Report No. 29500-BA Bosnia and Herzegovina Country Economic Memorandum May 2005 Poverty Reduction and Economic Management Unit Europe and Central Asia Region Document of the World Bank CONTENTS ACKNOWLEDGMENTS ........................................................................................................... .I EXECUTIVE SUMMARY ....................................................................................................... i INITIAL ACHIEVEMENTS OPPORTUNITIES AND ............................................................................ i11 CHALLENGES.............................................................................................................................. IV OPTIONS FOR THE W A Y AHEAD ................................................................................................. vi1 CONCLUSION................................................................................................................................ x 1. SUSTAININGA GROWTH-ENABLINGMACROECONOMICENVIRONMENT1 ......................................................................................................................... 1 B.POST-CONFLICTECONOMIC A.INTRODUCTION TRENDSAND RECENTDEVELOPMENTS ........................................ 2 c.MACROECONOMIC POLICIES-TRACKRECORD AND OPTIONS FOR THE FUTURE .................... 14 D THEMACROECONOMIC ......................................................................................... 24 . OUTLOOK E.CONCLUSION.......................................................................................................................... 26 2. INTEGRATINGINTOEU-BASEDSUPPLY CHAINS-A STRATEGYFOR BOOSTINGEXPORTS? ........................................................................................................ 29 A.~TRODUCTION ....................................................................................................................... 29 B.TRADEPERFORMANCE ANDPOLICYDEVELOPMENTS 1996......................................... SINCE 30 c.EXPORT PERFORMANCE: CAUSES AND CONSTRAINTS ............................................................ 33 D.POLICYRECOMMENDATIONS: THEWAYFORWARD ............................................................... 46 E.CONCLUSIONS......................................................................................................................... 54 3. ENTERPRISESECTOR STRUCTURE AND PERFORMANCE ............................. 55 A.LYTRODUCTION ....................................................................................................................... 55 B.CHARACTERISTICS OFTHE ENTERPRISE SECTOR INBH .......................................................... 55 c.DIMENSIONS DETERMINANTSENTERPRISE AND OF PERFORMANCE ........................................ 63 D.CONCLUSION.......................................................................................................................... 67 4. REMOVINGIMPEDIMENTSTO ENTERPRISEDEVELOPMENT ..................... 69 A.INTRODUCTION ....................................................................................................................... 69 B.THEOPERATINGENVIRONMENT ............................................................................................. 70 c.THE ROLEOFCORPORATEGOVERNANCE ............................................................................... 83 D.SOFTBUDGET CONSTRAINTS................................................................................................. 89 E.ENTERPRISE ................................................................................ 94 F.REMAINING RESTRUCTURING AND EXIT CHALLENGESAND POLICYRECOMMENDATIONS ................................................. 96 5. INVIGORATINGTHE LABORMARKET ............................................................... 101 ..................................................................................................................... B.THELEGACYA PARTICULARTRANSITION....................................................................... A.LYTRODUCTION 101 OF 102 c.AMORE COMPREHENSIVE LOOK THE LABOR AT MARKET ................................................... 105 D.ANALTERNATIVELOOKAT LABORMARKETFLEXIBILITY .................................................... 112 E.EDUCATION .................................................................................................. 117 F.THEROLEOFEMPLOYMENT AND TRAINING SERVICES INBOSNIA AND HERZEGOVINA ................................ 124 G.CONCLUSIONSAND POLICYRECOMMENDATIONS ................................................................ 126 ANNEX 2 1 FTAS CONCLUDEDBY BOSNIAAND HERZEGOVINA . ANNEX 2.2 BH OUTWARD PROCESSINGEXPORTS TO THE EU. 1997-2001 .............................132 131 ANNEX 3 .1: BEEPS ............................................................................................................. 133 ANNEX 3.2: 1000LARGE ENTERPRISES'SURVEY DETAILS .................................. 135 ANNEX 3.3: DATA FORTHE ANALYSIS OF DETERMINANTS OF ENTERPRISE PERFORMANCE .................................................................................................................. 136 Tables TABLE1:l COMPARATIVEOUTPUTDECLINE ................................................................................ 4 TABLE1:l BH.KEY ECONOMIC INDICATORS. 9 TABLE1:2CBBHRESERVESAND LIABILITIES (INKMM) .......................................................... 1999-2004........................................................... 16 TABLE1:3 BREAKDOWNCONSOLIDATED GOVERNMENT OF 20 TABLE1:4BH:GROSS EXTERNAL DEBT(us$MILLION, END2004) .......................................... EXPENSES. ........................... 2002 24 TABLE1:5 KEY EXTERNAL DEBT RATIOS `'............................................. 24 TABLE1:7BH:REFORM SCENARIO 2005-2009.......................................................................... FOR CEECOUNTRIES 26 TABLE1:8 BHEXTERNAL FINANCINGREQUIREMENTS .............................................................. 27 TABLE2:l BHTRADEINGOODS 1996-2003.............................................................................. 31 TABLE2:2 AVERAGE MFNAPPLIEDTARIFFS,2002(PERCENT) ................................................. 31 TABLE213 COMPARISON OFTRADEDATAFROMDIFFERENT SOURCES, 2000-2003 ...................33 TABLE2:4 BHEXPORTSPERCENTOF GDPIN1990AND 2002 .............................................. AS 34 TABLE2:5 IMPORTANCEOFEU-DESTINED EXPORTS BHAND COMPARATOR COUNTRIES,- 35 FOR TABLE2:6 BH's EXPORTSEND-USE OF PRODUCTSTO EUMARKETS, 1996-2002..................... 36 TABLE2:7 BOSNIA'S RCAINDICESAND OTHER SELECTED COMPARATORCOUNTRIES INEU MARKETS .................................................................................................................................... 38 TABLE218 EXPORTS THE EUUNDER LOHN TO TABLE3:l OUTPUTIN BH, 1998-2003....................................................................................... 56 ARRANGEMENTS, 1996-2002(IN MILLIONUS$) 40 TABLE3:2: BUSINESS CREATION, 1999-2003............................................................................. 57 TABLE3 3 : STRUCTURE OF SMESBY SECTOR, 2001.................................................................. 59 TABLE3:4 SIZE OFTHE INFORMALECONOMY, 2000 .................................................................. 60 TABLE3:5: FOREIGNDIRECT INVESTMENT 1997-2002.61INFLOWS INSOUTH EASTERN EUROPE, TABLE3:6 SALES GROWTH FIRMSBY OWNERSHIP, 1998-2002............................................. OF 64 TABLE3:7: CAPITAL PRODUCTIVITY AND LABOR PRODUCTIVITY, 2002 .................................... 64 TABLE3:8: AVERAGE GROSS PROFITMARGIN NUMBEREMPLOYEES BY OF .............................. 65 TABLE4:l BOSNIA AND HERZEGOVINA: REFORM PROGRESSINDICATORS, ................................ 70 TABLE4:2 OPENINGA BUSINESS, 2004 ...................................................................................... 74 TABLE43: AVERAGE .................................................................................................................................................... 75 SALES ON CREDIT AND THROUGH ADVANCE PAYMENT BY OWNERSHIP, 2002 TABLE4:4: EFFECTIVENESSTHE LEGAL OF SYSTEM BY ATTRIBUTE, PERCENTOF FIRMS INTOTAL76 TABLE4:s TAXRATES FOR SELECTED CENTRALAND SOUTHEASTERN EUROPE COUNTRIES ....79 TABLE4:6 SOURCES OFFINANCING,2002.................................................................................. 81 TABLE4:7: FINANCIALINSTITUTIONS ......................................................................................... 81 TABLE4:8 MARKETSHARE BY SECTOR, 2002 (PERCENT) .......................................................... TABLE4:9 DOWNSTREAM INTEGRATIONBY SECTOR, OWNERSHIP TYPE,AND SIZE, 2002 .........84 84 TABLE4:10ARREARS AND SUBSIDIES, AS A PERCENTOFTOTALANNUAL SALES, 2002 ............90 TABLE4:11 FIRMRESTRUCTURING. 94 TABLE4:12 BANKRUPTCY FBHBY TYPEOF OWNERSHIP. ENDOF 2002 ......................... 2002................................................................................... INTHE 95 TABLE5:1FORMALEMPLOYMENTY SECTOROF ACTIVITY.1991AND 1997-2003...............104 B TABLE5:2 AVERAGE WAGES AND EMPLOYMENT ..................................................................... 105 TABLE5:3 NET GROSS AND WAGES.2003 ............................................................................... 105 TABLE5:4GROsS WAGES INMANUFACTURING I/..................................................................... 106 TABLE5:5 SAMPLE DESCRIPTION, LSMS,BH2001................................................................. 107 TABLE5:6 EMPLOYMENT DISTRIBUTION BH,2001AND 2002 ............................................. IN 108 TABLE5:7 EMPLOYMENT UNEMPLOYMENT RATE, BYAGEAND EDUCATION, 2001AND AND BH 2002 (INPERCENT) .................................................................................................................... 109 TABLE5:8 MAINLABOR MARKET INDICATORS ........................................................................ 110 TABLE5:g COMPOSITION OF EMPLOYMENT (2001) AND OF EMPLOYMENT CREATION AND DESTRUCTION (2001-2002) ....................................................................................................... 111 TABLE5:lO WAGEREGRESSIONRESULTS, BHLSMS2001AND 2002.................................... 118 TABLE5:11DISTRIBUTIONWORKERSRECEIVINGARREARSORNOTUSUALWAGES, B Y ENTITY OF FOR 2001AND 2002 (PERCENT) ................................................................................................. 119 TABLE5:12 IMPLIEDPRIVATE I m S BY TYPEOF EDUCATION BH,2002........................... FOR TABLE5:13 TRENDSINENROLLMENT DIFFERENT OF mNDSOFSECONDARY SCHOOLS ...........121 122 TABLE5:14ESTIMATED DETERMINANTSPROFITABILITY: OLSREGRESSIONS.................... OF 137 Figures FIGURE1:lREAL BHGDPANDFOREIGNAID ............................................................................. 5 FIGURE1:2REAL GDPGROWTH ENTITY(ANNUAL INPERCENT) ............................................ BY 5 FIGURE1:3 FBH:SECTORCOMPOSITIONOFGDP(INPERCENTOFNOMINALGDP)..................... 6 FIGURE1:4RS:SECTORCOMPOSITION OFGDP (INPERCENTOFNOMINALGDP)....................... 6 FIGURE1:5 SECTORCONTRIBUTION OFNOMINAL (INPERCENT).......................................... GDP 6 FIGURE1:6INDUSTRIAL PRODUCTION.BOTH ENTITIES(1999=100) ............................................ 7 FIGURE1:7CONSUMER PRICEINFLATION..................................................................................... 7 (PERCENTCHANGE YEAR-ON-YEAR) ............................................................................................. 7 FIGURE1:9REMITTANCESANDPRIVATETRANSFERS ................................................................... 8 FIGURE1:10GROSS INFLOWSINBHAND COUNTRIES OFTHE REGION(INPERCENTOF GDP)8 FDI FIGURE1:11GROSS OFFICIAL RESERVES (INMONTHS OF IMPORTSOF GOODS AND SERVICES) .....8 FIGURE1:12DISTRIBUTION THE POPULATION(LEVEL OFPERCAPITA CONSUMPTION) ..........10 OF FIGURE1:13 INEQUALITY INSELECTED TRANSITIONCOUNTRIES(GIN1COEFFICIENT) ...............10 FIGURE1:14LENDING-DEPOSIT SELECTEDCOUNTRIES(INPERCENT) ........................... SPREAD. 11 FIGURE1:15BHREALINTERESTRATES (PERCENT. DEFLATEDWITH SAME YEAR CPIINFLATION) 11 FIGURE1:16EBRDTRANSITIONINDICATORS EBRD -BH.SELECTEDSEE.CEECOUNTRIES INDEX RANGINGFORM l(W0RST) TO 4+ (BEST) FIGURE1:17COMPOSITION OFTHE CURRENT ACCOUNT FIGURE1:18RECENT TRENDSINSAVINGSAND INVESTMENT(INPERCENTOF GDP) ................14 ...................................................................................13 BALANCE (IN PERCENT OF GDP) 14 FIGURE1:19RECENT TRENDS INREALEFFECTIVEEXCHANGE RATES ....................................... 16 FIGURE1%) FISCALTRENDSINTHE FBH(INPERCENTOFENTITYGDP)................................... 17 FIGURE121FISCALTRENDSINTHE RS (INPERCENTOFENTITYGDP)..................................... FIGURE1:22WAGE BILL INSELECTED SEE COUNTRIES (INPERCENTOFTOTAL SPENDING) .....17 18 FIGURE1:23 RATIOOFWAGE BILL TO OPERATIONS AND MAINTENANCE .................................. 18 FIGURE1:24FBH:PROJECTEDALLOCATIONOFTAXREVENUESBETWEENENTITYAND SUB- ENTITY(INPERCENTOFTOTAL) .................................................................................................. 21 FIGURE2:l FOREIGNTRADEPERFORMANCE.1996-2004........................................................... 30 FIGURE2:2 BHEXPORTS FTAPARTNERS.1997-2003.......................................................... TO 36 FIGURE2:3 BHEXPORTS THEEUBYFACTOR TO INTENSITY ..................................................... 37 FIGURE2:4BHEXPORTS CROATIA BY FACTOR TO INTENSITY .................................................. 37 FIGURE2:5 SHAREOFOPTINTOTALEXPORTSTHE EU........................................................ TO 41 FIGURE2:6 NET INFLOWSINTOSUCCESSORSTATESOFYUGOSLAVIA (US$ MILLION) .......44 FDI FIGURE2:7NET INFLOWSINTOSUCCESSORSTATESOFYUGOSLAVIA(PERCENT OFGDP).44 FDI FIGURE2:8 FDIINFLOWS 1994-2003BY SECTORS..................................................................... 45 FIGURE3:l CHANGES INAVERAGE EMPLOYMENTOWNERSHIP IN........................................ BY 58 BHAND SEE8, 1999-2002......................................................................................................... 58 FIGURE312:PROFIT-TO-SALESRATIO BYFIRMCHARACTERISTICS, 2001(PERCENT OFTOTAL FIRMS)......................................................................................................................................... 63 FIGURE4:1 ENTERPRISE PERCEPTIONSOF THE BUSINESS CLIMATE, 1999-2002 ......................... FIGURE4:2 BARRIERS THE OPERATIONAND GROWTH BUSINESSES TO OF (PERCENT OF FIRMS) 73 .72 FIGURE4:3 BARTER,BILLSOFEXCHANGE, DEBT AND SWAPS ................................................... 76 FIGURE4:4 SHARE OFFIRMSWITHN OPAYMENTDISPUTE RESOLVED IN .................................. 77 COURT, BY SIZE .......................................................................................................................... 77 FIGURE4:5 ACCESS FINANCEBY SIZEAND OWNERSHIP, 2002.............................................. TO 82 FIGURE4:6 STRUCTUREOF COMMERCIAL BANKS' LOAN PORTFOLIOS ...................................... 83 FIGURE4:8: WAGE ARREARSTHE FBHAND THE RSBY FIRMSIZE, 2001AND 2002 .............93 IN FIGURE5:1WAGE DISTRIBUTION, FORMAL FIGURE5:2 MINIMUMWAGE AS SHAREOF THEAVERAGE WAGE, TRANSITIONCOUNTRIES ......116 AND INFORMAL SECTOR, BY ENTITY,2001- 2002114 FIGURE5:3 IMPORTANCE OF SKILLSAS RANKEDBYEMPLOYERS ............................................. 120 Boxes Box 1:l STATISTICS INBH:INSTITUTIONALANDMETHODOLOGICAL ISSUES3 .............................. BOX 2:1 DISCREPANCIES INTRADESTATISTICS FROMDIFFERENT SOURCES .............................. 33 BOX 2:2 FLEXIBLEPRODUCTION INTHE APPAREL INDUSTRY ..................................................... 43 BOX 2:3 BUSINESS CLIMATE AS A CONSTRAINT ON EXPORTS ..................................................... 48 BOX 2:4 PAN-EUROPEAN CUMULATION OF RULES ORIGIN OF .................................................... 52 BOX 3.1 FOREIGN-FUNDED PROJECTSCROWDINGOUT DOMESTIC ONES................................... 62 Box3:2 DETERMINANTS OF FIRMPERFORMANCE ...................................................................... 67 Box4:l DATA SOURCES............................................................................................................. 71 BOX 4:2 OBSTACLES TO PRIVATIZATIONOF STRATEGIC FIRMS .................................................. 87 Box4:3 POLITICAL APPOINTMENTS ........................................................................................... 88 Box4:4 INFLEXIBLE ARREARS ELECTRICITY PRICING GENERATES ............................................ 92 BOX 4:5 DEALING WITH EXCESS LABOR ..................................................................................... 93 Box4:6 CLOSINGA BUSINESS (2003) ........................................................................................ 96 Box4:7 OUTOF COURT DEBT RESTRUCTURING/CONCILIATION INBH ..................................... 97 Map:IBRD32760 ACKNOWLEDGMENTS This report was prepared by a World Bank team. The preparation o f the overall report was led by Jean-Luc Bemasconi (task manager) andMarcelo Bisogno. Ardo Hansson and Ivailo Izvorski led the completion o f the report. The chapter on macroeconomic policy was prepared by Jean-Luc Bemasconi and Irina Smimov with input from ChristianBodewig andJim Anderson. The chapter on international trade and export promotion was prepared by Beata Smarzynska, Francis Ng, with key input from Alia Mouyabed and comments from Costas Michalopoulos and initial input from Bart Kaminski. The chapters on enterprise development and reforms was prepared by Gallina Andronova Vincelette and Harry Broadman, with input from Robert Gourley and Gregory Jedrzejczak. The chapter on labor markets was prepared by Marcelo Bisogno and Jean Fares, with input from Jim Stevens and Toby Linden on skills development as well as Christian Bodewig on active labor market policies. The team benefited greatly from guidance received from peer reviewers Daniela Gressani, Helena Tang (concept stage) and Vladimir Kreacic (decision stage); and from comments received on the concept paper for the report from numerous World Bank colleagues. The report was prepared under the overall supervision o f Bemard Funck (Sector Manager). Its production would not have been possible without the essential support from Mismake Galatis, Kathryn Rivera and Zeljka Meheljic in processing the document. Emily Evershedprovided editorial support. EXECUTIVESUMMARY 1. This Country Economic Memorandum (CEM) has three main objectives. First, it takes stock o f economic developments in Bosnia and Herzegovina (BH) since the end o f the war in 1995, with a particular focus on the most recent period. Second, the report discusses potential sources o f high and sustained growth that BH could exploit, as European Union integration is advanced. Finally, the report formulates policy recommendations that would help maximize the potential o f BH economy to achieve higher real GDP growth and generate more jobs. In doing so, the C E M focuses on four priority policy areas: macroeconomic management, international trade policy, business environment and enterprise reform, as well as labor market policies. The latter includes an assessment o fBH's education policies. 2. The report conveys three key messages. First, ten years after the end of the war, BH still suffers from the legacy of a conflict that destroyed a significant part of the country's physical productive capacity, depleted human resources and shattered institutions and social capital. Time lost during the war, combined with subsequent difficulties in coping with the double challenge of post-conflict reconstruction and the transition to a market economy have left BH far behind the most successful countries o f central and eastern Europe (CEE). While post-conflict reconstruction has been successful, mainly because o f large aid flows, the analysis in the report shows that BH lags on many of the structural reforms that have accompanied economic transition inthe region. These include strategic privatization, product and labor market liberalization and enterprise restructuring. 3. As the recent decline in aid flows i s likely to continue, delays in advancing structural reforms constrain BH's self-proclaimed goal o f emerging as a vibrant, productive economy integrated in global markets. The outstanding challenges are many. Strong and sustainable economic growth i s yet to be established after the post-conflict recovery. The job market is anemic and unemployment is high, and the informal economy appears to be gaining further ground. The country's external imbalances remain large and need to be corrected to limit long- term risks. Many companies, especially those that are state-owned or mass-privatized, continue to operate under soft budget constraints, leaving space for inadequate managerial behavior that can range from a weak profit motivation to asset stripping. BH's most precious resource, its people and especially its young population, i s still longing to leave the country. The uncertainties surrounding the country's economic future and the related apparent lack o f opportunities are an important cause o f continuing brain drain. However, the mindset in decision - making, be it at the political level, within the public administration or inside many companies, is still geared towards preserving short-term privileges over establishing grounds for sustained output expansion. 4. As an important caveat, the political and administrative structure which emerged from the Dayton Peace Accords (DPA) makes the task o f catching up more complex. Although it i s 1 outside the purview o f this report or the mandate and expertise o f the World Bank to make any assessment o f the political and institutional relevance o f the DPAs, which the report takes as given, it should b e noted that these arrangements carry a significant cost. The level o f complexity underlying administrative arrangements throughout the country brings direct fiscal costs. It also results in less tangible economic costs because major reforms have to be sanctioned by a mandatory consensus among BH's constituent peoples. Reforms which may appear to put one or another group at a disadvantage often cloud the policy debate and crowd out the reform's likely larger, longer-term development impact. This problem is compounded by weak technical capacity and policy coordination failures across the various levels o f government. The recent vacillations over higher education reforms provide a telling case o f the complexities o f BH's political economy. Not anticipating or suggesting any profound modification to the DPA, this reports signals that the political commitment for achieving reforms in a reasonable time frame needs to be even stronger in BHthan elsewhere in the region in order to overcome the system's inherent inertia. 5. The report's second key message i s that BH can indeed catch up with the more advanced economies in central and Eastern Europe. Opportunities for growth exist and the authorities have in several instances demonstrated the political will and persistence to engage in difficult reforms. Some very important institutional reforms been achieved with a measurable initial impact, even ifnot on the scale and at the pace that would be desirable. BH firms can export, foreign investors see business prospects in BH, and new jobs are being created. Inorder to scale up these initial results, however, a number o f critical challenges need to be addressed. These mainly relate to the business environment, enterprise reform and labor market reform. The authorities' Medium-Term Development Strategy (MTDS) identifies most o f these challenges and provides a vision consistent with that o f this CEM: the objectives are not necessarily higher growth and investment than the fast pace in the aftermath o f the war, but a different type o f growth and investment, where the private sector is inthe driver's seat. To achieve this objective, the MTDS recommends a wide array o f policy measures. Unfortunately, the M T D S fails to prioritize policy actions interms o f their likely impact for boosting real GDP growth. The C E M attempts to fill in this gap by singling out the most pressing issues requiring an urgent policy response inthese areas. 6. The third key message relates to the optimal characteristics of the policy response: selectivity in the choice of priority actions, and flexibility of the policy stance, ranging from holding firm in some policy areas to intervening smartly in others. As noted, BH authorities have not waited for this C E M to identify pressing policy issues and have even made plans in several areas requiring action. Often, however, the envisaged policy response is biased towards active and wide-ranging intervention by government. Moreover, envisaged actions are too numerous to be credible and to serve a signaling function, including those prescribed in the MTDS. The CEM, by contrast, takes a more nuanced view. The underlyingpolicy objective to stimulate the "new" type o f growth i s primarily to set the right incentives for efficiency and effectiveness. Insome areas, such as the trade regime or exchange rate policy, this translates into staying the course. In many other cases, setting the right incentives requires governments to minimize their presence or even get out o f the way. Finally, there are opportunities for active policy intervention, in particular with respect to reorganizing public expenditure in order to reform the education system or provide more adequate safety nets for those who need them the .. 11 most now and in the future as the economy transforms. Within this framework, the remainder o f this summaryreviews the most salient conclusions ofthe report. INITIAL ACHIEVEMENTSAND OPPORTUNITIES 7. Bosnia and Herzegovina has been successful in achieving macroeconomic stability. Inflation has been brought down to industrial country levels. Fiscal accounts have improved and the consolidated fiscal deficit has narrowed markedly. The confidence inthe currency is strong, as evidenced by steadily rising Kh4-denominated deposits. Successful rescheduling agreements, aided by lack o f access to capital markets, have helpedreduce external debt to sustainable levels. Despite a large and persistent current account deficit, foreign reserves have increased. Inflows o f foreign direct investment (FDI) more than doubled than since 2000 to 6 percent o f GDP in2004. Total FDIsince the end o f the war is estimated to have surpassedUS$1.6 billion. 8. The track record in structuralpoliciesis uneven and the agenda largely unfinished, as described above. Still, BHhas managed to develop a competitive banking market, including regulatory and supervisory functions that appear to be enforced. The successes in the financial sector do not come without challenges, however, with credit to households growing at a rapid pace and spilling into imports. Similarly, while the divestiture o f the state from many small- and medium-sized enterprises is completed, the use o f the mass privatization method has had an adverse effect on corporate govemance. 9. In the area of foreign trade, BH has adopted the most liberal trade regime in South EasternEurope (SEE) and has concludedbilateral free trade agreements with all its main tradingpartnersinthe sub-region. These agreements generally meet the standards o f coverage recommended by the Stability Pact. Policymakers and other observers, however, have generally expressed disappointment with the lack o f supply response to this favorable trade regime. As noted earlier, the data presented in this report confirms that export performance has been lackluster. However, these concerns are not entirely justified. Despite the extremely difficult initial conditions of the war's immediate aftermath, BH merchandise exports have increased by 10 percent percent a year in euro terms, while the export coverage o f imports has risen from less than one-tenth to one-third. More importantly, BH seems to have achieved some success in redirecting its trade towards the market which bears the most promise for its future, namely the EU. BHwas also able to gain trade shares with its second trading partner, Croatia. Participation inretail-driven commodity chains, such as in the furniture sector, shows that BH firms can be present in a very competitive market place. Also encouraging i s the progress achieved with curbing corruption at border posts, an important dimension for facilitating trade. 10. While the overall performance o f the corporate sector remains lackluster, the subset o f small and medium-sized enterprises (SMEs) appears to show promise. Business practices o f small, de novo firms are geared towards profit seeking and they have adopted a business culture where contracts are honored, and where suppliers and workers are paid. This contrasts starkly with large enterprises, especially state-owned firms, which still operate on the principles o f the pre-war self-management economy: over-employment, soft budgets, and poor govemance. Many crucial reforms are underway, although some are implemented very slowly. In some instances, the government has shown the resolve needed to advance the reform agenda. The submission o f the bankruptcy legislation in both entities - the Federation o f Bosnia and Herzegovina (FBH) 111 ... and the Republika Srpska (the RS) - and the subsequent successful campaign by the FBH government to convince Parliament to retract the amendments that it had introduced and that would have undermined the implementation o f the law, demonstrated that difficult reforms can be achievedinBH. 11. This duality is also reflected inthe labor market.Even ifBH displays low participation and employment rates overall, the SME sector shows that there i s potential for a dynamic labor market. Initial reforms o f the labor market institutions, including o f the labor code and the employment institutes, have gone in the right direction, but the labor market remains the area where challenges are perhaps the most significant. CHALLENGES 12. As the outstanding challenges are daunting, their analysis should lead to some prioritization in order to guide policy. The single most importantmacroeconomicimbalance i s the persistentcurrent accountdeficit. At close to 17 percent o f GDP, it i s not sustainable in the long run. The analysis presented in this report suggests that it is the result o f low domestic savings rates rather than high investment. The rate o f investment is not above the average for transition economies, but the share o f domestic savings that can finance it is particularly low. In a context where external savings in the form o f aid flows are declining and where investment levels need to remain high to accompany the economic transformation, macroeconomic and structuralpolicies should be geared to stimulating domestic savings and attracting foreign capital o f a new kind. 13. The need for higher domestic savings raises the question of the adequacy of BH's current fiscal policy stance to serve economic growth: can BH reduce the share of public expenditure, while increasingthe efficiency of public services?The challenge may at first seem insurmountable but this report argues that it can be tackled. Despite the gradual reduction inpublic expenditures, the weight o f government in the economy, while considerably reduced inthe last couple o f years, i s still too large at almost 50 percent percent o f GDP. Further consolidation i s called for, and will have to depart from recent efforts focused on investment spending. Curtailing recurrent spending, despite all o fthe inertia which affects such spending, is a top priority. Given the unmet funding needs for maintenance and operations, the public sector wage bill and some transfer programs are the prime candidates for rationalization. Increasing expenditure efficiency is also key to establishing a growth-enabling environment and providing better safety nets for those who will not be able to redeploy to new activities. 14. In the mediumterm, however,fiscal adjustmentalonewill not generatethe required increasein domestic savings. A substantialcontributionwill also needto come fromhigher domestic private savings. Inthe household sector, the recent period has been marked by high access to credit to finance construction and consumption. Inthe framework o f the currency board arrangement that governs exchange and monetary policy, this is believed to have been accommodated by sizeable inflows o f private savings from nationals abroad, including worker remittances. While a drying up o f these resources would automatically restrain credit, the short- term imperative is to rein in credit growth to households by resorting to the limited monetary policy tools that are available under the currency board regime. These include hiking reserve requirements and tightening prudential regulation and banking supervision. In the longer run, iv however, the real challenge is to boost domestic savings in the corporate sector, i.e., the profitability o f companies operating inBH. This will require a series o f actions to strengthen the business environment and corporate governance, stimulate enterprise restructuring and increase the flexibility o f the labor market. Only then will BH firms, whether locally or foreign owned, be able to capture large shares inEuropean and world markets, and contribute to an improvement o f the country's external position. 15. In the area of trade, the recent sluggish performancecan hardly be attributed to a policystancethat haspursuedliberalizationand integration.The real constraintsto a more dynamic export performance lie in the business environment and investment climate. In fact, BH's trade environment offers many opportunities that remain untapped. For example, merchandise exports are much too low, representing a little over 10 percent percent o f GDP. Exports are dominated by raw materials and products intensive inunskilled labor. Opportunities for participating in international supply chains, especially with the EU, are insufficiently taken advantage of. In sectors where BH firms are present, such as textiles and leather products, the long-run sustainability o f their participation i s threatened by tough competitive pressures, calling for an upgrading o f production to higher rungs on the value ladder. This type o f investment will be best promoted ifthe business environment becomes more attractive to foreign investors. Also, opportunities on the growing neighboring market o f Serbia and Montenegro ( S A M ) are not sufficiently exploited. BH has demonstrated that it can successfully export products from the ago-food sector to Croatia. It has however failed to capture the immense EU market for such products, not the least because it lacks an adequate institutional infrastructure for product certification. BH's own domestic market further needs developing into a single economic space that will also stimulate cross-border trade. 16. Despite the destruction o fproduction capacity duringthe war, and in spite o f the pre-war industrial structure that i s not easy to reconvert, some sectors appear to have potential for future growth, such as forestry and manufacturing o f wood products, textiles and leather, other small- scale manufacturing, some agriculture sub-sectors or the electricity sector. However, this report makes the case that policies should focus on ensuring maximum flexibility by removing barriers to investment or doing business and enhancing human capital. Through broad-based incentives, BH can further develop into a multi-sector economy. In contrast, "picking winners" through targeted industrial policy is unlikely to produce any better results in BH than it has in other countries. 17. The challenges of broad-based enterprise reforms essentially center around two main issues (i)removal of structural barriers to the creation and functioning of competitive markets; and (ii) establishment o f robust rules, procedures and related institutions that create incentives to facilitate day-to-day business transactions. 18. Removingbarriers to entry is essential to stimulatingfurther enterprise creation as well as to the growth of recently created de novo firms as this segment o f the enterprise population offers the best opportunities for growth and employment generation. Several bottlenecks should be removed as a priority. 19. First, the report finds that economic policy uncertainty and corruption are among the most powerful obstacles to foreign investment, business creation and development. Second, the V cost and access to finance, tax rates and administration, anti-competitive practices, and the functioning o f the legal system in protecting property rights and contracts place a significant burden on business development and investment. The lack o f strong and competitive enterprise sector in BHmakes financial institutions unwilling to lend to companies. As a result, firms rely on informal sources for working capital and investment needs. Difficulties inaccessing capital as well as the limited security o f contract rights makes businesses transact in primitive ways by resortingto on-the-spot deals. Third, administrative obstacles such as land-titling, access to land, labor regulations, permits, and licensing requirements are seen as a persistent challenge to entry and market competitiveness. It still takes too long and costs too much to open a business inBH. Business registration reform i s ongoing, but the pace of the reform is slowed not so much by the technical difficulties related to the establishment o f an integrated system, but also by protracted debates o f a political nature. Administrative and bureaucratic hassles also increase the costs o f runninga business. The commercial court system suffers from aheavybacklogo fpending cases. In addition, despite recent progress, the country still cannot be described as offering a nation- wide level playing field with respect to taxation levels and collection. 20. There is a compelling case for offering opportunities for longer-established firms to restructure and reorganize. The widespread practice o f accumulation o f arrears and reliance on subsidies hampers such restructuring and needs to cease. The accumulation of arrears indicates a willingness o f governments at all levels to tolerate and protect loss-makers from market exit as well as to accept a practice o f delayed payments among the firms. For state-owned enterprises, the options are clear: privatize strategically or liquidate. The report demonstrates, however, that the reliance on insiders and the state in most o f the completed voucher-privatization deals has left fragmented ownership structures, poor incentives for restructuring and development and weak corporate governance structures. This has limited inflows o f new capital and know-how in the newly privatized firms and has caused investors to be hesitant in bringing in greenfield foreign investment. The government will need to address the challenge o f streamlining and accelerating privatization so that the business prospect o f the tendered company i s the main factor in the investment decision. In mass privatized companies, frequently minority-owned by Privatization Investment Funds (PIFs), profit-seeking is constrained by the firm having to maintain excess labor. Because o f wage and contribution arrears, total severance costs exceed cash flow possibilities so that firms are constrained to maintain surplus workers even if this limits their investment and growth strategies. However, for most such enterprises, the main weakness lies in the dilution o f ownership and the lack o f corporate discipline imposed by owners. 21. Finally, market exit has been largely nonexistent. BH is a country where the concept of bankruptcy i s virtually untested, and where modem bankruptcy laws have only recently been adopted. An effective bankruptcy regime i s crucial for providing the institutional tools for corporate restructuring to restore a firm's economic viability, or to facilitate its exit. It provides an opportunity to release assets that could eventually lead to new business activities and jobs, even ifliquidation andjob losses happen insthe short term. Inmany cases, jobs inunviable SOEs and other companies were only virtual, not related to any productive activity or not giving rise to any wages or contributions being paid, but only accruing fictitious arrears in the company's books. Such companies being kept on life support at the expense o f their workers, suppliers or the taxpayers need to be fundamentally restructured or liquidated, both o f which the new bankruptcy laws can achieve. However, the use o f the bankruptcy process inBHi s still hindered vi by: (i)political reluctance, especially on the part o f the tax authorities, which have only very recently considered initiating a bankruptcy in either entity; and (ii) social pressure combined with a very negative perception o f bankruptcy. In such an environment, the effective implementation o f the new bankruptcy frameworks is entirely dependent on the political will to close down unviable companies. 22. Challenges in the labor market are numerous. Unemployment is high and rising, particularlyfor women and the young. New jobs being created are too few, and mainly in the informalsector. Participationrates remain low, even iflabor supply seems more responsive to job opportunities than previously assumed. The causes o f the lack o f flexibility inthe (formal) labor market are difficult to pin down with accuracy. Despite relatively flexible labor legislation, separation and remuneration practices show persisting signs o f rigidity, and wage increases appear to consistently outpace productivity. The relative weakness o f employers' associations in collective bargaining may explain apparently unsustainable employment practices. The problem i s compounded at the firm level by weak market discipline and corporate govemance as highlighted above. Workers are often not ready to enforce their claims on firms by initiating bankruptcy, even when it i s long overdue. Firms rarely take difficult decisions related to rightsizing the labor force and resistingautomatic wage increases, even when their survival i s at stake. The governments themselves appear to fail inmany aspects related to labor market reform. They prove to be a weak bargaining partner in labor negotiations, underestimating the signaling and spill-over effects o f public sector agreements. Instead o f focusing on core labor market intermediation, employment institutes have channeled their surplus funds into direct economic intervention. Education policies, especially for upper secondary and tertiary, are delinked from the needs o f the market. OPTIONS FORTHE WAY AHEAD 23. The BHauthorities have recognizedmost o f the challenges outlined above, as reflected in the MTDS. The authorities have also taken numerous initiatives to meet these challenges. Action plans for economic revitalization are discussed, social programs are prepared and solutions are sought to ease the financial plight o f troubled companies. Too often, however, the policies envisaged seek to increase government intervention or to achieve short-term relief by offering unsustainable subsidization o f some sort, including write-offs. In other instances, as in any country, governments are being pressured by interest groups to change policies that work or to delay reforms that could bring needed change. Repeated calls for increased protectionism are a case inpoint. As the trade chapter o fthe report concludes, there i s nothing fundamentally broken with the trade regimethat requires fixing. The real bottlenecks are elsewhere. 24: To achieve the overall policy objective o f an improved incentive environment for economic growth, this report therefore argues that policy needs to focus on the priorities identified above while the nature o f the policy stance may vary form case to case. First, in some areas, such as international trade or monetary and exchange rate policy, the policy choices have been adequate. Insuch cases, the best policy option is to stay the course. This means keeping the same policy direction, in some cases even accelerating the pace, while constantly monitoring the policy's impact. Maintaining the currency board for instance, does not mean that the authorities should not keep a close watch on the evolution o f the real exchange rate or credit growth, and vii weigh the options again in due course. At the present moment, the priority appears to be to rein ingrowth o fcredit to households. 25. The second type o f policy priorities should focus on improving the environment for the private sector to perform. It generally translates into streamlining the administrative framework surrounding private sector activity, orjust freeing up space and resources for it to grow. 26. Finally, in a few cases, active intervention i s needed. Clearly, this applies to all core areas o f the public domain, including reallocation and rationalization o f government spending.. Sometimes even a market intervention is advocated, such as a strict incomes policy for public sector companies. The remainder o f this section summarizes the report's key policy recommendations. Bank on RecentSuccesses 27. A key policy ingredientto support further growth will beto maintain the macroeconomic stance adopted over the past few years: rely on the currency board arrangement as the anchor for macroeconomic policy and support it through continued fiscal consolidation. The main vehicle for achieving this latter objective i s reinforced control o fthe government sector wage bill. Until the country establishes its creditworthiness, the authorities should maintain prudent external borrowing strategies. 28. With respect to trade policies, the report finds that all in all the current trade regime is favorable and is not the weak link o f exports. Reverting to protectionist policies would backfire, as it has inmany other countries. On the contrary, if BHis to amend its tariff schedule, it should liberalize tariffs further, and especially lower rates which remain unusually high, such as for some investment goods. Resorting to Tariff Quota Rates (TQRs), i s not likely to achieve productivity gains, even in the agro-food sector. Given the administrative complications they entail, this report recommends completely foregoing TQRs. If they are implemented for some goods, the tariff rates should be kept as low and the licensing procedure as transparent as possible. Promotethe RightIncentivesandFree up Spacefor Growth 29. One o f the most fundamental reforms on the authorities' agenda is the introduction o f a Value-Added Tax (VAT). This reform will have far-reaching implications for public finance, the business environment, the strengthening o f domestic competition and the promotion o f better govemance. The institutional arrangements for the adoption o f VAT have been put in place, starting with the establishment o f the State-level Indirect Tax Authority and fundamental pieces o f legislationregarding rates, structure, exemptions. Legislationregardingrevenue allocation and other issues needs to be adopted urgently so that the subsequent efforts can concentrate solely on buildingthe capacityo fthe tax administration and raisingthe awareness o ftaxpayers. 30. Improved incentives for trade are linked to the implementation o f key institutional reforms and the rapid establishment o f a regulatory framework. Priority policy measures should focus first and foremost on meeting the conditions for engaging in negotiations for a Stabilization and Association Agreement (SAA) as soon as possible. Advancing negotiations on the S A A will need to be accompanied by legal harmonization that will entail consolidation o f viii BH's single economic space. The SAA will also pave the way for BHto participateto the "Pan- European Cumulation of Origin." This would inturn raise BH's attractiveness as a participant in producer-driven commodity chains and to foreign direct investment more generally. Efforts should also focus on further pursuing trade facilitation policies; including streamlining o f procedures at border crossings and continuing the fight against border post corruption. 31. To provide the right incentives for new businesses to be created and to grow, the authorities will need to accelerate the implementation o f long-agreed reforms. The new business registration system must be implementedcompletely to ensure its smooth functioning. Plans to streamline business inspection, moreover, should also be implemented, to limit the regulatory burden on companies. Vigorous enforcement o f competition policy is also needed to safeguard vibrant and contestable markets. 32. Fundamental change in corporate ownership rules, as well as the rules for privatization investment funds (PIFs) is an important area o f reforms to improve incentives. The new rules should encourage PIFs to make a strategic choice: become a mutual fund or a management company. 33. Changes in privatization regulations in both entities are necessary, to allow for unbundlingand financial restructuring o f SOEs outside o f the bankruptcy regime. Inboth state- owned and privatized firms, a proper ownership structure should be in place for the effective restructuring o f the enterprise sector. For already privatized firms, the key issue i s encouraging the formation o f concentrated owners with incentives to operate in a business environment following strong corporate governance principles and profit motivation. In addition, the governance structure of all firms needs to engender incentives for managers and owners to maximize efficiency gains. Reforms that will promote such objectives include: the clear separation o f functions between management and supervisory boards; elimination o f political appointments to ensure the independence o f first-tier boards from management, employees and large shareholders; simplification o f the corporate boards systemthrough abolishing the third tier o f boards in BH, namely, the "audit board" in FBH and the "supervisory board" in the RS; harmonization o f company laws inbothEntities to establish equal rights for all shareholders; and the public disclosure o f firms' financial statements. Furthermore, adoption o f the accounting regulation standards compliant with international standards and independent extemal audits can be powerful tools for improving corporate performance, andbuildinginvestor confidence. 34. Increased labor market flexibility will be achieved, inter alia, by the government seeking to gradually reduce the tax wedge on labor, starting perhaps with a reduction in the employment insurance contribution in the FBH. Such reduction should be facilitated by broadening o f the social security contribution base, including through reducing the number o f workers exempt from making contributions. The government should actively pursue the abolition o f the automatic feedback loop between minimumand average wage inthe FBH. With respect to labor relations, the authorities should free space and play a facilitating role only for employers to become a more active partner inthe social dialogue. ix InterveneSmartly 35. While reducing overall public expenditure as a share o f GDP, the authorities should also continue to alter the composition o f spending, such as reallocating savings from the wage bill and social transfers to operations and maintenance and investment. This will require decisive action. Wage bill caps determined within medium-term fiscal targets need to be adhered to in annual budgets while salary determination rules should be overhauled and brought in line with standard Europeanpractice. 36. Within core public sector services, such as education or social protection, expenditure allocation should be drivenby efficiency and equity considerations. Inparticular, vocational and higher education financing are in urgent need o f reform, as are welfare transfers to households. Also, the planned policies to contain fiscal risk through the settlement o f the large overhang of domestic claims should be implementedas foreseen. 37. Accelerated privatization requires more active but concerted government action, Streamlining the processes by introducing a simplified and transparent auctions procedure is recommended. Similarly, the institutional and technical leadership for the privatization process should be clarified and strengthened. The Privatization Agencies could play a key role, if given more independence and protection from political interference either in the transactions themselves or inthe agencies' budgets. 38. For restructuring efforts to succeed, however, exit should be facilitated through the reduction o f subsidies and arrears, unsound firms should be liquidated, labor rigidities should be drastically reduced, and corporate governance and ownership structures should be significantly improved. These policy actions would lead to hardening o f budget constraints and result inmore effective restructuring. While bankruptcy i s largely an issue related to incentives and the business climate, the government could show the way by initiating bankruptcies for unviable companies with large tax and contribution arrears. 