BOSNIA AND HERZEGOVINA Social Protection Situational Analysis This project is funded by the European Union BOSNIA AND HERZEGOVINA Social Protection Situational Analysis © 2022 International Bank for This work is a product of the staff of The World Bank with external contri- Reconstruction and Development / butions. The findings, interpretations, and conclusions expressed in this The World Bank work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. 1818 H Street NW, Washington DC 20433 This publication was produced with the financial support of the European Telephone: 202-473-1000 Union. Its contents are the sole responsibility of World Bank staff and do Internet: www.worldbank.org not necessarily reflect the views of the European Union. The World Bank does not guarantee the accuracy of the data included in Rights and Permissions this work. The boundaries, colors, denominations, and other information The material in this work is subject shown on any map in this work do not imply any judgment on the part to copyright. Because The World of The World Bank concerning the legal status of any territory or the Bank encourages dissemination of endorsement or acceptance of such boundaries. 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 World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. ii BOSNIA AND HERZEGOVINA Social Protection Situational Analysis BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Table of Contents: LIST OF FIGURES iv LIST OF TABLES iv ACKNOWLEDGEMENTS v LIST OF ACRONYMS vi 1. Introduction 1 2. Poverty, labor market outcomes and demographics 3 3. Brief overview of the social protection system 6 in Bosnia and Herzegovina 4. Social assistance: Cash transfers and in-kind benefits to support 10 the poor, the vulnerable, and people with disabilities 5. Social services 16 6. Pensions 18 7. Employment and ALMPs 21 8. Learning from the COVID-19 pandemic response 24 9. Reform priorities and knowledge gaps 26 REFERENCES 29 iii BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Figure 1: Labor market indicators, Western Balkans (2018 & 2019) 4 Figure 2: Labor force productivity growth has stagnated 4 List of Figures: Figure 3: Bosnia and Herzegovina receives substantial remittances 5 Figure 4: Remittances dropped significantly during the COVID-19 5 pandemic and have slowly recovered Figure 5: Social protection coverage of total population, by 8 program type (2015) Figure 6: Expenditure in social protection programs, selected 9 countries (% of GDP) Figure 7: War veteran benefits spending as a share of total 11 social assistance expenditure is high in Bosnia and Herzegovina Figure 8: Spending on last-resort social assistance is particularly 11 low in Bosnia and Herzegovina Figure 9: Coverage of the poor (bottom quintile) with social 12 assistance is relatively low in Bosnia and Herzegovina compared to neighboring countries Figure 10: Adequacy of permanent social assistance transfers and 13 disability allowances Figure 11: Benefit incidence of last-resort income programs 14 (percentage of total benefits accruing to the poorest quintile) Figure 12: A large share of beneficiaries of the last-resort income 14 programs in BiH are just below or above the 20 percent consumption cut-off Figure 13: Support Ratio in EU and Western Balkans, 2016 or 19 Recent (Contributors per 100 Pensioners) Figure 14: Average Gross Pension Benefit Ratio, EU and Western 20 Balkans, 2019 or Recent Figure 15: Expenditure on labor market policies as a percent of 21 GDP, 2018 Figure 16: Unemployment benefits expanded as a response to the 25 crisis List of Tables: Table 1: Overview of the social protection system in Bosnia and 6 Herzegovina Table 2: Pension parameters in 2021 19 Table 3: Reform priorities 26 Table 4: Knowledge gaps 28 iv BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Acknowledgements This situational analysis has been prepared by a World Bank team composed of Alicia Marguerie, Gozde Meseli Teague, Zoran Anusic, Stefanie Brodmann, Meliha Kozaric Fanning, Cornelius von Lenthe and William Shaw. The team is grateful to Sarah Coll-Black, Mirey Ovadiya and Jamele Rigolini for providing comments on draft versions of the note and to Leonardo Ramiro Lucchetti and Carlos Gustavo Ospino Hernandez for supporting the poverty analysis. The team would like to thank the authorities in Bosnia and Herzego- vina for sharing feedback and views on findings and recommenda- tions. In particular, the team would like to thank the Federal Ministry of Labor and Social Policy, the Republika Srpska Ministry of Health and Social Protection, the Federal Employment Institute and the Republika Srpska Employment Bureau. Finally, the team appreciates the validation of findings and comments conducted by development partners at a consultation of the situational analysis in June 2022. The analysis was made possible by the generous funding from the Europe 2020 Trust Fund by DG NEAR. The note was prepared under the guidance of Linda van Gelder (Country Director for the Western Balkans), Christopher Gilbert Sheldon (Country Manager for Bosnia and Herze- govina) and Cem Mete (Practice Manager, Social Protection and Jobs Global Practice, Europe and Central Asia Region). v BOSNIA AND HERZEGOVINA Social Protection Situational Analysis List of acronyms ALMPs Active labor market programs BAM Bosnian Mark BiH Bosnia and Herzegovina CSW Center for Social Work EC European Commission ECD Early Childhood Development EU European Union FBiH Federation of Bosnia and Herzegovina FEI Federal Employment Institute GDP Gross Domestic Product HBS Household Budget Survey ILO International Labor Organization LFS Labor Force Survey NEET Not in employment, education or training PAYG Pay-as-you-go PES Public Employment Services PPP Purchasing Power Parity PSA Permanent Social Assistance RS Republika Srpska RSEB Republika Srpska Employment Bureau SPEED Social Protection Expenditure and Evaluation Database UNDP United Nations Development Programme vi BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Introduction While recovering from the COVID-19 crisis, Bosnia this note assesses each category of social protec- and Herzegovina (BiH) is struggling with persistent tion, namely: social assistance, social services, unemployment and a predicted rise in poverty. social insurance (specifically pensions), and labor Gross domestic product (GDP) is estimated to have market programs, in terms of program coverage, increased by 2.7 percent in the first quarter of 2022, equity, sustainability and effectiveness. Based on compared with a 4.0 percent increase in 2021 and this analysis, it proposes policy priorities and future 3.2 percent contraction in 2020.1 The unemploy- areas of focus. The note is based on an analysis of ment rate, which averaged 15.7 percent in 2019 administrative data and the most recent Household (before the crisis), remains high, exceeding 16.4 Budget Survey (HBS 2015) and Labor Force Survey percent in third quarter of 2021.2 Persistently high (LFS 2019), a literature review, and an engagement unemployment threatens to further increase the with technical experts in ministries and employment share of the population living in poverty, which was institutes. 16 percent in 2015 (most recent available estimate) and could increase by 4 to 10 percentage points as This analysis finds that unequal resource alloca- a result of the crisis. The poverty reduction chal- tion, limited targeting, administrative inefficien- lenge is exacerbated by the acceleration of inflation, cies and institutional fragmentation (Federation especially the increase in food prices which tends to of Bosnia and Herzegovina) reduce the impact of disproportionately affect poor households.3 social protection services in Bosnia and Herzego- vina. While considerable sums are spent on social To further reduce the incidence of poverty, meas- assistance, the poverty impact and cost-effective- ures to create more job opportunities for all BiH ness of social assistance spending are relatively low. citizens must be complemented by an effective Benefit levels are low and the coverage of the poor is social protection framework. Social protection is a limited, while administrative fragmentation reduces fundamental pillar of social policies. It is essential to cost-effectiveness and results in large disparities in protect people from falling into poverty and desti- benefits across political subdivisions. Last-resort tution; to help people cope with adverse idiosyn- income support programs and poverty alleviation cratic or covariate shocks and smooth consumption transfers have a particularly low cost-benefit ratio. over their lifetime; and to promote human capital An inadequate institutional framework impairs the accumulation and ensure better access to jobs provision of social services for people with disabili- for those who need support. Social protection ties, the elderly and children. While pensions should comprises a variety of policy tools, including cash be financially sustainable for the next 10-15 years, transfers to the poor and people with disabilities, further policy changes will be required to maintain social care services for vulnerable populations, labor sustainability beyond this period. Moreover, pension market programs and unemployment insurance, adequacy is low and declining, while the ageing of and old-age pensions.4 While each benefit or service the population and outmigration limit the ability addresses a specific need against which its perfor- to support pensioners. Few resources are devoted mance is assessed, it is also important to analyze to employment and labor market policies; the whether the system as a whole fulfills its objectives. coverage of active labor market programs (ALMPs) is limited; and programs are insufficiently targeted This note presents a situational analysis of the at vulnerable groups. Efforts to protect the popula- social protection system in Bosnia and Herze- tion from the COVID-19 pandemic were impeded by govina. It assesses the extent to which the social limited budgetary resources and a lack of flexibility protection system fulfills its purpose; it also in some social assistance measures. proposes policy priorities as well as areas for reform in the short, medium, and long term. To these ends, Several recommendations arise from this analysis. Back to table of contents 1 1 World Bank. 2022. “Western Balkans Regular Economic Report” Fall 2021. 2 Labor Force Survey, QIII of 2021. 3 Introduction paragraphs are based on previous World Bank reports mainly, World Bank “Western Balkans Regular Economic Report” Fall 2021, Spring 2021 and Spring 2022 as well as Systematic Country Diagnostic Update 2020. 