Report No. 78505-ECA South East Europe Regular Economic Report | No.4 From Double-Dip Recession to Fragile Recovery June 18, 2013 Poverty Reduction and Economic Management Unit Europe and Central Asia The World Bank This report is produced twice a year by staff economists at the World Bank Europe and Central Asia Region Poverty Reduction and Economic Management Department (ECA PREM). This and previous reports may be found at the www.worldbank.org/eca/seerer website. The team of authors is led by Abebe Adugna and Željko Bogetić: the following team members have thematic assignments: Dilek Aykut (global developments and outlook); Simon Davies (fiscal and debt); Agim Demukaj (external sector, Kosovo); Sandra Hlivnjak (Bosnia and Herzegovina); Caterina Ruggeri Laderchi (special topic note on jobs); Erjon Luci (SEE6 countries database manager, Albania); Anil Onal (research assistance), Sanja Madzarević-Sujster (inflation developments, Montenegro); Lazar Sestović (real economy, Serbia); Birgit Hansl and Bojan Shimbov (labor trends and financial sector, FYR Macedonia). Maria Andreina Clower and Mismake Galatis provided team assistance. Dissemination of the report and external and media relations is managed by an EXT team Lundrim Aliu, Denis Boskovski, Ana Gjokutaj, Jasmina Hadzić, Andrew Kircher, Vesna Kostić, Mirjana Popovć, John Mackedon, Krystin Schrader, and Dragana Varezić. The team benefitted from guidance and advice from Satu Kähkönen (Sector Manager, ECA PREM2) and Yvonne Tsikata (Sector Director, ECA PREM). In this report, “South East Europe� refers to the six countries of the Western Balkans (SEE6): Albania, Bosnia and Herzegovina, Kosovo, FYR Macedonia, Montenegro, and Serbia. Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. . Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978- 750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Table of Contents CHAPTER 1: RECENT MACROECONOMIC DEVELOPMENTS ................................................................. 1 GLOBAL CONTEXT: GROWTH PICKING UP, BUT EUROZONE STILL IN RECESSION ........................................................1 GROWTH IN SEE6: OUT OF A DOUBLE-DIP RECESSION?............................................................................................3 INFLATION DEVELOPMENTS––FOOD PRICE PRESSURES .............................................................................................4 POVERTY AND INEQUALITY––NOT IMPROVING .........................................................................................................6 TRADE AND EXTERNAL DEBT – SIGNS OF IMPROVEMENT ..........................................................................................8 FISCAL POLICY – DIVERGENT PRESSURES ...............................................................................................................13 FINANCIAL SECTOR––SLOW- CREDIT GROWTH .......................................................................................................16 LABOR TRENDS––HIGH UNEMPLOYMENT ACROSS THE REGION ..............................................................................21 SEE6 OUTLOOK––FRAGILE RECOVERY...................................................................................................................24 CHAPTER 2. SPECIAL TOPIC: REKINDLING JOB CREATION IN SEE6 ................................................ 27 THE 2000S: A DISAPPOINTING DECADE FOR JOB CREATION ....................................................................................27 AN AGENDA TO REKINDLE JOB CREATION ..............................................................................................................30 SUPPORTING ENTREPRENEURSHIP AND REMOVING BARRIERS TO JOB CREATION ....................................................31 MORE INCLUSIVE LABOR MARKETS, LOW- PARTICIPATION, AND EMPLOYABILITY ..................................................33 Addressing Emerging Skill Gaps ........................................................................................................................33 Managing Internal and International Mobility ...................................................................................................36 Addressing Disincentives to Employment ...........................................................................................................38 CONCLUSIONS ..........................................................................................................................................................40 REFERENCES ................................................................................................................................................... 41 ANNEX: KEY INDICATORS ........................................................................................................................... 43 Figures Figure 1: Global Industrial Production and Trade: Strong Trade, Output Bottoming Out ........................... 1 Figure 2: Global Financial Markets—Continued Calm ................................................................................ 2 Figure 3: SEE6––Industrial Output, Quarterly Growth in 2012, q-o-q (Percent) ......................................... 3 Figure 4: Inflation in EU15, EU11, and SEE6 Countries, CPI (yearend 2011=100).................................... 4 Figure 5: Inflation in SEE6 Countries, CPI, ................................................................................................. 4 Figure 6: Food Price Inflation in SEE6 Countries, y-o-y (Percent) .............................................................. 5 Figure 7: Energy Price Inflation in SEE6 Countries, y-o-y (Percent)........................................................... 5 Figure 8: SEE6 Real Exchange Rates, CPI Deflated, (Aug. 2008=100) ...................................................... 6 Figure 9: SEE6 Key Policy Interest Rates (Percent) ..................................................................................... 6 Figure 10: Poverty Estimates for SEE6 countries (Percent) ......................................................................... 7 Figure 11: Growth and Redistribution--Shapley Decomposition of Poverty Changes (US$ 5 a day) between 2009 and 2010 ................................................................................................................................ 7 Figure 12: SEE6 CAD and Trade Balance.................................................................................................... 9 Figure 13: CAD by Country (Percent of GDP)............................................................................................. 9 Figure 14: Exports in 2012 (Percent of GDP)............................................................................................... 9 Figure 15: Exports Growth (Percent) .......................................................................................................... 10 Figure 16: Import Growth (Percent) ........................................................................................................... 10 Figure 17: Export and Import Growth, y-o-y (Percent) .............................................................................. 10 Figure 18: Worker Remittances 2010–2012 (Percent of GDP) .................................................................. 11 Figure 19: Net FDI 2009-12 (Percent of GDP)........................................................................................... 11 Figure 20: Net FDI by Country(Percent of GDP) ....................................................................................... 11 Figure 21: Average SEE6 External Debt (Percent of GDP) ....................................................................... 12 Figure 22: Total International Bonds Outstanding, Selected SEE6 Countries (US$ million) .................... 12 Figure 23: Total Public and Private External Debt 2012 (% of GDP) ........................................................ 12 Figure 24: Fiscal Deficits (Percent of GDP) ............................................................................................... 13 Figure 25: Changes in Revenue and Spending, 2011-12 (Percent of GDP) .............................................. 13 Figure 26: Revenue and Expenditures, 2012 and 2013 (Projected) (Percent of GDP) ............................... 14 Figure 27: Fiscal Deficit, 2012 versus 2013 (Projected) (Percent of GDP) ................................................ 14 Figure 28: General Government Debta........................................................................................................ 14 Figure 29: State Guarantees (Percent of GDP) ........................................................................................... 14 Figure 30: Local and Foreign Currency Debt due in 2013 (Percent) .......................................................... 15 Figure 31: Local Currency Debt due in 2013 (Percent of GDP)a ............................................................... 15 Figure 32: Real Credit Growth Rates (Percent Change) ............................................................................. 16 Figure 33: Real GDP Growth, Real Credit Growth, and Foreign Financing, 2011-12 percent) ................. 17 Figure 34: Interest Rates on Loans and Deposits, 2011–12 ........................................................................ 17 Figure 35: Funding Costs for SEE6 Countries............................................................................................ 17 Figure 36: Loan-to-Deposit Ratios ............................................................................................................. 18 Figure 37: Nonperforming Loans (Percent of Total Loans) ....................................................................... 19 Figure 38: Return on Equity ....................................................................................................................... 20 Figure 39: Capital Adequacy Ratios ........................................................................................................... 20 Figure 40: Unemployment Rates, 2012 ...................................................................................................... 21 Figure 41: Labor Participation and Unemployment.................................................................................... 21 Figure 42: Labor Force Participation in SEE6 and EU11 ........................................................................... 22 Figure 43: Changes in Labor Force by Gender, Q4 2008 and Q4 2012 (Percentage points)...................... 22 Figure 44: Unemployment by Age Group .................................................................................................. 23 Figure 45: Change in Unemployment by Age Group, 2008 to 2012 (Percentage points) .......................... 23 Figure 46: Share of Public Sector Employment in SEE6 ........................................................................... 23 Figure 47: SEE6––Real GDP Growth ........................................................................................................ 24 Figure 48: Change in Employment Associated with 1 Percent Change in Real GDP, 1995-2010 (Percentage Points) ..................................................................................................................................... 27 Figure 49: EBRD Transition Index, 2000 and 2012 ................................................................................... 28 Figure 50: Changes in Transition Index Components, 2000 to 2012.......................................................... 28 Figure 51: Net Job Creation, Selected SEE6 and EU11 Countries, 2002–09 ............................................. 30 Figure 52: Steps taken towards Business Startup (1990 –2010) and Actual Business Startup (Percent) ... 32 Figure 53: Latent versus Actual Entrepreneurship in SEE6, 2010 (Percent of working age population) ... 32 Figure 54: Skills and Education NOT an Obstacle, 2005 and 2008 ........................................................... 34 Figure 55: Skills Content by Job-holder Cohort, FYR Macedonia and Lithuania ...................................... 35 Figure 56: Functionally Illiterate 15-year-olds, SEE6 (Percent)................................................................. 36 Figure 57: Internal Mobility and Intentions to Migrate (Percent) ............................................................... 38 Tables Table 1: Global Growth Assumptions – Real GDP Growth (in percent)...................................................... 2 Table 2: SEE6––Real GDP Growth .............................................................................................................. 3 Table 3: SEE6: Sovereign Credit Ratingsa.................................................................................................. 15 Table 4: SEE6––Real GDP Growth ............................................................................................................ 24 BOX Box 1: Shared Prosperity in the SEE6 .......................................................................................................... 8 Box 2: Deleveraging in SEE6 Countries..................................................................................................... 18 Box 3: Reforms and Job Creation ............................................................................................................... 29 Box 4: SEE6 Demographic Trends and the Jobs Agenda ........................................................................... 31 Box 5: Latent Entrepreneurship in Serbia ................................................................................................... 33 Box 6: Tertiary Student Mobility ................................................................................................................ 