Portraits of Labor Market Exclusion 2.0 Portraits of Labor Market Exclusion 2.0 © 2017 International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org The findings, interpretations, and conclusions expressed here do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other infor- mation shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The European Commission support for the pro­ duction of this publication does dorsement of the contents which reflects the views only of the not constitute en­ authors, and the Commission cannot be held responsible for any use which may be made of the information contained therein. The opinions expressed and arguments employed herein do not necessarily reflect the official views of the Organisation for Economic Co-operation and Development (OECD) or of the governments of its member countries. Rights and Permissions The material in this work is subject to copyright. Because the World Bank encour- ages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Wash- ington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Photos: © World Bank Cover design/layout and editing: Nita Congress Contents Acknowledgments iv Background 1 The context 3 The target population 5 The barriers to employment 7 Patterns in labor market vulnerabilities 10 Snapshots of the labor market vulnerable 13 Identifying priorities 16 Policies and programs: reality on the ground 19 The challenges ahead 21 Notes 24 References 25  iii Portraits of Labor Market Exclusion 2.0 Acknowledgments T his report was produced by a World Bank team co-led by Aylin Isik-Dikmelik (Senior Economist) and Mirey Ovadiya (Senior Social Protection Specialist) and including Frieda Vandeninden (Economist), Natalia Millan (Economist), and Sandor Karacsony (Social Protection Specialist). The report synthesizes findings from 12 country policy papers produced under a joint project of the European Commission (EC), the World Bank, and the Organisation for Economic Co-operation and Development (OECD). Six papers (Bulgaria, Croatia, Greece, Hungary, Poland, and Romania) were produced by the World Bank under the rubric “Portraits of Labor Market Exclusion 2.0” (available at worldbank.org/en/region/ eca/publication/labor-exclusion-portraits); and six country policy papers (Estonia, Ireland, Italy, Lithuania, Portugal, and Spain) by the OECD under the rubric “Faces of Joblessness” available at www.oecd.org/social/faces-of-joblessness.htm). The report also draws on the project’s Joint Methodology Paper. The report would not have been possible without the financial and technical sup- port of the EC’s Directorate General of Employment, Social Affairs and Inclusion. Katalin Szatmari (Policy Officer) led the EC team, and Herwig Immervoll (Senior Social Policy Economist) led the OECD team. The EC team included Christian Aagaard (Policy Officer), Aurimas Andrulis (Policy Officer), Maria Baroni (Policy Officer), Graeme Belshaw (Policy Officer), Suzanne Conze (Policy Officer), Ines Delgado Alves (Policy Officer), Michela Di Donato (Policy Officer), Isabelle Eng- sted-Maquet (Deputy Head of Unit), Manuela Geleng (Head of Unit), Ioana-Maria Gligor (Deputy Head of Unit), Georgi Karaghiozov (Policy Officer), Dora Krumova (Programme Manager) Katharina Muhr (Policy Officer), Raya Raychinova (Program Assistant), Valentina Remida (Policy Officer), Balazs Rottek (Policy Officer), Alina Seciu (Policy Officer), Sirje Sepp (Policy Officer), Katalin Somody (Policy Offi- cer), Graham Stull (Policy Officer), Alexandra Tamasan (Policy Officer), Georgios Taskoudis (Policy Officer), Miriam Toplanska (Policy Analyst), and Iva Zelic (Policy Officer). The OECD team included James Browne, Nicola Düll, Rodrigo Fernandez, Dirk Neumann, Daniele Pacifico, and Céline Thévenot. The World Bank team is grateful to the EC and OECD teams for their close collaboration during this project. Andrew D. Mason (Practice Manager, Europe and Central Asia Social Protection and Jobs Practice), Arup Banerji (Regional Director, European Union) and Cem Mete (Practice Manager, Europe and Central Asia Social Protection and Jobs Practice) provided overall guidance to production of the synthesis report. Peer review comments to the country policy papers were received at various stages from Christian Bodewig (Program Leader), Aline Couduel (Lead Economist), Vic- toria Levin (Senior Economist), Matteo Morgandi (Senior Economist), Cristobal Ridao-Cano (Lead Economist), Victoria Strokova (Economist), Ramya Sundaram (Senior Economist), and Trang Van Nguyen (Senior Economist). Finally, both the OECD and World Bank teams benefited from extensive interaction and consultations with government officials in the countries under study. In par- ticular, the teams would like to thank the respective country representatives of the ministries of labor, social policy, and public employment services who provided guidance, data, and specific inputs in finalizing the country policy papers. iv  Portraits of Labor Market Exclusion 2.0 Background The people within and across countries characterized as labor market vulnerable are not all the same. Different life situations lead people to being unemployed, inactive, or precariously employed. Various personal constraints may induce people to take temporary or precarious employment, work a reduced number of hours, or earn very low incomes despite being engaged in full-time work. All of these are considered the labor market vulnerable, and a better understanding of their differences and similarities is essential to successful labor market inclusion. Governments need to know who the labor market vulnerable are and what barriers keep them out of work or in unstable employ- ment. This information will help countries meet the inclusive growth priorities of the Europe 2020 Strategy (EC 2010) and the European Pillar of Social Rights;1 it also aligns with recognizing the potential negative impacts of labor market vulnerability on long-term growth. While some statistics on broad groups (for example, youth) exist, deeper analysis—par- ticularly of the diverse barriers faced by age and gender—is needed. Only by recognizing and understanding the specific, diverse barri- ers to employment can labor market vulnerability be meaningfully addressed. In this context, Portraits of Labor Market Exclusion 2.0—part of a joint study between the European Commission (EC), the World Bank, and the Organisation for Economic Co-operation and Development (OECD)— aims to inform activation and employment support and social inclusion policy making and programs through an improved understanding of labor market barriers.2 This study builds on both the original 2014 EC–World Bank Portraits of Labor Market Exclusion initiative (Sundaram and others 2014) and other earlier analyses characterizing people with labor market difficulties (EC 2012; Ferré and others 2013; Immervoll 2013). Specifically, the present study updates and broadens the analysis in Bulgaria, Estonia, Greece, Hun- gary, Lithuania, and Romania and extends the analysis to Croatia, Ireland, Italy, Poland, Portugal, and Spain to map the diversity of profiles for the out- of-work and precariously employed in 12 countries as part of the World Bank Portraits of Labor Market Exclusion and OECD Faces of Joblessness projects. The study expands on the previous EC–World Bank analysis by