39. With respect to labor relations, and inline with the above mentioned reform o f the wage determination system for government workers, the government should exclude these workers from general collective agreements as it moves to a statutory regime o f employment. A strict incomes policy should be applied to key SOEs. Finally, the government should reconsider the mechanism governing the determination o f the minimum wage, especially in the FBH, and introduce a low minimum wage that would not jeopardize the competitiveness o f the country or hamperthe employment o fyouth and low-skilled workers. CONCLUSION 40. BH faces all o f the political economy challenges that other countries in transition have typically grappled with, in addition to its own. This calls for adequate sequencing o freforms and improved governance. The sequencing and pace o f reforms is essential for policies to reach their desired objectives. As in other transition economies, some o f the needed reforms will encounter resistance from various interest groups such as SOE managers, local government officials, or affected workers. In addition, given BH's recent history and its complex political structure, fundamental reforms often require difficult and protracted consensus building among the X country's constituent peoples. There are also important external imperatives that drive the reform agenda, most importantly the prospect o f entering a S A A with the European Union. All these factors will require the authorities to distinguish the urgent from the important and refrain from adding new tasks to an unfinished agenda. The Medium-TermDevelopment Strategy provides a useful framework for such sequencing, but further focus is still required. More importantly, improved governance and more effective anti-corruption policies constitute a necessary condition for gaining the support o f the broader population for reforms. In addition to helping build a broad constituency for reforms, improved governance is a fundamental determinant o f enhanced economic growth, as research by the World Bank and others demonstrates. x1 1. SUSTAININGA GROWTH-ENABLINGMACROECONOMIC ENVIRONMENT A. INTRODUCTION 1.1 Even when comparedto other post-conflicteconomies, BHhas recordedimpressive growth rates since the end of the war that broke up the former Federal Socialist Republic of Yugoslavia in the mid-1990s. Inthe immediate aftermath o fthe conflict, growth was mainly spurred by intense reconstruction and rehabilitation investment in public infrastructure and private housing. Spending has been predominantly financed by large aid inflows. Private transfers and worker remittances from abroad also played an important role in fueling domestic demand. 1.2 The positivegrowthtrack recordreflectsa reconstructionprogramthat is perceived as largely successful. Official donors disbursed an estimated US$5.1 billion in support o f the program, o f rebuildingand rehabilitating BH's economic and physical infrastructure. 1.3 While real GDP growth picked up last year after slowing in 2003, sustaining a rigorous economic expansion will become a challenge amid declining official aid and modest private investment. The trend o f decreasing aid flows is likely to continue, as the strategic focus underpinning donor commitments shifts from financing reconstruction and refugee return to supporting deeper structural and administrative reforms. BH's most pressing macroeconomic challenge, as signaled by the unsustainable current account deficit, is therefore to boost domestic savings, increase investment, especially inthe private sector, and attract more foreign savings inthe form of direct investment. Moreover, despite the generally positive growth trend, about one-fourth o f the population is estimated to live below the poverty line, while another third is vulnerable to fall into poverty in case o f an income shock. Unemployment, although overstated by administrative data, is very high, especially amongst women and youth. The informal economy is also pervasive and spreading. 1.4 Macroeconomic policies have been geared to consolidating BH's stability. BH has anchored its price and exchange rate stability in a strict adherence to the currency board regime. Fiscal policies have been increasingly supportive o f stabilization, showing a dramatic improvement in government savings over the past few years. External debt policies have successfully focused on reducing the debt burden o f BH through rescheduling and cancellation of pre-war debt, while new borrowing was sought only on very concessional terms. 1.5 While questions have been raised as to whether stabilization-oriented macroeconomic policies excessively constrain BH's potential, this chapter argues that continued focus on preserving stability is the best option for promoting competitiveness and sustainable growth in the medium term. In particular, given the country's unique characteristics, the currency board remains the most adequate monetary and exchange rate arrangement for BH. Fiscal policies should aim at further expenditure rationalization and reorganization to achieve greater efficiency in core public services. Efforts to boost public investment inorder to compensate for the shortfall inexternal project financing are justified, but selectivity and thorough economic and financial analysis should be the main drivers o f public investment decisions. On the revenue side, tax reform will focus on the introduction o f a value- added tax, a fundamental change expected not only to stimulate and stabilize fiscal revenues, but also to foster an improved competitive environment inthe domestic economy. More generally, a major challenge lying ahead i s buildinga self-sustained "fiscal architecture", where hndamental policy decisions, such as the target for consolidated budgetary balance, can be taken in a coordinated and timely fashion. 1.6 Rigiditiesconstrainingthe potentialof BH's economy are best addressedthrough an acceleration of key structural reforms, including through improving the business environmentand increasingthe flexibility of labor markets.Assumingthe reformprogram is implemented at a satisfactory pace, barring the occurrence o f external shocks and assuming continued political stability, such policies could help sustain real GDP growth rates o f about 5 percent. These developments should also reflect significant growth o f exports, relatively high levels o f foreign direct investment, and a gradual increase in domestic savings and investment. As a consequence, the vulnerability relatedto the current account gap would be lessened. Onthe contrary, delayed reforms would result in lower growth in output and exports, as well as a persistently high current account deficit that would leave the economy vulnerable to an even sharper downtum. 1.7 The remainder o f the chapter is organized as follows. The next section provides an overview of economic trends since the end o f the war and highlights some o f the more salient recent developments. Section C reviews the track record o f macroeconomic policies and structural reforms and provides recommendations as to the future policy stance. A quantified macroeconomic outlook i s presented inSection D. Section E concludes. 1.8 As a note of caution, much of the analysis presented inthis chapter relies on a very weak statistical base. The data are constructed from various sources that are not always consistent and are subject to frequent and sizeable revisions. Moreover the statistical system in BH remains fragmented and urgent institutional reforms are needed to improve the quality o f economic and social data (Box 1.1) describes the main challenges o f BH's statistical system and some o f the reforms currently envisaged to address them. B.POST-CONFLICTECONOMIC TRENDS RECENTDEVELOPMENTS AND 1.9 During the first half of the 199Os, BH experienced the most devastating economic collapse of any economy in Central and Eastern Europe since World War 11. Economic activity fell to less than 20 percent of its pre-war levels (Table 1.1). The war altered the social and economic map of the country and took a heavy toll on its population and physical structure. Out o f a pre-war population of 4.4 million, 250,000 people died or were considered missing by the end o f the war. More than 60 percent of the population had been internally displaced, while more than one million people left the country. 2 Box 1:l Statistics inBH: Institutional andMethodologicalIssues A s a part o f the former Socialist Federal Republic o f Yugoslavia, Bosnia and Herzegovina inherited a statistical system that resembled that o f centrally-planned economies, including the Soviet-style Material Production System (MPS) methodology to compute aggregate production. During the war, the system was not operational and data was lost, which explains the lack o f any statistics for the period 1992-1995. At the end o f the conflict, the Federation Institute for Statistics (FIS) took over most o f the staff and buildings form the former statistical agency, while the RS established its owned Institute for Statistics (RSIS) from scratch. This ledto the application of different data collection and processing methodologies: while the FIS attempted to introduce the standard international methodologies, the RSIS kept applying pre-war methodologies. A s a result, the system was not capable producing any country-wide data for a prolongedperiod. A State-level Agency for Statistics (BHAS) was established in 1998, with the strong support o f the international community, and with the coordination o f statistical activities as its primary mandate. Because o f the lack o f any enforcement powers, the B H A S failed until recently to reform the system, let alone to produce reliable consolidated statistics. The current challenge facing BH statistical system is twofold: institutional and technical. Only once these issues are resolving can one expect production of reliable State-level statistical data, including the data o nprices and national accounts. Institutional issues arise mainly from the lack o f political commitment to establish a single, harmonized statistical system. Following intense pressure from the international community, a State-level L a w o n Statistics was adopted inearly 2003, andthe existing Entity-level Laws have beenharmonized withit. The Law provides for mandatory data reporting, cooperation and methodological harmonization requirements for the Entities. K e y challenges at present include the need to adopt a work program that focuses on, inter alia, improving national accounts statistics. T o start the work, various working groups have been established, with support o f the European Union, to resolve outstanding technical issues. Indeed, the technical challenge is important. The only national accounts data available, also used inthis CEM, are a production-side GDP o f dubious quality. The Statistical Institutes are using firm level financial statements, which introduces serious flaws, since the reports do not provide any information o n the nature and respective share of registered activities (e.g., manufacturing, trade etc). On the expenditure side, consumption data is not yet available. A Household Budget Survey (HBS), currently being implemented with support o f Italian Statistical Institute, and scheduled to be completed by end-2004, should provide reliable data to that end, as well as foI determination o f consumer price weights (the SIs are currently using pre-conflict weights which have been adjusted on an ad hoc basis). Another problem, reflecting the lack o f property ownership data, i s failure to include imputed rents as a part o f consumption. Investment data collection methodology has recently been harmonized. with data collection forms having been developed jointly by the three institutions, which should lead to more accurate data from the enterprise sector. However, the measurement of household investment remains to be addressed. For Government consumption, the system is relying o n the Central Bank data, which can be regarded as satisfactory. Foreign trade statistics are o f poor quality (Chapter 2) and will improve only with the nem information system to be installed at the Customs Administration, including a centralized, state-level custom database. Needless to say, there i s no methodology for informal sector estimates (although the Central Bank produces some statistics). Finally, the lack o f any reliable population data (the last census was carried out before the war!) raises concerns about the quality o f any per-capita indicator. In summary, the formulation o f economic and social polices in BH suffers from a seriously flawed statistica system. Policy actions in this area are very urgent, so that the available technical assistance can be effectivelj mobilized to address capacity shortages. The European Union has appropriately recognized the importance of thiz issue, and made progress in the implementation o f the Statistical L a w one o f its core prerequisites for opening the negotiations on a Stabilization and Association Agreement. 3 1.10 By 1996, the return of peace, coupledwith social and politicalstability and massive internationalassistance, brought about the resumptionof economic growth. Since then, the country has made significant strides in re-building its economy. The transportation, communication and power networks have been rehabilitated while, in parallel, the institutional infrastructure has been revamped, although progress on this front has been much slower than expected. After the war the country established a national currency, the convertible mark (KM), and an independent central bank. In parallel, BHmanaged to reduce a large portion o f its share o f external debt inherited from the former Socialist Federal Republic o f Yugoslavia (SFRY). Table 1:l Comparative OutputDecline Cumulative output decline RealGDP2002 (percentof 1989) (1989=100) Bosnia and Herzegovina 83 76 Croatia 41 87 CzechRepublic 13 105 Hungary 18 112 Slovak Republic 25 111 Albania 34 114 Bulgaria 27 80 Serbia and Montenegro 59 50 FYRMacedonia 32 75 Romania 25 87 SouthEastem Europe 30 81 Central Eastern Europe & the Baltic states 18 115 Source: For years 1990-2001 EBRD(2002); for 2002-2003 own estimates. 1.11 Between 1995 and 2004, real GDP quadrupled, as exports rose tenfold and consumer price inflation stabilizedto industrial country levels. The reconstruction-induced activity contributed, inparallel, to reducing poverty, with real per capita consumption increasing for most households. 1.12 A pickup in exports helpedstrengthen real GDP growthto 5.7 percentin 2004 from 4 percent in 2003. Sustaining such a pace going forward will be challenging however, amid declining official flows and modest reform efforts. Growth inboth Entities has taken somewhat different paths, with the RS growth lagging during most o f the post-conflict period. This could be explained inpart by delayed access for the RS to foreign aid, as well as by more pronounced output losses in the agricultural and manufacturing sectors. More recently, Entity growth rates seem to have started to converge, signaling a better integration o f the domestic economy.' 'In addition to the caveats related to the weak statistical base, interpretation o f GDP inthe RS i s complicated by the fact that since 2002, the RS Statistical offices has made efforts to improve the coverage o f its data collection. The growth rates pictured inFigure 1.2 have been corrected for this structural break. 4 Figure 1:l Real BHGDP and ForeignAid Figure 1:2 Real GDP Growth by Entity (annualpercent growth;USD m) (annual inpercent) ,$ ,.9" ,.9" ++ 8%8%+$ 8 -Real GDP growth (%, left) +Foreign Aid (US$ m, right) OFBH mRS Source: IMF, own calculations. Source: IMF; BH, FBH, and RS Statistical Offices. 1.13 In addition to the vast humanitarian and peace-keeping assistance provided immediately after the war, the international community disbursed more than US$5.1billion through the end of 2004, or roughly 13 percent of GDP a year, mostly for reconstruction and institution building projects. Foreign aid spurred economic activity by supporting the vast reconstruction program that was fundamental in rehabilitating public infrastructure and rebuildingprivate housing. Grants accounted for three-fourths o f the total aid inflows. Aid inflows declined markedly in 1998 and then again in 2000, as donors were facing competing demands from other countries under stress while their own budgets came under pressure. Strategically, most international partners also felt that the reconstruction objectives had generally been achieved and that new aid should increasingly focus on technical assistance to support the economic transition. The FBH has received about five-sixths o f this aid, a substantially larger share relative to its population than the RS, mainly because political reasons delayed the start o f aid to the RS. 1.14 The compositionof GDP appears to have changedmarkedlyin the post-war period. According to official statistics, the share of agriculture has declined significantly while the share o f services has increased (Figure 1:3 and Figure 1:4). While the overall trends follow a similar direction in both entities, notable differences remain. The primary sector (agriculture, fisheries and forestry) has remained an important sector inthe RS (20 percent o f GDP) but dropped to less than 10 percent in the FBH. The secondary sector (industry and mining) maintained its share o f about one quarter of Entity GDP in the FBH. Deindustrialization was more marked in the RS. Industry accounts for only 16 percent o f GDP, with a third o f industrial value-added generated by electricity generation. In the services sector, while financial and business services have grown steadily, the government still retains a dominant position, especially in the FBH. This reflects both the more complex administrative structure and higher civil service salaries than in the RS. 1.15 Sectoral contributionsto real GDP growth reflect the increasing significanceof the services sector (Figure 15). Most recently, the decline in the contributions o f agriculture (RS) and industry (both entities) are particularly striking. Ofcourse these statistics do not fully capture 5 the informal economy, which i s believed to be one o f the largest in the region. Informal sector activity i s mainly concentrated in agriculture and trade and i s estimated to be growing. The importance o f informal activities in the enterprise sector and labor market i s discussed in detail inChapters 3 and 5. Figure1:3 FBH: Sector Compositionof GDP Figure 1:4 RS: Sector Compositionof GDP (in percentof nominalGDP) (in percentof nominalGDP) I 100% 100% 80% 80% 60% 60% 40% I 40% 20% I 20% 0% 0% 1995 1996 1997 1998 1999 2000 2001 2002 2003 1 1 1995 1996 1997 1998 1999 2000 2001 2002 2003 DAgnculture Industry Construction I I DGov services lcomm services 0Gov services ElCon" services I Source: FBH, BHStatistical Offices. Source: RS, BHStatistical Offices. Figure 1:5 Sector Contributionof NominalGDP(in percent) 30 0% 25 0% 20 0% 15 0% 10 0% 5 0% 0 0% 2000-03 1996-99 2000-03 1 FBH RS I i OAgriculture l!JIndustry WConstruction ElServices Source: Statistical Offices. 1.16 Strict adherenceto the currencyboardarrangementintroducedin 1997 helpedslow consumer price inflation to single digits and below one percent during 2002-2004. The inflation differential between the two Entities has also been reduced, signaling improved integration o f the two sub-national economies. Despite persistently large trade and current account deficits, foreign exchange reserves have increased and credibility in the domestic currency has risen markedly. Confidence inthe currency was demonstrated in 2001, when euro coins and notes were introduced. A large volume o f hoarded deutschmarks, the K M ' s previous anchor currency, was exchanged by citizens for KM-denominated bank deposits, a testimony also o f regained confidence in a reformed banking system. Due to this reverse currency substitution, non-government deposits grew significantly between throughout the period, 6 especially between 2000 and 2002 (from 6.5 percent o f GDP to 9.5 percent), fueling a credit boom in 2002, when credits to the household sector more than doubled. Successful measures to tighten credit policy were taken in 2003 in the framework o f the IMF stand-by arrangement (SBA) and in 2004. Perverse indexation mechanisms, however, have allowed accrued and paid wage growth to exceed productivity, with an adverse impact on employment and corporate profitability inboth Entities. Figure 1:6 Industrial Production,BothEntities Figure 1:7 Consumer PriceInflation (1999=100) Source: National statistical offices, various publications. Source: National statistical offices, various publications. 1.17 The large external current account deficit remains a key macroeconomic challenge. Predictably, post-conflict growth was accompanied inthe initial phase by large external deficits. Large imports fueled by reconstruction were only partially offset by exports, which, although growing quite rapidly, started from a very low base (Figure 1:s). While exports have grown faster than merchandise imports in euro terms since 2002, they still account for only 10 percent of GDP. The composition o f merchandise imports, equivalent to one-third o f GDP, has more recently shifted towards non-reconstruction goods, largely consumer goods and oil products. Domestic demand has been bolstered by sizeable increases in credit to households as well as by continuedhighlevels of workers' remittances and other private transfers (Figure 1:9). 1.18 As a result, the recorded current account deficit ballooned to around 22 percent of GDP in 2002. This was financed by aid flows, foreign direct investment, a small drawdown on reserves, as well as non-identified non-debt creating capital flows. The latter are mainly believed to have reflected private capital transfers.* In 2003 and 2004, tighter fiscal policy and banking supervision appears to have restrained import growth leading to a slight easing in the current account deficit to about 17 percent of GDP. Encouragingly, foreign direct investment increasingly represents an alternative to aid in financing the current account deficit, signaling increased international investor interest in response to continued macroeconomic stability. Net FDI inflows have posted consecutive increases since 2002, reaching some 5 percent of GDP in * The CBBH presents these inflows believed to finance housing investment in its capital balance while the IMF- reported balance o f payments shows a sizeable errors and omissions item (about 10 percent o f GDP in2002). 7 2004, allowing BH to begin catching up somewhat with the most attractive countries in the region (Figure 1:10 and Table 1:1). Due to increased inflows of FDI, sustained aid flows, and large, often unrecorded, non-debt capital inflows (such as private capital transfers), foreign exchange reserves have increasedsteadily (Figure 1:11). Figure 1:8 Trend in MerchandiseTrade Figure 1:9 Remittancesand Private Transfers (In euro, 1998=100) (In million euro, left scale; in percent of GDP, right scale) 2001 2002 2003 2004 1998 1999 2000 2001 2002 2003 2004 1 -Exports, IIRemittances (Eur m) EElPnvate transfers (Eur m) Total in %of GDP f.0.b. Imports,f.0.b. Reconstructionimports,f.0.b. ~ Source: IMF balance o f payments statistics, own Source: IMF balance of payments statistics, own calculations. calculations. Figure 1:lO Gross FDIInflowsin BHand Countri s Figure 1:11Gross Official Reserv s (In monthsof D of the Region(In percentof GDP) importsof goods and services) I 1998 1999 2000 2001 2002 2003 2004 I 0 1996 2002 I2003or latest(2002) - -Gross Official Reserves(Monthsof M of G8S) Source: SIMA database,,own calculations. Source: IMF. 8 Table 1:l BH Key Economic Indicators,1999-2004 - 1999 2000 2001 2002 2003 2004 (annual percentage change) Real GDP growth 9.6 5.5 4.3 5.3 4.0 5.7 Industrial productionindex 11.9 9.3 -2.0 11.5 3.0 12.4 CPI Inflation 2.8 5.0 3.2 0.3 0.6 0.4 Gross wages 11.4 12.9 6.8 6.9 8.2 2.8 (inpercent ofGDP unless otherwise noted) Gross Investment 21.0 20.6 18.8 20.4 19.9 20.3 National Savings 12.8 8.1 2.2 -1.3 2.2 3.0 General Govemment Revenues 59.9 55.4 53.8 49.7 49.6 48.9 General Govemment Spending 67.7 64.5 58.8 53.4 51.0 50.0 oiw Capital Expenditures 17.5 14.9 14.7 9.3 8.1 6.4 Balance (accruals, incl. grants) -7.8 -9.0 -4.9 -3.7 -1.3 -1.1 Exports of Goods and Services (In 1285 1309 1354 1808 223 1 millions of USD) Imports of Goods and Services (In 2749 2961 3418 4025 4716 millions of USD) Current account balance -9.3 -12.5 -16.6 -21.7 -17.7 -17.3 Public extemal debt " 65.5 41.6 39.8 37.7 33.3 31.1 Extemal debt service (Inpercent 10.1 7.5 6.1 9.2 8.2 6.5 o f exports of goods and services) Foreign exchange reserves (In 1.9 2.1 5.1 4.5 5.3 6.1 months of imports) Source: NHauthorities; IMF, Fourth SBA review, February, 2004; Article IV Staff report, January, 2004; Bank staff estimates. Note: l/Excludes the "performance amount" o f the London Club equivalent to some US$280 million.. 1.19 While per capita GDP has increased,the living standards of many Bosnianfamilies remain low. Despite the rapid economic recovery, reflecting the extremely depressed level o f post-war economic activity, BH's GDP i s estimated to be at about three-quarters o f its prewar level. Not surprisingly, BH remains one o f the poorest countries in South East Europe (SEE). According to the World Bank's poverty assessment, one-fifth o f Bosnian households live in poverty, (World Bank, 2003a). The survey found, however, that severe poverty is not prevalent. While these findings might conclude that the post-conflict recovery effort was relatively successful inmitigating pervasive poverty, further analysis highlights the vulnerability o f a large share of the population. In addition to those living below the poverty line, another 30 percent o f thepopulation is estimated to beinrisk o f falling below the poverty line (Figure 1:12). Nearly 70 percent o f the population i s poor inat least one dimension o f poverty. Poverty is higher inthe RS (25 percent) than inthe FBH(16 percent). Interms on inequality, BH seems to share a common feature o f SEE and CIS countries, Le., a large gap between income and consumption inequality (Figure 1:13). Given measurement problems related to income data, the consumption-based equality measure i s more likely to provide an adequate picture. There, the situation in BH appears to be comparable to that inother countries o f SEE. 9 Figure 1:12 Distribution o f the Population (level Figure 1:13 Inequality in Selected Transition of per capita consumption) countries (Gini coefficient) 4 97711 - ,1 2 3 4 5 6I 7 8 9 1 Share of populat poor-vch IIncowbased0Consunptlon-based Per capita consumptionand poverty line +/-50% Source: World Bank (2003). Source: World Bank (2003). The Track Recordin StructuralandInstitutionalReforms 1.20 Since the end of the war, progress in structural and institutionalreforms has been most pronounced in the financial sector, trade policy and in public finance. In fact, the EBRD 2003 Transition Report identifies BH as the country with the fastest progress in the Region over the past few years. Despite substantial progress, however, BH still lags behind, especially insecond generation reforms (Figure 1:16). 1.21 Anchored in the currency board, financial sector reformshave focused sequentially on the dismantling of the payment bureaus and the establishment of a modern payment system, the state divestiture from the banking sector and the successful entry of strategic investors,the consolidationof the bankingsector at low fiscal costs, and the establishment and gradual strengthening of the prudential and regulatory framework. Minimum capital requirements are being strictly enforced and the prerequisites for participation in the deposit insurance system may leadto further consolidation o f the sector. 1.22 The impact of these reforms is sometimes questioned, with concerns being voiced about the actualcompetitivenature of the bankingmarket, as well as the lendingbehavior of financialinstitutions,especiallywith respect to the corporatesector.Chapter 4 reviews in detail some o f the most important factors affecting firms' access to finance, and argues that access has improved and that the pace o f lending adequately reflects the risks associated with investment projects. As for the competitive structure o f the market, Figure 1:14 demonstrates that the interest rate spread narrowed sharply in recent years, but remain high by regional standards. Real interest rates have also come down grad~ally.~ Real lendingrates of about 9.9 percent in2004, were lower than inFYR Macedonia and only slightly higher than in Croatia. To be more accurate, the lending rate should be deflated by long term inflation expectations. Using a projectedrate of about2percent, would yield real lendingrates of about 7.5 percent. 10 1.23 Trade policy reforms have been very pronounced with the adoption of the most liberal trade regime in the sub-region and negotiation of bilateral trade agreements in South Eastern Europe in line with Stability Pact principles. Chapter 2 provides a comprehensive overview o f BH's trade regime. With respect to public finance, the introduction of a modem treasury system at all levels o f government (except municipalities) has contributed to improved spending discipline. Significant progress was also achieved in tax policy and administration, including countrywide harmonization and improved collection enforcement. Figure 1:14 Lending-deposit spread, selected Figure 1:15 BHRealinterest rates countries (in percent) (percent, deflatedwith sameyear CPI inflation) 30 00 I 25 00 20 00 I1500 I 1000 I I 500 I 0 00 2000 2001 2002 2003 2004 I ~ 0Deposit Lending 1.24 The disappointing supply response to the favorable trade regime indicates that structural reforms have yet to have a sizable effect in the enterprise sector and labor markets. While an important mass privatization o f small-scale businesses was undertaken shortly after the war, it resulted, as in other transition countries, in diluted ownership and weak corporate governance. The privatization o f larger state-owned enterprises has lagged, however, with BH initially being an involuntary slow starter because o f the conflict, and later losing ground due to lukewarm political commitment. 1.25 The business environment, while modestly improving in recent years, remains characterized by high barriers to market entry for new businesses, proliferation of bureaucratic interference and virtually no formal market exit. Chapter 4 elaborates on these fundamental areas o f reforms for BH's future growth. The Medium-Term Development Strategy (MTDS) calls for an acceleration o f privatization and improvement in the business and investment climate. Specific measures currently under preparation or initial implementation include reforms o f business registration, the establishment o f a single registry for movable pledges, and streamlining o f business inspection. Importantly, harmonized bankruptcy legislation was recently adopted inboth Entities. 1.26 Finally, labor market rigidities have persisted despite the important changes in labor legislation introduced in 2000, which laid the ground for substantially greater flexibility in labor relations. This included the facilitation o f more varied and flexible forms of employment, the simplification o f procedures and entitlements related to hiring and dismissal of workers, and the formal termination o f the "wait listing" process inherited from the former 11 SFRY. The pervasiveness o f collective labor agreements and even firm-level policies o f inflexible wage determination remain a substantial rigidity, increasing firms' wage costs, hampering the proper functioning o f labor markets and stimulating informalization o f the economy. Wage inflation has consistently surpassed productivity growth, with detrimental effects to competitiveness. Recent trends in the labor market and policy implications are discussed inChapter 5. The Key MacroeconomicChallenge:IncreaseDomesticSavings 1.27 The rapid decline in foreign assistance has so far not brought the current account deficit down to sustainable levels. Despite narrowing o f late, the external shortfall amounted to 17 percent o f GDP in 2004. The decline in reconstruction imports has been largely offset by a pickup in other imports, financed via remittances and drawdowns o f assets held abroad. These imports include mainly constructionmaterials, household appliances and furniture and consumer goods. Receipts from services exports have picked up modestly, as have remittances from BH workers abroad. A decline in gross domestic savings appears to be the main contributor to the increase in the current account deficit through 2003 (Figure 1:18). The ratio of total investment to the current account deficit i s much lower than in the most rapidly growing transition economies (below 2 for BH between 2000 and 2003 and 4-8 for most new EU member states). This highlightsthe key macroeconomic objective to increase domestic saving rates, bothinthe public and private sector. 1.28 Macroeconomicpolicies to help narrow the current account deficit should combine with continued fiscal consolidation and an acceleration of structural reforms to boost exports and enterprise performance. As Table 1:l shows, the government has moved to reduce the fiscal deficit, but more could be done given growing pressures on fiscal resources. The macroeconomic framework presented in Section D anticipates further gradual increases in public savings to about 4 percent o f GDP by 2007. Within the budget, the challenge ahead i s to consolidate the achievements in fiscal moderation by achieving overall surpluses while expandingself-financed public investment inline with further gains insavings. 1.29 Available national accounts data do not allow for the disaggregation o f private savings between the household and enterprise sectors. However, the analysis o f firm-level surveys presentedinChapters 3 and 4 suggests that corporateperformance,while still lackluster, has improved,andthat the deteriorationof the savingrateover the past severalyears maywell be explained by lower household savings rates. In the short-term, reining in household consumption growth will be needed via a further tightening of credit policies, within the framework o f the currency board. As an important share o f household spending on imports is estimated to be financed through remittances, private transfers and other non-debt creating private capital inflows (such as draw down in savings held abroad), a drying up o f these financing sources will lead to an automatic adjustment in imports. In addition, in line with the recommendations stemming from the labor market analysis (Chapter 5), the government should consider selective use o f incomes policy, notably to cap wages inthe public sector. 12 Figure 1:16 EBRD TransitionIndicators-BH, Selected SEE, CEE countries EBRD IndexRangingform l(worst) to 4+ (best) PriceLiberalization Trade and ForeignExchange System Small-scale Privatization Large-scale Privatization 5 I 4 3 I ~ 2 .2m1 02003 1 0 Governanceand EnterpriseRestructuring CompetitionPolicy Source: EBRD(2003). 13 Figure1:17 Composition of the Current Figure1:18 RecentTrends inSavingsand Account balance(in percentof GDP) Investment (in percentof GDP) 20. , I '0° 20 0 10 0 0 0 -100 ,-20 0 - - - ~ ~ I -GoodsI 0FactorIncome0Transfers-CAB SeMces :BTotal Investment BNationalSavings CII Domestic Sawngs Source; IMFbalance of payments statistics, own Source: Own calculations. calculations. 1.30 However, the turnaround in private savings will be gradual, as it has been in most transition economies. The macroeconomic framework elaborated in Section D o f this Chapter foresees a recovery to 2.7 percent by 2007. Even though modest on the surface, such a turnaround in savings, and especially in corporate performance, would still require immediate implementation of priority reforms in the business sector, as laid out in Chapter 4. The establishment o f an effective bankruptcy process, the reduction o f administrative bamers to the operation o f firms and further strengthening o f the single economic space within BH will be hndamental to help reinvigorate enterprise profitability. Restoring profitability will also require reforming the labor market to make it capable o f flexibly reallocating the existing work force to the new demands while pricing existing employment in line with productivity (Chapter 5). The next section provides and overview o f the macroeconomic policy environment best suited to accompany the required structuralreforms. c.MACROECONOMIC POLICIES -TRACKRECORD AND OPTIONS FORTHE FUTURE Monetary and Exchange Rate Policy: The Case for Maintainingthe Currency Board 1.31 The currency board arrangement was established by the L a w establishing the CBBHinAugust 1997. The Currency Board Arrangement (CBA) can be considered as a "pure" CBA, i.e., one where there i s a legal commitment to exchange domestic currency against an anchor foreign currency at a fixed rate and where domestic currency is issued against foreign assets and remain fully backed by foreign asset^.^ As such, the C B A sets a fixed exchange rate peg and a full 100 percent backing o f the central bank's KM liabilities with foreign exchange reserves.5 For a review o f the main features of CBAs see Baliiio and Enoch (1997). Article 2 of the Law establishing the CBBH provides that the CBBH issue "domestic currency at a one to one exchange rate with the Deutschmark (DEM) with full backing in freely convertible foreign exchange." In strict compliance with this provision, the conversion rate o f the DEMto the euro (Le., 1.95583) was retained for the KM. 14 1.32 The choice of the CBA as an exchange and monetary regime has been well justified in the quest for institutionaland financial stability in BH's post-conflictenvironment. An almost exclusively rules-based regime was also best tailored for BH's peculiar political and administrative set-up where decision-making remains the outcome o f complex interactions and compromises. The CBA has been strictly adhered to (Table 1:2), except in the first year, as transactions o f the former National Bank of BH were still conducted outside the purview o f the CBBHwhile the newly established CBBHwas operatingwith minimalinitial capital.6 1.33 To date, and when measured against the objectives assigned to it, the CBA should be regarded as a major success. Inflation was brought down to industrial country levels. The KMestablished itself as a credible currency and stood the test ofthe introduction ofthe euro as noted above. Less tangible achievements include a broad public support for the regime, probably because o f its simplicity, as evidenced by opinion survey data. Finally, the CBA has allowed the CBBH to establish itself as a credible, perhaps the most credible, public institution in BH, allowing it to assume typical central banking functions with respect to financial market and banking supervision or the operationo f the payment ~ y s t e m . ~ 1.34 The CBA by definition constrains monetary and credit policies, but voices calling for more flexibility have emergedin BH, as they have in other CBA countries. Shortfalls o f CBAs are well researched and documented. They include the difficulty to respond to exchange rate misalignment. In the case o f BH, the real effective exchange rate has shown no signs o f appreciation, as measured relative to consumer prices, whether or not the Serbian dinar i s included in the basket (Figure 1:19). Relative to the Serbian dinar alone, the KM appreciated significantly in real terms during the first half o f 2000, which was then rapidly offset by mid- 2001. BH firms, especially in the RS, may have been affected by the temporary loss o f competitiveness. While deflating the exchange rate with unit labor costs provides a more accurate picture o f competitiveness, data weaknesses do not allow this analysis to be carried out. 1.35 Another argument used against CBAs relates to the loss o f the central bank function and in particular the impossibility to use monetary policy as an "policy instrument for economic development."' The domestic credit trends shown in Table 1:l signal that the C B A has not been a constraint to credit growth inBHto date. Onthe contrary, as highlighted earlier, the C B B H has had to revert to monetary policy instruments available under the CBA, such as reserve requirements or foreign exchange exposure rules, to moderate credit growth. 1.36 In summary, while developments of competitiveness need to be closely monitored, barringthe occurrence of a major external shock, the CBA remains the best policy option over the short and medium term. The hrther need to establish a sustainable institutional structure, with international "political" supervision phased out and the reform agenda driven increasingly by the process o f Stabilization and Association with the EU, will require fundamental reform, testing the capacity o f policymakers. Keeping a rules-based monetary policy in such a context i s an advantage. More importantly, fiscal discipline needs to be '*The See Nicholl(2003) for an account o f the CBA's early experience. L a w grants the CBBH a large degree o f independence. This argument is relatively commonly used in some political statements in BH. It has also been put forward in other countries with a CBA, such as Bulgaria (see Vincelette, 2004). 15 sustained (see below) and, as a result, commitment to the C B A provides an incentive to stay the course. As for other transition countries with a CBA that have engaged in the EU integration process (e.g., ,Bulgaria, Estonia and Lithuania), the question of an exit strategy will eventually be posed in terms of the adoption of the euro as legal tender. An unwarranted devaluation would automatically deplete the credibility capital accumulated to date, and would likely result in a sharp output contraction. Conversely, the C B A can only be sustained if it is supported by adequate fiscal policy. Table 1:2 CBBHreserves and liabilities Figure 1:19 Recent trends in RealEffective ExchangeRates (in KMm) (January 1999= 100, increase=appreciation) Foreign Monetary Assets liabilities 110 00 105 00 1997 144.1 160.3 100 00 1998 283.3 253.9 1999 865.7 836.7 95 00 2000 1021.4 973.2 90 00 2001 2696.5 2591.6 85 00 2002 2464.4 2345.2 9%9' 9%0' 0' Q' Q' Qz Qz QzQ3 Q3 0' 2003 2780.6 2626.3 46 5@'46 +*< 9 +*<0' 50Q ' 6 . $ <5 9Q'+ @ Q\' & $ &5% + ,6+<:eQ, a P+<;eQ'Qb aQbaOb 2004 3457.5 3283.5 -REER (excludes Serbian Dlnar) Source: CBBHQuarterly Bulletin. Source: IMF. The Fiscal Policy Reform Agenda: Consolidation, Reorganization and Mitigation of Fiscal Risk 1.37 Output recovery in the first phase after the end of the war was accompanied by large fiscal deficits. In addition to donor-funded public investment programs fueling public expenditure, recovery in domestic revenue collection triggered parallel increases in recurrent spending, especially inpublic sector wages. As shown inTable 1:1, fiscal stabilization efforts led to the elimination of the fiscal deficit, which peaked at near 8 percent of GDP in 1999. The preliminaryconsolidated fiscal accounts posted a small deficit of 0.6 o f GDP in2004 (accruals, including grants but excluding one-off bank recapitalization outlays that amounted to 0.5 percent o f GDP). Both Entities recorded negligibledeficits in2004. 1.38 The reduction in total spending from close to 70 percent of GDP in 1999 to 50 percent in 2004 was significant, with most of the adjustment due to a decline in externally- financed public investment and a shift from arrear accumulation to repayment. The reduction also reflected a cut ininterest outlays to 0.7 percent o f GDP. 1.39 The M T D S recognizes that further reduction of spending relative to GDP is a key to sustaining macroeconomic and financial sustainability. The corresponding medium-term target in terms of a primary surplus (excluding all grants and interest payments) that would be compatible with sustainable levels o f indebtedness i s estimated at around 1percent of GDP. This 16 can only be achieved through further expenditure rationalization because o f the need to decrease the prohibitive tax and social security contributions burden. In addition, new expenditure pressure could arise, even on a temporary basis, in relation to possible downsizing costs in the public sector, the implementation o f the domestic claims settlement plan, the implementation o f social mitigation measures to accompany corporate restructuring, or further institutional strengthening o f the state inthe process o fEuropeanintegration. Figure1:20 Fiscaltrends inthe FBH Figure1:21 Fiscaltrends inthe RS (inpercentofEntityGDP) (inpercent ofEntityGDP) =Revenues (incl Grants) m v e n u e s (lnd Grants) BExpendRures on a cOmmRment basis OBalance on a cnmmitment basis ,OBalance on a commitment basis Source: IMF; own calculations. Source: IMF; own calculations. 1.40 Giventhat overall investment will need to remain robust (around 20 percent o f GDP) to sustain real GDP growth, the declinein externallyfinanced publicinvestmentwill haveto be at leastpartly offsetby highercapitalspendingout of domestic resources.Similarly, inorder for this investment (as well as past capital spending) to achieve the expected economic rates o f return, expenditure on operations and maintenance would also have to be raised. Finally, even though tax reforms and improved tax administration are expected to yield additional revenues, these gains should, in the medium term, be primarily used to reduce tax rates or social security contribution rates. 1.41 Consequently, the bulk of the expenditure adjustment will have to be borne by a reduction (relativeto GDP) in other areas of recurrentspending, mainly the wage bill. The fragmented structure o f public administration, compounded by weak vertical coordination, has led to both inadequate staffing and duplication o f functions. Patronage in appointments and the absence o f effective systems for civil service training and development have also contributed to uneven levels o f professionalism in the administration. Structural and organizational problems, low civil service capacity, and the absence o f analytical information all adversely affect the formulation and implementation o fpolicies. 1.42 While the number of government employees relative to populationis not out of line with regionalaverages, higher wage rates make the public sector wage billunaffordable.In the Federation o f Bosnia and Herzegovina (FBH), public sector employees account for 20 percent o f the formal sector workforce, while the public sector wage bill represents close to 30 percent o f total spending and 16 percent o f Entity GDP. In Central and Eastern Europe, by 17 comparison, the wage bill averages about 17 percent o f total spending and only 7 percent o f GDP. 1.43 Inresponse to this situation, the authorities adopted in2003 a broad policy platform for public administration reform (PAR), which lays out the steps to the formulation o f a comprehensive PAR strategy. To inform the formulation o f the PAR strategy, the authorities have launched with EU support a series o f functional reviews. In addition, among the initial policy steps already taken, the state and Entity parliaments have approved new civil service laws that enshrine the principles o f political independence, transparency, professionalism, impartiality and effectiveness. Moreover, these laws clearly define the scope o f the civil service, mandate competitive recruitment, and introduce merit-based promotion. Although the laws also regulate civil service pay, they have thus far failed to introduce real reforms and merely confirmed the previously prevailing system. 1.44 The public sector pay system is characterized by rigid formulas driving the determination of a base wage which is then supplementedby a host o f allowances. General administration employees, as other categories o f public sector personnel (e.g. teachers, judiciary staff, and police), are also covered by collective agreements which further establish wage determination rules. While the principles o f the systems are the same throughout the country, the parameters applied vary widely from one level o f government to the other. At the level o f the State, some recently established agencies have developed their own salary structure without reference to the current wage rates in general administration. All in all, the wage structure i s complex, disparate, and generally compressed, not lending itself to reward performance or managerial responsibility. Research by the World Bank has shown that average (paid) wages in the public administration are higher than inmost other sectors (World Bank, 2002b). Figure 1:22 Wage Billin Selected SEE Figure 1:23 Ratio of Wage Billto Countries (in percentof total spending) Operations and Maintenance o 1996-2001I2002 Source: IMF: own calculations. Source: IMF; own calculations. 1.45 Wider public administration and civil service reform measures will only succeed if the issue of public sector pay is addressed. The RS authorities have designed a grading system for public wages and drafted a new law on public sector salaries. The State and the FBH are committed to engage on a similar path. Reform o f public sector wages i s absolutely essential not 18 only to hold public sector wages in check, but also to send a strong signal to the private sector that increases innominal wages cannot exceed increases inproductivity. 1.46 The second axis for fiscal reforms focuses on the reallocation of expenditure across and within spending categories. An expenditure review conducted by the World Bank highlightsmany areas o f inefficiency (World Bank, 2002a). The observed expenditure patterns shown in Table 1:3 reveal some hnctional categories o f expenditure which stand out as prime candidates for rationalization, in particular defense and public order and safety. Within certain sectors, reallocation and efficiency gains are called for. For instance, policy and institutional reforms inthe education sector to reduce fragmentation and improve efficiency and equity would yield better outcomes at a lower cost. Education spending in BH is characterized by the consumption o f excessive resources with respect to output (internal inefficiency), an inadequate contribution to the economy and to living standards generally (allocative inefficiency), and inequity inspending at the higher levels. Giventhe sector's importance for long-term growth and poverty reduction, a program has been devised under the framework o f the M T D S to reform education financing through increasing internal efficiency, with any gains realized to be reinvested in expenditure restructuring that would boost the effectiveness and equity o f the system. Of particular importance for growth and labor market outcomes in the context o f BH's economic transition i s the poor efficiency at the tertiary and secondary levels, as highlighted in Chapter 5. 1.47 Social transfers remain large relative to GDP and overall spending, but are very unevenly spread. While spending on social and child protection accounts for between 0.8 and 1.3 percent o f Entity GDP, direct spending on veterans' benefits alone amounts to a staggering 3.5- 4 percent of Entity GDP in recent years. Therefore, the reform agenda is about reorienting the structure o f social spending away from entitlement-based programs for those frequently not in need (particularly in war veterans' programs) to those most in need (particularly those receiving inadequate social welfare benefits and services), while at the same time reducing the overall fiscal pressure from transfer programs. Recently adopted new veterans' benefit legislation aims at rationalizing expenditure while protecting the most vulnerable more effectively. At the same time, the Entities will need to adopt new social welfare legislation which centralizes at least one cash benefit at Entity level (social welfare inthe RS and child protection inthe Federation) to help improve equity andprovide policy tools to mitigate the impact on the poor o f adverse developments. 1.48 A significant effort of downsizing the armed forces undertaken since 1998 resulted in a reduction by about 15,000 soldiers in various rounds of demobilization through 2002. Another round was initiated in 2004 in the context o f BH's commitment to accede to NATO's Partnership for Peace program. Similarly, analysis o f the intra-sectoral composition o f spending andexisting inefficiencies inthe area o fpublic order and safety is ongoing with the assistance o f the European Union Police Monitoring Mission (EUPM). An initial program o f police force downsizing has already been completed in the RS. Efforts in these areas will need to be consolidated inthe future to address still substantial overstaffing 19 Table 1:3 Breakdown of Consolidated Government Expenses, 2002l (Percentof BHor EntityGDP) BHTotal Federation of BH Republika Srpska (incl. State (incl. municipal (excl. municipal Government) expenses) expenses) Economic classification Current expenditures 42.6 45.0 37.4 o/w: wages & salaries 14.6 16.4 9.3 goods & services 4.4 4.0 5.O subsidies & transfers 22.0 23.3 20.4 interest 1.6 1.3 2.7 Capital expenditure 1.7 2.0 0.8 Total 43.9 47.0 38.2 Functional classification Generalpublic services 5.5 4.8 2.3 Defense 4.3 4.9 3.5 Public order & safety 4.2 4.3 4.2 Health 6.1 6.6 5.9 Socialprotection 15.3 16.9 13.5 Education 5.6 6.4 4.3 Economic affairs 3.O 2.8 3.6 Other 2.2 1.8 2.9 Source: World Bank Poverty Expenditure and Institutional Review (2002). 'Figures not directly comparable to those in table 1.2, which includes final data for 2002. This table, by contrast, is based on earlier estimates, with more recent figures o f the functional breakdown unavailable. The total includes expenditures by the Entities, state and extrabudgetary hnds but exclude foreign-financed public investment (estimated at 8.5 percent o f total BH GDP in 2002). The FBH data include estimates for municipal expenditures but the RS data excludes municipal expenditures (estimated at 7.9 percent o f the R S GDP and2.3 percent o fBHGDP in2002.) 