37 SUMMARY After the double-dip recession, as a group the six South East European countries (SEE6)––Albania, Bosnia and Herzegovina, Kosovo, FYR Macedonia, Montenegro, and Serbia––are now making a fragile recovery. Last year the recession in the Eurozone had adverse impact on external demand and foreign direct investment (FDI) in SEE6 and the severe winter and a summer drought crippled agriculture and affected trade, energy, and economic activity overall. Now, output is beginning to bounce back. Exports are recovering in Serbia, the largest SEE6 economy; weather conditions are much improved; and in some countries dynamism is revving up in electricity, tourism, and related sectors. However, the recovery in SEE6 is still tentative. In some countries nonperforming loans, sluggish credit recovery, continued deleveraging, and fiscal consolidation are exerting a drag—and recovery in SEE6 is unlikely to accelerate as long as the Eurozone remains in recession. Although global economic and financial conditions have continued to improve, the Eurozone is expected to be in recession in 2013 (at –0.6 percent growth). Global GDP is now projected to expand by 2.2 percent in 2013, 3.0 percent in 2014, and 3.3 percent in 2015. While growth in high-income countries will be a feeble 1.2 percent in 2013 (slowly rising to 2.0 in 2014 and 2.3 percent in 2015), growth in low- and middle-income countries will be 5.1 percent in 2013, accelerating slowly to 5.6 percent in 2014, and 5.7 percent in 2015. Gross capital flows to low- and middle-income countries are 60 percent higher than what they were a year ago—pointing to an end to the most serious effects of Eurozone deleveraging on those countries, including the SEE6 economies. Within this context, the SEE6 region is projected to grow 1.7 percent in 2013, signaling the end of the 2012 double-dip recession. Even though growth will in general be fragile, it will be on the upswing in all six countries. Kosovo again is expected to have the highest growth (3.1 percent), thanks to major public investments and a significant inflow of remittances. Next in the growth line is Serbia, at a projected 2 percent, in part reflecting the base effect from last year’s recession. Since Serbia accounts for 45 percent of the region’s economy, growth there is crucial to the region’s performance. Serbia is expected to benefit from increased FDI, solid performance from FIAT, and a return to normal agricultural output, which dropped nearly 20 percent in 2012; and as investors become more confident based on possible opening of EU accession negotiations later in the year, more FDI can be expected. Albania is projected to grow at about 1.8 percent, as it did last year, supported by a steady export performance. FYR Macedonia’s economic growth is expected to be moderate and will come mostly from FDI exports and public investments. Modest growth is expected in Montenegro partly because electricity and agriculture are recovering but mainly because tourism is surging ahead. In Bosnia and Herzegovina growth is likely to be tepid this year; the projection is just 0.5 percent. Unfortunately, numerous issues related to the BIH business environment will continue to constrain FDI flows there, as well as the prospects for expansion of domestic businesses. Against the backdrop of this tentative and fragile recovery, SEE6 countries should, as argued in the last report, intensify their efforts to reform structural areas. Fiscal consolidation efforts should become easier now that the output and revenue outlook is improving. The investment climate needs to be improved substantially, especially in the main areas of weaknesses: construction permits and licenses, barriers to entrepreneurship, and skills and infrastructure. Its neighbors could learn from FYR Macedonia, which continues to have the most favorable investment climate in the region as measured by the Doing Business indicators. i One of the main worries in this nascent recovery is that SEE6 economies are plagued by high unemployment, especially youth unemployment, and they are not creating jobs fast enough to absorb new entrants into the labor force. In fact, the jobs situation is worse than the dismal unemployment figures suggest because so many leave the region to work elsewhere. In part, this is the legacy of periods when some SEE6 countries suffered significantly from regional dislocations that delayed reforms. Emigration continues as the current environment for doing business exacerbates the difficult labor market conditions. What SEE6 countries now need to do is to sustain the fragile recovery and push for job creation. This will require aggressive job-oriented policies. Recent World Bank research on jobs in low- and middle-income countries in Europe and Central Asia suggests that the policy agenda for job creation would best be targeted to four areas: fostering entrepreneurship, improving skills, managing internal and international mobility, and reducing institutional disincentives to job creation. Accelerating reforms in these areas is imperative if there is to be hope for more, better, and more diverse jobs in SEE6 countries. ii CHAPTER 1: RECENT MACROECONOMIC DEVELOPMENTS GLOBAL CONTEXT: GROWTH PICKING UP, BUT EUROZONE STILL IN RECESSION Global growth has picked up in early 2013 Despite pallid growth in high-income countries in the fourth quarter of 2012 (especially in the Eurozone), the output of low- and middle-income countries firmed up and continued to strengthen in the first quarter of 2013 (Figure 1). While high-income growth strengthened in the first quarter of 2013, growth in low- and middle-income countries eased in some regions while mostly remaining solid. Available industrial production data confirm this mixed picture, with activity rates slowing in East Asia & Pacific (to 8.6 percent), firming in Europe & Central Asia (at 2.4 percent), returning to a positive territory in Latin America & the Caribbean (0.5 percent), and easing somewhat in South Asia but still at a robust 7.7 percent.Trade also rebounded, with low- and middle income-country imports expanding at a 21 percent annualized pace through January 2013, supporting the acceleration in high-income exports during the same period at a 7.3 percent annualized pace. Most recently, business surveys see growth slowing in the second quarter, partly because of the lagged effects of fiscal tightening in the United States and a regression to more sustainable growth rates in East Asia and the Pacific. Figure 1: Global Industrial Production and Trade: Strong Trade, Output Bottoming Out (volume percent changes, 3m/3m saar) High-income Imports 40 Developing Imports % 30 Developing Industrial production High Income Euro-Area Industrial production 20 10 0 -10 -20 -30 2011M01 2011M04 2011M07 2011M10 2012M01 2012M04 2012M07 2012M10 2013M01 Source: World Bank’s Global Prospects Group. Global GDP is projected to expand by 2.2 percent in 2013 and thereafter head up gradually, to 3.0 percent in 2014 and 3.3 percent in 2015 (Table 1). Growth in high-income countries will remain a flabby 1.2 percent in 2013 but is projected to reach 2.0 percent in 2014 and 2.3 percent in 2015. Improved financial and generally loose monetary conditions are expected to stimulate activity in underperforming low- and middle-income countries. This should prompt a gradual acceleration of their growth to 5.1 percent this year, 5.6 percent in 2014, and 5.7 percent in 2015— roughly in line with their potential. The Eurozone, however, is projected to stay in recession in 2013 (with–0.6 percent growth). 1 Table 1: Global Growth Assumptions – Real GDP Growth (in percent) 2008 2009 2010 2011 2012e 2013f 2014f 2015f World 1.4 -2.2 3.9 2.8 2.3 2.2 3.0 3.3 High income 0.1 -3.5 2.8 1.7 1.3 1.2 2.0 2.3 Developing countries 5.8 1.9 7.3 5.9 4.9 5.1 5.6 5.7 Memo item Euro Area 0.3 -4.3 1.9 1.5 -0.5 -0.6 0.9 1.5 Source: World Bank Global Economic Prospects Group staff estimates (World Bank 2013c). Better financial conditions globally and robust flows to low- and middle-income economies The substantial improvement in global financial conditions since summer 2012 has persisted, with yields on Eurozone and low- and middle-income country sovereign debt generally constant since January (Figure 2). Gross capital flows to low- and middle-income countries have firmed up as equity and bank flows during the first five months of the year rose by 70 percent—confirming an end to the most serious effects of Eurozone deleveraging on low- and middle- income country finances. Spreads on the bonds of low- and middle-income countries have declined since June 2012. That said, their stock markets have been weak since the beginning of the year with the benchmark MSCI index marking a 3.3 decline year-to-date due to weak corporate earnings and country specific factors. Though the uncertainty generated by the Cyprus rescue effort was a stark reminder of how fragile confidence can be, it seems to have had little impact on conditions in the rest of the Eurozone: yields on high-spread Eurozone sovereign debt have risen by less than 50 bps and rates on banking-sector CDSs are up only 20 bps since February. Despite these improvements, however, the Eurozone remains in recession and its domestic demand is weak. But the hope is that these financial market improvements could translate into a gradual recovery toward year-end. Figure 2: Global Financial Markets—Continued Calm 1,800 Credit Default Swap Rates, basis points 1,600 1,400 SPAIN IRELAND 1,200 ITALY PORTUGAL 1,000 800 600 400 200 0 January-10 October-10 July-11 April-12 January-13 Source: World Bank’s Global Prospects Group. 2 GROWTH IN SEE6: OUT OF A DOUBLE-DIP RECESSION? As we projected in a previous report, the SEE6 fell into recession in 2012. Overall, SEE6 growth turned out to be as projected in the last report (–0.6 percent). Only in Albania and Kosovo did growth stay positive; the other economies experienced recession (Table 2). Albania and Kosovo were again the best performers, contributing a positive 0.4 percentage points to the region’s growth in 2012, but this was more than offset by the negative performance of Serbia, Bosnia and Herzegovina, FYR Macedonia and Montenegro. Table 2: SEE6––Real GDP Growth Actual Estimated Estimated 2012 H1 2012 H2 2012 ALB 1.6 0.9 2.2 BIH -0.7 -0.2 -1.1 KOS 2.3 3.6 1.1 MKD -0.3 -1.1 0.4 MNE -0.6 -0.9 -0.2 SRB -1.7 -1.5 -1.9 weighted av. -0.6 -0.5 -0.7 Source: World Bank staff. Throughout 2012 the situation seemed to be most worrying in Bosnia and Herzegovina, Montenegro, and Serbia. Not only were these countries in recession, but in some countries, the economic decline accelerated in the second half of the year. In Bosnia and Herzegovina, high frequency data indicate that the economy fell by 1.1 percent in the second half, compared to just 0.2 percent in the first half. 1 In Serbia, the decline in the second half was driven by the impact of severe drought on agriculture and the food processing industry and to a lesser extent confidence effects related to political uncertainty in an election year. In Montenegro, the main causes of the decline in the first half were adverse weather andlow electricity generation; recovery in the second half could not make up for the losses early in the year as well as the problems in metal production (in the second half). Figure 3: SEE6––Industrial Output, Quarterly Growth in 2012, q-o-q (Percent) 15.0 10.0 Q1 Q2 Q3 Q4 5.0 0.0 -5.0 ALB BIH MKD MNE SRB -10.0 -15.0 -20.0 -25.0 Source: World Bank staff. * Data for Kosovo are not available. 1 The real GDP growth for BiH in 2012 is still an estimate. 3 Industrial output for the region as a whole fell by 3.6 percent in 2012 (Figure 3).2 Albania is the only country where industrial output grew. In the other four countries3 industrial output dropped on average by 5.3 percent compared to 2011. Industrial output declined most (close to 7 percent) in FYR Macedonia and Montenegro; Serbia had the smallest decrease, largely due to a good last quarter. However, early data for 2013 suggest a fragile recovery across the region. This reflects the reversal of one-time factors such as the bad winter and the drought, which adversely affected agriculture, construction and energy production in many of the countries. The recovery is also driven by a rise in industrial exports, especially from the large FIAT factory in Serbia, the largest economy in the region. Montenegro and Bosnia and Herzegovina are also expected to make a modest recovery in 2013, while Albania and FYR Macedonia are projected to grow at a similar or higher rate than in 2012. Albania’s growth is supported by steady export performance. In FYR Macedonia, near-term growth prospects appear to mainly hinge on the implementation of public and foreign direct investment plans. The continuing recession in the Eurozone, however, will keep a lid on this fragile recovery. INFLATION DEVELOPMENTS––FOOD PRICE PRESSURES Inflation dynamics in SEE6 were mixed throughout 2012, reflecting differences in demand, administered prices, and food price conditions (Figure 5). Inflation rates ranged from 2 percent in Albania to over 7 percent in Serbia. In some countries (Bosnia and Herzegovina, FYR Macedonia, Montenegro), adjustments in energy prices contributed to inflation, while in others (Serbia), food prices were the driver. But overall, negative output gaps and high unemployment kept price pressures down in late 2012 and early 2013. Figure 4: Inflation in EU15, EU11, and SEE6 Figure 5: Inflation in SEE6 Countries, CPI, Countries, CPI (yearend 2011=100) y-o-y Source: Eurostat, national statistical offices, and Source: Eurostat, national statistical offices, and World Bank staff calculations. World Bank staff calculations. 2 Weighted average, where the weight is the share of a country’s GDP in total regional GDP. 3 No data are available for Kosovo. 4 Nevertheless, inflation in SEE6 continues to be much higher than in EU154 and EU115 (Figure 4). At 6.7 percent year-on-year (y-o-y) in Q1 2013, SEE6 inflation compared negatively to the 2.1 percent in the EU15 or the 2.3 percent in the EU11. This reflects a correction of administered prices and taxes in some SEE6 countries that are struggling to consolidate their public finances and continued food price pressures. Food prices are still a major driver of headline inflation in SEE6 but energy price pressures are declining (Figures 6 and 7). In Q1 2013 food price inflation accelerated to 9.3 percent y-o-y, higher than the peak of 9.2 percent in Q4 2012. With OECD consumption still in retreat, global demand for oil remains subdued, and the energy price pressures are easing in virtually all SEE6 countries. Figure 6: Food Price Inflation in SEE6 Figure 7: Energy Price Inflation in SEE6 Countries, y-o-y (Percent) Countries, y-o-y (Percent) Source: Eurostat, national statistical offices, and Source: Eurostat, national statistical offices, and World Bank staff calculations. World Bank staff calculations. Most SEE6 currencies have been generally stable during the past year, despite persistent concerns over the Eurozone sovereign debt crisis. Still, most SEE6 countries saw their real exchange rates appreciate in 2012, causing declines in price competitiveness against trading partners. From January 2012 the real exchange rates for Albania, FYR Macedonia, and Serbia appreciated in real terms (Figure 8). Only in Bosnia and Herzegovina did the convertible mark (KM) see a slight real depreciation by December 2012, implying greater price competitiveness with major trading partners. The key policy interest rates have remained broadly stable (Figure 9). 4 EU15 consists of Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden, and the United Kingdom. 5 EU11 consists of Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovak Republic, and Slovenia. 5 Figure 8: SEE6 Real Exchange Rates, CPI Figure 9: SEE6 Key Policy Interest Rates Deflated, (Aug. 2008=100) (Percent) Source: National banks and World Bank staff Source: National banks and World Bank staff calculations. A lower RER indicates appreciation. calculations. Note: For Montenegro, the liquidity loans rate. POVERTY AND INEQUALITY––NOT IMPROVING An estimated 33 percent of people in SEE6 live in poverty—and 8 percent live in extreme poverty. The recession has led to a protracted increase in moderate poverty —6defined as the share of individuals with consumption levels between US$2.5 and US$5 a day—from 22 percent to 26 percent. In contrast, by 2012 moderate poverty was declining in the EU11. Simple projections for 2011–15 suggest that while in general poverty will decrease slowly in both the EU11 and SEE6 countries, extreme poverty will decline to 5 percent in the SEE6 and to 1 percent in the EU11. SEE6 poverty rates at the national level are much higher than in the EU11. Figure 10 shows poverty rates for the SEE6 countries according to a methodology that allows for international comparisons.7 In all but Montenegro poverty incidence is higher in rural areas, and rural poverty is deeper than in urban areas. Except for Montenegro and FYR Macedonia, where income distribution is more balanced, most of the poor live in rural areas. Differences in poverty rates between men and women are negligible. 