2015 State and Enterprises 21 TOTAL 29,726 Notes: 11Basedon morerecent studies for countriesjoining in2004, adjustedfor Romania.21Basedon primary studies carriedout inRomania. Source: See Annex 7 for details of the directives Table 8.4 Estimates of Annual Investments Needed to FulfillEnvironmental Directives , 2004-15 (Mn.) 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Total Water 820 669 546 675 731 731 812 852 569 592 512 512 8,022 Air 51 131 528 609 370 439 368 284 0 0 0 0 2,780 Waste 119 97 226 362 790 871 871 872 905 704 671 671 7,157 IPPC 520 282 255 255 349 396 438 459 374 342 243 235 4,147 Other 106 109 548 2134 73 73 73 44 44 44 44 44 3,337 Total 1,615 1,288 2,103 4,035 2,313 2,509 2,562 2,510 1,893 1,682 1,470 1,462 25,444 Source: See Annex 7 for details o f the assumptions made in the calculations. The actual program runs to 2022 where approximately4.2 billion is spent in the last sevenyears. 8.15 Although these figures are rough estimates, they are sufficient to provide a clear indication that the level o f spending on the environment will have to increase very substantially with respect to past levels. The OECD database has information on Romania's environmental expenditures to 1998 (as well as for other CEECs). Investment expenditures are reported ineuros from 1993 to 2001 in Table 8.5. They show a steady increase over this period, fi-om 79 million to 238 million, but even at that level they are considerably below the levels required in the hture as given inTable 8.4. The "gap" innational spending is also evident from the comparison with per capita expenditures inPoland and the Czech Republic, which were higher in 1993 than Romania's expenditures in 2001.'72 In addition, as shown in Table 8.5 there is the decline in expenditure in the public sector (both central and local) over this period. The 2001 figures for both are about 70 percent lower than the 1993 figures, during a period in which investment expenditures should have been increasing inanticipation of EUaccession. The OECD database excludes investment expenditures on natural resource management. In most cases these figures are a relatively small part o f total investment. A similar table drawn up for total environment-related expenditure comes up with much the same conclusions - levels of spending in Romania as much as an order of magnitude lower inper capita terms than insuch countries as the Czech Republic. 176 Table 8.5. Total Environmental Investment Expenditures, 1993-2001 ('000) 1993 1994 1995 1996 1997 1998 2001 Public 45,116 38,707 57,210 51,711 75,919 107,029 13,297 Central 8,121 6,967 10,298 9,308 13,666 19,265 2,393 Local 36,995 31,740 46,912 42,403 62,254 87,764 10,904 Private 33,928 35,170 49,147 57,689 65,419 86,188 224,677 Total 79,044 73,877 106,357 109,400 141,338 193,217 237,974 Per capita(euros) Romania 3.47 3.25 4.69 4.84 6.27 8.59 10.62 Poland 18.54 20.43 25.82 46.32 51.16 59.46 CzechRep. 54.91 79.64 88.93 102.35 111.18 90.93 Note: Only the investment expenditures in 2001 include expenditures for nature/environmentalprotection. From 1993-98, "investment expenditures" = end-of-pipeinvestments+processintegratedinvestments. Source: OECD (1993-98 data), EnvironmentalStatistics of Romania(2001 data). See Annex 7 for some assumptionsmade in the calculationo fcentraland localgovemment spendingduring 1993-98. C.2. The Financing Gap after EUGrants Have Been Taken into Account 8.16 Before an analysis o f the likely financing gap can be undertakes, the investments have to be broken down into the following sectors: state budget, local budget, and non-budgetary. The last o f these includes state and private sector enterprises, where investments inthe former are not subsidized from the budget. This breakdown has been carried out based on the present institutional arrangements. It has also been carried out by allocating EU funds to the different directives according to plans prepared by the Government o f Romania (where available). Where no such plans existed, the EU funds (ISPA and Cohesion) have been allocated to the different directives inthe same proportions as was applied inthe 2004 selected countries (notably Poland). The resulting figures are given in Table 8.6 and show the required expenditures from the state government, local government and non-public sources by year. The state investment is expected to remain at around 130-150 million a year, except in 2006 and 2007, when Directive 97/43 (human health security against ionizing radiation) has to be implemented and the costs will rise considerably. Given the extraordinary nature o f this investment, some additional fbnds may be available from the EU for this purpose. Net annual local government investment spending will have to be inthe range o f 400-500 million to 2007, after which it will rise to around 700-800 million. Finally, non-budgetary expenditures will be 700-1,300 million to 2007, rising to 1.4- 1.6 billion over the next four years and then declining to 700-1000 million for the rest o f the period. 8.17 How feasible i s such an investment plan? The analysis o f this question is divided into two periods: the short term (2004-06) and the longer term (2006-15). Inthe short term the average annual net investment needs are as follows: state (133 million), local government (448 million) and non-budgetary (936 million).173The highest corresponding levels o f spending by these groups in recent years (between 1997 and 2001) for which data are available were in the For the state expenditure, investment for Directive 97/43 has been excluded on the assumption that this extraordinary item will be funded separately from external sources. 177 following range: state (20 million), local government (90 million), and non-budgetary (225 million). Table 8.6 Summary of Expenditures to Meet the Directives, by Sector, 2004-15 (Mn.) 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Total State Investment 182 169 624 2238 174 185 199 178 25 25 25 25 ofwhich EUFunded 49 40 33 40 44 44 48 51 0 0 0 0 Net State Investment 133 129 592 2198 131 141 151 127 25 25 25 25 Total Local Gov. Inv. 669 539 510 665 907 946 1007 1037 1,053 984 888 888 ofwhichEUFunded 145 118 112 147 200 209 222 228 233 217 197 197 Net Local Gov. Inv. 524 421 398 519 707 738 785 808 820 766 691 691 TotaiNon-Public Inv. 1,074 841 1,227 1,470 1,710 1,879 1,884 1,837 1,160 965 835 827 ofwhichEUFunded 178 154 154 203 289 304 319 327 177 142 139 139 Net Non-Public Inv. 896 687 1,072 1,268 1,420 1,575 1,565 1,510 983 823 696 688 TotalNet Investment 1,553 1,238 2,062 3,984 2,258 2,454 2,501 2,446 1,828 1,615 1,412 1,404 Total EUFunds 371 312 299 389 533 556 589 606 410 359 336 336 Source: See Annex 7 regardingassumptions made inthe calculations. 8.18 For the state budget, the financing gap relative to past investment levels will be on the order o f 130 million and hence some increase in budgetary allocations would be required to fill the gap. The allocation of central budgetary expenditures to "environment and waters" is around 0.4 percent o f the budget, but this has to cover current spending as well as investment. Since around half o f this allocation is for current expenditures the "baseline" allocation for 2004-06 for environmental investment will be approximately 30 million every year. This will have to be increased by 100 million if the state financing gap i s to be closed, which would make environmental spending closer to 1 percent o f government expenditures. Table 8.7 provides some o f the detailed numbers for this estimate. 8.19 Local government spending net o f the EU grants will have to be on the order o f 450 million a year for the period 2004 to 2006, which represents an increase o f 360 million over the levels inthe last decade. Most o f the investment is inthe water supply and sewerage sector (see Table 8.3), but some o f it is also needed for waste management. There are some indications that substantial increases have been made in recent years. The World Bank's upgrading o f the Bucharest water supply system is one ongoing project amounting to about 16.17 million, and the EBRDhas made loans available for water projects in 15 municipalities, worth approximately 98.3 million. Repayment o f any borrowing is the responsibility o f the companies operating the municipal services, which have provided the guarantees for the 10ans.l~~The SAMTID'75 program is beinginitiated to cover the small and medium-size municipalities. The program has a total value o f around E400 million, o f which 200 million will come from the EIB and EBRD. The rest will be raised from the PHARE ESC Fund and Romania's state budget through the National Fund. The program may take ten years to reach most ofthe targeted towns (i.e., 2003 to 2013). The pilot phase starts in 2003, where the duration o f each project i s stipulated not to `74Local authorities can provide guarantees for local investments up to 20 percent o f their revenues. They can also issue bonds andborrow commercially and a couple o f themhave done so. 17'Small andMediumTowns Infrastructure Development Program. 178 extend beyond 20 months. Contracts must be signed by the end o f November 2004 and must be disbursedby end o f the same month in2006. Table8.7 Estimatesofthe FinancingGapfor StateEnvironmentalExpenditures, 2001-06(Mn.) 2001 2002 2003 2004 2005 2006 Total Gov. Exp. 14,192 14,176 12,818 14,190 15,200 15920 Total Env. Exp 57 53 54 58 60 63 Env. Investment 2 27 27 29 30 32 Env. Inv. Gap 104 99 105 Memorandum Exc. Rate: Lei/ Euro 26027 31,255 40,054 50,954 55,030 58,882 Note: EnvironmentalInvestmenti s assumedto behalfof expenditure,based on figuresfrom 1996to -2001. Source:ROMLDB Projectionsfor Romaniaand Annual Statistics for Romania 8.20 These programs clearly increase investment by local authorities relative to the past levels, but it is unlikely that they will reach the target value o f 450 million a year for the next three years. An enhanced role for the private sector may help fill the gap in the medium term, but is unlikely to do so in such a short time Hence some change in the sequencing will be needed for the water sector investments. For the waste sector, additional funds have been mobilized through the Environmental Fund, which i s discussed inmore detail below. 8.21 Finally, there are the investmentsby the private sector and by state enterprises operating without budgetary support. In 2001 the private sector invested around 225 million for environmental protection and this has now to increase to about 885 mi1li0n.l~~ The increase o f 600 million i s very substantial by any standards. The government can mandate the investment through changes in the environmental laws and regulations, and indeed it has been doing this quite diligently. However, such regulatory changes do not guarantee that the industry will be able to make the changes required. A lack o f funds will continue to be a major constraint. To put these investments in perspective, the total private sector fixed investment in the period 2004- 2006 is estimated to be around 10 billion, or 3.3 billion a year. Of this, the environmental expenditures will take about 27 percent, which i s much higher than usual (typical values o f the environmental share are around 3-4 percent). There is no obvious answer to this difficulty except to say that the target i s ambitious, it is likely that it will not be met, and some o f the required investment will have to be postponedto later inthe accession period. 8.22 Having looked at the short term financing gap, we now turn to the medium term. For the state sector the required level o f investment remains more or less the same until 2011, after which it declines. Giventhe anticipated growth inthe economy, this suggests that the state sector will find it increasingly easy to fulfill its investment obligations vis-a-vis the environmental acquis. 176Inthe mediumto longterm, however, an enhanced role for the privatesector, through publiciprivatepartnerships ,is vital. Operating the operate on a commercialized footing can increase access to capital markets and reduce the demand for investment funds from central and local budgets. Progress inthese areas has been slow inRomania but, since we are looking over a relatively long time horizon, an increased role for non-budgetary financing should not be ruled out. See Table 8.5, average o f 2004-2006net ofnon-budgetary investment. 179 8.23 For local government, however, the picture is not quite as good. Investments will have to increase sharply in 2007/ 08, after which they will increase more or less in line with income growth until 2012 and then will decline slightly to 2015. Here, an enhanced role for the private sector will be critical if the targets are to be met. There is a concern, however about the affordability o f these investments in terms o f the implied tariff increases. While this will attenuate as growth takes place, it will remain a major factor that will have to be addressed (see Section D). 