4 Health insurance is not considered here, although it is a form of social insurance. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis A key goal must be to increase resource effec- tiveness, by expanding support devoted to social assistance, elderly- and child-care, and ALMPs. The coverage of the poorest households, in particular, should be increased in a cost-effective way. This entails improving targeting by shifting social assis- tance funds from war-related to poverty-tar- geted programs and shifting ALMPs from support for jobseekers (who are not “hard-to-serve”) to programs focused on unemployed “at-risk” popu- lations and their activation. In FBiH, incentives for formal employment could be increased by de-linking health insurance from unemployment and allowing social assistance beneficiaries to keep a larger portion of benefits as wages rise from low levels. Improved coordination and data exchange are needed across the entities, cantons and munic- ipalities that administer social protection programs, as well as with private sector providers of services. The management of ALMPs could be strengthened through greater focus on monitoring and evalua- tion, along with improvements in profiling, the use of technology in supporting job applicants and various business processes. Finally, disaster prepar- edness and response could be strengthened by putting in place contingency financing to fund both social assistance cash transfers and unemployment benefits and introducing the required legislation and implementation arrangements for emergency support programs for households at risk and newly unemployed. The remainder of the note is organized as follows. Section 2 reviews the country’s main poverty and labor market outcomes; section 3 provides a brief overview of the social protection system; section 4 looks at non-contributory cash transfers to support the poor, the vulnerable and people with disabilities; section 5 examines social care services; section 6 analyzes pensions; section 7 explores employment and active labor market programs; section 8 reviews the social protection response to the COVID-19 pandemic; and section 9 concludes by identifying knowledge gaps and policy priorities for reform. Back to table of contents 2 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Poverty, labor market outcomes and demographics Bosnia and Herzegovina faces significant labor market challenges. The job losses from the pandemic raised unemployment and increased poverty. Bosnia and Herzegovina has the second lowest ratio of employment to working-age population in the Western Balkans, while population, aging and outmigra- tion have reduced the number of working-age persons. A large gender gap impairs economic efficiency and equity. Productivity has stagnated and informality is high. Poverty remains an important challenge. While the gration, a low employment rate and an inactivity poverty rate in Bosnia and Herzegovina is low com- rate of 44.5 percent in the second quarter of 2019, pared to most other Western Balkan countries, the it is apparent that recent economic growth has not COVID-19 pandemic has resulted in significant job translated into more and better-quality jobs. losses and worsened poverty levels. Restrictions imposed in response to the COVID-19 outbreak A large gender gap in labor force participation is (banning of public gatherings, closure of education causing significant economic losses. In the second and hospitality facilities, travel restrictions, lock- quarter of 2019, women were more than 21 percent- downs, and stay-at-home orders) resulted in a large age points less likely to be employed than men, the number of layoffs, particularly in the service and second largest gap in the region (Figure 1, panel a). manufacturing sectors. According to official esti- Childcare is limited, which is a significant barrier to mates, in November 2020, the number of people in female labor force participation. Wages for women paid employment was down by about 1 percent year- are approximately 9 percent lower than for men. on-year (y-o-y), and the number of unemployed There are also significant barriers to women access- had risen by about 3 percent.5 Deeper, adverse labor ing productive assets, with gender gaps in property market effects stemming from the pandemic were ownership and access to bank accounts.10 Illustra- prevented by wage subsidy programs in both enti- tive of the concern, in a cross-country study, the ties and other policy measures targeted to affected low participation of women in the labor force is esti- economic sectors to prevent more job losses. Sim- mated to cost an average of 16 percent of national ulations undertaken at the beginning of the pan- GDP.11 demic indicated that, with falling consumption, an additional 35,000 to 85,000 individuals may have Labor market’s productivity growth has stalled, fallen into poverty.6 and informality remains high in some regions. Labor productivity has stagnated since 2015 Bosnia and Herzegovina has the second worst (Figure 2). Informality is at moderate levels for a labor market outcomes in the Western Balkans middle-income country but has not decreased over and a shrinking workforce. In the second quarter of the last decade, representing around 26 percent of 2019, only 46.4 percent of the working-age popula- total employment in 2017. Its prevalence is espe- tion were employed, the second lowest in the region cially high in rural areas, where 41.7 percent of the after Kosovo (Figure 1, panel a). Total employment employed are in the informal sector, in comparison declined from 820,300 in 2010 to 760,500 in the with 11.5 percent in urban areas. Informal employ- second quarter of 2019.7 The employment rate of ment is concentrated among people with primary the working-age population increased in the same education or less.12 time period, largely because of the fall in the work- ing-age population from 2.6 million in 2010 to 2.4 Bosnia and Herzegovina relies significantly on million in 2018 due to population aging and con- remittances in bolstering household income, which tinued outmigration.8 Emigration is most prevalent plunged during the pandemic. Transfers from the among the young, working-age population, with roughly 2 million Bosnian and Herzegovinian citi- 30 percent of those who left in 2018 having been zens living abroad make up about 9 percent of GDP between the ages of 18 and 35.9 With steady emi- (Figure 3). Remittances play an important role in Back to table of contents 3 5 World Bank. 2021. Western Balkans Regular Economic Report No. 19. Subdued Recovery. Spring 2021. 6 World Bank. 2020. Western Balkans Regular Economic Report No. 17. The Economic and Social Impact of COVID-19. Spring 2020. 7 Wiiw and World Bank. 2021. Western Balkans Labor Market Trends 2020. 8 SEE Jobs Gateway: https://data.wiiw.ac.at/seejobsgateway.html. 9 World Bank. 2020a. Bosnia and Herzegovina: Systematic Country Diagnostic Update. World Bank, Washington, DC. 10 World Bank. 2017. Promoting Women’s Access to Economic Opportunities: Bosnia and Herzegovina. World Bank. Washington DC. 11 Cuberes and Teignier. 2016. “Aggregate Effects of Gender Gaps in the Labor Market: A Quantitative Estimate.” Journal of Human Capital 10:1. 12 Informality figures based on LFS 2017 data, where informal workers are identified as employees having no pension or health insurance. Based on the definition used, informality varies between 25.4 percent (“no pension contribution and no health insurance”) and 27.7 percent (“no pension contribution or no health insurance”). BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Figure 1: Labor market indicators, Western Balkans (2018 and 2019) a. Employment rate b. Unemployment rate 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% WB6 ALB BIH MNE MKD SRB KSV WB6 ALB BIH MNE MKD SRB KSV Total 15-64 Male 15-64 Female 15-64 Total 15-64 Male 15-64 Female 15-64 Total 15-24 Male 15-24 Female 15-24 Total 15-24 Male 15-24 Female 15-24 c. Inactivity rate d. NEET rate, ages 15-24 90% 35% 80% 30% 70% 25% 60% 50% 20% 40% 15% 30% 10% 20% 5% 10% 0% 0% WB6 ALB BIH MNE MKD SRB KSV WB6 ALB BIH MNE MKD SRB KSV Total 15-64 Male 15-64 Female 15-64 Total Male Female Total 15-24 Male 15-24 Female 15-24 Source: SEE Jobs Gateway: https://data.wiiw.ac.at/seejobsgateway.html. Note: Data for employment, unemployment, and inactivity are for 2019 Q2; data for NEET are for 2018. Figure 2: Labor force productivity growth has stagnated 45 000 6% 44 000 5% 43 000 4% 42 000 3% 41 000 2% 40 000 1% 39 000 0% 38 000 -1% 37 000 -2% 36 000 -3% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 GDP per person employed growth (annual %) GDP per person employed (constant 2017 PPP $) Source: World Development Indicators (Accessed December 2021). Back to table of contents 4 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis supporting households with workers in the informal economy who do not have access to social insur- ance but also, due to their higher welfare, do not qualify for social assistance.13 Remittances dropped by 30 percentage points during the second quarter of 2020 (COVID-19 pandemic outbreak) relative to the 2019 average but have since slowly recovered to pre-pandemic levels (Figure 4). Government pol- icies focusing on the facilitation of remittances can help support the recovery. Figure 3: Bosnia and Herzegovina receives substantial remittances (2021) 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% ia ia na ia o vo gr on rb an vi so ne Se go lb ed Ko te A ze ac on er M M H th d or an N ia sn Bo Source: Global Knowledge Partnership on Migration and Devel- opment (KNOMAD). Figure 4: Remittances dropped significantly during the COVID-19 pandemic and have slowly recovered 140% 15% 130% 10% 120% 110% 5% 100% 0% 90% -5% 80% 70% -10% 60% -15% 21 1 1 1 2 2 21 3 1 2 19 3 4 2 4 20 1 20 2 20 3 2 4 20 1 Q Q 20 0 Q 20 Q 20 9 Q 20 1 Q 20 9 Q 20 Q 20 Q Q 20 Q 20 Q 19 20 20 Real GDP growth (RH) Remittances as % of 2019 average (LH) Source: Central Bank of Bosnia and Herzegovina (2019). Note: Remittances (left-hand axis) are estimates based on private transfers and compensation of employees as per BiH’s balance of payments. Right-hand axis corresponds to GDP growth. Back to table of contents 5 Kolev. A, and La, J. 2021. Financing the Extension of Social Insurance to Informal Economy Workers: The Role of Remittances. 