6 Unfortunately, the evidence does not allow for a current in-depth assessment of how the double-dip recession in the SEE6 affected poverty. Despite continued investment in improving statistics, household survey data are processed and made available with a lag. In Bosnia and Herzegovina and Albania the household surveys used to compute poverty rates (the Household Budget Survey in the former and the LSMS in the latter) are not collected every year; the latest figures for Bosnia and Herzegovina are from 2007 and for Albania from 2008. For the other countries poverty figures are available for 2010 (2011 was recently released for Montenegro). Based on those figures and the ECAPOV methodology, which is designed to obtain internationally comparable estimates, an estimated 33.2 percent of people in SEE6 (more than 6 million) live in poverty, and 7.7 percent (more than 1.4 million) in extreme poverty. Poverty incidence in the SEE6 is considerably higher than the average for Europe and Central Asia (ECA) middle-income countries (MICs), where poverty was 19 percent, and for the EU10, where it was 12 percent, with extreme poverty in both below 2 percent. 7 These estimates are based on standardized datasets of national household surveys kept by the World Bank (the ECAPOV database) and a consumption aggregate constructed consistently across countries. The poverty lines used are of 2.5 USD per capita per day in 2005 purchasing power parity (PPP) for extreme poverty and of 5 USD per capita per day PPP for overall poverty. These estimates differ from national poverty line estimates and are not intended to replace them. They are only intended to be used for international comparisons. 6 The recession has been accompanied by an increase in inequality in SEE6 countries. 8 While inequality in the region declined from 2005 to 2008 (from an estimated 38.9 percent to 36.6 percent), it appears to have been going up since 2008 (reaching an estimated 37.7 percent). These trends appear to be different from those seen in the EU11, and underlying them were different country dynamics. For example, in Kosovo inequality fell both before and after the crisis. In Serbia, inequality declined before the crisis, then increased somewhat after it. Figure 10: Poverty Estimates for SEE6 countries (Percent) 100.0 80.0 60.0 40.0 32.2 20.0 0.0 7.7 KOS ALB MKD SRB MNE BIH Extreme Poverty Moderate Poverty Extreme Poverty SEE6 Poverty SEE6 Source: World Bank staff estimates based on the ECAPOV dataset. Data are for the last available year: Albania 2008, Bosnia and Herzegovina, 2007, Kosovo 2010, FYR Macedonia 2010, Montenegro 2011, and Serbia 2010. Notes: Extreme poverty is defined as living on less than US$2.5 a day, poverty as living on less than US$5 a day. Poverty in BIH is probably significantly underestimated because of a problem with purchasing power parity estimates for that country. Worsening inequality pushed up Figure 11: Growth and Redistribution--Shapley poverty in the region because poorer Decomposition of Poverty Changes (US$ 5 a day) groups were more affected by the between 2009 and 2010 recession than richer ones. The dynamics were different in different 6 countries (Figure 11). For example, 4 between 2009 and 2010 poverty in Serbia 1.92 0.04 rose, and while the lack of growth 2 3.18 2.30 0.47 explains most of the increase, the rise in 0 inequality was also an important -2.61 contributor. In contrast, in Montenegro, -2 which has the highest per capita income among the SEE6, inequality did not have -4 a statistically significant impact on Serbia Montenegro Kosovo poverty trends. In Kosovo, growth Growth Redistribution accounted for the substantial reduction in poverty, even if the dynamics of Source: World Bank staff estimates based on ECAPOV dataset. inequality partially eroded the poverty- reducing impact of growth. 8 Inequality is defined as the Gini index for the whole region (considering all countries as one country). 7 Box 1: Shared Prosperity in the SEE6 The World Bank has adopted two new targets to make explicit its goal of “A world free of poverty.� The two targets are: ending extreme poverty by 2030 and promoting shared prosperity. Sustainability is an overarching priority in the pursuit of these goals: achieving them requires promoting environmental, social, and fiscal sustainability. The emphasis on reducing extreme poverty is in line with the commitment to halve extreme poverty enshrined in the first target of the Millennium Development Goals (MDGs), which were adopted globally in 2000. Shared prosperity focuses explicitly on the incomes of the poorer members of the population, and promoting shared prosperity is defined as maximizing income growth for the bottom 40 percent of the population. In the SEE6 countries, the bottom 40 percent is typically associated with living in rural areas and living in a household whose head is young, male, with little education (at most primary education) and is unemployed, retired, or a student. Being self-employed, rather than being a wage-earner, increases the probability of belonging to the bottom 40 percent in Bosnia and Herzegovina and Montenegro, but lowers the probability in Albania, FYR Macedonia, Kosovo, and Serbia. The shared proposerity goal is an explicit recognition that while growth is necessary for development, unless the benefits of that growth are shared widely and particularly reach the poorer members of the population, it cannot help to improve their living conditions. TRADE AND EXTERNAL DEBT – SIGNS OF IMPROVEMENT Because of the close economic linkage, the SEE6 external sector is highly correlated with developments in the EU. In 2012, the SEE6 experienced a deterioration in trade, current account deficits (CADs), FDI, and transfers. Albania outperformed the rest of the region, partly because of higher oil and mineral exports and export reorientation to new markets, particularly to China and Turkey. Serbian exports started picking up in the second half of 2012 mainly due to substantial FDI in the previous year. In the first quarter of 2013, exports, driven by FDI and improved EU economic performance, have recovered noticeably, bringing hope for a better external position in 2013. Current Account Deficits and Trade Balances While both current account deficits (CADs) and trade balances (TBs) deteriorated in 2012, they reversed direction in the first quarter of 2013. The decline in EU demand for SEE6 commodities led to a slide across the region that began in 2011 and continued in 2012 (Figures 12 and 13). In Serbia (particularly in the first half), Bosnia and Herzegovina, and Montenegro, exports plummeted while imports grew only moderately. In Kosovo, the deteriorating CAD and TB reflected high imports for construction of the Kosovo-Albania highway, a boom in private construction, and a steep decline in exports. Albania was the exception—both its CAD and TB improved in 2012 as a result of exports growth and weakening imports. 8 Figure 12: SEE6 CAD and Trade Balance Figure 13: CAD by Country (Percent of GDP) (Percent of GDP) CAD Trade Balance KOS MNE ALB SRB SEE6 BiH MKD 2009 2010 2011 2012 0 0 -5 -5 -10 -10 -9.3 -8.5 -10.0 -15 -10.8 -15 -20 -20 -25 -25 -24.0 -24.4 -23.3 -30 -30 -29.1 -35 -35 2009 2010 2011 2012 Source: Central banks, IMF WEO, and World Bank staff Source: SEE6 central banks. calculations. Exports and Imports Trade with the EU is central to SEE6 export Figure 14: Exports in 2012 (Percent of GDP) performance (Figure 14). Even with the economic problems caused by the Eurozone crisis and lower 60 Other SEE6 demand, the EU still accounted for 55 percent of all OtherEU Germany 50 SEE6 exports in 2012, though down from 58.8 9 Italy percent in 2011. Italy and Germany continued to be 40 the main SEE6 trade partners, accounting for 26.6 11 13 5 2 percent (down from 28.7 percent in 2011). 30 9 5 9 8 Although the region’s exports to EU as a whole 12 20 16 6 5 fell, Serbia’s trade with the EU went up. Similarly, 13 17 10 5 Albanian exports to the EU, mainly oil and 10 15 17 8 9 4 4 minerals, have also grown. At the other extreme 3 2 5 4 0 1 Kosovo’s exports to the EU have plunged, mainly MKD MNE SEE6 SRB ALB BiH KOS because of the drop in the prices of base metals, Kosovo’s main export commodity. Source: SEE6 central banks. Intraregional trade has grown in importance. That is true especially for Kosovo, Montenegro, and Serbia. The share in GDP of intraregional exports increased from 24.9 percent in 2011 to 27.0 percent in 2012 as exports to other non-EU countries fell. The growing intraregional trend was most noticeable in Albania, though starting from a small base. Expansion of exports to SEE6 countries enabled Kosovo to mitigate the steep decline in its exports to the EU when prices of metals dropped. 9 Figure 15: Exports Growth (Percent) Figure 16: Import Growth (Percent) SEE6 EU11 EU15 SEE6 EU11 EU15 30.0 30.00 20.0 20.00 10.0 10.00 0.0 0.00 2009 2010 2011 2012 Q1:2013 -10.0 2009 2010 2011 2012 Q1:2013 -10.00 -20.0 -20.00 -30.0 Source: SEE6 central banks and Eurostat. Source: SEE6 central banks and Eurostat. Note: 2013 shows exports of goods only; previous Note: Q1 2013 shows imports of goods only; full years years show goods and services. show goods and services. Though weak Eurozone demand and bad weather took a toll on SEE6 exports in 2012, things are looking up for 2013 (Figure 15). In 2012 SEE6 exports declined by 0.7 percent—after having increased by 14.0 percent in 2011—against EU11 export growth of 4.9 percent. The main cause of falling SEE6 exports was the worsening economic climate in EU, which lowered demand (Figure 15). Exports began to improve slowly in the second half of 2012 (y-o-y) as Serbian FDI-related trade smoothed out the drop in the first half of the year. As of the first quarter of 2013, there is a ray of hope that exports may be recovering across the region in 2013. SEE6 imports were flat in 2012 and again dropped in Q1 of 2013 (Figure 16). Terms of trade movements also likely play a large role in explaining the trends in import growth across the SEE6. Prices for imports of energy and other commodities fell sharply in 2009, rose considerably in 2010 and 2011, but flattened in 2012 and have declined in 2013. In 2012, Serbia’s imports grew while the imports of other countries fell. Imports in the second half of the year fell faster than in the first half in most countries as both domestic demand and industrial production continued to slide downward (Albania, FYR Macedonia, Bosnia and Herzegovina and Kosovo), as again happened in the first quarter of 2013 (Figures 16 and 17) due mainly to weak domestic demand and economic activity. Figure 17: Export and Import Growth, y-o-y (Percent) 2010 2011 2012 Q1: 2013 40 30 20 10 0 -10 -20 Exp Imp Exp Imp Exp Imp Exp Imp Exp Imp Exp Imp ALB BiH KOS MKD MNE SRB Source: SEE6 central banks. Note: In Q1 2013 bars show exports and imports of goods only, while for previous years they show both goods and services. Kosovo exports for September 2012 to January 2013 do not include exports of electricity, so the actual decline in 2012 was less pronounced. 10 Remittances and FDI Remittances have been relatively Figure 18: Worker Remittances 2010–2012 (Percent of GDP) resilient to the Eurozone crisis, though they have gone down 2010 2011 2012 somewhat in the past two years. While Serbia’s remittances sank 20.0 from 9.5 percent in 2009 to 6.8 15.0 percent in 2011 and remained flat in 2012, those of other SEE6 10.0 countries did not show a dramatic difference from 2009 (Figure 18). 5.0 0.0 FDI––which is important to KOS BIH ALB SEE6 SRB MNE MKD EU11 financing, investment, exports, and growth in SEE6––fell off Source: SEE6 central banks. noticeably in 2012, by 45.6 Albania and Bosnia and Herzegovina define remittances as including percent (2.6 percentage compensation of employees; Kosovo, Serbia, Macedonia and Montenegro points)—but there is hope of use narrower definitions. recovery in 2013 (Figure 19). The decline was larger in the first half of 2012 in all countries except Albania. Over the whole year, FDI moderated in Albania, and declined in FYR Macedonia,Kosovo and Serbia, mainly because of the negative impact of the Eurozone crisis, the high 2011 base in Serbia and FYR Macedonia, and significant outflows from FYR Macedonia and Serbia. In early 2013, however, there are some encouraging signs. FDI in Montenegro recovered in the second half of 2012 with a prospect of further growth in 2013. Similarly, the privatization of Telecom Company (PTK) in Kosovo in April 2013 and the entry into the market of Türkiye Iş Bankasi has boosted 2013 prospects for FDI in Kosovo. In Serbia, a better political environment and recent progress toward EU accession after the political agreement with Kosovo has boosted FDI prospects for 2013. FDIs in Albania and Bosnia and Herzegovina are, however, expected to remain flat. Figure 19: Net FDI 2009-12 (Percent of GDP) Figure 20: Net FDI by Country(Percent of GDP) SEE6 EU11 2010 2011 2012 5.4 5.7 6.0 15.0 4.4 4.0 3.1 10.0 1.8 2.0 2.0 2.0 1.5 5.0 - 0.0 2009 2010 2011 2012 MNE ALB KOS BIH SEE6 MKD SRB Source: SEE6 central banks. Source: SEE6 central banks. External Debt Total external debt in SEE6 shot up by 7 percentage points of GDP in 2012 as governments borrowed to compensate for declining revenues and weak public finances (Figure 21). After a 11 decline to 62.6 percent of GDP in 2011 from its peak of 65.5 percent in 2010, the average external debt of SEE6 hit a new record of 69.6 percent in 2012. From June 2009 to September 2012 four countries accessed international commercial markets by issuing Eurobonds (FYR Macedonia in 2009, Albania in 2010, Montenegro in 2010 and 2011, and Serbia in 2011 and 2012; Figure 22) or tapping loan markets with an IBRD guarantee (Serbia, FYR Macedonia, Montenegro). The upward trend was shared by all SEE6 countries. Serbia increased its external debt most by issuing a US$1.75 billion bond in 2012, bringing its total international bonds outstanding to US$2.7 billion (Figure 23). However, the largest external debt increase in 2012 was in Kosovo due to disbursements of the IMF Stand-by Arrangement, though Kosovo’s total external debt9 is still very modest at 8.5 percent of GDP. Montenegro and Serbia have the highest stocks of public and private external debt, well above the regional average (Figure 23). Figure 21: Average SEE6 External Debt (Percent Figure 22: Total International Bonds Outstanding, of GDP) Selected SEE6 Countries (US$ million) 3000.0 External debt o/w Gov. debt 80 69.6 2500.0 70 62.9 65.5 62.8 2000.0 60 50 1500.0 40 1000.0 30 20 500.0 23.3 25.0 28.6 10 19.6 0.0 0 ALB MK MNE SRB 2009 2010 2011 2012 Source: SEE6 central banks and ministries of finance Source: SEE6 ministries of finance. (MoF). Figure 23: Total Public and Private External Debt 2012 (% of GDP) 2009 2010 2011 2012 120 100 80 60 40 20 0 MNE SRB SEE6 MKD BIH ALB KOS Source: SEE6 central banks and ministries of finance; IMF; World Bank. Note: MNE and KOS external debts are estimates. 9 Unlike other SEE6 countries, Kosovo has very limited access to international financial markets and consequently little external debt. 12 FISCAL POLICY – DIVERGENT PRESSURES The fiscal deficits in SEE6 countries ticked upward in 2012, on average from 3.2 percent of GDP in 2011 to 3.9 percent in 2012 (Figure 24). The largest increases were in Serbia (2.6 percentage points of GDP) and FYR Macedonia (1.4 percentage points), the latter mainly because of the clearance of government payment arrears. Bucking the trend were Albania and Montenegro, where fiscal deficits declined. Across the region revenue and spending as a percent of GDP moved in the same direction in 2012. Two clear groups emerged: countries that boosted both revenue and spending (growing fiscal footprint) and those that saw both decline.  