8.24 For non-budgetary investments a major increase i s also predicted from 2006 to 2011, after which the figures decline. Again, growth will help, but some external funding will also be needed inaddition, o f course ,to what the EUwill provide inthe way o f grants. C.3.The EnvironmentFund 8.25 One new source o f investment resources is the Environment Fund, established in 2002, which raises revenues from environmental fees and charges, environment authorization taxes and environment impact assessment taxes."* As o f February 2004, revenues from the fund had reached about US$29 million (24.6 million) in2004. However, no firm basis has been provided for this figure wbich may be optimistic. The intention is to use these revenues to finance the priority investments required to comply with the acquis communitaire covering, in principle, investment in local government and the private sector. Funds will be made available partly on a grant basis and partlyon a soft loan basis, depending on what financial benefits the investments generate. Inthe first year (a pilot year) the focus will be on projects involvingwaste management and recycling (covering Directives 94/62 and 99/31). At present, about 100 projects have been accepted, with a total value o f about 9 million. 8.26 As in other CEECs, the Environment Fund can play an important role in financing investments but there are also some possible pitfalls. The experience with similar funds in the CEEC and NIS countries has been mixed. While such funds be a useful source for the financing of environmental investments, there have been questions about the transparency o f the procedures and the efficiency the allocation o f the funds. They have been responsible for large levels o f investments, but the use o f state-of-the-art techniques to assess the cost effectiveness and to estimate the social benefits o f different investments is often lacking. At a less sophisticated level, for virtually no country is there comprehensive information on the extent to which different projects are subsidized. The instruments used include direct grants, soft loans, "green equity" investments, interest rate subsidies on semi-commercial loans, etc. The "subsidy equivalent" to each party is rarely calculated, which makes any cost effectiveness exercise very difficult to achieve. The Romanian Environment Fundshould do its best to avoid these problems by ensuring that the procedures for evaluating environmental projects do indeed meet the standards expected o f all public sector investments. 8.27 As established, the Environment Fundwill not be extrabudgetary but will have access to earmarked sources o f revenue. The inclusion o f the Fund inthe consolidated budget i s desirable Charges are levied o n ferrous and non-ferrous wastes, air emissions, packaging, and sales of dangerous chemicals, and timber sales. A small tobacco excise tax also goes to the Fund. The environment authorization taxes are collected by the One-stop-shop Office within the Trade Register Office (GD-573/2002). 180 from a fiscal viewpoint and, while earmarking is generally not favored, in this case it be may justified, given the urgent need for investment funds and the lack o f any other public sources for these funds. Finally, it should be noted that at around 85 million euros in 2004 the Fund can make an important contribution to the financing gap but it will byno means close that gap. Other sources will also be needed. Indeed, there i s a case for looking at other economic instruments that could generate revenues for environmental investments while at the same time providing incentives to reduce the environmental burdens that are the reason for the investments inthe first place. D.KEYISSUES INTHE ENVIRONMENTAL SECTOR 8.28 This section looks at some o f measures that can be taken to address the challenges o f meetingthe environmental component o fthe acquis. These are: the following: 1. Ways inwhich the budgetary burdencanbereducedandmore internal resources mobilized; 2. Institutional Reforms and Administrative capacity constraints that have to be addressed 3. Affordability constraints to implementation. D.l. Ways inWhichthe BudgetaryBurdenCanBeReducedandMoreInternalResources Mobilized 8.29 Based on the experience o f the other CEECs, costs can be reduced substantially by following a least cost investment plan for energy-related investments and by under taking making reforms in the energy sector, especially in energy-related investments. InRomania the Ministry o f Industry regulates this sector and the national strategy is one o f reform in the direction o f competition and privatization. Broadly speaking, the more rapid the reforms are, the lower the required investments and the environmental costs will be. 8.30 The costs o f meeting the environmental objectives will also be lower in line with the rapidity o f the rate o f implementation o f the economic based instruments to that will provide polluters with incentives to reduce emissions. Such measures, which provide fiscal incentives to polluters to reduce the volume o f wastes, can have two impacts. The first impact is to reduce the amount o f investment and expenditure in cleanup and the second impact i s to provide some resources for to enable the public sector to pay for the cleanup. InRomania, some progress has been made in this regard but there is scope for the more rapid implementation o f some instruments o f the kind used in the EU. These instruments include taxes on solid waste and wastewater that reflect the full social cost o f their disposal, as well as taxes on carbon to discourage the use o f fossil fuels.'79 8.3 1 Table 8.8 provides a description o f the main environmental market-based instruments in use inthe EU, the potential role that they play there, and the scope for their use inRomania. As 179 Article 8 of Law no. 7312000 (amended and republished) specifies a number of economic charges (see Section D on the Environment Fund), but more charges couldbe considered. 181 the table shows, there is considerable potential for the use o f such instruments, but there are also important issues o f design and application that have to be addressed, and not all environmental charges have worked well. The key areas in which charges have a serious potential and are feasible are the carbon tax, expanded hazardous waste charges, and product charges. There are issues o f design and economic impact with regard to these as well but with access to experience inthe EUand elsewhere it shouldbepossible to introduce a charge that functions effectively to reduce the relevant emissions and to provide revenues to the government that can be allocated to investments that satisfy the key EUenvironmental directives. 8.32 Finally, there are a number o f administrative measures that can be taken to increase efficiency and reduce investment costs, as discussed below 0 Costs are also sensitive to the efficiency with which municipalities take investment decisions. Opening up procurement to international tender and undertakmg careful project appraisal in evaluating the design o f the schemes will reduce costs significant1y.lg0 0 Indesigning the investments, account should be taken of the lower demand for some services when future service charges will have to recover capital and operating costs. The World Bank hzs had experience with this inthe wastewater projects that it funded in the Baltic States, where the level o f capacity is turning out to be substantially in excess of demand as a result o f the large increase in volume-based charges. A related issue i s economies o f scale and the need to take advantage o f them. The financing arrangements for these investments often militate against the adoption o f economies o f scale, as soft loans from suppliers and ISPA grants support the status quo service delivery arrangements. Direct pressure from the central government is needed to overcome this bias and to resist the loss o f political patronage that it will imply. e The impact that new investments will have on total operating costs will be to reduce those costs, as the operations and maintenance needs o f new equipment are lower than those o f the old equipment. In addition, operations costs can be reduced through increased efficiency, the scope for which is considerable, through outsourcing and through reductions inover-manning inthe state controlled power and water utilities. 0 The present value o f the costs will fall if the more expensive items are delayed. Taking this into account in the sequencing o f investment i s all the more justified when the ensuing benefits are low. As the discussion in Section C shows, a careful assessment of the benefits against the costs supports the adoption o f most o f the directions related to air pollution,, but it supports being more selective o f the directives on waste management and wastewater treatment. Finally, it should be noted that delaying investments not only reduces the present value o f the costs, but it also reduces the burden o f investments on the poor. Another factor is discrimination. Subsidies available to public utilities are not given to private operators. Numerous experts and commercial banks have reported cases where city authorities are unwilling to accept private financing because o f a loss o f subsidy for the utility if it is operated as a private entity. 1.3 I a, % 0 E .e c1 in a, sc bo k0 9 e, I rl- E J 1- n aE: 184 D.2. InstitutionalandAdministrativeReforms 8.33 The EU 2003 Report on Romania's progress toward accession has noted that there are still major gaps in the implementation o f the acquis program generally, including in the area o f the environment. The areas inwhich the situation is still challenging are listedbelow. (a) Progress is needed in integrating the environment into other policies. The medium to long term energy strategy, for example, has been prepared but little attention has been given to the environmental dimension. In addition, the capacity for carrying out strategic environmental analysis i s weak andneeds to be built up. (b) N o developments havebeenmade inthe areao f environmental impact legislation, which is a key requirement under EUlaw. (c) Considerable work is needed on industrial pollution control and risk management. While legislation has been adopted, enforcement will need more staff than have been allocated to the National Environmental Guard. (d) N oprogress hasbeen made inthe field o fnoise. (e) The Ministry o f Environment and Water Protection has expanded its administrative capacity through the creation o f the National Environmental Agency (NEPA), eight Regional EPAs, and 34 local EPAs. However, the Ministry remains understaffed and the situation has been deteriorating since 2001. (f) Although the budget for environmental protection has been increased slightly, it is still low. 8.34 The Bank's own experience supports these findings. However, the World Bank has expressed its growing concem that in the rush to accession Romania is passing laws and establishing a regulatory framework inthe virtual absence o f effective consultation, and without either the capacity or the resources to carry them out. While the body o f legislation has grown considerably, staff levels have not increased and working conditions remain poor. Romania urgently needs to build up the capacity o f the local Environmental Protection Inspectorates (EPIs) in order to improve permitting, monitoring, inspection, and enforcement. Although they are progressing intheir environmental management capacity, the key ministries are not up to the levels o f the present accession group o f countries. 8.35 In view of this shared opinion, Romania should give higher priority to institutional strengthening in the environment sector. Inevitably, this will require that an increased budget be allocated to the Ministry. But this is essential if the objectives o f improved consultation, more effective enforcement, andbetter coordination o f environmental policies with the other ministries are to be achieved. E.CONCLUSIONS 8.36 In comparison with new EU members and candidate countries, Romania is starting out with a lower level o f environmental capital and a poorer state o f the environment in many dimensions. This is demonstrated by indicators such as the country's low rate o f access to improved water sources and sanitation, its highmortality rate for children under the age o f five, 185 and its low energy efficiency. Hence, the country will need to devote more resources and will need to receive more assistance to fulfill the environmental requirements necessary for accession to the EU. 8.37 The total investment cost required to comply with 17 directives that address environment- related concerns amounts to approximately 29.7 billion. This figure is about 29 percent higher than originally estimated by the EC and it makes Romania one o f the highest in this regard among the new EUmembers and candidate countries. The EU funds committed to help finance the investments during the period 2004 to 2015 are estimated to be about 6.9 billion intotal. Hence, the remaining 22.8 billion will need to be obtained from other sources, namely, from central and local government and the private sector. 