1.49 Cross-cuttingreformsin intergovernmentalfiscal relationswill be neededto further strengthen financial discipline and engender hard budget constraints at subentity level, reduce horizontal and vertical imbalances and contribute to improving the delivery of public services. This would be achieved by revising expenditure responsibilities in key sectors (education, health, social services), and, in parallel, reviewing tax revenue assignments and intergovernmental transfer mechanisms. Finally, strengthened expenditure management and public governance can be powerful tools to enforce medium-term fiscal adjustment and expenditure allocation reforms, in particular with respect to budget preparation and the determination o f a medium-term resource and expenditure framework; financial accountability and transparency in terms o f budget execution, control and audit; and limiting conflicts or interest and opportunities for conuption (especially inpublic procurement). 1.50 The proposedintroductionof a VAT is a key reformon several accounts:tax policy, intergovernmentalfiscal relations,domestic competitionpolicy and governance.The recent establishment o f a single, state-level Indirect Tax Authority (ITA) that will be responsible for the collection and administration o f all customs and indirect taxes in BH, as well as for the introduction of V A T by 2006, is a major institutional breakthrough. From the tax policy and administration perspective, the ITA and the associated reforms should not only bring BHcloser 20 to EU standards and practices, but will also lead to improved tax collection and remove some o f the distortions related to the current sales tax. 1.51 The introduction of the VAT is a unique opportunity to reconsider intergovernmentalfiscal relations among the State and the Entities as well as within the Entities. Revenues collected by the I T A accrue to a single account, from which three expenditure items will be financed off the top: (i)the state administrative expenses: (ii) the country's external debt service obligations (which are all assumed by the State): and (iii) the ITA'Sown operational budget. The remainder o f the tax proceeds collected by the ITA will be allocated among the Entities and Brcko District, probably according to predetermined criteria, such as the final point o f sales in the case o f VAT revenues in the first year. How revenues would then be allocated to the sub-entities remains to be determined. This poses a particular challenge inthe FBH, where the current tax sharing arrangement would leadto a gradual erosion o f the revenue base o f the Entity to the benefit o f sub-Entity governments (Figure 1:24). Before simply adopting the derivation principle for sub-Entity allocation, the FBH authorities need to closely examine expenditure responsibilities across levels o f government andreview the revenue sharing arrangements inthat light. Figure 1:24 FBH: ProjectedAllocation of Tax Revenuesbetween Entityand Sub-Entity (inpercent of total) 100.0% 50.0% II 0.0% 2002 2003 2004 2005 2006 2007 Cantons and municipalities 0 Entity1 Source: FBHMinistryof Finance- MTEFdocument. 1.52 Managingmedium-termfiscal risks in BHwill requirea successful settlement of the large stock of domestic claims. There have been no domestic bond issues in BH and until the end o f 2003 bank lending to the government was negligible, amounting to less than 0.25 percent o f GDP. However, there i s a large component o f potential domestic claims amounting to more than 200 percent o f GDP. Most of this potential debt relates to the breakup o f former SFRY and the ensuing war, including: claims against the government for the freeze o f foreign exchange deposits immediately before the war (equivalent to around 18 percent o f 2003 GDP); several large war claims related to material losses as well as non-material damages including death o f relatives and other damages suffered by veterans and victims of the war; claims inherited from privatized banks and enterprises; unused privatization vouchers; and unsettled government 21 arrears, including arrears to suppliers, on wages and on social benefits, including those to pensioners and war veterans. 1.53 The settlement of most such debts will remain contentious, but there is consensus among governmentofficials on the impossibility of doing so at face value. As a result, both EntityGovernments agreed in2004 to limit the total cost o f settling these claims to no more than 10percent o f 2003 GDP innet present value (NPV) terms. Quantitative frameworks for settling the claims have been established for both entities, including combinations o f cash payments, debt write-offs or securitization invery long-term bonds (20 to 50 years) with extended grace periods and virtually no coupon. Inparallel, moratoria on the processing o f pending lawsuits related to war damages were passed by Entity parliaments. While the moratoria are in place and claims verification progresses, Entitylaws are meant to enshrine the governments' settlement strategy. 1.54 A recent constitutional court ruling has raiseddoubts whether this strategy can be implemented as planned. This would have adverse fiscal implications. Inthe event o f settling debts at higher NPV than the 10 percent the government intended, higher taxes or substantially lower spending would be required to generate the primary surplus needed to ensure that government debt i s sustainable. ExternalDebt Strategy:PursuePrudentBorrowingon the Roadto Creditworthiness 1.55 BH's gross externaldebt fell from 70 percent of GDP in 1998 to 35 percent by the end of 2004.9Bank staff estimates the net present value o f total external debt at end 2004 to be equivalent to 25 percent o f GDP or 91 percent o f exports o f goods and services. As such, BH i s classified as a moderately indebted middle-income country. This remarkable achievement was supported by the successes o f the BH authorities in achieving successive rescheduling agreements, as well as the regularization and cancellation o f old debts denominated in non- convertible currencies, while pursuingprudent external borrowing policies." 1.56 BH's external debt appears to be sustainable in the short run, but the authorities must consolidate this achievement by implementing policies that will help narrow the current account deficit, and by maintaining prudent external financing strategies. Compared to other SEE countries and countries o f similar per capita income (lower-middle income countries), BH's external debt i s not large (Table 1:5). The ratio o f NPV o f debt to GNI The central bank has yet to account for borrowing abroad by commercial banks and companies. lo After normalizing its relations with the IFIs inthe immediate aftermath o f the war, BH renegotiated its debt with official bilateral creditors, reachng an agreement with the Paris Club in October 1998. The debt renegotiated included the entire "allocated debt" (debts owed or guaranteed by entities located inBH's territory) and 13.2 percent o f the SFRY unallocated debt. The Paris Club agreed to reduce this debt by 67 percent in net present value (NPV) terms and to reschedule the remaining obligations over 30 years with a six-year grace period o n principal repayment for commercial type credits, and over 33 years with no grace period for Official Development Assistance (ODA). Similarly, in December 1998, BH finalized an agreement with the London Club which became effective the following January. The London Club forgave some 73 percent o f the debt on a present discounted value basis while the remaining $400 million was to be divided into two components: a "basic amount" o f some $147.8 million and a "performance amount" for the rest. The basic amount was rescheduled over 20 years with a seven-year grace period. The performance amount would fall due once and if BH GDP per capita reaches US$2,800 (measured at 1997 prices) for two consecutive years before 2018. 22 is at the low end o f the SEE region, higher only than that o f Albania (Table 1.5). Due to fairly underdeveloped export capacity,'' however, the NPV of external debt as a share o f exports o f goods and services i s above the median, but only by one percentage point. Debt-service ratios also look favorable. Payments due on total external obligations were 6.5 percent o f exports o f goods and services in 2004. Sensitivity analysis seems to show a relatively good absorption capacity to shocks on the economic growth rate or the exchange rate. A marked shortfall innon- debt creating balance-of-payments financing (e.g., one third below the assumptions set forth in the reform scenario below), however, would lead to an increase inthe NPV/GDP ratio above 30 percent by 2007 or so. 1.57 The highly concessional terms of donor assistance provided to BH helped ensure a relatively comfortable position with respect to external debt servicing. The debt service profile deteriorated somewhat in 2002, as repayments under IDA loans and the IMF Stand-By Arrangement fell due. As a result, the debt service-to-exports ratio reached 9.2 percent in 2002 before easing substantially thereafter. Inthe full reform scenario described below, however, this ratio is expected to decrease below 6 percent by 2007. In addition, assuming the continuation o f prudent external borrowing, the external debt-to-GDP ratio would decline below 30 percent over the projectedperiod. 1.58 The external debt of BH exhibits some particularities not found in most other countries in the region. Due to the exceptional needs for soft financing in the post-war period and the particular role o f the international community in the reconstruction of BH, financial assistance to the country was on highly concessional terms. An estimated two-thirds of all assistance provided to BHup until 2002 was disbursed in the form o f grants. The average grant component of the borrowing part was quite high at about 40 percent (2002-03) leading to an effective interest rate o f 2.5 percent. Inaddition, virtually all o f BH external debt i s public debt, denominated mainly in euros and U S dollars.12 As the private sector has had limited access to foreign borrowing mainly due to poor creditworthiness, external private debt is very low, amounting to about 1 percent o f the total outstanding external debt and being concentrated among a few creditors such as EBRD and EIB.13 Only one-fifth o f this amount corresponds to direct investment while the rest has been provided in the form of publicly guaranteed loans, mainly to support the strengthening o f financial intermediation. 1.59 The vast majority of BH's external debt is held by official creditors. While this has allowed a high grant component, the high share o f debt with preferred creditors (58 percent) reduces the room for maneuver in the event of adverse shocks, as these debts cannot be rescheduled. BH's external debt i s largely denominated inU S dollars and euros. The volatility o f the euro/US$ exchange rate over the course o f the past three years has affected BH's debt indicators. Inparticular, with the KMpegged to the euro at a fix parity, the recent appreciation o f the Euro, combined with the high share o f dollar denominated debt (45 percent after decomposing the SDR by its currency basket) has at the same time expanded the dollar value o f I'BH'sexport-to-GDP ratio ranks at the very bottomo fthe selected group o fregional countries. l2There are indications o f large borrowing abroad by commercial banks, but official statistics have yet to record these. l3Some inter-bank lending from foreign commercial banks has taken place lately but there i s n o evidence o f direct foreign commercial bank lending to private enterprises. 23 the external debt while contributing to the improvement of solvency indicators such as debt to GDP and even some external liquidity indicators such as debt service to export ratios. Table 1:4 BH:GrossExternalDebt (US$ million. end2004) Nominal NPV Structure of Value NPV Total 2766 1625 100 Multilateral 1802 1173 72.2 OfWhich IDA 884 431 26.5 IBRD 531 452 27.8 Bilatera 782 382 23.5 Paris Club 661 331 20.4 Non Paris Club 121 90 5.5 Private 182 135 8.3 Note: Does not include external debt o f commercial banks Source: CBBH Table 1:5KeyExternalDebtRatiosfor CEE Countries I' Present Value (PV) PV ofEDT EDT Debt service Interest (Yo of GNI) (Yo ofXGS) (Yoof GNI) (Yo ofXGS) (Yo ofXGS) Bosnia & Herzegovina *' 21 93 36 8 3 Albania 19 58 28 3 1 Bulgaria 65 118 75 19 7 Croatia 55 111 55 31 6 Czech Republic 40 55 41 12 3 Estonia 58 62 56 8 3 Hungary 60 87 64 42 4 Lithuania 47 98 47 37 4 Macedonia 33 73 40 12 4 Poland 37 123 39 32 5 Romania 30 91 31 22 5 Serbia and Montenegro 122 378 123 4 2 Source: Own estimates for Bosnia and Herzegovina and Global Development Finance 2003 for all other countries. BHdata includes the "performance amount" o fthe LondonClub equivalent to some US$280 million. D.THEMACROECONOMIC OUTLOOK The ReformScenario 1.60 A quantifiedillustrativescenario describes the possibleevolution of the economy if most of the priority reforms mentionedin this CEM are implementedin a timely fashion. Specifically, it is predicated upon the following assumptions: (i)continued adherence to the 24 currency board; (ii) continued efforts to consolidate public expenditure; (iii) tightening o f a credit policy leading to a reduction o f the current account deficit; (iv) the acceleration o f structural reforms, especially with respect to the business environment, privatization and the labor market, that will gradually translate into higher private domestic savings; and (v) the absence o f an external shock. The projected trajectory of key economic variables under this illustrative scenario i s summarized inTable 1.8. 1.61 Real GDP growth would slow somewhat in2005, inthe face o f a credit tightening, before accelerating towards 5 percent in outer years. This growth target would be feasible under the assumption o f overall productivity gains, brought about by labor market reforms and increased investment in the private sector. Because o f the continued adherence to the currency board, inflation would remain close to euro area averages. Nominal exports would grow robustly in double digit figures throughout the period, while imports trends would be influenced by a slowdown in consumer and reconstruction imports and still relatively strong non-reconstruction capital imports. Accordingly, gross investment would remain at around 20 percent o f GDP, financed by further gains in domestic savings, initially from the public sector under the assumption of expenditure consolidation and successful revenue reforms. In the outer years, domestic private savings would reflect improved corporate profitability. As a result, the current account balance would improve gradually, but external financing requirements would remain sizeable. External savings would be increasingly mobilized in the form o f foreign direct investment which would continue the momentum observed inrecent years and stabilize around 6 percent o f GDP, a figure that i s inline with averages observed among "good performers" (but not the best) inthe Region.14 Financing Needs under the Reform Scenario 1.62 Projected financing needs and sources based on assumptions discussed previously are summarized inTable 1.9. The non-interest current account is projectedto narrow further.Annual average debt service requirements would double in nominal dollars in 2009-13 with respect to current levels, reflecting increased principal repayments. Foreign exchange reserves are assumed to continue to grow in nominal terms but to decrease in months o f imports and goods and services from the current highlevel. 1.63 Evenunder a reform-drivenmacroeconomic scenario, BH's total gross external financing needs over the period 2005-14 are projected to remain large at roughly US$13 billion. These requirements would need to be met by borrowing by the public and (gradually) by the private sectors, as well as by non-debt creating flows such as FDI and grants. The projection is complicated by data uncertainty regarding non-identified capital inflows, most o f which are assumed to be private transfers. These flows have represented as much as 10percent o f GDP in past years. This projection takes a cautious approach with respect to these inflows and assumes a significantly lower figure at the beginning o f the projection period, followed by a gradual reduction in outer year. International financial institutions are assumed to continue to play a key but declining role. Similarly, improved creditworthiness in line with a more favorable business l4 FDI flows typically show considerable year-to-year volatility, as they are often dependant on a small number of large transactions, especially insmaller-sized countries. Such variations are obviously impossible to forecast and the projections displayed here are meant to describe an estimate of a medium-term average. 25 environment will allow the emerging private sector to gradually gain access to commercial foreign financing. E.CONCLUSION 1.64 Bosnia and Herzegovina has made enormous progress from a totally devastated post-war economy to a functioning economy which has slowly started to position itself in the regional and international context. In particular, BH has been successful in achieving macroeconomic stability: inflation has been brought down, fiscal accounts have improved, the confidence in the currency i s strong, the weight o f external debt has been reduced and foreign 'reserves have increased. Macroeconomic stabilization has resulted in robust increases in investment and output growth. Substantial challenges remain, however, with the persistentlyhigh current account deficit representing the most important macroeconomic imbalance to be addressed. Low employment and a rising size o f the informal economy are also hinderingoutput growth. Table 1:7 BH: ReformScenario2005-2009 2005 2006 2007 2008 2009 (Percent change previous years) Real GDP growth 5.4 5.O 5.O 5.O 5.O Real GDP per capita 4.9 4.5 4.7 4.7 4.7 CPIInflation 1.o 2.0 2.0 2.0 2.0 Exports of Goods and Services 14.3 15.5 13.8 12.5 11.0 Imports of Goods and Services 6.8 6.5 5.8 5.3 5.O olw non-reconstruction 9.3 8.0 7.0 6.5 5.7 (Inpercent ofGDP) Gross Investment 20.8 20.5 20.5 20.8 21.0 National Savings 4.1 6.2 8.4 10.6 13.0 olw public savings 1.9 2.9 3.8 4.7 5.5 olw private savings 2.2 3.3 4.7 6.0 7.4 General government revenues 48.9 47.1 45.8 44.7 43.8 General government expenditures 50.0 47.3 45.5 44.2 43.2 o/w: capital expenditures 6.4 5.5 4.9 4.8 5.O Fiscal balance -1.1 -0.2 0.3 0.5 0.6 Current account balance -16.8 -15.2 -13.4 -11.6 -9.8 External debt 52.7 49.4 46.2 42.9 39.7 o/w: external public debt 30.0 28.2 26.2 24.2 22.1 External debt service (Inpercent of exports of goods and services) 6.4 6.9 5.8 5.O 5.2 Gross official reserves (Inmonths o f imports of goods and services) 5.4 5.2 5.O 4.8 4.6 26 Table 1:8 BHExternalFinancingRequirements Annualaveragefor eachperiod (In millions ofUS dollars) 2005-09 2010-14 FinancingRequirements 1,625 1,107 Current account deficit (incl. official transfers) 1,445 885 3. Amortization (total) 110 139 4. Change inNet ForeignAssets (incl. net IMF credit; +=increase) 70 83 FinancingSources 1,625 1,107 1. Official Capital Grants 180 100 2. Disbursements(total) 130 161 3. Other (net LT and ST flows) " 600 153 4. Private investment 2' 718 693 Source: Bank staff estimates Note: Includes short-term capital flows, and capital not elsewhere included.. 1.65 Much remains to be done, as a result, to create a self-standing, vigorous economy capable of sustaining robust inflows of FDI and generating jobs. Many areas o f the "traditional" transition agenda remain insufficiently addressed, most prominently corporate restructuring and labor market reforms. Reforms inthese areas are absolutely essential to secure a supply response to market opportunities and improvements in the business climate. The country will have to offset the ongoing decline in concessional aid inflows by boosting domestic saving and helping spur an increase in foreign direct investment. This should help limit dependence on official aid to cover fiscal and external imbalances. 1.66 The authorities should sustain spending restraint and adhere to the CBA. While building the state institutions to advance EU integration, the authorities should continue to improve the efficiency o f spending, including through reallocation o f outlays from wages to public investment, operations and maintenance. This will require decisive action. Within core public sector services, such as education or social protection, expenditure allocation should be driven by efficiency and equity considerations. In particular, vocational and higher education financing are in urgent need o f reform, as are welfare transfers to households. Policy options in these areas have been formally endorsed and now need swift implementation. Also, the planned policies to contain fiscal risk through the settlement o f the large overhang o f domestic claims should be implementedas foreseen. To ensure that the country establishes its creditworthiness, the authorities should maintainprudentexternal borrowing strategies. 1.67 In the longer run, however, it is policies supporting an improved environment for business creation, investment and trade that will generate the private savings and the export capacity needed to lift BH on a higher growth path while reducing external vulnerabilities. The challenges, opportunities and policy options for doing so are reviewed in the subsequent chapters of the report. 27 28 2. INTEGRATINGINTOEU-BASEDSUPPLYCHAINS- A STRATEGYFORBOOSTINGEXPORTS? A. INTRODUCTION 2.1 Despite a liberal trade regime and early pick up in export growth in 1996, BH's recent sub-par export performance and still frail record in attracting foreign direct investmentremain major bottlenecksto economic growth.As shown inChapter 1, for BHto move to a sustained growth path in the reform scenario, exports o f goods and services would needto increase from 25 percent o f GDP in2003 to 36 percent in2010. The aim o f this chapter is to analyze recent trends in trade flows, shed some light on BH's potential to achieve the targetedexport growth and explore options for international trade policy that would support such an outcome. 2.2 The chapter reviews the surge in exports immediately after the end o f the war and their subsequent stagnation. This altogether disappointing performance can hardly be attributed to a policy stance under which trade liberalization and integration have been pursued: import tariffs are among the lowest inSouth East Europe (SEE), and BHhas successfully concluded a series o f bilateral trade agreements while pursuing integration with the EU and WTO membership. Linking trade performance with the trade environment, some have questioned whether such a policy stance i s indeed adequate and should not be reversed. This Chapter concludes, by contrast, that resorting to a more restrictive trade regime would not only send adverse policy signals, but also not deliver the intended results. First, while trade volumes are still too low, BHwas able to redirect trading flows towards the markets that will matter for its hture, notably the EU. Opportunities exist, ranging from more efficiently trading natural resources to integrating into retail and producer-driven supply chains. Initial success demonstrates that such opportunities can be seized in BH. In terms o f policy responses, the Chapter recommends that while the liberal stance o f the regime should be maintained, if not strengthened, some concrete trade facilitation measures regarding export certification and financing should be acted upon swiftly. More fundamentally, the real constraints to a more dynamic export performance lie in the business environment and investment climate. The main obstacles to be tackled there are described inthe subsequentchapters o f this report. 2.3 The next section o f the chapter reviews BH's recent foreign trade performance in light o f the latest free trade agreements. Section C examines in detail the patterns o f exports, including by analyzing revealed comparative advantages, the level o f processing and the factor content o f exports. BH's participation in international production and distribution networks, the role that FDIcan play inboosting the country's performance and a discussiono f the conditions necessary for deeper involvement in international supply chains feature in section D. Section E presents a numbero frelated policy recommendations. B.TRADE PERFORMANCEAND POLICY DEVELOPMENTS 1996 SINCE 2.4 Followingvery positive developments after the war, BH's overall trade performance has deteriorated markedly in recent years. After a spectacular surge in both exports and imports o f goods and services in 1996-98, both imports (except in 2002 when their value increased by 12 percent) and exports `stagnated'. During 1999-2003, the average annual rate o f growth inthe euro value o f exports and imports o f goods and services was a mere 4 percent and 7 percent, respectively. Trade in services actually stagnated, with receipts falling from a peak o f US$560 million in 1998 to US$422 million in 2002 ( Figure 2:l). However, performance in merchandise trade was better, with the average exports growing at more than 9 percent a year in euro terms during 1999-2003. Figure 2:1Foreign Trade Performance, 1996-2004 Imports-Exports CAD , 5,000I - 35% 1996 1997 1998 1999 2000 2001 2002 2003 2004 2.5 Despite the rapid growth in goods exports (from an extremely low base), the merchandise trade deficit widened in dollar terms and remains outsized relative to GDP. Over the period 1996-2002, exports grew by 357 percent, or at an average annual rate o f 33 percent. As imports generally grew at a slower rate, the ratio o f exports to imports steadily increased from 9 percent in 1996 to 20 percent in 1997 and 37 percent by 2004. Nevertheless, the larger absolute size o f imports, combined with the persistent deceleration o f export growth since 1996, led to a substantial widening o f the trade deficit. Relative to GDP, the merchandise trade deficit decreased from 64 percent in 1996 to 37 percent in2000, but then temporarily reversedits downward trend, bouncing back to 41 percent in 2002, before easing slightly again to 35 percent of GDP in2004. 30 Table 2:l BHTrade in Goods 1996-2003 1996 1997 1998 1999 2000 2001 2002 2003 1996=100 Exports (million US$) 336 595 697 649 809 851 952 1,271 378 Imports (million US$) 1,941 2,400 2,781 2,568 2,533 2,708 3,233 3,890 200 Trade Balance(million US$) -1,605 -1,805 -2,084 -1,919 -1,724 -1,857 -2,281 -2,619 142 Growthrate ofexports (percent) 71.1 21.2 -6.9 24.7 5.2 11.9 33.5 na Growthrate ofimports (percent) 23.6 15.9 -7.7 -1.4 6.9 19.4 20.3 na Exportdimports (percent) 17 24 25 25 32 31 29 33 na Trade deficit/GDP 64 57 48 41 36 37 41 38 na Source: IMF. Note: Given the constraints in data availability, this chapter relies primarily on two sources: the IMF's Direction o f Trade database for information on total trade flows and their geographic distribution, and the U " s COMTRADE database for the information on trade by product. Inmost cases, we use the figures reported by the trading partners. 2.6 BH's slow recent trade performance cannot be explainedby distortions in its trade regime or externaltariff barriers to market access. Onthe contrary, BHhas pursued a sound trade policy by maintaining a liberal trade regime and advancing vigorous trade integration at the regional, EUand global level. The current tariff schedule was adopted at the State level in 1998. BH's average tariff rate o f 6 percent i s the lowest in SEE after Croatia. BH also maintains the lowest tariff rate on agricultural products among all SEE countries. The structure o f tariffs i s simple with four rates (0, 5, 10 and 15 percent) and a relatively moderate dispersion. Moreover, the use o f Non-Tariff Barriers (NTBs) is not prevalent inBH.Accordingly, the cost o f imported inputs used in export production i s lower than in many other countries. Unfortunately, the Table 2:2 Average MFNApplied Tariffs, 2002 reported use o f NTBs by neighboring (percent) countries may have constrained, in some All cases, the expansion o f BH exports to these Country Goods Agriculture Industrial markets. Albania 7.4 9.5 7 BH 6 5 6.2 2.7 By pursuing policy induced trade Bulgaria 9.7 16.7 8.6 integration under the Stabilization and Croatia 5 10.2 4.2 Association process (SAP), BH further Macedonia, FYR 12.6 20.8 11.2 liberalizedexternal trade and succeeded Moldova 5 10.1 4.2 in receiving improved access to the EU Romania 17.1 23.1 16.1 markets. Along with other SEE countries, Serbia 7.4 15.7 6.2 BH was a beneficiary o f the SAP launched Montenegro 6.1 12.2 5.2 by the EU in 1999, which provides a road EuropeanUnion 4.4 9.4 4.2 map o f reform aimed at ultimate EU Source: GEM (2003) and staff calculations. integration.. As of 1999, the SAP provided all WesternBalkan countries, including BH, with non-reciprocal free access to EUmarkets inthe l5The Stability Pact has mandated a study on the prevalence of NTBs inthe Region, which should inform further rounds o f trade reforms inmember countries. 31 context o f Autonomous Trade Preferences (ATPs). Access to EU markets i s free o f tariffs and quantitative restrictions for all but a few products regulated by preferential tariff quotas.I6 As new members acceded to the EU on May 1, 2004, BH exporters began enjoying these preferences in 10 additional countries. These concessions will currently last until the end o f 2005. Once concluded, a Stabilization and Association Agreement (SAA) would govern trade relations between the EU and BH. The S A A envisages a free trade area within 6-10 years," in addition to a wide range o f economic and institutional reforms, cooperation in politics and judicial affairs and harmonizationo f legislation with the EU. 2.8 In addition, bilateral Free Trade Agreements (FTAs) concluded by BH offer wide market access to markets of the neighboringcountries. BHpursuedwith vigor the regional dimension of EU integration project by following up on commitments taken under the Stability Pact Memorandum o f Understanding (MOU) on Trade Liberalization and Facilitation signed in June 2001. BHhas signed FTAs with all parties to the MOU. With some partner countries, such as Croatia, the FTA went into effect evenbefore the MOUwas negotiated and signed. The MOU calls for trade liberalization o f at least 90 percent o f all tariff lines as well as 90 percent o f the value o f bilateral imports. Appendix 2.1 summarizes the main features o f the FTAs signed with other SEE countries. First, most of the agreements are asymmetric as the concessions offered initially by the FTA partners are reciprocated by BH only with a delay. For instance, while Croatia, S A M , and FYR Macedonia provided virtually duty free access for BH exports to their markets at the date o f agreement's entry into force, BH liberalized only about a quarter o f its tariff lines at that time. Only agreements with Albania and Moldova, two economies with only modest trade with BH, were symmetric innature. Second, out of seven agreements listed in the table, four have fully liberalized 90 percent o f their imports with BH." The FTAs with Bulgaria, Romania and Albania, fall short o f the 90 percent import coverage benchmark as they are restrictedto trade inmanufacturing goods, leaving untouched trade inagricultural products. 2.9 BH also signed an FTA with Turkey. The FTA is also asymmetric with a transition perioduntil2007. Moreover, BHexports havehad a similar preferential treatment (although with some restrictions) with Switzerland, Norway, Japan, the United States, Canada and Russia. ?I6The preferential tariff quotas under the ATPs are as follows: baby beef (9,975 tons for S A M ) ; trout, and carp (120 and 140 tons, respectively, jointly for S A M , BH, Albania and FYR Macedonia); sea bream, sea bass, sardines and anchovies (115, 100, 70 and 960 tons, respectively, jointly for SAM, BH, and Albania); wine (152000 hectoliters for all five countries with access by Croatia and FYR Macedonia subject to the exhaustion o f their individual quota under their SAAs). Inaddition, quotas apply to textiles imports originating inSAM. "Thetransitionalperiodfor gradualremovalofcustomstariffsis amatterofnegotiationandcouldalsolast less than six years. It should also last less than 10 years, which i s the maximumWTO accepted transitional period for regional trade agreements to be compatible with Art XXIV GATT. The MOU stipulated two main conditions for trade liberalization: Mutual trade had to be liberalized up to 90 percent o f HStariff lines and 90 percent o f the trade value (Art. 1.2.2). Moreover, the liberalization was to include a large majority o f goods upon entry into force o f the agreement, with a transitional period, not exceeding six years, for the most sensitive products (Art. 1.2.3). 32 Box 2:1Discrepancies in Trade Statistics from Different Sources While the Bosnian customs system continues to be strengthened, discrepanciesremain between the national trade statistics and the figures reported by foreign trading partners. As demonstrated below, the figures presented by the Ministryo f Trade are roughly consistent with those from the IMF's Balance of Payments Statistics.However, the IMFreports much lower exports in2000 and 2001and thus a higher growth rate of exports. The figures from IMF's Direction of Trade Database, based on trade partners' reporting, are significantly lower than those from the two sources described above but indicate similar changes inexports and imports during the past few years. These matters are further complicated by the recent appreciation o f the euro vis-a-vis the dollar which, as indicated below, results in an overestimation o f the changes intrade flows. Regardless o f which source we employ, however, the figures point to an increase inboth exports and imports inthe past two years. Table 2:3 Comparison of Trade Data from Different Sources, 2000-2003 2002 2003 2000 2001 2002 2003* 2001=100 2002=100 IMFDirectionofTrade (million US$)' Exports 669 739 827 112 Imports 2,643 2,820 3,242 115 Trade Balance -1,975 -2,081 -2,415 ...... 116 ...... BHMinistryofForeignTrade andEconomicRelations (million US$) Exports 1,042 1,052 1,130 1,344 107 119 Imports 2,974 3,173 3,603 4,395 114 122 Trade Balance -1,932 -2,121 -2,473 -3,051 117 123 IMF BOP figures (million US$) '9` Exports 832 870 1,046 1,341 120 128 Imports 2,657 2,823 3,271 3,801 116 116 Trade Balance -1,825 -1,953 -2,225 -2,460 114 111 IMF BOPfigures (million euro) Exports 901 971 1,107 1,186 114 107 Imports 2,876 3,151 3,464 3,362 110 97 Trade Balance Notes: Estimate. Based on trade partners' reporting. a ' -1,975 -2,180 -2,357 -2,176 108 92 Import figures are not adjusted downwards to exclude imports by non-residents C. EXPORT PERFORMANCE:CAUSES ANDCONSTRAINTS 2.10 Against this background of increased market access opportunities, BH's supply response in trade with the EU and FTA partners was positivebut insufficient to generate significant growth in exports and employment.BH's exports remain very low in relation to GDP, and more than three times lower than before the war taking into account trade with other former Yugoslav republics and almost twice as low with countries other than ex-Yugoslavia (Table 2:4). This section analyses BH's export performance under existing trading arrangements. It identifies the main features o f BHexports by focusing on commodity composition, direction, 33 and factor intensities. It also looks at the extent o f integration o f BH's producers with international production networks. Table 2:4 BHExports as percent of GDP in 1990 and 2002 1990 2002 2002a/ Exports to other republics of former SFRY 57.3% 8% 11% Exports outside former SFRY 22.8% 12% 17% Total exports 79.6% 20.1% 28.4% Source: Derived from data for 1990 o f the BHStatistical Office (as quoted inMesinovic et al. 2002), IMFDirection- of-Trade database for 2002, and WDI 2003. the value o f total exports including services Directionof Trade andFree TradeAgreements 2.11 A major impact on BH's export came from the grantingof improved access to EU markets. BH displays a `standard geographical pattern' for European countries in terms o f reliance on EUmarkets for its exports with a share o f almost two-thirds. The share o f BHexports to EUmarkets rose steadily from 44 percent in 1997 to 66 percent in 2002, a redirection similar to that experienced by CEE countries. As EU markets are highly competitive and more demandingthan those o f South Eastern Europe, the ability to make significant sales there bodes well for BH exporters. The EU also dominates on the import side, having been the source o f 43 percent o fBHimports in2002. 2.12 BH's share in EUimports remains negligibleat one-quarter of one percent despite efforts to reorient exports towards the EU. Inthe 1996-99, BH outperformed the 10 Central European EU-candidate countries in EU markets with rates o f growth o f 180 percent, 108 percent, 51percent and 45 percent, respectively. The value of EU-bound exports rose 21 percent in2000, quiteimpressivebyitself, the same as inBulgaria, butwell-below Estonia(37 percent). Export growth eased thereafter, slowing to 9 percent in 2002, trailing the pace o f increase in exports to the EUfrom Romania (19 percent), Lithuania (17 percent), Slovakia (14 percent), and Czech Republic (13 percent). The slowdown inexport growth appears to have reflected primarily sluggish incorporate restructuring o f BHcompanies. 2.13 Except for trade with Croatia, the impact of bilateral FTAs on export expansion remains limited." In the cases of Croatia and Slovenia, an increase in exports was recorded after the conclusion o f the FTAs.~'The FTA with the FYR o f Macedonia also resulted in an increase in BH exports, albeit from a very low base. In contrast to the FTAs mentioned above, the agreement with S A M did not lead to an increase in exports. The latter remained stagnant during 2001-02 and declined slightly in 2003. At the same time, a decline in imports was registered in 2003. NTBs and distorted implementation o f the FTAs may in part explain the disappointing results.21Anecdotal evidence suggests that while firms from the RS can enjoy duty 19Free trade agreements with Albania, Bulgaria, Moldova, Romania and Turkey only came into effect in 2003. It is thus too early to discuss their effects inthis report. 2o The ratios have been calculated based on the IMF Direction of Trade Statistics for all years including 2002 and the figures providedby the Federation and RS Customs Admmistrations for 2003. 2' The FTA has applied to Serbia since July 1, 2002 but Montenegro was not covered until September 5, 2003. Furthermore, Kosovo has been excluded from the agreement. 34 free access to S A M ' s market, potential exporters from the FBH are discriminated against, for example, by being subjected to mandatory licensing for road transport. A similar pattem o f discrimination against the FBH exporters takes place in the case o f veterinary inspection. Moreover, inviolation of the FTA agreement, BH exporters have been charged transit fees while en route to Kosovo. The elimination of NTBs in trade among SEE countries i s currently on the agenda o f the Stability Pact Working Group on Trade Liberalization and Facilitation, and all FTA signatories will be expected to soon remove them. Table 2 5 Importance of EU-destinedExports for B H and Comparator Countries,- 1994-2002 Exports to EU/Total Exports (percent 1996 1997 1998 1999 2000 2001 2002 Albania 86 87.5 92.5 93.6 97.7 96.2 96.1 Serbia and Montenegro n.a n.a n.a n.a 40.7 46.5 43.6 Slovenia 64.6 63.6 65.5 66.1 63.9 62.6 79.4 Hungary 62.6 71.2 72.9 76.2 75.2 74.4 76.9 Slovak Republic 41.3 47.1 55.7 59.4 59.1 75.5 75.5 Romania 55.9 56.7 64.6 65.5 64 67.9 69.9 Bosnia and Herzegovina 44.4 44 49.5 61.3 66.8 68.1 66.2 Bulgaria 40 45 51.5 54.2 51.7 55.2 56.1 Croatia 51 51.3 47.7 49.4 54.9 54.3 52.3 Turkey 49.8 46.7 50.5 54 52.3 51.5 51.5 Lithuania 33.4 32.5 38 50.1 47.9 47.8 48.4 Macedonia, FYR 42.8 36.9 44.1 44.7 57 63.3 46.8 Moldova 9.8 10.3 12.5 20.6 21.7 21.4 33.5 Source: Based on country reporting inIMF,Direction of Trade Statistics. Figures for S A M are based on datafrom SAM's Federal Statistical Office. Level of Processing and Revealed Comparative Advantage 2.14 While BH achieved a substantial redirection of foreign trade, the impact of exports on real GDP growth also depends on export composition and prospects for the medium to long run. This section shows that BH's exports remain heavily reliant on products intensive in unskilled labor natural resources and identifies the related challenges for returning to rapid export growth. 2.15 BH's exports are dominated by products at a lower level of processing. Table 2:6 presents the breakdown o f BH exports to the EU by end-use products. First, consumer goods constitute more than one-half o f BH exports. Clothing and footwear are particularly prominent, accounting for almost 30 percent of total exports in 2002. Second, foreign sales o f automobiles and automotive parts, which have served as a driver o f exports in many Central European countries, e.g., the Slovak Republic and the Czech Republic), are negligible. Third, food exports have beennegligible as well. To a large extent this i s due to historical conditions, in addition to the lack o f adequate veterinary services (Appendix Table 2.2). 35 Figure2:2 BHExportsto FTA Partners, 1997-2003 600 500 - h 400 e 300 200 100 0 Table 2:6BH's Exportsof End-UseProductsto EUMarkets,1996-2002 End-UseProduct(SITC) 1996 1997 1998 1999 2000 2001 2002 Export Share (in percent) Food (0+1+22+4) 2.3 0.9 2.1 2.1 2.4 2.2 2.3 Agncultural raw materials (2-22-27-28) 5.O 9.3 15.0 19.1 15.9 12.6 13.7 Industrialraw materials (27+28+68) 12.0 11.0 9.7 17.0 27.6 23.9 21.2 Machinery & equip, excl. auto (7-78) 10.2 5.9 4.3 4.3 4.0 3.6 5.1 Auto & parts (78) 1.2 0.5 0.4 1.4 0.2 0.4 0.5 Consumer goods (5+6+8+9-68) 69.2 72.3 68.4 56.0 49.8 55.5 57.1 ofwhich: Textiles (65) 0.9 1.6 1.2 1.2 1.8 2.5 1.8 Clothing (84) 23.3 29.9 34.2 25.9 18.0 19.7 18.4 Footwear (85) 16.4 14.9 9.0 8.4 8.6 8.9 10.9 All goods ( 0 to 9) 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: Computations based on EUdata from UNCOMTRADEStatistics. 2.16 BH's revealed comparative advantage (RCA)22in EU markets is concentrated in products with low level of processing and is similar to those found in other countries in the region.23BH's major exports include raw hides and fir skins, wood, footwear, leather and leather products, non-ferrous metals, furniture and furniture parts, apparel and sugar and sugar preparations. With the exception o f the latter, BH's revealed comparative advantage has been consistent since the mid-1990s. Its revealed comparative advantage in the Croatian market i s *'The RCA is the ratio o f the share o f a given product inBH's total exports to the share o f that same product inthe EU's total external imports. An ESI with a value that exceeds unity suggests a strong specialization in the product (Balassa, 1965). An R C A value exceeding 1 implies that BH is more specialized in exporting the product to the EU than an average trading partner o f the EU. 23Other sectors with the RCA index above 1include wood manufactures other than furniture, metalliferous ores and metal scrap, other manufactures o f metal, and iron and steel. 36 focused on agri~ulture.~~While BHcould potentially have a comparative advantage inexporting animal products to the EU, just as it does to Croatia, the inability o f potential exporters to meet the requirements for obtaining the veterinary certifications by the EU, and complying with the sanitary and phytosanitary measures, obstructs such exports. This missing out on potentially profitable export opportunities is more a reflection of internal rather than external constraints. FactorContentof Exports 2.17 In addition to low levels of processing, BH exports to the EU are heavily concentratedin naturalresources-basedand unskilledlabor-intensiveproducts.The former i s the dominant category, accounting for 45 percent o f EU-bound exports in 2002. However, since the value of natural resource-basedproducts has remainedstable since 2000, their share in total exports to the EUhas declinedby 10 percentagepoints since 200225.On the other hand, the value of unskilled-labor-intensive exports to the EU has been steadily increasing, reaching 42 percent of the total in 2002. This category comprises apparel, footwear, furniture, textiles and others. Figure2:3 BHExportsto the EUby Factor Figure2:4 BHExportsto Croatia by Factor Intensity Intensity UCapital Intensive OSkilled Labor Intensive 2.18 Similarly, exports to Croatia are dominated by natural resource-based products, which account for 60 percent of shipments.. Products intensive in skilled-labor also perform quite well inthe Croatian market, accounting for almost one-fifth of BH's exports in 2002. The third largest export category in 2002 consisted of unskilled labor-intensive products which amounted to15 percent o f the total. Exports of capital-intensive goods to Croatia have beenvery small, remaining below US$lO million or 10percent of the total between 1997 and 2002. 24 For instance, the highest value of the RCA index (5) can be found in fixed vegetable oils and fats (SITC 42) and oil seeds (STIC 22). Not far below are live animals chiefly for food (SITC 00), dairy products and eggs (SITC 02), and meat andmeat preparations(SITC 01), all three with the index value exceedingfour. 25 This category includes products such as aluminum and aluminum alloys (SITC 6841 the largest exports group), lumber (SITC 2433), leather footwear parts (SITC 6123), other ferro-alloys (SITC 6715), and bovine and equine hides (SITC 2111). 37 Table 2:7 Bosnia'sRCAIndicesandOther SelectedComparator Countries inEUMarkets BH Croatia FYRM Slovenia Bulgaria Hungary Romania SITC-2 Product 2002 2002 2002 2002 2002 2002 2002 21 Hides, skins and fur skins, raw 29.35 6.96 3.97 3.14 1.45 0.33 1.59 61 Leather, leather manuf., n.e.s. 9.54 6.92 3.74 1.54 6.45 0.91 8.98 85 Footwear 10.68 4.64 5.40 0.64 3.19 0.96 9.17 84 Articles o f apparel and clothing ac 3.07 3.80 7.81 0.78 5.24 0.72 6.37 24 Cork and wood 13.86 8.79 0.41 1.39 0.52 0.87 1.30 67 Iron and steel 1.61 0.73 12.72 2.56 5.05 0.77 1.96 82 Fumiture and parts thereof 7.28 4.35 0.04 6.14 1.81 1.48 3.92 06 Sugar, sugar preparations and honey 3.03 10.77 3.45 0.40 0.81 1.11 0.58 68 Non-ferrous metals 9.13 1.36 3.08 2.14 1.07 0.79 0.89 63 Cork and wood manufactures 2.96 3.58 0.07 4.16 1.32 1.23 3.27 11 Beverages 0.22 0.68 9.96 0.47 3.04 0.61 0.40 12 Tobacco and tobacco manufactures 0.11 0.19 11.05 0.00 2.87 0.05 0.00 04 Cereals and cereal preparations 0.35 2.3 1 0.00 0.16 8.37 2.15 0.52 56 Fertilizers, manufactured 0.00 4.52 0.00 0.01 6.51 0.23 1.90 81 Sanitary, plumbing, heating and light 0.67 0.83 0.22 2.59 4.56 1.67 1.11 00 Live animals chiefly for food 0.00 0.59 0.79 0.12 1.53 3.70 4.59 01 Meat and meat preparations 0.03 0.92 2.76 0.73 1.75 4.11 0.21 65 Textile yam, fabrics, made-up art. 0.90 1.42 0.98 1.61 2.23 0.68 1.25 69 Manufactures o f metal, n.e.s. 1.63 1.60 0.25 2.45 0.89 1.28 0.87 62 Rubber manufactures, n.e.s. 0.03 0.23 0.15 3.86 0.76 1.54 1.80 35 Electric current 0.00 0.07 0.00 7.53 0.12 0.14 0.00 27 Crude fertilizers and crude materials 0.14 3.73 0.88 0.16 2.49 0.11 0.05 64 Paper, paperboard, articles o f paper 0.52 1.85 0.04 2.60 0.79 0.77 0.53 Memo: 28 Metalliferous ores and metal scrap 2.16 1.94 0.07 0.44 1.03 0.45 0.56 Source: Based on partners' data fromUNCOMTRADEStatistics. Note: The table includes only products inwhich the average RCA for the group exceeds unity. 2.19 Unlike experiences in other CEECs, BH witnessed limited restructuringin factor intensities of its exports. BH is not unique among transition economies in relying heavily on exports o f natural resource-based and unskilled labor-intensive products. For example, in 1999 these groups constituted 37 and 40 percent, respectively, of total Lithuanian exports to the EU. However, while during 1992-99 Lithuania managed to double the combined share o f skilled- labor-intensive and capital-intensive products to 24 percent (Smarzynska, 2001), there is little evidence o f such adjustment in BH. Second, while exports o f products intensive in capital and skilled labor doubled and quadrupled, respectively, since 1997, albeit from a very small base, their combined share inEU-destinedsalesremainedvirtually unchanged at about 13percent over this period. 2.20 The existing composition and factor intensity of exports may not prove a source of sustained export growthin the longrun. The factor content o f exports matters for two reasons. First, since industries intensive in skilled labor and capital tend to pay higher wages, a growth o f exports in these sectors would boost output growth and help improve living standards. Relying on exports o f natural resource-based products which involve little processing, such as lumber or bovine hides, by contrast, would not have the same effect on wages. Second, as illustrated in Chapter 5, labor costs in BH are relatively high, at least in the formal sector. Therefore, a high 38 share o f unskilledlabor-intensive exports, such as textiles and footwear, may not be sustainable inthe long run, given the growing competition from low-wage countries inAsia and elsewhere. It shouldbe noted, however, that Chapter 5 also highlights the recent increase inemployment for lower wage jobs, so that the labor supply may not be as inelastic as previously assumed. In consequence, this market segment could well represent a source o f additional growth in the mediumterm. 2.21 More importantly,few of the labor-intensiveproductsin which BH specializeshave exhibited dynamic growth in EU external imports in recent years. Appendix Table 3A, lists BH sunrise exports to the EU(Le., products for which the EUimport demand has experienced the fastest growth over the past decade. As illustrated in categories with at least US$lOO,OOO worth o f exports in2002, sunrise products accounted for 15 percent o f all EU-destined sales. The list was topped by furniture, which has been an engine o f export growth in many transition countries and in BH accounted for almost one-tenth o f sales to the EU. It is noteworthy that apparel and footwear, the key groups o f BH exports, are absent from the list. This suggests that other types o f products may offer better prospects for future BH exports to EU markets. As described below, one important factor explaining the features of BH's exports i s the pattems o f integration of BH's producers into international production and distribution networks. InternationalProductionandDistributionNetworks 2.22 The increasing globalization of the world economy and the emergence of internationalproduction and distribution networks have changed the economic landscape facing nations, industries and individual firms. Such networks, also known as commodity chains, refer to the whole range o f activities involved in the design, production and marketing of a product. There exist two distinct types o f suchnetworks: "buyer-driven" and "producer-driven" commodity chains (Gereffi, 1999,p.41-42). `Buyer-driven commodity chains ' 2.23 BH, along with other transition economies in the region, has been participating in EUbuyer-drivencommodity chains.These exist inindustries inwhich large retailers, branded marketers and branded manufacturers play the pivotal roles in setting up decentralized production networks in a variety o f exporting countries, typically in developing or transition economies. This pattern has become common in labor-intensive, consumer goods sectors such as apparel, footwear, toys, house wares, consumer electronics, etc. Production is generally carried out by tiered networks of developing andtransition country contractors that make finished goods to the specifications o f a foreign buyer. Companies inthe EUmaintain these types o f production and outward processing trade (OPT) arrangements, also known as lohn arrangements, with f i m s inEastern and Southern Europe, particularly inthe apparel and footwear industries. Under such arrangements, companies in transition countries use inputs obtained from the EU buyers to assemble a final product that i s then exported back to the EU. The patterns o f integration o f BH producers in intemational production and distribution networks detailed below partly explain the lack of restructuring and the dominance o f unskilled labor intensive exports outlined in the previous section. 39 2.24 While the involvement of BH companies in OPT trade has been limited so far, operations with the EU have been concentrated in apparel and footwear. As indicated in Table 2:8, lohn exports from BH increased between 1996 and 1999, followed by a small decrease in 2000, after which the volume more or less stabilized. Increasing wages in BH have made the OPT a less attractive option for EU buyers. In terms o f the share o f OPT in total exports to the EU, BHranked fourth behind Albania, FYR Macedonia and Romania. While OPT could provide an additional short-run impetus to export growth, some o f the most dynamic transition economies, such as Slovenia, Hungary and the Slovak Republic, have exhibited a below-average reliance on lohn arrangements, as they have been able to increase other types o f exports (Figure 2.7). Among 13 categories with lohn exports exceeding US$lO,OOO in 2001, three belong to the apparel/textiles group, which, together with footwear constituted a vast majority o f the OPT trade, accounting for over 90 percent o f categories listed in the table (Appendix Table 2). Table 2:s ExDorts to the EUunder Lohn Arrangements. 