Growing fiscal footprint: Although the fiscal deficits increased, revenues actually rose as a percent of GDP in Serbia (Figure 25). Serbia had made significant efforts to boost revenues, raising both the VAT and corporate income tax (CIT) rates. Unfortunately, although the two countries benefited from increased revenues, their spending also went up, by 1.5 percentage points of GDP in FYR Macedonia (linked to paying down arrears) and by 6.3 percentage points in Serbia. Figure 24: Fiscal Deficits (Percent of GDP) Figure 25: Changes in Revenue and Spending, 2011-12 (Percent of GDP) 10 8 Change in Revenue Change in Spending 5 6 Change in Deficit 4 0 2008 2009 2010 2011 2012 2 ALB BiH KOS 0 MKD MNE SRB ALB BiH KOS MKD MNE SRB -2 Source: World Economic Outlook, April 2013. Source: World Economic Outlook, April 2013, and World Bank staff calculations.  Lessening fiscal footprint: Albania, Bosnia and Herzegovina, Kosovo, and Montenegro saw revenues fall as a proportion of GDP, by 0.2 to 1 percentage point. VAT receipts in particular took a hit (e.g., by over 2 percentage points in Albania) due to slow- or negative economic growth. All of these countries worked to cut spending to keep their fiscal deficits sustainable. In Kosovo, spending cuts were limited and related to the completion of the R7 highway to Albania. In some countries, spending cuts may have amounted to the accumulation of arrears rather than real reductions. Fiscal deficits in the region are projected to slightly decline in 2013 (Figures 26 and 27). The (unweighted) average fiscal deficit is expected to decline from 3.9 percent of GDP in 2012 to 3.7 percent. Serbia leads the way, where it the government’s plan is to cut the fiscal deficit from 6.8 to 4 percent of GDP. Bosnia and Herzegovina is also expected to pursue a small deficit reduction (0.5 percent of GDP) and the FYR Macedonian deficit will hold steady as a percent of GDP. By contrast, deficits are projected to increase in Albania and Kosovo, in the former due in part to weakening revenues and in the latter because of spending on the new R6 highway from Pristina to Skopje. 13 Revenue increases are expected to lead the fiscal recovery. Revenues are projected to remain stable across the region at an average 35.2 percent of GDP. This average hides declines in Albania and Montenegro. Elsewhere, revenues are expected to increase thanks to a slowly improving economic climate and further revenue-raising measures. Figure 26: Revenue and Expenditures, 2012 Figure 27: Fiscal Deficit, 2012 versus 2013 and 2013 (Projected) (Percent of GDP) (Projected) (Percent of GDP) 60 8 Deficit 2012 40 6 Deficit 2013 20 4 0 2 ALB BiH KOS MKD MNE SRB 0 Rev 2012 Rev 2013 Exp 2012 Exp 2013 ALB BiH KOS MKD MNE SRB Source: World Economic Outlook, April 2013 and Source: World Economic Outlook, April 2013 and World Bank staff estimates. World Bank staff estimates. Public Debt and Short-term Risks Public debt has been creeping up in all SEE6 countries for the past few years (Figure 28). On average general government debt (excluding guarantees and central bank bills) went up from 37 percent of GDP at yearend-2011 to 42 percent at yearend-2012. Serbia’s increase was about 12.5 percentage points, from 39.7 to 52.2 percent of GDP excluding guarantees (with guarantees, Serbia’s gross public debt was above 60 percent in 2012). Further, Albania changed the law to allow it to breach its debt ceiling of 60 percent of GDP. Yet sovereign credit ratings have been stable across the region since mid-2012 with the exception of FYR Macedonia’s recent downgrade to BB-(Table 3). State-guaranteed debt, an issue for several countries, is still rising (Figure 29). Guaranteed debt is over 11 percent of GDP in Montenegro and over 8 percent in Serbia. While guarantees remain below 5 percent of GDP in Albania, FYR Macedonia, they all seem to be edging up slowly. Kosovo issued a recent guarantee to the Electricity Transmission Company, KOSTT, but its guarantees are less than 0.2 percent of GDP. Guarantees in Bosnia and Herzegovina are about 0.8 percent of GDP. Figure 28: General Government Debta Figure 29: State Guarantees (Percent of GDP) (Percent of GDP) 80 15 2010 2011 2012 60 10 40 5 20 0 0 ALB BiH KOS MKD MNE SRB MNE SRB MKD ALB BiH KOS Source: World Economic Outlook, April 2013 Source: National authorities and World Bank staff a Gross concept used. Excludes guarantees. estimates. 14 Table 3: SEE6: Sovereign Credit Ratingsa Dec Dec Dec May 2010 2011 2012 2013 ALB B+ B+ B+ B+ BIH B+ B B B MKD BB BB BB BB- MNE BB BB BB- BB- SRB BB- BB BB- BB- Source: Standard and Poor’s. a Long-term foreign currency debt as of May 8, 2013. Kosovo does not have a sovereign rating. Short-term debt, especially in domestic currencies, has come to dominate the debt portfolios of some countries. Even though at year-end 2012 two-thirds of FYR Macedonia’s domestic currency- denominated debt was due by the end of 2013, the authorities succeeded during first half of 2013 to extend the maturity of domestic portfolio drastically reducing 3-months T-bills, while increase of 6 months and 12 months T-bills and new issuance of 3 and 5 year T-bonds. About half of Albanian domestic debt is due by the end of the year (Figures 30 and 31). Although only about a third of Serbian domestic-currency debt falls due by yearend, short-term debt is increasing: 58 percent of bonds and T- bills issued in the local market in 2013 are for 53 weeks or less, compared with less than half in 2012. All the bonds and T-bills Bosnia and Herzegovina has issued in 2013 have been short-term, compared with 57 percent in 2012. Figure 30: Local and Foreign Currency Debt Figure 31: Local Currency Debt due in 2013 due in 2013 (Percent) (Percent of GDP)a 70% 18% 60% 16% 14% 50% 12% 40% 10% LCU 30% 8% FX 6% 20% 4% 10% 2% 0% 0% MKD ALB SRB KOS BiH MNE ALB MKD SRB BiH MNE KOS Source: Bloomberg; World Bank staff calculations. Source: Bloomberg; World Bank staff calculations. a Excludes Central Bank bills, where issued and guaranteed debt.  Most SEE6 external public debt instruments have long maturities. No country has more than 8 percent of its total foreign exchange (FX)-denominated debt due in 2013 and Bosnia and Herzegovina, Kosovo have none; FYR Macedonia has EUR 175 million FX- denominated debt due in 2013 although it was redeemed in early January 2013.Some countries are using short-term FX-denominated instruments, however. 15 Combining long-term external debt with short-term domestic debt has helped to limit direct government exposure to exchange rate volatility risk and give domestic banks the opportunity to invest in short-term debt instruments. Much of the short-term public debt is purchased by domestic banks, allowing those with liquid assets to earn higher interest than central banks pay on deposits. At a time when private bank deposits are rising faster than credit, many commercial banks welcome the opportunity to earn higher returns elsewhere. However, short-term domestic public debt exposes governments to rollover risks. Although commercial banks are currently liquid and tend to roll over short-term public debt, a change in the internal or external financial or economic environments could reduce their willingness to do so. It would therefore be wise to consider all such risks when designing fiscal and debt policies. FINANCIAL SECTOR––SLOW- CREDIT GROWTH Credit growth in the region over the past year had been stagnant or decreasing (Figure 32). In Albania, Kosovo, FYR Macedonia and Serbia credit growth was slower than in previous years. Credit growth in Bosnia and Herzegovina was weak (at 2.8 percent growth) and in Montenegro, despite some improvements, was still negative. Figure 32: Real Credit Growth Rates (Percent Change)10 Demand factors explain a good part of 50 the declining or stagnating credit growth in SEE6 countries. The low 40 credit demand reflected the muted 30 economic activity of businesses and the 20 private sector following the crisis. Also, households and companies that had 10 borrowed more in the boom period are 0 now trying to reduce indebtedness to a -10 sustainable level. In addition, interest rate -20 developments make it clear that credit demand has dampened. Interest rates in ALB BIH KOS MKD MNE SRB the SEE6 countries over the last year fell 2009 2010 2011 2012 Average (2006-2008) (Figure 34); an indication that more pronounced demand-side factors were Source: National authorities and World Bank staff affecting credit growth. calculations. On the supply side, financing became more expensive as financial market conditions deteriorated, especially in the first half of 2012. European banks came under intense pressure as their credit default swap (CDS) spreads started rising in mid-2011 and again in the second quarter of2012.11 The sovereign CDS spreads of SEE6 countries also started rising, which made financing more expensive.12 10 Montenegro’s average credit growth for 2006-2008 was 97.2 percent (not shown). 11 Parent banks CDS spreads include: Raiffeisen Bank Austria, Erste Bank Asutria, Banca Intesa Italia, UniCredit Italia, Societe Generale France, National Bank of Greece and Alpha Bank Greece. 12 A common practice is for inter-group funding cost to be set at the cost of funding of the parent plus the CDS spread of the sovereign where the subsidiary is located. 16 Figure 33: Real GDP Growth, Real Credit Figure 34: Interest Rates on Loans and Growth, and Foreign Financing, 2011-12 percent) Deposits, 2011–12 8 Change in credit growth rates (2011-2012) 7.2 0.4 0.3 6 Change in GDP real growth rates (2011-2012) 0.2 4 0.1 Change in BIS reporting banks external position vis-à-vis SEE6, as % of GDP (2011-2012) 0.0 2 1.4 0.4 0.7 -0.02 Percentage point 0 -0.2 change in Deposits -0.2 (2011-2012) -2 -0.3 -1.4 -1.5 -1.5 -1.6 -0.4 -0.4 -2.0 -0.4 -4 -2.7 -3.1 -0.4 -3.7 -3.6 Percentage point -4.0 -0.6 change in Loans -6 -5.3 -0.7 (2011-2012) -0.8 -8 -0.8 -7.9 -8.4 -1.0 -10 ALB BIH KOS MKD MNE SRB ALB BIH MKD MNE SRB Source: National authorities, Bank for International Source: National authorities and World Bank staff Settlements and World Bank staff calculations. calculations. Despite recent improvements, the banking sector is still Figure 35: Funding Costs for SEE6 Countries 3000 1000.0 troubled. The second half of 2012 saw better CDS spreads of 2000 800.0 parent banks and SEE6 countries than the previous period, thanks 1000 600.0 to measures taken by the ECB 0 and the U.S. Federal Reserve. 400.0 This lowered the cost of funding -1000 (Figure 35).13 Consequently, after 200.0 -2000 a significant process of deleveraging that resumed in the -3000 0.0 second quarter of 2011 (see Box Change in BIS reporting banks external position vis-à-vis SEE6 (lext axis) 2) and resulted in a decrease in Parent Banks CDS spread, SEE6 average (right axis) the position of BIS-reporting SEE6 sovereign CDS spread, bps (right axis) banks vis-à-vis the region (2.9 percent of average 2011 plus Source: Bloomberg, Bank for International Settlements and World Bank 2012 GDP), the third quarter of staff calculations. 2012 saw an increase in the external position of 0.4 percent of combined SEE6 2012 GDP. However, as the fourth quarter of 2012 proved, the financial sector must still deal with nonperforming loans (NPLs), deleveraging, and reigniting credit growth. The fourth quarter saw a small but negative change in the external position of BIS- reporting banks (0.05 percent of combined 2012 GDP). 13 Emerging Markets Bank Lending Conditions Survey reports improved conditions on international markets in the second half of 2012. 17 Box 2: Deleveraging in SEE6 Countries Since 2008 SEE6 countries have been exposed to the risk of disruptive withdrawal of funds by European banks. By the end of 2008, according to the Bank for International Settlements (BIS), bank claims against the region had reached US$20.7 billion—22.2 percent of combined regional GDP. However, there were large disparities between countries. Claims against Serbia, Bosnia and Herzegovina, and Montenegro ranged from 27 to 32 percent but in FYR Macedonia and Albania were only about 5 percent, which has proved to be a good buffer. Since most funding to the region was intermediated through foreign-owned local banks that were also systemically important, the banking system considered the risk of fast withdrawal of funds to be significant. To complicate matters further, most foreign-owned banks in SEE6 were headquartered in troubled EU countries like Italy, Slovenia, and especially Greece. So far the worst- case scenarios have not materialized (largely thanks to the actions of international financial institutions, which created the Vienna Initiative to support the region through the crisis). However, the risk is still present, including confidence-induced shocks steaming form parent banks coming from troubled Euro area countries. Figure B2.1 Foreign-Bank Claims against SEE6 Banks (US$ millions) 25000 20000 15000 10000 5000 0 2007-Q3 2009-Q4 2012-Q1 2006-Q1 2006-Q2 2006-Q3 2006-Q4 2007-Q1 2007-Q2 2007-Q4 2008-Q1 2008-Q2 2008-Q3 2008-Q4 2009-Q1 2009-Q2 2009-Q3 2010-Q1 2010-Q2 2010-Q3 2010-Q4 2011-Q1 2011-Q2 2011-Q3 2011-Q4 2012-Q2 2012-Q3 2012-Q4 Claims on all sectors Liabilities to all sectors Source: Bank for International Settlements. Figure 36: Loan-to-Deposit Ratios Improved access to local 180 funding, through either deposits 160 or capital markets, will be central to the resumption of 140 credit growth. The external 120 position of BIS-reporting banks 100 vis-à-vis SEE6 countries still 80 averages 11.5 percent of combined 60 GDP (as of 2012). A sizable 40 supply-side headwind may 20 therefore be confronted if the process of shifting to locally 0 ALB BIH KOS MKD MNE SRB funded banks intensifies, at least in countries that have high loan-to- December 2012 Peak level (2006-2012) deposit ratios (Figure 36). The new Source: National authorities and World Bank staff calculations. rule of the game is to try and tie growth in lending to growth in deposits (or domestic financing in general), especially in the more unbalanced economies. Countries that 18 start from an already favorable position and can manage to fund lending from local sources are better off. Albania, Kosovo, and Macedonia have a long track record of loan-to-deposit ratios below 100; Bosnia and Herzegovina and Serbia will have a much harder time. Montenegro has managed to significantly lower its ratio from its peak.14 In fact, the four countries that had loan-to-deposit ratios below 100 up to mid-2009 saw higher credit growth during the crisis than the two countries that previously had high loan- to-deposit ratios. However, the evolution of supply conditions also reflects factors beyond the cost. Some of the reported tightening is due to domestic supply factors, such as monetary policy, reassessment of the economic outlook by banks, and high NPLs.15 The latest Emerging Markets Bank Lending Conditions Survey indicates that credit conditions continued to tighten as NPLs rose. Figure 37: Nonperforming Loans (Percent of Total Loans)16 NPLs are on the rise once again (Figure 37). High NPL levels reflect a deterioration of loan quality and 30 the chronic difficulties banks have in 25 writing off NPLs because of weak 20 insolvency regimes. Even though NPL levels differ significantly 15 across the region, the second wave 10 of increases is a cause for concern. 