8.38 The estimated investments by the central and local governments imply a major increase from the present levels. An estimate made o f the investment needs by year over the period 2004- 15 shows that the central government will have to allocate more for the environment and that local governments will have to mobilize more funds than they have at present if the schedule is to be met. With realistic expectations o f what can be achieved, however, one has to conclude that it is unlikely that the central and local governments will be able to raise their expected shares o f the total, at least in the next five years. On a more optimistic note, the picture looks better after 2007, especially for the central government. For local government the medium term will continue to be challenging, and an enhanced role for the private sector will be critical if the targets are to be met. Specific constraints on the environmental side for private sector participation inthe supply o f local public services have been noted, and some suggestions, have been offered for dealing with them. 8.39 A major demand is also being placed on the private sector as a result o f the acquis, and it is by no means clear that the levels o f investment required can be met. External resources will be vital to the success o f the program here, and measures that would ease the path should continue to be pursuedactively by the government. 8.40 The government i s seeking to mobilize more funds for the environment by using economic instruments such as charges and allocating them through an Environment Fund. If Fund adheres to principles o f sound finance and expenditure control, this measure is to be welcomed. These environmental charges serve not only to raise revenues but also to reduce, in the first place, the emissions that are the source o f the investment needs. Examples o f potential instruments that can be adopted in Romania, in addition to those already introduced, are the carbon tax andmore product charges, andthese instruments should be investigated further. 8.41 Examples o f additional environmental benefits that can be, expected from the environmental investments are the reduction o f risks to health and the protection o f the ecosystem. Estimates o f these benefits have been reported for Romania and other countries. They show that Romania will derive the greatest benefits from air pollution reduction, followed by water and waste measures. However, there is considerable uncertainty about the magnitude o f these benefits, and more information is needed, especially at the local level, so that the data can be used in sequencing the investments. Inparticular, an effort needs to be made to have a better 186 picture o f the benefits o f the drinking water and solid waste directives, and to select those investments that will generate the greatest benefits first. 8.42 Finally the successful execution o f the environmental directives does not depend solely on the availability o f funds. An important need i s to strengthen the institutional capacity to implement the directives. Perhaps even more than inother countries, there is a need inRomania to further integrate environmental policies into it's the country's overall framework. Judicial, legislative, and data collection institutions must complement the environmental regulatory institutions as the enforcement and monitoring o f the environmental regulations will need to rely on the strength o f such complementarybodies. ANNEX 1:FISCAL SUSTAINABILITYANALYSIS Fiscal sustainability measures the government's ability to service its debt while maintaining macroeconomic stability.lgl Fiscal sustainability analysis is centered on the concept o f the government's lifetime budget constraint. It determines the size o f the primary surplus needed to sustain the initial debt level given assumptions about output growth, interest rates and inflation. The primary balance needed to keep the initial debt to GDP ratio constant is calculated using the following equation, assuming a constant GDP growth, constant primary balance and seigniorage inpercent o f GDP: - where r is the real interest rate on government debt, calculated using assumptions about GDP growth g and nominal interest rates r: - r =(r-g)/(1+g) (2) - - The initial debt to GDP ratio i s bt-1,and 0 is the real value o f seigniorage revenue as a percent o f GDP. The seigniorage is defined as the net revenue from printing money and is measured as the change inbase money. Real interest rate is computed from F(R- n)/(1+n) (3)7 where n represents the average inflation rate. Assuming a steady state inflation rate o f 3.5 percent in the medium term, an average output growth o f 4 per-annum, and an average cost o f borrowing o f about 10percent, we estimate the real interest rate r =2.2 percent. Seigniorage revenue as a percent o f GDP is calculated based on the following equation: - where i s base money as a percent o f GDP, which i s estimated at 4.8 percent o f GDP at end 2002. Using the assumptions about inflation and output growth, we estimate a seigniorage revenue o f about 0.3 percent o f GDP. "' Bumside, A. Craig, (2002) Fiscal Sustainability in Bulgaria, WorldBank.Processed. 188 - Using equation (1) with initial debt ( bt-1 ) equal to 28 percent of GDP at end-2002, we estimate that the primary surplus needed to sustain the initial debt level of 28 percent o f GDP i s 0.3 percent of GDP. Inthe interpretation of results, some data constraints andpossible measurement issues should be taken into account. Government debt here refers to gross public and publicly guaranteed debt while usually net government debt is used to estimate the sustainable fiscal deficits. Inaddition, this stock of debt includes temporary borrowing from the treasury resources, which have not been netted out in the total public debt. The government budget deficit i s calculated on a cash basis and does not take into account the accumulation of tax arrears. While the stock of debt includes both public and publicly guaranteed debt, interest payments in the budgetcover only the cost of servicing government debt, andthe cost of activated guarantees. 189 ANNEX 2: MODELINGOFAGRICULTURALPOLICY A BriefDescriptionof Simulation Methodology A partial equilibrium model o f the Romanian agricultural sector has been used to simulate a series o f scenarios on the adoption o f EU-type agricultural policies and to provide information about the likely impact on producers' value added (farmers' income), consumers' real income, poverty, and the state budget. Ten major agricultural products are considered inthe model, namely, wheat, barley, maize, sunflower, potatoes, milk, beef, pork, poultry, and eggs (-70 percent o f Romania's gross agricultural production). For each o f these activities the simulation approach identifies and quantifies the potential impact of the changes in protection under various scenarios (see Box 1). The analysis developed inthis study assumes that Romania is a "price taker" in world markets for the products analyzed. Wheat, barley, maize, sunflower seed, and potatoes are considered as exportable, while the other commodities are considered as importable. Border price equivalents are calculated as unit values o f imported or exported commodities, depending on Romania's trade position inthe commodity discussed. Inbrief, farmers' incomeis determinedas the difference between revenues from the sale o f agricultural product and tradable input costs, plus any non-price-related direct monetary transfers to farmers. Any change in agricultural policies affecting input prices, output prices, or direct monetary transfers (such as the CAP direct payments) translates into changes inthe value added o f agricultural production @e., farmers' income). Inthe medium to long term, the level o f production o f the various commodities i s responsive to the level o f profitability o f each commodity (approximated by value added). Changes in value added thus lead to changes in the total output, further compounding the initial changes in farmers' income. For each commodity, the extent o f the medium- to long-term supply response is determined by the elasticities o f supply, which in this study were consistent with the elasticities used in similar World Bank studies for other countries inthe regions.'82 Changes in agricultural output prices affect consumers via their impact on food prices. Higher food prices lead to an increase inhousehold food expenditures and, given a fixed nominal disposable income, the consumer i s forced to reduce the consumption o f non-food goods and services by the percentage increase in food expenditure multiplied by the share o f food expenditure intotal consumption. This reduction inthe consumer's ability to purchase goods and services i s equivalent to a reduction in real income. In the medium to long term, consumers adjust to the new set o f relative prices, moving away from the consumption o f foods that have See, e.g., Csaki, C., Nash, J., Matusevich, V., and Kray, H.A.(2002), Food andAgricultura1Policy in Russia: Progress to Date and the Road Forward, World Bank Technical Paper No. 523; and Csaki, C., Nash, J., Fock, A., and Kray, H.A.(2000), Food and Agriculture in Bulgaria - The Challenge of Preparingfor EUAccession, World Bank Technical PaperNo. 481. 190 become relatively more expensive as a result o f the policy changes. The medium- to long-term impact on income i s therefore expected to be more moderate compared to the short-run impact. The exact magnitude i s determined by the price elasticity o f demand for each product, which regulates the extent o f consumers' adjustments to the changes in food prices. To allow for the various effects, two types o f simulations are carried out for each scenario: one simulation with no supply and consumption changes (static short-run impact) and a second simulation allowing for a supply and consumptionresponse (medium- to long-run impact). The impact o f the agricultural policy scenarios on poverty in Romania has also been analyzed. The share o f the Romanian population classified as poor was determined based on the Total Poverty Headcount Ratio and the Extreme Poverty Headcount Ratio derived from a separate World Bank study on poverty inRomania.'83 For the analyses o f the potential impacts o f the CAP-introduction scenarios we used the Extreme Poverty Headcount Ratio as an appropriate indicator o f poverty in Romania. Based on these assumptions, we calculate the changes in the number o f poor resulting from the introduction o f the various EU integration scenarios. The trade-related effects o f the policy changes on tariff revenues and export refunds are estimated based on the premise that these are required to "protect" domestic price levels under CAP-type policies. In our computations o f trade-related effects we do not treat Romania as a non-member country implementing CAP-type policies on its own. We thus simulate the trade- related budgetary implications if Romania adopts CAP-type agricultural policies before accession. The calculated revenues from trade flows include trade with EU member countries. Not only would the revenues from tariffs and the expenditures for export subsidies be lower if the trade flows within the enlarged EU were subtracted, but, most important, they would be financed by funds from the CAP. Under our simulation, Romania does not receive any compensation from the EUbudget for introducing CAP-type agricultural policies. Romania's potential gains from EU Structural and Cohesion Funds (SCF) were estimated on the basis o f average current per capita payments to EU member states. Assuming that Romania falls under Objective 1,per capita payments would amount to 217 per inhabitant per annum (EUCommission, 2002). Therefore, before imposing the maximum limit o f4 percent of GDP on SCF payments, Romaniawould qualify for annual transfers o f 152 trillion ROL (4.8 billion), namely, about 13 percent o f the 2001 GDP. Romania's potential gains from SCF have been estimated on the basis o f the payments granted to Greece, Ireland, Portugal, and Spain. SCF payments are granted to countries in which per capita GNP is less than 90 percent o f the EU average. The total amount to be spent in 2002 i s about 45 per inhabitant. These estimates are fairly optimistic as they assume that Romania (and other new entrants) would be supported under the same conditions that apply to current EUmembers. The consumption levels correspondingto these poverty lines are 1,535,570 ROL and 1,060,658 ROLper capita, respectively; 28.9 percent o fthe Romanianpopulationfall below the first, higher poverty line and 10.9 percent fall below the second, lower poverty line inthe year 2002 (World Bank 2003: Romania: Poverty Assessment, Vol. 2, p. 9f). 