1996-2002 IIn million US! 2001 2002 2002 1996 1997 1998 1999 2000 2001 2002 997=100 1997=100 ;H= 100 Romania 701 868 1,106 1,230 949 998 1,042 115 120 2,105 Czech Republic 928 987 864 754 663 606 749 61 76 1,512 Poland 1,380 1,452 1,341 1,140 916 722 644 50 44 1,300 Hungary 981 1,086 1,013 951 783 711 616 66 57 1,243 Slovak Republic 273 326 282 297 239 242 210 74 64 424 Estonia 89 104 160 256 73 57 186 54 179 376 Bulgaria 193 229 301 307 324 358 161 156 70 325 Albania 122 113 142 148 146 178 152 157 134 306 Slovenia 167 216 179 131 101 96 141 44 65 284 Lithuania 146 215 242 230 183 138 113 64 52 227 Croatia 222 293 294 222 186 148 105 51 36 213 Macedonia, FYR 70 100 114 131 106 111 84 110 84 170 Latvia 50 85 97 96 92 85 69 101 82 140 Serbia and Montenegro 42 121 171 84 90 69 54 57 45 110 Bosnia- Herzegovina 5 33 55 65 48 46 50 142 152 100 Source: COMEXT database. 2.25 Industries operating under lohn contracts provide little evidence of quality upgrading26especially in the apparel and footwear industry. The unit values (the amount paid per ton) received by BH subcontractors participating inthe EU-centered lohn arrangements 26However, unit values are imperfect measures, as they may reflect a change in the composition o f exports rather than price received per unit, especially when used with broader product groups such as the two-digit H S classification. 40 in textiles do not compare favorably with those in other countries in the region.27 BH ranked fourteenth among 15 comparator countries in terms o f the unit value o f its lohn exports o f textiles inboth 1997 and 2001. While in2001 Slovenian subcontractors received on average US$ 38,000 per ton o f textile/apparel exports, BH companies were paid less than US$7,000 per ton. Further, all o f these countries except Latvia registered a decrease in unit values. This can be explained either by changes in product mix towards higher weight/lower value products or by growing competitive pressures associated with the depreciation o f East Asian currencies and the increasingpresence o f cheap exports from China and other Asian countries inthe world markets. While the data available do not allow us to distinguish between the two scenarios, the anecdotal evidence suggests that the latter i s probably the case (Box 2:2). Figure 2:5 Share of OPT in Total ExDorts to the EU I 100 , I u1996 8 1997 0 1999 2000 02001 8 2002 2.26 While in the short runBHhaspotentialfor increasingitslohntradewith the EUfor apparel, footwear and leather products, these sectors will only represent a longer term source of growth if BH products can move up the value chain. The patterns observed inthe comparator countries suggest that there i s room for increasing BH involvement in lohn arrangements with EUbuyers, particularly considering the country's proximity to the European fashion centers in Italy. However, BH subcontractors may expect that, if labor costs continue to rise, BHwill become a less attractive destination for OPT trade, as has been the case with many Central European countries. Other factors that will put competitive pressure on BH subcontractors inthe near future are the end o f the Multifiber Arrangement (MFA) on January 1, 2005, which is likely to result in increased exports from low wage Asian countries, as well as from improvements in the business climate in the former Soviet Union where the wage levels have remained lower than inCentral Europe. 2.27 BH participation in the EU-based production and distribution networks for furniture has also been strong.'* As illustrated in Appendix Table 4A, many comparator ''It should be noted that the values in Figure 2.7 differ from those presented in Table 2.8, as they are based on different definitions. In the former case, the data o n total imports entering the EU and qualifying for Statistical Procedure 7 (i.e., imports "after outward processing [textiles]") are used. Inthe latter case, the figures for particular three-digit H S product groups are employed. 28Furniture supply chains would clarify as buyer-driven commodity chains when the principal customer is a retailer (see example in B o x 2.3). They do not, however, qualify as lohn arrangements since the input (wood) originates in BH, 41 countries have actively participated in the network, the fumiture sector played an important role inpromoting exports from Easternand Southern Europeantransition economies, exporting final products as well as parts and components to the EU. BH is no exception in this respect having increased its exports o f hrniture from almost US$4.4 millionin 1998 to US$9.6 million in2002. Similarly, BH exports o f furniture parts and components (mostly parts o f chairs and seats) to the EUjumped from US$3.9 million in 1998 to US$47.7 million in 2002. In 2002, the network exports accounted for 16 percent o f BH sales o f manufacturing products (excluding chemicals) to EU markets, up from 4.3 percent in 1998. The corresponding figures for 2002 ranged from 0.01 percent in FYR Macedonia and 2.3 in Hungaryto 8.5 in Croatia and 9.7 in Slovenia. The high shares in the ex-Yugoslav countries reflect the long tradition o f fumiture production and exporting experience acquired before the break-up o f the country. Many o f the companies currently involved in the network supply products had long-term cooperation with the partner firms dating back to the early 1990s, or even earlier. Producer-driven commodity chains 2.28 In contrast to many of the comparator countries, BH's participation in the producer-driven networks has been very limited owing to small FDI inflows. Producer- driven commodity chains are those in which large multinational corporations play the central roles incoordinating the production process. They emerge as a result o f the presence of FDI and they are mainly present in capital-intensive and skilled labor intensive industries such as automobiles, aircraft, computers, semiconductors and heavy machinery. The automobile industry offers a classic example o f a producer-driven chain, with multilayered production systems that involve thousands o f firms, including parent companies, subsidiaries and subcontractor^.^^ Automobile production networks centered among multinational corporations have played a prominent role in shaping the exports o f Central European economies. Even though one o f the major players in the automotive industry has entered the country, BH's involvement in this network has beenvery limited to date. 2.29 Despite a positive trend in FDI inflows since 1998, the foreign presence in the country is still too small to allow for the emergence of producer-drivennetworks.The total amount o f FDI received between 1998 and 2003 was equal to US$l billion, which exceeds the amounts received by FYR Macedonia and S A M , where the corresponding figures were US$0.8 billion and US$0.9 billion, respectively. However, performance inthis area has been dwarfed by that o f Croatia, which during the same period was a destination for almost US$6 billion o f FDI. When FDIinflows are expressed as a share o f GDP, BH has again experienced a positive trend, increasing from 1.6 percent in 1998 to 3.8 percent in 1999 and 6.1 percent in 2002. Moreover, in 2002 the ratio was the highest among the four ex-Yugoslav republics included. .30 29For example, the average Japanese automaker's production system comprises 170 first-tier, 4,700 second-tier and 31,600 third-tiersubcontractors (Hill, 1989, p. 466). 30Based on recent survey data, the Central Bank o f Bosnia and Herzegovina estimates FDI flows to have reached 1.6 billionUSD by mid-2003, which lies above administrative and IMF estimates. 42 Box 2:2 FlexibleProduction in the Apparel Industry Competition from cheap producers in developing countries inthe textile and apparel industry has been increasing the pressures facing firms in industrialized countries. While relatively l o w wages currently assure the competitiveness o f transition economies in these sectors, preserving the competitive edge may become increasingly difficult as the labor costs continue to rise. Thus, rather than fighting an uphillbattle o f competing purely o n the basis o f costs, transition economies, including BH, would benefit from repositioning themselves into a higher value added segment of the industry. The apparel industry can be thought o f as consisting o f three tiers (i) lowest value added segment - standard the products, such as generic white dress shirts; (ii) the middle value added segment - time-sensitive products with some fashion content; (iii) the highest value added segment -top fashion designer clothing. The highest profits are to be made indesigning and marketing rather than in apparel production. However, it may take a while for producers intransition economies to buildup a reputation that will allow them to supply the second tier o f the market under their own brand names. Thus, an alternative strategy for these producers is to move from supplying the lowest value added segment to supplying foreign retailers inthe middle tier. One of the recent developments inthe middle tier o f the apparel industry is product proliferation and shorter product cycles, which are reflected in quickly changing styles and product differentiation. These developments have contributed to general demand uncertainty for both retailers and manufacturers, making demand forecasting and productionplanning more difficult. Ina world where producers must supply an increasing number o fproducts containing time-sensitive fashion elements, speed and flexibility have become crucial. Instead o f making planning and production decisions based on forecasts and guesses made months inadvance o f a selling season, firms now receive ongoing orders reflecting actual consumer purchases. Thus, suppliers must be able to provide frequent deliveries, in smaller quantities, and o f more diverse products. Moreover, suppliers are expected to do this with a far greater level o f accuracy in fblfilling orders and 1meeting delivery standards than in the past. In short, the retail revolution has altered the basic rules o f global competition for the apparel and textile industry. Inorder to participate inthis new system, a supplier mustbe able to: Label, track and respond to product orders inreal time on the basis o f style, color, fabric, and size Exchange (send and receive) information concerning the current status o f a retailer's products on an electronic basis Provide goods to a retailer's distribution center that can be efficiently moved to stores - that is, containers marked with bar codes concerning contents; shipments o fproducts ready for display inretail stores. There i s evidence suggesting that firms inMexico and the Caribbean have been able to take advantage o f the growing importance o f timeliness. A study by Evans and Hanigan (2003), based on product-level information from U.S. department stores, has demonstrated a shift inthe location o f producers away from lower wage locations in Asia and toward higher-wage locations inMexico and the Caribbean where timeliness is important. The proximity to the EUplaces BH at a great advantage and makes it a primary candidate for becoming suppliers tc middle-tier apparel retailers. Moving up this route requires some investment inboth physical and human capital. Yet. as illustrated in Box 4, this i s not beyond the reach o f BH companies. While following this route carries some risks. not doing so may be an even more uncertain path. Source: Based on Abemathy et al. (1999). 43 Figure 2:6 Net FDIInflows into Successor Figure 2:7 Net FDIInflows into Successor States States of Yugoslavia (US$ million) of Yugoslavia (percent of GDP) 1997 1998 1999 2000 2001 2002 1997 1998 1999 2000 2001 2002 1--CBiH +Croatia I ,--CBiH +Croatia 1 ~ ~ Macedonia =-a+*Serbia& Mn. Macedonia -* 9~Serbia & Mn 2.30 Yet evidence of interest of European firms in BH is emerging. EU-centered or regional production networks are likely to be created by FDI from the EU member states and neighboring countries. Twenty-nine percent o f all inflows between 1994 and 2002 came from three EU countries: Germany, Austria and the Netherlands. FDI from countries in the region accounted for one-third o fthe total. 2.31 Despite the limited presence of foreign investors in the country, there is anecdotal evidence of some positive developments. For example, a local company specializing in the processing o f h i t s and vegetables, privatized in 2002, has managed to become a supplier o f Europeansupermarket chains, thanks to the market research and contracts provided by its foreign shareholder. Realizing that the relatively small scale of production and the relatively high labor costs in BH would prevent it from competing primarily on price, the company chose to target niche markets and supply labor-intensive yet specialized products which command a price premium. Similarly, it was noted in an earlier section that two types o f automotive parts were found among BH sunrise exports to the EU. Most likely these parts have been produced by Volkswagen, which recently has engaged (albeit to a very limited extent) inthe manufacturing o f car parts and components for its sister companies inother Europeancountries. 2.32 The patternsof trade analyzedabove andthe experiences of other countriesof CEE suggest that by integratingwith bothbuyer-drivenandproducer-drivencommoditychains, BH could achieve more rapidexport growth. Through the former, BHcould aim at sustaining the growth in unskilled labor intensive products such as apparel and footwear. The latter however will be necessary for BHto restructure its export composition and move towards skilled labor intensive and capital intensive products. 44 2.33 The fact that a substantial share of FDI Figure 2:s FDIInflows 1994-2003 by Sectors inflows have gone into the Non-banking manufacturing bodes well financial for the creation of enterwises -Other I production networks in the - -Manufacturina future. At the same time, sizable inflows into services (17 percent o f FDIgoing into Banking! Trade banking and 6.8 percent into other services) (Figure 2.8) I I I -Tourism should help improve the Other LTransport business environment by services enhancing the quality o f the Source: Central Bank BH services infrastructure. Moreover, by increasingthe amount o f total FDIreceived, inflows into services and other sectors serve as a signal o fpositive developments inthe BHbusiness climate. 2.34 Nevertheless, the current pattern of "buyer driven commodity chains" exports dominatedby unskilledlabor and naturalresourceintensive product cannot alone deliver the targeted growth rate in exports. First, pressures for wage increases and mounting competition from cheap Asian goods (especially following the expiry o f the MFA) is likely to hamper developments o f labor intensive sectors in the long run. Second, weak certification capacity and inadequate management systems still undermine the ability of natural resource- intensive products from reaching reaping the benefits o f market access, especially to the EU. Overcoming these hurdles will require measures to increase the flexibility o f the labor market, improve the investment climate and establish effective institutions facilitating exports. Only by creating the enabling environment that would attract FDI inflows and move up the value chain, would BHbe able to sustain and even increase its exports. 2.35 Enhancingparticipationin the buyer-drivencommoditychains couldtake three not necessarily exclusive paths. Inorder to make exports o f apparel, textiles and footwear a viable strategy in the long term, BH companies will need to climb up the value added ladder. Such upgrading can take several forms (i) moving from cheap to expensive items, from simple to complex products, from small to large orders; (ii)moving from the mass production o f standardized goods to the flexible production o f differentiated merchandise; or (iii) shiftingfrom the simple assembly o f imported inputs to more integrated forms o f original equipment manufacturing (OEM) and original brand name manufacturing (OBM). The first option will probably have to involve cooperation with large European retailers or their suppliers. Retailing across the EU has been marked by substantial concentration during the past decade. And the share o f specialty chains, franchise networks and hypermarkets has been rising rapidly (Gereffi, 1999). The second option o f engaging in flexible production, described in detail inBox 2, holds the greatest promise for BH, given its geographical proximity to the EU. Flexible production aims at reducing the typical production cycle o f fashion merchandise from as much as nine months to a few weeks. Flexible production would allow BHproducers to stay competitive even as wages continue to increase. For example, most large American manufacturers o f apparel and 45 textiles manage to continue operating thanks to the flexible production programs that they offer their customers (Jones, 1995). The third option is to adopt the OEM model. Inthat model, the supplying firm makes a product according to the design specified by the buyer, and the product i s sold under the buyer's brand. However, the supplier and buyer are separate firms, and the supplier lacks control over the distribution. In contrast to loan arrangements, in this model the supplieris responsible for purchasingthe inputs. An admittedly muchmore difficult step up from the OEM model is to become an OBM, which involves the designing and selling o f ownbranded merchandise indomestic and foreign markets (Gereffi, 1999). 2.36 Attracting FDI is also necessary to integrate into producer-driven commodity chains and to facilitate exports o f skilled labor-intensive and capital-intensive products. Attracting FDI inflows may, therefore, serve as a way o f upgrading the production base o f the ~ountry.~'While both buyer-driven and producer-driven supply chains offer access to foreign markets and thus facilitate exports, participation in the latter may be associated with higher knowledge transfer to the host country. Local firms involved in lohn production arrangements are often engaged in very simple assembly operations and thus receive little knowledge transfer from their foreign buyers. FDI inflows, on the other hand, bring new technologies, as well as management and marketing know-how into the country. Further, the presence o f FDI may be associated with knowledge spillovers boosting the productivity o f domestic firms. Constraints and policy options to improve the investment climate are reviewed inthe subsequent chapters o f this report. D.POLICYRECOMMENDATIONS: THEWAY FORWARD 2.37 BH's participation in international production networks should be based on a two- pronged strategy: (i) climbing up the value chain by engaging in flexible production for exports o f apparel and footwear to remain sustainable in the future, and (ii) using FDI inflows to integrate into producer-driven commodity chains and facilitate exports o f skilled-labor-intensive and capital-intensive products. The report analyzes the priority policy measures and actions required for the implementation o fthis strategy. 2.38 The analysis in Section C highlightedthree main areas for which policy action i s needed to promote the integration o f BH's enterprises into EUbased supply chains and improve export prospects. First, removing constraints facing exports requires a focus on internal rather than external obstacles. Indeed, while BH benefits from unrestricted access to EU markets, in many cases local producers are unable to take advantage o f the preferences owing to constraints imposed by domestic policies and other behind-the-border measures. Second, attracting FDI and engaging in cross border operations requires assured market access and stable rules governing trade and investment. Third, expanding buyer driven commodity chain operations, including OPT trade, requires avoiding the wage escalation that could undermine the competitiveness o f BH's operators. This section suggestsconcrete actions and measures aimed at tackling the above challenges. 3' See Kaminski and Smarzynska (2001) for the evidence from Poland, Kaminski and Riboud (2000) for the case o f Hungary, and Kaminski and Smarzynska (2003) for the Slovak experience. 46 Lifting Internal Constraints 2.39 Domestic policies should focus on lifting internal constraints to exports with three priority areas. First, BH needs to maintain a liberal trade regime conducive to export growth. Second, it needs to improve the investment climate, including through further progress in establishing the single economic space. Third, there i s a need for facilitating trade and consolidating the capacities o f customs administrations. Fourth, BHneeds to continue upgrading the standards andnorms legal and institutional infrastructure. 2.40 Maintaininga liberaltrade andtariff regimeis essentialfor boostingBHexports.As explained in section B, BH enjoys a simple tariff and trade regime. Inthe context o f the ongoing reform o f the MFNtariff schedule, BHwould benefit most from keeping its tariff rates low and even considering reducing them further. Reductions will be particularly helphl for tariff rates on final products and main intermediate goods as a way o f avoiding tariff escalation or any other anti-export bias. 2.41 Improving the investment climate, and especially resolving problems related to ownershipand arrears, is necessary in order to boost BH exports.These issues are discussed ingreater detail inChapter 4. The lack of access to credit resultingfrom higharrears, inflexible labor laws and unresolved ownership status constrains the ability o f many companies to finance investment and produce higher value-added goods. Similarly, deficiencies in the business climate, such as inflexible labor laws, have driven some entrepreneurs into the informal sector which hampers their ability to engage in foreign sales. 2.42 Strengthening domestic competition and the single economic space is another key to achieving a stable and predictable business environment. As part o f the SAP, BH is working to harmonize with the EU while establishing a single economic space within its own borders. The SAP calls not only for a viable state capable o f implementing and enforcing laws and regulations compatible with the acquzs communautaire, but also requires the creation o f a single internal market. A single internal market can be established through synchronized convergence to the acquis at the level o f the two entities through the removal o f all measures impeding trade within BH as well as with the EU and with countries with which BH signed bilateral free trade agreements (FTAs). Proceedingrapidly in establishing the single economic space would remove any uncertainty and provide greater clarity on the legal and institutional framework for trade and investment inBH. 2.43 Strengthening institutions supporting trade facilitation will help ensure successful integration into global production networks to provide for cheap, efficient and rapid linkages required under flexible production systems. Trade facilitation encompasses the entire environment inwhich trade transactions take place.32Inaddition to the logistics o f moving goods through ports or the required documentation at customs posts at borders it also involves domestic policies and institutional structures. Given the relatively high import content o f BH's products, higher customs and logistics costs translate directly into higher costs that could undermine the competitiveness o f BH's products in both domestic and international markets. 32For further information, see World Bank(2003~). 47 Long delays at the border and high variance in clearing times make it difficult for potential exporters to commit to a particular delivery time. Corruption and the need to paybribes at border crossings increase the costs o f doing business, thus lowering the competitiveness o f local suppliers. Box 2:3 BusinessClimateas a Constrainton Exports The case o f a large furniture manufacturer inRepublika Srpska provides an illustration o f how the poor business climate can hinder exports. Labor regulations and unresolved ownership status are the key factors affecting the performance o f the firm in question. The company carries o n its books 550 employees o f whom only 150-250 actually work. The company has been in arrears on pension contributions since 2000, which makes its workers ineligible for retirement. Moreover, political constraints prevent the company from laying o f f employees. There is little hope that the company will be privatized, as its total debt exceeds the market value o f its assets by more than 50 percent. The company i s in debt to the state, the IFC, and a commercial bank. Moreover, it has wage arrears and owes money to its suppliers. Currently, a large portion o f its transactions takes the form o f barter. The unresolved labor situation leads to a vicious cycle. Keeping the waitlisted or idle workers o n the books raises the arrears vis-a-vis the state and thus decreases the chances of privatization. While the company's experience and reputation would allow it to receive additional orders from large multinationals, such as IKEA, hlfilling such orders would not be possible because o f a lack o f working capital. T o illustrate the situation, in the 1990s the company used to sell one container o f products daily to IKEA. Currently, two containers are shipped per month. Obtaining a bank loan is out o f question, given the high indebtedness o f the comDanv. 2.44 Many positive developments have taken place at the level of customs. Customs clearance time has been reduced. For example, the average time required to clear goods at the Grude terminal in the FBH dropped from 154 minutesinMay 2001 to 49 minutes inNovember 2003. Moreover, 6-17 percent o f trucks were cleared inless than 15 minutes during the last three data collection efforts, compared with none initially. At the Raca border crossingpoint inthe RS, the time required for clearance first increased from 89 minutes in June 2001 to 492 minutes in August 2002, but then improvedto 15 minutesinNovember 2003. 2.45 BH also compares favorably with countries in the region in terms of frequency and value of bribes paid at border crossings. In 2002, 32.5 percent o f trucks crossing the border made side payments in BH compared to 47.7 percent in FYR Macedonia and 42 percent in Albania. There remains, however, much room for progress, as for instance in Bulgaria only 11.4 percent o f trucks were allegedly involved in such an activity. The average payment made in BH was equal to 53 euros, which, while being a burden to exports, is still relatively low by regional practice (Figure 29). 2.46 A critical priority is to pursue customs reforms and ensure that the customs administrations of the FBH and the RS are able to transfer documentation and data between their systems electronically. Currently, owing to incompatibilities between the systems used by the two entities, the bulk o f documentation is exchanged in paper form, which causes delays o f up to 40 days in refunding the guarantee payments made by importers residing inone Entitybut transporting their productsthough the territory of the other Entity. Moreover, implementing the TIR agreement which was signed in the mid-1990s is also a priority. By 48 reducing formalities associated with crossing international borders, delays and costs to parties engaged ininternational trade would be reduced. Figure2:9 CorruptioninCustoms - % of trucks making payments + A 4 bribe paid (EUR) 1 Source: World Bank (2003). Note: Figures for Romaniaare basedon a differentmethodologyfrom that usedfor other countries. 2.47 Standardization and norms. BH exports to the EU have been hinderedby the lack of institutions able to certify potential exports as meeting the EU standards. This issue has been particularly relevant to exports o f animals and animal-based products, which have not been able to enter the EU, as the recently established State Veterinary Office i s not yet fully operational. The lack o ftesting facilities and other bodies issuing standards certifications increases the cost o f obtaining such documents for potential exporters. The fact that only four local companies currently have CE certifications is telling. 2.48 Upgrading the standards and norms infrastructure should be made a priority in order to avoid turning standards into a barrier to exports.As the EUconstitutes the largest market for BHproducts, and as neighboring countries are striving to comply with EU standards, BH needs to urgently upgrade its legal, regulatory and institutional framework related to standardization, conformity assessment and accreditation. First, it should make the State Veterinary Office operational and assure compliance with E C Food and Veterinary Office recommendations in order to facilitate exports o f animals and animal-based products to the EU. Second, BH needs to develop technical capacity in the field o f standardization and conformity assessment and to strengthen the institutional infrastructure by harmonizing legislation in line with the EU acquis communautaire. Third, creating the legal framework allowing for the establishment o f accredited laboratories able to conduct the required testing and issue certifications will be crucial for lowering the costs o f compliance on the part o f exporters. 49 Pursuing regional and global integration 2.49 Besides liftinginternalconstraints,the accelerationof externaltrade integrationhas become more important in light of recent changes in BH's external en~ironment.~~ Such integration encompasses three dimensions (i) a European dimension involving BH's relations with the EU in the context o f the SAP; (ii) regional dimension covering its relations with a neighboring SEE-8 countries; and (iii)multilateral dimension through accession to the WTO. a The challenge is to manage these intertwined processes ina coherent and coordinated fashion, by adopting the necessary reforms and policies to take advantage o f the opportunities provided while minimizing the possible distortionary effects o fpreferential trade arrangements. 2.50 Embracing the three processes will allow BH to gradually remove several of the constraintsidentifiedin the previoussection.First, the processes o f EUintegration and WTO accession will introduce a wide-ranging institutional, legal and regulatory reform that will improve the quality and capacity o f the institutions supporting exports, namely, the standards regime and customs. Second, attracting FDIfor producer driven commodity chains requires wide market access, as decisions by foreign companies to invest and locate will be driven by consideration o f the country's market size and its ability to serve other countries through free trade. By integrating regionally and multilaterally, BH would gain greater access to external markets and would thus increase its effective market size, providing greater incentives for FDI inflows. The EuropeanDimension 2.51 Progress on the SAA should accompany BH's recent advances in liberalizingtrade with neighboring SEES.Theory and evidence indicate that regional trade liberalization yields greater benefits if it i s accompanied by multilateral liberalization, and if it involves deeper integration with highly developed partners like the EU.34BH i s already implementing FTAs with Croatia, FYRMacedonia and SAM, and is inthe process o f ratifying others. The authorities should focus on their commitment to fulfilthe 16requirements for entering S A A negotiations set forth inthe Feasibility Study released bythe EC inNovember 2003. 2.52 While it will not necessarily extend the formal market access provided under the ATPs, the SAA provides for a range of other benefits and potential dynamic gains. The conclusion o f the S A A i s an important step that BH authorities need to pursue actively for two other reasons (a) the SAA i s a contractual arrangement providing more security o f market access than the ATP which involve a unilateral EU action; and (b) eligibility for the S A A involves commitments that would allow for the alignment and implementation o f harmonization o f BH legislation and regulation to those o f the EU, which would be helpful to the long runintegration o f BHinto the European structures. 33All western Balkan countries are signatories to the M O U for Trade liberalization and Facilitation, and are inthe process o f negotiating or implementing their SAAs with the EU. Moreover, with the exception o f S A M and BH, all countries are WTO members. 34World Bank, 2000. 50 2.53 The conclusion o f an S A A with the EU will depend on progress along a wide range o f political and economic issues closely intertwined with WTO accession. On the economic front, the commitments for harmonization o f legislation and regulations, as well as for liberalization o f markets and increased competition, will be compatible with and, inmany cases, identical to what the government would have to do to secure WTO accession. Thus, while participation in two concurrent but separate negotiations will undoubtedly challenge the government's institutional capacity, the actual economic reforms involved would be closely related. 2.54 Intra industry trade, exports and FDIinflows could be promoted if diagonal cumulation o f rules o f origins under the Pan European Cumulation o f Origin (PCO) is applied for BHin its exchanges with the EU.35The decision to apply such rules o f origin inBH's exchanges with the EU lies with the latter and is only possible if BH signs a bilateral free trade area, i.e., the SAA with the EU. Diagonal cumulation allows economic operators to use components originating from any o f the participating countries in the PCO without losing the preferential status o f the final product when exported to the EU. The absence o f diagonal cumulation will discourage exporters from countries associated with the PCO system from sourcing their inputs from BH because the latter does not qualify for local content o f the product. This will negatively affect the export prospects o f BH produced intermediary products and will hamper the development o f intra-industry trade.36 Moreover, without diagonal cumulation, BH exporters would be constrained when using inputs from their SEE partners, as only inputs sourced from the EU would be accepted as local content under the bilateral cumulation o f origin rules prevailing inthe majority o f the S A A (FYR Macedonia and Croatia). Agreement among SEE8 countries participating in the network o f bilateral FTAs to allow for diagonal cumulation among themselves, would further enhance the benefits o f the various FTAs, and further strengthen regional integration. 2.55 The application of the PCO system will only be available following concerted efforts to strengthen the capacities of both Customs and local BH producers. Rules o f origin under the PCO system are sophisticated and could be difficult to implement by local customs authorities and by small and medium enterprises o f EU-partner countries.37 Targeted support is needed in order to avoid making these rules an impediment to export growth. This includes streamlining the PCO system into the customs regime and strengthening the capacities o f the border administration in testing and certifying the application o f rules in a speedy and accurate manner in order to guarantee preferential access o f exports to EU markets. A parallel effort is needed to raise the awareness o f local producers and exporters about the requirements for proving origin, and to support them inbearing its cost. 35 EU preferential rules o f origin define the requirements for the local content o f raw materials, components and value added, including the minimum level o f transformation necessary for a product to qualify as originating and thus benefiting from preferential treatment in access to the EU market. Since 1997, a system o f European cumulation, based on a network o f free trade agreements and protocols, has been established among the EU, the EFTA countries, the Central and Eastern European countries and Turkey. Since 2002, the system was extended to the Southern Mediterranean countries as signatories o f Association Agreements with the EU. 36 "Study on the economic impact o f extending the pan-European system o f cumulation o f origin to the Mediterranean partners' part o f the Barcelona process", The Sussex European Institute, University o f Sussex. 37Brentonand Manchin (2003). The cost o f proving origin involves satisfying a number o f administrative procedures to provide the required documentation, in addition to the cost o f maintaining the accounting system an accurately accounts for imported inputs from different geographic locations. 51 Box 2:4 Pan-EuropeanCumulation of Rules of Origin Rules o f origin are used to determine the nationaliterritorial provenance o f goods for trade purposes. The EUhas two types o f such rules: non-preferential and preferential. Non-preferential rules define the origin of a good for the purpose o f such matters as trade statistics and import quotas. Preferential rules, which are often stricter, are defined by members o f regional free trade areas (or other countries that have signed up for preferential trade agreements) to ensure that only those goods which genuinely originate inone o f the member countries enjoy the l o w tariffs or other benefits laid down inthe agreement. All EU preferential agreements contain rules o f origin which must be fulfilled in order for goods to qualify for a preferential treatment. With the exception o f the rules goveming the Generalized System o f Preference, all are negotiated with the partners concemed. Such rules ensure that goods from countries not party to the preferential arrangements cannot be diverted via a preferential partner to circumvent the EU tariff. Under the rules products acquire origin if they are wholly produced or if they are sufficiently processed in the preferential partner. For manufactured goods the basic criterion is a change o f tariff heading at the four-digit level. However, for many products alternative criteria must be observed. For example, there may be a limit on the value o f non-originating materials used, often 40 percent (effectively a 60 percent added value rule), or a rule specifying a process. The concept o f cumulation is an important element o f preferential rules o f origin. It allows inputs originating in one party to a preferential agreement to count as originating in another when those goods are used in further processing. There are various types of cumulation. Bilateral cumulation is the simplest form and applies in preferential trade between the EU and a preferential partner. Diagonal cumulation applies in preferential trade between the EU and groupings o f countries where the parties to an agreement share identical rules o f origin. It allows materials originating inthe EU or in one or more o f the country groupings to count towards fulfillment o f the rules o f origin. Full cumulation represents an extension o f diagonal cumulation. It provides for the aggregation o f all processing, whether or not origin conferring, undertaken anywhere in the preferential area. The concept o f pan-European cumulation introduces diagonal cumulation across a Europe-wide free trade area encompassing the EU, EEA, EFTA, the Central and Eastern European countries, the Baltic States and Turkey. Source: UK Department o f Trade and Industry (http://www.dti.gov.uWewt/rules.htm). The RegionalDimension 2.56 BH should seek to maximizethe benefits available under FTAs. As shown in section B, SEE countries capture an important share o f BH's export markets. The existing and newly signed FTAs provide an opportunity to further consolidate BH's strong positioning in SEE markets and expand its presence beyond its closest next door neighbors. This would be possible by deepening trade integration with its SEE partners and improving further market access for agricultural exports to SEE. 2.57 While market access to other SEEShas significantly improved, gains from intra- regionaltrade hinge first on broadeningintegration.Despite the liberalization o f trade for up to 90 percent of tariff lines and value o f imports, the various FTAs provide for different preferential treatment, product coverage and transitional periods. This creates inconsistenciesand increases the risk o f trade deflection. Inthe short term, mitigating this effect requires broadening integrationthrough further harmonization o f the FTAsto ensure a similar product coverage. This would considerably reduce the costs o f business and the administrative strain on customs, enhancing the flow of BH exports and imports across boundaries. In the medium term, moving towards a multilateral option by establishing a single free trade area among the eligible SEES countries could be pursued. 52 2.58 BH needs to deepen integration through greater harmonization of national regulations and addressing "behind-the-border"barriers to trade. First, cooperation with customs authorities o f other SEE countries on border crossing and customs procedures should be strengthened. Second, and in addition to measures at the national level, regional cooperation in harmonizing technical regulations in line with EU and intemational standards would enhance regional integration. BH could substantially reduce possible technical bamers to trade arising from differing standards and conformity assessment procedures in SEE markets. Collaboration would also allow for a greater burden sharing o f costs in developing the legal and regulatory texts pertaining to standards and technical regulations. Rules goveming various trade areas under the FTAs should always ensure consistency with those o f the EU and the WTO, in order to ensure complementarity and synergy among the various processes. 2.59 Movingtowards a single free trade area would bring substantial benefit to BH and other SEES.Experience from other regional arrangements such as CEFTA and BFTA generated higher intra-regional trade flows and greater export growth when combined with the Europe Agreements (Adam, et. al., 2003). By adopting the measures laid down above for further broadening and deepening integration, SEE countries including BH would on the one had avoid the hub-and spoke inherent in the SAA agreements with the EU and provide greater incentives for attracting FDI. Moreover, a revised single free trade agreement could allow for the further liberalization in the agriculture sector, in services and harmonize legislation related to competition, contingent protection and standards necessary for the facilitation o f trade and investment flows among these countries. Accession to WTO 2.60 The government's pursuit of WTO accession is important for the integration of BH's economy in the world market. WTO membership will provide the authorities with an opportunity to proceed with economic liberalization and upgrade the policies and practices o f BH's institutions with WTO requirements. It will also make market access for BH exporters more secure (through the provision o f unconditional MFNtreatment on BHgoods) and will give BHthe opportunity to use the WTO Dispute Settlement Mechanism to strengthen its capacity to safeguard its trading interests. 2.61 The speed with which the accession process is completed depends largely on the acceding country. The WTO accession process is lengthy and complex and places heavy burdens on the government both in the preparation o f reports regarding the current situation and in the modification of legislation and practices of institutions affecting the conduct of international trade. The government has taken several positive steps by appointing o f a highlevel group to coordinate the overall accession process. The group should seek technical assistance where neededto support the process. 2.62 Compliance with the WTO requirementswill test the government's capacity.There are technical assistance needs in three specific areas (a) revising or introducing new legislation and regulations to ensure consistency with WTO agreements; (b) strengthening the capacity o f institutions dealing with technical standards, phyto-sanitary controls, intellectual property and other such matters to meet the WTO requirements; (c) strengthening o f human resources. 53 E.CONCLUSIONS 2.63 Despite recent improvements in BH's export performance, the merchandize trade deficit has remained overly large. BH's less than stellar export performance has not reflected constrains inmarket access. 2.64 The current structure o f foreign sales raises concerns about the future o f exports. The reliance on exports o f unskilled labor-intensive and natural-resource-based products may not be sustainable in the long run, as labor costs, already high relative to productivity by regional standards, continue to grow.. Thus, to assure the exports sustainability, BH should strive to shift to higher value added exports by taking advantage o f its proximity to the EUand becoming more involved ininternational production and distribution networks, particularly inflexible production arrangements. 2.65 However, such a shift will require improvements to two key aspects which are crucial to firms participating in international production chains. First, a stable and business friendly regulatory framework, which allows entrepreneurs to set up businesses without excessive red tape, provides for the protection o f private property and is characterized by enforcement o f laws and good governance is needed. Second, flexible labor markets are needed to enable firms to improve competitiveness in the face o f adverse shocks by adjusting employment. Third, provision o f efficient and low cost transport infrastructure and a streamlinedcustoms service will facilitate the fast movement o f goods across international borders and within the country. Without the ability to commit to on-time delivery to customers abroad, it will not be possible for local producers to engage inflexible production arrangements. 54 3. ENTERPRISESECTOR STRUCTUREAND PERFORMANCE A. INTRODUCTION 3.1 As the chapter on macroeconomicissues suggests, the low level of domestic savings in Bosnia and Herzegovina (BH) reflects in part disappointing corporate performance. Establishinga favorable investment climate is central to reversingthe recent slowdown of output expansion and building a basis for sustainablegrowth. This will require the creation o f an enterprise sector motivated to prosper by utilizing the opportunities offered by BH's trade regime, as emphasized in the chapter on trade. A major policy challenge for the authorities i s to create an incentive structure for new firms to enter and loss-making firms to exit the BH economy. The objective should be to ensure vigorous inter-enterprise competition takes place and the absence o f opportunities for monopoly power. To achieve these goals, the government needs to provide the proper incentives and market signals byreforming its own actions. 3.2 To understandthe sources of the weak performancein BH's enterprise sector, it i s important to analyze the underlying institutional fabric of the country's business environment. Thus, we organize this chapter as follows. Section B reviews in detail the structure and characteristics of BH's enterprise sector. In Section Cy several dimensions o f the competitive performance o f businesses are analyzed. The chapter concludes by presenting the results o f an econometric analysis o f the determinants o f firm performance in BH, which provides insights that set the stage for the analysis o f the impediments to business development andgrowth that is the focus o fChapter 4. B.CHARACTERISTICSOFTHE ENTERPRISESECTORINBH 3.3 Our analysis o f the enterprise sector in BH distinguishes between the formal and the informal sectors. Within the formal sector, we look at domestic and foreign firms. For the informal sector, we present available data on its size and characteristics and put it in a regional perspective. We thendiscuss the extent o f the international integrationo f BH's businesses. The FormalSector Domesticfirms 3.4 Sectoral and Size Attributes. The majority o f registered firms in BH operate in the services sector, which accounted for roughly two-thirds o f GDP in 2003 (Table 3.1). Moreover, most o fthe country's businesses are o f a small or medium size. 3.5 Firms in agriculture and agro-processing are stagnating, and overall agricultural output hasbeen declining.The agricultural infrastmcture is outdated andpoorlyintegrated, and the quality o f products i s poor. Firms struggling to comply with sanitary and phytosanitary standards are unable to utilize their export potential. In addition, uncertainties regarding land ownership further constrain the development o f the agricultural sector. There is an abundance o f unused land in both Entities: 50 percent of arable land in the FBH and 30 percent in the RS (European Commission, 2003). 3.6 The structure of the industrialsector in BH is quite different from that of the other former Yugoslav republics, remaining dominated by a small number of very large companies (Combinats)with many lines of business.The war affected the operation o f these industrial giants as production collapsed. Industrialproduction has yet to recover to its pre-war levels, and capacity utilization remains low. The decline in manufacturing has meant a deterioration of the quality and production capacity not only of the traditional sectors, such as armaments, steel, and the chemical industry, but also o f clothing, textiles, leather, and wood production. As a result, exports have shifted to low-value-added products and raw materials (European Commission, 2003). Overall, the production o f tradable goods has been poor, and imports haveprimarily provided for increased domestic consumption. Table 3:l Output inBH, 1998-2003 RealGDPGrowthby Sector Value Addedby Sector (percent (percent) GDP) 1999 2000 2001 2002 2003 1999 2000 2001 2002 2003 GDP growthannual percent) 9.6 5.5 4.3 5.3 4.0 Agriculture -3.4 -8.3 0.6 -1.8 -8.9 13.3 11.7 13.0 12.1 10.7 Industry 9.0 5.7 0.5 -1.9 11.1 24.4 24.9 27.8 26.0 27.8 otw: Manufacturing -0.5 7.2 1.8 3.8 6.5 14.9 15.1 12.5 12.3 12.6 Services 14.1 9.4 7.0 10.2 3.5 62.3 63.4 59.2 61.8 61.6 Source: EBRD (2003, p.56), LMF, WorldDevelopment Indicators (2003), national statistical authorities and staff estimates. 3.7 Ownership. Although most firms are privately owned (Table 3:2), the share of the private sector in output was only 45 percent in 2002.38This is a much lower level than in the more advanced reform countries o f central Europe and even the rest o f the SEE region (with the exception o f SAM).39BHhas been unable to utilize adequately private sector activity as a source o f growth despite the slight increase inprivate sector participation inoutput in 2003. 38 The private sector share o f GDP i s calculated by the EBRD using available statistics from both official (government) and unofficial sources. The share includes income generated from the formal activities o f registered private companies, as well as informal activities where reliable information i s available. The term "private company" refers to all enterprises in which private individuals or entities own the majority o f shares. Data from EBRD(2003), p. 126. 39 See Broadman et. al. (2004). For example, the private sector share in GDP in 2002 was 80 percent in the Czech Republic, Hungary, the Slovak Republic, and Estonia, and 75 percent inBulgaria and Poland. See EBRD(2003). 56 3.8 Many privately owned firms are privatizedcompanies.While privatization has been under way for several years, with the sale o f small- and medium-size firms mostly ~ompleted,~' the state still holds majority stakes in the remaining firms. Mass privatization techniques and cash auctions were used to transfer ownership, primarily in medium-size firms. However, the divestiture o f state ownership inthe large industrial firms has been slow. Only 24 percent o f the large SOEs were sold in the FBH (end-2002), and 42 percent in the RS. The privatization o f large-scale strategic enterprises (mining, tobacco, refineries) has yet to take off, with only 18 out o f 56 firms being sold inthe FBH, and 4 out o f 80 inthe RS (European Commission, 2004). Put differently, over 50 percent o f state-owned capital is still to be divested in some 70-80 large strategic enterprises, along with another group of over 350 mid-size firms in which the state is the majority owner. 3.9 Firm Growth.The numbero fregistered industrial firms inthe country increased steadily prior to 2001 (Table 3:2). Since there was a small change in the number o f large firms, the opening o f predominantly small private firms in BH has driven new business creation since 1999. In 2002, however, there was a significant drop in the number o f enterprises in the FBH. This dramatic change is due to an administrative initiative in2001 by the Statistical Office inthe FBHto clean its database o f inactive or fictitious firms. As a result, close to 20,000 firms were dropped from the statistical books inthe FBH. Table 3:2:BusinessCreation, 1999-2003 1999 2000 2001 2002 2003 RS Total number ofregistered firms, all sectors 29,902 31,269 32 245 34 247 na Total number ofregistered firms, industry 3 072 3 337 3 5 16 3 823 na sector ' Total number of employees inregistered 83,264 81,375 76,472 na na industrial firms FBH Total number of registered firms, all sectors 43,220 47,281 49,053 29,779 27,501 Total number ofregistered firms, industry 9 262 10400 10 975 4 220 3,914 sector Total number of employees inregistered 136494 136498 129 879 127 879 na industrial firms Source: FBHand RS Statistical offices. Note: Includes mining, manufacturing, electricity, gas, and water supply. ' 3.10 EmploymentGrowth.The increase inthe number o f enterprises inthe FBHhas not led to corresponding increases in employment in industry. As Table 3:2 indicates, there was a 6.3 percent decline in the number o f full-time employees in the registered industrial firms in the FBH during 1999-2002, and a slightly stronger decrease of 8.2 percent o f industrial workers in the RS during 1999-2001. 