5 After briefly stabilizing (or declining in some countries, such as 0 Montenegro), NPLs started rising ALB BIH KOS MKD MNE SRB again, especially in the second half Dec-12 of 2012. Albania and Bosnia and Herzegovina have been hardest hit. Peak during 2008-2013 period NPLs in these countries are Pre crisis NPL level (end 2008) currently at their peak since the beginning of the crisis. FYR Source: National authorities and World Bank staff calculations. Macedonia,and Kosovo also saw a second wave of NPL increases, though less intense and the current NPL levels are just below the peaks. In Montenegro, despite a substantial decrease from the peak due to large sales of problem assets to factoring companies, NPLs are still the second highest in the region and still rising as there has been limited progress of restructuring or resolution. The overall picture in the region is of high NPLs that have been growing in the first months of 2013, constituting a drag on economic activity and on the willingness of banks to lend. At the same time, NPLs are well provisioned in Serbia, Kosovo, Macedonia, and to some extent Bosnia and Herzegovina. Provisioning in Albania and Montenegro is below average. 14 Interestingly, all countries except Kosovo had peak loan-to-deposit ratios in a very short span between March and July 2009 (right at the beginning of the financial crisis). Serbia had another peak in April 2011, and Kosovo had its peak in June 2010. 15 An IMF report on dealing with high NPLs in Central Eastern and Southeastern Europe confirms the adverse effect of NPLs on credit supply (www.imf.org/external/region/eur/.../030112.pdf). 16 Data for pre-crisis NPL level (2006-2008) for Serbia refers to end of 2008 level only. 19 Figure 38: Return on Equity Figure 39: Capital Adequacy Ratios 25.0 25.0 21.9 18.4 20.0 19.8 14.0 20.0 15.0 10.5 17.7 17.5 16.7 10.0 15.8 5.1 5.1 6.2 4.2 2.9 15.0 5.0 1.6 13.0 13.0 0.0 10.1 10.2 -5.0 10.0 -10.0 -15.0 5.0 -20.0 -18.4 -25.0 0.0 ALB BIH MKD MNE SRB ALB BIH MKD MNE SRB Average 2006-2008, quartery Average 2009-2012, quarterly End 2012 level Average 2006-2008, quarterly Average 2009-2012, quarterly Source: National authorities and World Bank staff Source: National authorities and World Bank staff calculations. calculations. Note: Data for Serbia for “Average 2006-2008 Note: Data for Serbia for “Average 2006-2008 quarterly� refers to 2008 only. Data for Montenegro quarterly� refers to 2008 only. for end 2012 refers to September 2012. On average banks in the region emerged from the crisis much less profitable than before: in all countries their profitability was down by more than half. Return on equity (Figure 38) declined from an average of 10.6 percent in 2006–08 to 0.7 percent in 2009–12, again with some cross-country differences. Montenegro was especially hard-hit. The good news is that banks managed to keep their capital positions solid (Figure 39). Compared to the pre-crisis level the capital adequacy ratios of banks did not change much; this helped regional banks to withstand the Euro area woes. Also, capital adequacy ratios in the region are substantially higher than is required by local regulations (between 8 and 11 percent, depending on the country). However, the high NPLs in some countries, although still manageable, could be a source of risk, and the regulators will have to monitor the situation closely. All the SEE6 countries have made considerable progress on financial sector reforms over the past two to three years, although not to the same degree. They have all enhanced the laws governing their central banks and both financial and nonfinancial institutions and their supervision. However, throughout the region, certain measures still need to be given priority:  Comprehensive measures to bring down NPL levels. Even though NPL levels differ across the region, in all SEE6 countries they affect the quality of banking sector assets, financial results, lending costs, management focus, and ultimately willingness to lend. Streamlining the resolution framework and insolvency procedures and adopting procedures for swifter out-of-court settlements might prove helpful.  Closer coordination between home and host country regulators. Even though the deleveraging has not yet caused any major disruptions, it could pose a challenge to the system. Thus, intensifying coordination with EU regulators is still vital to preventing any disruption. Vienna Initiative 2.0 would enhance coordination.  Building up local funding and local capital markets. This might prove to be especially important to countries that still have high loan-to-deposit ratios as banks try to find a more balanced and sustainable way of funding.  Enhancing deposit insurance protection through recapitalization of the deposit insurance agencies (where needed) or via specific agreement with international financial institutions. 20  Refining comprehensive crisis management frameworks. As the financial crisis has shown, authorities need to be prepared for a variety of adverse scenarios, and regulators need to be ready to coordinate crisis measures. LABOR TRENDS––HIGH UNEMPLOYMENT ACROSS THE REGION17 During 2012 unemployment stabilized at above 20 percent (Figure 40). In 2012, unemployment rates in SEE6 remained close to their peak crisis levels (purple diamonds in Figure 40).18 The average for the region was about 22.8 percent in Q4 2012, more than double the 11.2 percent average for EU11 countries. Unemployment went down slightly in Albania and Serbia (green bars) and worsened in Montenegro (orange bar) while not changing much in the other countries. As expected, where unemployment rates increased most, activity rates had fallen more (Figure 41). Figure 40: Unemployment Rates, 2012 Figure 41: Labor Participation and Unemployment Change in unemplpyment rate rate between 2007 and 5.0 SER 4.0 3.0 2012 (percentage points) 2.0 KOS 1.0 MK MNE 0.0 -4 -3 -2 -1 0 1 2 -1.0 BiH -2.0 -3.0 ALB -4.0 -5.0 Change in activity rate between 2007 and 2012 Source: LFS data and ILO. No post-2009 data are available Source: LFS data and ILO. for Kosovo. Note: Q4 2012 data, except for Bosnia and Herzegovina (Q3 2011). Activity rates are lower in SEE6 than in the EU11 countries, and the informal sector seems to be absorbing fewer unemployed than previously (Figure 42). Since 2008, in all the SEE6 but Serbia more women than men have entered the labor force. Male activity has been stagnant or dropped (Figure 43). Lower labor participation rates can be partly explained by high levels of informality19 in SEE6, where a large number of the self-employment and farm workers are in micro enterprises without paid employees. While SEE6 countries lack informal sector data, information from some new EU member states 20 suggests that informality was highest in agriculture (80.6 percent) and construction (38.2 percent), and 17 This section draws heavily World Bank ,2013d. 18 See also World Bank 2013a. 19 Informality can take the form of people working for unregistered employers, without a formal employment contract, or without social security coverage. Labor regulations usually do not apply to these employees. 20 This refers to: Bulgaria, Czech Republic, Estonia, Latvia, Poland and Slovakia. 21 among the self-employed (83.9 percent).21 Though informality is often considered the choice of last resort for the unemployed during crises, this time there is evidence suggesting that informal employment did not expand to fill the gap in most of Europe (Hazans 2011 and Walewski 2011).22 Figure 42: Labor Force Participation in SEE6 Figure 43: Changes in Labor Force by Gender, and EU11 Q4 2008 and Q4 2012 (Percentage points) 65 10 60 8 7.5 55 50 6 4 45 4 2.0 40 2 1.2 0.6 0.3 35 30 0 Croatia Hungary Czech Rep. Albania Serbia FYRM Latvia Slovak Rep. Romania Montenegro BiH Poland Slovenia Bulgaria Kosovo Estonia Lithuania -2 ALB BiH MKD MNE SER -4 -1.2 -6 -4.5 -4.2 -4.3 2007 2012 female male Source: LFS data. Source: LFS data. Note: Q4 2012 or end-2012 data, except for Bosnia and Herzegovina (Q4 2011). Youth unemployment in some countries, such as Serbia (51.2 percent) and FYR Macedonia (53.0), is double the national rate (Figure 44). While in most of the SEE6 unemployment among youth rose during the crisis, FYR Macedonia and Albania seem to have been able to create jobs for workers aged 15–25 (and for all other age groups). Although quite high, the ratio of youth to adult unemployment in SEE6 is actually in line with the level for other European and Central Asian countries even though more young people aged 15–24 are staying longer in education and enter the labor market later. There is also a large number of idle youth who are not in education, not employed, and not looking for work. In some countries with high youth unemployment, such as Kosovo, youth bulges put downward pressures on employment and earnings. In other countries, education and training systems do not provide the skills needed. However, often demand rather than supply issues are the real problem. Conditions for easier business entry and job access can help overcome the problem of limited competition that reduces employment opportunities. On the other side of the age spectrum, the aging of the population in SEE6 countries is projected to advance much slower than in the EU11 countries so that the working age population will either continue to expand, change little, or even expand (Albania) over the next two decades ( World Bank 2013). 21 World Bank 2012a: In from the Shadow Integrating Europe’s Informal Labor Market by Packard, Koettl and Montenegro. 22 Hazans, M. 2011and Walewski, M. 2011. 22 Figure 44: Unemployment by Age Group Figure 45: Change in Unemployment by Age Group, 2008 to 2012 (Percentage points) 80 SER BiH MNE MKD ALB 57.9 60 53.0 51.2 20 43.7 40 15 26.2 29.9 23.3 20.6 22.1 23.6 10 20 18.6 15.3 10.1 12.2 10.5 5 0 0 BiH MKD MNE SER ALB 15-24 25-49 50-64 -5 15-24 25-49 50-64 -10 Source: LFS data. Source: LFS data. Note: Latest available 2012 data, except for Albania Note: Latest available 2012 data, except for Albania and and Bosnia and Herzegovina (2011). Age groups for Bosnia and Herzegovina (2011). Albania and Serbia differ; they are: 15–24, 25–54, and 54–64. The SEE6 face innumerable Figure 46: Share of Public Sector Employment in SEE6 employment challenges, but 100 creating jobs for those who want to work is the most pressing issue (see 80 Chapter 2 on jobs). Next on the 60 agenda are reallocating people to 40 better, more productive jobs and 20 improving aspects of the work 0 people do. In some SEE6 countries a ALB BiH MKD KOS MNE SER significant share of jobs (20–30 1989 (Incl. Public Adm.) 1989 (Excl. Public Adm.) percent) is still in state-owned 2010 (Incl. Public Adm.) 2010 (Excl. Public Adm.) enterprises (SOEs) and other parts of the public sector, despite a dramatic Source: Life in Transitions Survey 2006 (1989 to 2006) and 2010 reduction over the past two decades (2010). (Figure 46). However, because the SOE share is being reduced and fiscal space for expansion is narrowing, workers are “queuing� for public-sector jobs, i.e., waiting for specific jobs to open up. This may also partly explain the high share of young and well-educated among the unemployed. Experience worldwide shows that the private sector creates nine out of ten jobs, making it the true engine of growth— and underscoring the need for the SEE6 countries to promote private investment as well as private sector employment. SEE6 countries used different policies to mitigate the impact of the recent crisis on jobs, though the goal for all was to break the vicious circle of limited job opportunities, slow growth in labor earnings and living standards, slow growth in productivity, and eroding social cohesion. All the SEE6 countries used fiscal stimulus, but FYR Macedonia and Montenegro also implemented policies to stimulate labor demand through limited wage subsidies, credit support, and public works programs. Wage subsidies often consisted of a reduction in social security contributions and were often targeted to specific firms or vulnerable groups. World-wide, high-income countries spent more than half their resources on credit policies to create and protect jobs. Low- and middle-income countries spent most of their resources 23 on direct job creation measures and temporary public works programs. FYR Macedonia and Montenegro also invested in skills and employment services. SEE6 OUTLOOK––FRAGILE RECOVERY In 2013 the SEE6 region is projected to grow 1.7 percent, ending the double-dip recession of 2012 (see Figure 47 and Table 4). All countries in the region are projected to grow, with Kosovo again leading the pack (3.1 percent), this time because of high public investments and a significant inflow of remittances. Serbia is projected to have the second highest growth rate (2 percent), which is crucial for the region’s performance because Serbia accounts for 45 percent of the region’s economy. Serbia is expect ed to benefit from increased FDI, the performance of FIAT, and a return to normal agricultural crops, which dropped nearly 20 percent in 2012. FDI is expected to rise with investor confidence based on possible opening of the EU accession talks later in the year. Albania is projected to grow at a rate slightly higher than in 2012 (about 1.8 percent). It could probably grow even faster if the construction sector was not dragging the economy down. Drivers of its growth will remain the extractive industry (oil, minerals) and electricity production. Sluggish growth is expected to continue in Bosnia and Herzegovina, Montenegro, and FYR Macedonia at about 0.5–1.4 percent in 2013. Bosnia and Herzegovina is projected to suffer from less external demand, but domestic demand may become healthier. Numerous Figure 47: SEE6––Real GDP Growth business environment issues still act as ALB BiH KOS MKD MNE SRB bars to a more significant inflow of 10 FDI and to expansion of domestic businesses. In FYR Macedonia the 5 moderate economic growth expected in the first half of the year should 0 slowly pick up in the second half, but since there have been major -5 improvements in the business environment, the growth outlook may -10 become sunnier. In Montenegro, early 2008 2009 2010 2011 2012 2013 data indicate that growth recovered in Source : World Bank staff. the first quarter of 2013, mainly because of good weather and higher production of electricity, but imports remain sluggish and aluminum production and supply bottlenecks could continue to be a drag on the economy; projected GDP growth is slightly over 1 percent for 2013. Table 4: SEE6––Real GDP Growth Projected Actual Estimated Estimated Projected Projected 2012 2012 H1 2012 H2 2012 2013 2014 ALB 0.8 1.6 0.9 2.2 1.8 2.0 BIH 0.0 -0.7 -0.2 -1.1 0.5 2.0 KOS 3.6 2.3 3.6 1.1 3.1 4.3 MKD 0.0 -0.3 -1.1 0.4 1.4 2.5 MNE 0.2 -0.6 -0.9 -0.2 1.2 1.5 SRB -2.0 -1.7 -1.5 -1.9 2.0 2.7 Weighted av. -0.5 -0.6 -0.5 -0.7 1.7 2.5 Source: World Bank staff. 