considering a broader group of the labor market vulnerable. The study’s target population extends beyond the unemployed as defined by the International Labour Organization, on whom most studies typically  1 Portraits of Labor Market Exclusion 2.0 focus, to include all out-of-work and those in precarious employment, such as those with restricted hours and those with near-zero incomes. It also refines the analytical methodology of its predecessor study by applying an employment barrier framework—which defines the vulnerable population, the types of employment barriers they face, and indicators to proxy these barriers—to inform policy interventions in country-specific applications, and to provide a reference point for future methodological extensions. The study separates the target population into distinct groups with respect to the types of employment barriers they face. It does so using latent class analysis, an advanced statistical method, and an employment barrier framework. This approach facilitates discussion of the strengths and limitations of existing policy interventions for specific groups of the labor market vulnerable, and helps inform policy decisions on whether and how to channel additional efforts toward specific groups. Addressing the same barrier may require a different set of poli- cies based on the distinct characteristics of the identified groups. For example, while not having recent work experience is an employment barrier faced by many individuals, addressing it may require a different approach for inactive mothers than for young unemployed men. Beyond the barriers, the socioeconomic characteristics of each group also mat- ter. The study’s approach and findings highlight the importance of such thorough segmentation. Countries are resource constrained and not in a position to attend to the needs of all the labor market vulnerable. Thus, specific priority groups among the target population were identified in close consultation with the country counterparts. The study then delved into the results of the latent class analysis for each priority group (OECD and World Bank 2016). The findings enable a richer, deeper, multidimen- sional understanding of combinations of employment barriers facing specific groups than that discernible through traditional statistics. The analysis focuses primarily on supply-side constraints and corre- sponding policies. By focusing on priority groups for activation, the study puts in perspective the existing policies and programs potentially available for them given their needs. Potential gaps were identified for tailored activation policies through intensive consultations with countries. While the study rec- ognizes the essential role demand plays in improving labor market outcomes and explicitly includes an “opportunity barrier” to capture demand-side constraints, analysis of these constraints—which requires a comprehensive approach across multiple facets of the economy—is beyond its scope. Twelve country policy papers have been produced using this approach. These papers analyze the out-of-work and precariously employed population, along with existing employment support programs and policies. This synthesis report provides a summary of the trends and findings emerging from these country policy papers. 2  Portraits of Labor Market Exclusion 2.0 The context T his section looks at the labor markets in the 12 EU countries covered by this study and highlights a few key trends distinguishing these markets. Unemployment remains relatively high in most EU countries, and levels have increased over the last 10 years following the economic crisis of 2008–09. While most countries (with the exceptions of Croatia, Greece, and Spain) were well on the way to recovery by 2016, unemploy- ment and precarious employment remain challenges to be addressed. Long-term unemployment is largely prevalent across the group of countries under study. As of 2016, in all but Estonia, Hungary, Lithuania, and Poland, more than half of the unemployed have been unemployed for more than a year (figure 1). And unemployment overall has been particularly sizable in Croatia, Greece, Italy, Portugal, and Spain (above 10 percent). In most of the remaining countries, unemployment rates remain higher than pre-crisis levels. Figure 1  Unemployment in 2007, 2016, and long term 25 80 70 % of long-term unemployment Unemployed as % of active working-age population 20 60 15 50 40 10 30 20 5 10 0 0 EU Hungary Romania Poland Estonia Bulgaria Ireland Lithuania Portugal Italy Croatia Spain Greece 2007 2016 Long-term unemployment (2016) Portraits of Labor Market Exclusion 2.0  3 Unemployment tends to be more concentrated among certain specific segments of the working-age population. This is the case even in countries with relatively low unemployment rates such as Estonia, Hungary, Poland, and Romania. For example, youth unemployment is four times that of the overall working-age population in Romania. And youth unemployment rates are particularly high in Greece, Italy, and Spain, at about 47 percent, 38 percent, and 44 percent, respectively (figure 2). Figure 2  Unemployment: overall versus among youth and the poorly educated 50 Unemployed as % of active working-age population 40 30 20 10 0 Hungary Romania Poland Estonia Bulgaria Ireland Lithuania EU Portugal Italy Croatia Spain Greece Lower secondary education or less Youth Overall Individuals with low levels of education (those with less than a lower sec- ondary education degree)3 experience much higher unemployment rates in EU countries, especially in Bulgaria, Lithuania, Poland, Hungary, and Spain (figure 2). The gender gap in labor market participation is striking in several countries. The activity rate—that is, the ratio of the active population to the overall working-age population—for women is relatively lower than for men in all 12 countries as well as for the EU average (figure 3). Gender disparities are particularly apparent in Greece, Ireland, Italy, Romania, and Spain, where the activity rates for men and women differ by more than 10 percentage points. Together with youth and the less educated, women are a common vulnerable group in the labor markets of the 12 countries under study. Figure 3  Labor market participation by gender 90 80 working-age population 70 Active as % of 60 50 40 30 20 10 0 Italy Croatia Romania Greece Bulgaria Poland Hungary Ireland EU Portugal Spain Lithuania Estonia Male Female Total 4  Portraits of Labor Market Exclusion 2.0 The target population U nlike most labor market profiling exercises which focus on the unemployed only, the target population of the Portraits of Labor Market Exclusion 2.0 analysis comprises a much larger group of individuals. The target population includes not only those within the working-age population who are out of work (whether actively searching for a job or inactive) but also those who are precariously employed—that is, people with unstable jobs, restricted hours, and/or near-zero incomes (see box 1 for definitions of these terms).4 The size of the target population differs Figure 4  Process for defining the target population: depending on demographics and shares Bulgaria of individuals in education and mili- tary service. The share of individuals age 18–64 varies significantly across countries, Total population: from 60.3 percent in Poland to 66.4 