40 In2003, some 55 percent of the small and medium firms inRS and some 78 percent o f the small and medium firms inFBHwere privatized. 57 3.11 Data from the EBRD-World Bank Business Environment and Enterprise Performance Surveys-"BEEPSl", which pertain to 1999, and "BEEPS2," which pertain to 2002-can also be used to illustrate these points.41 The surveys shows that job creation in the sample of BEEPS2 firms was entirely concentrated in de novo firms, while the state-owned and privatized firms registered negative rates of employment growth from 1999 to 2002. Compared to the regional trend, surveyed firms in BH fare better injob creation inthe new private sector, but muchworse injob destruction inthe state andprivatized firms, on average. Given that the de novo firms are in general o f a small size, one would not expect a significant net gain injobs for the country. Figure 3:l Changes inAverage Employment by Ownership in BHand SEES, 1999-2002 I State-owned 1 50% 40% 30% 20% 10% 0Yo -10% -20% Bosnia and Herzegovina SEE8 I Source: BEEPS2. Note: SEE8 comprises Albania, BH, Bulgaria, Croatia, FYR Macedonia, Moldova, Romania, and Serbia and Montenegro. Annex 1describes the BEEPS dataset. 3.12 A complementary analysis o fa sample of enterprises inBHfinds that the overalljob flow rate in firms with more than 100 employees was negative inthe period 1997-99; on average, 4.2 jobs were created while 5.3 jobs were lost (World Bank, 2002b). The shedding o f industrial labor signals that some restructuring is ongoing among industrial enterprises. Nonetheless. job creation was insufficient inthe aftermath o f the war. 3.13 Small and Medium-Sized Enterprises (SMEs). These enterprises (firms with less than 250 employees) have played a prominent role in SEE countries, including BH (Table 3:3). Over 85 percent o f the SMEs inBHare micro enterprises, with 10 or fewer employees. While this is in line with the overall regional trend o f pronounced micro-enterprise participation inthe enterprise sector, the number of SMEs (including the micro firms) per capita in BH is the lowest in the region, especially in comparisonwith the more advanced central European countries (Table 3:3). There are only seven SMEs per 1,000 people in BH, compared to 84 in Poland. In terms o f sectoral composition, most of the SMEs in BH are registered in trade (45 percent), construction and manufacturing (27 percent), and services (12 percent). 4` Annex I contains a detailed description o f the BEEPS survey data and the caveats that need to be heeded in interpreting them. 58 Firms in the Informal Sector 3.14 An important characteristic of the enterprise sector in SEE is the comparatively large size of the informaleconomy. Estimates o f the size and characteristics o f informal sectors are notoriously difficult to make and vary widely. In general, the size o f the informal sector i s at least a third o f officially-measured GDP ( Table 354). InBH, the size o f the shadow economy is estimated at over 34 percent o f GDP (2000 estimates). For comparison, the comparable figure i s 16.8 percent for the 21 OECD countries and 8.8 percent inthe United States. Burdensome taxes and social security contributions, as well as onerous government regulations are the main causes for the informalization o f the economy.42 Table 3:3: Structure of SMEs by Sector, 2001 Number of Number Number of Number of Share Share of Share of Share SMEs and of micro large SMEsper of S M E s i n SMEsin of micro enterprise enterprises 1,000 SMEs services' construc- SMEs enterprises S1 1 inhabitant in tion & in 1 S1 trade2 Manufac- other turing' activiti es' Albania 56,237 54,145 253 18.1 52.1 17.8 12.5 17.6 BH 30,000 25,600 200 7.0 45.0 12.0 27.0 16.0 Bulgaria 224,2 11 207,643 741 27.6 50.8 28.9 18.0 2.3 Croatia 63,135 41,988 426 13.7 35.0 7.0 28.0 30.0 Romania 612,862 311,260 1,955 27.4 63.O 16.0 14.9 6.1 Serbia and 66,968 64,002 742 7.8 47.5 5.3 25.8 21.4 Montenegro Czech 876,990 830,601 1,671 85.1 35.0 19.9 23.9 21.2 Republic Hungary 858,981 827,806 1.041 58.9 26.0 50.2 19.9 3.9 Poland 3,368,367 3,206,452 6,589 87.0 35.8 25.6 34.2 4.4 Slovak 365,783 354,373 160 67.7 43.9 29.8 20.6 5.7 Republic Source: EBRD survey of national authorities, European Commission (2002), and UNiECE (2003) Notes: ' Data for Albania, Croatia, the Czech Republic and Hungary are for 2002; for BH, Poland, Serbia and Montenegro, and the Slovak Republic data are for 2001; for Bulgaria and Romania data are for 2000 Data for Bulgaria are for 2000. 3.15 Informal firms in BH tend to be of small or medium size operating in industries producing low value-added products or in services such as retail trade. While the large informal sector does provide jobs, alleviate poverty, and support economic activity, its downside goes beyond the loss o f revenues for the governments. It provides unfair competition to the formal sector and erodes the enforcement o f property rights and contracts. Firms in the gray economy have an advantage over those in the formal sector, as the former do not pay taxes and 42 See Schneider (2002) whose definition o f the informal economy includes "unreported income from the production o f legal goods and services, either from monetary or barter transactions - hence all economic activities which would generally be taxable were they reported to the state (tax) authorities." 59 are not subject to regulation. These disadvantages o f a large informal sector make the formalization o f informal activity a policy priority for the BHgovernment. Table 3:4 Size of the InformalEconomy,2000 Informal Country Economy, percent of GNP Albania 33.4 Bosnia-Herzegovina 34.1 Bulgaria 36.9 Croatia 33.4 Moldova 45.1 Romania 34.4 Yugoslavia 29.1 Source: Schneider ( 2002), p. 13. 3.16 Research to understandwhat drives SEE firms to operate inthe informal sector concludes that since the cost and the difficulty o f doing business are high, the benefits o f operating in the gray economy outweigh the efforts to comply with cumbersome regulations (Broadman et. al., 2004). Inorder to bypass formal regulations, businesses opt to pay bribes to public officials. Data from BEEPS2 confirm this observation: over 45 percent o f surveyed firms in BHreport having paid bribes to public officials in 2002 to obtain licenses, permits, and/or inspection certificates. This has cost firm managers, on average, between 8 and 10 percent o f their time (Gray, Hellman andRyterman, 2004). Foreign Firms 3.17 The limited presence of foreign firms in BH i s reflected in the low inflow of FDI compared to the other transition economies in CEE and even to some of the economies in SEE (Broadman et. al., 2004). While inflows o f FDI rose from US$130 million in 2001 to US$230million in 2002, inflows are still negligible compared to that in the Czech Republic (US$9 billion in 2002) or Poland (US$3.7 billion in 2002). On a per capita basis, BH also fares poorly in comparison with the more advanced transition economies. While the cumulative FDI inflow per capita between 1989 and 2002 was US$198 in BH, in the Czech Republic and Hungary it was US$3,554 and US$2,253, respectively. As Table 3 5 shows, the cumulative FDI inflows to BHare the lowest among the eight SEE countries involume and per capita terms. 3.18 As noted in Chapter 2, the low level of FDI limits the ability of BH firms to reach international goods markets. Low inflows o f FDIreflect not only a limitedpresence o f foreign firms inthe country, but also a dominance o f labor-intensive products inthe country's exports. It also implies the relatively poor participation o f BH's enterprise sector-in terms o f internationalizationand fragmentation o fproduction-in the global economy. 3.19 survey by the Central Bank of BH (CBBH).4 The survey presents a comprehensive snapshot The specifics of the presence of foreiFn investors in BH are revealed by a recent 43The survey covered more than 720 enterprises with foreign investors. Among these firms, 55 percent had foreign investment over KM 100,000. In the majority o f these firms, the foreign investor held at least 10 percent o f the 60 o f foreign-owned firms in BH. Several interesting facts emerge. First, Croatia appears to be the largest foreign investor in BH, accounting for one-third o f the total foreign investment in the sample. Along with Austria and Germany, the three top investor countries accounted for 61 percent o f the inflows o f FDI in BH. Austria alone accounted for 59 percent o f the total foreign direct investment inbanks. Finally, firms in manufacturing and banking are the main recipients o f foreign investment inthe country, followed by firms intrade and other services. Table 3:5: ForeignDirect InvestmentInflows inSouthEasternEurope, 1997-2002 Cumulative Cumulative FDIinflows, FDIinflows FDIinflowsper FDIinflows, Country US$ m per capita, capita, US$ percentof GDP US% 1989-2002 1989-2002 2001 2002 2001 2002 Albania 936 303 66 44 4.8 2.8 BH 753 198 34 61 2.7 4.4 Bulgaria 4390 560 79 55 4.7 2.8 Croatia 6296 1419 316 86 7.2 1.7 FYRMacedonia 935 476 221 50 12.9 2.7 Moldova 849 199 37 25 10 6.6 Romania 9008 415 52 50 2.9 2.4 Serbia and Montenemo 1717 206 20 67 1.4 3.6 Source: EBRDTransition Report 2003, p.65. 3.20 Despite the comparatively low level of FDI, foreign resources play a key role in the development of the enterprise sector through foreign aid and grants as well as remittances, as discussed elsewhere in this study. BH is among the top 20 countries in the world receiving remittances, with remittances equivalent to over 11 percent o f GDP. On a corporate level, the case studies suggest that foreign aid and grants have been a double-edged sword for business development. On the one hand, the resources they provide have helped with post-war reconstruction. On the other, local businesses are reluctant to pick up projects which are not foreign-funded or guaranteed. As a result, foreign-funded projects may well crowd out local projects (Box 3.1). International Integration of BH's Businesses 3.21 Partly as a result of low inflows of FDI, the enterprise sector in BH is characterized by weak competitiveness in both internal and external markets, as revealed by macroeconomic and enterprise data. The supply side response o f the economy has not been sufficient for narrowing the trade deficit and successfully integrating into the world economy. The lack o f productive capacity and the inability to comply with EU import standards prevents the country from taking full advantage o f the autonomous trade measures (ATMs) introduced by the EUinSeptember 2000 and the Free Trade Agreements (FTAs) with the countries o f SEE. equity (owner shares) in the enterprise; 127 enterprises (including 23 banks) accounted for 81 percent o f the total foreign investment inthe country. See www.cbbh.gov.ba. 61 Box 3.1 Foreign-FundedProjectsCrowding out DomesticOnes A Sarajevo-based construction company, established in 1990, was forced to disrupt operations during the war. The company re-started its construction, engineering, and design activities in 1996. The general manager o f the firm shares the fact that the trigger to start operations again came f i o m the financial support for projects which the international community provided for the reconstruction o f BH. This helped the company expand tremendously in the post-war period. Starting with 15 employees in 1990, the company now employs over 230 workers. Inthe post- war period, the company engaged inthe purchase o f three SOEs and opened a concrete-production factory. The company caters mainly to foreign clients, and has generally steered away fiom domestic private and state projects unless they pay in full and in advance. In fact, the general manager shared the fact that he turned down many domestic clients, as "the foreigners come with secure financing and pay more, while domestic clients complain about the costs." There are three to four other companies in Sarajevo o f the size and capacity o f the interviewed company which also work primarily with foreign clients with very similar pricing o f their services. The interviewed manager indicatedthat as long as his company has foreign clients it does not plan to reduce its prices 01 redirect its activities towards domestic customers. The company avoids dealing with the government, too, but for the following reasons: cumbersome bureaucracy, corruption, as well as confusing government procurement procedures make the state an unattractive client for the construction firm. Source: Broadman et al. (2004). 3.22 The enterprise sector of BH is poorly integratedinto internationalproduction and distribution networks. Firms in BH are primarily inward-oriented. For example, over 63 percent o f the surveyed firms inBEEPS2 relied on foreign sources for their supplies o f material input. At the same time, export receipts were 10.6 percent o f sales revenue in 2002, a number lower than the SEE8 regional average o f 12.5 percent. Only surveyed firms in S A M reported weaker export intensity among the eight SEE countries (Broadman et. al., 2004). 3.23 Surveyed firms in BH also fared worse than the average SEE firm regarding their activities innew international markets.44Every fifth surveyed SEE firm exported to new markets between 1998 and 2002, while only 6.6 percent o f the surveyed firms in BH had reached new foreign customers during the same period. The reaching o f new markets by BH companies between 1998 and 2002 i s similar to the international expansion o f firms in Albania and S A M , but considerably lower than that o ffirms from Romania andBulgaria. 3.24 The picture that emerges o f the enterprise sector i s therefore as follows: the share o f the private sector in formal GDP in BH i s among the lowest in the SEE region, reflecting the slow progress in privatization, enterprise restructuring, and new business creation. The vast majority o f private enterprises are small and medium-size, and the state presence in the economy i s appreciable. Unsurprisingly, foreign firms do not play a prominent role on the enterprise scene in BH. In contrast, the informal sector is relatively large. In addition, market integration of the enterprise sector in BH i s poor. This suggests that major structural and institutional barriers to the development o f a vibrant private sector needto be removed. 44Broadman et. al. (2004). BEEPS2 asked the surveyed businesses to respond to the following question: "Has your company exported to a new country since 1998?" 62 c.DIMENSIONSANDDETERMINANTSENTERPRISE OF PERFORMANCE 3.25 Patternsof EnterprisePerformance.To assess the performance ofthe enterprise sector in BH, we look at measures of corporate performance such as profit-to-sales ratios, operating profits/losses, and capital and labor productivity and try to link them with firm characteristics. BEEPS data from 2002 show significant variations in the distribution o f profits by ownership and size among the surveyed firms in BH. Small and de novo firms have significantly higher profit-to-sales ratios than the rest o f the surveyed firms. Moreover, data on the dispersion o f firm profitability in 2001 shows that close to 20 percent o f the surveyed firms in BH are loss-makers or are not profitable. Over one-third o f the SOEs and one-third o f the privatized firms in the sample fall into this category o f loss-makers, as opposed to less than 10 percent o f the surveyed de novo firms. Most o f the surveyed firms indicate a profit-to-sales ratio inthe 1to 10 percent range, regardless o f ownership. Figure3:2: Profit-To-SalesRatiobyFirmCharacteristics, 2001 (percent oftotalfirms) I 1 N e w 9.9 45.5 15 9 I Privatized 30.0 I -1 State 34.8 17.4 Negative or 1 1 - 1 0 % 0 1 1 - 3 0 % 0 More than none Source: BEEPS2 31% Note: Annex 3.1 describes the BEEPS dataset. 3.26 Using BEEPS2 data for 2002, we also perform differences in means tests for firms' profitability (profit-to-sales ratio) by firm size and ownership. We find that there is a statistically significant difference (at 1percent) between the average profitability o f de novo firms versus all other (privatized and state) firms. 3.27 In addition, among the BEEPS surveyed enterprises in 2002, state firms' sales, on average, increasedleast comparedwith the growthin sales of privateand new firms (Table 3:6). Among the firms inthe sample whose sales grew since 1998, the greatest average increase was achieved by the de novo enterprises followed by the privatized firms and then state-owned firms. Given the slow pace of privatization and the limited enterprise restructuring efforts, it i s not surprising that the state firms registered the greatest losses on average intheir sales, which i s in accord with the expectations of poor firm performance in the state sector in a transitional environment. 63 Table 3:6 Sales Growth of Firmsby Ownership, 1998-2002 Changes in Sales Growth since 1998 By Ownership Type: State firms Privatizedfirms De nuvo firms Average 1.o 4.9 12.7 Average Increase 24.75 32.17 34.82 Average Decrease (36.0) (25.8) (32.9) Source: BEEPS2. Note: Annex 3.1 describes the BEEPS dataset. 3.28 Measures o f capital and labor productivity also reveal the causes o f the differences inthe performance o f various types o f firms. Capital productivity is the measure o f how well physical capital (such as machinery and buildings) i s used by firms in providing goods and services. BEEPSdata reportedinTable 3:7 reveal that average capital productivity was almost three times lower in the surveyed state-owned firms than in the de novo firms in BH in 2002. Surveyed privatized firms also performed worse than their new counterparts, although they reported a capital productivity that was two times higher than the state firms in the sample. On average, labor productivity (that is, output per unit o f labor) was lowest inthe privatized firms andhighest inthe de novo firms (Table 3:7). Table 3:7: CapitalProductivity and Labor Productivity, 2002 Ownership Type Capital Labor productivity productivity State firms 1.7 16.9 Privatized firms 3.2 13.0 De novo firms 5.O 26.5 Source: BEEPS2. Note: Annex 3.1 describes the BEEPS dataset. 3.29 The results o f a different enterprise-level survey, conducted among the 1000largest firms inthe FBH and the 1000 largest firm inthe RS, provide another opportunity to disentangle the performance o f the enterprise sector. The data are based on accounting information for 2002 provided by these firms and described in detail in Annex 3.2. The survey focused on firms with over 122,000 in the FBH and more than 109,000 employees in the RS. The median firm employed some 25 (50) workers in the FBH (the RS). The largest enterprise had over 6.700 (3,500) employees inthe FBH(the RS). 3.30 In addition, the surveyed firms have very thin profit margins. If their official financial statements are to be believed, operating profit as a share o f total assets was approximately 2.6 percent in 2002 in the RS and 0.6 percent in the FBH. Roughly one-half o f the firms in the RS were in fact loss making, with average losses twice as large as average profits o f the profitable firms. 64 3.3 1 These data reveal a strong correlation betweenfirm size in each Entity and average gross profit margin.45Smaller firms tend to perform much better than their larger counterparts (Table 3:8). Inthe FBH, the average gross profit margin o f firms with less than 50 employees was 2.3 percent in 2002, a slight improvement over the 2001 average level o f 1.8 percent. Surveyed small firms were evenmore profitable on average inthe RS; their average gross profit margin in 2002 was 2.9 percent. In contrast, performance o f large firms inboth Entities, which generally tend to be still in state hands or recently privatized, worsened. In 2002 these firms reported an average gross margin o f -8.3 percent and -14.8 percent in the FBH and the RS, respectively. Surveyed medium-sized firms inbothEntities were also loss-makers. Table3:8: Average GrossProfitMarginbyNumberofEmployees (in percent) Firm Size, Numberof employees FBH RS 2002 Small (0-49) 2.3 2.9 Medium(50-249) -1.1 -6.7 Large (250 andover) -8.3 -14.8 2001 Small (0-49) 1.8 1.7 Medium(50-249) -1.2 -9.8 Large (250 and over) -6.3 -22.7 Source: AFIP. Note: See Annex 3.2 for data description. 3.32 While average sales increased by a little over 1 percent from 2001 to 2002 in both Entities according to the financial statements o f the firms inthe two data sets, sales revenues are not sufficient to sustain an efficient and healthy financial structure in these firms. Arrears to suppliers and for wages are a particular problem for these firms. For example, wage arrears amount to more than 4.5 percent o f sales among the 1,000 firms inthe RS. Overdue payments to suppliers o f the top 20 percent o f state-owned loss-makers are equivalent to 21 percent o f the total debt o f these firms. In addition, the international exposure o f the surveyed firms is very limited.Exports in2002 were only 8 percent o ftotal sales inthe 1000largest firms inthe RS. 3.33 Determinantsof Differences in Enterprise Performance.Based on the analysis o f the BEEPS survey data and the two data sets o f financial information on the largest firms inthe FBH and RS, we put forward and test the hypothesis that better firm-level performance is associated with an investor friendly institutional environment that reduces obstacles to entry, exit, and day- to-day business operations. The results can provide insights into the factors that account for the observed differences in profitability and the extent to which higher firm profitability is associated with elements o f market structure and firm conduct that reflect the potential to exercise market power, controlling for other firm characteristics. 3.34 Annex 3.3 describes in detail the specific dataset we employ and presents descriptive statistics. Usingordinary least squares (OLS) the following simple model is estimated: 45Gross profit margin for each surveyed company is the ration o f annual gross profit to total annual sales, based on BEEPS2. 65 Pi is the performance o f firm i as measuredby gross-profit-to-total sales ratio Ai i s the competitive structure o f the marketplace o f firm i measuredby market share, vertical integration, andhorizontal integration B measures the softness o f budget constraints o f firm i estimated through various types o f i arrears Ci is the perceivedsecurity o fcontract rights by firm i measured by sales on credit Diis the accessibility o ffirm i to financial resources Fi is the characteristics o f firm i, such as ownership category, capturing state, privatized and new firms as well as size measured through the volume of sales in2001 i s a standard error term. 3.35 The results o f the multivariate linear regressions are presented inAnnex 3.3 and in more detail inBox 3:2. The models explain between 28 and 33 percent o f the variation inprofits as a percentage o f sales. The results are statistically significant. Regression (4) i s the one with the best fit. 3.36 To summarize, regression estimations indicate the following: Entryand exit barriershave a significant negative impact on a firm's performance inBH,controlling for firm-specific characteristics. Vertical integration influences positively firm performance, and so does market share, after controlling for firm size and horizontal integration. Greater horizontal integration does not increase profitability. Arrears, as proxies for the softness o f budget constraints, are ingeneral negatively associated with firm profitability. Poor access to credit impedes the performance and growth o f firms, when firm- specific controls are inplace. Insecurity o f contract rights constrains the performance and growth o f firms. Confidence in contractual relationships i s positively associated with firm profitability. New private firms register a higher profitability than privatized firms, which in tumaremoreprofitable than SOEs. 66 Box 3:2 Determinants of FirmPerformance I The four regressionestimates suggest that both entry and exit barriers have a significant negative impact on a firm's performance in BH, controlling for firm-specific characteristics. The estimated positive and statistically significant coefficients on vertical integration suggest that the market structure significantly influences the overall performance o f the firms in the sample. Although not statistically significant, market share appears to be positively associated with profitability after controlling for firm size and horizontal integration. Horizontal integration is negatively associated with profitability, but it is not statistically significant at 10percent inthree o f the four cases. Nonetheless, a higher degree o f horizontal integration is counterproductive to profitability, ceterisparibus. Onthe exit side, we look at the levels o f different types o f arrears as proxies for the softness o f budget constraints in the sample. Arrears to suppliers and workers, which are the two dominant types o f overdue payments in BH, are negatively associated with firmprofitability. However, tax and utilityarrears are positively, although not statistically significantly, related to firmperformance. Poor access to credit impedes the performance and growth o f firms when firm-specific controls are inplace. In all cases, the estimated coefficients on access to financing are positive and statistically significant, thus suggesting that, for the sample o f firms in BH, the easier access to credit is, the greater the f m ' s profitability is, all other things being equal. Insecurity o f contract rights constrains the performance and growth o f firms. Sales on credit, a measure o f confidence in contractual relationships, is positively associated with firm profitability in a statistically significant way, all other things being equal. This indicates that firms that opt for on-the-spot transactions or require pre- payments, and that do not have faith incontracts, performworse than firms that use sales o n credit. Among the firm-specific characteristics that serve as controls in the regression, the estimated coefficients on the ownership variables are statistically significant with the expected positive signs in the different specifications. D e novo private firms register a higher profitability than privatized firms, which, inturn, are more profitable than SOEs. However, the variable through which we measure size does not appear to be statistically significant in explaining Drofitabilitv. D.CONCLUSION 3.37 The share o f the private sector relative to formal GDP in BH is among the lowest inthe SEE region, reflecting slow progress inprivatization, enterprise restructuring, and new business creation. Small- and medium-sized enterprises comprise the vast majority o f private enterprises inthe economy, while state presence remains substantial. Not surprisingly, foreign firms do not yet play a prominent role. Incontrast, the informal sector is relatively large. Inaddition, market integration o f the enterprise sector i s poor. This suggests that major structural and institutional barriers to the development o f a vibrant private sector need to be removed. 3.38 Empirical analysis o f survey data on BH firms presented above indicates that new private businesses register a higher profitability than privatized firms, which inturn are more profitable than SOEs. The survey data suggest that entry and exit barriers have a significant negative impact on firm performance. While market share and vertical integration appear to enhance surveyed firms' profitability, a high degree o f horizontal integration is counterproductive. Arrears as proxies for "soft budget constraints" reduce firm profitability, as do poor access to credit, and insecurity o f contract rights. While these results must be interpreted cautiously in light o f the quality and coverage o f the survey data, they do suggest the key impediments to business development and growth in BH. Analyzing these factors in detail is the subject o f the next chapter. 67 68 4. REMOVINGIMPEDIMENTSTO ENTERPRISEDEVELOPMENT A. INTRODUCTION 4.1 The previous chapter suggests that slow progress in market-based institutional reforms underliesthe poor investmentclimate and weak corporate performancein Bosnia and Herzegovina (BH). The country's weak institutional framework deters local and foreign investors, and consequently diminishes the opportunities for enterprise development and economic growth. Entrepreneurs are reluctant to open or expand businesses where there i s policy uncertainty, room for discretionary behavior and lack o f transparency. Corruption and transactional difficulties inday-to-day business operations are also important deterrents. 4.2 The persistence o f soft budget constraints among privatized and state owned firms (as reflected inweak financial discipline, the accumulation o f arrears, and the continued provision o f subsidies) reflects the poor corporate govemance structures that have emerged from the privatization program as well as continued dispersed ownership. The lack o f a credible threat to exit also keeps loss-making firms on the market. As a result, many loss-making firms languish at low levels o f activity and tie up assets that could otherwise be put to productive use while new private initiative strugglesto emerge. 4.3 Building on the preceding analysis, this chapter investigates in detail the most critical institutional constraints and incentives to enterprise restructuring and development,based in part on the findings of recent region-wide analysis of the business environmentin eight SouthEasternEuropean(SEE-8) economies.46The chapter is organized as follows. Section B examines the overall business environment inBH, drawing both on EBRD transition indicators and recent survey results from the BEEPS, and key barriers to new business creation and growth. The following two sections show how weak corporate govemance compounds the impact o f difficult external circumstances (Section C), and, how this, in conjunction with collusion among managers, suppliers, workers, and governments, allows privatized and state-owned firms to operate under soft budget constraints and with limited profit motivation (Section D). Section E shows that the hardening o f financial conditions inthe last few years was not until very recently accompanied by a parallel strengthening o f exit mechanisms. Section F concludes the chapter by focusing on areas where the government can introduce reforms to facilitate enterprise development. 46The eight SEE economies are Albania, Bosnia and Herzegovina, Bulgaria, Croatia, FYR Macedonia, Moldova, Romania and Serbia and Montenegro. B.THEOPERATINGENVIRONMENT Aggregate Assessment of the Business Environment in BH 4.4 Initial reforms were launched successfully in BH shortly after the end of the war. As shown in Table 4.1, the indexes for BH indicate major progress in "first-generation" reforms since 1996, most notably the near completion o f price, foreign exchange, and trade liberalization and o f small scale privatization. Table 4:1Bosniaand Herzegovina:ReformProgressIndicators, 1996.2000 and 2004 Bosniaand Herzegovina 1996 2000 2004 Initial reforms Price liberalization 2.0 3.0 4.0 Foreign exchange and trade liberalization 1.0 3.0 3.7 Small-scale privatization 2.0 2.3 3.0 Second-generation reforms Large-scale privatization 1.0 2.0 2.3 Enterprise reform 1.0 1.7 2.0 Competitionpolicy 1.0 1.0 1.0 Infrastructurereform 1.3 2.0 2.3 Banking sector reform 1.0 2.3 2.7 Non-banking financial institutions 1.0 1.0 1.7 Legal extensiveness (company law)a na 3.0 na Legal effectiveness (company law)a na 1.0 na Source; EBRDTR 2004, p.108 Note: Legal extensiveness measures the impact o f the law o n commercial transactions. Legal effectiveness o f legal reformmeasures the extent to which commercial rules are clear, accessible, and adequately implemented, both administratively andjudicially. Legal effectiveness and legal extensiveness indexes are not reported since 2001. 4.5 However, progress in large-scale privatization and all "second-generation" institutional reforms has been substantially weaker. International experience as well as empirical research on the trade and integration of transition economies demonstrates that price and domestic market liberalization alone are not sufficient to improve the economic performance and international integration o f these countries (EBRD, 2003, p. 77). 4.6 Reflecting this, the business environment in BH is still perceived as investor- unfriendly. To complement the EBRD's economy-wide assessment o f the overall business environment, it is useful to look at the results o f perception-driven surveys o f businesses in Eastern Europe, in which BH firms took part.47 Figure 4.1 presents a snapshot comparison 47Here we use the EBRD-World Bank Business Environment and Enterprise Performance Survey (BEEPS 1999 and 2002), to illustrate the point. Despite some caveats in comparing the 1999 and 2002 samples of surveyed firms discussed inAnnex 1 o f Chapter 3, the survey results from the BEEPS1 (for 1999) and BEEPS2 (for 2002) allow us 70 between the perception of enterprises regarding the investment climate inBHin 1999 and 2002. The broader set o f dimensions - financing, infrastructure, tax, regulations, corruption, rule of law, and the judiciary - indicate a notable improvement in each o f the aspects, especially in infrastructure. Overall, the business climate as perceived by enterprises operating in the country seems to have improved since 1999. However, institutional development and implementation capacity are still weak. Surveyed enterprises consider obstacles related to finance, perceived widespread corruption, and taxes as more restrictive o f development and day-to-day operations of the enterprise sector thanother factors. Box 4:l Data Sources T o analyze the constraints to investment and private sector development inBosnia and Herzegovina, we rely on the following sources o f information: (i) statisticsatthecountryandEntitylevel. Official (ii) fiomoriginalbusinesscasestudiesoffirmsoperatinginfive sectors(software development, retail Data trade, construction, textile, and metal working) o f the BHeconomy inthe fall o f 2002 and 2003. (iii) assessmentsbyanddatafrominternationalinstitutions, associationsandthink-thanksworkingon Recent these issues in BH, including the World Bank, EBRD, and the Heritage Foundation, data from the Doing Business project, and data from the Lex-Mundmarvard survey o f local lawyers. (iv) Data from several enterprise level surveys including: - The two rounds o f the EBRD-World Bank BEEPS, conducted in 1999 and 2002 (see detailed survey description inAnnex 1o f chapter 3), --The World Bank Early Warning System survey, Survey o f 1,000 enterprises in each Entity containing financial information for the surveyed firms (see detailed survey description inAnnex 2 o f Chapter 3), - FIAS Administrative and Regulatory Costs Survey o f 2002. - LSMS o f 2001 and 2002. Barriersto New BusinessEntry andOperation 4.7 How do firms perceivethe importanceof variousobstacles to businessdevelopment? The BEEPS2 asked surveyed firms to rate the extent to which a specific factor, such as macroeconomic instability, business regulation, taxes, corruption, infrastructure and others affect their business operations and prospects for firm development. Several key observations that emerge from the data are presentedinFigure4:2. 4.8 First, over 60 percent o f the 2002 BEEPS surveyed businesses in BH see economic policy uncertaintyand corruption as the most burdensomeconstraintsto developmentand operations.Unlike many countries ineastem Europe, perceptions of corruption as an obstacle to business have not declined since 1999 in BH. A recent World Bank study (Gray, Hellman and Ryterman, 2004) finds that administrative corruption is perceived to have increased in the country since 1999 inboth frequency and the amount paid inbribes. Within the SEE region, BH is among the countries with the highest levels o f "state capture", or the degree to which the sampled firms were able to effectively influence policy changes. A similar finding emerges from to pinpoint the major obstacles to enterprise development in BH, to evaluate the institutional progress since 1999, and to identify the current constraints indoing business inthe country. 71 the Administrative and Regulatory Costs Survey, conducted by FIAS in April 2002, in which unpredictability inthe regulatory regime and conu tion inthe public sector were cited as serious obstacles by almost halfo f all surveyed investors. 4: Figure4:l EnterprisePerceptionsofthe BusinessClimate,1999-2002 F i n a n c i n g E 3 B 1 H 9 9 B I H 0 2 I n f r a s t r u c t u r e R u l e o f L a w T a x C o r r u p t i o n R e g u l a t i o n s I I Source: BEEPS1&2 Note: The responses of the enterprises are averaged and are normalized on a scalefrom I to 4, with 4 representing the most severe obstacle. See Annex 1inchapter 3 for description o f BEEPS dataset. 4.9 Second, the cost of and access to finance, tax rates and administration, anti- competitivepractices, and the functioningof the legalsystem in protectingproperty rights and contracts place a significant burden on business development and investment. This group o f issues also presents a serious obstacle to investors in BH according to the FIAS Administrative and Regulatory Costs Survey (World Bank, 2003e). 4.10 Third, administrativeobstacles such as land-titling,access to land, labor regulations, permits, and licensing requirements are perceived by the surveyed firms in BEEPS2 as lesser obstacles to business development and operations in the country.49They are seen as a routine cost o f doing business rather than as pernicious barriers to entry and market competitiveness. However, this i s not to say that administrative bamers are insignificant for the development and growth o f businesses in the country (as per BEEPS2). Administrative bamers are ranked relatively lower when compared to the more hndamental institutional problems inthe country's business environment. 48 The Administrative and Regulatory Costs Survey represents the views o f 302 enterprises in both entities: 100 firms from RS and 202 firms from FBH.The sectoral distribution of the surveyed firms is as follows (percentages represent share o f firms in each Entity): Agriculture, hunting. forestry and fishing (RS-6.0 percent; FBH-4.5 percent) ;Manufacturing and mining (RS-30.0 percent ;FBH-28.2 percent) ; Construction (RS-7.0 percent ;FBH- 8.9 percent) Wholesale and retail trade (RS-33.0 percent 1; FBH-30.2 percent) ; Catering trade (RS-5.0 percent ; FBH-3.5 percent) ; Transport, storage and communication (RS-8.0 percent ; FBH-9.9 percent) ; Financial intermediation (RS-7.0 percent ;FBH-5.O percent) ;Other services (RS-4.0 percent ;FBH-9.9 percent). The sample was heavily skewed towards small firms, as some 75 percent and 79.2 percent o f the surveyed firms inR S and FBH, respectively, we firms with less than 50 workers. The survey covered areas such as registering with the authorities to begin operations, receiving permits and licenses, acquiring real estate, engaging in construction, equipment certification, on-site inspections, tax reporting and payments (as well as other issues relating to tax administration), impodexport procedures, appeals, and businesses' views o f government. See World Bank (2003e). 49 A similar result is found among the rest of the countries inthe SEE region. See Broadman et al. (2004). 72 Figure4:2 Barriersto the ODeration and Growthof Businesseshercentof firms) EcMMmic@icy uncertainty Cormpbon Cost of finanung Tax rates MacroeaxKxnicinstablity hbcompebbve practicesof other producers Tax administrabon OrganizedcnmelMafia FuncOoningof thejudiuary Accesstofinandng Street cnme/tkft/disorder Contradvlolationsbyw s t mandsuppliers Customs andtrade regulations Businesslicensingandpermits Laborregulations Slulls andeducatmof availablem-kers Transportatkon Electnaty kcessto land Teleaxnmunications TitleOT leasingof land ONoobsMe I M i m 0% 20% 40% 60% 80% 100% OModerate OMajcf Source: BEEPS2 Note: See Annex 1 inchapter 3 for description o f BEEPS dataset. 4.11 In reality,administrativebarriersstill presenta challengefor businessesin BH. For example, starting a commercial limited liability company in BH is still expensive, prolonged, and cumbersome.Data from the World Bank's Doing Business indicate that it takes 74 business days, the equivalent o f US$684, and 12 separate but independent procedures to legally start a business inBH (Table 4.2). Incontrast, it takes 2 days, 2 procedures, and less than 1percent o f annual income per capita to open a private company inCanada and New Zealand." BH fares worse than the SEE regional average as well as the more advanced central European countries on all o f the four indicators, and especially on the minimum capital required to start a new business. Entrepreneurs in BH must deposit at least 46 percent o f GNI per capita in a bank account to obtain a business registration number, compared with the average of 82.1 percent o f GNI for the central European countries. Additional difficulties arise from the multiplication o f registration requirements and regulations across constituent entities and across thefour levels o f government inBH(state, Entity, canton, and municipal). 4.12 Overall, the evidence suggests that the main barriers to de novo business creation are policy obstacles and institutionally rooted obstacles. The lack o f a business-friendly institutional environment deters investors, and also restrains firm performance and economic growth. 50World Bank (2004a), p. 18 73 Cumbersome administrative procedures and complex regulations also affect the start-up and the operation o f enterprises. Table 4:2 Openinga Business,2004 Openinga Business Bosniaand SouthEastEuropean CentralEuropean Herzegovina CountriesAverage a CountriesAveragea Number ofprocedures 12 11 9 Duration (days) 54 42 47 Cost (% of GNIper capita) 46.4 18.7 13.2 Minimum capital (% of GNI 339.4 99.2 82.1 per capita e Source: http:/rm.worldbank.org/DoingBusiness. Notes: 1. South East European countries comprises Albania, Bosnia and Herzegovina, Bulgaria, Croatia, FYR Macedonia, Moldova, Romania, Serbia and Montenegro. Central European countries comprise the transition economies o f the Czech Republic, Hungary, Latvia, Lithuania, Poland, the Slovak Republic, and Slovenia. 2. Procedures: any interaction o f the company founder with external parties (government agencies, lawyers, auditors, notaries, etc.). 3. Duration: recorded incalendar days. 4. Cost: costs associated with starting up a business as a percentage o f GNIper capita. 5. Minimum capital: the amount that the entrepreneur needs to deposit in a bank account in order to obtain a company registration number, as percentage o f GNIper capita. Security of Contracts 4.13 To make matters worse, regular commercial transactions are insecure and arduous. Difficulties in accessing liquidity and enforcing contracts limit the availability o f transaction instruments for day-to-day and long-term business operations. As a result, businesses transact in rudimentary ways by resorting to cash-on-delivery and especially to pre-payment deals. Over 80 percent o f sales by the surveyed firms in BH are on such basis. Over 62 percent o f the BEEPS surveyed firms in2002 mentioned the use o fpre-payment methods as a way ofhandling business transactions. On average, the surveyed firms required advance payments from clients for some 17 percent o f all sales. As shown in Table 4.3, prepaid sales are at higher levels for surveyed state-owned and de novo firms in BH than the SEE8 average. Conversely, firms do not commonly extend credit when transacting with their customers. Among the firms surveyed in 2002, the share o f sales sold on credit is only 12 percent, below the average of 21 percent for the SEE region. While the use of pre-payments reduces the likelihood o f payment disputes- especially for goods made to order, special customization, and sales o f goods to a few customers-it also limits the security of contract rights inbusiness transactions. 74 Table 4:3: Average Sales on Credit and through Advance - Payment by Ownership, 2002 Ownershiptype Sales sold on credit Pre-paid sales Bosniaand Herzegovina State-owned 21.1 19.7 Privatized 14.1 12.4 D e novo 10.0 17.6 SEE8 State-owned 16.2 16.2 Privatized 21.1 13.0 De novo 22.2 13.6 Source: BEEPS2. Note: See Annex 1inchapter 3 for description o f BEEPS dataset. 4.14 In the insecure contracting environment of the country, surveyed companies also use alternative means of payment, such as barter, bills o f exchange, and debt swaps. Some 27 percent o f the surveyed enterprises engage inbarter, although the level o f barter intotal sales is about 5 percent (Figure 4:3).51 Inaddition, case studies o f BH firms (described in Broadman et. al., 2004) reveal that businesses prefer barter transactions, as often this is the easiest way o f settling both purchases and sales in an environment o f insecure contract rights. All o f the case study firms use barter as a means o f settling business transactions. Some use barter rarely, while for other companies barter transactions comprise over 30 percent o f sales. The firms that engage extensively in barter transactions usually exchange their final products for raw materials, coal, and other inputs essential for their productiodservices. These firms tend to be o f a large size without differences o f ownership. Incontrast, small private firms tend to use barter less. 4.15 Debt swaps and offsets are popular among 42 percent o f surveyed firms in the country, and comprise an average o f 8 percent o f total sales (Figure 4:3). While in the advanced economies, debt swaps are considered an alternative instrument for settling transactions, the extensive use o f offsets or "multilateral compensations" among state firms and the government inBHhave ledto lesstransparency and soft budgets(Broadman et. al, 2004, p. 166). 4.16 An important factor in the general insecurity in contractual activity is the unreliability of the legal system to enforce contractual compliance and resolve business disputes. As Table 4:4 shows, the overall efficiency o f the judiciary inthe country i s very poor. For example, two-thirds o f the surveyed firms in 2002 indicate that the legal system is slow in resolving business disputes through the courts. In addition, almost one-half o f the interviewed '`Withthe improvement of the business environment in BH, one would expect to see a decline inthe use o f barter, bills o f exchanges, and debt swaps among the firms inthe country. The data inBEEPS1 and BEEPS2 reveal such a trend: in 1999 some 58 percent o f surveyed firms used barter, bills o f exchange, or debt swaps to settle their transactions, while in2002 the share o f firms fell to 47 percent. The trend is inline with the SEE regional decline in the use o f barter transactions, although BH is still among the countries with the highest share o f firms resorting to these measures, and i s surpassed only by Croatia (where 67 percent o f surveyed firms used such instruments for settling transactions in2002). 75 firms state concerns about enforcing decisions effectively and honestly, and over 41 percent "seldom or never" see fairness in legal judgments. Furthermore, a considerable 39 percent o f surveyed firms indicate that courts are expensive. Figure4:3 Barter, Billsof Exchange,and Debt Swaps Barter HBills ofexchange 0 Barter debt swaps or offsets Bills of exchange 1 I debt swaps or offsets 1 42 n I 8 7.9 6 ` 4.8 20 10 0 average percent of sales percent of firms Source: BEEPS2 Source: BEEPS2 Note: See Annex 1 inChapter 3 for description of BEEPS dataset. Table 4:4: Effectivenessof the LegalSystem by Attribute, percent of firms in total Fair Honest Quick Affordable Enforceable Never or seldom 41.72 46.58 66.45 38.97 44.6 Usually or always 13.91 12.33 6.46 11.03 16.21 Source: BEEPS2. Note: Intermediate responses are not shown, andhence, percentages do not sum up to 100. See Annex 3.1 inchapter3 for description of BEEPS dataset. 4.17 Courts in BH are expensive and slow, and discourage the small entrepreneur from effectively using them. The Lex-MundVHarvard survey asked local lawyers to assess how a creditor would go about recovering a payment from a client who is in violation o f a sales ~ontract.'~ InBH, it would take 630 days from the day of filing the claim incourt to the final enforcement o f the court's decision and would require a total o f 31 procedural steps, which both the litigants and the judge need to undertake in the process of resolving the business dispute. Court procedures (attorney fees and court fees) would amount to over 40 percent o f the claim value - a much higher level than in other SEE countries such as Bulgaria and Croatia, where the total costs come to about 13 percent o f the claim value. In addition, the average time for an administrative appeal i s between 60 and 270 days, with a tendency toward a shorter time period inthe FBH and longer time period inthe RS (World Bank, 2002~).Data from the BEEPS2 for surveyed firms inBH show that large firms tend to use courts more than small (often new ones) do, a tendency evident also on the regional level, although to a somewhat lesser degree (Figure 4:4). 52 The payment is in the amount of 50 percent of the gross national income per capita in each o f the surveyed countries. See rm.worldbank.org/DoingBusiness database for a detailed description of the methodology and access to data. 76 Figure4:4 ShareofFirmswith NoPaymentDisputeResolvedin Court,by Size loo 1 Small Medium Lame' .E 60 :E 40 8 20 0 Bosniaand Herzegovina SEE8 Source: BEEPS2 Note: See Annex 1 inChapter 3 for description o f BEEPS dataset. 4.18 The weak security of contract rights and the lack of trust in business transactions, combined with perceptions that the 1egaYinstitutionalenvironment cannot uphold and effectively enforce property rights and contracts, are common signs of a poor investment climate. Heritage Foundation (2003) gave BH the lowest rating on the protection o f property rights index in the SEE region. This means that the law does not protect private property effectively, and that the legal system cannot effectively enforce contracts. The poor quality o f the market institutions that support firms' day-to-day transactions alienates local and foreign investments and makes the prospect o f doing business in or with such a country unattractive to investors. Burdenand Complexityof Taxation 4.19 Firms in BH have to cope with heavy taxes and a complicated tax system. Direct taxation is high (Table 45). Until recently, moreover, firms have also had to contend with different indirect tax regimes inthe two entities. The recent establishment o f a single, state-level Indirect Tax Authority (ITA) that will be responsible for the collection and administration o f all customs and indirect taxes inBH, as well as for the VAT after its introduction in2006, is a major institutional breakthrough. The introduction o f VAT will also automatically lead to full harmonization o f indirect taxation in policy and practice, an essential element to further establishing the single economic space inBH. Access to Finance 4.20 Access to foreign capital emerges as one of the main perceived constraints to enterprise development in BH (Figure 4:l). Indeed, the primary source o f financing new investmentsis the firms' internal sources. As indicatedinTable 45, retained earnings are usedto finance new investments in 55 percent o f the surveyed firms inBEEPS2, and on average finance some 82 percent o f new investments. The surveyed state-owned firms use retained earnings as a source for an average o f 96 percent o f their new investments. 4.21 Surveyed firms have little recourse to investment credits from domestic or foreign banks. Informal sources o f credit, such as loans from family or friends, moneylenders and others, were a major source o f financing for new investments for some 15 percent o f the firms 77 1999, and financed over half o f the investment needs o f these firms in2002. Trade credit i s used to a very limited extent to finance new investment. A negligible 0.55 percent o f the interviewed firms use trade credit from customers, and only a slightly higher 5 percent resort to credit from suppliers for investmentpurposes; such firms use this source o f financing for about 48 percent of their new investments. 4.22 The sources for financing working capital have a similar structure to those for financing long-term investments. As presented inTable 4.6, short-term capital is obtained most frequently from internal funds (retained earnings). Some 77 percent o f the surveyed firms indicate that they use retained earnings for operational needs. At the same time, every fourth firm obtains capital from informal sources, including family and friends. Equity appears to be another major source o f short-term financing, although this i s an instrument for a very limited share o f the surveyed publicly-traded firms. Trade credit from suppliers is often used for the financing o f short-term capital needs, as indicated above. By contrast, credit from customers i s used in less than 2 percent o f the firms, but its average share inthese firms i s comparatively large. 