102089 EU REGULAR ECONOMIC REPORT 2 S U S TA I N I N G R E C O V E R Y, I M P R O V I N G L I V I N G S TA N D A R D S f all 2 0 1 5 The EU Regular Economic Report (RER), is a semi- Anil Onal, Barbara Cunha, Tu Chi Nguyen, Natalia Millan, notes and acknowledgments annual publication of the World Bank Group and Paul Andres Corral Rodas, Leszek Kasek, Raquel Letelier, covers economic developments, prospects, and Alena Kantarovich, Natalie Nicolaou, Tulu Balkir, Sanja economic policies in the European Union. The report Madzarevic-Sujster, Stella Ilieva, Jose Luis Sanchez, Marc uses four sub-groups that share broadly similar develop- Schiffbauer, Sonia Plaza, Andrei Silviu Dospinescu, Ekaterine ment patterns EU (see map): Central Europe comprises T. Vashakmadze, Emilia Skrok, Catalin Pauna, Pilar Salgado Bulgaria, Croatia, Czech Republic, Hungary, Poland, Otonel, Suzana Petrovic, and Magali Pinat. Tito Cordella Romania, Slovak Republic and Slovenia; Northern Europe provided excellent advice throughout the preparation of comprises Denmark, Estonia, Finland, Latvia, Lithuania this report. The dissemination of the report and media and Sweden; Southern Europe comprises Cyprus, Greece, relations are managed by an External Communication Italy, Portugal and Spain; Western Europe comprises team comprising Andrew Kircher and Anna Kowalczyk. Austria, Belgium, France, Germany, Ireland, Luxembourg, The team is grateful to Christian Bodewig and Jean- the Netherlands and the United Kingdom. While the Francois Marteau (Acting Country Directors, Central report covers the European Union, it provides additional Europe and the Baltic States), Satu Kähkönen (Director, information on European countries which had historically Macroeconomics and Fiscal Global Management), Ivailo a stronger operational engagement with the World Bank Izvorski (Practice Manager, Macroeconomics and Fiscal Group, in particular, Bulgaria, Croatia, Poland and Romania. Global Management), Andrew Mason (Practice Manager, The focus note of this RER covers “Welfare States and the Social Protection and Labor Global Management), Carolina Protection of the Poor in the European Union”. Sanchez (Practice Manager, Poverty and Equity), the peer This report is prepared by World Bank staff economists reviewers Hans Timmer, Tatiana Didier and Margaret working on EU countries. The team of authors comprises Grosh and the Country Management Team for their guid- Doerte Doemeland (Task Team Leader), Enrique Aldaz- ance. The team would also like to thank the Central Banks Carroll, Theo Thomas (Task Team Leader), Matija Laco, and Ministries of Finances of Bulgaria, Croatia, Romania Gabriel Inchauste (lead author - poverty), Ramya Sunda- and Poland for their comments. This and previous EU RERs ram (lead author –focus note), Aylin Isik-Dikmelik (lead can be found at: www.worldbank.org/eurer. author –focus note). The team benefited from inputs from Finland Sweden Northern Europe Estonia Latvia Denmark Ireland United Lithuania Netherlands Kingdom Poland Belgium Germany Luxembourg Central Europe Western Europe Czech Rep. Slovak Rep. France Austria Hungary Slovenia Croatia Romania Portugal Spain Bulgaria Italy Southern Europe Greece Malta Cyprus ta b l e o f c o n t e n t s eu regul ar economic report  01 E XECUTIVE S UMM ARY  05 RECENT ECONOMIC DEVELOPMENTS 06 After a prolonged crisis that led to an increase in povert y…  10 …THE RECOVERY IS GAINING STRENGTH, FUELED BY PRIVATE CONSUMP TION AND SUPPORTED BY LOWER OIL PRICES, A WEAKER EURO AND THE ECB’ S QE  13 LABOR MARKETS ARE IMPROVING  17 FISCAL CONSOLIDATION IS SLOWING  19 COUNTRIES CONTINUE TO IMPROVE BUSINESS REGULATIONS IN AN EFFORT TO BOLSTER GROWTH  25 OUTLOOK  26 THE EU NEEDS TO TURN TAILWINDS INTO SUSTAINED GROWTH  28 MEDIUM-TER M RISKS ARE SUB STANTIAL  31 S POTLIG HT ON CENTR AL EUROPE  35 FOCU S NOTE: WELFARE STATES AND PROTECTION OF THE POOR IN THE EUROPEAN UNION  36 EU MEMBER STATES ARE AMONG THE BIGGEST SPENDER S ON SOCIAL PROTECTION IN THE WORLD…  40 …NOT ALL COUNTRIES PROVIDE EFFECTIVE PROTECTION FOR THE POOR  44 MOST WELFARE STATES CUT SOCIAL ASSISTANCE SPENDING DURING THE CRISIS…  46 …WITH DIFFERING OUTCOMES FOR THE POOR  49 EU MEMBER STATES WOULD NEED TO PROVIDE ADEQUATE PROTECTION AND SOCIAL INVESTMENT FOR THE POOREST WHILE ENSURING FISCAL SUSTAINABILIT Y  52 REFE RENCES f igur e s ta b l e o f c o n t e n t s   /   fi g ur e s  06 [ FIG . 1 ] THE CRISIS LED TO A SIGNIFICANT INCREASE IN POVERT Y IN SE VER AL EU COUNTRIES  07 [ FIG . 2 ] YOUNG AND LESS SKILLED PEOPLE WERE MOST AFFECTED BY THE INCREASE IN POVERT Y  0 8 [ FIG . 3 ] HOUSEHOLD INCOME HAS DECREASED MORE THAN GDP IN GREECE  0 9 [ FIG . 4 ] THE DECLINE IN LABOR INCOME AND EMPLOYMENT CONTRIBUTED MOST TO THE POVERT Y INCREASE  1 0 [ FIG . 5 ] GROWTH HAS PICKED UP PACE ACROSS ALL REGIONS FUELED BY CONSUMP TION  11 [ FIG . 6 ] CONFIDENCE IS IMPROVING 1 2 [ FIG . 7 ] THE COST OF BORROWING FOR ENTERPRISES AND HOUSE PURCHASES HAS REACHED A HISTORIC LOW  1 3 [ FIG . 8 ] THE EU HAS LOST FEWER JOB S THAN THE US BUT FIR MS PERFOR M WOR SE  14 [ FIG . 9 ] UNEMPLOYMENT R ATES ARE REACHING PRE-CRISIS LE VELS IN MANY EU COUNTRIES, e xcept in the South  1 5 [ FIG . 10 ] THE CUMULATIVE UNEMPLOYMENT RESPONSE TO A GDP GROWTH SHOCK IS LARGEST IN SOUTHERN EUROPE  16 [ FIG . 11 ] YOUTH UNEMPLOYMENT IS DECLINING IN MOST EU COUNTRIES  1 7 [ FIG . 12 ] SIGNIFICANT FISCAL ADJ USTMENT HAS BEEN ACHIE VED THROUGH DIFFERENT APPROACHES  2 0 [ FIG . 13 ] DIFFERENCES IN THE BUSINESS ENVIRONMENT IN THE EU VARY SIGNIFICANTLY WITHIN AND ACROSS REGIONS AND OVER TIME  2 2 [ FIG . 14 ] LARGE DIFFERENCES IN REGULATORY PR ACTICES DEPEND ON WHERE SMES LOCATE THEIR BUSINESS  26 [ FIG . 15 ] GROWTH R ATES ARE PROJECTED TO RISE ACROSS THE EU THOUGH INVESTMENT GROWTH REMAINS MODEST  3 2 [ FIG . 16 ] POVERT Y IS PROJECTED TO DECLINE f igur e s eu regul ar economic report  37 [ FIG . 17 ] EU MEMBER STATES SPEND ON AVER AGE MORE ON SOCIAL PROTECTION THAN OECD COUNTRIES 3 8 [ FIG . 18 ] THERE ARE FOUR DISTINCT EUROPEAN WELFARE STATES 40 [ FIG . 19 ] SOCIAL ASSISTANCE SPENDING VARIES WIDELY ACROSS EUROPEAN WELFARE STATES 40 [ FIG . 20 ] WITH VARYING IMPLICATIONS FOR THE COVER AGE OF BOTTOM 20 PERCENT 41 [ FIG . 21 ] CATEGORICAL PROGR AMS DOMINATE SOCIAL ASSISTANCE SPENDING ACROSS EUROPEAN WELFARE STATES 42 [ FIG . 2 2 ] HIGHER COVER AGE OF BOTTOM 20 PERCENT IS ACHIE VED MOSTLY THROUGH FAMILY BENEFITS AND SOCIAL E XCLUSION PROGR AMS 42 [ FIG . 2 3 ] SOCIAL ASSISTANCE LEAKAGE TO THE RICH IS SUB STANTIAL IN MANY EUROPEAN WELFARE STATES 43 [ FIG . 24 ] LEAKAGE TO THE RICH IS MAINLY DUE TO FAMILY BENEFITS 43 [ FIG . 2 5 ] LARGE BALANCED WELFARE STATES ARE THE MOST SUCCESSFUL IN POVERT Y ALLE VIATION THROUGH SOCIAL ASSISTANCE 43 [ FIG . 26 ] ONLY FEW COUNTRIES CAN AFFORD HIGH COVER AGE AND REASONABLE ADEQUAC Y 44 [ FIG . 27 ] SOCIAL ASSISTANCE SPENDING FELL IN SE VER AL COUNTRIES DURING THE RECESSION WHILE SOCIAL INSUR ANCE SPENDING WAS PROTECTED 45 [ FIG . 28 ] THE DECLINE IN SOCIAL ASSISTANCE SPENDING WAS MOST COMMON IN TRUNCATED AND LIMITED WELFARE STATES 47 [ FIG . 29 ] CUTS IN SOCIAL ASSISTANCE SPENDING HAD DIFFERENT OUTCOMES ACROSS WELFARE STATES IN TER MS OF COVER AGE OF THE POOREST 47 [ FIG . 3 0 ] INCREASE IN COVER AGE WAS ACHIE VED PARTLY THROUGH IMPROVEMENTS IN TARGETING ACCUR AC Y boxe s fi g ur e s   /   b o x e s   /   ta b l e s  0 8 [ BOX . 1 ] POVERT Y TRENDS IN GREECE  1 5 [ BOX . 2 ] UNEMPLOYMENT ADJ USTMENT TO A GDP SHOCK IN EUROPE  21 [ BOX . 3 ] DIGITAL DIVIDE OR DIVIDEND?  2 2 [ BOX . 4 ] THE IMPORTANCE OF LOCAL BUSINESS REGULATION: THE CASE OF POLAND AND SPAIN  29 [ BOX . 5 ] THE EU AND THE REFUGEE CRISIS  37 [ BOX . 6 ] SOCIAL PROTECTION FUNCTIONS 39 [ BOX . 7 ] T YPOLOGY OF SOCIAL PROTECTION SYSTEMS ta b l e s  2 0 [ TABLE 1 ] EASE OF DOING BUSINESS R ANKINGS IN THE EU  eu regul ar economic report Standard Disclaimer: This volume is a product of the Copyright Statement: The material in this publication is staff of the International Bank for Reconstruction and copyrighted. Copying or transmitting portions or all of Development/The World Bank. The findings, interpreta- this work without permission may be a violation of the law. tions, and conclusions expressed in this paper do not The International Bank for Reconstruction and Develop- necessarily reflect the views of the Executive Directors of ment/The World Bank encourages dissemination of its The World Bank or the governments they represent. The work and will normally grant permission promptly to repro- World Bank does not guarantee the accuracy of the data duce portions of the work. For permission to photocopy presented in this work. The boundaries, colors, denomi- or reprint any part of this work, please send a request with nations, and other information shown on any map in this complete information to the Copyright Clearance Center, work do not imply any judgment on the part of The World Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, tele- Bank concerning the legal status of any territory or the phone 978-750-8400, fax 978-750-4470, http://www. endorsement or acceptance of such boundaries. 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. a c r o n ym s a n d a b b r e vi at i o n s a c r o n y m s a n d a bbr e vi at i o n s   /   l i s t o f c o u n t ri e s   ABSPP ASSET -B ACKED SEC U R I T I ES   PPP PURCHASING POWER PARIT Y P U R C H ASE P R OG R AM   PSPP PUBLIC SECTOR PURCHASE PROGR AM   BRICS B R A Z I L , R U SS I A , I ND I A AND C H I NA   RER REGULAR ECONOMIC REPORT   CAB C UR R ENT ACCO UNT B ALANCE   SME SMALL AND MEDIUM ENTERPRISE   CBPP COVERED BOND PURCHASE PROGR AM   SILC S URVE Y ON INCOME AND   EAPP E XPANDED ASSET PURCHASE PROGR AM LIVING CONDITIONS   ECB EUROPEAN CENTR AL BANK   UK UNITED KINGDOM   EU EUROPEAN UNION   US UNITED STATES   GDP GROSS DOMESTIC PRODUCT   QE QUANTITATIVE EASING   ICT INFOR MATION AND COMMUNICATION   VAR VECTOR AUTOREGRESSION TECHNOLOGY   VAT VALUE-ADDED TA X   IMF INTERNATIONAL MONETARY FUND   WEO WORLD ECONOMIC OUTLOOK   MTO MEDIUM TER M OB JECTIVES   OECD ORGANIZ ATION OF ECONOMIC CO-OPER ATION AND DE VELOPMENT l i s t o f c o u n t ri e s   BG BULGARIA   CY C YPRUS   HR CROATIA   EL GREECE   CZ C ZECH REPUBLIC   IT ITALY   HU HUNGARY   PT PORTUGAL   PL POLAND   ES SPAIN   RO ROMANIA   AT AUSTRIA   SK SLOVAK REPUBLIC   BE BELGIUM   SI SLOVENIA   FR FR ANCE   DK DENMARK   DE GER MANY   EE ESTONIA   IE IRELAND   FI FINLAND   LU LUXEMBURG   LV LAT VIA   NL NETHERLANDS   LT LITHUANIA   UK UNITED KINGDOM   SE SWEDEN Executive Summary eu regul ar economic report The European Union (EU) is gradually emerging this measure - surpassed all other EU countries and – from a protracted economic downturn that led to an measured by an absolute poverty line of USD5 in PPP terms increase in poverty in many countries. Poverty rates – exceeded poverty in several Central European countries, – as measured by the share of the population with an including Hungary, Poland, Slovenia and Slovak Republic. income of less than 60 percent of the 2008 median in- The increase in poverty in the EU has been mainly due to a come – increased between 2008 to 2012, particularly in decline in labor incomes, employment and, to a lesser Southern European countries that were most affected by extent, social assistance spending. the downturn. By 2012, poverty in Greece - according to Many EU countries cut social assistance spending increase in poverty. Building on the lessons from the crisis, during the crisis, which, combined with the design important reforms for making social protection systems of the programs, sometimes contributed to an in- more effective include the introduction of guaranteed crease in poverty. While EU countries are among the minimum income (GMI) programs, maintaining the effec- biggest spenders on social protection in the world, some tive coverage and adequacy of existing social exclusion EU countries are unable to provide effective protection programs, and reducing the leakages of social assistance for their poorest citizens. In some, cuts in social assistance transfers to the rich. spending during the crisis did not help ameliorate the Fueled by consumption, the EU recovery is gaining from 2014. After leading the recovery in 2014, robust strength across all EU regions. Growing wages and private consumption growth was sustained in the first half employment, supported by low consumer-price of 2015 across all EU regions, as real disposable household inflation, have raised household incomes and sup- income grew strongly. That was largely due to higher ported households’ willingness to consume. Real wages and employment combined with exceptionally low GDP growth rose by a moderate 1.8 percent year-on-year consumer price inflation, driven by low energy prices. Un- in the first half of 2015, below the 2.2 percent OECD average employment among the young – who suffered the largest and half the global pace of expansion. The economy of increase in poverty during the crisis – is finally declining, Central Europe expanded by 3.5 percent, the highest re- but remains high in Croatia, Slovenia and Southern Europe. gional rate. Southern Europe’s growth rate, at 1.5 percent, Estimates suggest that the recovery has led to a small was the slowest, but was still a significant improvement decline in poverty in Bulgaria, Croatia Poland and Romania. 2 e x e c u t iv e s u m m a ry Investment remains the weak link to a more vigorous exports contributed positively to growth in the first half of recovery. Unlike the US, investment as a share of GDP is 2015 in all regions of the EU except Southern Europe, as the not yet trending upwards across the EU. Investment by Euro depreciated. After years of consolidation, the fiscal firms remains weak despite increasing confidence and policy stance across the EU has become broadly neutral as favorable financial conditions supported by the ECB’s fiscal deficits in 15 out of 28 EU member states fell to below Quantitative Easing (QE). Despite declining world trade 3 percent of GDP. and weakening import demand from China and Russia, net The recovery is expected to gain strength over the and labor markets continue to improve. However, growth medium term, fueled by private consumption but is projected to remain below pre-crisis levels as investment investment is likely to be constrained by structural remains subdued as private and public sector deleveraging rigidities. Private consumption is expected to continue continues. The output gap is projected to close slowly over to support the expansion, as inflation remains subdued the medium-term. The EU needs to turn current tailwinds into self- increase the coverage of the poorest and most vulnerable sustaining growth through continued structural groups, while ensuring that spending is efficient and reform. Countries need to continue to remove con- affordable over the long-term. Well-designed social as- straints on businesses by reducing rigidities in labor and sistance systems also facilitate labor market activation of product markets and ensuring labor has the right skills poor and vulnerable groups, supporting poverty reduction through investing in education and training. In this con- and long-term growth. text, social assistance systems also need to continue to Significant risks to a more robust recovery remain. growth and higher financial volatility could all exacerbate The crisis has left a legacy of high public and private-sector these vulnerabilities and undermine the current outlook. debt. A prolonged period of low investment growth and Moreover, regional tensions could undermine confidence below-target inflation, a slowdown in emerging-market in the recovery. 3 chapter one RECENT E C O N O MIC DEVELOPMENTS 06 A F T E R A P R O LO N G E D C R I S I S T H AT L E D TO A N I N C R E A S E I N P OV E R T Y… 10 …THE RECOVERY I S G AINING STRENGTH, FUELED BY PRIVATE CON SUMP TION AND SUPPORTED BY LOWER OIL PRICES, A WE AKER EURO AND THE ECB’ S QE 13 L ABOR M ARKET S ARE IMPROVING 17 FISC AL CON SOLIDATION I S SLOWING 19 COUNTRIES CONTINUE TO IMPROVE BU SINES S REGUL ATION S IN AN EFFORT TO BOL STER GROW TH eu regul ar economic report A f ter a prolonged cr isis t hat led to a n i ncrea se i n pover t y… The 2008-09 global financial crisis hit the EU hard, went through an exceptionally steep and sustained eco- leading to an increase in poverty in many EU coun- nomic downturn, suffered from a very large increase in tries. 1 As the EU endured two contractions between poverty which in 2012 exceeded poverty in all EU coun- 2008 and 2012, poverty, as measured by the share of the tries. 