percent 7.3 million in Hungary. The share of those in educa- tion or military service excluded from the working-age population also varies, ranging from 3.9 percent of the total population in Working-age population: 4.6 million Bulgaria to 6.2 percent of the total population in Hungary. The complete set of working-age Students and military: individuals considered in the analysis ranges 0.3 million from 55.7 percent of the total Irish population to 60.9 percent of all Romanians. Remainder (from which to The working-age population is next clas- extract target population: 4,4 million sified into two main categories (figure 4): (1) those who have no labor market con- No labor market constraints: straints; and (2) the target group, consisting 2.8 million of those who are persistently out of work and those who are precariously employed. TARGET: Greece and Ireland have the largest share 1.6 million of the target population, accounting for  5 Portraits of Labor Market Exclusion 2.0 almost half of the respective working-age population; followed by Croatia and Spain (45 and 43 percent, respectively). In contrast, the target popula- tion represents only about 29 percent of Estonia’s working-age population and about 33 percent of Lithuania’s. In general, the precariously employed population is much smaller than the out-of-work population. The highest proportions of pre- cariously employed are found in Spain (15 percent of the working-age population), Ireland (14 percent), and Estonia (13 percent). The composition of the out-of-work category varies greatly across countries. Those persistently unemployed (for at least a year) represent the main bulk of the target population in Bulgaria, Croatia, Greece, Portugal, and Spain. In some countries, there are very few persistently unemployed, but instead a large proportion of retirees, such as in Romania and Hungary. There is also a wide variation in those performing domestic tasks—from 3 percent in Poland, Hungary, and Lithuania to 14 percent in Italy (figure 5). Figure 5  Composition of target population by country 50 % of working-age population 40 Out of work Other inactive Performing domestic tasks 30 Incapable of work Retired Persistently unemployed 20 Precariously employed Unstable jobs Restricted hours 10 Near-zero earnings 0 Greece Ireland Croatia Spain Romania Italy Portugal Poland Hungary Bulgaria Lithuania Estonia Box 1  Study target population: definitions nn The out-of-work: people who report being persistently unemployed, retired (that is, early retirees), or inactive (including those incapable of work, those engaged in domestic tasks, or other inactive) throughout the reference period (12 consec- utive months and at the time of the interview); these individuals were also not working at the time of the survey interview nn The precariously employed: people who belong to one of more of three nonmutually exclusive groups: • Those with unstable jobs: individuals reporting work activity but only for a limited number of months during the refer- ence period (maximum 45 percent of potential working time) and those who report no work activity during the income reference period, but report being employed at the time of the interview • Those with restricted working hours: people reporting less than 20 hours of work a week for most or all the reference period; excluded from the target population are individuals working 20 hours or less because they were in school or in train- ing programs or because the number of hours they were working is considered to be a full-time job in their field of work • Those with negative, zero, or near-zero labor incomes: people reporting some work activity during the income refer- ence period but negative, zero, or near-zero earnings; to enable comparison across countries, the same low-earnings threshold is used for all countries (€120/month in purchasing power parity with EU as the reference) Source: OECD and World Bank 2016. 6  Portraits of Labor Market Exclusion 2.0 The barriers to employment T he next step in the analysis is segmenting the target population into distinct groups according to labor market barriers and socio- economic characteristics. To this end, a framework identifying three types of employment barriers—insufficient work-related capabilities, weak economic incentives to look for or accept a “good” job, and scarce employment opportunities—that prevent individuals from being partially or fully active in the labor market was formulated (figure 6). As these barriers cannot be directly observed, a set of eight indicators was developed to capture them using survey data.5 The barriers and their associated Figure 6  Employment barrier framework indicators are listed in table 1.6 These indicators provide a means for categorizing the labor market vulner- BETTER QUALITY able, who face various combinations EMPLOYMENT and severities of employment barri- ers. The indicators enable segmenting of the target population by the combina- POLICY INTERVENTIONS tions of barriers they face. Through this Lack of approach, policy interventions can be tai- financial incentives lored to improve employment outcomes. Lack of Lack of job opportunities work-related capabilities Some barriers are more predomi- Near-zero income Out-of-work nant in some countries, but there can also be large variations in inci- �  dence of barriers across countries  Unstable jobs (figure 7). Overall, the most salient barriers are low education and low rela- Restricted hours  tive work experience—with an average  TARGET GROUP: of 42 and 41 percent across 12 countries People with potential labor market difficulties Portraits of Labor Market Exclusion 2.0  7 Table 1  Employment barrier indicators Barrier Indicator 1. Low education: if an individual has an education level lower than upper secondary education in the International Standard Classification of Education (ISCED)-11 classification) 2. Care responsibilities: if an individual lives with someone who requires care (i.e., children 12 and under receiving under 30 hours of non-parental care a week, incapacitated household members or elderly with health limitations) and is either the only potential care giver in the household or is reported as inactive or working part time because of care responsibilities Insufficient work- 3. Health limitations: if an individual reports some or severe self-perceived limitations in daily activities related capabilities due to health conditions 4. Low relative work experience: if an individual has worked less than 60 percent of the time since they left full-time education 5. No recent work experience: those who have worked in the past but have no recent work experience (have not worked for at least 1 month in the last semester of the reference period or at the month of the interview); or those who have never worked Weak economic 6. High non-labor income: if household income (excluding that from the individual’s work-related activi- incentives to look for ties) is more than 1.6 times higher than the median value in the population of working age or accept a “good” 7. High replacement benefits: if earnings-replacement benefits (excluding categorical social benefits) job are more than 60 percent of an individual’s estimated potential earnings in work Scarce employment 8. If an individual is estimated to have a high probability of being