13 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Brief overview of the social protection system in Bosnia and Herzegovina Many institutions, at the entity, cantonal, and local level, are involved in the design and provision of social protection. The social protection system is highly fragmented: programs are similar but not identical across entities and across cantons. About half of the population benefits from a social protection program, with the largest coverage by social insurance. Social protection expenditures are less than half the European Union (EU) average (as a share of GDP) but higher than in neighboring countries. Table 1 provides an overview of the BiH social objectives that range from income security in old protection system based on the four broad cate- age, to assistance for people with disabilities (work gories of social protection (social assistance, social related or otherwise) and income support to war services, social insurance and labor markets) and the veterans and other special categories arising from benefits and programs under each. Here, the term the war with limited contributions. These pensions “pension” is applied to a number of programs with are funded largely from the state budget. Table 1: Overview of the social protection system in Bosnia and Herzegovina Category Benefits and programs Federation of Bosnia and Herzegovina Republika Srpska (FBiH) (RS) • Permanent and One-off financial • Permanent and One-off financial assistance assistance Cash transfers to • Civil victims of war • Civil victims of war support the poor, • War veterans’ benefits • War veterans’ benefits (including the vulnerable • Family and child allowances veterans’ supplement) and people with • Funeral support • Family and child allowances disabilities • Funeral support Social assistance • Scholarships • Disability benefits • Scholarships • Disability benefits • Subsidies for housing, heating and • Subsidies for health health In-kind benefits • Food distribution • Free schoolbooks • Placement in a Social Welfare Institution • Placement in a Social Welfare Institution for adults and children in need for adults and children in need. Social services • Foster care • Foster care • Special training for people with • Home assistance and care for the disabilities in social welfare institutions elderly Pensions and • Old-age pension • Old-age pension other contributory • Survivor pension • Survivor pension Social insurance benefits • Disability pension • Disability pension • War veterans’ pension • War veterans’ pension Non-contributory • Maternity allowance • Maternity allowance • Unemployment benefits • Unemployment benefits Employment and • Wage subsidies • Wage subsidies labor market • Entrepreneurship support • Entrepreneurship support programs • Vocational training • Vocational training • Public works • Public works Source: Authors’ elaboration. Back to table of contents 6 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis The Federation of Bosnia and Herzegovina (FBiH), FBiH and RS have similar contributory and non-con- the Republika Srpska (RS) and the District of Brcko tributory programs (Table 1). In both entities, con- are responsible for social protection, with state- tributory pensions include old-age, disability and level institutions playing a coordinating role.14 In survivor pensions. In terms of non-contributory FBiH, the social protection system is decentralized social assistance benefits, both entities offer two across the ten cantons, while RS has a centralized last-resort social assistance programs based on social protection system. the frequency of the payment: permanent social assistance (PSA) and one-off financial assistance. In • In FBiH, the main institutions responsible for FBiH, permanent social assistance is paid from can- social protection and labor policies are the Min- tonal budgets while one-off financial assistance can istry of Labor and Social Policy (FMoLSP), the be paid from cantonal, city and municipal budgets. Federal Employment Institute (Public Employ- Both entities offer various family and child benefits ment Services (PES)) and the Ministry for Issues as well as comprehensive benefits for veterans and of The Veterans and Disabled Veterans. The civilian war victims.17 RS provides a non-contributory FMoLSP implements and coordinates laws and veteran supplement. On employment, both entities policies that include the payment of permanent offer active labor market programs for unemployed social assistance and maintaining a unique regis- target groups (such as wage subsidies or self-em- try of social beneficiaries for all non-contributory ployment programs), pay unemployment benefits payments. Programs for war veterans are under to eligible unemployed and offer vocational training. the mandate of the Federal Ministry of Veterans FBiH and RS also offer social services for adults and Affairs along with eight Cantonal Ministries and children in need. two Directorates for Issues of the Defenders. In addition, responsibility for labor and social protec- One in two individuals in Bosnia and Herzego- tion extends to cantons with ten Cantonal Minis- vina benefited from at least one social protec- tries for Social Policy, ten Cantonal Employment tion program in 2015 (Figure 5).18 This calculation Institutes (cantonal PES) and Centers for Social includes coverage by social insurance, social assis- Work (CSWs)15 which implement the social welfare tance and labor market programs (Box 1). Cover- support and coordinate social services. Although age was below the Europe and Central Asia (ECA) the federal ministry has a regulatory role, the average of around 60 percent. Total coverage is cantonal ministries have authority over adjust- slightly lower in FBiH, at 48 percent, than in RS, ing eligibility rules and benefits levels for existing at 56 percent. Coverage is mostly driven by social programs as well as introducing new programs. insurance programs, with lower levels for social Cantons can also partially or completely finance assistance programs and almost non-existent cov- programs. erage for labor market programs. In FBiH, social assistance programs reach 13 percent of the total • In RS, the main institutions responsible for social population in comparison to the ECA average of protection and labor policies are the Ministry of around 30 percent. While social assistance reaches Health and Social Welfare; the Ministry of Labor, 25 percent of the RS population, these programs War Veterans, and Disabled Persons’ Protection; have targeting issues. RS Employment Bureau with six regional insti- tutes; and CSWs in 51 municipalities. The first two oversee the design, administration, and deliv- ery of programs in the entity. The municipalities develop annual and mid-term social welfare pro- grams based on an analysis of the social status of citizens in their area of jurisdiction. The programs are financed by RS budget with co-financing from the municipalities, and a dedicated Public Fund for Child Protection finances family and child allow- ances. CSWs are the essential institutions of social welfare; they implement the majority of social welfare measures and coordinate the activities of all social services in local communities, including non-government and government sectors.16 Back to table of contents 7 14 Due to limitations of this analysis, the independent social protection system of the autonomous District of Brcko was not assessed separately. 15 In FBiH, CSW’s are funded by cantons and cities. There are 2 cantonal CSWs and 71 municipal CSWs. 16 Sucur-Janjetovic, V., Kurta. A. and Oruc, N. 2017. Social Protection System in Bosnia and Herzegovina: Country Brief. A Western Balkan Regional Initiative, December 2017. 17 Those benefits slightly differ across entities and the level of benefit is not uniform across FBiH cantons. Family and child benefits include paid maternity leave for employed mothers, maternity allowance for non-working mothers, child benefits and newborn assistance (in FBIH, one-time support for newborn equipment and in RS, a one-time cash benefit for each newborn child and a one-off allowance for every third and fourth child). In 2019, RS also introduced a caregiver’s allowance for parents of children who need constant care. In 2021, FBiH developed a new draft Law on Support to Families with Children proposing a uniform maternity allowance (for non-working mothers) and a uniform child assistance benefit across cantons, both being means tested. 18 While the most recent household survey available is from 2015, the country’s social protection system has not changed markedly since then. Thus, the 2015 results are still relevant. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Figure 5: Social protection coverage of total population, by program type (2015) 70 63 60 56 51 50 48 47 41 Coverage (%) 40 40 40 31 30 25 20 17 13 10 1 1 0 1 0 Total FBiH RS Brcko All social protection All social insurance All social assistance All labor market programs Source: World Bank Social Protection Expenditure and Evaluation Database (SPEED) using HBS 2015 data. Box 1: Data limitations for poverty assessment and the measurement of social assistance performance This analysis of the main drivers of poverty and the performance of social protection programs, including the impact of the recent COVID-19 pandemic, has been constrained by a lack of up-to-date comparable data in Bosnia and Herzegovina (BiH). BiH is the only Western Balkan country without an EU Statistics of Income and Living Conditions (EU-SILC) survey; a national round of the survey was never implemented in the country. The EU-SILC measures inter- nationally comparable income poverty data in the other Western Balkans and EU economies. As such, the EU-SILC survey would allow BiH to use a welfare and poverty measure comparable to that of other coun- tries in the region. The BiH Statistical Office (BHAS) has published poverty reports using relative poverty measures for 2004, 2007, 2011 and 2015. These reports use a standard international consumption-based methodology. Nationally and entity-representative Household Budget Surveys (HBS) have been implemented to measure consumption poverty since 2004. These surveys have been spaced about four years, with the 2015 HBS being the latest poverty data available. The country recently finished collecting the new 2021/22 HBS survey, but the results are not yet publicly available. Data on households along with their level of consumption and benefits received, as provided in the HBS, are therefore used to analyze the cost-efficiency, coverage and targeting of different social protection measures. The results of the 2015 survey are still valuable, as the social protection programs and structures have not changed markedly. To better understand the performance over time of the overall population in general, and the less well-off, regular and broadly available comparable household data, including