2 In contrast, some Central European countries, such population with an income of less than 60 percent of the as Poland and the Slovak Republic, achieved significant and 2008 median income, increased in most EU countries sustained declines in poverty during the same period fig. 1 . and, particularly, in Southern European. Greece, which Q fig. 1 The crisis led to a significant increase in poverty in several EU countries Poverty anchored at 60 percent of median income in 2008 in adult equivalent terms 60 50 40 30 20 10 0 AT BE FR DE IE LU NL UK EL IT PT ES BG CZ HU PL RO SK SI DK FI SE EE LV LT West South Central North 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: World Bank estimates using EU-SILC. Data for Croatia prior to 2010 is not available. / Note: Incomes have been measured in USD 2011 PPP terms. Income year 2012 is the latest year for which survey data is available. 1 The EU typically uses a relative poverty line set at 60 percent of median adult equivalent income. In order to compare poverty over time, we anchor this poverty line at the 2008 value. For EU targets see: http://ec.europa.eu/social/ 2 In 2012, poverty in Greece – measured by an absolute poverty line of USD5 in PPP terms – exceeded poverty in several Central European countries, including Hungary, Poland, Slovenia and Slovak Republic 6 r ecent economic de velopment s Young people and children were particularly affected where poverty increased substantially more following by the poverty increase. In 2012, poverty rates were the the crisis, particularly for children and young people (see highest for individuals aged 15-24, followed by those Box 1) fig. 2B . Poverty among the elderly declined on aver- under 15 fig. 2A . This was especially true in Southern Europe, age in all regions, except in Southern Europe. Q fig. 2 Young and less skilled people were most affected by the increase in poverty Poverty anchored to 60 percent of median income in 2008 in adult equivalent terms. A. Poverty by Age in 2012 B. Change in Poverty by Age, 2008-2012 45 20 40 15 35 30 10 25 5 20 15 0 10 -5 5 0 -10 West South Central North West South Central North Ages 0-14 Ages 15-24 Ages 25-54 Ages 0-14 Ages 15-24 Ages 25-54 Ages 55-64 Ages 65+ Ages 55-64 Ages 65+ Source: World Bank estimates using EU-SILC. Note: Incomes have been measured in USD 2011 PPP terms. 7 B OX 1 . P OV E RT Y T R E N D S I N GR E E CE eu regul ar economic report Greece has suffered from an exceptionally steep poverty, measured by the share of the population with and long economic downturn that translated into an income below 60 percent of the median income an even steeper decline in household incomes. anchored in 2008, more than doubled from 19.7 percent Between 2008 and 2012, real GDP per capita plummeted to 45.9 percent over the same period. Greeks at the low by 21 percent, or 5.7 percent per year on average- by far end of the income distribution suffered the most. While the largest economic decline of any EU country in recent average incomes dropped by 37 percent, the income of years. During this period, household income per capita the bottom 40 percent fell by 43.1 percent fig. 3B . fell 37 percent, or 10 percent a year fig. 3A . As a result, Q fig. 3 Household income has decreased more than GDP in Greece A. Household income and GDP B. Growth Incidence Curve, 2009-2012 120 -30 100 Change in income per capita (%) -40 Index (2004=100) 80 -50 60 -60 40 Real GDP per capita -70 20 Real household income per capita -80 0 1 21 41 61 81 Percentiles 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Source: Real GDP per capita from Eurostat; EU-SILC UDB surveys 2009-2013. Changes in employment and pensions explain the middle-income households. Pension incomes declined relatively steep decline in incomes of the bottom 40 relatively little up to 2012, protecting pensioners and the percent. First, young and less-skilled people constitute relatives who lived with them. As a result, poverty among a significant share of the population with relatively low the elderly – which was relatively low at the onset of the incomes. They were the most affected by the increase in crisis – barely increased. At the same time, social transfers unemployment after 2008: their unemployment rates were small, accounting for less than 5 percent of income doubled, peaking at 58.3 percent for the young and 29.8 of the bottom 40 percent of the income distribution – percent for the less-skilled in 2013. Second, labor earnings and covered few low-income households, thus playing a fell significantly after 2008, particularly in sectors that limited role in protecting them from falling into poverty employ a relatively large share of lower-skilled workers. (see Focus Note). Third, patterns in social protection spending insulated 8 A decline in labor incomes, employment and in some by 6.8 percentage points between 2008 and 2011. Cuts in r ecent economic de velopment s cases government transfers contributed to the rise employment added an additional 5 percentage points. In in poverty. Labor income makes up the largest share of Slovenia, declining labor incomes and employment contrib- gross household income in EU countries, ranging from 55 uted a total of 3.9 percentage points to the increase in the percent in Ireland to 67.7 percent in Poland in 2012. Labor poverty rate fig. 4. By contrast, Poland saw its poverty rate income declined significantly during the crisis, particularly decline as a real GDP growth of about 3.4 percent led to an in- among those in the bottom 40 percent of the income crease in labor income. Cuts in social assistance contributed to distribution. 3 This decline increased Greece’s poverty rate a rise in poverty in many EU countries (see also Focus Note). Q fig. 4 The decline in labor income and employment contributed most to the poverty increase Contribution to change in poverty per capita between 2008 and 2011 in percentage points 20 15 10 5 0 -5 -10 AT BE FR LU NL UK EL ES IT BG CZ HU PL RO SI DE EE FI LV LT West South Central North Share of income net of taxes Share of adults Share employed Labor income per employed adult Pensions Social Assistance Other Total Source: World Bank estimates using EU-SILC panel date. Data for Croatia is not available. Note: Poverty anchored at 60 percent of the median income in 2008, measured in USD2011PPP. Poverty estimates for 2014 and 2015 for some Central with a focus on those factors that are most likely to affect European countries suggest that the EU recovery is the evolution of poverty across the EU, such as household likely to alleviate poverty in many countries. In fact, income, labor market development and social assistance poverty in Bulgaria, Poland and Romania has likely declined spending. The focus note will discuss how well-designed in 2014 and 2015 (see Spotlight on Central Europe). The social protections systems across the EU are for protecting next section will look at recent economic developments the poor. 3 The labor income of the bottom 20 percent also declined steeply in Southern Europe, particularly in Spain and Greece, but increased in some EU member states, including in the Netherlands, the UK, Poland, and the Czech Republic. 9 eu regul ar economic report ...t he recover y is ga i n i ng st reng t h , f ueled by pr ivate con su mpt ion a nd suppor ted by lower oi l pr ices, a wea ker Eu ro a nd t he ECB ’s QE The EU growth recovery is gaining strength but GDP growth in the first half of 2015 at 3.5 percent driven by remains moderate. Real GDP growth quickened to 1.8 a solid growth outturn in the Czech Republic, Poland and percent (y-o-y) in the first half of 2015 from 1.3 percent in Romania. Flash estimates indicated that growth remained the first half of 2014 FIG. 5A . The recovery has taken a firmer strong in the third quarter of 2015, with the EU continuing root in all regions, but particularly in Southern Europe, to grow at an annual rate of by 1.9 percent, despite contrac- where real GDP growth increased from 0.1 percent in the tions in Greece, Estonia and Finland. This year’s economic first half of 2014 to 1.5 percent in the first half of 2015, largely acceleration notwithstanding, growth remained below driven by stronger growth in Spain and to a lesser extent the OECD average of 2.2 percent and half the global pace Portugal. The Central Europe experienced the highest real of expansion. Q fig. 5 Growth has picked up pace across all regions fueled by consumption A. Year-on-year chain-linked GDP volume growth B. Contributions to growth in nominal disposable income (in percentage points, not seasonally adjusted) in the EU 28 (in percentage points, seasonally adjusted) 4 6 Taxes 3,5 Social transfers in kind 3 5 Social benefits 2,5 2 Net property income and other current transfers 1,5 4 Gross operating surplus and mixed income 1 Wages 0,5 3 Adjusted gross disposable income 0 -0,5 -1 2 H1 2014 H1 2014 H1 2014 H1 2014 H1 2014 H1 2015 H1 2015 H1 2015 H1 2015 H1 2015 EU28 West South Central North 1 0 Private consumption Public consumption Fixed investment Change in inventories Net exports Real GDP growth -1 2014Q4 2014Q2 2013Q4 2014Q3 2014Q1 2013Q2 2015Q2 2013Q3 2013Q1 2015Q1 -2 Source: Eurostat; World Bank calculations. Note: Contributions are calculated based on the Eurostat methodology described in Compiling Annual and Quarterly National Accounts Main Aggregates for the European Union 10 The recovery has largely been fueled by private con- r ecent economic de velopment s in labor markets, which led to higher wages and employ- sumption as household disposable incomes increased ment. Higher earnings of the self-employed as well as due to a rebound in labor markets and lower oil prices. property incomes also supported household income Growth in private consumption contributed 1.2 percentage growth.4 Contributions from social transfers in kind and points to real GDP growth in H1 2015 and accelerated in all social benefits remained broadly constant FIG. 5B . Real dis- four regions FIG. 5A . In fact, private consumption growth posable income grew even stronger as consumer price significantly outpaced real GDP growth in the first two inflation remained below zero throughout the first quarter quarters of 2015. Nominal disposable household incomes of 2015 and turned negative again in September 2015 due increased by 5 percent (y-o-y) in H1 2015, driven by a rebound to falling energy prices5 as oil prices reached a multi-year low. The Euro depreciated strongly, supporting net export the EU accelerated despite declining world trade and growth. The Euro depreciated by over 10 percent in real weakening import demand from China and Russia and net effective terms between August 2014 and March 2015, exports contributed positively to growth. The EU current before appreciating by 3.7 percent between March and account surplus reached a historical high of 2.2 percent of September 2015. As a result, extra-EU export growth across GDP in June 2015. QE has helped the recovery by increasing confidence Q fig. 6 Confidence is improving and easing financial conditions. After reducing policy Economic sentiment indicator long-run average=100 rates to near zero in the wake of the global financial crisis, the ECB started to implement its quantitative easing policy in March 2015. In the six months through September 2015, the ECB purchased securities issued by Euro-area member states and selected institutions that amounted to €343.3 billion 6, as most economic confidence indicators have janv.-13 avr.-13 juil.-13 oct.-13 janv.-14 avr.-14 juil.-14 oct.-14 janv.-15 avr.-15 juil.-15 oct.-15 gradually improved. The overall economic sentiment for the EU remains above its long-run average FIG. 6 . Construction confidence indicator Retail confidence indicator Consumer confidence indicator Services confidence indicator Industrial confidence indicator Economic sentiment indicator (right) Source: Eurostat 4 The Central Europe was the only region that saw household incomes decline in Q2 2015, as wage growth and net property incomes slowed down. 5 Consumer price inflation peaked at 0.3 percent in May 2015 Core inflation increased slightly to 1 percent in October 2015 as a result of the pass-through of the Euro’s depreciation and the continued recovery in domestic demand. 6 Under its expanded asset purchase program (EAPP or “QE”), which comprises the “covered bond purchase program” (CBPP3), the “asset-backed securities purchase program” (ABSPP) and the “public sector purchase program” (PSPP), the ECB buys around €60 billion a month (currently planned to last until September 2015). 11 Despite accommodative monetary policy and fa- declined in every country in the euro area, with the cost of eu regul ar economic report vorable financial conditions, investment remains borrowing ranging from 1.6 percent in France to 4.8 percent subdued as investors remain concerned about the in Greece, as of July 2015.7 The average interest rate strength of the recovery, and households and corpo- charged on new mortgages has also declined in all Euro rations continued to deleverage. Despite accommodative area countries, except in Ireland, with the cost of borrow- monetary policy and favorable financial conditions, invest- ing ranging from 1.4 percent in Finland to 3.3 percent in ment remains subdued as investors remain concerned Cyprus. The Euro area bank lending survey Q3 2013 (ECB about the strength of the recovery, and households and 2015a) confirms that the interest rate level was an important corporations continued to deleverage. Rising confidence factor in the increase in credit demand for households and and favorable financial conditions have led to an increase enterprises within the EU. Credit growth was highest in in asset prices, contributing to improvements in private Central Europe, closely followed by Western Europe, but sector balance sheets. In fact, the net financial wealth of remained subdued in Southern Europe, though underlying households and non-financial corporations in the Euro demand and supply factors differed across countries.8 area has started to exceed pre-crisis levels. The cost of bor- Despite improvement in private sector balance sheets and rowing for both households and corporations has reached favorable financial conditions, investment growth remains historic lows in both nominal and real terms, supporting sluggish in the EU and is not yet trending upwards as a share credit demand FIG. 7A & 7B, respectively. For enterprises, the of GDP. Investment by firms remains weak (see also FIG. 8D ). average interest rate charged on new bank loans has Q fig. 7 The cost of borrowing for enterprises and house purchases has reached a historic low Annualized Agreed Interest Rate on new loans in percent B. House purchases A. Enterprises 7 5 min. of Jan-05 thru Jul-14 4,5 min. of Jan-05 thru Jul-14 6 juil-14 juil-14 4 5 juil-15 juil-15 3,5 4 3 2,5 3 2 2 1,5 1 1 0,5 0 0 EA19 AT DE NL CY LV PT SK FR EE ES EL EA19 IT IE FI SI AT DE CY NL LV PT SK EE FR ES EL IT IE SI FI Source: European Central Bank. / Note: The rate is the average Annualized Agreed Interest Rate / Narrowly Defined Effective Rate charged on new loans during the period Not only has the cost of bank loans declined, but corporate bond spreads have also reached historically low levels. 7 While in Italy demand for credit picked up and supply conditions (credit standards and terms and conditions) eased, suggesting some signs of recovery, demand for 8 credit and supply conditions remained unchanged in Spain. 12 r ecent economic de velopment s L abor ma rket s a re i mprov i ng Despite the prolonged crisis, the EU has lost fewer participation rates in Q2 2015 are higher than at the begin- jobs than the US since 2007. In 2007, the employment ning of 2008 in most European countries, even in those rate9 was 63 percent in the US and 53.1 percent in the EU. In with still high unemployment rates10. Second, Europe has 2014, the US rate was still 4 percentage points below its relied more on reducing hours worked per job than on level in 2007, but only 1.4 percentage points lower in the EU shedding jobs. With trends in unit labor costs similar be- FIG. 8.A . There are two main reasons for employment declining tween the US and Europe FIG. 8B , but differing job losses, less in the EU. First, labor force participation declined total employment compensation of firms in Europe seems steeply in the US after the crisis. By contrast, labor force to have been less reactive to the crisis and may have con- participation rates increased consistently in the EU28 as tributed to divergent trends in the operating surplus of previously inactive groups, in particular elderly workers firms and investment FIG 8C & D. and women, joined the labor market. In fact, labor force Q fig. 8 The EU has lost fewer jobs than the US but firms perform worse A. Employment rate B. Unit labor cost (percent of population aged 15+, rebased 2007=100) (deflated by CPI, rebased 2007=100) 101 102 100 100 99 98 98 97 96 96 95 94 94 92 2009 2008 2007 2014 2010 2012 2013 2011 2009 2008 2007 2014 2010 2012 2013 2011 US EU28 US EU28 US (adjusted by hours worked) EU28 (adjusted by hours worked) Source: OECD. 9 Employment rate denotes the number of employed relative to the working-age population. 