unemployed or involuntarily working opportunities part time due to their age, gender, education, and region of residence Source: OECD and World Bank 2016. respectively, of the target populations facing these barriers. On the other hand, only 15 percent of the target population faces this barrier in Lithu- ania, and up to 73 percent in Portugal, with Poland closest to the average at 43 percent. The range is smaller for low relative work experience, with more than 30 percent of the target populations being affected by this bar- rier in almost all countries; the outlier here is Croatia, where this is a barrier for 59 percent of its target population. Incentive barriers are not as great an issue overall, both for nonlabor income and earnings replacement; how- ever, certain specific subgroups of the target population are an exception, as discussed in the next section. Figure 7  Comparison of incidence of barriers across 12 countries 80 Portugal 70 % of target population 60 Croatia 50 Italy Poland Bulgaria Estonia 40 Hungary Italy Lithuania 30 Romania Portugal Ireland Portugal 20 Croatia Greece Estonia Bulgaria Poland Bulgaria Lithuania 10 Romania Estonia Portugal Croatia 0 Low No past work Low relative Health High care High nonlabor High earnings Scarce job education experience work experience limitations responsibilities income replacements opportunities 8  Portraits of Labor Market Exclusion 2.0 Barrier incidence reflects structural policy challenges, cyclical labor market trends, and/or country-specific constraints as well as existing policy interventions. The labor market recovery in Estonia, for instance, to some extent drives the low incidence of work experience barriers (for example, no past work experience) as well as of the scarce job opportunities barrier. On the other hand, even though countries have a high prevalence of a particular type of barrier, it does not necessarily mean they have no policies effectively dealing with them. In Poland, the prevalence of the care barrier is at 15 percent, which is on par with the 12-country average; the country has existing initiatives expanding child care coverage. Significant shares of the respective population face multiple, albeit varied, barriers across the 12 countries (figure 8). The share of the target population that faces three or more barriers ranges from 33 percent in Estonia and 38 percent in Bulgaria to 81 percent in Greece. In contrast, about 25 percent of the target population faces only one barrier in Estonia, Ireland, and Lithuania compared to only 4 percent in Greece and 12 percent in Croatia. Figure 8  Number of barriers facing target population by country 100 90 80 with employment barriers Share of target population No major barrier 70 Single barrier 60 2 barriers 50 40 3 barriers 30 4+ barriers 20 10 0 Greece Croatia Poland Romania Italy Average Hungary Ireland Portugal Spain Lithuania Bulgaria Estonia Portraits of Labor Market Exclusion 2.0  9 Patterns in labor market vulnerabilities L atent class analysis (LCA) is a statistical tool that allows the target population to be clustered into different groups using the employment barrier framework. Applying the clustering methodology results in the classification of the target population into Box 2  Latent class analysis separate homogeneous groups (box 2). nn Enables characterization of a categorical unobserved Examples of such clusters are individuals latent variable: in this case labor market vulnerability; with unstable jobs, workers with low edu- cation, low-educated young mothers with nn By analyzing relationships among observed variables: no work experience, early retirees with in this case “employment barrier” indicators; high replacement income, or young men in urban areas with high education but no nn In doing so, it segments a heterogeneous population work experience. into meaningful distinct groups (similar within groups and different between groups) The total number of groups identified varies across countries, depending nn Exploits interrelations of the employment barriers and on the heterogeneity of the labor the joint determination of observed outcomes (unlike market vulnerable and the richness regression analysis) of available data. In each country, the a  groups are distinct from each other in      �  terms of size, demographic and socio- economic characteristics, and mix of      �  barriers faced. For instance, in Greece,     � the application of employment barriers  using LCA leads to eight distinct groups,      �  each of which faces similar barriers and  possesses similar socioeconomic charac- b  teristics (figure 9). But in every country,  the same principle holds true: individuals are homogeneous within a group with 10  Portraits of Labor Market Exclusion 2.0 respect to the employment barriers they Figure 9  Greece's eight groups identified through LCA face, but different across groups. The approach yields rich, informa- tive, and useful results. By beginning  from the explicit employment barriers   � themselves, the methodology allows for  all the barriers affecting a group to be Out-of-work and precariously considered together. The barriers are employed, with heterogeneous the main drivers for the groupings. In characteristics contrast, the traditional approach of exam- ining drivers of labor market vulnerability � entails isolating one dimension while  holding the others constant. a 24+19+17116z The result for policy makers is the opportunity to take a broad view: to 8. Stay-at-home married mothers with see the full range of needs and the 7. Urban middle- care responsibilities aged part-time distinct subgroups involved. This van- workers 1. Low-income tage point enables “big picture” planning 6. Rural middle- 6% middle-aged aged in unstable 6% unemployed with and budgeting for the different types of jobs 24% work experience 6% activation and employment support pro- 5. Well-off grams needed. (More detailed design and relatively educated implementation will require further data (early) 11% retirees and analysis.) 19% 2. Older inactive 11% with past work Although the LCA iteration process 4. Relatively experience and educated long- health limitations leads to different numbers and types term single 17% unemployed of groups in each country, common NEETs trends have emerged (table 2). All 3. Low-educated inactive women with 12 countries have at least one group of no work experience older individuals—usually early retirees— facing some sort of health issue. Women with care responsibilities are present in all countries except Romania. Long-term unemployed facing a particular set of discernible and discrete barriers can be found in all countries except Poland and Spain. Note that this does not mean that long-term unemployment is low in Poland or Spain, but that the long-term unemployed do not share similar barriers and characteristics in those countries and therefore do not cluster as a single group. Groups with those in precarious employment and youth with low work experience are also present in most countries. In contrast to these groups common to all or most countries, there are some groupings that occur only within a specific country. Examples include “parents (both male and female) with a high-income partner” in Ireland, “low-educated inactive middle-aged women with no work Portraits of Labor Market Exclusion 2.0  11 Table 2  Common groups across countries Lithuania Romania Portugal Bulgaria Hungary Estonia Croatia Greece Ireland Poland Spain Italy 1. Older inactive adults with health problems 2. Mothers with care responsibilities 3. Long-term unemployed 4. Low-intensity/precarious employment/working poor 5. Youth with low relative work experience 6. Disabled Note: Size of circle indicates prevalence. experience” in Romania, and “middle-aged relatively educated (mostly) men living in urban areas” in Bulgaria—just to name a few. The ability of the LCA methodology and the employment barriers approach to highlight similarities and differences across and within countries speaks to added value of this analysis. It goes beyond a standard categorization of the labor market vulnerable to enable a highly tailored approach for their activation. 