4.23 The predominantly short-term structure o f deposits in the commercial banks largely explains the low share o f firms using conventional sources o f credit. Long-term lending options are limited on the local market and real interest rates are much higher than inother economies o f the region especially compared with world rates. Although lending rates have come down inthe last few years to 11 percent in2003 (from over 30 percent in2000), the interest rate spreads (the difference between deposit and lending rates) remain large (Table 4:7). However, the decline in lending rates in the past three years is partly due to some improvement in the quality of borrowers, not to a better business environment inthe country.5354 4.24 After privatization, commercial banks have become quite conservative and reluctant to engage with risky clients, despite the banks' improved liquiditypositions. Loans to private businesses comprise only 43 percent and 46 percent o f total loans inthe FBHand the RS, respectively, while credit to SOEs in both Entities notably declined. Financial intermediation remains very low (Table 4:7) with domestic credit to the private sector in 2002 at only 14.6 percent o f GDP. 53 Moreover, commercial banks still charge high commissions to hedge against risks - a sign o f lack of competition among the banks, which has muchroom for improvement (European Commission, 2004). j4Up-front fees for business loans are around 1percent o f the loan. BEEPS-surveyed firms in 2002 also indicate that short-term bank credits are expensive. Enterprise-level data for 2002 show that the average cost o f such loans i s 11.64percent with an average approximate value of the collateral requirement o f 81percent o f the loan value, and a median duration o f 12 months. 78 vi 3 3 m: ~~ 4- ......................... ..L ............ w ........................................................................................................................................................................................................................................... ..... "r " c d t . w a m . - o s - - 0 0 0 2 -m! m ? ec?m 0 m a cd" E od 0 00 Table 4:6 Sources ofFinancing,2002 New investment Working capital Frequencyof Average share of firms using Average share of Frequencyof new investment each source of working capital firms usingeach Sources financedfrom capital, financed from source of each source of percentof total each source of capital, percent capital, percent firms capital, percent of total firms Internal funds 54.4 80.8 76.9 New equity 55.0 1.1 51.7 1.6 Local private banks 39.1 12.6 32.7 17.6 Local state banks 56.4 7.7 38.9 13.7 Foreign banks 70.0 3.3 37.9 7.1 Family/friends 46.8 13.7 30.6 23.1 Other informal sources 23.3 1.65 15.6 4.4 Trade credits from 48.3 4.95 35.5 15.4 suppliers Trade credits from 30.0 0.55 31.7 1.6 customers Credit cards 10.0 0.55 12.5 2.2 Leasing 46.7 1.65 33.3 1.6 Government 18.3 1.65 30.0 2.2 Source: BEEPS2. Note: See Annex 1inChapter 3 for description of BEEPS dataset. Table 4:7: FinancialInstitutions Financialinstitutions 1999 2000 2001 2002 2003 Asset share o f state-owned banks (in%) 75.9 55.4 17.3 6.3 5.2 Non-performing loans (in% o ftotal loans) 58.7 15.8 20.7 11.5 8.3 Domestic credit to private sector (inper cent o f 14.6 GDP) 8.9 7.4 9.5 11.9 EBRD index of banking sector reform 2.3 2.3 2.3 2.3 2.3 EBRD index of reform of non-banking financial 1.7 institutions 1.0 1.0 1.0 1.7 Deposit rates (inpercent) na 14.67 6.07*/ 4.53 4.00 Lending rates (inpercent) na 30.50 16.60*/ 12.69 10.87 Source; EBRD Transition Report 2004, p. 126 for all indicators but deposit and lending rates. The latter comes from IMF, International Financial Statistics and the Central Bank of Bosnia and Herzegovina. Notes: */break inseries. 4.25 There are several key reasons for this. First, banksprefer to engage with households rather than firms because of difficultieswith the enforcement of loan contracts in cases of default. Lending to households is thus prominent (48 percent of total loans in the FBH and 41 percent inthe RS). Anecdotal evidence suggests that a good portion of the households' credits i s 8 1 directed to the informal sector, as partly reflected in the BEEPS 2 data (Table 4:6). A better understanding o fhouseholdlendingi s neededto untangle the direction and use o f the funds. 4.26 Second, the collectiontime of non-performingloansin BH is long(taking over several years) - a fact partly explained by the weak bankruptcy institutions and the poor enforceability o f contracts described in the previous section. In addition, the lack o f a central register o f delinquentborrowers ineither Entitymakes banks cautious inlendingto enterprises. 4.27 Third, commercialbanksare unwillingto lendto firms with large overdue liabilities. At the same time, firms find it easier to accumulate arrears as an alternative to conventional sources o f finance, instead o f complying with commercial banks' requirements for the creditworthiness o f enterprise borrowers. 4.28 Small and medium-size firms are particularly affected by the limitations in accessing affordable financing. In BH, a considerably larger share o f surveyed small and medium-size firms indicate that it i s very difficult or impossible to receive short-term financing in comparison with the average for the surveyed large firms (Figure 45). Banks impose additional requirements on SMEs for extending a loan. For example, some banks require at least six months of prior successful operation, two legal Entity guarantors, and a mortgage over property valued at 150 percent o f the loan amount for loans over KM 20,000 (World Bank, 2004b). Figure 4 5 Access to Finance by Size and Ownership, 2002 t 30% 5 25% 3. -d 00 20% w 2 c %l!i0/o a - -55 ~ 1 0 % $ 3 . 4. 5% 6 0% > Source :BEEPS2 Source: BEEPS2. Note: See Annex 1inChapter 3 for description o f BEEPS dataset. 4.29 This is however only halfthe story. With the emergenceof a healthyfinancialsector, the comPosition of commercial banks' portfolios has begun to slowly shift to support growth.5 Figure4:6 shows an expansion o f credit to the private sector, especially in the FBH. As a sign o f the changing times, commercial banks have become more reluctant to lend to heavily indebtedand illiquid state-owned firms. This i s not only a first step toward the enforcing o f financial discipline and the hardening o f enterprises' budget constraints, but is also an 55 Data on the credit portfolio composition from World Bank (2004b). 82 effective mechanism for shifting financial resources from unproductive uses to profitable ventures. Figure 4:6 Structure of Commercial Banks' LoanPortfolios 100% FBiH RS 75% 50% 25% 0% Dec -00 Dec-03 Dec -00 Dec-03 Gov and Public Enterprises Primte enterprises Banks and Fin Institutions Citizens NGOs and other ~~ ~ Source: World Bank, 2004 (forthcoming), "Bosnia and Herzegovina Banking Sector Review," Finance and Private Sector Development Department, ECA. Integration 4.30 One way for firms to survive in this difficult market environment is to integrate vertically or horizontally and thereby internalize product markets transactions as well capital markets. While the degree o f horizontal integration (as measures by the combined market share o f firms supplying more than 5 percent o f their markets) i s comparable to the one observed in other SEE countries, vertical integration is much more prevalent (Table 4.9). As the BEEPS data reveal, on average, over 28 percent o f the surveyed firms' sales transactions are conducted internally or are made to related parties. This is the highest degree o f downstream integration among the countries inthe SEE region, where the regional average is over two-and-a- halftimes lower than that inBH (11percent o f sales).56Table 4.10 presents the degree o f vertical integration among surveyed firms inBHby sector and ownership. c.THEROLEOF CORPORATEGOVERNANCE Ownership Structures 4.3 1 As corporate performance attests, ownership change has not resulted in effective enterprise restructuring in BH. Most privatized firms are not profit-motivated because o f the way the privatization process was carried out (see below). Empirical research on other transition economies has demonstrated that not only i s privatization strongly correlated with efiterprise restructuring, but the method o f privatization i s related to the benefits derived from corporate restructuring. For example, Djankov and Murre11(2002) find that: (i) average, privatization to on outsiders i s associated with 50 percent more restructuring than privatization to insiders; (ii) the effect of privatization through sales to foreigners is, on average, 10 times as strong as sales to 56See Broadman et al. (2004), p. 116-117. 83 diffuse owners; (iii) all other types of ownership, state ownership i s the least conducive to among restructuring. Table 4:8 Market Share by Sector, 2002 (percent) Sector Bosnia and SEESb Herzegovina Mining 30.00 31.23 Construction 13.40 23.41 Manufacturing 33.33 32.11 Transportation, storage, and 25-00 41.89 communications Trade 23.64 23.27 Business services 20.00 33.11 Hotels andrestaurants 19.00 19.86 Other 36.17 39.97 Note: Covers only surveyedfirms indicating market shares greater than 5 percent. a Emptycell indicates no firms surveyed inthat sector. SEE8 comprises ofAlbania, Bosnia and Herzegovina, Bulgaria, Croatia, FYRMacedonia, Moldova, Romania, Serbia and Montenegro. See Annex 1inchapter 3 for description o f BEEPS dataset. 4.32 Unfortunately,privatizationin BH has relied primarily on some of the less effective methods. Mass privatization techniques encompassed the largest number o f privatizations in both entities (68 percent o f all privatized firms in the FBH and 45 percent in the RS). Privatization in BH has given a prominent role to three groups of players: (i)Privatization Investment Funds(PIFs); (ii) insiders (i.e., managers and workers); and (iii) state. the Table 4:9 Downstream Integration by Sector, Ownership Type, and Size, 2002 (average percentageof an enterprise's sales madeto parentfirm or affiliatedsubsidiary) Average Average Average Sector share Ownership share Size share (percent) (percent) (percent) Mining 3.3 State 29.1 Large 25.0 owned Construction 15.8 Privatized 18.8 Medium 32.9 Manufacturing 30.1 De novo 31.5 Small 27.4 Transportation, 29.0 storage, and communications Trade 30.7 Business services 35.7 Hotels andrestaurants 20.0 Other 28.8 Source: BEEPS2. Note: See Annex 1inchapter 3 for description o f BEEPS dataset. 4.33 Furthermore, the design of the mass privatization programs has favored domestic voucher-holdersover foreigners.Foreign investors, as a result, were limited to acquiring only 30 percent o f the shares with cash in contrast to 55 percent o f the shares held for voucher holders. Most vouchers in the RS (two-thirds o f all vouchers) or o f certificates in the FBH were exchanged with shares inthe PIFs inboth Entities, which made the PIFs the largest shareholders 84 inmass privatized companies. Inmid-2004, there were 11PIFs inthe FBH and 13 in the RS. Obtaining shares ina large number o f companies was typical for the PIFs inboth Entities. Some PIFs obtained shares in over 200 enterprises in the RS, and the largest PIF inthe FBHobtained shares in 111companies. 4.34 Despitethe widespreadpresenceof PIFsin the ownershipstructureofthe privatized firms, PIFs never command a concentrated or controlling state. The FBH law limits the stake a single PIF may hold in a company to 30 percent, restricting the PIFs ability to act as an effective owner. Inaddition, the incentive structure that would enable PIFs to effectively manage the privatized firms intheir portfolios i s missing. As a result, PIFs have de facto become passive portfolio management companies. 4.35 Another factor in the passivity of the PIFs is the structure of their compensation. PIFs receive a 1 percent management fee in the FBH and a 2 percent fee in the RS. The management fee i s calculated as a percent o f the net asset value o f the PIFs' portfolio companies. The original value o f assets, however, was based on book value without a proper mechanism for adjusting the valuation to its market level, which greatly inflates the numbers.57This discourages PIFs from taking any action to recognizethe real value o ftheir holdings, which would negatively affect their management fees. The incentives to sell shares in a transparent way or to engender restructuring efforts in troubled companies are missing. Thus, the institutional environment preventsPIFs from playing a meaningful managerial role inthe companies intheir portfolios. 4.36 Insiders-managers andworkers-are the second prominent group of shareholders in the privatizedfirms. Most o f the employees o f the companies offered for mass privatization used their vouchers to acquire shares in"their" firms. These purchases o f shares were in addition to shares they previously obtained under the 1988 Enterprise Law (the so-called Markovic Law), which allowed employees to obtain shares o f the enterprises in which they worked. The proportion o f insider ownership varies, although Entity-based statistics reveal that individuals (a group believed to comprise almost exclusively insiders) have shares in close to 50 percent o f the mass-privatized firms in the FBH. In most o f the mass-privatized firms, management has remained unchanged, which limits the prospects for enterprise restructuring, partly because o f the lack o f incentives to release redundant workers. 4.37 At the same time, the state has maintaineda considerableshare of ownership in the mass-privatizedfirms, an average o f about 15 percent inthe FBH and 30 percent inthe RS. In addition, 10 percent o f the shares are allocated to the state-controlled Pension and Disability Fund and another 5 percent to the Restitution Fundinthe RS. The state has become the largest ownership group in BH, retaining the right to appoint the majority o f the members o f the board o f directors o f the newlyprivatized firms. 4.38 Apart from the mass-privatizedfirms, governmentshave retained a large portfolio of wholly state-owned firms. These include: (i)utility and infrastructure companies; (ii) 79 strategic companies (some utilities, large companies o f national importance and others): and (iii)numberofenterprisesofallsizesthatwereeithernotincludedinthemass-privatization a j7At present, the value of assets is calculated based on guidelines from the Securities Commission, which are still linkedto the book but not the market value of the assets. 85 program or were included but the sale was unsuccessful. Currently, the government i s not planning ownership divestiture o fthe utilities and infrastructure companies, although introducing competition and private sector participation in these would be key to their restructuring and profitability. By contrast, rapid privatization o f the strategic enterprises has been on the government's agenda, but progress has been mixed and slow. By mi-2004, only 22 strategic companies were privatized. Part o f the problem is institutional, as the privatization agencies have insufficient power to fully administer and finalize the sales, and the process is associated with cumbersome procedures. There are twelve privatization agencies inBH(one for each o f the two Entities and one for each o f the FBH's ten cantons), with often overlapping powers. Firms with operations in more than one canton or with strategic significance can fall within the purview o f several cantonal agencies and the federal agency. Unsuccessful attempts at privatization, meanwhile, are made inthe thirdgroup where the state is still the dominant owner. 4.39 Indeed, the governments are facing major performance problems across various firms in which they have retainedmajor ownership and controlrights.For example, among a sample o f 39 strategic enterprises in the FBH, some 75 percent need major financial and/or industrial restructuring efforts to revive their operations. Among the 39 strategic firms, only one- quarter are financially viable and suitable for speedy privatization. These firms are primarily in the telecommunications, public utilities, metal processing and tools sectors. Firms inthe leather and footwear sectors as well as in wood processing also have good prospects for early privatization should the financial restructuring o f these sectors materialize. In addition, major restructuring efforts or the exit o f non-viable strategic enterprises are needed, especially in the agricultural and agro-processing sectors and also among firms in the steel, casting, and tools sectors, intourism, and inthe oil/petrol sectors. 4.40 A similar assessment o fthe viability o f 51strategic enterprises was conducted in2003 for the RS. Two-thirds o f the enterprises are inneed o f major restructuring, and among them, over 27 percent o f the enterprises were deemed non-viable and thus are probable cases for bankruptcy or liquidation. Among the non-viable companies inthe RS are enterprises inthe agricultural and consumer product sectors, and inthe textile, mining, oil, and others sectors (Box 4:2). 4.41 In sum, the reliance on insiders and the state in the mass privatization has left fragmented ownership structures, poor incentives for restructuringand development and weak corporate governance structures. This has limited inflows of capital and know-how into the newly privatizedfirms. The privatization process failed to bring fresh capital through new owners (outsiders) and to create the incentives for effective corporate governance in the mass-privatized firms. Overall, enterprise restructuring has yet to change the way firms are runin the country. Weak performance, as noted above, is linked to slow privatization and the establishment o f an inappropriateincentive structure inthe newly privatized firms. 86 IBox 4:2 Obstacles to Privatization of Strategic Firms I A strategic enterprise inthe textile industry, which employs over 500 workers, has been struggling for the last few years to find an investor to buy its assets and restructure its operations. Less than 5 percent o f the company's share capital is privately owned. Founded over 25 years ago, the company used to be the major employer intown and was active o n both the domestic and foreign market inthe pre-war years. Inthe post-war years, the firmhas operated at 15 percent o f its capacity, and sold approximately 80 percent o f its production to the government. According to the manager o f the interviewed firm, the company i s losing ground in this segment o f "guaranteed" contracts due to competition from private firms, which are more prone to bribe officials to buy their products. The interviews with the management o f this strategic enterprise suggest that the obstacles to selling the state assets inthe company are both structural and institutional, and can be outlined as follows (i) investors are not strategic interested in purchasing just about 65 percent o f the share capital at a minimumpre-set price, when the state and insiders retain the rest; (ii) the poor state o f financial accounts and liabilities o f the firm alienates outside investors; (iii) obsoleteconditionofequipment wouldrequiremajorinvestmentsinfixedassetsandtechnology, which the would come in addition to the price paid by the external investor without major contributions from the rest of the shareholders. At the same time, the minority shareholders will retain significant control over the management o f the company. 1Source: Broadman et al., (2004). OtherAspects of CorporateGovernance 4.42 Other aspects o f corporate governance do not help. First, the governance structure of joint-stock companies in BH is particularly cumbersome, as it involves three tiers of managing boards with blurry responsibilities and definition o f purpose. Inthe FBH, ajoint-stock company has a management board, a supervisory board, and an auditing board, while in the RS the three-tier structure comprises a managing board, a board of executive directors, and a board o f supervision. There is no clear separation o f functions among the boards, which makes their roles o f monitoring and oversight o f the corporations difficult to fulfill and the election o f members sitting on the different boards i s not well regulated. Inaddition, inthe RS, for example, the functions o f the shareholders' assembly are not clearly defined by law, which affects the boards' election process. Moreover, the process for appointing o f members to the management boards is often politically motivated. Box 4:3 illustrates the problems with examples from case studies. 4.43 Similar governance weaknesses also directly affect the PIFs, which fall under the regulation for joint-stock companies. The existing corporate governance mechanisms are ineffective and ill-applicable to the PIFs, some o f which have thousands o f small (individual) shareholders. Instead o f protecting unsophisticatedsmall shareholders, the rules practically allow the management o f the PIFs to exercise full control o f the funds. An annual meeting or a shareholders' assembly i s impossible to organize in companies with thousands of dispersed shareholders and a legal requirement for at least 30 percent presence. For example, at a recent annual shareholders' meeting o f a PIF with 200 000 shareholders, none came to the meeting. At the same time, it is expensive and difficult for small shareholders wishing to divest from PIFs to sell their shares. This prevents investors from accumulating concentrated ownership and control positions inthe PIFs and from providing strong governance. 87 Box 4:3 PoliticalAppointments Two state-owned companies interviewed inthe the RS, operating in two different industries, described the way members are appointed to their management bodies. The Management Board consists o f three external and two internal members. The external members are political appointments by the government o f the RS. Inone o f the interviewed f m s the three external appointees were not involved in the government, while in the other enterprise the three outside members were politically active. The other two members o f the Management Board inbothenterprises were employees o fthe companies. The Management Board selects and appoints the Director o f the company, although not without approval from the government. The Director could be any one o f their choosing regardless o f technical expertise or management skills. The Director's appointment is open-ended. In contrast to the Management Board, all three members o f the Supervisory Board are government appointees. The political parties inthe Entity suggest nominees, which the government appoints. Inthe interviewed company, all three members o f the Supervisory Board were politically active at the time o f the interview. I Source: Broadman et al., (2004). ' 4.44 Second, minority shareholders' rights inbothentities are unprotected,and are often counterintuitive and inconsistentwith contemporary standards of corporate governance. Minority shareholders' rights provide an important role for these shareholders on issues such as voting rules, board elections, disclosure o f information, and other matters. But in both Entities increasinga firm's share capital does not require a qualified majority. Yet, a qualified majority i s needed for corporate decisions with respect to ownership asset divestiture. Furthermore, minority shareholders are not protected with respect to stock repurchase rights, are not provided with derivative suit rights or given cumulative voting rights.58 4.45 Third, disclosure and transparency in the enterprise sector are particularly poor. This i s one o f the key factors in the failure o f dispersed privatization to establish effective corporate governance structure^.^^ Without the proper institutions to provide adequate and transparent information on the viability o f firms, investors and shareholders cannot assess the prospects for business development o f any company, cannot make informed decisions, and cannot hold managers responsible for their actions. Adoption o f Intemational Financial Reporting Standards (IFRS) for public interest entities6' and statutory audit requirements consistent with the acquis communautaire6' would be a first step toward increasing financial transparency inthe country. 58 Derivative suit rights are actions brought by a shareholder on behalf o f a corporation to protect the company's legal rights. 59 Details are available in the assessment of the accounting and auditing environment under the World Bank Report on the Observance o f Standards and Codes (ROSC). The ROSC report includes policy recommendations to address the issues discussed inthis section. 60 Public interest entities are enterprises in which the general public has an interest by virtue o f the nature o f their business, their size, their number o f employees, or their range o f stakeholders. Examples of public interest entities include banks, insurance companies, investment funds, pension funds, listed companies, and large enterprises. This definition i s consistent with the concept discussed and developed during the preparation of the Recommendations on Quality Assurance and Auditor Independence inthe EU Committee o n Auditing. The proposal for a new EighthEU Company Law Directive includes hrther guidance on this matter. 61 The acquis communautaire comprises the relevant Directives and Regulation, including the Fourth, Seventh, and Eighth European Union Company Law Directive but also the "soft acquis" of the European Commission 88 4.46 Disclosure o f financial information should also extend to wider range o f economically important firms such as the state-owned companies. Currently, the identity o f the large shareholders, and the ownership and remuneration o f management personnel are not disclosed. Case studies presented in Broadman, et al., (2004) reveal that at present firms do not publish their financial statements, in spite o f external pressure (legal requirements). This causes general mistrust among the shareholders and the market in general. The mere adoption o f laws and regulations will not be sufficient if the firms do not recognize the benefits associated with improving financial transparency and disclosure. D.SOFT BUDGETCONSTRAINTS 4.47 Weak corporategovernanceand collusionwith suppliers,workers and governments allow firms continue to operate under soft budget constraints and with limited profit motivation.This impedes enterprise restructuring. Most large firms and mass-privatized mid- size firms experience over-indebtedness to the tax authorities, the social funds, andtheir workers. The heavy burden o f arrears hampers a firm's ability to effectively use the institutional system to resolve such claims. On the other hand, arrears have been used as an easy means o f financing a firm's activities when access to affordable financing emerges as a major issue for all firms inBH (see above). Figure4:7 IncidenceofArrears andSubsidies (percentFirmsin Total), 1999 and2002 Utilities -2002 60, , c .c 1999 Source; BEEPS1 and BEEPS2 Note; See Annex 1inchapter 3 for description o f BEEPS dataset. recommendations on auditors' independence (May 16, 2002) and on quality assurance for the statutory audit (November 15, 2000); as well as the "looming acquis" flagged inthe Communication of the European Commission to the Council and the Parliament "Reinforcing the Statutory Audit inthe EU", dated May 21, 2003. 89 Arrears to Suppliers 4.48 Although there has been a decrease in the share of firms operating under "soft budgets," arrears to suppliersremainthe most widespread problemon the liability side of the firms. In2002, every fifth surveyed firm in BH reported payments overdue by more than 90 days to material input suppliers. The size o f suppliers' arrears among the BEEPS surveyed firms with overdue payments to suppliers was 9.5 percent o f total annual sales in 2002 (Table 4:lO). Case studies in Broadman et. al. (2004) suggest that difficulties in having customers pay on time and inreceiving goods and services from suppliers on time are among the most common impediments to business development. Table 4:lO Arrears and Subsidies, as a percentof Total Annual Sales, 2002 Type of Arrears Share inTotal Sales Suppliers arrears 9.5 Wage arrears 7.7 Tax arrears 7.8 Utilities arrears 6.8 Subsidies from the federal government 8.3 Subsidies from the regional, local, or other 9.7 government Source: BEEPS2. Note: See Annex 1inchapter 3 for description of BEEPS dataset. Arrears to the Government 4.49 Tax arrears are not insignificant.Among the BEEPS surveyed firms with overdue payments to the government o f 90 days or longer, the incidence o f tax arrears was close to 17 percent in 2002, or 7.8 percent o f total annual sales on average (Table 4:lO). A similar message comes from the review o f 1,000 largest firms in the FBH: as o f the end o f April 2004, there were 8,404 companies in tax and contributions arrears, amounting to KM511million. The reported overdue payments were concentrated in large SOEs and large mass-privatized companies that consider themselves immune from enforcement action. Inthe RS, the situation i s similar: large SOEs firms are the main holders or arrears. However, the majority of SMEs in Bosnia and Herzegovina do not long remain in tax arrears as enforcement i s much stricter for them. Arrears to Utilities 4.50 Although unpaid obligations to the utilities companies has also decreased sharply since 1999, over 15 of the BEEPS2 surveyedfirms reportedunpaidbillsfor over 90 days in 2002. Indeed, the incidence o f arrears to utilities in BH is the largest among the firms interviewed inthe SEE region (Broadman et. al, 2004). Arrears to utilities amount on average to 7 percent of the annual sales o f the firms with such overdue obligations (Table 4:lO). Over 35 percent o f the interviewed state-owned enterprises and over 25 percent o f the privatized enterprises with arrears to utilities do not pay utility providers on time, compared to some 6 90 percent o f de novo firms. These observations reveal the emergence o f a strong uneven playing field across different ownership categories. 4.5 1 Official data also confirm the extent o f the problem. For example, inthe power sector the cash collection rate i s 86 percent and the commercial losses (defined as non-billed consumption) are 25 percent (Broadman et. al, 2004). Accumulation o f arrears to the utility companies is also stimulated by inflexible pricing'policies, as Box 4:4 indicates. WageArrears 4.52 Arrears to employees are also high, reflecting the failure of enterprises to pursue effective labor restructuring. Labor rigidities materialize in wage arrears, a category that encompasses unpaid salaries and social contributions as well as severance payments to workers. The LSMS data presented in Figure 4:8 reveal that the problem o f delayed salary payments, especially in the medium and large enterprises, has been persistent in the FBH.62In fact, 26 percent and 28 percent of employees working in medium and large firms, respectively, reported such delayed payments. Moreover, delayed payments to workers in firms with over 200 employees increased between 2001 and 2002. Despite a slight improvement in payments in 2002, the situation inthe RS i s even worse. Some 46 percent and 45 percent o f workers in large and medium firms, respectively, reported delayed payments for over two months. Of these, 31 percent o f workers in firms with over 200 employees were not paid in four or more months. Smaller firms in both Entities indicate a significantly lower share o f delayed payments in comparisonwith firms with over 51 workers. 4.53 Several factors explain the reluctance of firms to dismiss excess workers. On the one hand, labor dismissal costs for the employer are perceived to be high, although the legal requirementsto clear wage and contribution arrears prior to the dismissal o f an employee pertain only to certain groups o f workers. On the other hand, the reluctance to shed excess labor might have cultural roots, as managers perceive themselves as being responsible for the social welfare o f their employees. There appears to be a tacit agreement between management and employees that the latter agree to long-term non-payment, provided they are kept formally employed and receive social benefits, while they work informally somewhere else. 4.54 As a result, the formal sector exhibits artificial over-employment. Firms interviewed through the BEEPS2 were asked to indicate the level o f optimal employment intheir firms (as a percent o f the existing workforce). assuming no restrictions existed. Not surprisingly, the interviewed medium and large firms indicated that they over-employ. On average, to reach the optimal level o f employment correspondent to production levels, they would need to cut at least 10 percent o f their workforce. In contrast, the interviewed small firms revealed a need for the expansion o fworkers by 12 percent on average over their exiting level o f employment. The data, as well as evidence from case studies, also show that firms in industry generally tend to over- employ, while firms in services tend to have a small shortage o f workers and slight willingness to hire. The case studies, however, signal that the size o f the problem might be even greater than i s shown by the enterprise-level survey (Box 4:5) 62 See the chapter on Labor for a detailed description o f the LSMS data. 91 Box 4:4 Inflexible Electricity Pricing Generates Arrears Enterprise level case studies in BH reveal that firms are unhappy with the differences in power pricing between residential and industrial users. Firms are required to pay a fixed monthly amount for a fixed amount o f electricity. No compensations are made for smaller than the fixed amounts used. For example, one o f the interviewed companies, working in the textile industry, indicated that the firm uses only one-third o f the fixed pre-paid amount o f electricity. Payments for utilities, most notably for electricity, comprise 50 percent of operational costs o f this fm. Electricity over this fixed amount i s charged at a much higher rate. Such an inflexible pricing arrangement hurts both the small and the big users o f electricity, and encourages the accumulation o f arrears and inefficiencies. Electricity is muchmore expensive on a unit basis for the small users, and due to the quick price acceleration large users find themselves with "unconceivable electricity bills," as one o f the interviewed fm described the problem. None o f the interviewed firms expressed satisfaction with the quality o f the power services in the country. They described frequent interruptions and voltage fluctuations as the main problems regarding the quality o f power supply. N o company receives compensation from the electricity supplier for losses due to blackouts. T o cope with the problem, several o f the interviewed firms have purchased their own back-up electricity generators, which are expensive but seen as worth the investment. The utilitycompanies treat unpaid liabilities fiom their enterprise customers unevenly. A private mid-size firm reported that power supply is stopped every time the company does not pay within 15 days. At the same time, a large state-owned firm operating in the metal-working industry shared during the interviews that in the winter o f 2001 the company was over 30 days late in paying its electricity bill. The manager wrote to the Electric Company indicating that if the power supply i s terminated, the large enterprise will shut down. Inthe same letter, the manager advised the utility supplier that before turning o f f the power to the firm, they should Icheck with the Prime Minister. Power was never cut off. Source: Broadman et al. (2004). Subsidies 4.55 Alongside the gradual hardening of payment terms since 1999, the share of firms receiving formal subsidies from the government increased among the surveyed firms since 1999. State firms are the main beneficiaries. According to BEEPS in 2002, they receive on average 10 times more direct subsidies as a share o f sales revenues than do privatized firms. Across sectors, the largest recipients o f subsidies are firms in services, in infrastructure sectors such as telecommunications and transport, and inmanufacturing. 4.56 Subsidies come from all levels of government.As Table 4:lO shows, subsidies fiom the state govemment amounted to 8.33 percent o f annual sales and subsidies from other levels o f government were equivalent to 9.7 percent. This indicates a willingness o f the govemment at all levels to tolerate and protect any loss-makers from market exit. Subsidies in any form (direct or indirect) perpetuate soft budgets and distort markets. Insum, the accumulation o f arrears and the reliance on subsidies exacerbate the restructuring problem faced by a majority o f firms. 92 Figure 4:8: Wage Arrears inthe FBHand the RSby Firm Size, 2001 and 2002 F B l H iI I I I I I I I I I Box 4 5 Dealingwith ExcessLabor Case studies reveal the severity o f the over-employment problemin BH. Two examples illustrate the magnitude o f the problem. A large state-owned company, operating in the steel industry, formally employs 420 workers at present. This i s the city's largest employer. The company was three times as large inthe pre-war period, but was forced to close parts o f its operations and dismiss over 950 workers at the start o f the war. Despite this significant labor shedding, the manager o f the company estimates that for the current operation capacity (which is about 20 percent o f its installed capacity) at least 50 percent of its current employees need to be laid off. H e i s unwilling to dismiss the excess labor, as he believes that since the state cannot offer a social solution to the problem, the company needs to carry the burden regardless o f the fact that his workers' pay i s delayed for months. Another private company shared a similar story. The company currently has 1000 formally employed workers, butpracticallyuses only 200 o f them. The other 800 workers either rotate with the 200 that actually work or are just kept on the company's books in order to maintain their social benefits. According to the manager o f the firm,the majority ofthe fictitiously employedhavejobs inthe informal sector, thus delayed wage payments are not a real issue for the company. 93 E.ENTERPRISE RESTRUCTURINGAND EXIT EnterpriseRestructuring 4.57 There has been no mechanismor incentiveto force loss-makingcompanies to either restructureor disappear.As a result, many of themlanguishat low levels of activity andtie up assets that could otherwisebe put to more productiveuse.Data from BEEPS2presented inTable 4:11point to the limited enterprise restructuringrecord ofthe firms inBH.The BEEPS sample o f firms i s skewed toward small- and medium-sized firms, meaning that this i s not a representative sample o f the population o f firms inBH(Annex 1inChapter 3). Nevertheless, the data are insightful. Overall, only 11percent of the surveyed firms have discontinued a product and some 9 percent have closed a plant since 1998. For example, more than 80 percent o f the surveyed state and privatized firms have not discontinued the production o f even one product in the last three years. A complete closure o f plants has occurred in some 17 percent o f the surveyed SOEs and in 14 percent o f the surveyedprivatized firms. These are rather low shares, giventhat many firms inBHare not commercially viable. Table 4:11 FirmRestructuring, 2002 Type of firms by Share of firms which have Share of firmswhich ownership discontinuedat least one haveclosedat leastone product plant State firms 21.74 17.39 Privatized firms 20.93 13.64 New firms 5.41 5.41 All 11.30 8.99 Source: BEEPS2. Note; See Annex 1inchapter 3 for description o f BEEPS dataset. BankruptcyandBusinessExit 4.58 Continued presence of soft budget constraints is associated not only with weak financialdiscipline and the accumulationof arrears, subsidies, poor corporategovernance structure and state ownership, but also with a lack of a credible threat to exit that keeps loss-makingfirms on the market.As Table 4:12 indicates, by the end o f 2002 some 144 firms were in the process o f being liquidated or reorganized through bankruptcy. The reorganization technique was popular among all types o f firms in the FBH, while almost all o f the firms that have gone through liquidation are private firms. Overall, a few state-owned firms have used the bankruptcy procedure despite the acute needfor restructuring or exit. 4.59 As of 2003, experts working in the field estimate that it took almost two years to complete an insolvency procedure. While this i s not a long period in comparison with other countries, the process rarely leads to an efficient economic solution (meaning an insolvency process which results intimely foreclosure, liquidation, or a successful rehabilitation of the firm, Box 4:6). 94 4.60 Well-functioningbankruptcyinstitutionsare only now established.Recently adopted new bankruptcy laws inboth Entities are yet to be underpinnedby well-functioning commercial courts. This has limited the scope and pace o f bankruptcy implementation and contributed to hampering corporate restructuring. . Table 4:12 Bankruptcyinthe FBHby Type of Ownership, endof 2002 State- Socially- Bankruptcy owned owned PrivateCoopera- Mixed Other Total Method firms firms firms tives firms firmsFirms Liquidation 2 0 58 0 0 0 60 Reorganization 19 2 25 1 2 8 57 Other 1 0 5 0 0 21 27 Total 22 2 88 1 2 29 144 Source:FBHStatistical Office. 4.61 Every bankruptcy involves two essential parts. The first is the sale or redeployment of the assets, which should take place as soon as possible. The second is the determination o fthe claims and o f who is entitled to the proceeds o f sale or reorganization. This often takes much longer than the sale part, as the courts are generally very slow in finalizing any sort o f dispute. The inefficiency o f the bankruptcy procedures in BH is due in part to delays related to court procedures (especially in sorting out disputes on creditors' claims) and to other implementation problems, such as the lack o f a registry for pledges and specialized bankruptcy courts, and the scarcity o f well-trained practitioners. In the past, the legal framework in which the enterprise sector operated was only somewhat supportive o f the partial reorganization and liquidation o f unsound firms. The laws were primarily geared towards fast liquidations, with weak, if any, rehabilitation opp~rtunities.~~Box 4:7 summarizes the difficulties involved in undertaking "out o f court" restructuring o f state claims on enterprises and its low likelihood o f success inBH. 4.62 The major limitingfactor on the effectivenessof the bankruptcyprocedurein BH is the legal or institutionalframework, but rather the underlyingpolitical-economyfabric of the country. Specifically, the use of the bankruptcy process is hindered by: (i) political reluctance, especially on the part o f the tax authorities, which have never initiated a bankruptcy application ineither Entity, and (ii) social pressure, combined with a very negative perception of bankruptcy. In such an environment, the effective implementation o f the new bankruptcy framework inboth Entities i s entirely dependent on the political will to exit unviable companies. 4.63 In addition, unique problems stem from the poor coordination between cantons and municipalities and the differences in views regarding their future in a unified economic space. The various levels o f government continue to apply complex rules, which often duplicate or contradict each other. Similar to other institutions, the bankruptcy institution i s administered at the Entitylevel, which further complicates the implementation o f exit regulation. 63 There is a procedure (coercive settlement) that corresponds to a reorganization effort, which allows the conversion o f debts into ownership for creditors or the same for the debtor to maintain the debtor's business activities. 95 Box4:6 ClosingaBusiness(2003) Central Closing a business Bosnia and European The ability o f courts to resolve insolvencies in Herzegovina Averagea BH is shown in the accompanying table. The Actual time (inyears)b 1.9 3.7 Goals o f Insolvency Index for BH is 51, compared with the Central European average o f Actual cost (% o f estate)c 8 21.7 55.4 and the OECD average o f 77, where a Goals o f Insolvency Indexd 51 55.4 higher score indicates a more efficient insolvency Court Powers Indexe 67 52.6 system. An indicator o f the power o f the courts Notes: a Central European countries comprise the transition during the insolvency process is also provided. A economies o f the Czech Republic, Hungary, Latvia, Lithuania, higher value indicates more court involvement in Poland, the Slovak Republic, and Slovenia. the process, usually an impediment to insolvency resolution. BH has a value o f 67, compared with the Central European regional average o f 52.6 Actual time: recorded incalendar year. and the OECD average o f 36. Overall, the table Actual cost: the cost o f the entire bankruptcy process, shows that the insolvency procedure in BH is including court costs, insolvency practitioners' costs, the cost often inefficient, and requires a heavy o f independent assessors, lawyers, accountants, etc. involvement o f the courts and bureaucracy. At Goals of Insolvency Index: calculated as the simple average almost two years, the length o f time for a o f the cost o f insolvency (rescaled from 0 to 100, where higher company to complete a bankruptcy proceeding scores indicate less cost), time o f insolvency (rescaled from 0 compares favorably with the Central European to 100, where higher scores indicate less time), the observance average o f 3.7 years. In addition, the actual cost, o f absolute priority o f claims, and the efficient outcome as a percent o f the estate value is much lower on achieved. The total Index ranges from 0 to 100: a score o f 100 average in BHthan in central Europe. This might indicates perfect efficiency, 0 means that the insolvency be partly explained by the fact that a modem system does not function at all. bankruptcy framework in both Entities has only Court Powers Index: documents the degree to which the recently been adopted and firms have not yet court drives insolvency proceedings and i s scaled from 0 to gone through all o f the necessary procedures. 100, where higher values indicate more court involvement in the insolvency process. Source: See http://rru.worldbank.org/DoingBusiness/database for a detailed description of methodology and access to data F.REMAININGCHALLENGESAND POLICYRECOMMENDATIONS 4.64 Three key findings emerge from the discussion above. First, despite recent progress, the operating environment for firms remains hostile. Second, large, loss-making firms still dominate, hindering the development o f a competitive and contestable market environment inthe country. Third, partly as a result of weak corporate governance arrangements, restructuring efforts in the state and privatized firms have been poor and have yet to deliver increased profitability. 4.65 Progress in second-generation institutional reforms has not been sufficient in BH. Sluggish progress has impededbusiness development and growth, and delayed the international integration ofthe country's companies. To advance these reforms, the government needs to focus on the removal o f structural impediments to the creation and functioning o f competitive markets. It also needs to strengthenthe institutional framework to facilitate business development. Several areas require priority attention, as discussed below. 96 Box 4:7 Out of CourtDebt RestructuringKonciliation in BH Out o f court solutions to financial restructuring are almost always preferable to a court determined solution. However, at present neither Entity government has the information or the tools to effectively participate in out-of- court restructuring which involves a significant value or number o f other creditors. Any restructuring schemes in the future would have to be implemented against the background o f the widespread failure of previous rescheduling o f tax and contributionarrears. Government-led restructuring (presumably by the Tax Department) i s possible where the government is one o f a small group o f large creditors, where each agrees to make similar concessions and to forbear from action against the debtor enterprise during the period o f the agreement. Creditors outside this group o f consenting creditors are not affected, and there i s no "cram down." This i s the classic London Rules Approach which has been copied many times to help resolve other crises. For its part, the debtor is required to undertake a remedial course o f action to repair its financial capability. At present, there are no reliable statistics to help the governments identify the major creditors, nor a basis for determining whether out o f court restructuring i s a possible solution in any particular case. Typically, the debtor needs expert technical assistance to formulate an acceptable plan. It i s unlikely that BH enterprises can do this without assistance. The possibilities o f using this approach, at least inthe RS, may be constrained by the previous rescheduling failures, which indicate that the managements are incapable o f capitalizing on the "breathing room" created. The failure to make current payments o f tax and contributions or the installments specified in the rescheduling program is a separate issue from the profound potential impact of the penalty interest which accrues during the life o f the RS rescheduling program. Source: World Bank, 2004: ECSPF project preparation material. 4.66 For restructuring efforts to succeed, however, exit should be facilitated through the reduction o f subsidies and arrears. Unsound firms should be liquidated, labor rigidities ought to be reduced, and corporate governance and ownership structures should be significantly improved. These policy actions would lead to hardeningo fbudget constraints and more effective restructuring. For this to happen, the government needs to act and signal its commitment to corporate restructuring. Freeing resources from unproductive uses allows for their better reallocation and utilization, and hence promotes the growth o f the economy. Hard budgets stimulate the evolution o f a competitive market environment, which i s currently absent inBH. EnhancingCompetitionand ReducingBarriersto Entry 4.67 Markets inBH are characterized by weak competition and contestability. In addition, the enterprise sector has yet to realize its potential and to serve as an engine for growth and economic development. Regulatory reforms that allow for quick and efficient entry are warranted. In this light, several recommendations could simplify business entry, including the following: (i) Completing the introduction o f a unifiedVAT system across the entire country. (ii) Streamlining business registration and licensing procedures through the introduction and implementation o f a unifiedbusiness registration system. (iii) Removingtheprocessofbusinessregistrationfromthecourtstoadministrative agencies, which would lower the time and cost o f entry. (iv) Enforcing competition policy. Most fundamentally, as the experience o f other transition economies has shown vigorous enforcement o f competition policy i s neededto safeguard vibrant and contestable markets. To this end, the creation o f an independent competition agency in BH (on the Entity or state level) could 97 improve the implementation and enforcement o f pro-competition rules and vigorously guard the competitive nature o f the enterprise sector. On M a y 1,2004, the Competition Council was established. The Council consists o f Council Members who serve for six years. However, to make the Council operative, the government needs to buildhuman capacity and expertise in the agency, and also to allocate the necessary budgetfor its operation. Enhancingthe Securityof Contracts 4.68 The enforcement o f contracts is a serious challenge for enterprises in BH. Meeting this challenge requires regulatory changes to develop an efficient and transparent judiciary free o f political interference. The experience o f other transition economies suggests several reforms that the government of BH could employ to improve contract enf~rcement.~~These reforms are as The Government could remove transactions and procedures that are not business disputes (for example, business registration) from the courts. It should also simplify court procedures, including reducing the opportunities for defendant appeal and delay, minimize written procedures and move toward oral ones, limitingthe need for legaljustification, allowing more discretion on the part o fthe judge inadmitting evidence, and improving court notification. The Government could establish specialized courts (such as courts for small claims and specialized commercial courts) in order to reduce procedural complexity, costs, and time. It could also expand the use and implementation o f alternative dispute resolution mechanisms. The Government could adopt legislation establishing a single High Judicial and Prosecutorial Council for BH as part o f the Copenhagen criteria from the S A A process. The Government needs to assume full national responsibility for the State Ombudsman and to make progress on the merger o f the state and Entity Ombudsmen. Inthis light, the government needs to draA and enact the appropriate legislation, but also needs to ensure the financial independence o f the State Ombudsman. FosteringEnterpriseRestructuringand the Exit of UnviableFirms 4.69 With regard to enterprise restructuring and the exit o f value-subtracting firms, the government needs to implement a comprehensive agenda with a focus on several issues: strengthening corporate ownership, improving corporate governance, reducing the labor rigidities perpetuated to a great extent by the government's passivity, and fostering bankruptcy. 64Based on a detailed review of the practice of commercial dispute resolution in SEE8. See Chapter 6 of Broadman et al., (2004). 