information on income, are necessary. Recent shocks, such as the COVID-19 pandemic, illustrate the need to collect surveys on a regular basis and to make the information available publicly in a timely manner. Back to table of contents 8 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Total social protection spending is above the Western Balkan countries (Figure 6). Overall, roughly Western Balkans average due to higher levels of two-thirds of social protection spending is sourced spending on pensions and on veteran and civilian from FBiH while the remaining spending is sourced war victim benefits. In 2017, Bosnia and Herzegovina from RS. This correlates with the population differ- spent around 12.3 percent of GDP on social protec- ence between the two entities: 2.3 million in FBiH tion benefits, including 2.7 percent on social assis- compared to 1.3 million in RS. The highest spending tance, 9.1 percent on social insurance and 0.5 percent is dedicated to pension programs, such as old-age on labor market programs. This spending level is pensions, in both entities. For non-contributory ben- lower than the EU average of 26.5 percent, while still efits, the largest spending is for veteran benefits, fol- higher than the 10.8 percent of GDP average among lowed by civilian war victim support. Figure 6: Expenditure in social protection programs, selected countries (% of GDP) 16% 14% 12% 10% % of GDP 8% 6% 4% 2% 2.7% 0% SVK KSV ALB MNE MKD BGR BIH SRB ROU WB6 2017 2017 2019 2019 2019 2017 2018* 2019 2017 Average - latest year*** Labor market programs (%) Social assistance (%) Social insurance (%) Social care services (%) Source: World Bank SPEED (Accessed January 2022). Note: *BiH uses 2018 social insurance and labor market data and 2017 for social assistance and social care services due to data availability. *** Simple arithmetic average of the total SP spending (%) among the six Western Balkans countries on the chart: Albania, Serbia, Montenegro, North Macedonia, Kosovo and Bosnia and Herzegovina. Back to table of contents 9 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Social assistance: Cash transfers and in-kind benefits to support the poor, the vulnerable, and people with disabilities19 Total social assistance spending in Bosnia and Herzegovina is high relative to other countries in the region, but most resources are devoted to pensions and veteran and civilian war victim benefits. Coverage of and benefit levels for the poor are low. In fact, the cost-benefit ratio of the last-resort income programs responsible for poverty alleviation (permanent and one-off social assistance) is the lowest in Europe and Central Asia, pointing to inefficiencies in the targeting of these benefits. The fragmentation of the pro- grams further contributes to the low cost-effectiveness of social assistance spending and creates dispar- ities across cantons and entities. A comprehensive vision of social assistance spending should be devel- oped, including a new poverty-targeted approach across programs to increase the cost-effectiveness of and to reduce the inclusion of non-poor in social assistance. A highly fragmented social assistance system Bosnia and Herzegovina does not have one com- differing programmatic design parameters—such as prehensive social assistance system; rather its eligibility rules and benefit levels. entities and Brcko District BiH have distinct social assistance programs. FBiH adopted five laws that The social assistance systems in FBiH and RS share comprise its system: Law on Social Protection, Pro- similar objectives and instruments, with a strong tection of Civilian Victims of War and Protection focus on programs for war veterans and their fam- of Families with Children;20 Law on the Rights of ilies. Both systems aim to address similar risks and Defenders and Members of their Families;21 Law on meet similar objectives, including: (i) poverty reduc- the Rights of Medal Holders;22 Law on Foster Care;23 tion, (ii) assistance to families with children, (iii) assis- and Law on Parents Caregivers.24 Under these laws, tance to individuals with disabilities (including war the ten cantons have the authority to expand the veterans with disabilities) and (iv) assistance to war policies (e.g., special programs like food distribution, veterans and their families and to civilian victims of heating subsidies and school transport subsidies) war. In total, there are close to 80 different social and to redefine eligibility criteria but they do not assistance benefits, for which rules differ across have the authority to lower benefits. RS adopted entities and within FBiH across cantons (see Table 1 three laws that comprise its system: Law on the for an overview). The two entities differ on measures Rights of Veterans, War Invalids and Families of supporting families and veterans or victims of war.27 Killed Veterans of the Defense and Patriotic War of the Republika Srpska; Law on Social Protection of Spending on social assistance skews Republika Srpska; and Law on Child Protection of the Republic of Srpska. Brcko District administers to war veteran and disability benefits social assistance based on separate laws.25 BiH spends more on social assistance than most other ECA countries. In 2017, social assistance In FBiH, the decentralized structure of institutions expenditure, including cash transfers and in-kind generates administrative inefficiencies and results benefits, but excluding social services, represented in different benefits across cantons and municipal- 2.7 percent of GDP (of which 1.8 percent in FBiH and ities. At present, only programs covering disability 0.9 percent in RS) (Figure 6). This makes BiH the benefits (including care and orthopedic benefits) for second-largest spender on social assistance in the war veterans and civilians26 are financed at the entity Western Balkans (after Kosovo) and the sixth largest level and have harmonized rules across the ten can- in the ECA region. tons. A large number of local ministries manage the other social assistance programs with mixed sources Spending on war veteran benefits as a share of of financing —entity, canton and municipal—and total social assistance expenditure in BiH is the Back to table of contents 10 19 This section draws on World Bank (2018b). 20 Law on Social Protection, Protection of Civilian Victims of War, and Protection of Families with Children was further amended in 2004, 2006, 2009 and 2016. 21 Law on the Rights of Defenders and Members of their Families was adopted in 2004, amended in 2005, 2007, 2010 and 2017. 22 Law on the Rights of Medal Holders was adopted in 2005, amended in 2006, 2010 and 2017. 23 Foster Care Law was adopted in 2017. 24 Law on Parents Caregivers was adopted in 2021. 25 UNICEF. 2017. Situation Analysis of Children with Disabilities in Bosnia and Herzegovina. November 2017. 26 Program names in the local language are ‘Prava ratnih vojnih invalida i prava članova porodice poginulog, umrlog, nestalog branioca i umrlog RVI’, ‘Prava dobitnika ratnih priznanja i odlikovanja’, ‘Temeljna prava osoba sa invaliditetom’, and ‘Prava civilnih žrtava rata’, which include disability benefits, long-term care and orthopedic benefits. Also, a program for civilian victims of war, which is financed in the proportion of 70 percent FBiH and 30 percent cantons, has harmonized rules in FBiH. Since the beginning of 2018, there have been ongoing discussions on harmonizing the child allowance at the federation level. 27 RS has a categorical war veteran allowance. Regarding family benefits, assistance for newborns differs: in FBIH, one-time support for newborn equipment and in RS, a one-time cash benefit for each newborn child and a one-off allowance for every third and fourth child. Since 2019, RS also introduced a caregiver’s allowance for parents of children who need constant care. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis highest in the ECA region (Figure 7). Spending on assistance spending has declined as a share of GDP assistance to war veterans, war veterans with disa- since the mid-2010s. While recent, comprehensive bilities and their families represented 1.6 percent of data are not available, social assistance spending GDP in 2017, representing more than 61 percent of was roughly constant in real terms from 2014 to total social assistance spending.28 Disability bene- 2017 but fell from 3 percent of GDP to 2.7 percent of fits fell to about 0.6 percent of GDP in 2019. Spend- GDP. More recent data show that disability benefits ing on all other programs represented about 0.3 and family and child allowance programs continued percent of GDP in 2017. to fall as a share of GDP through 2019. Spending on last-resort social assistance, which is targeted at Targeted benefits in BiH represent a small minority the poor in the form of permanent social assistance of social assistance spending, and social assistance and one-off financial assistance, is particularly low spending, in general, has been declining. Social (Figure 8). Figure 7: War veteran benefits spending as a share of total social assistance expenditure is high in Bosnia and Herzegovina 70% 61% 60% 50% % of SA spending 40% 36% 30% 26% 19% 19% 20% 10% 3% 3% 3% 2% 1% 1% 0% 0% 0% 0% BIH KSV KGZ MNE AZE MDA TUR ARM KAZ ROU ALB GEO SRB MKD 2017 2017 2017 2012 2014 2017 2016 2017 2017 2017 2019 2018 2019 2019 Source: World Bank SPEED (Accessed January 2022) and World Bank calculations. Figure 8: Spending on last-resort social assistance is particularly low in Bosnia and Herzegovina 50% 45% 45% 40% 35% 32% Percent of GDP 30% 26% 24% 25% 25% 21% 20% 15% 10% 10% 5% 0% WB6 BIH 16 MNE 17 MKD 19 SRB 20 ALB 20 KSV 19 Source: World Bank staff estimates based on national statistical data (2021). Note: Last-resort social assistance measures include permanent social assistance and one-off financial assistance for both enti- ties. WB6 is a simple average of the six Western Balkan countries. Back to table of contents 11 28 More recent data on spending are not available for war veteran benefits. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis The poor are not benefiting from the veterans and non-war veterans) cover 5.5 percent high spending on social assistance of the poorest quintile (and only 0.8 percent of the richest quintile), explained by the high correlation Only about a third of the poor are covered by social between disability and poverty in Bosnia and Her- assistance in BiH. Coverage of the poor with social zegovina and, in general, in the Europe and Central assistance is higher in Bosnia and Herzegovina than Asia region.31 in neighboring Western Balkan countries, although lower than in most other ECA countries (Figure 9). Even poverty-targeted programs have extremely About 32 percent of those in the poorest quin- low coverage of the poorest quintile. The last-re- tile benefited from at least one social assistance sort income programs (PSA and one-off financial program, which placed Bosnia and Herzegovina assistance) and the child allowance cover a small 16th among the 24 ECA countries. share of the total population, at 1.9 percent and 4 percent of the BiH population, respectively.32 More Pro-poor spending represents a small fraction of importantly, those programs are designed to reach social assistance spending in BiH. The two largest the poor (poverty-targeted) but exhibit low cover- programs in terms of spending, namely war veteran age rates of the bottom quintile. Child allowances benefits and disability benefits, are not pover- cover 10.3 percent of the poorest quintile (versus ty-targeted. While spending on war veteran bene- 1.4 percent of the richest quintile). Coverage of the fits has been declining over time, a significant share last-resort income programs is 4.8 percent of the of individuals continues to receive benefits.29 The poorest quintile (versus less than 1 percent in each war veteran allowance (present in RS only) has the of the richest three quintiles). In comparison, cover- largest population coverage of social assistance age by last-resort income programs of the poorest benefits (5 percent of BiH population and 14.1 quintile is significantly higher in other ECA countries percent of RS population), followed by war veteran (20 percent in Albania and 10 percent in Serbia, for benefits (4.5 percent of the population in BiH) which example).33 The child allowance (in both BiH enti- mainly consist of disability, orthopedic and long- ties) and permanent social assistance (in FBiH) are term support benefits for veterans. However, those means tested (except for the child allowance in two benefits cover only 4.8 percent and 8.5 percent Canton 2 which is universal).34, 35 In RS, permanent of the poorest quintile, respectively, while covering social assistance is means tested and restricted to 5 percent and 3.2 percent of the richest quintile, individuals unable to work and households with no respectively.30 The disability transfers (both for war members able to work.36 Figure 9: Coverage of the poor (bottom quintile) with social assistance is relatively low in Bosnia and Herzegovina compared to neighboring countries 100 90 80 70 60 Coverage 50 40 30 20 10 0 SVK-2009 HUN-2007 GEO-2018 ROU-2016 UKR-2018 LVA-2010 LTU-2008 BGR-2007 MDA-2018 POL-2016 HRV-2008 KAZ-2018 TUR-2019 AZE-2015 ARM-2018 BIH-2015 SRB-2018 MKD-2017 KSV-2017 KGZ-2018 ALB-2017 MNE-2015 TJK-2011 BLR-2018 Source: World Bank SPEED (accessed 2022). Notes: Non-contributory basic pensions and war-related pensions are not included. Figures for Kosovo represent the Social Assis- tance Scheme only. Figures for Albania represent the Ndihma Ekonomike only. “Bottom quintile” refers to the bottom quintile for consumption per capita. Back to table of contents 12 29 Data presented are based on HBS 2015, which partially captured (comparing with administrative 33 World Bank SPEED (Accessed January 2022). spending) the SA programs. While the child allowance and poverty alleviation are relatively well- 34 FBiH: Benefit formula and eligibility rules vary canton by canton in FBiH and are presented in detail represented, the spending on war veterans and disability benefits are not well-represented (indicating in World Bank (2018), annex 3. In most cantons, eligibility to the child allowance is based on the potential underreporting). total household revenues not exceeding a certain threshold (percentage of the base, again varying 30 Beyond the coverage of individuals, the spending is also not directed on the poor, with 25.7 percent by canton). Eligibility to Permanent Social Assistance is based on the difference between the total of the spending on war allowances received by the richest quintile and, likewise, 15 percent of the household revenues and a minimum household revenue (subsistence), set by each canton, as well as spending on war veterans benefits received by the same quintile (based on HBS 2015). ineligibility conditions related to verifiable assets (possession of land/house and of any motor vehicle). 31 In Bosnia and Herzegovina, looking by quintile, the highest share of people with disabilities is in the 35 With the recent adoption of the Law on Financial Support to Families with Children in FBiH (published first quintile—representing 12 percent of the total population. Additionally, Bosnia and Herzegovina July 1st, 2022), it is expected that the coverage of the child allowance will increase (with relaxed (with Bulgaria) has the highest level of disability in the Europe and Central Asia region (Mete 2008). As eligibility criteria) as well as its adequacy (raising the allowance to the monthly amount of 103.17 BAM) reviewed by Braithwaite and Mont (2009), people with disabilities have higher risk of poverty in most with a shift from cantonal to Federation budget financing. European and Central Asian countries. 36 RS: Eligibility for a household is based on the number of dependents and on the condition that there 32 The overall coverage of last-resort income programs by entity is 1.3 percent in FBiH and 2.4 percent are only non work-able adults in the household. In addition, the household total income should be in RS. below a certain (subsistence) threshold and no alternative sources of subsistence. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis The coverage and level of benefits vary significantly is despite the fact that in 2019, the minimum wage across and within entities. According to 2015 in RS (BAM 450 or approximately USD 235) was data, coverage of last-resort income programs higher than in FBiH (BAM 407 or approximately was highest in Canton 2 (Posavski, 6 percent of USD 213). In both entities, the disability allowance population) and RS (2.4 percent of the population), corresponds roughly to 25 percent of the minimum but only 0.3 percent of the population in Canton wage while the ratio changes with the degree of 7 (Hercegovacko-Neretvanski). In terms of benefit disability. levels, there does not seem to be any correlation between average consumption level in each Low targeting performance limits political subdivision and the size of the transfers, nor the poverty impact of social indications of deliberate decisions to invest more in the poorest areas.37 The likelihood of the benefit assistance programs lifting individuals out of poverty is thus uneven Bosnia and Herzegovina performs worse than most across the country. Similarly, the income eligibility Western Balkan countries in terms of the share criteria for child allowances varies in each canton, of benefits that goes to the poor. The poorest leading to wide disparities in coverage across 20 percent received 61.5 percent of the spending cantons.38 Despite the regulatory framework and on poverty alleviation benefits (PSA and one-off laws with clear objectives, the implementation of financial assistance) (benefit incidence) (Figure 11); benefits at the canton level is driven by availability 50 percent of program beneficiaries were in the of funds rather than actual needs, which creates poorest quintile. This means that half of beneficiar- disparities in terms of access to social assistance.39 ies of last-resort social assistance are not among Permanent social assistance (PSA) acts as a last- the poorest households in Bosnia and Herzegovina. resort income, but the transfers are inadequate. The child allowance and disability benefit share The level of PSA and disability allowance benefits similar performance than poverty alleviations ben- compared to the minimum wage and the old-age efits (respectively 59 percent and 57 percent of the pension levels provides insights on the (in)adequacy spending received by the poorest 20 percent). The of the transfers to support the basic needs of the best-targeted program according to 2015 data is poorest households (Figure 10). In 2019, the average the maternity benefits and newborn package,40 PSA transfer per household represented between 12 where 70 percent of total spending on benefits and 35 percent of the minimum wage in FBiH (11 to was received by the poorest quintile. However, the 32 percent of average old-age pensions), varying by program is quite small, reaching only 0.38 percent of canton. The PSA benefit level is more substantial in the population. Changes were subsequently intro- RS where for a two-person household, the benefit duced in RS to improve child protection (new Law on represents 40 percent of the minimum wage. This Child Protection adopted in 2017) and require more Figure 10: Adequacy of permanent social assistance transfers and disability allowances a. FBiH - Average PSA per family as a share of minimum wage b. RS - Average PSA and disability allowance as a share of and average old-age pension, 2019 minimum wage and average old-age pension, 2019 40% 80% 35% 70% 30% 60% 25% 50% 20% 40% 15% 30% 10% 20% 5% 10% 0% 0% one person two people three people four people 70 percent disability 100 percent disability na na zla oj je ia a na vo 10 tv sn in ob Sa vi vi je Tu on t H ere dr Bo sa go ra -D - nt o na Sa Po -N ze -P ca l Ca ra U er na ni ia nt sn Ze vi Ce go Bo es ze W er H Permanent Social Assistance Disability allowance (household type) (by degree) Percent of minimum wage Percent of average old age pension Percent of minimum wage Percent of average old age pension Source: Authors’ calculations based on legislation on benefits and on pension and minimum wage levels as of 2019. Back to table of contents 13 37 FBiH: To date, benefit formula and eligibility rules vary canton by canton in FBiH and are presented in detail in World Bank (2018b), Annex 3. In most cantons, the child allowance is a percentage of a cantonal base, usually the average cantonal salary of the previous year. Both the base and the allocation percentage vary canton by canton. The PSA benefit is set as a percentage of a base (average cantonal salary of the previous year), and this percentage is set by each canton and varies. Each canton set additional rules adjusting the benefit level to certain conditions (e.g. disability, number of household members). 38 In 2015, cantons 1, 2, 7, 8, and 10 did not provide child allowance benefits, while the benefit levels in cantons 5, 6, and 9 were around 200 BAM compared to 154 in canton 3 and 60 in canton 4 (cantons 2 and 7 began implementing the program in 2017, but we lack information on benefits levels). Recently adopted legislation (Law on Financial Support to Families with Children, published July 1st, 2022), with child allowance being funded at Federation level, will reduce geographical disparities and guarantee a monthly entitlement of 103.17 BAM per child for eligible households. 