12  Portraits of Labor Market Exclusion 2.0 Snapshots of the labor market vulnerable P eople in very different circumstances can face similar challenges. For instance, the group of mothers in Poland, while facing similar care responsibilities, might be very different from their counterparts in Ireland—in terms of level of education, work experience, marital status, residence in a remote location, and so on. All of these specifications have policy implications; policy makers must take a holistic view in identifying the needs of the labor market vulnerable in their country. This section underscores the multidimensional richness of the results of this analysis by telling stories of representative individuals from different countries. It elaborates on the complexity of labor market vulnerability by juxtaposing paired stories of individuals who face similar barriers.7 Although nominally similar, they are actually quite different. These stories further illustrate the importance of having a differentiated, customized menu of program support to address barriers to employment for the labor market vulnerable. Mothers with care barriers Emily lives in Ireland with her husband and two children. She is 42: she obtained a bachelor’s degree 20 years ago, and worked about 13 years thereafter. Once she had her first child, she stopped working to take care of him and has, since then, been economically inactive, as her husband provides sufficient income for the whole family. Her main barriers to employment are low incentives due to husband’s income level and child care, as her children receive less than 30 hours of nonparental child care each week. Lena is a 29-year-old single mother in Hungary, with 2-year-old child. She lives in a rural area of a region with scarce employment opportunities. She  13 Portraits of Labor Market Exclusion 2.0 did not complete upper secondary school and has worked only a couple of years; she has not been working recently. She lives in material deprivation, and even though she receives family and social exclusion benefits, she is within the country’s poorest income quintile. Even though Emily and Lena are both mothers with care responsibili- ties, they face different barriers. While Emily has a tertiary education degree, Lena has only completed lower secondary education. Lena lives in a remote area where job opportunities are scarce. Compared to Lena who lives in poverty, Emily is relatively wealthy because of her husband’s income. This circumstance means that she has little incentive to work, which is her main employment barrier after child care responsibilities. These differences in their profiles require different activation policies. Emily is not looking for a job, does not benefit from earning replacement benefits, and is therefore not registered for public employ- ment services. The measures that would be available for her if she were registered—Ireland’s Vocational Training Opportunities Scheme and Springboard courses and Jobs Plus subsidies—are not relevant given her high-skill profile. Improving access and decreasing the opportunity cost of child care would be a key element for Emily’s activation, in order to com- pensate for the disincentive effects of her husband’s contribution. Several instruments should be combined to effectively activate Lena. A combined approach that focuses on job search assistance and wage subsidies may help address her low level of education, low amount of work experience, and scarce job opportunities. The approach should simultaneously address the care responsibilities barrier and inactivity in the labor market. Given that she receives social exclusion benefits, the lat- ter could be used as an outreach measure. Youth with low work experience Ivan is Bulgarian and lives in Sofia. He has been unemployed for more than a year. Twenty-seven years old, he is single and lives with his parents. He would like to live on his own, but cannot find a job, even though he is actively looking for one. He is relatively well educated, having finished sec- ondary education, but lives in an area with few job opportunities. He faces three main barriers: a lack of recent work experience, scarce job opportuni- ties, and disincentives related to his parents’ income. Costa is also among the long-term unemployed and lives in Athens, Greece. He holds a bachelor’s degree but has not yet found a job. Like Ivan, he is single and lives with his parents. 14  Portraits of Labor Market Exclusion 2.0 In light of their relatively well-educated status, Ivan and Costa would ben- efit from job search assistance/counseling, combined with intermediation and employment subsidies. Although they have at least upper secondary education, their low work experience implies that they may lack the voca- tional skills needed for many jobs. Skill-building activities—particularly training activities closely linked with employers, such as apprenticeships or on-the-job training—and preferably integrated with job search assis- tance or intermediation would aid them in obtaining much-needed work experience. Actions to activate Ivan and Costa need to be tailored to their specific group characteristics—which are quite similar—within the policy context of their respective native countries. In Bulgaria, even though active labor market program spending is very low (0.17 percent of GDP in 2014), there are employment subsidies and training targeted to those under age 29 (for example, employment in public administration for young people with university degrees and first job programs). However, only 10.7 percent of active labor market program beneficiaries are under age 25—a low percentage in view of the high unemployment rate among this age cohort. In addition, only 13 percent of beneficiaries are among the long-term unemployed. Reinforcement of current programs and increased linkage with job search programs tailored to long-term youth unemploy- ment would be key elements to activate this group in Bulgaria. In Greece, the few programs that explicitly target unemployed youth focus mostly on training. One example is the traineeship program targeted to youth (intended to reach 3,000 beneficiaries age 25–29 and 10,000 benefi- ciaries age 18–24). At the same time, access to job search assistance and counseling (which is exacerbated by limited institutional capacity) and employment subsidies are limited.  