98 4.70 Reforming Corporate Ownership. To stimulate restructuring, the state needs to accelerate the privatization o f firms with current state ownership, and to allow for active and motivated owners in the privatized firms to emerge. Changes in the privatization laws in both entities are necessary in the direction o f the unbundling and financial restructuring o f SOEs outside o f the bankruptcy regime. In both broad groups o f state-owned and privatized firms, a proper ownership structure should be in place for the effective restructuring o f the enterprise sector. 4.71 For already privatized firms, the key issue i s to encourage the formation o f concentrated owners with the incentives to establish and operate in a business environment with strong corporate governance and profit motivation. To achieve this, the appropriate steps include the following: (i) Liftthe limits onownership. (ii) Allow more outsiders to purchase more shares inthe firms. (iii) Encourage the state to divest ownership stakes in the privatized firms by selling its shares on the stock market. 4.72 Privatization o f the still state-owned firms, including the strategic enterprises and the public enterprises, needs to be accelerated. To this end, the government should focus on the following: (i) Assessing the enterprises' viability and drawing a clear line between the firms (or their healthy parts) that are suitable for privatization and those that need to be taken offthe market. This i s key to a successful and speedy privatization, (ii) Empowering the privatization agencies to handle the privatization process through streamlined tender and auction procedures and ensuring that the government intervenes inthe process to the minimum. (iii) Changing the current legislation in the FBH to allow sales through auctions o f a firm or its parts. This is currently possibleinthe RS. 4.73 StrengtheningIncentivesfor Sound CorporateGovernance.The governance structure o f the privatized and state-owned firms needs to engender incentives for managers and owners to maximize efficiency gains. Reforms that will improve the corporate governance structure in these firms include the following: Clear separation o f functions between management and supervisory boards. Elimination o f political appointments to ensure the independence o f the first tier boards from the management, employees and large shareholders. Simplification o f the corporate boards system through abolishing the third tier o f boards inBH, namely, the "audit board" inthe FBH and the "supervisory board" inthe RS. Harmonization o f company laws inboth Entities so as to establish equal rights for all shareholders. Public disclosure o f firms' financial statements. This can become a valuable source for shareholders and prospective investors in evaluating the value o f the firms. Improving financial transparency builds trust in the financial position o f 99 firms in the eyes o f banks, investors, and regulators, and improves access to new financial resources. Furthermore, adoption o f the I A S and independent extemal audits can be powerful tools for improving corporate performance, and building investor confidence. 4.74 ReducingSubsidies and Arrears. To overcome the problem, the government needs to strengthen financial discipline in the country, harden budget constraints, and explore the institutional options for exit and for viable restructuring. Tolerating arrears, especially in the large state-controlled companies, sends the wrong signals to the rest o f the market participants: companies are allowed to accumulate arrears to their creditors without financial consequences. 4.75 Facilitating Bankruptcy Procedures. Institutional efforts to improve bankruptcy implementation are pertinent for BH. Improving and harmonizing the bankruptcy institutions in both Entities would lower the barriers to the exit o f unsound firms, would allow the development and growth o f productive ventures, andwould also attract investment and stimulate employment. 4.76 In addition, improving the capacity of the bankruptcy courts is critical to providing a functioning mechanism for the enforcement o f creditor rights and, consequently, for accelerating the transition o f non-productive assets to productive use. 4.77 Lengthy and costly bankruptcy procedures have discouraged companies from using the institutional mechanisms o f exit. The recent adoption of modem bankruptcy laws provides creditors these means to creditors motivated to initiate bankruptcy procedures. However, dispersed and numerous creditors, and the need for their coordination in initiating bankruptcy, might present a key problem. Finding the right incentive structure to motivate the parties is important, and the role o f the government (not only as a major creditor for poorly performing firms, but also as a regulator) i s critical to the success o f bankruptcy. The government should make a serious effort to speed up the implementation o f the bankruptcy procedures and to ensure their enforceability. 4.78 Reducing Labor Rigidities. The next chapter will discuss labor market arrangements and industrial relations and outline ways to make them more conducive to firm restructuring and growth, as well as to employment generation. 100 5. INVIGORATINGTHE LABORMARKET A. INTRODUCTION 5.1 Labor mobility and redeployment have been low in BH by regional standards, especiallyin in the formal sector (World Bank, 2002b). This sector, particularly the budgetary, state and socially-owned sectors, still maintains some o f the features that characterized the labor market during socialism. Lack o f labor mobility, a compressed wage structure, low labor force participation and wage determination mechanisms that allow strong wage growth with little regard to changes in productivity, were all characteristics o f the pre-transition labor market in BH. Given the relative weight o f the formal labor market, many o f its features characterize the labor market a whole. 5.2 These features can be traced to thepre-transition behavior of the labor market in BH. Hoping to obtain severance pay or feeling unable to cope with changes, workers in the formal sector usually get attached to their old enterprises. Job preservation i s also facilitated by managers that see their influence over employees as a way to preserve their own managerial positions. Acting with little regard to profits, SOE managers often show little interest inmeeting the labor restructuring needs o f their enterprises. Wage, tax, and utilities arrears, coupled with subsides, have concentrated among large public sector enterprises, softening their budget constraints, and perpetuating the existence o f such schemes. However, these mechanisms are progressively diminishing in importance as the sector o f protected enterprises is shrinking. Not surprisingly, the sector's share intotal employment i s decreasing accordingly. 5.3 Enterprises in the informal sector seem to operate differently, particularly in the way they recruit, pay and retain personnel. These enterprises are generally smaller in size, recruit mainly younger workers, and on average pay slightly lower but more timely wages. During2001-02, this sector increased its overall employment and its share intotal employment. The sector's lead in net and gross employment creation results from much higher accession and separation rates than in the formal sector. Inother words, the chances o f finding and losing ajob are much greater inthe informal sector. The informal sector i s also less associated with political use o f employment, receives lower subsidies and has a stronger profit seeking orientation. It i s also outside collective labor agreements. 5.4 The labor market in BH shows a noticeable duality between the functioning and performance of its formal and informal elements. Its average performance reflects the combination o f a poorly functioning and relatively rigid formal labor market dominated by large and state-owned companies, and a very flexible informal labor market based primarily on small and medium enterprises. Not surprisingly, BHhas one o f the largest informal sectors in Europe. This i s not a sustainable situation, either from a labor market perspective or from a fiscal point o f view. Informal sector companies have limits on their ability to develop and grow. They also face obstacles to export and are more difficult to integrate into the European production chains. In addition, workers in the informal sector pay no social security contributions and do not accrue pension and healthcare rights. 5.5 This study finds a labor market which is also dual in several other dimensions. Particularly important i s the duality between a relatively flexible labor market in the private sector and a rigid labor market inthe budgetary, state and socially-owned sectors. There is also a striking duality in the way men and women approach the labor market, as well as between poorly-educatedand well-educatedworkers and between younger and older workers. 5.6 The objectives of this chapter are to analyze the recent performance of BH's labor market and to identifypolicy measuresto improveits functioningas a meansof supporting growth in output and employment.The assessment o f labor market policies and performance is challenging, especially in an economy undergoingprofound structural change. This is because labor market outcomes can be influenced by a broad range o f macroeconomic and structural policies, external and behavioral factors at the branch or even firm level. In addition, since the benefits o f labor market reforms (revamped labor laws were introduced inboth Entities in2001) are likely to materialize only over the long term, and since BH introduced most such measures relatively recently, their impact i s not yet fully discernible. 5.7 This chapter starts with a historical ovewiew o f the labor market in BH and the transformation the economy had undergone in the last two decades. The evolution o f wages, employment and productivity i s analyzed using time series data from the formal sector. In the early 2000s, nominal wages grew at a pace unmatched by the growth of productivity. The resulting increase in unit labor cost adversely impacted the competitiveness o f BH firms. The recent release o f two previously unavailable household surveys allows for a more comprehensive analysis, including o f the informal sector and employment dynamics. The chapter reviews the institutional setup and its relevance for wage determination, minimum wages and presents an empirical analysis o f earnings. A special section i s devoted to skill development inBH, covering education, training and active labor market programs. The last section o f the chapter summarizes the evidence and the policy recommendations relevant to the labor market inBH. B.THELEGACYA PARTICULARTRANSITION OF 5.8 BH was born out of the disintegration o f Yugoslavia. The Yugoslav self-management system was based on kocial' ownership o f enterprises, effectively that o f managers and employees. The incentives embedded in the system resulted in persistent wage pressures and poor allocation o f resources. Inefficiencies at the micro level were covered by financial arrangementsthat permitted non-viableenterprises to survive. Enterprises owned banks and were consequently able to influence their lending decisions, softening their budget constraints and "socializing" losses, thereby artificially extending the lives o f otherwise non-viable firms. Enterprise losses were financed through bank loans that were ultimately monetizedby the central bank. The prevalence o f soft budget constraints turned inflation into a chronic problem. In fact, inflation inBHaveraged near 70 percent annually inthe twenty years preceding independence. 102 5.9 Despite its flawed incentive structure, the Yugoslav system was far more market- oriented than those of the Central and Eastern European countries (CEECs) and the former Soviet Union, and did not rely on central planning. Inthe so-called `self-management' system, enterprises were free to make most o f their own decisions regarding pricing, external trade and investment.Banks were owned by enterprises and managed ina decentralized manner. The system allowed greater competition than in the CEECs and generally avoided the shortages o f consumer goods that ofien characterized those economies. 5.10 While generally more flexible, the self-management system did not spare the Yugoslav republicsfrom several of the distortions that saddled factor markets in socialist economies elsewhere.Inparticular, the labor market inBH during socialism was characterized by large employment in the social sector (especially industry) and little geographic mobility o f the labor force. High government and worker involvement in wage determination resulted in a fairly compressed wage structure. However, there were also differences with the more centralized forms o f socialism. The labor market in the self-management system was relatively more flexible. Unemployment existed inBHwell before the beginningo f transition65although it was repressed by political pressures exerted on enterprises to hire beyond needs, resulting in considerable overemployment. Systemic pressures to induce labor force participation were far more moderate than in the more centralized socialist economies. Consequently, activity rates in the SFRY were lower compared with the more centralized socialist economies inEurope.66 5.11 The historical perspective is particularly important in the case o f BH as many of these labor market features remain entrenchedand can be relatedto slow reformin many areas. Infact, activity rates for women remain low nowadays, government and worker intervention in wage determination high and the resulting wage distribution compressed. The public sector continues to be an important employer. Although industrial employment has fallen significantly, labor hoarding continues to be an issue, particularly in protected, state-owned industries, while labor mobility remains low inthe formal sector. The FormalLaborMarket ThroughoutTransition:Structureof FormalEmployment 5.12 An evaluation o f the BHlabor market would ideally be based on an analysis o f extended trends inlabor force participation, employment andunemployment. However, the lack o f survey- based labor market data before 2001 constrains the coverage o f longer periods primarily to the formal sector. The picture o f the post-war period which emerges i s one o f limited labor mobility, important recovery o f wages and a timid pickup o f employment at a pace far below real GDP growth. 5.13 As demonstrated inChapter 1, BHhas seen significant economic growth since the end o f the war. The drop in economic activity was much sharper than in other transition economies6' and so was the rebound. Postwar employment in the formal sector in BH was 65 Unlike other socialist countries, the former SFRY recognized the existence of unemployment in the 1950s. Unemployment remained low untilthe 197Os, however. 66 Activity rates in BH were among the lowest in SFRY and consequently, among the lowest among the socialist countries. 67 At one point duringthe war, recorded GDP fell tojust 17percent o fits 1989 level. 103 drastically reduced, with formal employment only 59 percent o f the comparable 1991 level in 1997 (the reduction was pronouncedinboth entities; Table 5:1), and some 65 percent in2003. Table 5 1FormalEmployment By Sector of Activity, 1991and 1997-2003 (in thousands) 1991 1997 1998 1999 2000 2001 2002 2003 A. Bosnia andHerzegovina TotalI' 976 575 639 628 641 627 637 634 Agriculture " 36 n.a. n.a. 21 21 21 21 20 Manufacturing" 498 n.a. n.a. 256 255 247 239 235 Services" 441 n.a. n.a. 351 365 359 372 369 B.Federation Total 638 373 395 408 410 407 394 387 Agriculture 21 n.a. 10 10 10 11 11 10 Manufacturing 325 n.a. 161 161 161 156 151 146 Services 292 n.a. 224 237 241 241 232 231 C.Republika Srpska Total 338 202 244 221 228 220 238 237 Agriculture 15 n.a. n.a. 11 10 10 10 10 Manufacturing 173 n.a. n.a. 96 94 92 88 89 Services 149 n.a. n.a. 114 123 118 140 138 Notes: 1/ Totals for 2001 and 2002 include employment inthe Brcko district that are excluded inthe sectoral breakdown. The sectoral breakdown, as a result, does not add up to the total. 5.14 Formal employment has also undergone some important changes in structure. Above all, the postwar formal workforce has become significantly older on average. The share o f less educated workers in employment fell substantially. In comparison to the prewar years, employment has shifted away from manufacturing, where it fell in both absolute terms and relative to the total. From 1991 to 2003, the share o f manufacturing employment fell by 12 percentage points, a similar decline to that in Bulgaria and Romania, but steeper than the reduction in Croatia, FYR Macedonia and SAM (WIIW, 2000). With agriculture stagnating, employment in services commensurately increased. In comparison to the FBH, in 2003 the RS had a higher share of workers employed in manufacturing (7 percentage points more), and in agriculture (2.5 percentage points more). The Formal Labor Market Throughout Transition: The Growth of Wages 5.15 The economic recovery was also accompanied by significant general wage inflation in both Entities. Table 5:2 shows the evolution o f aggregate nominal net wages during 1998- 2003 period. Net wages grew from KM 296 a month in 1998 to KM 486 in 2003, a cumulative increase o f more than 64 percent. Even after controlling for consumer price inflation, which averaged around 2.2 percent per annum over the same period, real wage increases remain significant. This wage inflation spiral appears to be led by relatively highpaying firms and the government sector, with systemic spillovers to the rest o f the economy. 104 Net Wage Employment Net Wage Employment Net Wage Employment monthly growth Growth (%) monthly growth Growth (%) monthly growth Growth (%) wages (%) wages (%) wages (%) Year (KM) (KM) 1998 329 170 296 1999 374 13.7 0.8 216 27.1 -9.6 343 15.9 -3.1 2000 414 10.7 0.3 277 28.2 3.2 374 9.0 1.5 2001 444 7.2 -1.4 309 11.6 -3.4 443 18.4 -2.3 2002 484 9.0 -3.8 347 12.3 6.7 446 0.7 1.9 2003 530 9.5 -0.7 385 11.0 0.0 I 486 9.0 -0.6 5.16 Rapidwage growthhaspositionedBHon the highside of the distributionof average wages across the countriesin the region.As shown inTable 5:3, wages inBH are second only to wages in Croatia and above average wages in Serbia. Interestingly, the relative position o f average wages inBHin2003 replicates closely the one before transition; the exception beingthe relative position o f BH wages vis-a-vis Serbia.68However, the structure o f GNI per capita has changed dramatically, with the relative position o f BH deteriorating markedly. That is, wages in the formal sector inBHrecovered their relative position vis-a-vis the other countries, outpacing the recovery o f economic activity. Table 5:3 Net and Gross Wages, 2003 (In euro per month) FYR Bulgaria Romania Macedonia Croatia Serbia FBH RS BH Net Wages 115 130 194 521 176 268 194 247 Gross Wages 145 179 327 743 255 394 294 367 Source; National statistical offices and central banks. c.AMORE COMPREHENSIVE LOOK THE LABOR AT MARKET 5.17 The data reported above show developments inthe formal part o f the labor market. With the release o f the 2002 survey Living in Bosnia and Herzegovina (LiBH), labor market data covering more than one year are now available for the first time.69This enables more insights into two previously not well analyzed aspects o f the functioning o f the labor market: the informal sector and employment dynamics (Table 5:5). This section examines the characteristics o f the labor market in BH, including skills and age composition using micro level data. It also '* Wages in Serbia before transition were on average higher than inBH. 69This allows analysis o f two cross sections o f the Living Standards Measures Survey (LSMS), conducted in2001 and 2002. The LSMS i s a household level survey, with several components including a detailed labor market module. In 2001, the sample covered 9,445 individuals, o f which 7,313 were selected and surveyed again in 2002. Individuals living inRS account for 55 percent o f the sample, while women account for almost 52 percent. The two waves form a panel data allow for a unique opportunity to address key questions regarding labor market outcomes in BHingeneral, and employment and wage determination inparticular. 105 investigates the employment and wage structure across different sectors, including changes in this structure between 2001 and 2002. The surveys reveal some interesting developments in the labor market that may challenge to a certain extent the prevalent understanding o f BH labor market inareas such as labor force participation and overall labor market flexibility. 5.18 In other words, relatively high wages by regional standards are not backed by a proportionately higher labor productivity. Using GNI per capita as a proxy for labor productivity, wage differentials between BH and other SEE countries are larger Table 5:4Gross Wages inManufacturing 11 than productivity differentials (Table Monthly GNIper Gross wages/ 5:4). Put differently, productivity gross capita GNI per capita adjusted wages are high inBH compared wages (percent) with other transition countries. For (US$) example, manufacturing wages in BH 219 1,230 18 Croatia are about 125 higher than inBH; Bulgaria 109 1,520 7 however, productivity in Croatia almost Croatia 494 4,620 11 four times greater than in BH. This implies that despite higher wages, unit Czech Republic 341 5,250 6 labor costs are lower in Croatia than in Estonia 281 3,580 8 BH. Similarly, while manufacturing Greece 1432 11,960 12 wages in Bulgaria and Romania are Hungary 312 4,7 10 7 about half those in BH, labor Ireland 1804 22,660 8 productivity inthese countries i s higher. Latvia 225 2,920 8 Lithuania 247 2,930 8 5.19 The high unit labor costs in BH Poland 457 4,190 11 reflect both a large tax wedge and high Portugal 679 11,120 6 net wages, primarily the latter. The tax wedge in BH i s broadly similar for the Romania 112 1,670 7 two entities. But while the tax wedges Slovak Republic 255 3,700 7 have recently been declining, average net Slovenia 793 10,050 8 wages have increased. In fact, the strong Spain 1461 15,080 10 real wage growth inBH over the past six Source: Wages: Yearbook of Labor Statistics 2001, ILO; Gross years has significantly contributed to the National Income: World Bank Atlas, 2002; Exchange rate: current high unit labor costs.70 Factors World Development Indicators, The World Bank, 2002. which may have contributed to these developments include the strong bargaining position o f workers with `protected' jobs, and the hnctioning o f the wage negotiation process (Collective Agreements) in the formal sector. The latter seem to have an impact on both the aggregate and inter-sectoral evolution o f wages. There is evidence of spillover effects, where wages continue to move together in the labor market through mechanical adjustments mandated by the Branch Collective Agreement. To understand the factors behind this wage evolution, we analyze workers' mobility in the next section and the wage mechanisms and institutions in the subsequent section o f this chapter. 'OWages inthe FBH, nevertheless, are significantly higher than in the RS. 106 Table 5:5 Sample Description, LSMS, BH2001 (in percent) Entity Gender Federation 44.69 Male 48.2 RS 55.31 Female 51.8 Age Education 15-18 Unfinished 7.63 Elementary 15.12 19-24 11.44 Elementary 29.43 25-34 15.94 Vocational 31.04 3 5 - 4 4 17.78 High School 17.66 45-54 17.29 College (2Y) 3.32 55-64 13.71 College (4Y+) 3.37 65 + - . 16.21 Source: LSMS2001. Notes: The estimatedfrequency distributionsare all weighted. Employment,Unemployment, andLabor Force Participationin2002 5.20 Based on the 2002 LSMS survey, the following emerge as the salient features about aggregate labor market stocks and participation rates inBH: Employment rates are very low, below 43 percent of the working age population, with lower rates for the FBH-about 39 percent- than for the RS-about 49 percent. The share of employment in the informalsector is an extremely high37 percent. This share is 32 percent inthe FBHand an even higher41 percent inthe RS. The proportion of women in informal employment is practically the same as their share intotal employment. At 21.4 percent in 2002, the unemployment rate is one of the highest in the region. It is similar inboth the FBH and the RS, but it is higher inboth Entities for women than for men. The level of youth unemployment is alarming,reaching 56 percent for 18 years old or younger and close to 44 percent for the 19-24 years old. Female labor force participation is low at about 39 percent of the working age population, one of the lowest in the region. Generally, participation rates are significantly higher in the RS than in the FBH, including for the female workforce. Therefore, while the participation rates in the RS are in line with regional averages, BH's low overall participation rates reflect low rates in the FBH, particularly for women. 5.21 The age and education distributions reveal a relatively young and well-educated population. About 20 percent o f the working population are less than 24 years old and almost 50 percent have a high school degree or vocational training. Vocational training i s the most common among individual educational attainments. 107 EmploymentandUnemploymentStructure 5.22 The public sector continues to provide a large portion of employment in BH (Table 5:6).71However, its share declined from almost 49 percent in 2001 to 42 percent in 2002. The private sector's employment share increased in 2002, but to a still very low 36 percent. As mentioned earlier, the informal sector employed a significant close to 40 percent o f all workers in both years of the sample. More than 50 percent o f the employed are in the agriculture, manufacturing or construction sectors. 5.23 Table 5.7 breaks down the employment rate and unemployment rate by age and education groups in both 2001 and 2002. The unemployment rate shows consistent patterns with the employment rate, with the average increase inthe unemployment rate from 2001 to 2002 spread across different age and education groups. Unemploymentis the highestfor the young and the poorly educated. It decreases significantly with age and educational attainment. The unemployment rate for the young i s very high, 56 percent and 46 percent for those between 15- 18 years or age and 19-24 years of age, respectively. These rates are substantially higher than the overall unemployment rate o f 21 percent in2002. 5.24 Similarly, the employmentrate rises with age until 55 years, before dropping very rapidly.The employment rate also varies across different education groups. Itis very low for the uneducated: 20 percent in 2001 and 25 percent in 2002 for individuals with no elementary education. It increases with educational attainment with significant differencesfor university graduates. The increase in the aggregate employment rate between 2001 and 2002 i s most pronounced for older and less educated individuals, while the young appear not to have benefited from this increase inemployment. Table 5:6 EmploymentDistributioninBH, 2001 and 2002 EmploymentShare (in ) 2001 2002 2001 2002 Public 48.83 42.17 Agriculture and Fishing 19.43 20.86 Private 33.43 36.18 Mining 2.97 3.17 Others 17.74 21.65 Manufacturing 20.05 19.16 Utility 2.5 2.52 Informal Sector 37.22 39.74 Construction 11.34 10.26 Formal Sector 62.78 60.26 Trade 9.18 9.45 Hotels and Restaurants 4.27 4.06 Storage and Communication 6.08 5.89 FIRE 1.6 1.74 Others 7.97 9.94 Public Administration 7.16 5.5 I Education 4.13 4.08 Health 3.31 3.37 Note: Total employment inthis tables includes workers inand out working age (15-64). 71The definition o f Public Sector in this section encompasses the Consolidated General Government as well as publiclyowned enterprises. 108 Table 5:7 EmploymentandUnemploymentRate,by Age andEducation,BH 2001 and 2002 (inpercent) UnemploymentRate EmploymentRate 2001 2002 2001 2002 Age 15-18 53 56 5 6 19-24 36 46 29 27 25-34 19 28 48 48 35 4 4 11 15 58 60 45-54 10 11 54 57 55-64 6 8 22 27 Education UnfinishedElementary 14 13 20 25 Elementary 20 28 26 29 Vocational 18 23 50 51 High School 15 21 44 45 College 2Y 8 8 58 58 University 4Y 2 5 75 76 LaborDynamics, 2001-2002 5.25 The panel structure o f the LSMS 2001 and the LiBH 2002 also allows further investigation of some dynamic aspects o f employment in BH. Inthis section, we describe some basic statistics on job changes. We then analyze labor market flexibility from a micro perspective; that is, that related to labor market stock and flow analyses of employment, unemployment and non-participation, as well asjob creation andjob destruction. 5.26 The evolution o f the labor market between the two panel waves7* is relatively positive. Despite a severe jump in the unemployment rate by about 5 percentage points, which understandablyattracted considerableattention, employmentevolved in a positivefashion over the two years. Extrapolting the survey results, an estimated 50,000 jobs were created on a net basis, resulting in a 5 percent growth in employment. There was a significant increase in private sector employment (particularly in small and medium enterprises) and a moderate one in the public sector. There have also been significant changes in the sectoral composition o f employment and in the weight o f public sector employment. All these changes are in line with the expected changes duringtransition. 5.27 Positively, and in contrast to previouslyheld beliefs, labor market participationin BH has also provento be very responsiveto changes in available employment.An increase o f close to 12 percent in the number o f active persons injust one year shows that individuals are 12 The 2001 LSMS involved 5,400 households. The LiBH Wave 2 sample is made up o f over 3,000 households drawn from the LSMS, using a corrected sample that enhanced the representative nature o f the data. For the purpose o f analyzing the dynamics o f the labor market in BH during 2001-02 we use a panel o f only those households interviewed inboth waves inorder to avoid variations introduced by the rotationo fpart o f the sample. 109 willing to participate even at the available wage rates if and when job opportunities become available. On the negative side, employment growth was driven by the informal sector. Formalizing at least part o f this job creation without hurting overall employment remains a key challenge inthe years to come. Table 5:8 MainLaborMarket Indicators 2001 2002 Unemployment rate 16.2 21.4 Employment rate 40.7 42.7 Participation rate 48.5 54.3 GrowingInformalityandFallingPublicSector Empl~yment'~ 5.28 An estimated 242,000 gross jobs were created from 2001 to 2002. This is defined as the number of working age persons that were either unemployed or inactive in 2001 and that were employed at the time o f the following wave. In other words, an important 23 percent o f those employed in 2002 were inactive or unemployed in the previous period. This indicator (which omits labor-to-labor mobility) shows a non-negligible employment creation. In the opposite direction, the level o fjob destructionwas also important, albeit to a lesser degree. Some 193,000 people within working age that reported to be employed in 2001 were either unemployed or inactive by 2002. 5.29 Does this new evidence contradict the characterization o f the labor market in previous studies as rigid and with little capability o f introducing the necessary changes required by transition? As earlier studies were based on data collected from pension institutes, they could only analyze the dynamics o f the formal labor market. The second wave o f the household survey provides for the first time a view o f the dynamics o f the whole labor market. The resulting far more dynamic picture o f BHlabor markets by no means contradicts previous studies. 5.30 In fact, the view of a rather static formal labor market remains valid in light of the limited mobility encountered in the formal sector. It is in the informal sector where the majority of job creation and destruction occurs. The formal sector in BH shows severe symptoms o f sclerosis, including low labor sheddingandjob creation. Only one-eighth o f the net employment gain during the period o f analysis was in the formal sector. While formal sector employment increased by a non negligible 2.8 percent, employment in the informal sector grew by 9.3 percent. A similar picture emerges for labor shedding, even thoughjob destruction inthe formal sector was unusually high during this period as a result o f an important contraction in public sector employment. 5.3 1 Of those previously unemployed or inactive individuals that became employed between2001 and 2002, some 71 percentfoundjobs in the informalsector.It was also inthe informal sector where the majority o f the job losses occurred. However, the incidence o f the l3In this section, the definition of public sector includes both the consolidated general government and public enterprises. 110 informal sector injob destruction was somewhat smaller. In fact, some 67 percent o f those who lost or left ajob between the two surveys had ajob inthe informal sector at the time of the 2001 survey. Nevertheless, within the context o f an overall expansion in jobs, the informal sector created more than 87 percent o f all netjobs gained in2001-02. 5.32 The estimatedshare of informalityin total employmenthas increasedfrom roughly 35 percent to 37 percent in just one year; a phenomenon with important consequences for the quality of employment as well as for fiscal performance. Interestingly, the share of youth in informal employment has decreased, that o f older workers has increased and that of prime-age workers has remainedrelatively constant. EnterpriseSize andLabor Mobility 5.33 Labor mobilityis highly concentrated among small and medium enterprises.While enterprises with fewer than 51 employees accounted for half o f total employment in 2001, they created more than four-fifths o f the newjobs and destroyedalmost three-quarters o f the oldjobs. The cohort of SMEs was the only one with net job creation during 2001-02. The remaining cohorts generated a net contraction of employment while their share in total labor creation and destruction was below their share in total employment at the beginning o f the period. This difference i s connected to the ownership structure o f enterprises, the creation andperformance o f de novo enterprises and the status o f the privatization process in BH. In fact, even though privatization has been underway for several years, BH has managed to transfer ownership primarily in small- and, to a lesser extent, medium-sized firms, with the state still holding majority ownership stakes inthe remaining companies. Table5:9 Compositionof employment (2001) and of employmentcreation and destruction(2001-2002) (byenterprise size) Employment Empl. Empl.Destruction 2001 Creation 2001-02 2001-02 1--50 50.4 81.4 73.7 51-200 25.6 13.0 13.6 201 + 24.0 5.6 12.7 100 100 100 5.34 Alternative evidence arising from the BEEPSZ, coveringthe 1999-2002period, also shows that netjob creationwas entirely concentratedamong the newly created firms while employment contracted among state-owned and privatized firms during this period. Complementary analysis o f a sample o f enterprises inBHshows that the overall job flow rate in firms with more than 100 employees was negative during 1997-1999, with 4.2 jobs created on average and 5.3 jobs were lost (World Bank, 2002b). Labor Force ParticipationandUnemployment 5.35 Low labor force participation, comparedwith other countries in the region, is a key problem characterizing BH's labor markets. While male labor force participation is on the 111 low side, female participation has been extremely low, and among the lowest in the entire Europe.74However, according to the latest data coming from Labor Force Surveys, the increase in employment seems to have triggered a rush towards participation, which has surprisingly reached all age groups, even the elderly. 5.36 The recent increase in participation has surpassed the increase in employment, however, resultingin a significant increase in the unemploymentrate. Consequently, injust one year and in the context o f an employment expansion, the unemployment rate jumped from 16 to 21 percent. Every single age group has seen its ratio o f employment to working age increase, along with participation and unemployment. D.ANALTERNATIVE LOOKAT LABORMARKETFLEXIBILITY InstitutionalFramework 5.37 So far we have analyzed the flexibility o f the labor market from a micro perspective. It is also important to explore labor market flexibility from a macro perspective with a focus on the flexibility underpinning labor market institutions and wage determination mechanisms. This involves the analysis o f labor laws, labor taxation, minimumwage, unemployment benefits, and active measures. Improving the macro flexibility o f labor markets i s one o f the key preconditions for addressing key weaknesses o f BH's labor market. A labor market with strict rules to protect job security tends to have more stable jobs but also more long-term unemployment, greater non- participation and larger informal employment. Employers may be reluctant to hire workers if they face constraints in dismissing them for business reasons. Together, restrictive hiring and firing regulations increase the protection o f incumbent employees but reduce access to formal employment. Excessive informal activity is evidence that the formal labor market i s not functioning properly. Formalization o f informal employment would help improve both the fiscal position o f social security funds and prospects for those currently employed in the informal sector. 5.38 BH's significant recent progressin reformingrigidlabor market legislationneedsto be preserved and deepened.A new Law on Labor, adopted inDecember 2001, provided more flexibility both in the firing and hiring o f workers. These changes are the basis for the efficient reallocation o f labor to increase labor productivity. The InstitutionalSetup for Wage Determinationinthe FormalSector 5.39 In Yugoslavia, wages were determined under the self management system. The government set the firm's wage bill, with the objective to minimize differences across firms. Within the firm, individual wages were set by workers. This resulted ina very compressed wage distribution, both across and within firms. In 1988, the Yugoslav Law on Enterprise formally ended the self management system by transferring wage decision making from the workers to the employer (equity owner). 74Labor force participationinBHwas one of the lowest amongall Yugoslavrepublicsduringthe SFRY period. 112 5.40 Today, BH's labor market is governedbyvery elaborateinstitutionalarrangements, with variation across entities, sectors, and cantons.The institutional framework that regulates the determination o f salaries and payments is defined by the Entity labor codes. According to these codes, the government, trade unions, and employers' associations negotiate and sign a general collective agreement (GCA) whose outcome i s obligatory for employers represented in the bargaining process. However, the ministry o f labor could extend the coverage of the collective bargaining agreement to employers which did not participate in the conclusion o f the collective agreement. Collective agreements and the rule book regulate the minimum wage and the terms and methods o f its harmonization. Labor codes also set a range o f entitlements including severance pay, termination notice periods, annual leaves, overtime pay, and work experience premium. 5.41 In the FBH, until very recently, only state-owned enterprises participated in bargaining. For private enterprises, collective bargaining agreements have been non-binding since the 2000 legislation. The public administration also has separate collective bargaining agreements for the Federal government and the different cantons. Trade unions in both entities have significant bargaining power that extends to the pre-war era. This power was reflected in the process and outcomes o f the current GCAs which predate the approval o f the labor codes in both entities. With the government and the Chambers o f Commerce (representing mostly the management o f SOEs) heavily concerned about social outcomes at the time the GCA is concluded, little heedwas paid to the interest o f the owners. However, this has started to change and as the private sector continues to expand, newly formed employers' associations are expected to play a more significant role infuture negotiations. 5.42 On the basis o f the GCAs and the labor codes, Branch Collective Agreements (BCAs) grouping related activities in production are negotiated among governments, trade unions and employer association representatives. The BCAs follow the GCA, but may specify more generous conditions. More importantly, they determine a base wage and coefficients for the complexity o f the specific task. Workers are classified into nine categories based on the level o f education, training and personal qualifications. The base wage, the complexity coefficients and the experience premium are used in the calculation o f paid wages. Rule books o f operations at the firm level tailor sectoral arrangements to enterprises' particularities. 5.43 In each Entity, governments, trade unions and employers' associations are engaged, within the framework o f the recently established Social and Economic Councils, in ongoing discussions on reforming the wage determination system, including indexation formula and the minimumwage (see below). Impacton Wage Formation 5.44 Usingindividuallevel data, we try to answer a set of important questionsregarding wage determination.To what extent does the institutional setup introduce rigidity in the labor market? Is the wage distribution overly compressed? What i s the effect o f minimum wages on the overall wage distribution? Do wages reflect market forces and individual characteristics? 5.45 We start by investigatingwage inequality and how it changed between 2001 and 2002, distinguishingbetweenthe formal and informal sectors. These differences could shed additional light on the effect o f wage setting institutions in the labor market in BH. Figure 5.1 113 displays the kemel estimates o f the log wage distributions in the formal and informal sectors in BH, in 2001 and 2002. In2001, the differences between formal and informal sector wages are shown inthe higher density inthe left tail o fthe informal sector wage distribution, and the higher mean for the formal sector wage distribution. In 2002, these differences almost disappeared and the overall wage distribution appears to have become more compressed. 5.46 Most the differences between formal and informal sector, as well as the changes between 2001 and 2002, are attributedto wage changes in the FBH. In 2001, wages in the formal sector inthe FBHwere considerably more compressed than wages in the informal sector. In2002, these differences narrowed significantly. Figure 5.2 shows the differences in(log) wage distributions between the formal and informal sectors in2001 and 2002 for the FBH and the RS separately. Inthe RS, the two wage distributions in 2001 and 2002 are very close. On the other hand, formal and informal wages in the FBH are distributed very differently. In 2001, the gap between the left tails of the two distributions is very large, the mode for the informal sector distribution is considerably lower, and the right tails o f the two distributions are virtually identical. These features o f the distibutions appear to reveal the binding effect o f the minimum wage inthe FBH. Figure51Wage Distribution,FormalandInformalSector,byEntity,2001--2002 6 - 6 ' I- += -s 0 - . Le+- 7 2 4 6 6 10 2 4 6 8 10 Wage Dstr~butiY,%s2001 Wage Distributiony%iH 2001 114 The Role of the MinimumWage 5.47 One possible explanationfor the differences across sectors and across entities is the effect of the minimum wage. The minimum wage in the FBH is set by the 2000 GCA at 55 percent o f the net average wage inthe economy, adjusted regularly based on publisheddata from the statistical office. In the RS, the minimum wage in 2000 was 68KM for workers in budget financed enterprises, and 8OKM for workers in other enterprises. In 2001, the ratio o f the minimumwage to average wage is estimated around49 percent inthe FBHand 26 percent inthe RS. Inbothentities, the BCAmay specify abase wage that exceeds the minimumwage. 5.48 The minimum wage i s not expected to affect wage setting in the informal sector. However, in the formal sector, a binding minimum wage i s expected to constrain wages from falling below a certain level. One would expect a mass in the wage distribution around the minimumwage as many employers are bound to pay the minimum wage. Spillover effects and mechanical escalation clauses could also influence wages for workers earning above the minimumwage. A non-binding minimumwage shouldnot have suchimplications. 5.49 To test the effect o f the minimumwage on the wage distribution, we need to compare the actual wage distribution to the one that could describe wages inthe absence o f a minimumwage. The informal sector is often thought o f as a good candidate for such a counterfactual wage di~tribution.~~ 5.50 The differences between the two sectors inthe two entities, shown inFigure 5.2, support the argument that the minimum wage influences the level of wages in the FBH, with substantial spillover effects. The graph for the RS shows a very small gap between the two distributions. 5.51 The effect o f the minimumwage on wage distribution can be further studied by looking at international evidence. Compared with other transition economies, the level o f the minimum wage as a proportion o f the average wage i s the highest in the FBH. Together, Figures 5.2 and 5.3 suggest a significant effect o f the minimumwage on the wage distribution, particularly inthe FBH. EmpiricalAnalysisof Wage Determination 5.52 entities, gender, educationalgroups and sector^.'^ They also suggest an important institutional An empirical invetsigation of wages reveals significant differences across the influence on wage setting. 75 Infact, the minimumwage could influence wages inthe informal sector through its effect on workers' reservation wage. This means that the minimumwage should be thought o f as a lower bound for wages. 76 In the earning model retained here, net monthly wages are the dependent variable. The explanatory variables include individual characteristics, like gender, age, educational attainment, as well as firm characteristics like firm size, sector o f employment, industry o f occupation. The model also include Entity and job tenure indicators. The model i s estimated separately for the two survey years. Regression results are reprodcued inAppendix 5.3 115 5.53 As shown in Appendix Table 5.1, average wages in the RS are significantly below average wages inthe FBH. Holding everything else constant, workers inthe RS earned about 39 percent less than their counterparts in the FBHin 2001. While this gap has narrowed slightly in 2002, it remains large and significant. 5.54 A gender gap also seems to persist, with women estimated to have earned around 21 percent less in 2001 and 16 percent less in 2002 than their male counterparts. This gap is statistically significant and i s also in line with that in other countries in the region. However, combined with the low participation rate (uncharacteristic for the region) and low employment rate, this pay gap highlights the important disadvantages faced bywomen inBH. Figure5:2 Minimumwage as share ofthe averagewage, TransitionCountries 50 40 30 20 10 0 5.55 The estimated age earning profile reveals systematic differences between the formal and informal sectors. In the informal sector, the estimates suggest a weakly inverted U shape profile, reaching a maximum for workers between 25 and 34 years old in 2001, and an almost flat profile in 2002. In the formal sector, the results show a positive and increasing age effect. Wages are estimated to be more than 18 percent (13 percent) higher for workers 55-64 years old compared to the wages o f workers 25-34 years old in2001 (in2002). 5.56 Returns to education are estimated to be large and significant. In 2001, compared to workers with elementary education, workers with vocational training earned 10 percent more; those with high school diploma 23 percent more; graduates o f two-year colleges 37 percent more and university graduates 72 percent more. All these differences were statistically significant, and increased even further in 2002. In contrast to the steep profile for the returns to education, the effect o f firm tenure on wages seems to be negligible and statistically insignificant. Taken together with the age-eaniing and education-earning profile, the small effect o f tenure may indicate that firm-specific skills are valued less than general skills. 116 5.57 Workers in state-owned enterprises (SOEs) are estimated to have earned between 25 and 19 percent less than workers inthe public administration, health and education in2001 and 2002, respectively. Workers in the private sector are estimated to have earned 18 and 9 percent more than government workers in 2001 and 2002, respectively. All these differences are statistically significant. The compression o f wages across these sectors between 2001 and 2002 i s also consistent with the overall wage compression shown inthe distributions inthe previous section. 5.58 The estimated model also shows significant differences across different firm structures. Wages in small size firms are between 35 and 20 percent lower than wages in large firms. The size effect on wages i s common in many other countries. Given the sizable informal sector in BH, however, it i s hard to disentangle the effect of size from the effect o f formality. WageArrears 5.59 Wage arrears are widespread among firms in both Entities. Faced with high labor costs and/or deteriorating revenues, a number o f firms either delay or reduce wage payments. The top panel of Table 5.11 shows the shares o f workers inboth entities that were accumulating arrears in2001 and 2002. Inthe RS, about 42 percent o f workers have not received their pay for at least 2 months, down to one-third in 2002. While arrears are less common in the FBH, the share o f workers accumulating arrears continues to be large and significant. The lower panel in Table 5.11 shows that about a quarter of workers in both entities are receiving a different remunerationthan their contractedpay. E.EDUCATION TRAINING AND 5.60 Educationand training opportunitiesfor BH's youth havenot changed significantly since the Yugoslav era, and in the case of adult training, opportunitieshave substantially decreased. The focus o f the country's politicians since the end o f the war has been more on issues o f political control o f education institutions, than on thinking about what sort o f skills, competencies and aptitudes the country's youth and workforce will need in Europe in the 21" century. 5.61 The overall level o f educational attainment in the workforce is high and the age distribution is biased towards younger people. Moreover, according to the BEEPS, enterprises report that the education and skills of workers is not currently seen as a significant barrier to operations and growth, though this might change as the economy shifts). However, the formal labor market i s rigid and has become relatively inaccessible to young people and those with low skills; and unemployment among youth remains very high. Finally, as would be expected, individuals' wages rise as their educational attainment increases. Moreover, while employers value general skills more than specific ones, approximately halfo f those inthe labor market have secondary vocational training as their highest educational qualification. This calls for a profound change ineducational strategies, especially for the upper-secondary system. 117 Table 5:lO Wage RegressionResults,BHLSMS 2001 and 2002 2001 2002 Coefficient T-Statistic Coefficient T-Statistic Constant 5.969 77.160 5.974 68.530 R S -0.397 -14.020 -0.388 -13.300 Informal Sector Age 15-18 -0.081 -0.420 0.053 0.250 19-24 0.138 1.770 0.022 0.250 35 - 4 4 -0.168 -2.240 0.086 1.090 45-54 0.038 0.410 0.000 0.000 55-64 -0.354 -2.130 -0.088 -0.530 Formal Sector Age 15-18 0.071 0.220 0.362 0.810 19-24 -0.028 -0.360 -0.071 -0.940 35-44 0.069 1.510 0.021 0.450 45-54 0.163 3.210 0.062 1.200 55-64 0.184 2.450 0.138 1.970 Female -0.219 -7.250 -0.165 -5.270 Tenure 1 -0.117 -2.130 -0.039 -0.630 2 0.032 0.570 0.134 1.580 3 -0.012 -0.220 -0.020 -0.400 4 0.063 1.150 -0.067 -1.350 5 -0.050 -0.900 -0.134 -2.570 6 -0.159 -2.740 -0.218 -4.040 Education Unfinished elementary 0.089 1.170 0.164 2.110 Vocational 0.101 2.560 0.154 3.880 HighSchool 0.235 5.230 0.368 8.020 College 2Y 0.377 5.560 0.53 1 7.510 University 4Y 0.719 11.350 0.898 13.690 Sector SOEs -0.256 -6.070 -0.192 -4.340 Private 0.185 3.310 0.099 1.910 Others -0.715 -4.360 0.086 0.470 Informal 0.067 0.910 -0.034 -0.470 FirmSize 1-10 -0.357 -6.040 -0.228 -3.270 11-100 -0.078 -1.660 -0.037 -0.610 101-500 -0.215 -4.500 -0.194 -3.130 Adjusted R-Squares 0.31 0.29 Note: LSMS2001 and 2002. Dependent variable i s the net monthly wage inlog form. The excluded group are men, 25 to 34 years old, with elementary education, working in the public formal sector in a large 500+ firmsize. Sample weights were used inall regressions. 