10 Portugal is one of the few countries were labor force participation in Q2 2015 was slightly lower than at the beginning of 2008. 13 C. Non-financial corporations operating surplus D. Non-financial corporations fixed investment eu regul ar economic report (share of GDP, rebased 2007=100) (share of GDP, rebased 2007=100) 102 100,5 100 101 99,5 100 99 99 98,5 98 2007 2008 2009 2010 2011 2012 2013 2014 98 2007 2008 2009 2010 2011 2012 2013 2014 US (operating surplus and mixed income) EU28 (operating surplus and mixed income) US EU28 US (compensation of employees) EU28 (compensation of employees) Source: OECD. Wages are on the rise and unemployment is declining educated (European Commission 2015b).11 Given the impor- in most EU countries. Wages have been increasing in all tance of labor income for poverty reduction, these labor sub-regions during the last three quarters. Unemployment market patterns suggest that poverty is again declining in fell in all regions, responding very strongly to the moder- most EU countries. Moreover, there were around 2 million ate recovery, and reached 2007-08 levels in Germany, unfilled vacancies in the EU in June 2015, which compares Czech Republic, United Kingdom, Malta, Hungary and to 23 million unemployed. Filling these vacancies by enhanc- Poland FIG. 9. The biggest gaps in unemployment remain in ing labor mobility within the EU could further improve Southern Europe (Box 2) and Croatia. Unemployment fell living standards. for all education groups, but on average less for the less Q fig. 9 Unemployment rates are reaching pre-crisis levels in many EU countries, except in the South Seasonally adjusted unemployment rate 30 25 20 15 10 5 0 EL ES HR CY PT IT SK FR BG IE LV LT SI FI BE PL SE HU NL RO DK EE LU AT MT UK CZ DE US 2007-08 Q1/Q2 2015 Source: Eurostat; World Bank calculations. Long-term unemployment showed some timid signs of decline, falling from an EU-wide average of 5.1 percent in Q2 2014 to 4.7 percent in Q2 2015, but it remained high com- 11 pared to 2.8 percent in 2008. Long-term unemployment rates were particularly high in Greece (18.2 percent), Spain (11.8 percent), and Croatia (9.8 percent) as of Q2 2015. 14 r ecent economic de velopment s B OX 2 . U nemployment adj ustment to a GDP shoc k in Europe Growth shocks appear to have a larger negative Southern Europe. Yet, Southern Europe showed the most longer term impact on employment in Southern persistent increase in unemployment. In fact, after 5 years Europe. Estimates show that the initial response to a (20 quarters), the cumulative effect of a given negative one-percent negative real GDP growth shock is on average GDP growth shock was 50 percent higher in Southern than the highest in Western Europe and, initially, the smallest in in Western Europe.1/ Q fig. 10 The cumulative unemployment response to a GDP growth shock is largest in Southern Europe A. Change in unemployment rate (in percentage points) B. Cumulative change in unemployment rate 0,014 0,09 0,08 0,012 West South 0,07 0,01 Central 0,06 North 0,008 0,05 0,006 0,04 0,03 0,004 0,02 West South Central North 0,002 0,01 0 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Source: World Bank staff simulations. Several factors may explain this difference across labor market rigidities than Southern European states, regions. Theoretical and empirical literature on labor according to two key measures of labor markets flexibility: market institutions suggests that the initial response of hiring and firing practices, and redundancy costs.2/ Alterna- unemployment is strong in countries with high labor tively, it has been argued that if wages were set by bargaining market flexibility while the unemployment response is between insiders and firms, the equilibrium unemployment more persistent in countries with low labor market flexi- rate may increase after a shock (the so called hysteresis in bility. In fact, Western European countries have lower unemployment, see Blanchard and Summers, 1987). 1/ Estimates are based on a panel VAR with country fixed effects and quarterly data from Q1 1999 to Q2 2015 using the ‘Helmert procedure’ (Arellano and Bover, 1995). This procedure removes only the forward mean, i.e. the mean of all the future observations available for each country-quarter, and preserves the orthogonality between transformed variables and lagged regressors. Lagged regressors are used as instruments and coefficients estimated by system GMM. This methodology deals successfully with the Hurwicz-type bias associated to the use of fixed effects in a dynamic panel framework. 2/ Source: Global Competitiveness Report 2014-2015 and Doing Business 2015. 15 Youth unemployment is also declining but remains along with other improvements in public employment eu regul ar economic report very high in some countries along with high levels of services for young people as well as vocational training and inactivity, suggesting that the recovery alone is not education systems. Yet, youth unemployment remains enough. Europe’s youth faced the steepest surge in poverty very high in Greece, Spain, Croatia, Italy, Portugal, Cyprus, among all age groups in the wake of the 2008 global financial and Slovakia. Moreover, while youth unemployment is fall- crisis as unemployment in several EU member states ing rapidly in many countries the rate of young people soared FIG. 11A . Young people are often the first to be fired neither in education, employment or training (NEET) remains once a crisis hits, and also first to be hired in a recovery. In stubbornly high at 12.5 percent in 2014, compared to a peak fact, youth unemployment has started to decline rapidly in of 13.2 percent in 2012, requiring the implementation of most EU countries and in particular, in Croatia, Estonia, sound strategies in high NEET countries that focus on Ireland, Latvia, Lithuania and Spain FIG. 11B . The decline in retaining young people in formal education and training youth unemployment has been supported by EU-wide and enable them to acquire marketable skills. policy initiatives, such as the Youth Guarantee schemes Q fig. 11 Youth unemployment is declining in most EU countries Seasonally adjusted unemployment rate A. Change in youth poverty and youth unemployment B. Change in youth and total unemployment rates rates 2008-2012 (in percentage points) Q2 2013 – Q2 2015 (in percentage points) 40 4 IT FI Change in youth unemployment rate 35 2 EL AT 30 Changes in total unemployment rate ES 0 FR -6 -5 -4 -3 -2 -1 DE 0 1 2 25 NL -2 BE MT DK SE 20 RO IE -4 BG PT LT 15 SK IT LV LV LT EE PL CZ LU -6 10 CZ SI BG UK EE HU ES SK PL IE UK DK PT HU -8 5 FR RO SI CY FI SE BE NL AT LU EL -10 0 HR -10 -5 0 5 10 15 20 25 30 35 40 45 Changes in youth -5 Change in youth poverty rate -12 unemployment rate Source: Eurostat; World Bank staff calculations. / Note: Youth poverty is calculated as the percentage of young people with an income below the 60 percent median income as of 2008, measured in US$2011 PPP terms. 16 r ecent economic de velopment s Fisc a l con sol idat ion is slow i ng Governments have slowed fiscal consolidation as procedure). Both the headline and the structural deficits many EU countries seek to implement a more neutral (the latter adjusted for the economic cycle and for one-off fiscal policy stance. Saddled with high public debt, EU fiscal measures) changed little in the first half of 2015. The countries have reduced fiscal deficits sharply from a peak substantial fiscal adjustment during the last years masks of 9.7 percent of GDP in 2010 to 2.7 percent in 2014 FIG. 12A . considerable differences among EU countries. The Western The average deficit is now below 3 percent for the first time and Central European countries relied more on expendi- since 2008 (15 out of the 28 countries have deficits below ture reduction while countries in the Southern Europe the 3 percent limit but 9 are currently in an excess deficit focused largely on revenue increases FIG.12B . Q fig. 12 Significant fiscal adjustment has been achieved through different approaches Contribution to fiscal deficit reduction, 2010-2014, percent of GDP A. Structural versus cyclical adjustment B. Revenue versus expenditure adjustment 15 Structural Cyclical/one-off Deficit change 12 Rev Exp Reduction in deficit 10 10 8 6 5 4 0 2 0 -5 EU28 West South Central North -2 EU28 West South Central North     Source: EU Ameco; World Bank calculations. While most countries across the EU witnessed an in- tance spending, which is primarily designed to support crease in total social benefits during the crisis, social vulnerable families and individuals, actually declined in assistance spending actually declined in the Central most countries in Central and Southern Europe between and Southern Europe. Social benefits in percent of GDP 2008 and 2012 in real terms, reflecting sensitive political increased across the EU during the crisis. Yet, this increase economy considerations associated with reforms to pen- was to a significant extent driven by an upward trend in sion and disability systems and different types of social contribution-based social protection spending, in particu- protection systems (see Focus Note). This decline in lar, pensions that cannot easily be adjusted in times of crisis social assistance benefits has likely contributed to an and, in some EU countries, a declining GDP. Social assis- increase in poverty. 17 Despite reductions in fiscal deficits in recent years, of debt accumulation in the EU slowed markedly in 2014, eu regul ar economic report public debt levels remain elevated and are among the and flattened in the first quarter of 2015, reflecting better highest in the OECD, particularly in Southern Europe. fiscal positions of many countries and higher economic While the overall increase in indebtedness in 2010-2015 in growth. Nonetheless, the significant debt overhang will the EU was very similar to that in the US, there were sig- continue to constrain governments’ ability to relax fiscal nificant variations between countries. In particular, public policy over the long-term, and heightens their exposure to debt in the Southern Europe increased by 30 percentage increased costs from interest rate movements, slow points of GDP and its level is now well above 120 percent of growth and low inflation. GDP, driven significantly by large output declines. The pace Countries with the largest output gaps continue to have est. Combined with the modest recovery in much of the highest levels of public debt, limiting their ability to Southern Europe, this leaves few options but to continue use discretionary fiscal policy to boost growth. The to improve the efficiency of public spending, including of scope for fiscal stimulus is therefore constrained in many social protection systems, and implement reforms to bol- countries, particularly in Southern Europe, where output ster private-sector activity in order to boost growth and gaps are the largest and poverty increases were the steep- reduce debts to more manageable levels. 18 r ecent economic de velopment s C ou nt r ies cont i nue to i mprove bu si ness reg u lat ion s i n a n ef for t to bolster g row t h Structural reforms, in particular in Southern and Cen- De-regulation in the services sector significantly boosted tral Europe, have supported the growth recovery. Yet, total factor productivity of firms in the manufacturing and the pace of reform implementation has slowed. Business services sectors that use these services (World Bank 2015a, environment reforms12 and the implementation of the EC’s Box 3 ). Yet, despite empirical evidence that the impact of Services Directives have boosted labor productivity in structural reforms between 2008 and 2013 have helped affected sectors by 4-9 percent in Portugal, Spain, Italy and boost productivity and export performance, the pace of Greece, respectively (Varga, Werner and t’Vled 2013). reform implementation has recently slowed (ECB 2015b). Though EU countries rank high in terms of Doing compared to less than a year in Lithuania, Luxembourg and Business, there is ample room for further reforms to Sweden. A construction license is issued on average in the strengthen the EU’s long-term growth potential. In EU within 23 days. Yet, it takes more than 50 days in Belgium, 2015, one third of the top 20 countries in terms of ease of Croatia and Slovenia. It takes more than 400 hours in Bulgaria doing business were EU member states13 and all EU coun- and the Czech Republic to compile taxes, compared to 96 tries, except for Malta, Luxembourg and Greece, rank in hours in Luxembourg, Estonia, Ireland and Finland. Table 1 the top 50 of the World Bank’s Doing Business Ranking highlights how countries perform within the EU, and where (World Bank 2015b). Yet, differences between the best and reforms might help to boost countries competitiveness. worst scores vary significantly FIG. 13A . For example, it still Countries are also increasingly looking at the implemen- takes on average 590 days in the EU to enforce contracts, tation of reforms within a country (see Box 4 ). 12 According to the World Bank’s Doing Business (2015) Southern European countries improved their regulations and administration for enforcing contracts, minority investor protection, paying taxes and getting electricity. Regulations on insolvency and taxes were significant areas of improvement in the Central Europe. In contrast, Northern Euro- pean countries made it easier to start a business by simplifying pre-registration and registration formalities (Estonia, Sweden) or introducing more online procedures (Den- mark, Lithuania). Similar to the Southern Europe, but to a lesser extent, the Western Europe made broad-based improvements, notably in starting a business, protecting mi- nority investors and paying taxes. 13 Singapore and New Zealand remained at the top of the ranking, and the US remained seventh. The top-ranked EU member states in 2015 were Denmark (third), United King- dom (sixth) and Sweden (eighth). 19 Q fig. 13 Differences in the business environment in the EU vary significantly within and across regions and over time eu regul ar economic report Doing Business Scores and Distance to Frontier A. Overall ease of doing business, distance to frontier B. Contribution to change in doing business scores by (100 = best practice) country grouping 100 5 Average score West South Central North 95 4 Min. score 90 3 Max. score 85 2 80 1 75 70 0 Starting a business Dealing with construction Getting electricity Registering property Getting credit Minority investors protection Paying taxes Trading across borders Enforcing contracts Resolving insolvency 65 60 55 50 EU28 West South Central North Source: World Bank Doing Business. Q table  1 EASE OF DOING BUSINESS RANKINGS IN THE EU STARTING DEALING WITH GETTING REGISTERING GETTING PROTECTING PAYING TRADING ENFORCING RESOLVING A BUSINESS LICENSES ELECTRICITY PROPERTY CREDIT MINORITY TAXES ACROSS CONTRACTS INSOLVENCY INVESTORS BORDERS DENMARK 8 1 5 2 6 6 3 1 13 4 SWEDEN 7 5 2 5 18 3 9 16 14 11 UNITED KINGDOM 3 7 3 16 2 1 2 26 22 8 FINLAND 12 12 4 12 13 24 4 24 10 1 PORTUGAL 2 13 8 6 23 24 18 17 6 3 AUSTRIA 23 9 7 10 16 11 21 1 2 12 ESTONIA 9 4 17 4 6 26 8 20 9 19 LITHUANIA 1 6 18 1 6 15 12 19 5 26 GERMANY 26 3 1 22 6 16 19 23 3 2 NETHERLANDS 10 24 12 7 20 20 7 1 21 6 LATVIA 5 8 22 8 2 17 6 21 11 20 SLOVENIA 16 16 14 19 26 2 11 1 19 7 FRANCE 6 15 6 26 20 9 22 1 7 16 IRELAND 4 18 10 15 6 5 1 27 23 14 SLOVAK REPUBLIC 17 11 15 3 13 27 20 1 20 17 POLAND 27 14 13 17 2 18 14 1 16 18 SPAIN 21 25 24 13 16 9 13 1 12 15 CZECH REPUBLIC 20 26 11 14 6 19 27 1 25 10 BULGARIA 13 21 27 18 6 3 26 18 18 22 ROMANIA 11 20 28 20 1 21 17 14 17 21 BELGIUM 14 10 19 28 23 23 23 1 15 5 ITALY 18 17 21 9 23 13 28 1 24 13 HUNGARY 25 23 25 11 2 22 25 15 8 25 CROATIA 24 28 23 25 18 7 15 13 4 24 GREECE 15 19 16 27 20 14 16 22 28 23 CYPRUS 19 27 20 21 13 8 24 28 27 9 LUXEMBOURG 22 2 9 24 27 28 10 1 1 27 MALTA 28 22 26 23 28 12 5 25 26 28 Source: Doing Business 2016 database. / Note: The intensity of the color reflects the rank (out of 28) of a country for an indicator. The lower the number, the better the performance. Colors range from green (strong performance) to yellow (average performance) and red (low performance). 