15 Portraits of Labor Market Exclusion 2.0 Identifying priorities T his analysis offers very rich data in terms of understanding the diversity of the labor market vulnerable. However, policy makers face finite resources and countless limitations and constraints. In close consultation with the 12 governments included in this study, hard decisions were made regarding which of the groups identified in the respective target populations would be considered a priority for activa- tion and employment support policies. The rationale for selection of these groups was based on several criteria, reflecting the current social policy goals and objectives of the respective countries, along with salient features of their labor market. Figure 10 illustrates countries’ divergent priority group choices with respect to labor market vulnerability and the relative size of each group, in accordance with their respective policy objectives and labor market challenges. This section details a few country cases to highlight country prioritization processes. At the conclusion of these processes, more in-depth identification needs for priority groups will occur, along with an assessment of whether existing policies correspond to or are adequate to address these needs. In Hungary, the driving factors for selecting priority groups were youth and low level of skills, based on high unemployment among those groups, and in accordance with the objective of reducing poverty and the risk of social exclusion. Key considerations included the low labor force participation rate among youth, other factors foreshad- owing productivity constraints of the future labor force—such as aging and demographics; the high share of youth neither in employment, education, or training (NEET); and a high share of early school leavers—high incidence of poverty and material deprivation in families with children, and the EU 2020 employment target and associated flagship initiatives (including the 16  Portraits of Labor Market Exclusion 2.0 Figure 10  Country priority groups Ireland 100 Hungary 90 Poland 80 % with 3+ employment barriers 70 60 50 40 Estonia 30 20 10 0 0 5 10 15 20 25 30 35 40 Size of the group (% of out of work or precariously employed) Agenda for New Skills and Jobs, and the European Platform Against Pov- erty and Social Exclusion). As a result, among the six identified groups in the target population in Hungary, the priority candidates for activation and employment support policies are the poorest groups: poor, low-edu- cated, unemployed, or inactive individuals, some with health limitations ; poor, middle-aged, mostly male, long-term unemployed with past work experience ; and poor, low-educated, inactive young mothers with care responsibilities . This last group is relatively small, but represents a real challenge in terms of the multitude of barriers its members face. Taken together, the latter two groups represent less than 10 percent of the target population and face multiple constraints (upper-left portion of figure 10) which might make activation harder. However, focusing on other, larger groups—such as relatively wealthier women or older individuals—does not respond to the country’s social inclusion priorities. In Ireland, the prioritization process led to a focus on small groups with large numbers of barriers as in Hungary—with some different drivers, however. The logic behind the selection process in Ireland was to follow the government’s stated ambition to extend the scope of activation programs to those not on the Live Register of claimants of unemployment benefits, as part of the current phase of the Pathways to Work strategy—therefore excluding groups with unemployed and older  17 Portraits of Labor Market Exclusion 2.0 workers as part of the prioritization. The groups selected were mothers with limited work experience and with care responsibilities ; and parents with higher-income partners and care responsibilities , economically vulnerable parents without any past work experience and with care respon- sibilities ; these represent 4 percent of the target population or 2 percent of the reference population, which reflects the labor market structural issues related to child care. In Poland, current demographic challenges drove the prioritiza- tion. Even though Poland is characterized by lower unemployment, the activity rates among women are relatively low (only 62 percent). Given the rapidly aging population and low fertility rates (one of the lowest in the EU), the country’s focus has been on reducing barriers to employment for inactive women. However, the activation of large groups such as inactive women may be politically difficult for the policy agenda. Another concern in Poland is the surge of temporary contracts and resulting job precarious- ness. The duality of Poland’s labor market thus calls for a focus on those in precarious jobs, which led to the selection of three groups: middle-aged inactive women with care responsibilities and low relative work experi- ence, middle-aged and young men in precarious jobs with low relative work experience, and younger relatively educated unemployed women in areas of scarce jobs . The selection process resulted in a choice of larger groups with high numbers of barriers (upper-right panel of figure 10). In Estonia, the selection process led to a focus on large groups, but with few concurrent barriers. One of the drivers of the selection process in Estonia was to improve social inclusion. Two large groups with high poverty rates were selected, given that their reintegration in the labor market is of significant broader social concern: older labor market inactive adults with health limitations, low skills, and limited work experience ; and working poor. Another driver for priority selection include a current policy interest in Estonia, stemming from recent changes in the parental leave scheme with the aim of increasing the labor market participation of moth- ers. Therefore, unskilled mothers with care responsibilities and limited work experience was selected as a third priority group. The labor market is tight in Estonia, and there is widespread agreement—notably among employers—that additional groups need to be brought closer to the labor market. In sum, a country’s policy context and objectives for labor market inclusion are the drivers of prioritization, rather than the size of the groups and the difficulty/ease of activating the selected groups, as illustrated on figure 10. 