39 World Bank (2018b). 40 These programs benefit mothers of newborn children and include both means-tested (some maternity benefits) and non-means-tested (one-time support for each newborn and a one-off allowance for every third and fourth child) components. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Figure 11: Benefit incidence of last-resort income programs (percentage of total benefits accruing to the poorest quintile) 90% 74.7% 76.6% 77.1% 80% 73.5% 70% 61.5% 60% 52.3% 50% 40% 30% 20% 10% 0% MKD 2016 BIH 2015 SRB 2018 KSV 2017 MNE 2015 ALB 2018 Source: World Bank SPEED (2016-2018). Notes: “Benefit incidence” refers to the percentage of total benefits accruing to the poorest quintile households. Non-contributory basic pensions (“social pensions”) and war-related pensions are not included. Figures for Kosovo represent only the Social Assis- tance Scheme (SAS). recent data to be assessed. Veteran allowances in lowest compared to last-resort income programs in RS have the worst targeting of the poor among all other ECA countries. Mediocre targeting accuracy is social assistance programs according to 2015 data. the main driver. Almost 40 percent of households receiving the poverty alleviation transfers (PSA and Overall, the poverty impact of each social assis- one-off financial assistance) are above the poorest tance programs is low due to poor targeting, quintile (Figure 12). In addition, many households limited coverage of the poor or low benefit level. fall just below the poverty line, resulting in small per At the national poverty line of 205 BAM (approx. capita poverty gap reductions. Taken together, this USD 107) per capita per month, in the absence of all leads to low cost-effectiveness of the BiH’s last-re- social assistance transfers, the poverty rate would sort income support. increase by 1.9 percentage points. The highest impact on poverty is achieved by the war veteran Figure 12: A large share of beneficiaries of the benefits (including disability, survivor, long-term last-resort income programs in BiH are just below care and orthopedic): in the absence of the program, or above the 20 percent consumption cut-off 0.5 additional percent of the population would fall into poverty.41 However, war veteran benefits are not targeted to the poor, so the high level of spend- ing (total and per capita) implies a high cost for the poverty reduction achieved, and therefore a low cost-benefit ratio (0.18 for the war veteran bene- fits and even lower at 0.06 for the war allowance).42 The highest cost-benefit ratios, which specify how much the poverty gap is reduced by each BAM spent on the program, are found for the maternity and newborn package (cost-benefit ratio of 0.5) and child allowance (cost-benefit ratio of 0.45). Surprisingly, last-resort income programs (PSA and one-off financial assistance) are among the Source: Bosnia and Herzegovina HBS 2015 results based on least cost-effective social assistance programs in authors’ calculations. BiH. The cost-benefit ratio of the last-resort income Notes: Welfare aggregate is calculated using poverty team’s programs was 0.2 in 2015, implying that only 20 aggregate gallT minus the specified program on the figure to simulate the absence of the specific program. Two histograms percent of this spending went directly to reduc- look at the beneficiaries of last-resort income support (LRIS) ing the poverty gap. This cost-benefit ratio is the and those in Q1 that do not receive LRIS. Back to table of contents 14 41 Based on mechanical statistical simulation, which does not take into account behavioral responses toward such changes. 42 The poverty gap reduction for each 1 unit spent on the social program. The cost-benefit ratio looks at how much a unit of spending reduces poverty, that is, a score of 1 means that BAM 1 spent is reducing the poverty gap by BAM 1. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis A poverty-based approach can increase the cost-effectiveness of social assistance spending in BiH. In RS, the PSA only targets households whose members are unable to work; the PSA does not target households whose members are working (i.e., the working poor).43 In FBiH, PSA eligibility is meas- ured by total household revenues and a threshold (i.e., minimum household revenue).44 In terms of beneficiary identification, improvements can be made to effectively identify the poorest. In countries with limited administrative data and high levels of informality, a proxy means test (PMT) formula has proven to be more effective than cat- egorical targeting or targeting based on reported income.45 To improve the cost-effectiveness of social assistance spending, a PMT formula would be appropriate. A PMT formula can be applied across a number of social assistance programs, including last-resort social assistance, family and child allow- ances and war-related benefits.46 Further, the FBiH and RS single registries need to be updated and maintained. Each entity has devel- oped its own unique registry for non-contributory payments.47 The objective is to centralize infor- mation on all non-contributory cash payments to individual beneficiaries (citizens) who are paid from government and institutional budgets in order to improve efficiency, effectiveness, coverage and tar- geting in social protection. The COVID-19 crisis demonstrated the value of an up-to-date single registry, which was not the case in the BiH entities. In FBiH, the social registry was established based on the 2017 Law on Unique Reg- istry of Cash Payments Beneficiaries. In 2019, the Federal Ministry of Labor and Social Policy began implementing the law, and across all levels of FBiH government. The ministry entered into an agree- ment with the USAID E-Governance Project to cen- tralize mandatory identification documents, data on income and household property, expenditures, and household social and health needs, with the objec- tive of providing more efficient and accurate social assistance. One current limitation of the unique social registry in FBiH is that there is no legislation making compulsory the registration of benefits pro- vided at the municipal level (e.g., one-off financial assistance). Back to table of contents 15 43 See footnote 36. 44 See footnote 34. 45 Simulations for FBiH, however, show that the low coverage of the last-resort income support programs (PSA and one-off financial assistance) precludes a PMT to work efficiently. Using a PMT formula for targeting beneficiaries makes economic and administrative sense if the program coverage ranges between 15 percent and 20 percent. 46 World Bank. 2016. Formula or no formula? That is the question: Targeting social assistance beneficiaries in the Federation of Bosnia and Herzegovina. Washington DC. 47 In each entity, the social registry is referred as “SOTAC”. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Social services Bosnia and Herzegovina lacks an institutional framework to guarantee adequate social services for children and people with disabilities. In particular early childhood development services are inadequate and access is unequal. While foster and institutional care services for children have improved, more emphasis is required on preventive services and avoiding the unnecessary institutionalization of children. The coverage of people with disabilities by services is unequal, with inadequate services for individuals, particularly children, whose disability was not caused by war. Services for the elderly are limited and not commensurate with the growing share of elderly in the population. The delivery of social services at the local level is ipal budgets, creating geographical inequalities as underdeveloped and affected by the low level municipalities which are less economically devel- of financial resources, impacting the amount of oped allocate lower and irregular funds to pre- human resources and the availability of services school education. Overall, the absence of regular, for those in need. Social services are overburdened reliable or appropriate funding mechanisms struc- because of limited time and resources. This affects turally undermines the sustainability of early child- their quality and accessibility, especially in rural hood education. Multi-sector financing and pub- areas and for marginalized groups (Roma minorities, lic-private partnerships have not yet been estab- among others). The financing of social services is lished. only about 1 percent of social protection spending, and most of the funding goes to institutionalized Foster care and institutional care for children have care. Access to community-based social services seen significant improvements over the past few is lower in rural areas and in smaller communities.48 years, but preventing institutionalization needs While case management is in place in some CSWs, greater attention. The development of standardized it is not legally mandatory in FBiH. An assessment case management tools for Centers for Social Work of social services carried out by UNDP in 2020 (CSWs) has improved the social services for children. concludes that the current funding arrangements In both FBiH and RS, guidelines for Child Protec- cannot ensure the equitable provision of essential tion Case Management were developed in 2018 social services for all. and have been rolled out in all RS municipalities and several FBiH cantons. The foster care system has Access to early childhood development (ECD) ser- been significantly improved through legislation and vices and in particular early childhood education regulations, professionalization of the social service remain unequal, and more investment is needed to workforce in foster care and education, and foster cover all children. ECD services are not fully institu- parent education and certification. Child and family tionalized and have significant inequities in access, welfare services have started replacing long-term with lower attendance rates among families in rural institutional care. Yet, there remains a lack of pre- areas, families with unemployed parents and Roma ventive services. Unnecessary placement into alter- families. While FBiH integrated ECD into the primary native care needs to be reduced, especially among healthcare package in 2018, there is wide variation children with disabilities who are not referred to across the ten cantons because the cantons are appropriate services or care arrangements. More responsible for funding the package. RS adopted investments are needed to deinstitutionalize chil- the Multisectoral Programme for Early Childhood dren through preventive and reintegration efforts. Development 2016-2020, with the commitment to focus on marginalized groups. There have been Social services for people with disabilities are improvements in pre-primary programs, with close unequal. Bosnia and Herzegovina lacks a clear to 80 percent of children entering primary school social model for disability, including no standard- enrolled in pre-primary education in 2018/19.49 But ized methodology to assess disability. There are preschool education is often financed from munic- three categories for the consideration of disabili- Back to table of contents 16 48 UNICEF. 