20 Labor market and product market reforms could the EU’s Single Market14 – notably by implementing fully r ecent economic de velopment s strengthen the EU recovery. In particular, further re- the EU Services Directive, easing financial fragmentation forms which enhance labor mobility within and across EU through the Capital Markets Union and enhancing labor member states and reduce labor market rigidities would be mobility and the deregulation of professions – as well as important. Moreover, a continued investment in high quality complementary measures to boost investment through general education and marketable skills remains impor- the so-called Juncker Plan to leverage private resources for tant. Continued product market reforms will be essential investment, including for SMEs. to support long-term growth. These include strengthening B OX 3 . Digital divide or dividend ? In October 2015, a ridesharing service arrived in petition and the efficiency of transport services. In San Croatia, which local taxi drivers vowed to fight. Francisco, the birthplace of ridesharing companies, taxi Many cities around the world require drivers to hold a use fell 65 percent between January 2012 and August 2014. license to operate a taxi; entering the market is often as- No wonder that many local taxi drivers feared income sociated with high fixed costs since licenses are issued losses. Yet, taxi companies in various cities responded by rarely and must be bought from current owners. In New enhancing the quality of their customer services to com- York City, costs for taxi licenses skyrocketed from about pete with the ridesharing services, such as developing $400,000 in 2004 to more than $1,100,000 in the begin- joint smartphone applications enabling online payment, ning of 2013. They started to decline only after the intro- rating and vehicle tracking in real time. duction of ridesharing services, which can enhance com- Service providers now rely on digital technology developing countries. The Estonian startup TransferWise that create significant opportunities for growth, and the U.S. startup Xoom match requests for interna- but that faces severe opposition in sectors most tional currency transfers online, saving direct and indirect protected from competition and innovation. transaction fees by clearing reciprocal currency transfer The internet has created new types of startups that base requests. The two startups reduce regulatory rents by their business model entirely on the web but offer tradi- reducing the prices of international currency transfers by tional services, such as retail trade, finance, transport, up to 90 percent. Postmates and Parcel provide local lo- logistics, tourism, media, publishing and advertising. gistic services in U.S. urban centers and have started to Airbnb, for example, operated in more than 40 countries compete with traditional service providers such as FedEx in 2014, enabling owners to let their homes for short-term but also with existing e-commerce platforms by matching rents, putting competitive pressure on the hotel and tour- customers demanding any type of locally available goods ism industry, which has frequently enjoyed high rents due with a pool of couriers. to local market segmentation or exclusive contracts in Digital technology can create unprecedented oppor- in highly protected sectors like retail and wholesale tunities for growth. They are often missed because trade, finance, transport or public utilities, that digital they are largest in sectors that are typically the most pro- technology can increase productivity and, consequently, tected from competition and innovation. Yet, it is precisely growth the most. Source: World Bank, World Development Report 2016 “Digital Dividends” (forthcoming) See EU Communication on upgrading the Single Market: http://europa.eu/rapid/press-release_IP-15-5909_en.htm 14 21 B OX 4 . T h e importance of local business regulation: eu regul ar economic report t h e case of P oland and Spain At the national level, Poland and Spain have been are determined by local authorities. Coordinating different reducing the complexity and cost of business regu- levels of government and institutions is essential to reduce lation and administration in line with global best the regulatory burden for companies. From an entrepre- practices. Since 2008, both countries have made consider- neur’s point of view, it is irrelevant whether a requirement able progress in reducing the regulatory and administrative comes from the municipality, the region or a national in- burdens faced by companies, moving closer toward the stitution. This is a particular problem for SMEs, who employ frontier of good practices in many areas, as measured by a significant number of people and are responsible for a the World Bank’s Doing Business survey. However, at the large share of net job creation in the EU. local level within each country, the picture is more nu- anced as many regulations and administrative measures In both in Poland and Spain, the regulatory envi- and a half months, while in larger cities, such as Krakow, ronment varies significantly at the subnational Poznań and Warsaw, more than six months. With the ex- level, suggesting the need for greater policy coordi- ception of Kielce, all Polish cities do better than average nation. No single city benchmarked (out of the 18 in Poland on at least one indicator. Similarly, all but two Spanish re- and 19 in Spain) does equally well in all Doing Business gions obtained an above-average score in at least one indicators (see Fig. 14 ). Starting a business in Seville, Spain, area, and all have at least one area where they rank below- for example, takes only 7 steps, but 12 in Pamplona. In Po- average, suggesting a large potential for inter-city learning. land, obtaining a construction permit in Opole takes four Q fig. 14 Large differences in regulatory practices depend on where SMEs locate their business. Procedures to deal Time to start a Cost to register property with construction permits business (number) (days) (% property value) All Poland 0.15-0.3— — 0.3 Switzerland — — 2.5 Portugal — — — 5.5 Italy — — 6 UK Polish best 8 — — — — — — 11.6 EU Avg. — Spanish best 3.1 — — Spanish best 14 — — 7 Sweden — 14.5 Germany — 8 France; Germany — 9 UK — Spanish best 10 — — 10 Italy — 11 Malta — 4.4 Italy; 4.5 EU Avg. — — — — — 12.6 EU Avg. — — 14 Portugal — 6.1 France — Spanish worst 30.5 — — — — Spanish worst 17 — — — 7.3 Portugal — — 35 Malta Polish best 19 — — — — — — Polish worst 42 — — Polish worst 22— — — — Spanish worst 10.1— — 10.1 Luxembourg Source: Doing Business 2015 data. 22 There are many reasons that explain differences Spain grapple with higher costs. Spain’s average cost is r ecent economic de velopment s between subnational regions. In Poland, variations more than twice that of the EU average for dealing with often result from the local interpretation and implemen- construction permits, and 75% higher for getting electricity tation of national laws. Spanish entrepreneurs face regula- and registering property. Poland typically has lower costs tory complexity and red tape as they interact with three than the EU average. While Spain’s average cost to transfer levels of government – national, regional and municipal property is nearly 8 percent of the property’s value, it’s – each with its own competencies and legislation. While only 0.32 percent in Poland. Similarly, to deal with con- entrepreneurs in both countries face regulatory complexity struction permits, entrepreneurs pay 5 percent of the and different levels of administrative efficiency, SMEs in warehouse value in Spain but only 0.22 percent in Poland. Subnational Doing Business surveys thus provide ease of doing business across different locations within more information on how government might sup- the same country may thus help drive reform, since it is port good practices within a country and provides a difficult for local governments to justify why doing busi- tool for specific locations to share their experiences. ness in their city or province is more burdensome than in Because cities in the same country operate under a com- neighboring locations. Uneven performance can guide mon legal framework, the good practices of the best local policymakers to areas where improvements are performing cities can usually be replicated, although this possible, and these administrative improvements can may require support and shared experiences from more make a big difference in the life of an SME. successful jurisdictions. Sharing comparable data on the Source: World Bank Subnational Doing Business. 23 chapter two O UT L O O K 26 T H E E U N E E D S TO T U R N TA I LW I N D S I N TO S U S TA I N E D G R O W T H 28 M E D I U M -T E R M R I S K S A R E S U B S TA N T I A L eu regul ar economic report T he EU need s to t u r n t a i lw i nd s i nto su st a i ned g row t h Continued easing of financial conditions, improved Investment is projected to pick up in 2016 as confidence confidence, low commodity prices and a reduced drag improves and financial conditions remain favorable fig. 15B . from fiscal consolidation will continue to support Yet, with slowing global demand and continuing downside growth in 2015 and 2016. The EU’s modest recovery risks, investment growth is expected to remain subdued. looks set to continue to strengthen with more balanced Residential construction is likely to increase on average growth between regions fig. 15A . Still, growth in the EU is as real disposable household incomes and house prices expected to remain modest at 2.0 percent in 2015 and continue to rise and mortgage rates remain low. Growth of 2.1 percent in 2016. Despite the moderation in activity in investment in non-residential construction and equipment China and the uncertain outlook for Greece, consumption is likely to remain sluggish as firm-level profits remain will continue to fuel growth as employment and wages grow. subdued and corporate deleveraging continues. Q fig. 15 GROWTH RATES ARE PROJECTED TO RISE ACROSS THE EU THOUGH INVESTMENT GROWTH REMAINS MODEST A. Real GDP growth 2015-2016 B. Total investment, Index 2008 = 100. 3,5 120 EU28 West South 2014 2015 F 2016 F 3 3,3 3,2 110 Central North U.S. 100 2,5 90 2 2,3 2,1 2,0 2,1 80 2,0 1,8 1,5 70 1,6 1,4 1 60 0,5 50 2008 2009 2010 2011 2012 2013 2014 2015 F 2016 F 0 EU28 West South Central North Source: Eurostat, World Bank staff estimates and projections. Fiscal policy is expected to remain broadly neutral, recovery gains strength across Europe, further structural as high debt levels will continue to restrict public reforms and favorable financial conditions are expected spending. Governments across Europe have significantly to increasingly support the consolidation effort. The pace lowered fiscal deficits over the past few years, with most of adjustment and its drag on overall economic growth is countries’ deficits expected to be below the formal expected to moderate. Maastricht criteria of 3 percent of GDP in 2015. As the 26 Inflation across the EU is likely to gradually rise as outlook stability objective of below but close to two percent in output and employment gaps narrow. The latest ECB 2017. Core inflation is likely to rise as output gaps close, survey of inflation forecasts suggests that EU inflation is while petroleum and commodity prices look set to remain likely to fall from 0.4 percent in 2014 to around zero in relatively subdued over the medium-term as global de- 2015, before very gradually rising toward the ECB’s price mand remains weak. The economic recovery is likely to ease poverty support the vulnerable and help families and individuals find rates across the EU (see Spotlight below). However, jobs as the economy recovers (see Focus Note). Poverty the rate at which poverty is reduced will depend on both forecasts for Bulgaria, Croatia, Poland and Romania suggest the speed and the depth of the recovery in labor markets as improvements in poverty rates between 2014 and 2017 well as the efficiency of the social protection systems that (see Spotlight on Central Europe). While the QE has provided a strong tailwind for spend their gains (Carroll, Slacalek and Tokuaoka, 2014). the recovery, it is unlikely to be enough to put Europe Only sustained improvements in labor markets will be able on a high, self-sustaining growth path. Financial net to consolidate consumption growth. While QE has signifi- wealth has increased, but it is unlikely to fuel a large and cantly helped improve financing conditions, investment sustained increase in consumption, as those who gained growth is expected to remain moderate. the most from the increase in net wealth are less likely to A significant structural reform agenda needs to be continuing to reduce labor market rigidities and promoting advanced to raise the productive potential of the EU. the skills needed for dynamic job creation and innovation Despite major reform efforts in recent years, significant (World Bank 2013); fully implementing the capital market impediments to growth remain. The EC estimates that union to help deal with the remaining financial market full implementation of the Services Directive could add fragmentation and to boost the availability of venture 1.8 percent to EU GDP (EC 2015). Similarly, an emphasis capital ; and implementing the EU’s Digital Single Market on improving the business environment at both the na- Strategy to improve access and reduce Internet costs for tional and subnational levels is welcome, as discussed in businesses and consumers and by encouraging more the earlier section (World Bank 2015c), although the EU efficient networks. This will need to be combined with could go further with the aim of making the EU the most affordable social policies that help to protect the most business-friendly global region over the medium term. vulnerable, while promoting greater social and labor Other high-priority structural reforms include the following: market inclusion. 