191 ANNEX 3: IMPACT OF CHANGESINAGRICULTURAL POLICY I.Absolutevalues Scenario Gross Agricultural Farmers' income''' Poverty Net budget output [% of popu- revenues lation below fromEU' [billion ROLJ [billion ROL] poverty line'? [bln ROL] Timehorizon' SR LR SR LR SR LR A: Baserunwith current policies 168,938 168,938 65,676 65,676 10.9 10.9 0 B1:Agenda 2000 without direct payments" 171,501 154,259 64,181 58,541 13.7 12.7 35,856 B1:Copenhagen with 100%dir.payments" 211,334 220,626 104,013 114,611 13.7 12.7 75,688 B3:Copenhagen with 25% direct payments" 182,390 170,064 75,069 71,772 13.7 12.7 46,744 B4:Copenhagenwith SIPS (25% dir.paym.)" 178,622 165,227 71,301 67,559 13.7 12.7 46,744 C1: Council Compromise with decoupled direct payments"' 162,134 143,509 54,813 48,772 12.5 12.1 79,410 C2:C. Compromise with 100%dir. payments " 205,687 214,980 98,366 108,965 12.5 12.1 79,410 D:Romaniannegotiation position 213,524 243,724 110,262 130,638 13.7 12.7 81,363 E:Non-intervention scenario 160,520 153,301 65,569 64,773 10.5 10.6 0 11.Percentage changes relative to ScenarioA Scenario Gross Agricul- Farmers' Household food Household tural Output income expenditure income Timehorizon' SR LR SR LR SR LR SR LR Bl: Agenda 2000 without direct payments'' 1.5 -8.7 -2.3 -10.9 7.1 4.5 -3.9 -2.5 B2:Copenhagen with 100%dir.payments" 25.1 30.6 58.4 74.5 7.1 4.5 -3.9 -2.5 B3:Copenhagenwith 25% direct payments" 8.0 0.7 14.3 9.3 7.1 4.5 -3.9 -2.5 B4:Copenhagenwith SIPS (25% dir.paym.)" 5.7 -2.2 8.6 2.9 7.1 4.5 -3.9 -2.5 C,: Council Compromise with decoupled direct payments"' vi -4.0 -15.1 -16.5 -25.7 4.0 2.8 -2.2 -1.6 C2:C. Compromise with 100% dir. payments I' 21.8 27.3 49.8 65.9 4.0 2.8 -2.2 -1.6 D:Romaniannegotiationposition 26.4 44.3 67.9 98.9 7.1 4.5 -3.9 -2.5 E:Nonintervention scenario -5.0 -9.3 -0.2 -1.4 -1.7 -1.2 1.0 0.7 Notes: i SR = ShortRuneffect, LR = LongRuneffect ii Thesearethedirectpaymentsgrantedto EUfarmersascompensationfor thepricecutsintroducedundertheCAPreformsin 1992 and 1999. iiiIncomeeffectsderivingfrommarketpricesupportandincomeeffectsderivingfromdirect(compensation)paymentstofarmersare consideredjointly. iv Referring to the Extreme Poverty Headcount Line of 1,060,658 ROL according to World Bank (2003): Romania: Poverty Assessment, Vol. 2, p. 9f. v Net revenue is calculatedas Total Receiptsless Total Contributionsfrom the EUBudget, assumingthe application of the restriction on maximum annual receipts from EU Structuraland Cohesion Funds: Currently, maximum annual transfers from EU Structural and CohesionFundsto any member country shouldnot exceed 4 percent of national GDP. vi Our simulationscover activity-based,i.e. commodity-specific,direct payments only. Inthe event of adecouplingof direct payments, these are granted not to individual commodities but to the farm as a whole. Therefore (dueto the nature of the analyticalapproach) the results displayed above are correctly indicating the value-added as proxy for the income generated from the analyzed activities, but underestimatethe total farm income. While these decoupled paymentshave not been included in the farm income calculations, the net revenue indicated above includes the direct payments granted to the sector as a whole (net transfer without direct payments: ROL 35,856 billion). Source: Own calculations. 192 ANNEX 4: EARNINGSFUNCTIONFORMALESAND FEMALES IN2000 Earnings Function for Males and Females in 2000 2000 Male Female Male-Female Coef. t-value Coef. t-value D-coef t-value Experience 0.0274 11.85 0.0276 12.87 -0.0002 -0.08 Exp. squared -0.0005 -8.84 -0.0004 -7.99 0.0000 -0.34 No. o f years o f schooling 0.0469 14.49 0.0409 11.52 0.0061 1.26 Married 0.1106 7.05 -0.0251 -2.08 0.1357 6.85 Industry:Manufacturing is the omitted category Farming -0.1438 -4.56 -0.0750 -1.98 -0.0688 -1.40 Extractive industry 0.4113 12.04 0.2264 4.51 0.1850 3.05 Electricity, Water, Gas RA 0.1796 7.08 0.1721 4.92 0.0075 0.17 Construction -0.0357 -1.78 -0.0654 -1.74 0.0298 0.70 Trade, Hotel and restaurants -0.0998 -4.37 -0.1128 -5.5 0.0129 0.42 Transport, Storage, Communications 0.0201 1.07 0.1325 4.94 -0.1123 -3.43 Finance, Banking, Insurance 0.0410 0.82 0.1841 5.71 -0.1430 -2.40 Real Estate -0.0526 -0.97 0.0156 0.28 -0.0682 -0.88 Public administration, Defense 0.0585 2.2 -0.0770 -2.6 0.1355 3.40 Education -0.2632 -8.33 -0.1805 -7.54 -0.0826 -2.08 Health and social assistance -0.1488 -4.12 -0.0764 -3.33 -0.0724 -1.69 Other social andpersonal services -0.1404 -5.87 -0.1522 -6.16 0.0118 0.34 Household activities -0.3789 -1.55 -0.1316 -0.88 -0.2473 -0.86 International organizations -0.1136 -0.27 0.1840 0.89 -0.2976 -0.63 Occupation: Skilled workers i s the omitted category Politicians, chief and senior executives -0.2498 -3.51 o f institutions 0.5263 13.1 0.7761 13.21 Experts and intellectual activities 0.2665 10.26 0.3876 13.48 -0.1211 -3.12 Technicians 0.1299 6.13 0.2113 9.63 -0.0813 -2.67 Office workers 0.0435 1.39 0.1274 6.06 -0.0839 -2.22 Operational workers inservices and 0.0168 0.52 trade -0.0881 -3.6 -0.1050 -4.88 Farmers -0.0998 -1.01 -0.1956 -2.04 0.0958 0.70 Plant and machine operators 0.0474 2.82 0.1288 5.42 -0.0814 -2.80 Unskilledworkers -0.2659 -10.6 -0.2094 -9.4 -0.0565 -1.68 Armed forces 0.4631 11.88 0.4468 3.98 0.0163 0.14 Ownership o fthe firm: state firmis the omitted category Private firm -0.0600 -4.4 -0.0275 -1.91 -0.0325 -1.64 193 2000 Male Female Male-Female Mixedfirm 0.0250 1.36 0.0270 1.34 -0.0019 -0.07 coop. -0.3852 -6.13 -0.2036 -5.12 -0.1816 -2.44 Public o f nationaland local interest -0.0013 -0.05 0.0363 1.59 -0.0376 -1.10 Other 0.0048 0.04 -0.0916 -0.76 0.0965 0.56 Nationality: Romanian is the omitted category Hungarian -0.0449 -1.93 0.0262 1.24 -0.0711 -2.26 Gipsy 0.0593 0.72 0.0207 0.26 0.0386 0.34 German 0.3149 2.9 0.1267 1.47 0.1881 1.36 Other 0.0224 0.38 -0.0436 -0.54 0.0659 0.66 Population o fthe locality: between 15 and 50 thou. is the omitted category Population less than 5 thou. -0.1135 -1.06 -0.0208 -0.23 -0.0927 -0.66 Populationbw. 5 and 15 thou. 0.0202 1.01 -0,0201 -1.01 0.0403 1.43 Pop b w 50 and 150 thou. 0.0613 3.76 0.0688 4.39 -0.0075 -0.33 Pop. bw. 150 and 500 thou. 0.1202 7.67 0.0810 5.35 0.0392 1.80 Pop. above 500 thou. 0.0319 0.26 -0.0037 -0.02 0.0357 0.16 Regions: North East i s omitted South East 0.0918 4.45 0.0163 0.83 0.0756 2.66 South East 0.0171 0.84 -0.0304 -1.55 0.0475 1.68 SouthWest 0.0913 4.41 0.0038 0.19 0.0875 3.03 West 0.0702 3.2 0.0120 0.58 0.0582 1.93 NorthWest 0.0348 1.66 0.0152 0.77 0.0196 0.68 Center 0.0239 1.12 -0.0081 -0.41 0.0320 1.10 Bucharest 0.2191 1.79 0.1781 1.1 0.0410 0.20 Const. 13.3295 275 13.2287 262.2 0.1008 1.44 No. of observations 6,540 6,214 R-squared 0.4117 0.4326 Source: Staff computations using Integrated Household Survey data. 194 ANNEX 5: NON-BANKING FINANCIALSERVICES NON-BANK FINANCIAL SECTOR ISSUES1s4 1. CapitalMarkets The capital market in Romania is one of the smallest in Eastern Europe, with an estimated market capitalization in 2002 of around 11 percent o f GDP, half the average in EU accession countries (Figure 6.4). Capital market diversification i s limited to a relatively small range o f instruments, consisting of stocks listed on the two stock exchanges, BSE and RASDAQ. In2002, the market capitalization on both exchanges was US$4.5 billion, out of which the BSE accounted for US$2.7 billion. The daily trading volume o f both exchanges was US$330 million over 2002, out of which US$213 million was on the BSE. The fixed income market mainly consists o f government securities (Treasury bills and bonds) and a limited number of municipal bonds. The corporate bond market is very small. In April 2003, the outstanding value of government securities was about 56,000 billion ROL (denominated in ROL), and US$424 million (denominated in US$), with daily traded volumes of 1,300 billion ROL and US$4 million, respectively. Stock market activity has been highly Figure4. Stock MarketCapitalization (2002) (% of GDP) concentrated in a few equities. The top 10 listed companies have accounted for over 85 -"'" 1 1 percent o f the market capitalization, 70% compared to an average of 60 percent in EU 60% accession countries. On the BSE, the most 50% 40% active exchange, over 90 percent of the 30% trading volume has consisted o f the 5 closed- 20% end investment funds (SIFs), a few large 10% banks and companies. Notwithstanding 0% growth in market capitalization on the Bulgaria Romania Poland Czech Slovakia Hungaq Russia EU Area Republic Alerage Bucharest Stock Exchange (BSE) o f about Source: EstimatedGDP and IFC EmergingMarket Database This section is extracted from the Joint Bank-Fund Financial Sector Assessment Program (FSAP). See Romania: Financial System Stabilitv Assessment. Including Reuorts on the Observance of Standards and Codes on the following touics: Banking Supervision, Securities Regulation, Anti-Monev Laundering, and Combating, the Financing o f Terrorism, Monetaw and Financial Policy Transuarencv, Country Report No. 031389, December 16, 2003. 195 120 percent in 2002 and on the RASDAQ o f about 70 percent, activity on these two markets remains insufficient to assure their viability. Exploratory discussions are under way about their possible merger. Compliance with EU directives could reduce the number o f traded securities even further, as a result o f high minimal capital requirements for listed companies and a requirement that all listed companies have a public "float" o f at least 25 percent o f outstanding shares. Capital markets in Romania are disproportionately small, in good part as a consequence o f lack o f investor confidence caused by, inter alia, the elements that follow. * There was the collapse o f two large open-ended money market mutual funds- SAFI (1996) and FNI (2000). First, in 1996, SAFI, an ~ p e n - e n d e d 'money ~ ~ market mutual fund failed after the C N V M required it to re-value the net asset value o f its portfolio downward according to a new valuation methodology. The re-evaluation triggered a crisis ininvestors' confidence, causing a runon the fund. Second, in 2000, the National Investment Fund (FNI), which was an open-ended fund with over 300,000 investors and US$l50 million assets, also suffered from massive redemption demand due to doubts about the overvaluation o f the assets in the fund. The capital market has been subject to frequent legal and regulatory changes, adding to the uncertainty. In 2000 alone, there were five emergency ordinances issued affecting the balance among shareholders' interest. Following the 2000 collapse o f FNI, a number of reforms were initiated to strengthen the regulatory and supervisory environment and to strengthen the transparency o f the capital market.ls6In2002, new securities laws were enacted regardingthe structure o f the C N V M (Law N o 514/2002); securities, financial investment services and regulatedmarkets (Law 525/2002): collective investment intransferable securities (Law 513/2002), and commodities markets and derivative financial instruments (Law 512/2002). However, inan effort to comply with EUdirectives, C N V M has initiated another legal drafting process, to combine these new laws into a single securities law. There is insufficient protection o f investors who do not have managerial control o f the corporations that issue securities (outside investors). Romania still suffers from important forms o f abuse o f minority shareholders' rights including dilution o f minority shareholders' rights through capital increases without prior revaluation o f existing capital, abusive allocation o fprofits (including transferring The form and structure o f open-ended funds is that of a type generally known as "contractual," that is to say that the management o f the pool o f assets is covered by a civil law contract between the management company and the investor. The legal form of closed-ended funds is that o f a joint stock company and therefore its directors are bound by the duties imposed by thejoint stock company legislation and its shareholders have the same rights as shareholders in any other joint stock company. A "code o f conduct" for listed companies in accordance with OECD was introduced to improve transparency and corporate governance. Some o f the listed companies are in the process o f adopting this code to become eligible for the Transparency Plus tier listing on the stock exchange. 