2017. Situation Analysis of Children with Disabilities in Bosnia and Herzegovina, November 2017; UNDP. 2020. Social Inclusion in Bosnia and Herzegovina: 2020 National Human Development Report – Extended Summary; UNICEF. 2020. Situation Analysis of Children with Disabilities in Bosnia and Herzegovina, March 2020. 49 Enrollment of all children of pre-school age (3-6 years old) is much lower: 25 percent in 2018/19. (UNICEF. 2020. Situational Analysis of Children in Bosnia and Herzegovina.) BOSNIA AND HERZEGOVINA Social Protection Situational Analysis ties—war veterans, civilian victims of war, and civil- ians with non-war-related disabilities—and the separate legislation for the three categories leads to different levels of benefits, procedures and treat- ments for individuals who have exactly the same level of impairment. While the three share the same needs, the third group is not given priority in terms of funding and is left without adequate coverage. This affects particularly children with disabilities, who have a disability not caused by war. In addition, children with disabilities are not sufficiently included in mainstream education, due to lack of required infrastructure and insufficient training of teachers. In particular, schools lack assistants who are special- ized to support children with disabilities and their teachers. Services for the elderly are scarce even though the population is aging. The share of the popula- tion aged 65 or above rose from 6.5 percent in 1991 census to 14.2 percent in the 2013 census, and the total fertility rate fell from 1.65 in 1996 to 1.26 in 2017.50 According to a 2020 representative survey led by the United Nations Development Programme (UNDP), 33 percent of citizens think that there are no care facilities for the elderly and 11 percent report that the prices of elderly care facilities are too high. While laws at the entity and canton level prescribe services for the elderly—such as home care and assistance, and day care and institutional care—the kinds of services and levels of benefits differ sig- nificantly. Community services remain underdevel- oped and are often provided by private organiza- tions where support tends to be project-based and financial limitations can impair service delivery.51 Based on the most recent data (2015), BiH had lower spending on elderly care than any other country in Europe. BiH spent 0.1 percent of GDP on long-term care within the health system and 0.1 percent of GDP on long-term care within the social protection system.52 Back to table of contents 17 50 Agency for Statistics Bosnia and Herzegovina. 2018. Demography 2017. Thematic Bulletin 02, Sarajevo, pp. 19-22. 51 Jusić, M. 2019. Bosnia and Herzegovina’s long-term care challenge. ESPN Flash Report 2019/38.CY 52 Jusić, M. 2019. Bosnia and Herzegovina’s long-term care challenge. ESPN Flash Report 2019/38.CY BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Pensions Prior to the outbreak of COVID-19, the pension systems in the two entities were estimated to remain fiscally sustainable over a period of 10 to 15 years. Pandemic-related pension measures temporarily widened the pension deficits and shortened the expected sustainability period. Further changes to the rules governing years of service for eligibility and early retirement will be necessary to maintain sustainability beyond this period. Moreover, both pension systems are structurally challenged by population aging and outmigra- tion, which have led to low support ratios (contributors compared to pensioners). The adequacy of pension benefits is low and declining (while the pension contribution rate is already substantial) and can be organ- ically improved in the future through further reforms of the parameters and by encouraging participation in voluntary pension programs. FBiH and RS operate two separate public pension systems face similar challenges, related to their fea- systems covering old-age pension, disability pen- tures. They are characterized by low support ratios sions and survivors’ pensions. In January 2022, the (1.21 contributors per pensioner in FBiH and 1.16 in system covered 430,365 beneficiaries in FBiH and RS) (Figure 13), high pension expenditures and low 269,139 in RS, with around 60 percent receiving the pension adequacy. The average benefit ratio in old-age pension (58 percent of pensioners in FBiH, both pension systems is among the lowest in the 60 percent in RS). Around 15 percent of beneficiar- ECA region and Europe (Figure 14).55 In December ies receive a disability pension and remaining 25 2021, the average gross replacement rate stood at percent a survivor pension. The average monthly net 30 percent in FBiH and 28.1 percent in RS.56 The pension is similar across entities, BAM 490 (approx. pension contribution rate, however, was 23 percent US$ 255) in FBiH and BAM 431 (approx. US$ 225) of gross wage in FBiH and 18.5 percent of gross in RS. Total expenditures in pandemic-distressed wage in RS, which is above the rates in many other 2021 increased to 10.5 percent of GDP in FBiH and countries. In both FBiH and RS, the share of disa- 11 percent of GDP in RS. bility pensioners in 2021 was relatively high: 15.4 percent in FBiH and 13.1 percent in RS. This reflects While the two pension systems are expected to be the historical context of disability pensions for war sustainable in the medium run, parametric reforms veterans and civilian victims of the war and the rela- will be required in the longer run. Before 2011, the tively generous definitions of disability. two systems were similar. In 2011, RS launched a reform to restore its fiscal balance and to reduce Currently, both reforms are focused on cushion- pension contribution arrears. In 2018, FBiH launched ing the consequences of the COVID-19 pandemic a reform to strengthen the link between benefits and emerging inflation through pension supple- and contributions and to improve the long-run sys- ments and alternative pension indexation for- temic adequacy while maintaining fiscal sustain- mulas. Longer-term policies should, however, be ability.53 With the 2018 reform, the parameters of revisited, including specific pension system issues. the two systems were again similar. Both reforms This includes early retirement in hazardous occupa- introduced a point system54 with clear valorization tions, privileged pensions and the disability assess- and indexation rules (which differ across entities) ment process; improving administrative capacity and tightened early retirement rules as well as dis- and user-friendliness of pension administration; ability pension rules (Table 2). The parameters of and addressing and reducing the stock of unpaid both systems should stabilize COVID-19 pandemic pension contributions.57 impacts over the next few years; however, further reforms will be required. Eventually, in both entities, further parametric pension reforms to improve adequacy organically The pension system will need to tackle low and by extending service periods will have to be imple- gradually declining pension adequacy. Both pension mented. Service periods can be extended by raising Back to table of contents 18 53 In FBiH, the Pension and Disability Insurance Law (PDIL) was adopted by the Parliament in February 2018 and implemented in March 2018. 54 An insured individual earns pension points in proportion of her or his salary to the national average wage. Average wage implies one pension point. At retirement, the number of earned pension points is multiplied by a unique point value to obtain the pension benefit. 55 The average benefit ratio is the share of the average gross pension in average gross wage. 56 The replacement rate corresponds to the gross pension divided by gross (pre-retirement) earnings. 57 Unpaid pension contributions in both entities have been mounting over the past three decades, creating eligibility obstacles for insured individuals and financial problems for the fiscal authorities. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Table 2: Pension parameters in 2021 Federation of Bosnia and Republika Srpska Herzegovina 65 years male/female with 15 years of 65 years male/female with 15 years of contribution contribution Retirement age Any age with 40 years of contribution Early retirement: 60 years with 40 years Early retirement phased out by 2026 of contributions (men); 58 years with 35 (men) and by 2036 (women) years of contributions (women) Accrual rate/point value 1.6% of net wage = 1 point 1.33% of net wage = 1 point Sum of points over full career Sum of points over full career Pensionable base Minimum point value = 0.5 Maximum point value = 4 Maximum point value = 5 50% of CPI change + 50% of real GDP 50% of CPI change + 50% of nominal Indexation post-retirement growth average gross wage growth Eligibility for disability Contributed for one-third of working life Minimum 2-5 years of service and at pension from age 20 to age of disability least 35 years of age Same level as old-age pension (and not Same level as old-age pension (and not lower than the level for 15 service years) lower than the level for 15 service years) Level of disability pension In case of work injury, not lower than the In case of work injury, not lower than the level for 40 service years level for 40 service years Age 45 or 50, incapacity or caring for a Age 50 years, incapacity or caring for a child in case of widow child in case of widow Age 60, incapacity or caring for a child in Age 60, incapacity or caring for a child in Eligibility for survivor case of widower case of widower pension Children ages 15 years or under and until Children ages 15 years or under and until age 26 years if students or lifetime if age 26 years if students or lifetime if incapacitated incapacitated 70%-100% of disability pension 70%-100% of disability pension Level of survivor pension depending on number of survivors depending on number of survivors Contribution rates 23% of gross salary 18.5% of gross salary Figure 13: Support ratio in EU and Western Figure 14: Average gross pension benefit ratio, EU Balkans, 2016 or recent (Contributors per 100 and Western Balkans, 2019 or Recent pensioners) 350 70 300 60 250 50 200 40 150 30 100 20 50 10 0 0 DK RO HR FBIH (2018) BG RS (2018) SRB (2018) FR SI PT IT EE FI DE EU27 SK BE MNE (2018) AT LV HU EL CZ PL MAK (2018) ES NL MT LU SE IE CY LV ALB LT BG IE RS EE FBIH SI HR RO SRB SE MNE SK NL HU MKD CZ FR DE EU EA DK PL MT BE FI LU AT NO PT CY ES IT EL Source: European Commission (EC). 