24 Even this fragile recovery faces several downside risks. First, if the Euro area growth turns out weaker than projected, the SEE region will likely be affected, with the recovery even slower than projected. Second, in the financial sector, although there has been considerable progress, the SEE6 still face the risk of confidence-induced shocks stemming from weak (euro-area) parent banking institutions--a low- probability but nevertheless high-impact risk for many countries in the region—with significant consequences for growth. Finally, further increases in payment arrears (especially towards the private sector) and expansion of off-budget spending could undermine fiscal consolidation and complicate the liquidity conditions for the private sector, with palpable effects on macroeconomic stability and growth. These risks will need to be managed proactively if recovery is to take hold. 25 26 CHAPTER 2. SPECIAL TOPIC: REKINDLING JOB CREATION IN SEE6 This note summarizes the findings of the forthcoming World Bank report on jobs for the SEE6 economies (World Bank 2013, forthcoming). This report finds that up until the 2008 crisis, the SEE6 saw proportionally fewer jobs created, and more jobs destroyed, per unit of GDP growth than other countries in Europe and Central Asia (ECA) that were further along in reforming their economies. The delay in implementing reforms due to the turmoil of the 1990s meant that when the crisis hit countries had just started reaping the benefits of their reform efforts. Rekindling job creation in the region will require encouraging firms to create job, on the one hand, and addressing factors that dampen participation in the labor market and the employability of workers on the other. Salient elements of this agenda are (1) deepening reforms that would support the growth of young firms; (2) building up the quality of education to improve the provision of basic skills while laying the basis for acquisition of the skills needed in the “new economy�; (3) capitalizing on external migration to foster investment and innovation while removing barriers to internal mobility and reaping the benefits of migration; (4) creating more inclusive labor markets by lowering institutional barriers, such as high labor taxes and barriers to specific groups, such as women, older workers, and ethnic minorities. THE 2000S: A DISAPPOINTING DECADE FOR JOB CREATION Figure 48: Change in Employment Associated with 1 Before the crisis hit in 2008, the SEE6 Percent Change in Real GDP, 1995-2010 (Percentage was experiencing annual growth of 4.8 Points) percent, yet net employment creation had stagnated. This contrasts with the EU11, where annual growth of 6.4 percent led to positive job creation. 23 After the crisis hit, the relationship between growth and employment strengthened, with the economic contraction causing significant job losses. There were significant disparities (see Figure 48) in the way growth led to employment creation. A recent report by the World Bank (2013) identified a number of countries (almost all of the EU11) that were early implementers of reforms as “advanced modernizers.� Before the crisis, these advanced modernizers had managed to translate growth into employment creation faster Source: World Bank calculations based on ILO 2011, 2013; than the averages for the ECA region and World Bank 2013. Note: SEE6 (Albania, Bosnia and Herzegovina, the SEE6. Note that in these countries the FYR Macedonia - 1995-2010; Serbia – 2006-2010). Advanced Modernizers (Bulgaria, Czech Republic, Estonia, Hungary, Latvia, recession resulted in loss of proportionally Lithuania, Poland, Romania, Slovak Republic, Slovenia, Turkey) fewer jobs than in the SEE6. 23 This positive performance of the EU11 was much less impressive than the performance of other regions in the world over the same time period. 27 Why was the SEE6 slower to translate growth into job creation (and more vulnerable to job losses) than the advanced modernizers? The timing of reforms is an important factor.24 Because of the turmoil of the 1990s, SEE6 countries began introducing reforms much later than advanced modernizers. By the end of the 1990s Bosnia and Herzegovina, Serbia, and Montenegro had fallen significantly behind them in terms of the EBRD Transition Index (Figure 49). 25 By 2012, while advanced modernizers had continued to make progress, those three SEE6 countries managed to narrow the gap noticeably (FYR Macedonia and Albania continued to be ahead, as they had been for a decade.) The catching up was possible thanks to a decade of intense reforms (Figure 50). During the 2000s the SEE6 registered more improvement than any other group of countries in the region on all the profiles captured by the Transition Index. Only on one indicator did one country, Russia, improve more than the SEE6. Figure 49: EBRD Transition Index, 2000 and Figure 50: Changes in Transition Index 2012 Components, 2000 to 2012 5 4 3 2 1 0 2012 2000 Source: Based on World Bank (2013), calculations using data from EBRD (2013). As the SEE6 embarked on reforms, there is some evidence that countries further along on the transition path reaped the lagged benefits of reforms put in place earlier and moved on to second- generation reforms. Econometric analysis suggests that the positive performance of advanced modernizers in the 2000s reflects the pay-off for earlier reforms (Box 3). In particular, it seems that reforms that lowered the cost of restructuring (e.g., privatization and enterprise restructuring); leveled the playing field in product markets (e.g., competition); and improved the governance structure had the most impact on creating jobs. Reforms that directly tackled rigidities and imperfections in the labor and capital markets also had positive impacts, but mostly in countries that had already accomplished first-generation reforms. 24 While the timing of reforms is likely to have played an important role, others are also likely to have played an important role. For example the large outflow of skilled labor in the 1990s documented by Beine et al (2006) is likely to have slowed down considerably the process of structural transformation, hindering the shift of both capital and qualified labor to those industries with higher productivity increases. As it will be discussed below this underscores the importance of both closing educational gaps and tightening the links with the diaspora. 25 The EBRD Transition Index is an aggregate score averaging performance in terms of large and small scale privatization, competition policy, liberalization of trade and the foreign exchange system, price liberalization, and governance and enterprise restructuring. 28 Box 3: Reforms and Job Creation To analyze the direct impact of reforms on employment creation, Richter (2013) regressed employment growth on both GDP growth and the level of different reform indicators, such as the Doing Business (DBI) and the Transition (TI) indicators and their components. Parameter estimates suggest that a number of reforms can boost employment: Large-scale privatization, reforms that make it easier to close a business, and governance and enterprise restructuring are positively and significantly correlated with employment growth. In other words, other things being equal, the deeper the reforms in any of these three areas, the more jobs are created. Employment growth is also positively correlated with several governance indicators, including more control of corruption, better quality regulation, more effective government, and higher levels of voice and accountability. Certain reforms also have a positive impact on employment growth for some groups of countries but not others; their effectiveness might depend on their cumulative effect with reforms already undertaken. For instance, more flexible hiring regulations and banking reform are associated with more jobs created among advanced modernizers, and better competition policy and improved governance led to higher employment creation in intermediate and, especially, late modernizers. In examining the indirect impact of reforms on employment creation, in all cases the deeper and more advanced the reform, the closer the relationship between growth and employment. As with the results of the direct impact of reform, this effect is significant for reforms related to large- and small- scale privatization, governance and enterprise restructuring, and competition policy. However, parameter values and significance levels are quite sensitive to the model specification. Source: Richter (2013) prepared for World Bank (2013, forthcoming). SEE6 countries may have been just starting to benefit from their push toward reforms and restructuring but had not yet had a chance to reap the benefits of the reforms when the global crisis hit. This seems confirmed by the firm level evidence available for two of the SEE6 (Figure 51; firm evidence is available only for Bosnia and Herzegovina and Serbia among SEE6, presented here for purely comparative purposes with the Czech Republic and Poland, both advanced modernizers). Between 2002 and 2006 in both Bosnia and Herzegovina and Serbia, more jobs were being destroyed than created. By 2007 and 2008 that trend had reversed in both countries, but the global crisis that reached them at the end of 2008 sent it back down again. In contrast, the Czech Republic and Poland had the opportunity to enjoy more years of positive job creation, even if they were not spared the employment problems the crisis brought.26 26 It is worth underscoring that is analysis is very sensitive to data quality as measurement error could bias the estimates of job creation flows. 29 Figure 51: Net Job Creation, Selected SEE6 and EU11 Countries, 2002–09 Bosnia and Herzegovina Serbia 15.4 13.6 9.8 9.5 9.7 12.8 8.4 7.6 6.4 6.9 6.3 9.2 8.9 8.4 9.2 7.3 6.7 7.2 2.1 3.8 1.2 -0.2 -1.2 -1.2 -0.3 -3.0 -2.7 -1.2 -1.7 -2.5 -4.3 -6.2 -6.2 -5.6 -7.6 -7.5 -8.3 -9.0 -9.4 -10.0 -9.6 -9.7 -9.6 -10.1 -10.3 -11.0 -11.1 -12.8 2002 2003 2004 2005 2006 2007 2008 2009 2002 2003 2004 2005 2006 2007 2008 2009 Czech Republic Poland 18.2 6.0 5.7 4.5 4.9 4.7 12.4 4.1 4.0 3.3 3.4 9.5 2.8 1.8 2.1 6.7 1.5 6.4 4.8 5.5 0.4 0.2 0.1 3.8 4.5 2.9 2.9 1.4 -0.2 -2.3 -2.0 -2.2 -2.3 -0.9 -0.6 -2.5 -2.7 -3.2 -3.5 -3.0 -3.9 -5.8 -5.0 -6.6 -6.1 -5.6 -11.7 -8.5 2003 2004 2005 2006 2007 2008 2009 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: World Bank (2013, forthcoming). Note: The analysis includes the full sample of observations for each country (for these countries similar patterns hold when focusing on the more restricted sample for which panel information is available). See World Bank (2013) for details on the methodology and data limitations. AN AGENDA TO REKINDLE JOB CREATION What can countries do to ensure that growth translates into jobs? Action is needed to encourage firms to create jobs and to address factors that dampen participation in the labor market and the employability of workers. These factors vary by demographic group. They ultimately depend on the skills and mobility of prospective workers and on incentives for and barriers to work. Ethnic minorities, women, young workers, or older ones might be particularly affected by certain constraints. The different demographic profiles of the SEE6 countries will help to shape the exact content of the changes necessary in each (Box 4). 30 Box 4: SEE6 Demographic Trends and the Jobs Agenda While the demographic prospects for some of the SEE6 are favorable for the next few decades, others will see job creation complicated by demographic factors (Figure B4.1). Bosnia and Herzegovina and Serbia are expected to see their working age populations decline; others countries, particularly Albania and Kosovo, will see them increase. And while all countries will see more workers aged 55 –65 in the labor force, in FYR Macedonia, Serbia, and Bosnia and Herzegovina that group will be significantly larger than the group of youngsters (15–25). This suggests that while all countries should start assessing the implications of an aging population on their labor market prospects, in some more than others measures to ensure that older individuals remain active and keep their skills up-to-date needs to be a major priority. Similarly, while all countries should ensure that the young people that enter the labor market can find pathways to develop the right skills and be productive, this will be a particularly high priority in younger countries that will be depending significantly on the productivity of these workers for their growth. Figure B4.1: Composition of the Working Age Population, SEE6, 2030–11 (Percent) 14.0 12.0 10.0 10.0 8.0 6.3 6.0 3.9 2.8 3.3 4.0 1.8 2.0 0.0 -0.9 -0.9 -0.9 -0.2 -0.5 -2.0 Bosnia and Serbia FYR Macedonia Montenegro Albania -4.0 Herzegovina -2.6 -6.0 -4.2 -4.2 -8.0 Difference between 2030 and 2011 working age population (15+) -10.0 Difference between (55-64) and (15-24) in working age population, 2011 -12.0 -14.0 Difference between (55-64) and (15-24) in working age population, 2030 -12.7 Source: World Bank calculation based on UN population projection data, 2013. SUPPORTING ENTREPRENEURSHIP AND REMOVING BARRIERS TO JOB CREATION Net job creation in the EU11 is typically led by a small group of firms (about 20 percent) that grow faster than the others. While the sectoral concentration of these job creators varies by country, where the regulatory environment is less burdensome and there is less corruption firms grow faster, with significant improvements in employment and profitability. Greater competition, access to higher-quality infrastructure, and the efficiency of the courts are also associated with better performance. In the SEE6 countries for which evidence is available, that pattern seems to hold. In Bosnia and Herzegovina, for example, during 2006–09 a small group of about 17 percent of firms growing at over 20 percent a year created 70 percent of net jobs. Most of these were construction firms. In Serbia, a similar fraction of firms growing at over 20 percent accounted for 67 percent of net jobs in those years. However, while the fastest-growing job creators in EU11 are young firms, that is less true of the SEE6. One possible explanation is that new entrants into the market may be less able to mobilize resources like financial capital to finance growth, or that they might be dealing with disincentives to grow, at least in terms of the formal workforce. The gap in employment growth rates of new and older 31 firms is narrower than in the EU11.27 The same factors that make growth more difficult for younger firms—potentially the vehicle for innovation and new technology—to enter the market might, as elsewhere, have made those firms more vulnerable after the crisis.28 Basic regulation may be discouraging new businesses from emerging. Indeed, only a small percentage of those who declare to be interested in starting a business (“latent entrepreneurship�) take some steps in this direction. However, over two thirds of those who take some steps to start a business manage to succeed, other than in Kosovo, where this is true for less than 40 percent of those who take some steps (Figure 52). Interestingly, however, for the region as a whole the gap between latent and actual entrepreneurship is the same than in the EU11, even if there are differences in these gaps between countries (Figure 53). Figure 52: Steps taken towards Business Figure 53: Latent versus Actual Startup (1990 –2010) and Actual Business Entrepreneurship in SEE6, 2010 (Percent of Startup (Percent) working age population) 80 72.8 73.3 35 32.5 67.4 66.5 69.1 70 30 60 25 23.0 50 20.8 38.4 40.5 39.2 20 40 14.8 15.3 15 14.4 14.0 27.8 12.8 12.6 30 22.4 10.1 20.2 10 20 6.2 11.6 10 5 2.9 0 0 KOS BIH SRB MNE MKD ALB KOS BIH SRB MNE MKD ALB Took step, % Latent Actual, % took steps Latent, % Working Age Actual, % Working Age Source: LITS 2010 database; World Bank staff Source: LITS 2010 database; World Bank staff estimates. estimates. SEE6 has the lowest levels of latent entrepreneurship in the ECA region.29 Only 19 percent of those aged 18–64 report being willing to start their own businesses, against 21 percent in the EU11 and 22 percent in four Western European countries for which there are comparable data. At the individual level a number of characteristics, including experience in the private sector and active membership in organizations (for women respondents) are related to stated desire to become an entrepreneur. The still relatively high prevalence of public sector employment in some countries (Box 5) may help explain the low- rates of latent entrepreneurship. 