27 eu regul ar economic report Med iu m-ter m r isk s a re subst a nt ia l Risks to the outlook for the EU are tilted to the and rebalance growth toward domestic demand, downside as divergent global trends increase or the imminent monetary policy tightening in the uncertainty over external demand. Key downside US, could slow the EU export growth. A protracted risks include: recovery could hamper market confidence, slow investment and adversely affect growth prospects (i) Structural rigidities: and financing costs; Continued weakness in the recovery of investment would constrain the sustainability of the EU’s growth (iii) An upsurge in financial-market volatility related over the medium term, once the impact of the current to the timing of U.S. interest rate rise, set in a context tailwinds have abated. Despite the ECB’s quantitative of divergent monetary policies in the EU and Japan, easing and favorable financial conditions, credit growth could trigger further swings in exchange rates and remains weak. Deleveraging pressures and a high capital markets and reduce the predictability of share of non-performing loans remain in some EU funding for investment; countries. Private sector investment remains sluggish. Although investment activity could be bolstered (iv) Regional geopolitical tensions also pose economic by the 315-billion-Euro European Fund for Strategic risks for the EU and could undermine confidence in Investment (EFSI), also known as the Juncker Plan, the recovery. A strong European commitment will which is intended to stimulate growth and create jobs, be important for building confidence. In particular, implementation arrangements are still being put in the recent refugee and migration flows calls for place and it may take considerable time before its common solutions (Box 5). impact is felt in many countries. Reducing structural rigidities in product, labor and capital markets would On the upside for the EU, improvements in efficiency help boost private-sector performance and investment; of public spending and further progress in the develop- ment of the banking union may help bolster the outlook (ii) Slower external demand resulting from a continued for growth over the medium term, as may low commodity slowdown in emerging markets, most notably China prices and the weak euro. as it seeks to restructure local government financing 28 B OX 5 . T H E EU A N D T H E R E F UGE E CR I SI S r ecent economic de velopment s The European Union is encountering an unprece- number of refugees then diminished, but shot up again dented influx of refugees and migrants as forced in the past couple of years. The vast majority of interna- displacements accelerated significantly in 2014, turning tional refugees remain in countries close to their own. it into the year with the highest number of internally Refugees equal less than 8 percent of the total of more displaced on record and the highest annual increase in a than 250 million international migrants, and 0.2 percent single year  (UNHCR 2014). Historically, the number of of the global population. The number of Syrian refugees refugees increased from the 1950s (after the refugees has reached 3.8 million, the largest in absolute numbers involved in the huge population transfers that followed currently, but less than the 5.6 million Afghan refugees World War II were resettled) until the peak of 17.8 million in 1989. persons following the collapse of the Soviet Union. The As a result of the larger influx of refugees, the number Yet, the Nordic and Baltic EU countries, which have the of people seeking asylum in Europe increased highest share of first-time asylum applicants, are about dramatically. In Q2 2015, the number of first-time asylum 0.2 percent of the total. Between April 2011 and October applicants reached 213 200. This was an increase of 85 2015, a total of 679,240 Syrian asylum applications were filed percent compared to the same quarter in 2014 of which in Europe, equal to 0.13 percent of the EU28 population at 69.7 percent was due to applications from Syrians, Afghans, the end of 2010. This number remains very small compared Albanians and Iraqis. The highest number of first-time asylum to Syrian refugees registered in neighboring countries. applicants in the second quarter of 2015 was registered The total number of registered refugees in Turkey, in Germany (38% of total applicants in the EU), Hungary Lebanon, and Jordan amount to 3.249 million, and make (15%), Austria (8%) and Italy, France and Sweden (about up 10.4 percent of their 2010 populations. The number 7% each). These six Member States together accounted for of Syrian refugees in Turkey, Lebanon and Jordan alone is more than 80% of all first applicants in the EU-28. In fact, eight times higher than those in all European countries Hungary saw its number of asylum seekers jump more (400,000 Syrian refugees in 2014). than 13 times compared to the same quarter of 2014. While the refugee crisis is likely to bring costs in most other countries that have received a large influx of the short term, there are potential benefits in the immigrants, Europe has the resources, the experience medium to long-term. In the short-term, this influx to and the capacity to turn the refugee influx not only into the EU is likely to put a strain on public services provided an opportunity for the refugees but also for domestic in some transit and destination countries and may lead economies. Economic inclusion including access to to a moderate increase in spending as a share of GDP in employment will facilitate, not only integration, but also, the most affected countries, notably for delivery of public ultimately, larger fiscal contributions of refugees. Policy services and integration support. In the medium-term, the responses should focus on helping to contain the crisis net fiscal impact and net economic benefit of immigration in the source country, alleviate capacity constraints in is likely to turn positive, in particular for rapidly ageing transit countries and develop monitoring mechanisms societies, provided efforts to integrate refugees into and policies and regulations to accelerate the integration domestic labor markets are successful. Compared to of migrants into domestic labor markets. 29 c h a p t e r t hr e e SPOTLIGHT ON C E N TRA L E UR O P E eu regul ar economic report s p o t l igh t o n c e n t r a l e ur o p e The Central European countries continue to enjoy which are all projected to grow 3 percent or more in some of the fastest growth rates in the EU, though 2015. The differing speeds of recovery, and particularly the expected pace of recovery, and thus poverty employment, will impact the rate of poverty reduction. reduction, varies widely. 15 Economic growth in the region is driven by domestic demand and continuing robust export growth, mainly due to demand from the rest of Q fig. 16 POVERTY IS PROJECTED TO DECLINE Poverty US $5/day (PPP) percent 2008-17 the EU. Consumption has generally been bolstered by 40 improving employment and wage growth, low inflation 35 from petroleum and food products, and easing credit 30 conditions. The acceleration in the absorption of EU 25 20 Structural and Cohesion Funds (before the deadline for 15 use of budgeted amounts for 2007–13) also boosted 10 growth in Central Europe in 2015, but the impact is ex- 5 pected to moderate in 2016 as the new budget funds 0 start to be spent. In most countries, unemployment rates 2008 2009 2010 2011 2012 2013 E 2014 E 2015 F 2016 F 2017 F have declined to their lowest levels since 2009. The pace Bulgaria Croatia Poland Romania of recovery is slower in Bulgaria and Croatia than in the Czech Republic, Poland, Romania and the Slovak Republic, Source: World Bank projections. Bulgaria’s recovery is projected to be modest, which to decline to 2.8 percent of GDP, following the increase to is likely to translate into a slow pace of poverty 5.8 percent in 2014, mainly due to one-off factors related reduction. Growth is projected at 2.9 and 2.2 percent in to the payout of guaranteed deposits by the Bulgaria De- 2015 and 2016 respectively. Growth in 2015 is likely to be posit Insurance Fund. The government is also expected to higher than initially expected, supported by strong exports further reduce the fiscal deficit by about 0.5 percentage to the EU in the first half of 2015, the recovery of invest- points of GDP per year in 2016 and 2017. Increasing child ments, mainly public ones from improved implementation benefits and heating subsidies for children and the elderly, of EU-funded projects, and better labor market perfor- as well as minimum wage increases and the Youth Guarantee mance. However, inflation was minus 1.2 percent in October Program to help youth in finding jobs are expected to 2015 with the annual rates having been negative since July contribute to poverty reduction. However, given the 2013. The substantial fiscal consolidation planned for 2015 modest pace of recovery, poverty – measured at US$5 per is likely to be achieved due to buoyant revenue performance day in PPP terms – is expected to remain above pre-2008 while some of the expected savings in spending are not levels and decline slowly to 14.1 percent in 2016. likely to materialize. The deficit (accrual basis) is projected This section looks at economic and poverty developments in the Central European countries, where the World Bank has historically had the strongest engagement. It provides 15 additional information on selected countries, in particular Bulgaria, Croatia, Poland and Romania, where the Bank is supporting efforts to bolster growth and shared prosperity. 32 In Croatia, economic activity is expected to recover s p o t l i g h t o n c e n t r a l e ur o p e and 9.0 percent in 2016. One of the most pressing problems at a very modest pace, after contracting for twelve is the high fiscal deficit, which averaged 6 percent over the quarters, with poverty projected to decline only last six years. The general government fiscal deficit was marginally. The economy is expected to grow by 1.5 5.6 percent in 2014, and is expected to remain high at percent in 2015, with the construction sector finally 4.9 percent in 2015, the highest level in the EU, as public growing, supported by a doubling of EU funds two years debt has more than doubled from 2008, to 85.7 percent after accession, together with a gradual rise in personal of GDP by August 2015. Over the medium-term, government consumption and exports. Unemployment remained high, consumption is projected to remain subdued due to a at about 17 percent in the first half of 2015 (four-quarter much-needed fiscal consolidation required under the EU moving average), despite active labor market policies and Excessive Deficit Procedure (EDP) that has set ambitious reduced labor shedding with slower corporate restructuring. targets to reduce the deficit. Although the European Council While real wages grew by 3.7 percent year-on-year in in June 2015 decided not to activate any corrective action the first nine months of 2015, real pensions declined by against Croatia for its large deficit, it pointed out that the 1 percent annually by September 2015, with adverse “level of ambition remains below expectations in a number poverty impacts on households dependent on pension of areas,” and has given six new recommendations that income. Real per-capita income remained flat in 2014 at the Croatian Government should meet in 2015 and 2016. 8 percent below its pre-crisis level. It is estimated that Credit growth also remains subdued, with high levels of the poverty rate, measured at $5/day PPP 2005, remained household (12.1 percent in September 2015) and corporate essentially unchanged in 2014 at 9.8 percent. In the absence non-performing loans (31.1 percent) and public debt ham- of stronger private employment and wage growth, poverty pering the financial sector. is projected to decline only marginally, to 9.3 percent in 2015 In contrast to the previous two countries, growth medium term, with negligible internal and external imbal- in Poland is projected to reach 3.5 percent for ances. Domestic demand is likely to remain the main 2015 and remain above 3.5 percent over the medium driver of growth amid robust private consumption (around term, supporting a pickup in poverty reduction. 3.5 percent), as employment and wages continue to rise, Private consumption in the first half of 2015 continued to and solid investment growth. Having exited the EU Excessive benefit from higher real disposable incomes as a result of Deficit Procedure in 2015, a year ahead of schedule, Poland is improved labor market conditions as the unemployment projected to decrease its headline deficit modestly, from 3.2 rate declined by almost 2 percentage points over the first percent in 2014 to 3.0 percent of GDP in 2015, before increasing half of the year, to 7.4 percent in the 2nd quarter of 2015. to 3.2 percent of GDP in 2016. Social expenditures are poised It also benefited from relatively strong credit growth to to rise with the introduction of new social benefits and households, from consumer price declines and from higher programs supporting families with children, the elderly wages (private sector wages have been growing at 2.5 per- or more generous pensions’ indexation resulting from cent and pensions at 2.3 percent). Investment was supported commitments made during the current election year and by solid corporate profits, growing confidence, record low the budget will be amended to reflect these changes. interest rates and final disbursements from the EU’s previous Underpinned by these developments, poverty, measured financial perspective (budget). Since February 2013, headline by the US$5/day in 2005 PPP, is expected to decline at a inflation has been below the Central Bank’s 1.5-3.5 percent faster pace than recent years and reach 4.6 percent in target range and since July 2014, consumer prices have fallen 2015, down from an estimated 4.8 percent in 2014, and to as energy and food prices declined. Poland’s economy fall to 4.2 percent by 2017. is expected to grow at a rate above 3.5 percent over the 33 Romania’s economy is expected to grow at 3.6 percent eu regul ar economic report by the Parliament in September 2015 envisages several in 2015 and 3.9 percent in 2016, contributing to tax cuts, starting in January 2016, which could amount to further declines in poverty. Economic growth has been 1.1 percent of GDP, according to the Fiscal Council, in- led by a strong rise in private consumption, which reached cluding a cut in VAT rates from 24 percent to 20 percent. an annual growth rate of 5.4 percent in the first half of Poverty is projected to have declined from 35.8 percent 2015 on the back of increases in wages and a decline in in 2011 to 28.4 percent in 2014, using the US$5.00/day PPP the VAT rate for food from 24 percent to 9 percent in June poverty line, driven by improved labor market conditions 2015, which contributed to inflation falling to negative 1.9 and increased support to vulnerable categories. These percent in August 2015. The construction and industrial include increased allocations to the minimum guaranteed sectors have recovered strongly, but public investment income, to family benefit and heating benefit programs, has remained subdued, reflecting lower-than-expected as well as increases in the minimum wage. Continued absorption of EU funds. Despite the positive developments, recovery of domestic demand and growth in employment growth for 2015 is projected to slow slightly to 3.6 percent and real wages, aided by low inflation should boost real for the entire year, due in part to a drought-induced decline incomes and lead to further declines in poverty incidence. in agricultural production. GDP growth, however, is expected The US$5.00/day PPP poverty rate is projected to decline to rise to 3.9 percent in 2016, supported by strong domestic to 22.1 percent in 2017. consumption and investment. The Fiscal Code approved 34 F O CU S N O T E w e l fa r e s t a t e s 16 a n d protection of the poor in the eu rope a n u n ion 38 EU MEMBER STATES ARE A MONG THE BIGGEST SPENDER S ON SOCIAL PROTEC TION IN THE WORLD… 42 …HOWE VER NOT ALL COUNTRIES PROVIDE EFFECTIVE PROTEC TION FOR THE POOR 46 MOST WELFARE STATES CUT SOCIAL A S SI STANCE SPENDING DURING THE CRI SI S… 48 …WITH DIFFERING OUTCOMES FOR THE POOR 51 EU MEMBER STATES WOULD NEED TO PROVIDE ADEQUATE PROTEC TION AND SOCIAL INVESTMENT FOR THE POOREST WHILE EN SURING FI SC AL SU STAINABILIT Y eu regul ar economic report EU member st ates a re a mong t he big gest spender s on socia l protec t ion i n t he world … How well positioned are European countries to high in several EU economies. Youth unemployment has protect the poor in the context of Europe’s fragile been found to significantly depress future earnings; long economic recovery? While the European Union’s growth spells of unemployment are likely to affect the minimum recovery is gaining strength, poverty across EU still remains contribution periods needed to become eligible for retire- high, especially in Southern Europe (see Recent Economic ment pensions, increasing the likelihood of future old-age Developments). Higher incidence of poverty, especially poverty. Risks to the outlook for the EU remain tilted to the among the young, undermines education outcomes and downside, raising the question how fiscally constrained bodes ill for future labor market prospects, with adverse governments can provide an effective protection. To address consequences for long-term growth. While unemployment these challenges, EU member states need to focus on rates have declined to their pre-crisis level in most EU providing well-targeted and adequate protection to the countries, the youth unemployment and inactivity remain poorest most vulnerable population. Seventeen out of 28 EU member states spend the working age populations are projected to decline, more than the OECD average on social protection continued long-term growth requires investment in children, expenditures fig. 17. A large proportion of this expenditure youth, and otherwise weaker segments of society in order is on social insurance (see Box 6 ), and in particular on old to fully maximize their productive potential. This highlights age pensions, in part reflecting the high median age of the importance of robust social protection systems. its populations in many countries. As the relative size of Social assistance expenditure constitutes a small a high of 33 percent in Denmark and 35 percent in Sweden. fraction of overall social protection spending but Although expenditures are small when compared to the has a particularly important role in protecting and social insurance pillar, social assistance programs are very investing in the poor. Expenditure on social assistance important in terms of cushioning the chronic poor and programs