18  Portraits of Labor Market Exclusion 2.0 Policies and programs: reality on the ground A cross the 12 countries studied, 78 percent of the target population faces at least two barriers, and 49 percent faces three or more. Moreover, as the previous sections have highlighted, addressing any given barrier may require a different set of activation policies accord- ing to the other barriers faced and characteristics of the identified priority group. Thus, tailored policies are critical in addressing the complexity posed by multiple constraints to employment for different priority groups. In order to inform the development and refinement of such policies and programs, the study examined the current “supply” of activation policies in each country to highlight potential gaps with respect to priority group needs. This section summarizes the common themes emerging from that exercise. nn Activation and employment support programs are fragmented. In several countries, including Croatia, Greece, and Poland, the study found a high proliferation of programs with low coverage and limited capacity for some key services such as child care. This reveals potential fragmentation and the need for a more coherent approach to activation policies. A lack of coordination between the various institutions involved in activation—specifically, the central and regional governments, as in Italy and Spain; but also among private providers and nongovernmen- tal organizations—can reinforce this fragmentation. Though nascent in many, some countries such as Estonia and Ireland are beginning to adopt a more coherent activation strategy with menus of coordinated programs and services. Overall, however, there is room for a more sys- tematic approach to be taken to labor market policies. nn Integration of employment support policies with social assis- tance, public employment services, and social services is at  19 Portraits of Labor Market Exclusion 2.0 different stages across the countries studied. A holistic approach with interlinkages between employment and social policies is par- amount in addressing the complexity posed by multiple barriers to employment. In most countries, this link between activation and other social policies could be improved. In some countries, such as Portugal, the social assistance pillar is related to activation measures (for exam- ple, by making eligibility for the guaranteed minimum income scheme conditional on activation policies); in most, however, this approach is still relatively underdeveloped. Similarly, there are numerous sup- ply-side constraints that should be addressed in parallel with activation policies: for example, part-time measures targeted to young mothers are ineffective if the child care support provided is insufficient. Registration for public employment services should be made more accessible, using various outreach strategies to include all the labor market vulnerable and allow them to participate in activation measures. nn There is a relatively low level of national spending on active labor market programs. Croatia, for example, has one of the lowest shares of active labor market program spending against total labor market expenditure in the EU—and its total labor market expenditure is already a third that of the EU average. Spending is also particularly low in Italy, Poland, Romania, and Spain. Low spending levels translate into low coverage, and mean that many who need the support—even if iden- tified as a priority—cannot be covered by existing programs. nn Active labor market program spending is not always targeted to priority needs. The study found that spending in many cases was skewed toward relatively well-educated/skilled youth who are short-term unemployed as opposed to low-educated long-term unemployed with low work experience. Further, in some countries, despite the availability of a variety of active labor market programs, the majority of measures do not appear to address complex needs. In Bulgaria, for instance, the majority of employment promotion measures are limited to targeting the registered unemployed; and the number and availability of programs addressing complex needs such as mobility, access to child care, and access to work experience simultaneously appears to be limited. nn There are few monitoring and evaluation best practices in evi- dence. Few countries invest in monitoring and rigorous evaluation to improve the design of active labor market programs. Without systematic impact evaluation, it is not possible to distinguish effective measures from those that are not; this prevents efficient use of scarce resources. Countries tend to lack a national framework for identifying outcomes and assessing impacts. There is also little data exchange between insti- tutions that can aid in effective and efficient monitoring. 20  Portraits of Labor Market Exclusion 2.0 The challenges ahead W hile differences in policies and needs exist, common challenges emerge for countries in creating employment and social policies and programs that address their priority and other labor market vulnerable groups. These challenges fall into four interrelated policy areas, as summarized below and shown in figure 11. nn Targeting and tailoring the design of activation and employ- ment support policies and programs. Policy makers must take into account the different needs of, and barriers faced by, the individual at-risk Figure 11  Challenge areas groups in the labor market when they develop policy tools or program-level interventions. This is especially cru- cial in that the funding available for active labor market policies is limited. Rigorously comparing the coverage Targeting of current activation and employment support policies with the needs of individual groups will reveal scope to recalibrate spending on measures in Outreach  line with priority group needs. Needs Monitoring  can be assessed through a variety of profiling methods; the important thing is to ensure methodological consis- tency across programs, policies, and institutions. Integration The diversity of barriers and groups highlighted in this profiling analysis demands more effective targeting  21 Portraits of Labor Market Exclusion 2.0 of both employment services and activation and employment sup- port policies and programs. First, employment services need to be strengthened to respond to potential capacity constraints and improve the accessibility of employment measures. Programs should then be Targeting tailored with respect to the needs of the identified groups, supported by an adequate profiling system. Most of the time, policies target large groups, such as educated young people, without considering the unique mix of barriers they face—potentially lessening their efficiency. nn Outreach. Improving outreach, employment promotion, and individual- ized services are vital in reaching—in particular—those far from the labor market, such as inactive women and youth in remote and rural areas. A first step in so doing is to ensure they have information about and access to public employment offices, and that they register as job seek- ers so as to benefit from counseling services and available programs. Special attention should be given to the most vulnerable and social assistance beneficiaries depending on policy objectives. Links between  Outreach existing benefits (such as guaranteed minimum income or targeted child benefits) could be used to improve access for such individuals and therefore aid in their activation. Notably, joint outreach activities at the local level on employment promotion and specialized services targeting priority groups living in rural or remote areas could provide better cov- erage and consistency. Where public institutions do not have adequate capacity, private provision can be explored to boost outreach and deliv- ery capacity. Ongoing, demand-driven active labor market policies and services should be a priority. nn Integrated model. A menu of programs should be established in a systematic manner, avoiding duplication and overlaps an the one hand, and scarcity on the other. In several countries, there are numerous small and ad hoc programs that may target the same populations without  any coordination between them. A clear activation strategy—based on profiling exercises—with flagship programs and clear targets would help reduce potential fragmentation. Coordination among agencies at Integration the local and central levels should be reinforced, along with coordina- tion with institutions that provide services to vulnerable populations and the working poor. Given the profile of the priority groups and the multiple barriers they face, activating them will require a concerted effort and an integrated package of benefits; programs; and services such as improved job assistance, mobility incentives, training, flexi- ble work opportunities, and access to care facilities. Data and delivery linkages between institutions should be strengthened, and innovative approaches to integrated delivery implemented more systematically. 