27 Over 2004–07, according to the BEEP surveys in SEE6 the annual employment growth of firms created after 2003 was 9.7 versus 7.3 for those created before 2003; in the EU10 those rates were 11.1 versus 3.9 28 Detailed evidence on the negative impacts of the crisis on younger firms exists for other parts of the ECA region but not for SEE6 countries. 29 The region as a whole already displays much lower rates of latent entrepreneurship than other parts of the world. 32 The findings suggest that (a) there is a relatively high share of home-grown entrepreneurs who manage to create successful start-ups, but (b) these new firms are not providing sufficient new jobs to make a significant dent on unemployment. While some of the reforms already undertaken may be creating an environment that is relatively conducive to setting up new firms, more might be needed to remove constraints to their growth. The educational system might also deserve investigation in terms of whether it is creating new skills that might support entrepreneurship or that innovative firms might demand (as discussed below). Finally, expectations about the relative roles of the public and the private sectors in providing employment also need to be addressed. Box 5: Latent Entrepreneurship in Serbia Figure B5.1: Employment Sector Preferred In April 2012 a special module was No added to the Serbian Labor Force response / Survey (April 2012) to explore Refuse to attitudes and perceptions of employed answer Do not workers (Figure B5.1). The majority of 3% know 19% respondents overwhelmingly preferred Self- to work in the public sector. Among employed workers surveyed 61 percent preferred Public 11% this option, with just 10 percent sector Private enterprise choosing self-employment and 8 enterprise 65% percent private enterprise. More than 2% 30 percent of those surveyed were formal private sector workers who, if they could choose another job, would want to work for a state-owned Source: LFS survey with special module. enterprise (SOE). Interestingly, those Note: Answers to the question, “If you had to leave this job for another job, what would be the preferred sector?� already working for an SOE (36 percent of respondents) considered it comparatively less attractive than the civil service (particularly women and workers in the SOEs administered by the Privatization Agency and therefore likely to be either restructured or liquidated). MORE INCLUSIVE LABOR MARKETS, LOW- PARTICIPATION, AND EMPLOYABILITY Addressing Emerging Skill Gaps According to firm-level data, though in SEE6 countries skills are not the most serious constraint to doing business, their importance was on the rise before the financial crisis (Figure 54). In Kosovo and Albania the problem is more visible; more than 70 percent of firms there consider lack of skills to be a constraint. In all countries other than Montenegro and Kosovo, the share of firms that do not see skills as a problem declined between 2005 and 2008 (Figure 54). This may be a sign that SEE6 countries are starting to see changes in the demand for skills, particularly the growing demand for “new economy� skills (ability to analyze data and information, to think creatively and communicate with others, and other interpersonal skills) that some EU11 countries are already witnessing. Because education and training systems have not yet been able to adapt to these shifts, skills gaps are beginning to limit the employability of both younger and older workers. 33 Figure 54: Skills and Education NOT an Obstacle, 2005 and 2008 80 70 2005 2008 60 50 40 30 20 10 0 Source: World Bank 2011, Challenges to enterprise performance in the face of the financial crisis: Eastern Europe and Central Asia. Analysis of the skills content of the jobs held by different cohorts of workers shows that the SEE6 might also be seeing new skills demands (Figure 55). The pattern in FYR Macedonia (the only SEE6 country for which this analysis is available) is less marked than for countries where the process is more advanced, like Lithuania. Still, it appears that firms are increasingly demanding more non-cognitive non- routine skills (“new economy skills�). Younger generations are better able to access these jobs. As these trends continue, older workers are going to find their skills increasingly more obsolete, raising the dilemma of how to ensure that they remain employable. 34 Figure 55: Skills Content by Job-holder Cohort, FYR Macedonia and Lithuania a. Cohort born after 1974 b. Cohort born before 1955 FYR Macedonia, Cohort Born before 1955 FYR Macedonia, Cohort born after 1974 70 70 Mean Skill Percentil of 2007 60 Mean Skill Percentil of 2007 Skills Distribution 60 Skills Distribution 50 50 40 40 30 2007 2008 2009 2010 2011 30 2007 2008 2009 2010 2011 Lithuania c. Cohort born after 1974 d. Cohort born before 1955 Lithuania, Cohort born after 1974 Lithuania, Cohort Born before 1955 80 New 80 Economy Skills New 70 70 Mean Skill Percentil of 2002 Mean Skill Percentil of 2002 Routine Economy Skills Skills Distribution Skills Distribution cognitive 60 60 Routine 50 cognitive 50 Manual Skills 40 40 Manual Skills 30 30 2002 2003 2004 2005 2006 2007 2008 2009 2010 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: World Bank 2013. Note: New economy skills are defined as non-routine cognitive skills (analytical, required for analyzing information, thinking creatively, and interpreting information for others; and interpersonal, required to establish and maintain personal relations, and guide, direct, and motivate subordinates and coach others). Routine cognitive skills are those required to repeat the same tasks, be exact and accurate, and change between structured and nonstructured work. Manual skills are those required for both routine tasks, such as controlling machines and processes, and nonroutine tasks, such as operating vehicles. More fundamentally, there are serious concerns on the educational systems in the SEE6. The PISA reading test provides sobering evidence that current systems might not be equipping potential labor market entrants with even the basic skills (Figure 56). In Serbia 33 percent, in Montenegro 50 percent, and in Albania 57 percent of 15-year-olds appear to be functionally illiterate (unable to locate information in a simple text or make a simple connection between information in the text and everyday experience). 35 This compares with 24 percent for the EU11 and 19 percent in EU15. Functional illiteracy in Albania is among the highest recorded worldwide, and 11 percent of Albanian 15-year-olds score so low that PISA does not capture their score (the EU 11 average of such low scores is less than 2 percent). Figure 56: Functionally Illiterate 15-year-olds, SEE6 (Percent) 70 59.6 60 49.5 50 40 32.9 30 23.5 19 20 10 0 EU15 EU11 Serbia Montenegro Albania Source: PISA 2009; World Bank staff estimates. Improving employability will require improving the quality of education and addressing the constraints that disadvantaged groups face in learning. It will also require ensuring that young generations are learning the skills the new jobs require (e.g., acquiring analytical skills for problem-solving rather than learning facts). As the workforce ages, SEE6 countries should also put in place a system for adult continued learning and coordination mechanisms and regulations to implement it. Managing Internal and International Mobility Internal and international mobility can foster growth and job creation by opening up opportunities to aggregate economic activities and to better match jobs and workers (World Bank 2012b). Internal mobility can also be a force for structural transformation, both in terms of the shift from agriculture to manufacturing and services, and by fostering links between leading and lagging areas through remittances or local investments.30 International and national mobility are linked; for example, recent evidence from Albania suggests that international migrants might be providing the resources and encouragement for those left behind to move to more prosperous areas (Çaro et al. 2013). Albania, Bosnia and Herzegovina, and FYR Macedonia are among the top emigration countries in the world, with Albania in the top 10 (World Bank 2011). Migratory flows differ across SEE6 countries. More than half the migrants from Albania and Serbia were low-skilled (against one-third for the region as a whole, OECD 2012),31 while 29 percent of migrants from FYR Macedonia and 24 percent of those from Bosnia and Herzegovina had tertiary education (World Bank 2011). The high levels of emigration explain why remittances on average account for 10 percent of GDP for the SEE6 against 2 percent in the EU11. Remittances range from 4 percent in FYR Macedonia to 30 An aspect of internal mobility that seems particularly relevant at the present juncture is that it can affect the way economies adjust to negative regional shocks. In much of Europe these appear to take place often through reductions in labor force participation or persistent unemployment (WDR 2012). In contrast, in a very mobile country like the US labor mobility enables adjustment in both unemployment rates and real wages. 31 This might have been a self-reinforcing pattern as there is some evidence that in the case of Albania migration might have provided incentives for students to drop out to migrate toward low-skilled and higher-paid jobs abroad (World Bank 2010). 36 17 percent in Kosovo (2010; WDR 2013).32 As remittances generally support consumption rather than investment, the challenge for policy-makers now is to create incentives for more productive investment of remittances and tighten links with the diaspora. Closer links between the diaspora and the local economy might also build more circular patterns of migration that could favor the local economy and employability. A recent study of migrants returning to Albania, for example, found that migration experience promotes upward labor mobility (Carletto and Kilic 2011). And to the extent that high levels of tertiary student mobility (Box 6) result in these workers acquiring skills and experience abroad and coming back home, student migration offers the potential to significantly increase growth prospects. Public policies can make transitions to the home country easier, for example by making it easier to retain social benefits, buy property, and start a business. Box 6: Tertiary Student Mobility The SEE6 has the highest mobility of tertiary students in Europe. The EU has as a goal that by 2020 for “at least 20 percent of higher education graduates should have had a period of higher education -related study or training (including work placements) abroad, representing a minimum of 15 ECTS credits or lasting a minimum of three months.� So far mobility for tertiary education within Western Europe is just 2.3 percent of students enrolled Figure B6.1: Number of Studying Abroad, 2010 (Percent of (UNESCO 2010, no data are available Total Tertiary Enrolment) for graduates). 25 Student mobility in SEE6, on the other 19.5 hand, is much higher: 12.2 percent on 20 16.5 average, but with consistent variation between among countries. At the 15 lower end, just 5 percent of Serbian 12.2 11.8 students are enrolled in tertiary 10 8.5 education abroad (the same as EU11 5.8 5.9 countries). Albania has already 4.7 4.8 5 reached the Europe 2020 benchmark 2.3 1.4 with a student mobility share of 19.5 0 percent, and Montenegro is close, at 16.5 percent. The numbers are 8.5 percent for Macedonia and 11.8 percent for Bosnia and Herzegovina. Source: UNESCO Institute of Statistics (2013); World Bank staff Such high mobility can be seen as a estimates. positive sign. At the European level student mobility is encouraged to foster the development of more inclusive societies and for the benefits it offers for both those who move and those who stay such as greater self-confidence, adaptability and capacity for teamwork. Moreover, migrant students are found to have a greater sense of initiative and entrepreneurial skills.a a (European Commission, 2011, Mid term review of the Lifelong Learning Programme, COM(2011) 413). 32 The importance of those flows was underscored by the fact that at least one-third of households in Serbia, Kosovo, and Bosnia and Herzegovina reported the decrease in remittances as the main way they were affected by the global crisis over the first two years that its effects were felt. 37 SEE6 countries have less internal mobility than either the EU11 or the EU15. The exception is Albania, where internal mobility was repressed until 1990 so that now almost a quarter of the adult population seems to have moved to a different place in the last 20 years. Although previously mobility had been low in SEE6, in 2010 high shares of respondents declared a willingness to migrate, both nationally and abroad (the latter indicator being particularly high for FYR Macedonia, Figure 57). Whether these intentions materialize is likely to depend on the extent to which barriers to internal mobility are addressed, such as rigidities in factor markets (particularly housing 33 ); the portability of social benefits and other regional policies; and the existence of information and networks that could facilitate matching of workers to jobs. Urban development policies might also significantly influence the costs and benefits of internal migration. Figure 57: Internal Mobility and Intentions to Migrate (Percent) 50.0 46.5 40.0 35.8 35.3 30.632.3 27.729.5 27.9 27.0 26.7 30.0 24.5 23.5 20.0 20.0 15.3 16.5 17.2 11.2 12.6 10.0 0.0 MKD MNE KOS SRB BIH ALB % Adult population who moved in the last 20 years % Adult population who is willing to migrate inside their country % Adult population who is willing to migrate abroad Source: LITS (2010); World Bank staff estimates. Addressing Disincentives to Employment34 Policy makers in SEE6 are concerned that low labor market participation might be due to welfare program disincentives. Except for Albania and Bosnia and Herzegovina, which are focusing on first- generation reforms (better targeting and administration of safety nets), countries in the region are starting to tackle employability and welfare dependence as part of social assistance reforms. A series of studies are currently looking at how design of social assistance benefits might be interacting with barriers to employability and participation to generate dependence (World Bank 2013e). Given the limited coverage and attenuated generosity of SEE6 safety nets, they are unlikely to be a major factor in the current high levels of labor market inactivity, but some design features might lead to long-term dependency for certain groups, particularly if the coverage and generosity of the scheme were expanded without program design changes. Today, while half the recipients of safety net programs 33 In SEE6 home ownership is above 90 percent in all countries other than Montenegro, a very high share (LITS 2010). Owning a home, combined with shallow rental markets (particularly for “official� rentals, in a context where registration requirements make it attractive to rent through informal transactions) and limited access to residential mortgages are likely to significantly restrict mobility. 