range from a low of about 9 percent of overall the most vulnerable against potential risks across their social protection expenditures in Poland and Portugal, to lifecycles. Last resort social assistance programs help 16 The welfare state is usually defined as consisting of cash benefits like social insurance, social assistance and labor policies, and in-kind benefits like health, education, housing and other social services (Barr, 2012). For the purposes of this note, the welfare state is defined more narrowly with a primary emphasis on its ability, through non- contributory social assistance programs, to protect the poor. 36 welfare states and protection of the poor in the european union alleviate chronic poverty and potentially activate those children; disability benefits support the disabled that that are able to work, enabling for investment in human may be excluded from the labor market, and can help capital; child benefits protect and invest in families with connect others back to jobs. Q fig. 17 EU MEMBER STATES SPEND ON AVERAGE MORE ON SOCIAL PROTECTION THAN OECD COUNTRIES Social protection spending, Europe and comparators as a share of GDP in 2012 35 30 25 21,6 20 15 10 5 0 France Belgium Denmark Finland Italy Austria Sweden Spain Greece Germany Portugal Netherlands Slovenia Luxembourg Japan United Kingdom Hungary Ireland Norway OECD - Total New Zealand Czech Republic Poland Switzerland United States Australia Slovak Republic Iceland Canada Estonia Israel Turkey Chile Korea Mexico Source : OECD. / Note : The definition of social protection expenditure differs across various sources. The data for this graph derive from the OECD SOCX database. The rest of the note uses Social Protection data (including breakdown into individual components) from ESSPROS with 2012 as the most recent data available, with adjustments as defined in Annex 1. B OX 6 . S ocial Protection F unctions Social protection can be divided into three main pillars: poverty and investing in people to promote their escape from chronic poverty/vulnerability. They can include 1. Social insurance aims to cushion workers in the formal guaranteed minimum income programs (which are labor market against risks of sickness, disability, or old- generally captured in Eurostat surveys under the “social age poverty. It is mostly financed through contributions. exclusion not elsewhere classified” function, see below), Old-age pensions are often the largest expenditure housing benefits, heating benefits or child or family element in the social insurance pillar; and the goal of benefits. Based on the objective of the program they old-age pensions is to provide some form of replace- can be categorical (e.g. disability benefits) and/or means- ment of labor income when a worker retires. tested (e.g. guaranteed minimum income programs). 2. Social assistance primarily aims to support the poor 3. Labor market policies play both an insurance role and vulnerable families or individuals and is financed (in the case of passive measures such as unemployment from the general government budget. The social assis- benefits) and an active role to facilitate (re-)entrance tance pillar has the dual objectives of protecting the into the work force through employment services, poor and vulnerable from shocks, but also tackling wage subsidies and public works programs. 37 There is no single European welfare state. Social eu regul ar economic report but high social assistance coverage; and limited welfare protection systems across the EU are classified using two states with low social protection spending and low social key dimensions that are intermediate outcomes of the assistance coverage. These groups have fairly distinct char- underlying social contract between the state and the acteristics as described in (Box 7) and analyzed in this note. citizen. The first dimension is the social protection spending as a share of GDP, which is a proxy for the overall Q fig. 18 THERE ARE FOUR DISTINCT EUROPEAN WELFARE STATES resources a society allocates to the provision of social Social protection spending and social assistance coverage, 2012 protection. Yet, it says little about how willing a society is 100 Small balanced MT Large balanced to protect its poorest citizens. The second dimension is welfare states IE NL welfare states 90 LU Social assistance coverage of the bottom 20 the coverage of the poorest 17 by overall social assistance SK HU UK FR FI 80 which is a proxy for the effectiveness of allocated resources SI SE DK LV or the types of instruments that can be potentially leveraged 70 LT AT RO BE to tackle chronic poverty and protect the poor in a crisis 18 60 CY BG PL (see Annex 1 for a definition of coverage). Countries that EE HR PT 50 CZ spend more than 16 percent of GDP in 2012 on social pro- IT tection are classified as high spenders 19; and countries 40 EL that cover more than 60 percent of the poorest 20 percent 30 of the population as countries with high social assistance ES 20 Limited Truncated coverage fig. 18. On this basis, countries can be divided into 4 welfare states welfare states 10 categories: Large balanced welfare states with high social 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Social protection spending as share of GDP protection spending and high social assistance coverage; truncated welfare states  20 with high social protection Source: World Bank staff calculations using ESSPROS 2012 and EU-SILC 2013. / spending, but low social assistance coverage; small bal- Note: Labor market spending data for 2012 for Greece are from 2010, for Cyprus anced welfare states with low social protection spending, and the UK are from 2011. Germany is not ranked, since we do not have coverage data for Germany. 17 For the purposes of this note we assume that the poorest 20 percent of the population (bottom quintile) are representative of the most vulnerable segment of the population in each country. In fact, the relative at risk poverty rates in 2012 (from SILC 2013, latest year data is available) range between 9 to 21 percent with 16 percent as the simple average for the EU. 18 When well-targeted and well-functioning programs with a reasonably high coverage of the poor already exist, countries are better placed to leverage such programs to respond quickly and efficiently to a crisis. In contrast, countries with smaller coverage and without an effective poverty targeted program often have had to rely on less efficient instruments (such as providing subsidies or using pensions as the main crisis response instrument), which may make crisis response prohibitively expensive and leave the population with little protection. 19 The median (and mean) expenditure on SP among all member states in 2004 was 16 percent and is used as the cutoff point for the typology. 20 As outlined in Fizsbein (2004) truncated welfare state refers to systems where public social protection is offered to a certain group (in most cases formal sector workers) and the protection drops sharply at the lower parts of the income distribution. 38 B OX 7. T Y P OL O G Y OF S O CI A L PRO T E CT ION S YST E M S welfare states and protection of the poor in the european union Large balanced welfare states are characterized by Large balanced welfare state Small balanced welfare state Truncated welfare state Limited welfare state high social protection spending that goes hand in hand Large balanced welfare state Small balanced welfare state with high social assistance coverage of the poorest Truncated quintile welfare state Limited welfare state and includes most Western European countries – Austria, Belgium, Cyprus, Denmark, France, Finland, Hun- gary, Ireland, Luxembourg, the Netherlands, Slo- venia, Sweden, and the United Kingdom. Averaging 21 percent of GDP in 2012, their social protection spending is relatively high. They have a relatively more balanced breakdown between social assistance and social insurance spending, enabling them to offer protection and income replacement (upon retirement) to formal sector workers while also providing safety nets for the poor and vulnerable. These countries are able to cover a significant portion of the bottom quintile (around 80 percent on average) through their social assistance programs. The social protection system includes a more balanced mix of programs with basic elements that buffer against different risks (e.g., risk Large balanced welfare state designed Small balanced welfare state to protect vulnerable groups, such as the young Truncated welfare state Limited welfare state of unemployment, of disability, of poverty). It also includes and the poorer segments, of society against risks. instruments to connect people back to opportunities, such as providing job search assistance, counselling, social Small balanced welfare states are characterized by services, and so on. It is important to note that large low social protection spending, but high social assistance balanced welfare states are expensive. Fiscal sustainability coverage of the bottom quintile. Latvia, Lithuania, of the system provides the key risk in terms of whether Romania, Slovakia, and the island of Malta , fall such countries can continue to provide protection during under this group. Most have reformed (and continue to future crisis episodes. A fall in GDP could necessitate cuts reform) their social protection systems since the 1990s in expenditure, and the quality of fiscal consolidation transition and have steered them to a relatively more will determine if these states are able to become leaner, balanced approach which offers a mix of programs that but still maintain protection of the vulnerable. cover various risks and covers a large share of the poor- est population mostly through categorical/universal Truncated welfare states are characterized by low programs but still at relatively low costs. coverage of the poorest quintile by social assistance despite high levels of social spending. The Southern European Limited welfare states are characterized by low social countries, Greece, Portugal, Spain, and Italy, fall in this protection spending and low coverage of the poorest group. Their overall social protection spending around quintile by social assistance. Bulgaria, Croatia, Czech 21 percent of GDP is similar to that of large balanced Republic, Poland, and Estonia fall in this group. Similar welfare states. Yet, a relatively large share of their spending to the small balanced welfare states, these countries is dedicated to social insurance. This system thus focuses started to reform the social protection systems they had on covering formal sector workers and managing the inherited from socialist times. Yet, they chose a different lifecycle risk of old age poverty. Overall social assistance direction by limiting the protection offered by the public programs, or indeed coverage, of the non-formal sector, sector, resulting in minimal protection of the poorest is not well-developed. The system is therefore not well segments against various shocks. 39 eu regul ar economic report …not a l l cou nt r ies prov ide ef fec t ive protec t ion for t he poor Higher spending on social assistance does not auto- yet they achieve much better coverage of the bottom 20. matically mean higher coverage. Small balanced welfare Even limited welfare states, which spend the least on social states spend less on social assistance than truncated welfare assistance (1.7 percent), manage to cover a larger share states (2.2 percent vs 2.9 percent of GDP, respectively) fig. 19, of the poor at 56 percent fig. 20. Q fig. 19 SOCIAL ASSISTANCE SPENDING VARIES WIDELY ACROSS EUROPEAN WELFARE STATES… Social assistance spending as a share of GDP in 2012 10 9 8 7 6 5,7 % GDP 5 4 2,9 3 2,2 2 1,7 1 0 Average Average Average Average MT HU RO DK HR UK AT CY DE BG NL LV LU PT CZ BE SK FR EE LT SE EL ES PL IT IE FI SI Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states Source: World Bank staff estimates using ESSPROS Q fig. 20 ...with varying implications for the coverage of the bottom 40 percent Social assistance coverage of the bottom 40 percent in 2012 100 90 80,4 78,1 80 70 60 55,5 50 39,5 40 30 20 10 0 Average Average Average Average MT HU RO UK DK HR AT NL CY BG LU LV PT CZ BE SK FR EE LT SE PL EL ES IT IE FI SI Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states Source: World Bank staff estimates using EU-SILC. / Note: Germany is not ranked, since we do not have coverage data for Germany. 40 The composition of spending matters; large and small welfare states and protection of the poor in the european union due to their universal nature. Large balanced welfare states balanced welfare states achieve higher coverage also spend more than 1 percent of GDP on means-tested through a combination of family benefits (which housing and social exclusion benefits; however this is are mostly universal) and “social exclusion n.e.c.” much smaller (0.3 percent of GDP) among small balanced programs (which are mostly means-tested) 21. Most welfare states. Both large and small balanced welfare welfare states spend a larger proportion of social assistance states achieve high coverage of the bottom 20 through expenditures on categorical programs than on means-tested the social exclusion benefits (22 percent and 40 percent, programs fig. 21 . The largest spending sub-category across respectively fig. 22). Coverage of the bottom 20 through all programs is the universal family/child benefits in both family benefits is also high in the large (61 percent) and large balanced (1.76 percent of GDP) and small balanced small (59 percent) balanced welfare states. (0.68 percent of GDP) welfare states, which are expensive Q fig. 21 CATEGORICAL PROGRAMS DOMINATE SOCIAL ASSISTANCE SPENDING ACROSS EUROPEAN WELFARE STATES Means-tested vs. categorical spending, by program in percent GDP in 2012 4,0 0,16 3,5 3,0 1,76 2,5 2,0 0,54 0,13 1,5 0,97 0,43 0,18 0,68 0,51 1,0 0,30 0,63 0,12 0,80 0,58 0,42 0,36 0,5 0,30 0,90 0,27 0,33 0,38 0,12 0,33 0,36 0,38 0,0 0,15 0,08 0,06 0,06 Large balanced Truncated Small balanced Limited Large balanced Truncated Small balanced Limited Means-tested Categorical Disability Old-age Family/children Housing Social exclusion n.e.c. Source: World Bank staff estimates using ESSPROS. The quality of means-testing and who is targeted national poverty targeted program, and this significantly matter for coverage. Even though truncated welfare reduces their ability to cover the poor. For instance, Italy states spend more than 1 percent of GDP on family benefits spends 4.2 percent of GDP on social assistance and only (combined between means-tested and universal programs), spends 0.1 percent of GDP on means-tested social exclusion they achieve much lower coverage compared to lower benefits. In contrast, Slovakia spends 2.9 percent of GDP spending small balanced and even limited welfare states on social assistance, of which 0.4 percent goes to the fig. 21 & 22 . This raises questions about the types of means-tests main poverty focused program, Benefit in Material Need. used to target various benefits. Similarly, truncated welfare In addition, the disability benefit is also income-tested. states spend 0.18 percent of GDP on means-tested social Consequently, Slovakia covers more than 80 percent of exclusion benefits, they achieve even lower coverage the poorest through its social assistance system whereas (10 percent) than limited welfare states, which spend a Italy covers less than 50 percent fig. 20. meagre 0.12 percent of GDP. Both Greece and Italy lack a 21 Social exclusion not elsewhere classified (n.e.c) includes programs that are not classified under the non-contributory old age and disability benefits, family benefits, and housing benefits. They mostly include guaranteed minimum income programs and other types of last resort social assistance programs, and are mostly targeted towards the poor. 41 eu regul ar economic report Q fig. 22 HIGHER COVERAGE OF BOTTOM 20 PERCENT IS ACHIEVED MOSTLY THROUGH FAMILY BENEFITS AND SOCIAL EXCLUSION PROGRAMS Social assistance coverage of the bottom 20 PERCENT by program in 2012 80,4 78,1 80,4 78,1 61,2 58,8 61,2 58,8 55,5 55,5 48,5 48,5 39,5 40,1 34,3 39,5 40,1 34,3 29,8 29,8 22,0 22,0 16,6 9,9 16,6 12,2 9,9 7,1 12,2 3,2 7,1 3,2 Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states Overall social assistance Family benefits Housing Social exclusion not elsewhere classified Overall social assistance Family benefits Housing Social exclusion not elsewhere classified Source: World Bank staff estimates using EU-SILC. / Note: Germany is not ranked, since we do not have coverage data for Germany. Leakage of social assistance benefits to the rich is similar pattern – covering 67 percent of the poor, but 44 significant in several EU member states. Containing percent of the rich. By contrast, the United Kingdom is leakage could provide fiscal savings. Universal family effective in containing leakage to the rich – 86 percent benefits lead to good coverage of the poorest in large of the poor in the UK receive a social assistance transfer; and small balanced welfare states. At the same time, only 20 percent of the rich get such a transfer. Likewise, many in the richest 20 percent of the population also Poland minimizes leakage– only 8 percent of the rich get receive social assistance benefits in these countries any social assistance. However, Poland achieves this at fig. 23 . Much of the coverage of the rich comes from family the cost of also excluding some of the poorest, with only benefits fig. 24 22 . The leakage to the rich is particularly 57 percent of the poor covered by social assistance. Better acute in some Baltic states – for instance, Estonia covers targeting of social assistance, especially categorical programs 56 percent of the poorest 20 percent of the population such as family benefits, would free up resources, which with social assistance, but also 51 percent of the richest can then be channeled into savings or investments in 20 percent. Similarly, Latvia covers 70 percent of the countries with high coverage of the poor, or redirected poorest, but also 48 percent of the richest. Some large to increase coverage of the poorest in countries with balanced welfare states, such as Belgium, also show a insufficient protection of the poor. Q fig. 23 SOCIAL ASSISTANCE LEAKAGE TO THE RICH IS SUBSTANTIAL IN MANY EUROPEAN WELFARE STATES Social assistance coverage of (leakage to) the richest 20 percent in 2012 70 70 60 60 50 50 36,7 38,8 40 36,7 38,8 40 30 30 19,9 20 19,9 20 10,4 10 10,4 10 0 Average Average Average Average MTMT HUHU RORO DK DK HRHR UK UK AT AT CY CY BG BG NL NL 0 LU LU LV LV PT PT CZ CZ BE BE SK SK FR FR EE EE LT LT PL PL SE SE EL EL ES ES IT IT IE IE FI FI SI SI Average Average Average Average Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states states Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states states Source: World Bank staff estimates using EU-SILC. / Note: Germany is not ranked, since we do not have coverage data for Germany. 