196 profits outside the company), delays in the payment o f dividends, fraudulent transfers o f company's assets to thirdparties, and limitedaccess to information.'*' e Corporate governance i s weakened by the continued govemment shareholding in privatized companies. The state privatization agency (APAPS) holds shares inan estimated 90 percent o f the companies traded on the BSE and the over the counter RASDAQ e Financial reporting i s weak. While IAS will soon be required for all publicly traded companies (and all financial institutions), the Ministry o f Public Finance has issued annual instructions which for 2002 require that annual financial statements be prepared in accordance with IAS except for IAS-29, Financial Reporting in Hyperinflationary Economies. Non-application o f IAS-29 may generate corporate governance abuse when in-kind contributions (using current market values) are used to fund increases in share capital. The result is that existing shareholders may see their interests unfairly diluted. e While supervision o f capital markets has improved, C N V M should be further strengthened in this respect. CNVM still lacks sufficient resources and trained staff to adequately monitor activities and common business practices inthe capital markets. 2. Insurance The insurance market plays an insignificant role within the Romanian economy, as both total gross premium income (GPI) and assets in the industry represent less than 1 percent o f GDP. GPIas a share o f GDP inCEE countries varies from 3 to 5 percent andthe average for EU countries is 8 percent. Important progress has been made in modernizing the regulatory framework o f the insurance sector, but critical areas such as corporate governance, internal controls, reinsurance, and prudential rules on assets and liabilities need further strengthening. The Insurance Supervisory Commission (ISC) was recently established (2000), but it still lacks a formal and transparent supervisory process and an adequate capacity in critical areas o f supervision. Aware that further consolidation o f the market is required, the ISC plans to achieve it through increased minimumcapital requirements, the implementation o f newly introduced solvency rules, and the further strengthening o fprudential rules on asset and liability. While the insurance market is small, a rapid growth o f about 40 percent in 2002 and an expected growth o f between 30 and 40 percent in 2003 suggest that it has a good potential for promoting financial intermediation and facilitating economic transactions through risk transfer. "'Theabusesof shareholders' rights are comprehensively described, besides other corporate governance issues, in the OECD report on "Corporate Governance in Romania" o f 2001. 197 3. Pensions The Romanian public pension system consists exclusively o f one unhnded pillar, which suffers from major structural weaknesses. These include: (i)a very high contribution rate (35 percent o f nominal eamings out o f a total payroll tax o f 58 percent) that stimulates evasion, narrows the contribution base, and causes economic inefficiency, as workers switch to the informal labor market which operates with less capital and has lower productivity; (ii) a very highdependency ratio'88as a result of evasion, high unemployment, and low retirement ages; (iii)the aging ofthe Romanianpopulation(a significant decline ofthe total population is projected over the medium term), which will further worsen this ratio; and (iv) a low gross replacement rate @ension over wage) o f 36 percent (2001), despite highcontributions. Systemic pension reform is essential, as the current system will not be financially sustainable in the long run, given its very high contribution rate and the widespread evasion. Reforms should include a multi-pillar pension system with funded and un-hnded components, public and private sector management, and effective regulation and supervision. Fundedpension systems are more conducive to growth, as an,increase innational savings, properly intermediated through the financial sector, may lead to more capital accumulation. As a matter o f fact, a privately funded pillar for the pension system could strongly support the development of local financial markets. However, there should be a regulatory and supervisory framework ensuring that these schemes do not assume more risk than they can effectively manage. 4. FinancialInvestmentFunds The privatization investment funds (SIFs) were created from the coupon privatization programs o f the 1990s. The five SIFs are the largest institutional investors in Romania and their portfolios include significant holdings in the privatized banks, including BRD. The SIFs were exempted from the 2002 law on collective investment funds and thus operate under the special purpose 1996 legislation approved for their creation and the prior 1993 investment law. The SIFs are therefore exempt from the substantial governance provisions o f the 2002 investment funds law, and there is little transparency in their portfolio o f assets. To simplify the securities legislation, the authorities may choose to include SIFs under the provisions o f the investment funds legislation, with additional regulations as needed to meet the specific requirements o f SIFs and do allow for an orderly transition to the new law on collective investment funds. 5. Credit Unionsand Credit Cooperatives The total volume o f savings in the credit union system is.small, at approximately US$170 million, which i s not o f systemic significance, although the soundness o f these entities is important for public confidence in financial institutions in general and for market development. Credit union assets are limited to loans to members up to a level related to the savings of the borrower and government securities. Credit unions (ECUs) are not technically credit institutions, as they do not take deposits from the general public. However, membership is open to all who have permanent employment.' 89 The Ministryo f Public Finance has overall responsibility for the credit union system. This is the ratio ofbeneficiaries over contributors. In practice, ECUs may be,close to collecting deposits from the general public and thus close to meeting the definition o f a bank, as the law permits membership by anyone with an employment contract. Romania had a 198 Credit unions are subject to supervision exercised by the National Union o f Credit Unions, which, in tum, reports to the Ministry o f Public Finance. The National Union has recently expanded its supervisory staff and has made organizational changes to support better supervision. The legal framework for credit unions i s limited, not sufficiently robust to handle a potential expansion o f this type o f intermediary, and permits incorporation with negligible capital and by a few individuals and companies. It i s not clear what the regulatory situation o f ECUs would be should they decide to secede from the National Union. Therefore, more restrictive regulations may be needed. The Ministry o f Public Finance should have clear authority to intervene inthe management and operations o f credit unions as needed. The main four networks o f credit cooperatives have combined assets o f US$96 million, which i s not o f systemic significance. However, as with credit unions, they are important for confidence inthe system and for market development. Credit cooperatives take deposits from the general public and are beginning to be supervised by the NBR, although a form o f delegated supervision through the Association o f Credit Cooperatives is applied. Thus, there is a need for monitoring o f the Association's performance by supervisors. The NBR should retain ultimate regulatory and supervisory powers with respect to credit cooperatives. Licensed credit cooperatives are covered bythe Deposit Guarantee Fund. 6. HousingFinance Housing finance is negligible relative to the size o f the economy. By the end o f 2002, the outstanding reported portfolio was equivalent to US$lOO million, or less than 2 percent o f total private credits or 0.22 percent o f GDP. Despite the recent fast growth of mortgage lending, it remains low compared to intemational benchmarks (35 percent o f GDP in the EU, and between 4 percent and 7 percent inmost other EUaccession countries). So far, mortgage loans have been conservatively underwritten and have performed well, although many households conclude foreign exchange denominated mortgages due to affordable initial repayments and easier refinancing. This could become a major factor o f credit risk for middle-income borrowers incase o f devaluation vis-a-vis the euro. The size o fthe mortgage market i s limitedby the incomplete legal framework and lack o f long maturity mortgage securities. Currently, banks are dependent on long-term external lines. In that way banks cannot transform much o f their short-term resources or finance high levels o f volatile adjustable rate mortgages. The legal framework o f mortgage loans is incomplete and unclear, with several layers o f laws and ordinances. Harmonization is needed across different lenders. Priority should be given to the introduction o f rules on (i)prepayment fees; (ii) acceptable indexes to adjust lei loans; (iii) common on-line NBR reporting standards applied to banks and mortgage companies; (iv) comparative summarized information to borrowers; and (v) consistent prudential rules for mortgagecredit insurance. negative experience when L a w 10911996 allowed the creation o f Banca Populara, which engaged in financial intermediation activities outside the regulatory authority of the NBR. The result was the failure of numerous comparatively large intermediaries o f the type in 1999. 199 ANNEX 6: THEATHENS MEMORANDUM Inrecognitionofpotentialgains from increased trade, andaspart of awider movement to deeper regional integration, the governments o f Albania, Bosnia and Herzegovina, Bulgaria, Croatia, the Former Yugoslav Republic o f Macedonia, Greece, Kosovo, Romania, Serbia and Montenegro, and Turkey signed, in November 2002, the "Athens Memorandum - 2002" and agreed to develop a South East Europe Regional Electricity Market (SEE REM). Under the Athens Memorandum, participating countries have made a commitment to undertake the following steps: Policy 0 Energy strategy to be adopted. Effective tari$s and affordability 0 Power tariffs andpayments discipline to be such that effective tariffs cover costs. 0 Social safety net for the power sector to be inplace. Regulation 0 Independent regulator to be set up. 0 Gridcodes to be adopted andimplemented. 0 Transmission tariff methodologies to be adopted. Industry commercialization and restructuring 0 Increased utility transparency to be achieved through application o f International Accounting Standards (IAS) and best practice on corruption abatement as advised by areputable intemational bodyto be adopted. 0 Transmission system operator to be set up. This entity should be independent at least in terms o f its legal form, organization and decision making from other activities not relating to transmission. 0 Distribution system operator(s) to be set up. This entity should be independent at least in terms o f its legal form, organization and decision making from other activities not relating to distribution. 200 Market development Information exchange betweennational dispatch centers to commence, moving to tele- information exchange among dispatch centers. e Commercial codes to be developed e Open network access and liberalization o fthe market for non-household consumers to take place ina phased manner from 2005. Regarding inter-regional trade, the Memorandum states that the energy market in SEE will be integrated into the European Union's internal energy market. The 2003 revision o f the Athens Memorandum also includes provisions relating to gas market development, and proposes the establishment o f a South East EuropeRegional Energy Market. Requirements for the gas sector largely mirror those for the power sector and include: the setting up o f an independent gas regulator; the unbundling o f different gas industry functions with legal separation o f transmission and distribution from other industry functions; open access to networks and storage facilities; the opening o f the gas market in a phased manner from 2005, defining eligible customers at that time to include power generators, and ensuring that,20 percent o f the market is liberalized and that there is separate ownership o f transmission and distribution. For countries in SEE where gas penetration is currently less than 10 percent, a gas expansion planto raise the use o f gas above this threshold before 2010 mustbe adopted and implemented. Since signing the Athens Memorandum, the Government of Romania has developed a Road Map for the Energy Sector, adopted inJuly 2003, mapping out steps to align the Romanian energy industry with the EU acquis in the energy field. The Road Map covers energy demand and supply, investment requirements, and the energy sector institutional framework (e.g., regulation, market design). Without attempting a full summary, key aspects o f the Road Map highlyrelevant to the present chapter are given as follows: e The Government states as a main objective accession to the regional energy market and the wider European market. e The Government recognizes the need to align energy tariffs with economic costs and to maintainpayments discipline. e The power sector will be liberalized on the basis o f a bilateral contracts model with a day ahead market.lgO The World Bank is working with the Romanian market operator to develop and implement this framework. e The Government's intention is that investments will be delivered by the private sector. In particular, it is hoped that strategic investors will purchase assets in power generation and distribution. e Required capacity additions in the power sector are estimated by the Government to be 1300 MW between 2003 and 2005 and 4400 MW between 2006 and2010. 