22  Portraits of Labor Market Exclusion 2.0 nn Monitoring and evaluation. Rigorous evaluations are critical if the targeting and design of activation and employment support policies and programs are to be properly and appropriately adjusted. In general, national employment agencies and ministries of labor have underdevel- oped systematic impact evaluation processes, including capacity to exchange data. This deficiency prevents them from identifying which Monitoring measures are more effective and represent better deployment of scarce resources. One potential solution lies in the administrative data already being collected in most countries. These data may be narrow in cover- age, but rich in other aspects, enabling conduct of cost-benefit and net impact analysis on a regular basis. Designing interventions with a robust results evaluation framework will allow identification of those design and implementation elements that work for particular target groups—and enable adjustment of existing programs accordingly. Taking a systematic approach to evaluating and monitoring activation policies would allow for improved cost-effectiveness and avoid pitfalls to generate better evaluation outcomes (such as investing in those who are closer to the labor market and likely to find jobs without any sup- port). Additionally, sharing and discussing experiences with other agencies within and outside of Europe will have cross-country benefits.  23 Portraits of Labor Market Exclusion 2.0 Notes 1. Information on the European Pillar of Social Rights priority pol- icy area can be found here: https://ec.europa.eu/commission/ priorities/deeper-and-fairer-economic-and-monetary-union/ european-pillar-social-rights_en 2. The overarching Understanding Employment Barriers to Inform Policy project is financed through separate agreements between the EC and the World Bank and the EC and the OECD, respectively. The EC–World Bank effort comprises six country policy papers (Bulgaria, Croatia, Greece, Hungary, Poland, and Roma- nia) under the rubric “Portraits of Labor Market Exclusion 2.0”; the EC-OECD initiative consists of six country policy papers (Estonia, Ireland, Italy, Lithuania, Portugal, and Spain) under the rubric “Faces of Joblessness.” 3. This definition of individuals with low levels of education as being those with less than a lower secondary education degree is as per the International Stan- dard Classification of Education (ISCED2). 4. The survey data used in the analysis were European Union Statistics on Income and Living Conditions (EU-SILC) data, where the reference period is equal to the previous calendar year. 5. While these indicators can capture broad aspects of the three main types of employment barriers identified in this framework, they do not offer a compre- hensive view of the labor market barriers. Rather, they represent the barriers that can be captured using EU-SILC data. 6. See OECD and World Bank (2016). 7. The stories of individuals presented here are composites derived from the empirical results, and individuals represent different groups. 24  Portraits of Labor Market Exclusion 2.0 References EC (European Commission). 2010. “Europe 2020: A strategy for smart, sustainable and inclusive growth.” EC, Brussels. http://eur-lex.europa.eu/LexUriServ/Lex- UriServ.do?uri=COM:2010:2020:FIN:EN:PDF Eurostat. 2017. Labour Force Survey. http://ec.europa.eu/eurostat/web/lfs/data/ database Ferré, C., H. Immervoll, and E. Sinnott. 2013. “Profiling of People with No or Lim- ited Labor-market Attachment, Latvia: Who is Unemployed, Inactive of Needy? Assessing Post-Crisis Policy Options.” World Bank, Washington, DC. Immervoll, H. 2013. “Profiles of Vulnerable People in Turkey.” In D. Finn, R. Grun, K. Herrera-Sosa, H. Immervoll, C. Ridao-Cano, G. Uysal, and A.L. Yener, Activating Vulnerable People into Good Jobs in Turkey. Washington, DC: World Bank. OECD (Organisation for Economic Co-operation and Development) and World Bank. 2016. “Understanding Employment Barriers: OECD–World Bank Joint Methodological Paper.” World Bank, Washington, DC. Sundaram, R., U. Hoerning, N. De Andrade Falcao, N. Millan, C. Tokman, and M. Zini. 2014. “Portraits of Labor Market Exclusion.” World Bank, Washington, DC. World Bank and OECD (Organisation for Economic Co-operation and Develop- ment). 2016. “Joint Methodology Paper.” Country Policy Papers (available at worldbank.org/en/region/eca/publication/ labor-exclusion-portraits and www.oecd.org/social/faces-of-joblessness.htm) Browne, James. 2017. “Faces of Joblessness in Estonia: Main Results and Policy Inventory.” Organisation for Economic Co-operation and Development, Paris. —. 2017. “Faces of Joblessness in Ireland: Main Results and Policy Inventory.” Organisation for Economic Co-operation and Development, Paris. Düll, Nicola, and Céline Thévenot, 2017. “Faces of Joblessness in Portugal: Main Results and Policy Inventory.” Organisation for Economic Co-operation and Development, Paris. Isik-Dikmelik, Aylin, Natalia Millán, and Mirey Ovadiya. 2017. “Portraits of Labor Market Exclusion 2.0: Understanding Employment Barriers. Romania Country Policy Paper.” World Bank, Washington, DC. Karacsony, Sandor, and Natalia Millán. 2017. “Portraits of Labor Market Exclusion 2.0: Understanding Employment Barriers. Hungary Country Policy Paper.” World Bank, Washington, DC. Karacsony, Sandor, Frieda Vandeninden, and Mirey Ovadiya. 2017. “Portraits of Labor Market Exclusion 2.0: Understanding Employment Barriers. Bulgaria Country Policy Paper.” World Bank, Washington, DC. Millán, Natalia, Mirey Ovadiya, Aylin Isik-Dikmelik. 2017. “Portraits of Labor Mar- ket Exclusion 2.0: Understanding Employment Barriers. Greece Country Policy Paper.” World Bank, Washington, DC.  25 Portraits of Labor Market Exclusion 2.0 Ovadiya, Mirey, and Frieda Vandeninden. 2017. “Portraits of Labor Market Exclu- sion 2.0: Understanding Employment Barriers. Croatia Country Policy Paper.” World Bank, Washington, DC. Pacifico, Daniele. 2017. “Faces of Joblessness in Italy: Main Results and Policy Inventory.” Organisation for Economic Co-operation and Development, Paris. —. 2017. “Faces of Joblessness in Lithuania: Main Results and Policy Inven- tory.” Organisation for Economic Co-operation and Development, Paris. Rodriguez, Fernando. 2017. “Faces of Joblessness in Spain: Main Results and Policy Inventory.” Organisation for Economic Co-operation and Development, Paris. Vandeninden, Frieda, and Karolina Goraus-Tan´ska. 2017. “Portraits of Labor Mar- ket Exclusion 2.0: Understanding Employment Barriers. Poland Country Policy Paper.” World Bank, Washington, DC. 26  Portraits of Labor Market Exclusion 2.0