22 Universal family benefits may have an additional social policy objective of increasing fertility. However, there’s mixed evidence on pro-natalist policies on fertility and labor force participation rate of women. Purely financial measures are 42 welfare states and protection of the poor in the european union Q fig. 24 LEAKAGE TO THE RICH IS MAINLY DUE TO FAMILY BENEFITS Social assistance coverage of (leakage to) the richest 20 percent by program in 2012 38,8 36,7 38,8 36,7 32,3 32,3 30,3 30,3 19,9 19,9 17,3 17,3 10,4 10,4 6,6 6,9 6,6 6,9 2,3 2,6 2,3 0,9 2,6 0,7 0,7 0,8 0,9 0,7 0,7 0,1 0,8 0,1 Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states Large balanced welfare states Truncated welfare states Small balanced welfare states Limited welfare states Overall social assistance Family benefits Housing Social exclusion not elsewhere classified Overall social assistance Family benefits Housing Social exclusion not elsewhere classified Source: World Bank staff estimates using EU-SILC. / Note: Germany is not ranked, since we do not have coverage data for Germany. In addition to coverage, the adequacy of transfers is assistance, have had the greatest success in reducing poverty important to lift people out of poverty; yet not every through social transfers 23, reducing the poverty rate from 22 country can afford high coverage and reasonable ade- percent to 14 percent through social assistance transfers, an quacy. The adequacy of benefits, measured as the share of impressive 8 percentage point reduction fig. 25. Spending less, transfers relative to the total disposable income of the ben- the small balanced welfare states achieve high coverage, but at eficiary households, is a complementary dimension of the lower adequacy levels fig. 26. Therefore, they are not able to pull extent of protection offered to the poor. Large balanced as many people out of poverty. Still, they obtained a 4 per- welfare states, which also have the highest adequacy of social centage point reduction in the “at risk of poverty rate” in 2012. Q fig. 25 LARGE BALANCED WELFARE STATES ARE THE MOST SUCCESSFUL IN POVERTY ALLEVIATION Q fig. 26 ONLY FEW COUNTRIES CAN AFFORD HIGH COVERAGE AND REASONABLE ADEQUACY THROUGH SOCIAL ASSISTANCE Coverage and adequacy of overall social assistance for the At risk of poverty -before and after transfers, 2012) bottom 20, 2012) 25 80,4 80,4 25 78,1 78,1 22 22 22 22 22 22 20 20 20 20 20 20 18 18 17 55,5 17 55,5 15 14 15 14 38,7 39,5 39,5 38,7 10 28,0 10 27,8 27,8 28,0 22,5 22,5 5 5 0 0 Large balanced Large balanced Truncated welfare Truncated Small balanced Limited welfare Large Large balanced balanced Truncated Truncated welfare welfare Small Small balanced balanced Limited welfare Limited welfare welfare states stateswelfare Small welfarebalanced states Limited welfare states welfare states welfare states states states welfare welfare states states states states welfare states states welfare states states Before social assistance After social assistance Coverage Coverage Adequacy Adequacy Before social assistance After social assistance Source: World Bank staff calculations using ESSPROS. Source: World Bank staff calculations using EU-SILC 2013. Note: Germany is not ranked, since we do not have generosity data for Germany 23 Social assistance programs are typically not designed to fully lift families out of poverty – their purpose is to mitigate poverty and connect people to opportunities that can lift them out of poverty. A transfer given to an extremely poor family will enhance their welfare, and reduce the poverty gap. Transfers given to families closer to the poverty line are more likely to have an impact in terms of fully lifting them out of poverty and therefore reducing the poverty headcount. 43 eu regul ar economic report Most wel fa re st ates cut socia l a ssist a nce spend i ng du r i ng t he cr isis… Discretionary policy choices did not allow social increased across all social protection systems between assistance programs to play the role of “automatic 2004 and 2008, particularly in the small balanced welfare stabilizers” over the economic cycle. It is expected that states. Only 4 countries cut social assistance spending expenditure and coverage of social assistance programs during the boom period, these were the United Kingdom, will automatically expand during a downturn as more Germany, France, and Slovenia. This changed during the people become poor, and contract as the economy crisis years when most European countries entered a recovers. In other words, social assistance programs that phase of fiscal consolidation. Social assistance spending are responsive to shocks and the resulting economic as a share of 2008 GDP fell on average in all social protec- needs are often counter-cyclical. However, contrary to tion systems but the large balanced welfare states expectations, real social assistance spending in the EU between 2008 and 2012. The latter even expanded social over the last decade was pro-cyclical in all but the large assistance spending after 2008 though their social assis- balanced welfare states. Due to policy changes, average tance spending was already on average far higher fig. 27B . real social assistance spending in terms of 2008 GDP 24 Q fig. 27 SOCIAL ASSISTANCE SPENDING FELL IN SEVERAL COUNTRIES DURING THE RECESSION WHILE SOCIAL INSURANCE SPENDING WAS PROTECTED Change in Real Social Insurance and Social Assistance Spending in 2004-2012 A. Change in Social Insurance Spending B. Change in Social Assistance Spending (as share of 2008 GDP) (as share of 2008 GDP) 3,5 14 0,8 6 3 12 5 0,6 Percentage point change 2,5 10 Percentage point change 4 0,4 2,0 8 3 1,5 6 0,2 2 1 4 0 0,5 2 1 0 0 -0,2 0 Large balanced Truncated Small balanced Limited welfare Large balanced Truncated Small balanced Limited welfare welfare states welfare states welfare states states -0,4 welfare states welfare states welfare states states -1 2004-08 2008-12 SI spending in 2008 as % of GDP (right axis) 2004-08 2008-12 SA spending in 2008 as % of GDP (right axis) Source : EU-SILC. / Note : Germany is not ranked, since we do not have generosity data for Germany with adjustments as defined in Annex 1. 24 After the crisis in 2008/2009, many EU countries suffered a decline in GDP. Spending, measured as a share of GDP, may therefore no longer capture the real spending. Real spending (in constant 2005 Euros) is anchored at the share of 2008 GDP – the peak GDP for many countries in recent years – to compare the actual spending over time while taking into account the country differences on the level of the spending. 44 Expenditure on social assistance fell between welfare states and protection of the poor in the european union Czech Republic, i.e. Bulgaria, Croatia, Poland, and Estonia 2008 and 2012 in 11 EU member states, including FIG. 28 . Thus, only 3 countries exhibited a true countercyclical in every truncated welfare state. Eleven countries pattern, with expenditures declining during the boom experienced a decrease in social assistance expenditure period and increasing in the recession –the United King- during the recession period, (a) Slovenia among the large dom, Germany, and France. By contrast, every single trun- balanced welfare states; (b) all 4 countries in the truncated cated state cut social assistance spending between 2008 welfare states – Greece, Spain, Italy, and Portugal; (c) Ro- and 2012; the largest cut was in Greece, where expendi- mania and Latvia among the small balanced welfare states; tures declined by 0.77 percentage points in the midst of a and (c) all countries in the limited state except for the severe recession, from an already low base. Q fig. 28 THE DECLINE IN SOCIAL ASSISTANCE SPENDING WAS MOST COMMON IN TRUNCATED AND LIMITED WELFARE STATES 25 Social assistance spending change as percent of 2008 GDP 2 2 1 Percentage point change 1 0 DK SE FI CY LU UK HU NL AT BE DE IE FR SI IT EL ES PT SK MT LT RO LV BG CZ HR PL EE -1 -1 Large balanced Truncated Small balanced Limited welfare welfare states welfare states welfare states states -2 2004-2008 2008-2012 Source: World Bank staff estimates using ESSPROS By contrast, expenditure on social insurance were cuts in pensions were reversed by the Constitutional Court protected during the recession between 2008 and reflecting the difficulty of cuts in social insurance pro- 2012. Social Insurance expenditures are financed largely tection spending during periods of fiscal consolidation. through contributions, and are determined by longer term In fact, real spending on social insurance increased as a factors such as demography, retirement ages, and stage share of 2008 GDP on average across all four groups fig. 27A . of development of the private pension systems. In several The increase was smallest in the limited welfare states. 26 European countries (Romania, and Lithuania, for instance) 25 The real spending is measured in constant 2005 Euros. Exchange rate changes during this time explain a significant part of the changes in spending in EU member states with flexible exchange rates, in particular in Romania. 26 Therefore, it is important, during periods of economic expansion, to carefully consider expansions to pensions and social insurance programs as these expenditures are often difficult to cut during a period of recession or slow economic growth. 45 eu regul ar economic report …w it h d i f fer i ng outcomes for t he poor Cuts in social assistance expenditures resulted in in 2012 is just 40 percent of expenditure in 2004. As a result, lower coverage for the limited welfare states. In the both the coverage and the adequacy of social exclusion limited welfare states, consolidation (cut) of social assistance benefits to the poorest 20 percent of the population expenditures resulted in lower social assistance coverage declined (ECA SPeeD). In Poland, the number of recipients of the poorest quintile, with coverage falling from 61 percent of means tested family benefits declined sharply, from in 2008 to 56 percent in 2012 fig. 29. This is due to a decline 3.8 million in 2008 to about 2.3 million in 2013—the drop in the coverage of all types of social assistance benefits. in entitled families is partially explained by the absence In particular, tight eligibility thresholds, which have not of indexation of the income threshold in a context of been updated to reflect the economic growth in good mostly positive GDP growth and slightly declining fertility times, have led to gradual marginalization of targeted (World Bank 2015d). The marginalization of means-tested benefits. For instance, in Bulgaria, overall social assistance programs had already begun in the pre-crisis period, expenditures increased between 2004 and 2012; however with an average decline in social assistance coverage by the expenditure on the Guaranteed Minimum Income nearly 11 percent between 2004 and 2008 in the limited program (that mitigates poverty among the extreme poor) welfare states. Among countries in small balanced welfare states, Malta, and Lithuania expanded spending as well as coverage Slovakia, Malta, and Lithuania increased social during this time period. As a result, the small balanced assistance spending as well as coverage from 2008 welfare states expanded coverage from an already high to 2012. On the other hand, Latvia and Romania cut 76 percent in 2008 to 78 percent in 2012, by expanding the spending during the recession; the cut was very large in poverty targeted housing and social exclusion programs fig. 29 the case of Romania, from 2.7 million Euros in 2008 to . Not only did coverage expand, but this expansion 1.6 million in 2012 27. As a result, there was a large drop in has been achieved along with an increase in targeting overall social assistance coverage of the poor, from 89 accuracy fig. 30A – resulting in making the systems more percent in 2008 to 68 percent in 2012. Latvia also cut cost effective in reducing poverty. For instance, real expen- expenditures marginally over this time period; however ditures on the Guaranteed Minimum Income program in there was an increase in real spending on means-tested the Slovak Republic in 2011 was 20 percent higher than in programs, with the result that overall coverage expanded 2008; and the number of beneficiaries had increased by in Latvia, with coverage of social exclusion benefits doubling 12 percent (Sundaram, Strokova, Gotcheva, 2012). from 13 percent in 2008 to 27 percent in 2012. Slovakia, 27 Denoted in constant 2005 terms. 