118 Table 5:11 Distribution ofworkers receiving arrears or not usualwages, By Entity for 2001 and 2002 (percent) 2001 2002 Share of workers receiving arrears RS 42 33 FBH 19 16 Share of workers receiving actual pay differentfrom contracted pay RS 24.1 25.6 FBH 25.6 27.2 Note: Arrears are defined a5 a delay insalary payment of 2 months or more A New Perspective on Skills 5.62 What sort of skills, competencies and aptitudes are r quired to supp rt economic growth? This question should be examined in the context o f BH's transition from a largely protected economy dominated by large, socially-owned industrial enterprises, to an economy that must compete in a broader European and world markets and must depend on the development o f the service sector and the growth o f small companies. Inthe global economy, BHwill not be able to compete on the basis o f its comparatively low-wage workforce. Private enterprises o f all sizes will be looking to fill knowledge-intensive, value added, and higher-earning jobs in part to pursue exports and to attract foreign direct investment. Evidence from other countries (de Ferranti et al. 2003) has shown that investment decisions, both foreign and local, depend on investors' perceptions o f whether or not the local workforce has the skills, knowledge and attitudes to utilize and exploit the new investments. 5.63 A recent study (Birks Sinclair IBHI, 2003) sheds some light on the changing demand for skills in BH. When 102 employers were asked to rank the importance o f various skills, only about half reported that technical skills gained from formal education were essential or very useful, while over 70 percent identified as essential or very useful skills gained from specialized training outside the education system. Even more revealing are the skills that employers found essential (Figure 5:3): good communication skills (identified by 59 percent of employers as essential), business methods (marketing) (37 percent), word processing skills (35 percent), and foreign languages (29 percent). At the same time, only just over 20 percent found technical skills essential (whether gained inthe formal system or the private sector). 5.64 The same study found a worrying gap between what employers are looking for and the perceptions of teachers and secondary education graduates. When asked to identify the most important skill that employers are looking for, about half o f the 152 teachers and over 1,000 recent secondary graduates surveyed chose technical skills gained o f a subject in the formal education system. There was little recognition among teachers or students o f the importance employers give to communication skills andto foreign languages. 119 Implications for Secondary Education 5.65 Other evidence also suggests serious problems with the orientation o f BH's education system. For example, the study also found that the private rate o freturnto all forms o fthree-year secondary vocational schooling in BHwas zero, compared with a 4 per cent return to secondary general education. These relative returns are a function o f the higher costs o f vocational education and the fact that vocational graduates have a higher tendency to be unemployed or employed in low wage jobs (which are not always in the occupational field in which they were trained). Many EU and transition countries are rapidly moving away from a manpower planning model in which the public sector attempts to predict what specific jobs and occupational skills will be indemand inthe future and then fine tune the secondary education curriculum to produce workers who are fully trained for those jobs. This i s both because their economies are moving too quickly for the formal system to catch up and, in any case, because individuals need broader skills to adapt as their jobs change not simply specific skills for their first jobs. Incontrast, fully 80 percent o f secondary students in BH continue to be enrolled in technical or vocational programs and one-third o f all secondary students are enrolled in vocational programs o f three years or less, none o f which offer access to tertiary education and only a very few o f which will lead to eventual employment in that occupation. Moreover, many young people do not receive their first choice o f occupational program or choose a program mainly because a nearby school happens to offer it. Figure 5:3 Importance of Skills as Ranked by Employers 60 50 40 30 20 10 0 Source: (Birks Sinclair IBHI, 2003) 5.66 In addition to offering only limited employment prospects, vocational education is also expensive. Evidence from the Sarajevo and Tuzla cantons shows that vocational programs in BH are up to one third more expensive per student than general education programs (Catic 2003, Levacic 2003). Given all this evidence, it is not surprising that there has been declining 120 parental and student demand for vocational programs and increasing demand for four-year secondary programs (Table 5:13), inkeepingwith trends inOECD countries (OECD 2001). 5.67 The nature of vocational education in BH is gradually changing in a progressive direction. With EUfunding, the curriculum in several programs i s being broadened to include a wider range o f more generic work-related skills and increased general education components. This clearly should be encouraged. However, significant shifts away from vocational education (especially three-year courses), which this report advocates, will require a careful information campaign by government to explain the rationale. At present teachers, parents and students were not overly dissatisfied with current three-year vocational programs compared to general programs, except to note that they are under hnded (World Bank 2003a). Table 5:12 ImpliedPrivateI R R s by Type of Educationfor BH, 2002 Secondary IRR YO HigherEducation IRR Yo Gymnasium 4 General Programs 0 Secondary technical 3 Education 0 Secondary Vocational 0 A r t s and Humanities 16 University 8 Social Sciences, Economics, 9 L a w Science 17 Technical Industry 9 Construction Agnculture 6 Health 0 Source; Birks Sinclair IBHI2003 Notes: (1) All secondary school I R R s have been derived from the comparison with no schooling because there was no significant impact of primaryeducation. (2) The IRR for secondary vocational education and for some higher education programs is zero reflecting no significant impact for these programs. The IRR for university has been derived from a comparison with gymnasium since these schools supply the most university entrants. The IRR for all higher education degrees has been derived from a comparison with gymnasium since these schools supply the most university entrants. I R R s for higher education have been derived assuming that all courses last 6 years on average. Implications for tertiary education77 5.68 As in most countries, tertiary education in BH remains a good investment for the individual. Rates o f retum average about 8 percent, with generally higher rates o f retum for the sciences and humanities and lower rates o f retum for agricultural sciences and other technical subjects (Birks Sinclair IBHI, 2003). University graduates are less likely than those with lower levels o f attainment to be unemployed and much more likely to be employed in the formal sector, particularly the public sector, which makes up a large share o f formal sector employment (World Bank 2003b). Interestingly, however, those with a university degree are also more likely 77 This Report uses the term `tertiary' rather than `higher' education in line with current intemational practice (OECD 1998), to cover all types o fpost-secondary education includingbutnot limited to universities. 121 to be discouraged workers, Le., without work but not actively searching for employment. This may be because these graduates can wait for what they perceive to be the right job, since they can expect higher wages and in any case they usually come from more affluent families who can support them while they wait for work. Table 5:13 Trends in Enrollment of Different Kinds of Secondary Schools School Gymnasium YO Technical % Vocational YO Total Year Schools School FBH 2000101 26118 23.24 42545 37.85 40781 36.28 112399 2001/02 27139 23.70 44211 38.60 40567 35.42 114523 2002103 27496 23.78 45478 39.33 38992 33.72 115632 RS 200112002 9,242 18.0 24,424 47.7 17,581 34.3 51,247 200212003 9,419 18.1 25,965 50.0 16,564 31.9 5 1,948 Source: Federation and RS Statistical Institute 5.69 Like the secondary level in BH, the tertiary education system seems to be failing to provide a steady supply of graduates with appropriate skills in a dynamic labor market. Although the evidence i s more anecdotal innature than with secondary (more research is clearly needed), students, university leaders, some professors and many local stakeholders report that tertiary programs suffer from outdated curricula, teaching methodology, reading lists and mono- disciplinary programs that are highly theoretical and unrelated to labor-market reality. However, few university professors identifiedthe relationshipbetween the labor market and the curriculum as an important issue and most were focused only on the need for more funding, higher salaries andbetter working conditions (World Bank 2003b). 5.70 The reasons for the disconnect between demand for more relevant programs and the perpetuation of traditional teaching and learning at BH's universities lie in the fragmented nature of the tertiary education system, without effective management or accountability. The autonomous-faculty system leads to costly duplication in teaching (with each faculty organizing its own courses in subjects outside its specialization), administration and services. The responsibility for tertiary education lies with cantonal governments in the FBH, which because o f small revenue bases cannot provide significant investments in quality improvements and engage inethnically-based political interference. Moreover, students, many o f whom study free o f charge, are allowed to churn for years in the system. For example, o f all students who began their university studies inthe FBHin 1997/98, less than 3 percent graduated on time (the EUaverage i s roughly 50 percent). At the Universityo f Sarajevo, the ovenvhelming levels o f repetition mean that the costs per graduate in 2002 were near 13 times the annual costs per student (Catic 2003). These flaws are widely recognized throughout the education system and there i s broad consensus on the need to shift responsibility for tertiary education to the level o f the Entity in the FBH or the State. However, political maneuvering has so far prevented adoption o f a modern tertiary education legal framework. 122 Implicationsfor Adult Training 5.71 The inadequaciesof BH's formal education system are compoundedby weaknesses in the supply of quality adult educationandtrainingopportunities.As a result, those exiting the formal system with inflexible skills and a weak aptitude for continued learning have great difficulty overcoming these shortcomings. The proceeds o f an earmarked unemployment benefits payroll tax are devoted largely to passive programs, such as those providing unemployment benefits, rather than active employment facilitation measures, such as job-based and private sector training, job search assistance and counseling. When training does occur, monitoring and evaluation for quality assurance and the tracking o f participants does not take place. Job search assistance, one o f the most cost-effective active labor market interventions, is underdeveloped in BH. 5.72 Although training by enterprises was the principal source o f adult training in the former Yugoslavia, very little job training by enterprises takes place in BH today. For example, as suggested by a recent survey o f 95 adults found that only about a quarter undertook some form o f adult training within three years o f leaving secondary school in 2000 (Birks Sinclair IBHI, 2003). This i s despite a well meaning but non-enforced law that obliges enterprises with more than 50 employees to provide training for workers with work experience over 6-12 months. For the limited amount o f enterprise-based training that does take place, there i s no adequate institutional structure for recognizing certificates, diplomas or transcripts. To the extent training is regulated, it is done on the basis o f setting physical input standards (size o f classroom, number o f qualified trainers, etc.) rather than by attempting to measure what trainees are learning. This i s a weakness shared by the formal education system. 5.73 Training offered by private training providers and NGOs is almost exclusively limitedto computer training and foreign languages, and even for these there is no system for degree or diploma recognition. NGOs are prohibited by law from charging for training courses, limitingtheir ability to expand their services. The relationshipbetweenpoverty and educationin BH 5.74 There is a strong relationshipbetweenpoor educationalopportunityand povertyin BH (World Bank 2003b). BH's 93 percent net primary enrollment rate is the same for those from poor and non-poor households, but it is low by EUstandards, including the EU's 10newest members who have net enrollment rates very close to 100 percent. Although the primary enrollment rate was not found to vary significantly across gender or urban and rural areas, it i s thought that non-enrollment in primary education is heavily concentrated among refugees, the disabled and the Roma. BH's pre-school enrollment rate o f only 4.3 percent is strikingly below European standards for all income levels and very few children from poor families have access to this head start in education. As access to pre-school and kindergarten programs inevitably rises in the coming years, one o f the most pro-poor policies the govemment could embrace, given international experience, would be to provide funding for pre-school and kindergarten programs targeted to the poor and the disadvantaged. 5.75 O f equal concern, only 73 percent o f the 15 to 18 year age group is enrolled in secondary school, reflecting some incidence o f dropouts and the disproportionate share (40 percent) o f 123 students in three-year vocational programs. Most European countries have net secondary enrollment rates o f 85-95 percent. Among the population in BH falling below the poverty line, the share o f secondary enrollment falls to only 57 percent. To make matters worse for poor children, those that are enrolled in secondary schools have a much greater tendency to be enrolled in vocational programs, a development that reduce labor market flexibility and usually leads to unemployment or low eamings. 5.76 Three-year secondary programs do not provide access to tertiary education, compounding the problems faced by the poor in gaining access to higher education. Whereas the net enrollment rate for the non-poor in the tertiary age group (age 19-23) i s 27 percent, it i s only 9 percent for youth from households below the poverty line. Because public higher education institutions in BH offer free admission only to the best performing students (who tend to come from higher income households) and higher education studentstend to benefit from a number o f non-education subsidies such as housing, board and transport, public funding for higher education inBH i s highly skewedto benefit households that needpublic subsidies the least. The top 20 percent o f households inBHinterms o f overall consumption benefited from 41 percent o f all public spending on higher education, while the bottom 20 percent benefited from just 6 percent o fhighereducation spending (World Bank 2003b). 5.77 BH is in danger of seeing declines over time in the educational attainment of the poor as privatecontributionsrise and ifpublic spendingis not re-orientedto support those disadvantagedstudents. However, even ifthe country manages to keep the poor in school, it is only improvements inthe quality and relevance o f their educationthat will help break the vicious circle o fpoverty and make a positive contribution to overall economic growth inBH. F.THEROLEOFEMPLOYMENT SERVICES INBOSNIAAND HERZEGOVINA 5.78 Low job reallocation in the formal sector and existing mismatches between labor supply and demand underscorethe importance of the efficient use of employment services and passive and active labor market policies.However, despite significant resources allocated to these initiatives, inparticular inthe FBH, and a wide network o f offices spanning the country, employment services have proved largely irrelevant in facilitating job flows inthe formal sector, especially inthe face o f a large and absorptive informal sector. InstitutionalArrangements 5.79 The legal frameworks governing the activities of the Employment Institutes (EI) were substantially overhauledin both Entities in 2000, but continueto face a large reform agenda in the area of active labor programs. Employment services are dispensed at both Entity and sub-Entity levels. Inthe RS, there is a single EI, with 6 regional branch offices and 55 municipal offices. The FBH employment service network comprises the Entity-level Federal Employment Institute and ten cantonal Employment Services with 79 municipal offices.78Ofthe resources collected inthe FBH, 70 percent are administered by the cantons and 30 percent by the '*The formation of the Federal Institute structure eliminated previously separate and ethnically-based employment services inFBH. 124 Entity.79Recent reforms have introduced an Employment Agency at the State level which is tasked with compiling country-wide employment statistics and representing BH in international andbilateral labor affairs. Programs 5.80 Apart from registering the unemployed, the Employment Services in BH are tasked with administering unemployment insurance and active labor market programs for the unemployed. Financing unemployment benefits i s the priority inboth Entities and active programs are funded from the excess resources not taken up bybenefits. 5.81 Unemployment benefit coverage in BH is low by regional standards. With cash benefits available only to unemployed who have made unemployment contribution payments (and whose employer has made payments) and rampant contribution evasion, cash benefits accounted for only 16 percent of total E1 expenditure in the FBH and 23 percent in the RS in 2003. Consequently, significant resources remain for the financing o f active labor programs, in particular in the FBH. Due to important differences in contribution rates for the unemployment insurance (2.5 percent in the FBH and 1 percent in the RS),*' the financial positions o f the two EIs are nevertheless very different. 5.82 The current mix of active labor programs in both Entities is very limited, with an emphasis on programssuch as subsidizedwager or credits to firms. However, international evidence suggests that these two are among the least cost effective programs (Betcherman, et. al. 2004). At the same time, simpler and more cost effective programs such as job counseling and basic job brokerage functions have been non-existent or seriously neglected. Inparticular, funds for active programs have been directed to favored state firms with negligible monitoring or accountability for use o f funds. A recent audit and business process review of the Entity EIs identifiedpoor business processes and lax financial management inboth Entities, inaddition to a serious lack o f monitoring o f internal operations and evaluation o f E1programs. 5.83 The resulting lack o f accountability for resource use, exacerbated untilrecently by lack o f appropriate regulations, has been a serious concern, particularly in the FBH. As the FBH spends significant public resources on credit programs (around 0.5 percent o f Entity GDP in2003 out o f a total expenditure o f 1 percent o f GDP, with an additional 0.4 percent o f GDP in unallocated reserves), this i s an unsatisfactory outcome, and one that Entity Governments recognize in the MTDS must improve giventhe highlevels o fregistered unemployment and significant corporate restructuring challenges ahead. Some o f these issues are being addressed; in fact, the RS has recently stopped its credit programs, while the FBHE1i s inthe process o f winding them down. 19Entityfunds are available for redistribution to the cantons on a needsbasis througha solidarity mechanism. 80 Excess financing in the FBH E1 have triggered recent reform discussions about a possible reduction in the unemploymentinsurance contributionrate. 125 Table 5:14 Expendituresof the EntityEmploymentInstitutes, 2003 (percent) RS FBH Administration andcapital investments 33 15 Assistance to the unemployed 26 30 - Unemploymentbenefits 23 16 - Healthinsurance contributions 1 12 - Pensioninsurance contributions 1 2 Labor marketprograms 41 54 Reserves Total 100 100 Total Expenditures(percent of Entity GDP) 0.3 1 Source: Entity Employment Institutes, labor market programs in the FBH include credit programsworth KM39m. 5.84 Inthemunicipalemploymentoffices, simpleregistrationactivitiesarecrowdingoutother activities such asjob referral and counseling. Since the employment services inboth Entities also finance health and pension insurance contributions for the registered unemployed, receiving these benefits i s often the only incentive to register This also applies to a significant number o f people employed inthe informal sector or engaged inagriculture who register as unemployed for that reason (which accounts for the significant gap between figures for registered unemployment and ILO-consistent estimates based on the LSMS). 5.85 The inadequacy of E1 activities is confirmed by surveys of the unemployed which indicate low faith in public employment services to provide them with a useful service when looking for work.81 Moreover, surveyed job-seekers rely on informal networks in their communities to find employment, mostly o f an informal nature, rather than to go to the municipal employment office for help. This i s compounded by perceived better employment opportunities in the informal than the formal sector, according to anecdotal evidence. The flip side o f this unsatisfactory job referral performance i s that, on the labor demand side, the employment services are equally unattractive to employers seeking further hires.82 Employers see advertising and personal contacts as more useful ways to recruit. Failure to keep integrated Entity-wide (or let alone country-wide) records o f unemployed by set o f skills means that the employment services have little to offer employers intheir strategic search. G.CONCLUSIONSAND POLICYRECOMMENDATIONS What DidWe Learn? 5.86 BHhas seen significant growth over the last several years, fueled by aid flows supporting reconstruction after the end o f the war. In the labor market, this growth was reflected in the 81See Birks Sinclair (2002) for surveyresults o f theunemployedfrombothEntities. 82Ibid. 126 early 2000s by increased number of entrants, and increase in employment opportunities. It was also reflected inoverall highwage inflation inboth the RS and the FBH. 5.87 The labor market in BH is characterized by a dual structure, with a dynamic informal sector covering small- and medium-sized enterprises and the self-employed, and a formal sector still dominated by public workers and large enterprises. Employment creation duringthe early 2000s was primarily inthe informal sector, while the formal sector continues to struggle with significantly highlabor cost and debt overhang. 5.88 The BH labor market is characterized by a very low participation rate, a low employment rate, a high unemployment rate and a large informal sector. Between 2001 and 2002 the labor market was not able to absorb all the new entrants into the labor force As a result, the unemployment rate increased to an even higher 21 percent by 2002. The unemployment rates for young men and women were around 3 times the national average, while the employment rate for older workers is significantly below the national average. 5.89 Concerns about equity in the wage determination, possibly due to cultural and historic reasons, are clearly reflected in a compressed wage distribution and (accrued) wage inflation driven by automatic indexation mechanisms. As a result, these mechanisms intorduce distortions and keep formal employment artificially low. Informal employment i s stimulated, by contrast, underpinned by incentives to underreport and underinvest. While the number o f "waitlisted" people decreased significantly inthe early 2000s, wage arrears and other distortions continue to hamper corprorate restructuring, including privatization. 5.90 Preliminary evidence at the industry level points towards a wage inflation spiral led by relatively high paying firms, but systematically spread over other sectors in the economy. Wage rigidity in general, and the minimum wage in particular, could become more binding as growth slows down, and the economy goes through severe restructuring and adverse demand shocks. 5.91 At the micro level, databetween2001 and 2002 suggest a lack of flexibility in the labor market hinders resource allocation and employment creation. Women and youth appear to have suffered the most inthe labor market, with a particularly low female participation rate and a significantly high youth unemployment rate. Older workers seem to have also been differentially affected by relatively highwages leading to a very low employment rate. 5.92 Returns to education are very large and are reflected inhigh earnings and favorable employment outcomes. While the age earning profile i s rather flat in the informal sector, it i s steep in the formal sector. Wages in the SOEs are lower than those paid to employees in the public administration, health and education. Policy Implications and Recommendations Reducing rigidities in theformal sector 5.93 As shown in Table 5.14, a large fraction o f workers were accumulating arrears in both Entities during 2001 and 2002. More that one-quarter o f workers are paid below their usual pay 127 levels. In turn, accumulated wage arrears and pendent severance payments have created a debt overhang problem that is hindering privatization. Rigidities in the formal sector are anchored in the behavior o f managers that act with disregard for profits and show a strong orientation towards protecting existingjobs. Accelerated privatization would help break this vicious circle. Minimum Wage 5.94 The evidence in this report indicates a high relative minimumwage in BH compared to other transition economies, and a bindingminimumwage reflected in a rigid wage structure and compressed wage distribution inthe formal sector covered by collective bargaining agreements. To reduce the unwanted effect o f the minimumwage on the wage distribution, it will be essential to untie the minimum wage from the average wage in the FBH. De-linkingthe minimumwage from the average wage would contribute to arrest wage inflation and reduce wage compression in the formal sector. Youth Employment and the Minimum Wage 5.95 An estimated 53 percent (56 percent) and 36 percent (46 percent) o f teenagers and young adults were unemployed in 2001 (in 2002). The corresponding employment rates are very low. Because young workers tend to be hired in low-paying jobs, minimum wage policies are very relevant to their employment outcomes. In fact, the binding minimum wage shown in the previous section might be having an adverse differential effect on youth employment outcomes inBH, and could explain the highunemployment rates for youth in2001 and 2002. To facilitate the entry o f younger workers into the formal sector, new entrants to the labor market and young workers need not to be covered by the collective bargaining agreements and their wages should be allowed to drop below the minimumwage set by these collective bargaining agreements. CollectiveBargaining 5.96 Inorder for the collective agreements to reflect the interests of all concerned parties, it will be important to empower the private sector inthe B C A negotiations and to reduce the scope o f coverage o fthe BCAs to only firms represented inthe bargainingprocess. 5.97 Recognizingthe different nature o f the state as employer, the wage setting process for the public administration should be removed from the collective agreements that determine the salaries o f other formal sector workers. The public sector should move away from BCAs into a statutory setting o f wages for workers in the public administration. In parallel, the Government should improve its control over the enterprises it owns. Income policies could be introduced -on a limited time basis- to arrest wage inflation in a selected group o f firms, particularly among the group o f large enterprises. 5.98 Allowances currently represent a disproportionate share o f total earnings for employees inthe formal sector. Allowances should beprogressively integratedinto the base wage. Old Workers 5.99 Incontrast to younger workers, the regression results above show that wages for older workers in the formal sector are very high. Unlike most estimates o f the earning profile in other 128 countries, the estimated wage age profile for the formal sector inBHdoes not have an invertedU shape. The 0.6 percent experience premium mandated by the labor code in both entities could provide an explanation. However, the low employment rate for older workers (22 and 27 percent in 2001 and 2002, respectively), might be a consequence o f considerably more expensive older workers for employers. I nformality 5.100 The compression inthe wage distribution and the rigidities highlightedabove underlie the relatively high share o f informality in the labor market. While the informal sector seem to be highly dynamic and responsible for the creation of most new jobs between 2001 and 2002, the implications in terms o f fiscal repercussions and social protection for workers are alarming, makingany potential employment recovery unsustainable. 5.101 Policy discussions should focus on reducing the sources o f rigidities as a means to support a gradual shift o f employment to the formal sector. Labor market institutions, particularly payroll taxes, contribute to the co-existence o f a dual labor market inBH. A large tax wedge, particularly in the FBH, produces clear disincentives for SMEs to formalize their operations. Improving the efficiency o f the social security institutions and reducing payroll contributions when their finances so allow it, will help reduce incentives for informality. Education 5.102 In a global economy, BH will not be able to compete on the basis o f its comparatively low-wage workforce. Private enterprises o f all sizes will be looking to fill knowledge-intensive, value added, and higher-earning jobs inpart to pursue exports and to attract direct foreign direct investment. Evidence from other countries (de Ferranti et al. 2003) has shown that investment decisions, both foreign and local, depend on investors' perceptions o f whether or not the local workforce has the skills, knowledge and attitudes to utilize and exploit the new investments. The effects on the poor are disproportional. BH i s in danger o f seeing declines over time in the educational attainment o f the poor as private contributions rise and ifpublic spending is not re- oriented to support those disadvantaged students. However, even if the country manages to keep the poor in school, only improvements in the quality and relevance o f their education will help break the vicious circle o f poverty and make a positive contribution to overall economic growth inBH. 5.103 As access to pre-school and kindergarten programs inevitably rises in the coming years, one o f the most pro-poor policies the government could take, given intemational experience, would be to target funding for pre-school and kindergarten programs on poor areas and disadvantaged groups. While there i s broad consensus on the need to shift responsibility for tertiary education to the level o f the Entityinthe FBH or the State, political maneuvering has so far prevented adoption o f a modem tertiary education legal framework. Employment Services 5.104 The inability o f Employment Institutes to facilitate labor reallocation through active labor programs, in particular through job referral and counseling, calls for urgent further reforms, 129 given the large pool o f unemployed and likely additional demands as enterprise restructuring gathers pace. Some progress is already being made. Both Entities are developing more effective policies on ALPSwhich are consistent with fiscal realities and administrative capacity o f the employment services. A framework for revised programs has been developed with Entity and cantonal employment services, and is being used as a basis for further program development. This reform is informed by recent experience which showed that job referral and counseling as well as focused on the job-training can have a significant positive impact on job reallocation in BH. 5.105 To put this agenda into practice, the employment services and inparticular the municipal offices as main service providers are in need o f capacity buildingto improve their job referral andjob search assistance capacities. Inaddition, the financing of healthandpensioninsurance for the unemployed is in need o f reorganization, in order to re-orient the employment offices' central task from registering people to active programs and to remove the adverse incentive to register as unemployed simply to have access to health coverage. Finally, excess financing inthe FBH E1makes it possible to explore a reduction in the pay-roll unemployment insurance contributionratefiomthe current 2.5 percent closer to the 1percent inthe RS. 130 ANNEX 2.1 FTAS CONCLUDED BY BOSNIA AND HERZEGOVINA Trade`Overage (art' Quantitative 1.2.2) Liberalization pace (art. 1.2.3)restrictions (art. 1.2.1) FTA Share o f HS mutual Share of Share o f HS Endof Numberof (year o f lines trade lines with Year o f entry into Country liberalized liberalized ,upon entry lines transitional Q R ~upon abolition? force) period entry into (%) (Yo) into force (%) force ' BH-CRO Bosnia Herzegovina 100.0 100.0 25.6 1Jan. 2004 (2001) Croatia 100.0 100.0 100.0 entry into force BH-S&M Bosnia Herzegovina 100.0 100.0 25.6 18 2 1 Jan. 2004 (QR on Not (2002) Serbia and Montenegro 100.0 100.0 99.9 exports) specified BH-MAC Bosnia Herzegovina 100.0 100.0 25.6 1 Jan. 2005 (2002) Macedonia, FYR 100.0 100.0 99.9 2 31 Dec. 2003 BH-ROM Bosnia Herzegovina 91.8 83.7 27.1 1Jan. 2005 (2003) Romania 88.7 89.9 88.7 entry into force BH-BUL Bosnia Herzegovina 91.5 75.9 27.2 1 Jan. 2005 (2003) Bulgaria 88.9 95.6 88.9 entry into force ALB-BH Bosnia Herzegovina 93.0 88.6 26.6 1 Jan. 2008 (2003) Albania 91.0 91.7 4.7 BH-MOL Bosnia Herzegovina 100.0 100.0 26.2 1 Jan. 2006 (2003) Moldova 100.0 100.0 44.9 Source: Messerlin and Miroudot (2003). Data: national statistics, bilateral imports 2002 (2001 for SAM). Imports for BHare replacedbypartner countries' exports. Whenmissing, bilateral imports are replacedbyworld imports. 131 ANNEX 2.2 BHOUTWARD PROCESSINGEXPORTSTO THE EU, 1997-2001 Value of BHOutward ProcessingExports to the H S Industry EU(1000's US%) 2001 1997 1998 1999 2000 2001 1997=100 62Apparel articles and accessories, not knit etc. 20,803 36,313 40,397 27,483 27,696 133 64Footwear, gaiters etc. And parts thereof 6,890 12,048 20,283 14,359 11,555 168 61Apparel articles and accessories, knit or crochet 661 3,793 1,705 2,156 2,952 446 83 Miscellaneous articles o fbase metal 1 127 2,44 1 90Optic, photo etc, medic or surgical instruments etc 641 1,294 1,349 1,181 1,079 68 63Textile art; needlecraft sets; worn text art 1,229 28 162 95 248 20 84Nuclear reactors, boilers, machinery etc.; parts 161 703 96 643 163 01 87Vehicles, except railway or tramway, and parts etc 231 46 23 10 41Raw hides and skins (no furskins) and leather 115 63 88 41 17 15 73Articles of ironor steel 1 3 74 72 16 1,422 99Special import provisions, nesoi 14 14 13 39Plastics and articles thereof 22 31 31 7 12 54 85 Electric machinery etc; sound equip; tv equip; pts 264 249 128 10 10 4 Unitvalue (1000'sUS$per ton) 2001 1997 1998 1999 2000 2001 1997=100 62Apparel articles and accessories, not knit etc. 31 29 26 21 23 74 64Footwear, gaiters etc. And parts thereof 27 16 15 12 10 38 61Apparel articles and accessories, knit or crochet 6 17 14 15 20 352 83 Miscellaneous articles o fbase metal 7 12 90Optic, photo etc, medic or surgical instruments etc 4 9 9 9 9 207 63Textile art nesoi; needlecraft sets; worn text art 24 1 3 16 11 46 84Nuclear reactors, boilers, machinery etc.; parts 16 3 14 92 16 101 87Vehicles, except railway or tramway, andparts etc 6 3 12 211 41Raw hides and skins (no furskins) and leather 16 21 22 14 9 52 73Articles o f ironor steel 5 6 5 99Special import provisions, nesoi 14 14 39Plastics and articles thereof 22 10 10 7 4 18 85Electric machinery etc; sound equip; tv equip; pts 29 12 9 3 21 8 Source: COMEXT database. Only products with exports above $10,000 in2001 were included inthe table. ANNEX 3.1: BEEPS The EBRD-World Bank Business Environment and Enterprise Performance Survey (BEEPS) instrument was developed to study the environment in which businesses in 26 countries o f Central and Eastern Europe operate. The survey was carried out twice, in 1999 and2002, which allows a comparisonover time within the participating countries. Over 180 firms from BH took part in the survey in 1999 and 2002. The firms operate in industrial sectors such as mining, construction, or manufacturing, or are active in services such as transportation, trade, real estate, business services, tourism, or other. The survey instrument was designed with intendedoverrepresentation o f smaller firms in all 26 countries. As per vintage and ownership, the respondents were chosen to match sample quotas, so there are state-owned firms, privatized firms, and newly established firms. The profile o f the surveyed businesses from BH in terms o f sector, size, age, and ownership i s presented in Annex 1Table 1. A major caveat o fthe survey is that the sample o f firms interviewed across the 26 countries i s not representative o f each national economy. Instead, the samples are stratified to cover firms with different characteristics within a country. This allows us to make comparisons among different firm characteristics. Yet the samples o f firms interviewed in 1999 and 2002 do not have identical characteristics. For example, the BEEPS1 sample includes some firms in sectors such as farming, fishing, and forestry, power generation, and financial services that are not covered in the subsequent round o f the survey in 2002. In addition, BEEPS1 does not include firms from the hotel and restaurant sector, other community and social services, and real estate services, while enterprises from these sectors take part inthe second round of the survey in2002. Also, sole proprietorships did not take part inBEEPS2, but did participate inBEEPS1. 133 Annex 3.1 Table 1 Sample Structure: BEEPSl and BEEPS2 Numberof Number of BEEPS2 Sample firms" BEEPSl Sample firms* Industry Industry Farminglfishing/forestry 1 Miningand Quarrying 3 Miningand Quarrying . 1 Construction 12 Construction 23 Manufacturing 68 Manufacturing 58 Power generation 1 Services Services Transportation, Storage, Transportation, Storage, Communications 12 Communications 5 Wholesale, retail, repairs 57 Wholesale, retail, repairs 74 Real Estate and Business Service 7 Real Estate and Business Service 26 Hotels and Restaurants 15 Financial Services 3 Other services 8 Personal Services 0 Size Size Small (2-49)*** 111 Small 135 Medium (50-249) 42 Medium 35 Large (250-9999) 29 Large 22 Ownership Ownership State-owned 23 State-owned 56 Private, o f which: 159 Private, o f which: 135 D e novo 114 D e novo 133 Privatized SOE 45 Privatized SOE 2 Vintage Vintage Oldest 1885 Oldest 1885 Youngest 1999 Youngest 1999 Average Age 1984 Average Age 1988 MedianAee 1995 Median Age 1994 Source: BEEPS1&2. Note: The number oftotal firms by categorydoes not match owing to missingvalues. * 134 ANNEX 3.2: 1000LARGEENTERPRISES' SURVEY DETAILS The datasets were obtained to estimate the extent and distribution o f financial distress in different sizes o f enterprises, and in enterprises with different ownership structures. It i s essential to develop a better understanding o f the strategies usedby different types o f firms to cope with what has apparently been substantial financial distress sustained over a number o f years. It is also critical that information be gathered to assist the government and the international community in estimating the projected cost o f possible social programs to deal with the perceivedhighdegree o f excess labor inthe real economy. The datasets contain the financial status o f 1,000 enterprises in the FBH and 1,000 enterprises in the RS as o f 31 December, 2001 and 2002. These datasets were obtained in spring 2004 from the Agency for Financial and Informatics (AFIP), the remnant o f the old payment bureau. Companies are submitting their year-end statements to this agency. Firms in the database cover a very wide range of sectors, industries, and sizes, while incorporating private, mass privatized and solely state-owned enterprises. The datasets are based on reported annual statements o f companies and not on actual financials. The dataset for the FBH i s much less complete then that for the RS, simply because balance sheets and income statements in the FBH are less detailed then those in the RS. Finally, not all o f the companies that existed in 2001 are in the 2002 database (because o f ownership change, closed business i s a significant drop insales revenues or assets, etc.). ANNEX 3.3: DATAFORTHE ANALYSIS OFDETERMINANTSOFENTERPRISE PERFORMANCE Table 2: Descriptive Statistics Std. Variable Observations Mean Dev. Min M a x Notes 1:Negative; 2: 0%; 3: [1-lo%]; 4: Gross Profit-to-Total [11-20%]; 5: [21-30%]; 6: [31-401%; Sales Ratio 164 3.21 1.15 1 7 7:[more than 40%] Sales2001mln 120 2.01 5.73 0.006 53.32 Million US$ Ownership 182 1.50 0.71 0 2 0: SOE; 1: Privatized; 2: De novo % o ftotal market share inthe main Market share 139 17.37 21.75 2.5 100 product line o fbusiness 0: no difficulties; 1:difficulties in Access to finance 160 0.48 0.50 0 1 accessing finance Sales on credit 175 12.41 23.70 0 100 % o ftotal annual sales % o ftotal annual sales to parent6 Vertical integration 182 26.51 38.76 0 100 company or affiliated subsidiary Horizontal integration 182 2.30 2.92 1 20 Suppliers arrears 182 1.36 4.20 0 30 % o ftotal annual sales Worker arrears 182 0.71 2.81 0 20 O/Oo ftotal annual sales Taxes arrears 182 1.02 3.21 0 20 % o ftotal annual sales Utilities arrears 182 0.71 2.87 0 20 % oftotal annual sales 136 Table 5:14 Estimated Determinants of Profitability: OLS Regressions Gross Gross Gross Gross Profit-to- Profit-to- Profit-to- Profit-to- Total Sales Total Sales Total Sales Total Sales (1) (2) (3) (4) Sales2001mln -0.023 -0.020 -0.022 -0.010 (-0.51) (-0.44) (-0.49) (-0.24) Privatized firm 0.425 0.426 0.404 0.578 (1.020) (1.040) (0.980) (1.430) New firm (2.27) * 0.863 * (2.38)** 0.906 0.868 (2.29) ** 0.846 (2.28) ** Market share 0.001 0.000 0.001 0.002 (0.160) (0.020) (0,100) (0.320) Access to finance 0.510 0.541 0.53 1 0.502 (1.94)* (2.09)** (2.04)** (1.99)** Sales on credit (2.86) ** 0.014 * (2.99) *** 0.015 0.014 (3.3 3) *** 0.016 (2.90) *** Vertical integration 0.006 0.006 0.006 0.007 (1.94)* (1.97)* (1.87)* (2.16)** Horizontalintegration -0.062 -0.061 -0.064 -0.081 (-1.44) (-1.42) (-1.4) (-1.89)* Worker arrears -0.012 (-0.34) Suppliers arrears -0.074 (-2.21)** Taxes arrears 0.010 (0.160) Utilities arrears 0.055 (0.900) Constant 2.254 2.178 2.240 2.299 (5.83)* ** (5.5 9)* ** (5.8 1)*** (6.14)* ** Observations 87 87 87 87 R-squared 0.28 0.29 0.28 0.33 Notes: Absolute value of t-statistics inparentheses. *significant at 10%;** significant at 5%; *** significant at 1%. 137 References Abemathy, F., J. Dunlop, J. Hammond and David Weil. 1999.A Stitch in Time. Lean Retailing and the Transformation of Manufacturing - Lessons from the Apparel and Textile Industries. Oxford University Press: New York andOxford. Adam, Antonis, Theodora Kosma andJimmy McHugh.2003. "Trade LiberalizationStrategies: What Could South Eastem Europe Leam From CEFTA andBFTA?" IMFWorking PaperNo. 031239. Balassa, Bela. 1965, "Trade Liberalization and `Revealed' Comparative Advantage," Manchester School 33, pp. 99-123. Balino, Tomas and Charles Enoch. 1997. "Currency BoardArrangements: Issues and Experiences", IMF Occasional Paper 151. Betcheman, Gordon, Karina Olivas and Amit Dar. 2004. "Impacts of Active Labor Market Programs: New Evidence from Evaluations with Particular Attention to Developing and Transition Countries." World Bank Social Protection Discussion PaperNo. 0402. BirksSinclair. 2002. "Labor Markets inBosniaandHerzegovina." BirksSinclair IBHI.2003. "Qualitative StudyNo 2. BHSkills Gaps andTrainingNeeds: Current Status and Policy Implications," DFID Project Report, Development o f Labour and Social Polices, Sarajevo. Brenton, Paul and Miriam Manchin. 2003 "Making EU Trade Agreements Work: The Role of Rulesof Origin." The WorldEconomy, Vol. 26, pp. 755-769. Broadman, Harry, James Anderson, Constantijn Claessens, Randi Ryterman, Stefka Slavova, Maria Vagliasindi, and Gallina Vincellete, 2004. "Building Market Institutions in South Eastem Europe: Comparative Prospects for Investment and Private Sector Development." The World Bank:Washington, DC Catic, Lejla. 2003. "Report on Education Finance inBH'. Manuscript. World Bank Europe and Central Asia HumanDevelopment Unit. D e Ferranti, David et al. 2003. Closing the Gap in Education and Technology, Washington D.C., World Bank, Latin American and Caribbean Studies Series. Djankov, Simeon and Peter Murreep. 2002. "Enterprise Resturcutimg in Transition: A quantitative Survey." Journal of Economci Literature, Vol. 40, No. 3, pp. 739-92. EuropeanBank for Reconstruction andDevelopment. 2002. Transition Report. EuropeanBank for Reconstruction andDevelopment. 2003. Transition Report. 138 European Bank for Reconstruction and Development and World Bank. 2002. The BEEPS I1 Interactive Dataset: Enterprise SurveyinTransition, 2002. http://info.worldbank.org/governance/beeps2002/ European Commission. 2002. "Highlights from the 2001 Survey: The European Observatory for SMEs." Report 2002/No. 1. European Commission. 2003. "Report from the Commission to the Council on the Preparadness o f Bosnia and Herzegovina to Negotiate a Stabilization and Association Agreement with the EuropeanUnion." Brussels European Commission. 2004. "European Economy: Western Balkans in Transition", Occasional PapersNo.5. Evans, Carolyn L., and James Harrigan. 2003. "Distance, Time, and Specialization," NBER Working PaperNo. 9729. Gereff, Gary. 1999. "International Trade and Industrial Upgrading in the Apparel Commodity Chain", Journal of International Economics, 48: 37-70. Gray, Cheryl, Joel Hellman and Randi Ryterman. 2004. Anticorruption in Transition 2 - Corruption in Enterprise-State Interactions in Europe and Central Asia 1999-2002. The World Bank. Heritage Foundation. 2003. Index of Economic FreedomDatabase. Hill,R.C. 1989. "Comparing Transnational Production Systems: The Automobile Industryinthe USA andJapan," InternationalJournal of Urban and Regional Research. 13(3): 462-480. InternationalMonetary Fund.2004a. Bosnia andHerzegovina: Article IV Staff Report, January. International Monetary Fund, 2004b. Bosnia and Herzegovina: Fourth Stand-By Review, February. Javorcik, Beata Smarzynska, and Bartolomiej Kaminski. 2004. "`EU Factor' and Slovakia's Globalization: FDIandtheir MultipleImpact," Unpublishedmanuscript, The World Bank. Jones, J. 1995. "Forces Behind Restructuring in US Apparel Retailing and Its Effect on the U S ApparelIndustry," Industv, Tradeand Technology Review. March, 23-27. Kaminski, Bartlomiej, and Beata Smarzynska. 2001. "Foreign Direct Investment and Integration into Global Production and Distribution Networks: The Case o f Poland," Post-Communist Economies, Vol. 13, No. 3. 139 Kaminski, Bartlomiej, and Beata Smarzynska. 2003. "Never too Late to Get Together Again: The Czech and Slovak Customs Union-A Stepping Stone for EUIntegration?" Policy Research Working Papers Series 2954, The World Bank. Kaminski, Bartolomiej andMichelle Riboud. 2000. "Foreign Investment and Restructuring: The Evidencefrom Hungary." World Bank Technical Paperno. 453. LevaEiC, Rosalind, 2003, "Financing schools to achieve efficiency and equity: a review of principles and practices and their application to Bosnia and Herzegovina". Manuscriptproduced inpreparationofthe EducationRestructuringProject. Mesinovic, Hans, Zlatko Hurtic and Mersida Suceska. 2003. "Country Study - Bosnia and Herzegovina." In Trade Policies and Institutions in the Countries of South Eastern Europe in the EUStabilization andAssociation Process. World Bank Report No. 24460. Messerlin, Patrick, and Sebastien Miroudot. 2003. "Trade Liberalization in South East Europe: From a Network of 23 FTAs to a Full Free Trade Area?" Groupe D'Economie Mondial, Sciences Po, September24. Nicholl, Peter. 2003. "Five years of the CBBH." Presentation for the CBBH Conference "The Monetary Policy Role of Currency Boards: History andPractice." April. OECD. 1998.Redefining Tertiary Education, Paris. OECD. 2001 Knowledge and Skills for Life: First Results from PISA 2000, Paris Schneider, Friederich, 2002: "Size and Measurement of Shadow Economies o f 110 Countries Around the World." Mimeo. Smarzynska, Beata. 2001. "Does Foreign Direct Investment Increase the Productivity of Domestic Finns? In Search o f Spillovers Through Backward Linkages." Policy Research Working Paper2923. The World Bank. The Sussex European Institute. 2003. "Study on the economic impact o f extending the pan- European system of cumulation o f origin to the Mediterranean partners' part of the Barcelona process." University of Sussex. UN/ECE. 2003. "Small andMediumSizedEnterprises inCountries inTransition." Vincelette, Gallina Andronova. 2004. "Challenges to the Bulgarian Monetary Policy on Its Way to United Europe," in Antoaneta Dimitrova (ed.) Driven to Change. Manchester University Press: Manchester and New York. pp.98-111. World Bank. 2000. `Trade Blocs' 140 World Bank. 2002a. "Bosnia and Herzegovina - From Aid Dependency to Fiscal Self-Reliance: A Public Expenditure and InstitutionalReview." October. World Bank. 2002b. "Bosnia and Herzegovina - Labor Market in Postwar Bosnia and Herzegovina: How to Encourage Businesses to Create Jobs and Increase Worker Mobility." November. World Bank. 2002c. Bosnia and Herzegovina - Administrative and Regulatory Cost Survey. Foreign Investment and Advisory Service (FIAS). World Bank. 2003a. Bosnia and Herzegovina - Poverty Assessment, Main report (Volume I). November. World Bank. 2003b. `Bosnia and Herzegovina - Beneficiary Assessment for Secondary and Higher Education Reform'. Manuscript available from Europe and Central Asia Human Development Unit. World Bank. 2003c."Trade Facilitation in a Development Context"- Trade Note No.12, December. World Bank. 2003d. Global Development Finance. World Bank. 2003e. "Reducing Administrative Barriers to Investment in Bosnia and Herzegovina: Progress Report of FIAS Project to Implement Monitoring Reforms." World Bank. 2003f. "The Qualityof the Business Environment: Early Warning System - Europe and Central Asia." World Bank. 2004a. "Doing Business in2004: UnderstandingRegulation." World Bank. 2004b. "Bosnia and Herzegovina - Banking Sector Review." 141 Report No. 29500-BA Bosnia and Herzegovina Country Economic Memorandum May 2005 Poverty Reduction and Economic Management Unit Europe and Central Asia Region Document of the World Bank IBRD 32760R 16° To Zagreb 18° To Osijek 19° To Zagreb To Zagreb C R O A T I A To Vinkovci Danube To Osijek To Glina Bosanska Novi Bosanska Gradiska Sava To Belgrade (Novi Grad) (Srp. Gradiska) Bosanski Brod 45° (Srp.Brod) Bosna B.Samac 45° Cazin Prijedor Derventa Modrica K A Brcko* Una P S Gradacac Bosanska S R Prnjavor Karlovac Bihac´ Krupa Banja oT Luka A Bijeljina Sanski Most Doboj L IK Gracanica Spionica B Kotor Varos Teslic´ Bosanski PVrbas U Petrovac Tesanj Tuzla Loznica Maglaj Lukavac Kljuc E R Skender Vakuf Zivinice Kljuc Zavidovici ´ (Ribnik) Banovici ´ Drvar Mrkonjic Zvornik F Grad E Sakovica ´ Jezero Jajce Drina CROATIA D E Travnik Kladanj Sipovo R Zenica Bratunac A Pucareve Vares Vitez T Kakanj Vlasenica I Olovo O Donji Vakuf Srebrenica N Busovaca Bugojno Knin O Glamoc Visoko 44° Fojnica 44° F Semizovak B Sokolac O To S Gornji Vakuf N To I Kresevo Sarajevo Belgrade A A To Zadar Livno N D H E R Z E G O V I N A Pale Rogatica Visegrad Uzice Prozor Duvno Sibenik Gorazde Jablanica Konjic Trnovo A K Foca S Neretva (Srbinje) P To Posusje Kalinovik R Prijepolje Split S Mostar A K I Nevesinje L B To Capljina Gacko Podgorica U Stolac P SERBIA AND 43° Ploce 43° E R MONTENEGRO Bileca ´ This map was produced by A d r i a t i c the Map Design Unit of The World Bank. The boundaries, Trebinje colors, denominations and S e a any other information shown Dubrovnik on this map do not imply, on the part of The World Bank To Group, any judgment on the BOSNIA AND HERZEGOVINA Podgorica legal status of any territory, or any endorsement or a c c e p t a n c e o f s u c h DAYTON AGREEMENT LINES boundaries. 19° REPUBLIKA SRPSKA FEDERATION OF BOSNIA AND HERZEGOVINA 30° 50° MUNICIPAL BOUNDARIES LUX. 10° 20° CZECH POLAND UKRAINE MAJOR ROADS GERMANY REP. RAILROADS FRANCE SLOVAK REP. RIVERS MOLDOVA SWITZ. AUSTRIA SELECTED CITIES BOSNIA AND HUNGARY HERZEGOVINA NATIONAL CAPITAL 42° SLOVENIA ROMANIA 42° CROATIA INTERNATIONAL BOUNDARIES SAN MARINO Adriatic BOSNIA AND HERZEGOVINA SERBIA Black AND Note: The municipality of Brcko is depicted as a condominium ITALY Sea MONTENEGRO BULGARIA Sea of both the Federation of Bosnia and Herzegovina and the Republika Srpska. Area of map MACEDONIA FYR ALBANIA 40° 40° 0 25 50 75 Tyrrhenian Sea GREECE Aegean 30° KILOMETERS Sea TURKEY 16° 17° 18° 10° 20° JULY 2004