190Discussed indetail below inthe sectionMarket Rules. CA z n .-cm E 0 4 E .- 2 I .-N .3 g c .3 2m 9 L .L 5C 7e .L r ft u c .L 22L ccC .r t c 0 Q :c. a I r C m m o\ n 00 0 - . . --. 3 3 0 0 hl a .-x e 4- 8m r 203 AssumptionsUsedinthe Calculations 1. Regarding Annex Table 8. 2: "Figures are based on Poland's weighted expenditures for the corresponding directive, where the weight i s the ratio o f Romania's populationto Poland's population. Weight =0.58. bFigure includes costs administered by Termoelectrica and the Ministry o f Administration and Interior (MAI).Note that the MA1counterpart =155.4 million, comes from the State Budget & Environmental Funds. Termoelectrica's counterpart amounts to about 1.2 billion. This directive amends Directive 881609. `Costs for Aperchim and Petrobazi refineries = 104+36 = 140 million. Administrative costs are about 160,000. dScenario 5% - protected areas extend over 5% o f country surface; minimumvalue. eThe required investment effort for the whole energy sector from 2003 to 2015 is US$10,385 million, which was converted to euros (US$1 = 0.87). The share o f private and state-owned companies in the total investments are about 43% and 57%, respectively. Furthermore, the environment investment cost i s estimated at 10% o f the total investment efforts. These figures were derived from the "Road M a p for the Energy Sector o f Romania" prepared by the Ministry of Economy and Commerce of the Government o f Romania. These estimates should be interpretedwith caution as there i s a wide range o f possibilities. The expenditures are charged to the EC budget and are allocated as 860 m, 93 1 m and 1,002 m respectively, for years 2004,200, and 2006. 2. Regarding Table 8.4 o fthe MainText: At the time this chapter was prepared, only Directive 2001/80 had detailed information about the annual investments that were necessary. For the other directives, annual values were estimated based on available similar information from Poland. The available flows o f funds for different sectors (air, water, waste, IPPC, other) inPoland covered 17periods, and were used as weights to derive Romania's annual figures. Adjustments were made to directives whose complianceperiod ends before or beyond 17years. UsingDirective 761464with a complianceperiodof 12years, as an example: Period 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Total WATE 15860 RP 1021 832 680 840 910 910 1010 1060 1067 1110 960 960 960 960 910 860 810 Let WATERP-flow o ffunds for directivesrelatedto water quality inPoland x = 2WATER<. 1 7 j=9 Y = WATERP, C1 / j=1 Z = Y - X 204 To estimate the funds allocated for the first year of meetingthe directive, the following formula i s used: =1021 2.18 -*933 * * 8597 =241 where 933 i s the aggregate value o f investment to fulfillDirective 76/464. 3. RegardingTable 8.5 o f the Main Text: Complete data are only available for year 2001. Based on 2001 values, the shares o f the local and central governments in the total public investments are estimated to be about 82 percent and 18 percent, respectively. 4. RegardingTable 8.6 o f the Main Text: Step 1. The EU Funds (ISPA and Cohesion Funds) for Romania were estimated usingthe EU funds for Bulgaria. The European Commission prepared a document entitled "Roadmaps for Bulgaria and Romania" (COM(2002) 624 Final). The report provides the EU grants for both countries from 2004 to 2006, and states that the division o f allocation between Romania and Bulgaria i s set at 70/30. This ratio was used to estimate how much of ISPA and Cohesion Funds are available for Romania from 2004 to 2015. Step 2. To determine the directives to which the EUFunds are allocated, similar directives inPoland were examined. Based on Poland's information, the Funds are distributed to only 5 directives: 76/464, 91/676, 91/271 (Water); 99/3 1(Waste); and 2001/80 (Air). Step 3. Polanddata from Step 2 were used as weights to derive the Romanian annual figures. 205 STATISTICALANNEX 206 Table 1.1: Population by Area (mid-year) Total Urban Rural Total Urban Rural Urban Rural (Thousand o f inhabitants) (Annual % growth) (Percent of total) 1990 23,206.7 12,608.8 10,597.9 54.3 45.7 1991 23,185.1 12,552.4 10,632.7 -0.1 -0.4 0.3 54.1 45.9 Census 1992 22,810.0 12,391.8 10,418.2 -1.6 -1.3 -2.0 54.3 45.7 1992 22,789.0 12,367.4 10,421.6 -0.1 -0.2 0.0 54.3 45.7 1993 22,755.3 12,406.2 10,349.1 -0.1 0.3 -0.7 54.5 45.5 1994 22,730.6 12,427.6 10,303.0 -0.1 0.2 -0.4 54.7 45.3 1995 22,681 .O 12,457.2 10,223.8 -0.2 0.2 -0.8 54.9 45.1 1996 22,607.6 12,411.2 10,196.4 -0.3 -0.4 -0.3 54.9 45.1 1997 22,545.9 12,404.7 10,141.2 -0.3 -0.1 -0.5 55.0 45.0 1998 22,502.8 12,347.9 10,154.9 -0.2 -0.5 0.1 54.9 45.1 1999 22,458.0 12,302.7 10,155.3 -0.2 -0.4 0.0 54.8 45.2 2000 22,435.2 12,244.6 10,190.6 -0.1 -0.5 0.3 54.6 45.4 2001 22,408.4 12,243.7 10,164.6 -0.1 0.0 -0.3 54.6 45.4 2002 21,794.8 11,578.8 10,216.0 -2.7 -5.4 0.5 53.1 46.9 Census 2002 21,698.2 11,436.7 10,261.4 -0.4 -1.2 0.4 52.7 47.3 2003 a 21.772.8 11,580.3 10,192.5 -0.1 0.0 -0.2 53.2 46.8 aAs of January 1st. Note: 1990-2002: at July 1st; 2003: at January 1st. Source: Statistical Yearbook of Romania 20001and NIS data. 207 Table 1.2: Population b y Gender (mid-year) Total Male Female Total Male Female Male Female (Thousand of inhabitants) (annual % growth) (Percent of total) 1990 23,206.7 1,449.1 11,757.6 49.3 50.7 1991 23,185.1 1,435.3 11,749.8 -0.1 -0.1 -0.1 49.3 50.7 Census 1992 22,810.0 1,213.8 11,596.3 -1.6 -1.9 -1.3 49.2 50.8 1992 22,789.0 1,200.7 11,588.3 -0.1 -0.1 -0.1 49.1 50.9 1993 22,755.3 1,176.4 11,578.9 -0.1 -0.2 -0.1 49.1 50.9 1994 22,730.6 1,156.8 11,573.8 -0.1 -0.2 0.0 49.1 50.9 1995 22,681 .O 11,124.0 11,557.0 -0.2 -0.3 -0.1 49.0 51.0 1996 22,607.6 11,080.9 11,526.7 -0.3 -0.4 -0.3 49.0 51.0 1997 22,545.9 11,041.4 11,504.5 -0.3 -0.4 -0.2 49.0 51.0 1998 22,502.8 11,012.1 11,490.7 -0.2 -0.3 -0.1 48.9 51.1 1999 22,458.0 10,984.5 11,473.5 -0.2 -0.3 -0.1 48.9 51.1 2000 22,435.2 10,968.9 11,466.4 -0.1 -0.1 -0.1 48.9 51.1 2001 22,408.4 10,944.5 11,458.9 -0.1 -0.2 -0.1 48.9 51.1 2002 21,794.8 10,642.5 11,152.3 -2.7 -2.8 -2.7 48.8 51.2 Census 2002 21,698.2 10,581.4 11,116.8 -0.4 -0.6 -0.3 48.8 51.2 2003 a 21,772.8 10,627.7 11,145.1 -0.1 -0.1 -0.1 48.8 51.2 aAs of January 1st. Source: StatisticalYearbook of Romania2001;2002 and 2003: National Institutefor Statistics. 208 Table 1.3: Populationby Age Group and Gender (thousand,mid-year) Age group 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Total Total 23,206.7 23,185.1 22,789.0 22,755.3 22,730.6 22,681.0 22,607.6 12,545.9 22,502.8 22,458.0 22,4352 22,408.4 0 - 4 1.804 6 1,726.9 1,561.7 1,447 4 1,344 6 1,230 1 1,2104 1,176 8 1,1644 1,147 I 1,144.8 1,135 5 5 - 9 1,698 3 1,661 4 1,663 0 1,7127 1,748 4 1,745.6 1,660.1 1,545 4 1,432 9 1,3307 1,218 3 1,198 7 10-14 1,965 8 1,940 6 1,877 5 1,787 3 1,709.8 1,668 7 1,629 3 1,653.6 1,702 9 1,737 2 1,735 0 1,651 4 15-19 1,879 4 1,885 6 1,940 4 1,983 8 1,989.3 1,963.5 1,931 4 1,867 3 1,778 7 1,701.9 1,661 8 1,623 5 20-24 1,974. I 2,089 3 2.053 2 1,864.0 1,811 7 1,821 7 1.865.7 1,925 2 1,971 5 1,978 8 1,955.4 1,924 9 25-29 1,394.4 1,344 0 1,282 2 1,553.7 1,722 5 1,8609 1,9663 2,023 8 1,840 9 1,792 8 1,806.7 1,854 9 30-34 1,708 4 1,642 3 1,509 8 1,423 3 1,358 1 1,304 7 1,253 4 1,261 8 1,530 8 1,698 3 1,837 5 1,945 7 35-39 1,720 0 1,741 0 1,7166 1,713 9 1,703 5 1,629 5 1,559.6 1,481 3 1,397 9 1,335 0 1,285 2 1,237 9 40-44 1,437 I 1,518 1 1,558 o 1,622 0 1,6168 1,641.3 1,6599 1,674 4 1,672 6 1,663.8 1.594 8 1,530 5 45-49 1,197 4 1,1969 1.1844 1,208 I 1,280.0 1,3795 1,4534 1,507 4 1,569 6 1,565 4 1,592.4 1,613 7 50-54 1,440 8 1,376 5 1.311 4 1,243 9 1,1920 1,127.6 1,1274 1,131 2 1,1542 1,224 2 1,322.5 1.396 4 55-59 1,353 6 1,353 9 1,352 7 1,352 6 1,352 6 1,346.7 1,281 6 1,229 2 1,165 9 1,118 3 1,060 7 1,063 3 60-64 1,2193 1,229 8 1,245 1 1,247 5 1,245 0 1,2403 1,2407 1,237 0 1,236 5 1,236 8 1,234 5 1,178 0 65-69 981 4 1,020 8 1,0366 1,054 3 1,065 3 1,077 0 1,084 5 1,096 2 1,098 5 1,096 3 1,095 1 1,099 5 70-74 488 3 531 5 599 7 705.5 775 1 807 I 838 6 853 9 869 1 878.9 891 8 904 7 75-79 534 3 500 1 444 1 368 9 342 9 3574 3895 446 8 521 7 570 5 598.0 626 6 80-84 282 5 293.7 308 2 3164 3170 3163 2913 262.3 2167 203 7 219 I 245 9 85 andover 127 1 132.9 1444 149.9 I56 0 I62 8 I64 6 I72 3 178 1 178.3 181 4 I77 5 Male Total 11,449.1 11,435.3 11,200.7 11,176.4 11,156.8 11,124.0 11,080.9 11,041.4 11,012.1 10,984.5 10,968.9 10,949.5 0 - 4 920 8 881 4 797 7 739 3 688 1 6305 621 2 604 4 598 1 589.1 587 9 582 9 5 - 9 868 3 849 8 848.1 874.3 891 5 8900 8466 788 7 731 5 680 8 624 2 E14 9 10-14 1,004.9 991.4 958.1 911.7 872 5 851.7 831.1 842 9 868 9 885 2 883 9 841 5 15-19 960.2 963 8 998 6 1,019.2 1,018 7 1,0033 985 7 952 1 906 6 867 8 847 5 827 6 20-24 1,006.1 1,064 4 1,035.1 941 6 917 9 9265 955 I 990.0 1,012 2 1,012.7 998 5 981 7 25-29 713 6 687 4 652 0 789.2 875 5 9456 9946 1,019 6 929 7 908 1 918 7 949 2 30-34 867 6 836 0 762 3 719 3 686 8 6595 6346 640 1 776 4 862 3 932 9 982 7 35-39 867 2 877 9 860 7 858 8 853 4 8168 782.5 744 0 702 8 671.9 646 9 624 2 40-44 719 7 760.9 778 9 809 7 806 5 818 1 826 4 832.1 830 3 825 9 792 6 761 8 45-49 589 4 589 5 586 5 598 1 633 7 682.6 7181 743 7 172 9 770 2 783 4 793 4 50-54 699 3 666 1 636 2 603.3 578 8 5477 5473 549 6 560 5 594.7 642 4 677 8 55-59 653 8 652 9 648 6 646 3 643 8 638.4 6057 579 8 549 9 528 I 501 7 503 2 60-64 573.7 578 6 586 0 585.5 582 4 578 1 576 4 572 1 569 2 566.8 563.9 537 0 65-69 430 1 454 0 464 8 476.0 481 6 4865 489 1 493 3 492 5 489 9 487.7 488 4 70-74 I96 9 214.2 240 8 285.6 317.7 335.7 3525 362 2 371 0 375 8 381 6 386 9 75-79 213.9 197.1 I72 8 141.4 1300 134.8 I46 6 168.6 I98 0 219 3 233 1 247 3 80-84 1126 1163 1195 121.2 1199 118.1 IO7 0 95 4 77 5 71.8 17.2 86 4 85 andover 51 1 53 5 53 8 55 8 58 0 60 3 60.4 62.7 64 2 64 0 64 8 62 8 Female Total 11,757 6 11,749 8 11,5883 11,578 9 11,573 8 11,5570 11,526.7 11,5045 11,4907 11,4735 11,466 4 11.458 9 0 - 4 883 8 a45 5 164 0 708.1 656 5 5996 5892 572 4 566 3 558 0 556 9 552 7 5 - 9 830 0 811.5 8149 838.4 856.9 855.6 8135 756 6 701 4 649 9 594 1 583 8 10-14 960 8 949 2 9194 875 6 837 2 817 1 798 2 8107 834 0 851 9 851 1 809 8 15-19 919 2 921 7 941 8 964 6 970 6 9602 9456 915 1 872 1 834 0 814.3 795 9 20-24 968 0 1,024 9 1.018 1 922 3 a93 9 8953 910.6 935 2 959 2 966.2 957 0 943 2 25-29 680 8 656 6 630 2 764.5 846 9 915.3 971 7 1,004 2 911 3 884 7 888.1 905 6 30-34 840 7 806 3 747.4 704 0 671 3 645.1 618 8 621 7 754 5 836 0 904 6 963 1 35.39 852 7 863 1 855 9 855 1 850 1 8127 777 1 737 3 695 I 663 1 638 3 613 8 40-44 7174 757 2 779.0 812 3 810 3 8233 833.5 842 3 842 2 837 9 802 2 768 7 45-49 608 0 607 3 597.9 6100 646 4 697.0 7353 763.8 796 7 795 2 808 9 820 3 50-54 741.5 7104 675 1 640 5 613 2 5800 5802 581 5 593 6 629 5 680 1 718 6 55-59 699 8 701 0 704 1 706.3 708 8 708.3 6759 649 4 616 1 590 2 559.1 560 0 60-64 645 6 651.2 659 I 662 I 662.6 6622 6643 664 9 667 3 669 9 670 6 641 0 65-69 551 3 566 8 571.8 578 2 583 7 5905 5954 602.9 606 0 606 4 607 4 611 0 70-74 291 4 317.3 359 0 419 9 457.4 471 5 486 1 491.8 498 1 503 2 5102 517 8 75-79 320 4 302 9 271 3 227.5 212.9 222.6 2429 278 2 323 7 351 2 364.9 379 3 80-84 I69 9 I77 5 188 8 195 2 197.1 198.3 1843 166.9 139 1 131 9 141.9 1594 85 andover 76 0 79.3 90 7 94.1 98 o 102 6 IO4 2 1096 1139 114.3 116.6 1147 Source:RomaniaStatisdcal Yearbook (2002) Table 1.5 :Labor Market (ILO methodology) 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Thousand of persons) Active Population 11885 12120 11726 11757 11577 11566 11585 11447 10557 male 6361 6514 6379 6369 6296 6262 6254 6155 5768 female 5524 5606 5348 5388 5282 5304 5331 5292 4789 Employment 10914 11152 10936 11050 10845 10776 10764 10697 9673 male 5873 6027 5979 6004 5885 5799 5772 5719 5253 female 5042 5126 4956 5046 4960 4977 4992 4978 4420 Unemployed 971 968 791 707 732 790 821 750 884 male 488 488 399 364 410 463 482 436 515 female 483 480 392 342 322 327 340 314 369 Unemployment rate 8.2 8.0 6.7 6.0 6.3 6.8 7.1 6.6 8.4 male 7.7 7.5 6.3 5.7 6.5 7.4 7.7 7.1 8.9 female 8.7 8.6 7.3 6.4 6.1 6.2 6.4 5.9 7.7 Inactive population' 10863 10592 10906 10752 10937 10908 10860 10961 11238 male 4711 4629 4716 4655 4724 4731 4721 4794 4875 female 6152 5963 6190 6097 6213 6177 6140 6167 6363 ILOmethodology;for 2002: provisional. Notes: 1) ShareofILOunemployedto activepopulation2) 2002: estimatedas total population- activepopulation. Definitions: Active population:all peopleprovidinglabor force available for goods and servicesproductionduringthe reference. period,includingemploymentandunemployment Employment:all peopleaged 15andover who havecamedout an economic or social activityproducinggoods or services, with a duration ofonehour at leastduring the referenceperiod, withaview to obtainincomesinthe formof salaries, inkindremunerationor otherbenefits(workingpersons havingajob, unpaidfamily workers andselfemployed). Unemployed: people aged 15 andover, meetingsimultaneouslythe following criteria:they do nothaveajob, they are readyto start working duringthe next 15days. - N 4 0 0 N zr-" N m 0 E N 0 0 2 N 8 ? 