46 welfare states and protection of the poor in the european union Q fig. 29 CUTS IN SOCIAL ASSISTANCE SPENDING HAD DIFFERENT OUTCOMES across welfare states in terms of coverage of the poorest Change in Coverage of Bottom 20 in 2004-2008 and 2008-2012 8 90 6 80 4 70 2 Percentage point change 60 Percent of the bottom 20 0 50 -2 40 -4 30 -6 20 -8 -10 10 -12 0 Large balanced Truncated Small balanced Limited Overall social assistance 2004-2008 2008-2012 2012 (right axis) Source: EU-SILC. / Note: Germany is not included, since we do not have coverage data for Germany. Q fig. 30 INCREASE IN COVERAGE WAS ACHIEVED PARTLY THROUGH IMPROVEMENTS IN TARGETING ACCURACY Targeting Accuracy and Adequacy of Bottom 20 in 2004-2008 and 2008-2012ttom 20 in 2004-2008 and 2008-2012 A. Change in Targeting Accuracy B. Change in Adequacy 6 60 5 45 40 4 Percent of benefits going to the bottom 20 50 3 Percent of gross income of the bottom 20 Percentage point change 35 Percentage point change 2 40 1 30 0 30 -1 25 -2 20 20 -3 -4 10 15 -5 -6 0 10 Large Truncated Small Limited balanced balanced -7 5 Overall social assistance Large Truncated Small Limited balanced balanced 2004-2008 2008-2012 2012 (right axis) 2004-2008 2008-2012 2012 (right axis) Source: EU-SILC. / Note: Germany is not ranked, since we do not have targeting accuracy and adequacy data for Germany 47 Truncated welfare states, which were hit the hardest eu regul ar economic report crisis with a fairly effective Guaranteed Minimum Income by the economic crisis, lacked effective poverty program (the RSI), tightened eligibility conditions for the targeted instruments; that could have helped them program in 2010, making it harder for the poor to access scale up social assistance quickly and efficiently. the program. The number of program recipients declined In Greece, severe fiscal constraints and the need for con- from about 400,000 in January 2010 to less than 300,000 solidation in the midst of a severe recession (see Box 1 in by August 2012 (Portugal RSI, mimeo). Despite declining Recent Economic Developments) led to cuts in social coverage in Portugal, the average coverage increased in assistance spending (starting from a low base). The real value truncated welfare states between 2008 and 2012 mainly due of all transfers (family, housing benefits, old age benefits) to large increases from a very low base in the case of Spain. declined between 2008 and 2012. However, adequacy Spain increased coverage by overall social assistance to the increased slightly, as recipient household incomes de- poorest 20 percent of the population from 12.3 percent clined by even more FIG. 30B . Portugal, which entered the in 2004 to 17.7 percent in 2008 and 24.4 percent in 2012. The different responses across welfare states during There was a large drop in coverage of both social assis- the crisis contributed to differential effects on the tance and social exclusion benefits in Romania – coverage incomes of the poor. Declining labor income was of the poor by social exclusion benefits dropped from 53 clearly an important driver of decreasing income for the percent in 2008 to 31 percent in 2012. More recently, bottom 20, especially among truncated welfare states. Romania has started to take policy measures to reverse Social assistance transfers mitigated this negative impact these trends with increases in the budgets and benefit in many large balanced welfare states (Hungary and Ire- levels of the Guaranteed Minimum Income and Family land being exceptions with large declines 28 ), and in small Support Allowance (FSA) programs. The FSA benefit level balanced welfare states (with the exception of Romania). was doubled in October 2014. Social assistance transfers Hungary introduced various eligibility restrictions along did not mitigate impacts in the truncated welfare states with a nominal freeze on social transfers since 2008, and (with the exception of Spain), or in limited welfare states. the unemployment benefit period was cut back to 3 months. The Hungarian Government decided to spend 0.8% of GDP on public works program and plans to raise this spending to 1.2% until 2018. 28 48 welfare states and protection of the poor in the european union EU member st ates wou ld need to prov ide adequate protec t ion a nd socia l i nvest ment for t he poorest while ensuring f isca l susta inabilit y European member states are some of the biggest models: namely large balanced welfare states, truncated spenders on social protection in the world, but welfare states, small balanced welfare states, and limited some are unable to provide effective protection for welfare states. These models differ along the size of social some of their poorest citizens. Adopting a poverty protection spending; and on the extent of protection reduction lens, this note identifies four distinct European offered to the poor. The composition of spending matters for effective 20 percent of the population largely due to having well- protection. The types of programs, extent of means-testing, functioning guaranteed minimum income programs that and who is targeted through means-tested programs target poor families, and due to universal child benefits. matters in terms of coverage of the poor and in mitigating In contrast, truncated welfare states, despite spending poverty. For instance, small balanced welfare states spend more on family benefits, cover only a small share of the less on social assistance than truncated welfare states, bottom 20 percent, raising questions on the quality and yet they achieve much better coverage of the bottom types of means testing. Several EU member states cut spending on social unable to mitigate the effects. In limited welfare states, assistance during the recent recession. Contrary to poverty targeted programs have become marginalized expectations, real social assistance spending was pro- and thus the crisis response was less effective than in cyclical in all but the large balanced welfare states. This had large balanced welfare states. Small balanced welfare an impact on crisis response. Countries with truncated states reoriented resources during the crisis away from welfare states were hit hardest, but lacked poverty targeted categorical programs toward means-tested programs instruments they could scale up. Consequently poverty and were better able to both contain spending as well as increased sharply and the social protection system was provide protection. Looking ahead, the challenge for EU member states fiscal consolidation. With the challenging global economic is how to balance the sustainability and effectiveness outlook, aging populations, and declining working age of the welfare state with the constraints imposed by populations, demands on social protection systems will 49 grow even as fiscal space shrinks. At the same time, well targeted social assistance systems will become even eu regul ar economic report shrinking younger cohorts make it imperative to invest in more important in the future, as the number of self- opportunities in all youth, including through well-designed employed entrepreneurs increases, unemployment spells social assistance to the poorest. In addition, the world of become more frequent, and if fewer are covered through work is changing. Technological change, urbanization, the current pension systems (World Development Report and globalization have accelerated, creating unparalleled 2016 –forthcoming). The ongoing refugee crisis in Europe economic opportunities and challenges leading to rapid also poses a question on how welfare states can help labor market changes. Providing efficient protection through integrate refugees. Social assistance and labor market programs play unemployment benefits; do not have family members an important role in alleviating poverty, managing receiving pensions; and do not have children receiving risks, and supporting investment in the poor. They benefits. Tightening eligibility conditions during a crisis can improve individual productivity and income through (e.g. Portugal, Hungary) or marginalizing coverage during contributing to preserving and building human capital, periods of growth, as was the case in many limited welfare and through promoting access to better jobs and income. states (e.g. Poland, Bulgaria), need to be reversed to Building a robust and cost-effective social assistance pillar ensure robust coverage and adequacy. 29 is important in terms of investing in and promoting the poorest members of society. c. Reduce leakage of social assistance transfers to the rich: Several countries, particularly among the Going forward, important reform priorities include: large and small balanced welfare states, cover more than 40 percent of the rich through social assistance a. Introduce guaranteed minimum income (GMI) transfers – for instance, 61 percent of the rich receive programs that target poor families: This is espe- transfers in Malta; 58 percent in Ireland; 51 percent in cially relevant for truncated welfare states; the lack of Estonia; 48 percent in Latvia; 46 percent in Slovakia; a national poverty targeted program in Greece and Italy 44 percent in Belgium; and 43 percent in Demark. limits their ability to protect and invest in the poor. Such high coverage accruing to members in the richest The main challenge is to introduce and sustain efficient, 20 percent of the population is often through universal well targeted poverty programs that provide protection transfers that are categorically targeted (i.e. universal to the poorest while simultaneously consolidating the child benefits). Ensuring fiscal sustainability will involve large and fragmented plethora of programs. Greece is cutting back on spending and reducing leakages to currently piloting a guaranteed minimum income program, the rich. Better targeting of social assistance transfers with a plan to roll this out nationally starting in 2017. would free up resources, which can then be channeled into savings or investment (in countries with high b. Maintain effective coverage and adequacy of coverage of the poor), or redirected into increased existing social assistance programs ( especially coverage of the poorest (in countries with insufficient poverty-focused programs such as guaranteed protection for the poor). minimum income), particularly during downturns. Guaranteed minimum income programs are cost- d. Do not wait for an economic crisis to reform effective and provide mitigation against idiosyncratic social assistance; strengthening the protection as well as systemic shocks. Ensuring access to these capacity of social assistance programs requires time programs, particularly during downturns, is crucial for and political effort. families that lose jobs and that are not covered by` It is also important to facilitate the connection back to the labor market for beneficiaries of poverty-focused programs through better integration of employment services 29 with social assistance services. Several countries, such as Bulgaria, are taking initial steps towards better integration (World Bank Dimitrov and Duell 2014). 50 A N N E X 1 . DATA S OU RCE S A N D DE F I N I T ION OF I N DICAT OR S welfare states and protection of the poor in the european union Data on social protection spending comes from and labor markets (unemployment benefits). Given that the European system of integrated social protec- the spending and performance data come from two dif- tion statistics (ESSPROS), managed by Eurostat. ferent sources and that the income data do not necessarily It provides a coherent classification system of social reflect all the programs that are available, there is not a full benefits to households across European countries, thus alignment between spending and performance indicators. making cross-country comparison possible. ESSPROS in- cludes spending on both cash and in-kind social benefits The main indicators of performance of social for different functions (sickness/health care, disability, old assistance cash transfers include: age, survivors, family/children, housing, social exclusion not elsewhere classified, and unemployment). As this note Q Coverage: What portion of the population receives focuses on the social protection system in countries, we the transfers (focusing on the share received by the adjust social protection data to (a) exclude spending on poorest quintile)? some health care functions such as health care and social care services; (b) exclude overall administration costs; Q Targeting accuracy: What portion of social assistance (c) exclude unclassified spending; and (d) replace spending transfers goes to each quintile (with particular focus on on the unemployment function with labor market spending the share of transfers that goes to the bottom quintile)? reported from Eurostat’s Labor Market Policy database since the unemployment function of ESSPROS does not Q Adequacy (or dependency): How much is the transfer capture the expenditures on active labor market programs as a fraction of disposable income? If this fraction is (ALMPs) and public employment services. large, it would imply that the household is fairly dependent on the transfer. This could be either due Social protection performance indicators are esti- to (i) the transfer being large so that the household is mated from the EU statistics of income and living able to depend only on this transfer and does not conditions (EU-SILC) – a compilation of individual have to find other means of generating income or (ii) country survey data on personal and household the household finds it hard to generate any other incomes managed by Eurostat. It collects information income. In the latter case, it is particularly important on the transfers that individuals and households receive to additionally assess adequacy of provided income on old-age benefits, sickness benefits, disability benefits, support by comparing the size of the transfers to a survivor benefits, unemployment benefits, housing allow- more objective measure such as the poverty line. ance, family/child benefits, education allowance, and social exclusion not elsewhere classified. Similar to spending data, For the purposes of the analysis, individuals are ranked these benefits are also classified into social insurance on the basis of equivalised disposable income before all (which includes old-age benefits, sickness benefits, dis- social assistance cash transfers and then divided into five ability benefits, and survivor benefits), social assistance equally sized groups, representing 20 percent of the (housing allowance, family/child benefits, education population (“quintiles”) to form the bottom, second, allowance, and social exclusion not elsewhere classified), third, fourth, and top quintile. 51 R E F E R E NCE S eu regul ar economic report Q Arellano, M. and Bover, O. 1995. “Another look at the Employability: An assessment of the social assistance instrumental variable estimation of error-components system in the Slovak Republic.,” Washington, DC: models,” Journal of Econometrics, 68, pp. 29-51. World Bank. Q Barr, Nicholas. (2012). “Economics of the Welfare State.” Q The United Nations Refugee Agency (UNHCR). 2015. Oxford University Press. Global Trends: Forced Displacements in 2014. Q Blanchard, O.J. and L.H. Summers (1987) “Hysteresis Q Varga, J., R. Werner and J. t’ Veld. 2013. “Growth in unemployment,” European Economic Review 31, Effects of Structural Reforms in Southern Europe: 288-95. The case of Greece, Italy, Spain and Portugal”, European Economy – Economic Papers, No. 511, DG ECFIN, Q Caroll, Christopher, Jiri Slacalek and Kiichi Tokuika. 2014. European Commission December 2013. “The Distribution of Wealth and the MPC – Implications of New European Data,” ECB Working Paper Series Q World Bank (forthcoming). “Global divide or dividend?” No. 1648, March 2014. World Development Report, Washington, DC. Q Dimitrov, Yordan and Duell, Nicola. 2014. “Activating Q World Bank. 2015a. “Productivity Growth in Bulgaria: and Increasing Employability of Specific Vulnerable Trends and Options”, Washington, DC. http://www. Groups in Bulgaria: A diagnostic of institutional capacity” worldbank.org/content/dam/Worldbank/document/ Washington, DC eca/Bulgaria/Bulgaria-Productivity-6-26-15-eng-web.pdf Q ECA SPeeD. Europe and Central Asia Social Protection Q World Bank. 2015b. “Doing Business 2916 – Measuring Expenditure and Evaluation Database. Washington DC: Regulatory Quality and Efficiency,” Washington, DC. World Bank. http://www.doingbusiness.org/reports/global-reports/ doing-business-2016 Q European Central Bank. 2015a. “Bank Lending Survey, October 2015” Frankfurt. Q World Bank 2015c. “Bulgaria: Systematic Country Diagnostics,” Washington, DC. Q European Central Bank. 2015b. “Economic Bulletin”, Issue 2, 2015. Q World Bank. 2015d. “Spending Review of Low Income Family Support in Poland,” Mimeo, Washington, DC. Q European Commission. 2015. “EU Communication on upgrading the Single Market,” Brussels. http://europa. Q World Bank. 2013. “Creating More and Better Jobs in eu/rapid/press-release_IP-15-5909_en.htm Europe and Central Asia,” Washington, DC, http://www. worldbank.org/en/news/feature/2013/11/07/creating- Q Eurostat. 2008. “The European System of Integrated more-and-better-jobs-in-europe-and-Central-asia Social Protection Statistics (ESSPROS),” Luxembourg: Office for Official Publications of the European Com- Q World Bank (2011). “Latvia: Expenditure and Perfor- munities. mance Benchmarking Country Level. World Bank, Washington, DC.World Economic Forum 2015. “Global Q Fiszbein, Ariel. 2004. “Beyond Truncated Welfare Competitiveness Report 2014 and 2015,” Zurich. Status: Quo Vadis Latin America?” Washington, DC: World Bank. Q Sundaram, Ramya, Strokova, Victoria, and Gotcheva, Boryana. 2012. Protecting the Poor and Promoting Q Gauthier, A.-H. and Philipov, D. 2008. “Can policies Employability: An assessment of the social assistance enhance fertility in Europe?” Vienna Yearbook of Population Research 2008: 1-16. system in the Slovak Republic. Washington, DC: World Bank Q Sundaram, Ramya, Strokova, Victoria, and Gotcheva, Q World Development Report 2016 (forthcoming). Digital Dividends. World Bank, Washington, DC Boryana. 2012. “Protecting the Poor and Promoting 52