34 The analysis of social assistance and labor taxation in this section draws heavily, occasionally also verbatim, from World Bank 2013 e. 38 might be able to work (they are of working age, are not engaged in education or training, and are not incapacitated), only a small percentage of those who could work (2 percent in Montenegro, 11 percent in Serbia) actually receive those benefits and are therefore exposed to the disincentives.35 A different type of disincentive effect might be created by design features that imply that workers receive the difference between their income and a given threshold, as any additional income they might be able to earn will be deducted from the amount of the transfer. Since benefits are withdrawn for income levels lower than the minimum wage, it is unlikely that many will be affected by this disincentive effect. Perhaps more significant is that qualifying for social assistance benefits often automatically leads to eligibility for other benefits as well, which would magnify any disincentives to work. Although disincentives may not be much of a problem, positive incentives to work might need to be reinforced. In most SEE6 countries some work requirements are embedded into social assistance design. In Kosovo, for example, recipients must participate in public works programs; in FYR Macedonia recipients who could work must prove they are making efforts to search for jobs, and there are specific provisions to help individuals transition from social assistance to work. Compliance with the “actively searching� requirement, however, might be through acts, such as registering with the unemployment office, that are little more than formalities. From a policy point of view social assistance could be better designed by omitting some sources of work income from the determination of eligibility.36 In addition, since, by penalizing certain forms of asset ownership, the criteria for social assistance eligibility might create disincentives to earning, savings, and acquiring assets, it might be useful to review the criteria. Institutional reforms to better link the work of the Public Employment Services (PES) and centers for social work and increasing the capacity and financing of PES programs to provide activation measures on a large scale would also be essential to facilitate creation of more inclusive labor markets. While the design of social assistance design is not likely to be a major barrier to activation of recipients, high labor taxes, particularly for low-wage earners, are likely to create considerable disincentives to work. All the SEE6 countries except FYR Macedonia have “tax wedges� on labor (defined as the sum of costs of social contribution by employers and employees and of the personal income tax of employees, expressed as a share of total labor costs) that are higher than the OECD average. For example in Serbia, the minimum base for calculating social security contributions equals 35 percent of average salary, so part-time and low-paid jobs for which the employee’s monthly gross salary is below the threshold face a disproportionate tax burden. A similar floor for social contributions exists in FYR Macedonia. This is likely to create strong disincentives for workers to take up lower paying jobs, at least in the formal sector. Reviewing labor taxation could make formal jobs more competitive, creating incentives to expand the tax base—a tax base likely to shrink as the population ages. Several OECD countries and EU members have moved toward “in work� tax credits, benefits which reduce the tax burdens on some groups and reduce the number of claimants for unemployment benefits. Also worthy of consideration is the possibility of financing social assistance out of general taxation, rather than labor taxation, along the lines of reforms conducted by Poland over the last decade (World Bank 2013e). 35 The low participation rates of beneficiaries appear to be largely driven by their profile, with low skills, particularly among out-of-school youth, very prevalent. 36 Similar measures might be applied in the case of unemployment benefits, which are typically discontinued when an unemployed person takes up a job; beneficiaries would not accept jobs that would pay less than the unemployment benefit, thereby reinforcing the bias created by the tax system against low-paid or part-time formal jobs. 39 While the disincentives discussed affect all workers, several groups face specific barriers in the labor market. Such barriers span a variety of factors, ranging from access to information, networks, and productive inputs to flexible work arrangements and adequate work environments, the provision of care for the children and the elderly, and social norms. SEE6 countries as a whole, for example, score much higher than the benchmark countries in the EU15 on indicators that measure the importance of network and connections in getting good jobs in government or in the private sector. As a result of these barriers, access to jobs depends heavily on gender, parental education, and majority status (World Bank 2013b), all indications of inequality of opportunity as they do not depend on the efforts and abilities of the individual worker. Minority men and women have significantly lower employment rates than the general population. For instance, while 40 percent of Roma women are in the labor force in Serbia, only 9 percent work37—a much greater gap than for majority women. This suggests that there are significant barriers in accessing jobs. And the persistent cultural expectations that women perform household chores and provide care to children, the disabled, or the elderly are reinforced by the lack of alternative public provision of these services. In Serbia and FRY Macedonia, for example, women spend at least three hours more a day on household chores while men spend almost two hours more on paid employment. Further, even when these groups enter the labor market, wage gaps persist (gender gaps are estimated at 19 percent in Albania, 18 percent in FYR Macedonia, 16 percent in Montenegro, and 11 percent in Serbia). Finally, older workers are unlikely to find new opportunities for employment Some of the measures discussed, such as incentives for more flexible forms of work (e.g., part-time) or creating opportunities for skills upgrading and lifelong learning, are likely to help address the constraints that these groups face. Other specific measures might be to expand the provision of good- quality and affordable care for children and the elderly (particularly with an aging population). Finally, longer-run measures should address social norms and attitudes that lead to groups being excluded, especially as expressed in the education and legal systems, and working with the media to reverse stereotypes. CONCLUSIONS While ensuring the conditions for economic recovery and regain pre-crisis reform momentum will be essential, a broader set of interventions focused on both the demand and the supply of labor will be needed to rekindle job creation in the region. This will include efforts on the demand side, by continuing business climate reforms to eliminate impediments to business expansion and foster entrepreneurship, particularly as opportunities to work in the public sector decrease. On the supply side, making workers more employable will require ensuring the quality of their skills, removing disincentives and barriers to work, and better managing internal and international mobility. Continuing to invest in the collection and dissemination of labor market data is also an important priority. As shown in this chapter, coverage of different aspects of labor market performance is quite uneven across countries, and often access to the micro-data necessary for deeper analysis is restricted to the research community. Addressing these data gaps will be essential to strengthen program and policy design. 37 The World Bank, ECA (2010): Roma Inclusion: An Economic Opportunity for Bulgaria, Czech Republic, Romania and Serbia Policy Note. 40 REFERENCES Beine, Michel, Frédéric Docquier, and Hillel Rapoport (2006) Measuring International Skilled Migration: New Estimates Controlling for Age of Entry, World Bank Research Report. Carletto, G. and T. Kilic (2011), Moving Up the Ladder? The Impact of Migration Experience on Occupational Mobility in Albania, Journal of Development Studies, 47, pp. 846-869. Erka Çaro, Ajay Bailey and Leo J. G. Van Wissen (2013), Exploring Links between Internal and International Migration in Albania: a View from Internal Migrants, Population, Space and Place, forthcoming. Hazans, M. (2011), Informal Workers across Europe: Evidence from 30 European Countries. Background paper for: In from the Shadow: Integrating Europe’s Informal Labor. Policy Research Working Paper 5912, World Bank, Washington, DC. IMF, (2013), World Economic Outlook, International Monetary Fund, Washington, DC. OECD (2012), Connecting with Emigrants: A Global Profile of Diasporas, OECD Publishing. Richter, K. and B. Witkowski (2013), Does Growth Generate Jobs in ECA?, Mimeo, World Bank, Washington, DC. Walewski, M. (2011), An Attempt to Measure the Trends in Shadow Employment in Poland and the Transition Probabilities out and into Shadow Employment Using the LFS Data Augmented by the Results of Dedicated Survey Performed by CASE in 2007. Background paper for: In from the Shadow: Integrating Europe’s Informal Labor. Policy Research Working Paper 5910, World Bank, Washington, DC. WDR (2012): World Development Report 2012: Gender Equality and Development, World Bank, Washington, DC. WDR (2013): World Development Report 2013: Jobs, World Bank, Washington, DC. World Bank ( 2010) : Albania—The New Growth Agenda: A Country Economic Memorandum�). World Bank (2011): Migration and Remittances Factbook 2011, Second Edition, World Bank, Washington, DC. World Bank (2012a): In from the Shadow: Integrating Europe’s Informal Labor Market, by Packard, Koettl and Montenegro, World Bank, Washington, DC. World Bank (2012b): Golden Growth, Restoring the lustre of the European economic model, World Bank, Washington, DC. World Bank (2013a), ECA Employment Monitor, World Bank, Washington, DC. 41 World Bank (2013b, forthcoming): From Jobless Growth to Growing Jobs: Fostering Employment Creation in Eastern Europe and Central Asia, Human Development Unit and Poverty Reduction and Economic Management Unit, Europe and Central Asia Region, World Bank, Washington, DC. World Bank (2013c), Global Economic Prospects June 2013: Looking for stable ground, World Bank, Washington, DC. World Bank (2013d), World Development Report: Jobs, World Bank, Washington, DC. World Bank (2013e): Western Balkans Activation and Smart Safety Nets AAA Synthesis Note, Mimeo, World Bank, Washington, DC. 42 ANNEX: KEY INDICATORS Figure A. 1: Real GDP: Percentage Change since Pre-Crisis Peak Percentage change from 2008 Real GDP index (2002=100) Albania 160 Albania Bosnia and Herzegovina 150 Kosovo Kosovo 140 Macedonia, FYR Montenegro Montenegro 130 Serbia Macedonia, FYR 120 Bosnia and Herzegovina SEE6 110 EU11 Serbia 100 EU15 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 -5 0 5 10 15 Source: World Bank ECA database. Figure A. 2: Real GDP Growth Forecasts for 2013 Forecasted GDP growth in 2013, percent Percent Albania 12.5 Kosovo Bosnia and Herzegovina 10.0 Kosovo 7.5 Serbia Macedonia, FYR 5.0 Macedonia, FYR Montenegro 2.5 Serbia Albania 0.0 -2.5 Bosnia and Herzegovina SEE6 -5.0 EU11 Montenegro -7.5 EU15 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013f 2014f -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Source: World Bank ECA database and World Economic Outlook. Figure A. 3: Unemployment Rate 2012, percentage of labor force (aged 15-64) Percentage of labor force (aged 15-64) 50 Albania Kosovo 45 Bosnia and Herzegovina 40 Macedonia, FYR Kosovo 35 Bosnia and Herzegovina Macedonia, FYR 30 Serbia Montenegro 25 Serbia 20 Montenegro 15 Albania 10 EU15 2005 2006 2007 2008 2009 2010 2011 2012 0 5 10 15 20 25 30 35 40 Notes: Albania as of 2010; Bosnia and Herzegovina as of 2011. Preliminary 2012 estimates for Bosnia and Herzegovina show that unemployment rate among the active population aged 15 and over was 28 percent, up from 27.6 percent recorded in 2011. 2010 and 2011 estimates for Kosovo are not available. Source: National statistical offices and Eurostat. 43 Figure A. 4: Quarterly Youth Unemployment Rate 2012, percentage of labor force (aged 15-24) Percentage of labor force (aged 15-24) 80 Bosnia and Herzegovina 75 Bosnia and Herzegovina Kosovo 70 65 Kosovo Macedonia, FYR 60 Macedonia, FYR 55 Montenegro 50 Serbia Serbia 45 40 Montenegro 35 EU15 30 2005 2006 2007 2008 2009 2010 2011 2012 0 10 20 30 40 50 60 70 Note: 2010 and 2011 estimates for Kosovo are not available. Source: National statistical offices and Eurostat. Figure A. 5: Fiscal Balance 2012, percentage of GDP Percentage of GDP 7.5 Albania Bosnia and Herzegovina Bosnia and Herzegovina 5.0 Kosovo Kosovo 2.5 Macedonia, FYR Macedonia, FYR 0.0 Albania Montenegro -2.5 Serbia Montenegro -5.0 Serbia -7.5 SEE6 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013f 2014f -7 -6 -5 -4 -3 -2 -1 0 Source: World Bank ECA database. Figure A. 6: Public Debt 2012, percentage of GDP Percentage of GDP 100 Albania Albania 90 Bosnia and Herzegovina 80 Montenegro Kosovo 70 Macedonia, FYR 60 Serbia Montenegro 50 40 Macedonia, FYR Serbia 30 Bosnia and Herzegovina 20 SEE6 10 Kosovo EU11 0 EU15 2005 2013f 2014f 2015f 2003 2004 2006 2007 2008 2009 2010 2011 2012 0 10 20 30 40 50 60 70 80 90 100 Source: World Economic Outlook and Kosovo Ministry of Finance. 44 Figure A. 7: Exports as a Share of GDP 2012, percentage of GDP Real effective exchange rate (2005 = 100) 135 Albania 130 Bosnia and Herzegovina 125 Kosovo 120 Serbia Macedonia, FYR 115 Macedonia, FYR Montenegro 110 Bosnia and Herzegovina Serbia 105 100 95 SEE6 2010 2005 2006 2007 2008 2009 2011 2012 0 10 20 30 40 50 60 Source: International Financial Statistics and national statistical offices. Figure A. 8: Real Export Growth 2012, percent Real exports (2006 = 100) 250 Albania 225 Kosovo 200 Bosnia and Herzegovina Albania 175 150 Montenegro Kosovo 125 100 Serbia Macedonia, FYR 75 Bosnia and Herzegovina Montenegro 50 25 Macedonia, FYR Serbia 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 -25 -20 -15 -10 -5 0 5 10 15 20 25 2012 Source: World Economic Outlook. Figure A. 9: Current Account Balance 2012, percentage of GDP Percentage of GDP 0 Albania -5 Macedonia, FYR -10 Bosnia and Herzegovina -15 Bosnia and Herzegovina -20 Albania Kosovo -25 -30 Serbia Macedonia, FYR -35 -40 Montenegro Montenegro -45 -50 Kosovo Serbia -55 2006 2000 2001 2002 2003 2004 2005 2007 2008 2009 2010 2011 2012 -25 -20 -15 -10 -5 0 Source: World Bank ECA database. 45 Figure A. 10: Non-Performing Loans 2012, percentage of total loans Percentage of total loans 25 Albania Albania 20 Bosnia and Herzegovina Montenegro 15 Serbia Kosovo 10 Bosnia and Herzegovina Macedonia, FYR Macedonia, FYR Montenegro 5 Kosovo Serbia 0 2006 2000 2001 2002 2003 2004 2005 2007 2008 2009 2010 2011 2012 0 5 10 15 20 25 Source: World Development Indicators and national statistical offices. Figure A. 11: Basic Financial Sector Statistics Domestic credit to private sector, percentage of GDP M2, percentage of GDP 90 90 80 Montenegro 80 Albania 70 Bosnia and Herzegovina 70 Bosnia and Herzegovina 60 60 50 Serbia 50 Macedonia, FYR 40 Macedonia, FYR 40 Serbia 30 30 20 Albania 20 Kosovo 10 Kosovo 10 Montenegro 0 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: World Development Indicators and World Bank ECA database. Figure A. 12: Ease of Doing Business 2012, proximity to frontier (best practice = 100) Proximity to frontier (best practice = 100) 100 Albania Macedonia, FYR 95 Bosnia and Herzegovina 90 Kosovo Montenegro 85 Macedonia, FYR 80 Serbia Montenegro 75 Serbia 70 Kosovo 65 Albania SEE6 60 55 Bosnia and Herzegovina EU15 50 EU11 2005 2006 2007 2008 2009 2010 2011 2012 50 55 60 65 70 75 80 85 90 95 100 Source: Doing Business. 46