0 N vi 0 N m h B vi 9 N m m h z 0 2 N t- h z U 2 N - m '4 m r- 0 3 N a vi h -9 m 3 N a w h '9 0 2 m N en 01 '? 9 vi -. N N m h 2m r-. N - m -v; 9 m W- N 0 5 2 x 0 N ? 9 ' 4 N 0 N e w 0 N m r . - 0 N 0 0 N h h z co 01 9 c h 2 h h - w, E xI m h z h N 9 n x -z 2 2 2 QI r - w - h 0 z f5 0 N p? vi m 4 N 2 vi Vm N 39 0 N VI! N N 0 N 0 - m- N N 0 0 N 0 0 0 N e- e. 2 h 00 3 p. e- z ro z h m e- m N w - m m m ? . * 0 4 xh 0; m" e- m % N h 2 - m h C Y - ~ - % e- 0 2 2 2 2 % p 0 3 x c I 40 -0m0 9 ' 4 7 z 0 - -- 7 ? N Y 0 0 0 ? ? * = 0 - ' 4 0 ! ' 4 0 0 - 4 ' 4 4 9 f 2 0 0 ? n Y 9 - Y 4 i.0- - m E O 3 9 24 9 9 9 -Y0: VI0i.N"- 0 - p 1 2 2 N - 9 ' 4 4 m t ; - - n e - = ' ? = D9 - '4'4'4 E N 0 W v i Q m 0 s x -- m - T ; * h i . . - 0 0 N N - 6 :: y < i' N 0 0 N - 0 N 0 0 N m m 0' :; y G 0 0 0N m Q' m 00 0' (n r- m I .-E 0' --a I cf m w Q' (n V a N - 0 N 0 0 N 0 m- m 2 d m , 218 Table 2.9: Structure of IndustrialOutput, by Activities __ 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 (Billion RomanianLei) INDUSTRIAL OUTPUT 2768 6726 18091 44301 60339 92969 217340 260899 392878 632032 966445 inbillionUS$ 36.2 21.8 23.8 26.8 29.7 30.2 30.3 29.4 25.6 29.1 33.3 Memo"averageexchange rate inROL/US$ 22 76 308 760 1655 2033 3083 7168 8876 15333 21693 29061 ( Percentoftotal output) Mining andquarrying 9.0 8.9 7.5 6.8 7.1 6.9 6.7 7.9 6.8 5.3 5.6 5.7 Coal mining andpreparation 1.o 1.1 2.2 2.1 2.4 2.2 2.2 1.8 1.5 1.3 1.4 1.3 Petroleumand gas extraction 6.8 6.7 4.4 4.0 3.7 3.6 3.6 5.1 4.4 3.1 3.5 3.8 Other extractiveactivities 1.1 0.9 0.8 0.7 0.9 1.1 0.9 1.0 0.9 0.9 0.7 0.6 Manufacturingindustry 85.8 83.5 81.6 84.6 79.6 80.5 81.9 78.9 78.8 74.4 79.4 79.7 Foodandbeverages 14.4 14.8 14.7 19.4 15.9 16.3 17.1 16.1 18.2 13.9 16.1 14.3 Tobacco 0.5 0.5 0.4 0.6 0.5 0.5 0.7 1.1 1.5 1.8 1.5 1.6 Textile andtextileproducts 6.8 6.7 4.9 4.3 3.5 3.2 2.8 2.6 2.4 2.2 2.1 2.1 Textile, furs andleather wearingapparel 3.6 2.7 2.1 2.6 2.2 2.6 2.9 2.4 3.2 3.5 3.3 3.6 Leatherandfootwear 1.9 1.9 1.7 1.6 1.2 1.5 1.4 1.2 1.4 1.3 1.3 1.4 Woodprocessingindustry ( excl. fumitue) 1.6 1.8 2.0 1.8 1.8 1.9 1.9 1.8 2.0 2.6 2.5 2.5 Pulp, paper andcardboard 1.2 1.5 1.4 1.0 0.9 1.2 1.2 1.0 1.0 1.1 1.2 1.0 Publishinghouses, polygraphy 0.3 0.3 0.3 1.5 1.1 1.1 1.3 1.1 1.5 1.5 1.3 1.3 Crudeoil processing,coal coking and treatmentof nuclear fuels 6.9 6.5 6.7 8.6 7.7 7.7 6.4 8.3 6.3 7.9 10.1 10.8 Chemical industryandsynthetic and artificial fibres 7.3 7.8 8.9 7.4 7.5 8.7 7.8 7.2 5.8 5.8 7.0 6.3 Rubberandplasticsprocessing 2.6 2.6 2.2 2.3 1.6 2.1 2.1 1.7 1.7 1.8 1.7 1.7 Other productsof nonmetallic minerals 3.5 3.6 3.9 2.9 3.7 3.8 3.9 4.2 3.8 3.5 3.3 3.3 Metallurgy 8.5 9.6 10.6 8.8 9.5 10.4 10.4 11.6 9.7 9.0 11.4 13.2 Metallic constructionsand metal products 4.1 3.3 3.0 2.5 4.4 2.8 3.0 2.5 3.1 2.7 2.5 2.5 Machinesandequipments 9.3 7.8 7.5 6.2 5.4 5.8 5.6 4.6 4.4 3.8 3.6 3.7 Computers 0.5 0.1 0.1 0.3 0.1 0.2 0.4 0.4 0.4 0.3 0.2 0.2 Electric machineryand appliances 2.5 2.4 2.2 1.7 1.8 2.0 2.3 2.1 2.3 2.0 1.9 1.9 Radio,TV sets andcommunications equipment 0.8 0.7 0.4 4.1 3.2 0.9 1.3 1.0 0.9 0.8 0.7 0.6 Medical, precision,optical and watchmakinginstrumentsandapparatus 1.1 0.9 0.6 0.3 0.3 0.4 0.5 0.4 0.4 0.4 0.4 0.5 Roadtransportsmeans 3.7 3.5 3.1 2.7 2.9 3.0 4.2 3.3 3.9 3.5 2.5 2.5 Othertransportmeans 2.3 2.1 2.6 1.5 1.3 1.5 1.7 1.7 2.1 2.2 1.9 1.9 Fumiture and other nonclassifiedactivities 2.2 2.3 2.2 2.2 2.8 2.6 2.7 2.4 2.5 2.4 2.3 2.3 Waste recovering 0.2 0.1 0.1 0.3 0.3 0.3 0.4 0.2 0.3 0.4 0.6 0.5 Electricandthermalenergy, gas andwater 5.2 7.6 10.9 8.6 13.3 12.6 11.3 13.2 14.5 20.3 15.0 14.6 Electric andthermal energy, gas andhot water production,transport anddistribution 5.0 7.3 10.3 7.9 12.4 11.6 10.3 12.2 13.0 19.1 14.0 13.8 Water collection, treatment, distrib 0.2 0.3 0.6 0.7 0.9 1.0 1.0 1.0 1.4 1.2 1.0 0.8 The industrial output representsthe valueof finishedgoods deliveredor to be delivered, semi-fabsfrom own production, the value of processingrawmaterialsandcustomers'materials,work(services) with industrialcharacterrenderedto third parties, and remainingunfinishedoutput and semi-fabs. Source: StatisticalYearbooksof Romania 1996, 1999, 2001, 2002. 219 Table 2.10: Evolution of the Privatization Process Year 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Number o f Companies Prizatized) Total 265 860 1479 2791 3832 5073 6420 7761 Total 7956 8215 Small 239 711 1031 2010 2930 3823 4576 5646 Small and Medium 24 134 404 640 797 1066 1546 1793 Medium 7625 8012 Large 2 15 44 69 105 184 298 302 Large 311 346 (Equity Privatized inbillion Romanian Lei) Total CCs 64 575 3747 6854 10760 19180 28597 41370 Small CCs 32 121 354 779 1151 1705 2279 4756 Medium CCs 22 321 2176 3797 4835 6639 9930 15134 Large CCS 11 133 1216 2279 4774 10835 16388 21480 Notes: Small companies: with capital under 10billion Romanian Lei. Medium companies: with capital between 10 and 50 billion Romanian Lei. Large companies: over 50 billion Romanian Lei. Source: Romania privatization team and data available on M A P S site. 220 Table 2.11: Private Sector Share in Total GDP and in Value Added by Branches 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001' 2002' (Percent o fGDP) GDP, total 16.4 23.6 26.4 34.8 38.9 45.3 54.9 60.6 61.4 63.7 65.6 67.9 66.8 (Percent of Gross Value Added) GVA in Industry 5.7 9.2 11.8 17.4 23.3 29.9 38.5 42.1 46.0 53.7 68.4 76.0 78.6 GVA in Agriculture andForestry 61.3 73.9 81.7 83.5 89.3 89.0 90.1 96.8 96.3 96.7 98.6 98.9 97.1 GVA in Construction 1.9 16.1 21.0 26.8 51.6 57.8 69.3 76.6 79.3 81.9 91.7 96.3 96.4 GVA in Services 2.0 16.8 18.8 29.3 39.1 58.1 66.7 71.5 76.1 76.6 71.6 67.4 64.4 (Percent o f total) Investment 4.3 8.1 15.6 26.0 36.8 39.3 39.7 35.4 40.5 50.5 58.3 62.4 66.4 Construction 3.1 4.5 13 30.2 55.5 70.8 75.5 70.5 74.7 85.6 89.3 89.8 89.6 Export(FOB) 0.2 15.9 27.5 27.9 40.3 41.2 51.4 54.8 48.9 65.7 65.7 66.7 66.6 Import (FOB) 0.4 16.1 32.8 27.2 39.2 45.4 48.3 52.4 48.3 72.1 70.1 69.6 72.6 RetailTrade 70.1 71.9 74.3 79.9 86.7 90.8 92.0 93.3 na Servicesdeliveredto households .... .... .... 64.4 43.9 55.9 58.9 60.3 62.1 74.5 83.4 85.7 87.4 na Average number of employees 0.7 2.4 5.3 10.9 17.5 22.1 23.0 27.8 34.6 39.4 44.5 49.0 na a)Provisionaldata. Source: National Institute fur Statistics. 221 Table 3.2: Exportsby Major Commodities Section 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (million USdollars) Total Exports, f.0.b. 5775 4266 4363 1892 6151 7910 8085 8431 8302 8487 IO366 11385 13876 Live animals and animal products 23 148 185 162 219 170 155 199 94 116 126 131 145 Vegetable products 38 70 50 58 64 209 374 164 181 250 1I9 181 178 Animal or vegetable oils and fats 0 6 7 61 46 80 67 126 70 54 21 25 10 Food, beveragesand tobacco 21 36 48 47 68 72 108 106 91 64 72 97 105 Mineral products I136 621 572 574 714 731 692 638 507 502 822 784 1181 Chemicalproducts 301 285 423 345 489 724 691 560 336 327 519 503 481 Plastics and tubber 106 54 85 82 143 202 193 186 174 182 230 231 359 Hides and skins, dressedleather, furs 36 25 25 35 56 53 53 66 62 65 99 143 I73 Wooden products, excluding furniture 180 115 155 177 222 258 296 338 385 495 564 530 624 Paper and paper articles 33 22 17 19 34 85 58 64 46 48 81 105 125 Textiles and textile articles 573 404 455 785 1156 1570 1733 1942 2162 2197 2506 2979 3514 Footwear, headger, umbrellas, others 129 87 73 160 311 428 505 545 609 683 789 982 1166 Articles o f stone,plaster,cement 78 75 88 100 111 152 150 151 157 159 161 172 200 ceramics, glass and similar materials Basic metals and articles 949 634 735 959 1065 1437 1268 1557 1583 1310 1658 1516 1782 ,Machinery, appliances and electric 947 692 508 439 519 656 673 737 790 968 1451 1680 2175 equipments; audio and video Transportation means and materials 770 492 473 405 391 428 441 448 427 450 509 598 786 Miscellaneous goods and products 413 404 366 402 436 574 552 526 519 502 533 613 745 others (XNIXVIII+XIX, XXII) 42 96 98 82 107 81 75 78 109 115 107 I15 127 (as percent oftotal exports) Live animals and animal products 0.4 3.5 4.2 3.3 3.6 2.1 1.9 2.4 1.1 1.4 1.2 1.2 1.0 Vegetable products 0.7 I.6 1.1 1.2 1.o 2.6 4.6 1.9 2.2 2.9 1.1 1.6 I.3 Animal or vegetable oils and fats 0.0 0. I 0.2 1.2 0.7 1.0 0.8 1.5 0.8 0.6 0.2 0.2 0. I Food, beveragesand tobacco 0.4 0.8 1.1 1.0 1.1 0.9 1.3 1.3 1.1 0.8 0.7 0.9 0.8 Mineralproducts 19.7 14.6 13.1 11.7 11.6 9.2 8.6 7.6 6.1 5.9 7.9 6.9 8.5 Chemicalproducts 5.2 6.7 9.7 7.1 7.9 9.2 8.5 6.6 4.0 3.9 5.0 4.4 3.5 Plastics and rubber 1.8 1.3 1.9 1.7 2.3 2.6 2.4 2.2 2.1 2.1 2.2 2.0 2.6 Hides and skis, dressedleather, furs 0.6 0.6 0.6 0.7 0.9 0.7 0.7 0.8 0.7 0.8 1.0 1.3 1.2 Wooden products, excluding furniture 3.1 2.7 3.6 3.6 3.6 3.3 3.7 4.0 4.6 5.8 5.4 4.7 4.5 Paper and paper articles 0.6 0.5 0.4 0.4 0.6 1.1 0.7 0.8 0.6 0.6 0.8 0.9 0.9 Textiles and textile articles 9.9 9.5 10.4 16.0 18.8 19.8 21.4 23.0 26.0 25.9 24.2 26.2 25.3 Footwear, headger, umbrellas, others 2.2 2.0 1.7 3.3 5.1 5.4 6.2 6.5 7.3 8.0 7.6 8.6 8.4 Articles o f stone,plaster,cement I.4 1.8 2.0 2.0 I.8 1.9 1.9 1.8 1.9 1.9 1.6 1.5 1.4 ceramics, glass and similar materials 0.0 Basic metals and articles 16.4 14.9 16.8 19.6 17.3 18.2 15.7 18.5 19.1 15.4 16.0 13.3 12.8 Machinery, appliances and elecmc 16.4 16.2 11.6 9.0 8.4 8.3 8.3 8.7 9.5 11.4 14.0 14.8 15.7 equipments; audio and video 0.0 0.0 0.0 Transportation means and materials 13.3 11.5 10.8 8.3 6.4 5.4 5.5 5.3 5. I 5.3 4.9 5.3 5.7 Miscellaneous goods and products 7.2 9.5 8.4 8.2 7.I 7.3 6.8 6.2 6.3 5.9 5.1 5.4 5.4 others (xN+XVIII+XIX, XXII) 0.7 2.3 2.2 1.7 1.7 1.o 0.9 0.9 1.3 1.4 I.o I.o 0.9 Source: National Institute for Statistics. 222 Table3.3: Importsby Major Commodities Section 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (million USdollars) Total Imports, c.i.f. 9202 5793 6260 6522 7109 10278 11435 11280 11838 10557 13055 15552 17862 I Liveanimalsandanimalproducts 352 42 86 68 99 I29 74 65 208 i23 148 280 295 I1 Vegetable products 424 369 429 475 140 154 167 170 226 224 273 335 276 Ill Animalorvegetableoilsandfats 46 14 39 21 30 28 21 33 56 32 35 34 66 N Food, beveragesand tobacco 389 360 442 400 391 580 605 426 524 416 476 558 537 V Mineralproducts 3,896 2,626 2,028 1,872 1,906 2,484 2,689 2,408 1.687 1,252 1,893 2,237 2,272 VI Chemical products 655 410 414 514 563 932 990 940 1,028 975 1,077 1,218 1,503 VI1 Plastics and rubber 92 76 180 204 227 390 448 444 512 477 581 758 98I VI11 Hides and skins, dressedleather, furs 92 18 76 1I 4 149 211 263 278 306 302 362 509 620 I X Wooden products, excluding fumiture 75 35 33 30 27 56 52 53 59 68 83 1I O 149 X Paper and paper articles 44 39 91 96 135 235 274 255 311 264 291 345 419 X I Textiles and textile articles 264 246 562 659 810 1,209 1,343 1,565 1,825 1,937 2,131 2,500 2.930 XI1 Footwear, headger, umbrellas, others 15 45 42 48 67 106 124 157 198 190 222 254 291 XI11 Articles of stone,plaster,cement 1I 4 74 55 63 69 1I 4 145 143 163 154 179 230 291 ceramics, glass and similar materials xv Basic metals and articles 522 243 272 278 352 549 715 670 790 688 891 1,138 1,322 X V I Machinery, appliances and electric 1,321 838 914 1,149 1,450 2,113 2,506 2,593 2,723 2,483 3,216 3,527 4.093 equipments; audio and video xvn Transportation meansand materials 637 142 294 283 332 397 415 384 485 420 552 799 1,018 xx Miscellaneous goods and products 72 31 52 77 113 181 202 209 227 207 242 275 338 Others (XIV+XVIII+XIX, XXIQ 192 185 251 171 249 410 402 487 510 345 403 445 461 (as percent in total imports) I Live animals and animal products 3.8 0.7 I.4 1.o 1.4 I.3 0.6 0.6 1.8 1.2 1.I 1.8 1.7 II Vegetableproducts 4.6 6.4 6.9 7.3 2.0 I.5 1.5 1.5 1.9 2.1 2.1 2.2 1.5 111 Animal or vegetable oils and fats 0.5 0.2 0.6 0.4 0.5 0.3 0.2 0.3 0.5 0.3 0.3 0.2 0.4 N Food, beverages and tobacco 4.2 6.2 7.1 6.1 5.5 5.7 5.3 3.8 4.4 3.9 3.6 3.6 3.O V Mineral products 42.3 45.3 32.4 28.7 26.8 24.2 23.5 21.3 14.3 11.9 14.5 14.4 12.7 V I Chemical products 7.1 7.1 6.6 7.8 7.9 9.0 8.6 8.3 8.7 9.2 8.2 7.8 8.4 VI1 Plastics and rubber 1.o 1.3 2.9 3.1 3.2 3.8 3.9 3.9 4.3 4.5 4.5 4.9 5.5 VI11 Hides and skins, dressedleather, furs 1.o 0.3 1.2 I.7 2.1 2. I 2.3 2.5 2.6 2.9 2.8 3.3 3.5 I X Wooden products, excluding fumiture 0.8 0.6 0.5 0.5 0.4 0.5 0.5 0.5 0.5 0.6 0.6 0.7 0.8 X Paper and paper articles 0.5 0.7 1.4 1.5 1.9 2.3 2.4 2.3 2.6 2.5 2.2 2.2 2.3 X I Textiles and textile articles 2.9 4.2 9.0 10.1 11.4 11.8 11.7 13.9 15.4 18.3 16.3 16.1 16.4 XI1 Footwear, headger, umbrellas, others 0.2 0.8 0.7 0.7 0.9 1.o 1.1 1.4 1.7 1.8 1.7 1.6 1.6 XI11 Articles o f stone,plaster,cement 1.2 1.3 0.9 1.o 1 .o 1.I 1.3 1.3 1.4 I.5 1.4 I.5 1.6 ceramics, glass and similar materials 0.0 0.0 0.0 xv Basic metals and articles 5.7 4.2 4.3 4.3 4.9 5.3 6.3 5.9 6.7 6.5 6.8 7.3 7.4 XVI Machinery, appliances and electric 14.4 14.5 14.6 17.6 20.4 20.6 21.9 23.0 23.0 23.5 24.6 22.7 22.9 equipments; audio and video 0.0 0.0 0.0 xvn Transportation meansand materials 6.9 2.5 4.7 4.3 4.7 3.9 3.6 3.4 4.1 4.0 4.2 5.1 5.7 xx Miscellaneous goods and products 0.8 0.5 0.8 1.2 1.6 1.8 1.8 1.9 1.9 2.0 1.9 1.8 1.9 Others (XIV+XVIII+XIX, XXID 2.1 3.2 4.0 2.6 3.5 4.0 3.5 4.3 4.3 3.3 3.1 2.9 2.6 223 Table 3.4: Direction of Trade inExports a 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (million USD) Total Exports, f.0.b. 4892 6151 7910 8084 8431 8302 8487 10367 11385 13876 (percent of total ) Europe 65.5 67.7 71.0 74.0- 75.8 81.2 84.1 84.9 85.1 82.3 EuropeanUnion 1) 41.4 48.2 54.1 56.5 56.6 64.5 65.5 63.8 67.8 67.1 of which: Austria 1.6 1.6 2.0 2.1 2.1 3.0 2.9 2.4 3.0 3.0 France 4.5 5.1 5.8 5.7 5.5 5.9 6.2 7.0 8.1 7.6 Germany 14.3 16.1 18.1 18.4 16.8 19.6 17.8 15.7 15.6 15.6 Italy 8.3 12.9 15.7 17.1 19.5 22.0 23.3 22.4 24.9 24.9 Netherlands 4.3 3.5 3.0 4.2 3.1 3.8 3.9 3.2 3.4 3.1 UKofGreat BritainandNorthern 3.8 3.3 3.0 3.1 3.5 3.6 4.9 5.3 5.1 5.8 Other Europeancountries Bulgaria 2.1 1.7 0.9 0.9 0.7 0.9 1.6 2.8 1.8 1.3 HwwY 2.4 2.6 2.2 2.1 2.2 2.6 3.2 3.4 3.3 3.1 Poland 0.4 0.2 0.5 0.7 1.2 1.o 1.4 1.o 0.9 0.8 Republic of Moldova 1.9 1.o 1.o 1.2 1.5 1.6 1.2 1.4 1.o 0.8 RussianFederation 4.5 3.4 2.0 2.0 3.0 1 .o 0.6 0.9 0.7 0.3 Switzerland 2.4 0.7 0.8 0.5 0.5 0.6 0.7 0.6 0.5 0.5 Turkey 5.7 4.1 4.4 4.8 4.2 3.9 5.5 6.0 4.0 4.1 Ukraine 2.1 1.5 1.l 0.8 1.1 0.6 0.7 0.9 0.4 0.3 Yugoslavia 0.2 0.3 0.4 1.7 1.7 1.4 1.o 1.3 1.4 0.9 Asia 25.0 19.6 16.6 14.2 12.9 8.0 6.9 6.5 7.4 9.7 China 8.6 4.5 2.3 1.2 0.5 0.3 0.4 0.8 0.8 1.5 Africa 5.3 6.8 7.2 6.8 5.4 4.4 3.9 3.6 2.6 2.1 America 4.1 5.8 5.1 4.9 5.8 6.3 5.0 4.8 4.6 5.5 United Stateso fAmerica 1.4 3.1 2.5 2.4 3.8 3.8 3.l 3.1 3.1 4.3 Notes: According to 2002 structure. a)The breakdownby continents and countries includegoods for which customsdeclarations havebeen completedwithdestination country for export and origin country for imports. Source: National Institute for Statistics ;staffcalculations. 224 Table 3.5: Direction of Trade in Imports a 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Million USD) Total Imports, c.i.f. 6522 7109 10278 11435 11280 11838 10557 13055 15552 17862 (percent o ftotal ) Europe 71.5 75.8 76.6 77.3 77.6 82.6 82.6 80.7 81.6 82.5 European Union 1) 45.3 48.2 50.5 52.3 52.5 57.7 60.7 56.6 57.3 58.4 ofwhich: Austria 2.5 2.7 3.1 3.1 2.7 2.9 2.9 2.5 2.8 3.3 France 7.8 5.1 5.2 4.9 5.7 6.9 6.7 6.1 6.3 6.4 Germany 15.8 18.0 17.5 17.6 16.4 17.4 17.4 14.7 15.2 14.9 Italy 9.4 11.8 13.3 15.3 15.8 17.4 19.5 18.7 19.9 20.7 Netherlands 2.4 2.5 2.7 2.3 2.0 2.3 2.3 2.2 2.1 2.2 United Kingdom 2.6 3.1 2.9 2.9 3.4 3.4 4.2 4.1 3.5 3.8 Other European countries Bulgaria 1.1 0.9 0.7 0.6 0.5 0.4 0.5 0.7 1.o 0.8 Hungary 2.6 2.3 3.1 2.5 3.1 4.6 3.9 3.9 3.9 3.6 Poland 0.5 0.4 0.6 0.8 0.8 1.2 1.5 1.5 1.8 2.0 Republic o f Moldova 1.3 1.3 1.o 0.7 0.5 0.5 0.4 0.3 0.2 0.3 Russian Federation 11.7 13.8 12.0 12.5 12.0 9.0 6.7 8.6 7.6 7.2 Switzerland 2.3 2.1 1.9 1.7 1.3 1.1 1.2 1.2 1.1 0.9 Ukraine 2.0 2.1 1.9 1.6 1.6 1.4 1.0 1.5 2.1 1.9 Yugoslavia * * * 0.3 0.5 0.5 0.5 0.5 0.2 0.1 Turkey 2.3 2.1 2.4 1.9 1.9 2.3 2.2 2.1 2.4 3.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Asia 16.6 11.6 11.2 11.6 11.9 8.5 9.4 9.6 8.9 11.1 China 1.4 0.8 0.9 1.o 1.1 1.5 1.4 1.3 1.6 2.1 Africa 2.8 2.4 4.7 2.7 1.7 0.7 0.8 0.6 0.8 0.8 America 8.7 9.5 6.7 7.3 7.6 6.9 5.8 5.7 5.9 5.0 UnitedStates ofAmerica 5.7 6.5 4.1 3.8 4.1 4.2 3.5 3.0 3.2 3.0 Notes: According to 2002 structure. a)Thebreakdownby continentsandcountriesinclude goods for which customs declarationshavebeen completedwith destinationcountry for export andorigin country for imports. Source: NationalInstitutefor Statistics; staffcalculations. 225 Table 3.6: ForignDirectInvestmeent (subscribedcapital) Year No. of companies Amount of SubscribedCapital number % billionLei Y million US$ YO millionEuro Y O 1991-2002 90,749 100 132,774.8 100 9,120.4 100 8,197.6 100 1991 5,501 6.1 1,197.1 0.9 872.1 9.6 783.9 9.6 1992 11,834 13.0 584.8 0.4 544.9 6.0 489.8 6.0 1993 10,626 11.7 916.6 0.7 416.0 4.6 373.9 4.6 1994 11,097 12.2 2,357.3 1.8 911.7 10.0 819.4 10.0 1995 3,389 3.7 737.5 0.6 285.6 3.1 256.7 3.1 1996 3,620 4.0 2,293.6 1.7 587.0 6.4 527.6 6.4 1997 5,262 5.8 2,394.0 1.8 373.9 4.1 336.1 4.1 1998 8,841 9.7 7,360.7 5.5 731.3 8.0 657.3 8.0 1999 7,385 8.1 12,270.9 9.2 947.7 10.4 851.8 10.4 2000 8,558 9.4 18,191.2 13.7 816.9 9.0 734.2 9.0 2001 7,137 7.9 48,443.8 36.5 1331.7 16.8 1,376.8 16.8 2002 7,499 8.3 36,027.3 27.1 1,101.6 12.1 990.1 12.1 Source: RomanianTrade RegisterOffice. 226 Table 3.7: Foreign Direct Investment by Country of Origin (subscribedcapital as of April 2003) Country No. of companies Amount of SubscribedCapital number % billionLei YO millionUS$ % millionEuro YO ~ Total 93,016 100 144,782.0 100 9,435.0 100 8,48 1.O 100 Netherland 1,624 1.7 28,508.6 19.7 1,661.8 17.6 1,493.7 17.6 Germany 10,446 11.2 13,933.1 9.6 927.4 9.83 833.6 9.8 France 2,923 3.1 10,3 17.1 7.1 695.3 7.37 625.0 7.4 USA 3,613 3.9 6,162.2 4.3 677.2 7.18 608.7 7.2 Italy 13,046 14.0 6,754.4 4.7 593.2 6.29 533.2 6.3 Austria 2,609 2.8 12,673.2 8.8 578.2 6.13 519.7 6.1 Neth. Antilles 7 0.0 16,664.4 11.5 526.7 5.58 473.4 5.6 Cyprus 1,064 1.1 6,881.9 4.8 470.7 4.99 423.1 5.0 Turkey 8,406 9.0 3,437.4 2.4 402.2 4.26 361.6 4.3 UK 1,524 1.6 4,739.9 3.3 327.8 3.47 294.7 3.5 Greece 2,406 2.6 4,575.7 3.2 300.4 3.18 270.1 3.2 Switzerland 1,193 1.3 4,301.5 3.0 268.4 2.84 241.3 2.8 Hungary 4,111 4.4 3,936.0 2.7 234.4 2.48 210.7 2.5 SouthKoreea 79 0.1 478.4 0.3 218.3 2.3 1 196.2 2.3 Luzembourg 198 0.2 2,097.5 1.4 171.4 1.82 154.1 1.8 Spain 550 0.6 533.3 0.4 148.4 1.57 133.4 1.6 Virgin Islands 172 0.2 3,046.3 2.1 119.8 1.27 107.7 1.3 Portugal 79 0.1 3,561.7 2.5 108.9 1.15 97.9 1.2 Sweden 758 0.8 2,437.5 1.7 108.6 1.15 97.6 1.2 China 8,152 8.8 1,620.6 1.1 81.2 0.86 73.0 0.9 Source: RomanianTrade RegisterOffice. 2- r0a N 0 * O - r - N O a 0 0 m o m o b - N b m o a v i m 0'. r-"-" a o o m ~cf '4 9 m - N 0 0 N 0 a a r - r - r - ~ c m f -- a P r - o b m o r - b o m b f -- a t- c o o o o o o o o o f a \o \ 0 0 \ 0 0 0 N 9 ~ v i m b o m m v i o m 0" "" 0" '4 s a m - - a v, f a Q 2 8 m m m o c o ro\ f4 z z a 0 2 m p- '9 a. 9 09 P 9 0: '9-'9 9 9 9 P 2 N m m P ~ N = - w O N - N O o N N N 8 2 m N - m?o? -Po? o q- p N . m 3 N 0 0 5 8 m P N 0 m m 0 a 0 s m P m m C a m a s m P m m P b E 32 cE N 01 iD R 2 2 0 Q V, G. 2 2 N P Q a ... ... ... 0 r- P 2 % c1 ... . . a .. .. w s m 0 R . . . N -- a .. .. .. w s r0,4ywm.P w o N o _ m i - '9 p? -.0 f -z u m N 0 Q a ... . .. ... 23 M 0 m 5 ... ... ... m * 8 M -mc 5) a 8 3 .. z 8 F *m sb 3 0 0 N 0 0 0 0 0 0 0 0 0 0 0 N a a z W za t. a s \o m s 0 0 0 0 0 0 0 0 4 a m 0 0 0 0 0 0 0 0 n z 7 9 3 n d za 7 $J P N m 3 a 3 5 N Y 0 0 N z o\ 2 t. W M * 0 M m I- .-) 2 8 E td i; * s> 6EE0 L- 2.. W 2 3 m0 N 0 0 N m e 0 0 N 0 0 0 N ul o o o o o o o o w o o * o o * ul % $ ul e 3 m - Q\ Q\ o m e 3 s w - t- \ d + z ul t w z ul o o o o w b - o w o - m w o w In ul ul e - - - c e * ul z 232 Table 5.1 Prices (annual growth rates) Year CPI (average) CPI (eop) IPPIa (average) GDP deflator 1991 170.2 222.8 220.1 195.0 1992 210.4 199.2 184.8 227.3 1993 256.1 295.5 165 227.4 1994 136.7 61.7 140.5 139.1 1995 32.3 27.8 35.1 35.3 1996 38.8 56.9 49.9 45.3 1997 154.8 151.4 152.7 147.3 1998 59.1 40.6 33.2 54.2 1999 45.8 54.8 44.5 47.8 2000 45.7 40.7 53.4 44.3 2001 34.5 30.3 41.O 37.4 2002 22.5 17.8 24.6 23.5 IPPI= IndustrialProduction Price Index. Since 1998for output delivered on the doemstic market only. Source: National Institute for Statistics. 233 Table 5.2: Administered Prices 1996 1997 1998 1999 2000 2001 2002 Weights TOTAL PRODUCTS (Yo) 9.05 8.76 8.76 10.96 13.71 17.48 20.31 December-on-December percentage change 1996 1997 1998 1999 2000 2001 2002 Medicines 43.3 140.7 38.9 43.1 62.9 29.0 20.1 Electric Energy 58.7 156.1 141.6 92.6 62.5 35.6 16.3 HeatEnergy 58.0 267.0 19.4 140.9 52.2 57.2 45.5 Gas 57.5 265.1 95.7 100.0 68.2 99.6 8.4 Water Supply 69.2 189.6 62.3 53.9 71.1 30.7 39.1 SewerageServices 71.0 205.6 69.7 63.7 64.9 40.7 42.9 Hot Water 58.0 267.0 19.4 140.8 52.2 57.1 45.5 Postal Services 41.1 138.8 72.6 41.9 44.0 26.4 40.9 Telephone 85.6 249.5 145.0 125.0 34.3 29.5 18.4 Radio-Tv Subscription 33.5 660.1 26.7 94.5 59.0 48.4 5.3 UrbanTransport 46.7 184.4 36.1 49.2 58.1 45.7 30.3 RailwayTransport 104.2 265.1 41.4 45.6 53.4 108.3 0.0 Transport By Waterway 31.3 3 16.0 27.9 23.5 0.0 8.3 23.9 FeesForIssueOfIdentityCard, Driving LicenceAnd Passport 158.1 81.0 196.3 45.0 27.9 37.4 Source: National Institute for Statistics. - 0 0 N 0 0 0 N h 01 s h 0 9 a I. 9 h W 9 v, h 9 Yr - h h m h a e N h 3 e h 9 235 Table 5.3B: Real Wage Indices (inpercent) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002' 1990 100.0 81.7 71.3 59.4 59.4 66.5 72.7 56.3 58.2 56.0 58.6 61.5 62.8 1991 100.0 87.3 72.7 72.7 81.4 89.0 68.9 71.3 68.6 71.7 75.2 76.8 1992 100.0 83.2 83.3 93.2 101.9 78.9 81.7 78.6 82.2 86.2 88.0 1993 100.0 100.1 112.0 122.5 94.8 98.1 94.4 98.7 103.6 105.7 1994 100.0 111.9 122.4 94.7 98.0 94.3 98.6 103.5 105.6 1995 100.0 109.4 84.7 87.6 84.3 88.1 92.5 94.4 1996 100.0 77.4 80.1 77.1 80.6 84.5 86.3 1997 100.0 103.5 99.6 104.1 109.2 111.5 1998 100.0 96.2 100.6 105.5 107.8 1999 100.0 104.6 109.7 112.0 2000 100.0 104.9 107.1 2001 100.0 102.1 Note: Net salary eamings are calculatedby subtractingfromthe gross nominal salary eamings the tax and the employee's contributionfor unemployment, supplementary pensionand health insurance.Realnet salary eamingindex is calculatedas a ratio betweenthe averagenet nominalsalary earningsindex and the CPI computed for for the householdsof employees. '1 Provisional. Source: 1990-2001:StatisticalYearbook of Romania: 2002: NationalInstitute for Statistics. 236 Table5.4: InterestRate (period average inpercent) Year NBR reference CPI (ave) ratea Tbill Rate LendingRate Deposit Rate Spread inflation 1991 14.8 8.6 6.2 170.2 1992 50.5 49.6 28.3 21.3 210.4 1993 70.0 58.9 33.8 25.1 256.1 1994 65.3 91.4 58.9 32.5 136.7 1995 39.6 48.6 36.5 12.1 32.3 1996 35.0 51.1 55.8 38.1 17.7 38.8 1997 47.2 85.7 63.7 51.6 12.1 154.8 1998 38.0 64.0 56.9 38.3 18.6 59.1 1999 35.0 74.2 65.9 45.4 20.5 45.8 2000 35.0 51.9 53.5 32.7 20.8 45.7 2001 35.0 42.2 45.1 26.4 18.7 34.5 2002 a 20.4 27.2 35.2 18.7 16.5 22.5 a)Since February 1st, 2002 : official discount rate; for 2002: eop. Source: National Bank of Romania, %ill for 1996-2000: Intemational Finance Statistics, IMF. rate 237 Table 5.5 ExchangeRates (RomanianLei perUSD) End of Period Average End of Period Average BRER REER Romanian Lei /US$ Romanian Lei / Euro 1990 34.7 22.4 100.0 100.0 1991 189.0 76.4 76.1 93.1 1992 460.0 308.0 56.9 57.5 1993 1,276.0 760.1 1,445.6 884.6 79.7 19.6 1994 1,767.0 1,655.1 2,134.0 1,967.1 84.5 85.4 1995 2,578.0 2,033.3 3,299.0 2,629.5 88.5 83.5 1996 4,035.0 3,082.6 5,005.0 3,862.9 78.7 75.3 1997 8,023.0 7,167.9 8,867.0 8,090.9 84.3 87.6 1998 10,951.0 8,875.5 12,788.0 9,989.3 106.6 113.8 1999 18,255.0 15,332.9 18,330.8 16,295.6 88.0 96.1 2000 25,926.0 21,692.7 24,117.7 19,955.8 87.7 105.8 2001 31,597.0 29,060.9 27,881.2 26,026.9 85.6 107.3 2002 33,500.0 33,055.5 34,918.7 31,255.2 90.8 110.2 a)BRERas index o fnominal exchange rate against the US$adjusted by US$/Romania CPI; WB staffcalrul. b'IFS REER (as nominal effectiveexchange rate index adjusted for relative movements. in national price and selected main trade partner countries) Source: National Bank of Romania, Intemational Finance Statistics - W F . N 0 N 0 7- 0 0 N 0 0 0 N h N b O W W 9 00 01 - I - - o m m 9 - - I- 0 v) h h 'Z 50 0 e B h h * 9 a.-0 .-id 0 *0k Q .-8 Y0 vi 01 '0 C 9 gc 0 k Q .-0 Ba cr 3 h -Y rr 9 0 0 m 71 -.-.-E C 0 s e n W Q\ m 9 v ic 3 0 m - N h 01 --.-.-E e 0 v 01 9 0 01 9 * c w w c - G i m m o o MAP SECTION