2018. The 2018 Ageing Source: EC. 2021. The 2021 Ageing Report, pension agencies in Report, PAYG pension agencies in Western Balkans countries. Western Balkans countries. Gross benefit ratio is the ratio of Support ratio is the number of contributors / number of average pensions to average gross wages. pensioners. Back to table of contents 19 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis the retirement age and eliminating early retirement. Expansion of voluntary pension savings would also help cushion the eroding adequacy of public pen- sions, but the development of voluntary pension savings schemes has been slow and subjected to legal obstacles. Both entities have been working on a legal framework and on stimulating voluntary pension savings to improve the pension adequacy in the long run. In FBiH, the establishment of voluntary pension funds has been slow and affected by polit- ical constraints. Despite the required laws enacted in 2017, no license for pension fund management in FBiH was either issued or applied for until January 2022. In RS, the voluntary pension system has been operating since 2017. There is however only one pension fund in operation, covering around 15 percent of active workers. RS stimulates volun- tary pension savings through tax stimulus on both employee and corporate income. Contributions up to BAM 100 (approx. US$ 52) per month are free from personal income tax and social contributions, and, simultaneously, the amount is added to cor- porate expenses and thus deducted from corporate profit tax. RS has prepared a legal framework to introduce automatic voluntary opt-out enrollment system in 2020, but its implementation has been delayed due to the COVID-19 pandemic and not yet implemented in 2022. Back to table of contents 20 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Employment and ALMPs Similar to other Western Balkan countries, Bosnia and Herzegovina spends little of the social protection budget on employment and labor market policies (ALMPs). While intermediation services improved their business processes, the case load of counsellors remains high. The coverage of ALMPs is limited and the targeting of vulnerable groups can be improved. The menu of ALMPs remains narrow, mostly focused on wage subsidies. Efforts recently undertaken for evidence-based decisions on the design of ALMPs should be continued. There are several structural disincentives to go back to work and no linkages between Public Employment Services (PES) and Centers for Social Work (CSW) to activate social assistance beneficiaries. Labor policies and programs are implemented sep- addition to the agencies’ counselling activities and arately by each entity. Labor policies are developed intermediation services, they also offer a menu of at the ministry level in each entity (the Ministry of ALMPs, provide jobseekers with health insurance Labor and Social Policy of Federation of Bosnia and and pay out unemployment benefits. In 2018, they Herzegovina in FBiH and the Ministry of Labor, War spent the equivalent of 0.23 percent of GDP on Veterans, and Disabled Persons’ Protection in RS) ALMPs, largely toward wage subsidies for employ- and implemented by their respective PES agen- ers. In both entities, the PES budget supports the cies. In FBiH, the Federal Employment Institute payment of unemployment benefits.59 In 2019, an (FEI) coordinates the implementation of activities average of 10,732 beneficiaries received unem- in ten cantonal employment institutes, which are to ployment benefits per month in FBiH, whereas an a large extent financially independent and perform average of 2,148 beneficiaries received unemploy- all functions of job registration, job intermedia- ment benefits per month in RS. Spending on unem- tion and payment of benefits as well as implement ployment benefits accounted for the equivalent of ALMPs.58 In RS, there is one centralized PES, the 0.2 percent of GDP in 2018, around the average Republika Srpska Employment Bureau (RSEB), with spending on passive labor market measures in the six regional branches. Western Balkans but significantly less than in the EU (Figure 15). In 2019, the PESs paid health insur- In total, the two PESs represent the second highest ance for a total of 213,416 registered jobseekers, level of expenditures on ALMPs in the Western slightly more than half of the average 411,230 regis- Balkans, but considerably less than in the EU. In tered jobseekers in BiH. Figure 15: Expenditure on labor market policies as a percent of GDP, 2018 1.4% Total LMP (incl. administration, placement and related services) 2018 Expenditures (% of GDP) 1.19% 1.2% Active LMP (including Vocational Training) 1.0% Passive LMP (unemployment benefits) 0.8% 0.65% 0.6% 0.48% 0.43% 0.42% 0.4% 0.36% 0.2% 0.10% 0.11% 0.0% EU-27 7STEE Montenegro BiH North Serbia Kosovo Albania (2019) Macedonia (2021) Sources: EU-27/7STEE: Eurostat; North Macedonia: Administrative Data; Montenegro: MONSTAT; Kosovo: Paun Jarallah et al. 2019; Albania: Administrative Data NAES. Bosnia and Herzegovina: Bosnia and Herzegovina Statistics Agency; Serbia: World Bank SPEED (2022). Note: 7STEE refers to the seven small transition economies of Europe: Bulgaria, Croatia, Estonia, Latvia, Lithuania, Slovak Republic and Slovenia. Back to table of contents 21 58 Cantonal Employment Institutes are financed through cantonal tax contributions and non-tax revenues. Part of the financing for active labor market measures and unemploy- ment benefits is transferred from the Federal Employment Institute. The degree of transfer depends on a number of factors, such as: tax revenues of Cantonal Employment Institutes and number of registered unemployed persons. 59 In FBiH, unemployment benefits are paid in the amount of 40 percent of the average salary over the last three months prior to the unemployed person’s registration. Unemploy- ment benefits are paid for a period from 3 to 24 months depending on the duration of previous employment of the newly unemployed person. In RS, unemployment benefits are paid in the amount of BAM 416 to BAM 800 (approx. US$ 217 to US$ 418) for a maximum of 24 months, which depends on the duration of previous employment of the unemployed person. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Since 2018, PESs have been implementing reforms and receiving subsidies (in the form of reimburse- to strengthen counseling and intermediation ser- ment of social contributions) for the jobseekers vices to improve employment outcomes for job- employed, or jobseekers receiving vouchers to be seekers. Offices across Bosnia and Herzegovina face incentivized to find a job (the voucher decreasing different challenges in terms of caseloads and labor their cost of employment to employers for the first market tightness. A review of PES efficiency led in months). Both entities introduced a measure to 2016 by the World Bank found that the average support jobseekers interested to become self-em- effectiveness of offices was moderate and that they ployed, but for a very limited number of individuals. could better use their resources (both staff and Both entities are considering introducing coaching expenditures) to reach higher levels of jobs place- or mentoring to such programs.63 Finally, both enti- ments.60 The recorded case load is very high, partly ties also implement public works on a needs basis, because about half of the registered unemployed but not as regular annual programs. The Federal benefit from health insurance but do not really Employment Institute introduced a new program in search for a job. The ratio of unemployed workers to 2021 to place hard-to-serve youth and those aged counsellors ranges between 200 and 1,300 across 40+ with low skills into jobs and to benefit from the different branches and entities, far above the on-the-job training. The program is incentivizing recommendations of the International Labor Organ- employers to conduct three months of technical ization (ILO).61 The same review identified areas training for unemployed persons with an additional requiring improvements, in particular, the business 6-12 months of employment. This program provides processes to better serve jobseekers and employ- modest compensation to employers for the training ers and the relationship with the private sector.62 which also includes a stipend for the unemployed A series of reforms have been implemented since person/trainee for the entire duration of the train- 2018. These include a more rigorous assessment ing. Employers who participate in this program are of jobseeker needs, the systematic development obligated to keep at least 60 percent of the total of individual action plans for jobseekers, and using number of trainees formally employed upon com- improved methodologies to understand employer pletion of the training. The Federal Employment needs and labor demand. In RS, all jobseekers now Institute pays a one-time compensation to employ- have an action plan. The Federal Employment Insti- ers for every person they employ after the comple- tute is also rolling out a mandatory action plan for tion of the training. its registered jobseekers. The outreach to employ- ers has increased, leading to a 15 percent increase The performance of ALMPs has been improv- in the collection of vacancies for intermediation ser- ing over the last few years as they become more vices since 2017. The number of successful matches responsive to employer and jobseeker needs but thanks to the intermediation services of the PES remains insufficiently targeted. Improvements have increased by 38 percent between 2017 and 2020, included changes in the delivery and targeting of with 63 percent of the vacancy requests shared by ALMPs and changes in the services provided to employers being matched with a registered job- improve cost-effectiveness. The PES in both enti- seeker and leading to a job placement in 2020. ties have been conducting evaluations and adjust- Between December 2020 and July 2021, despite ing the design of ALMPs over the last four years, an weak labor market conditions, close to 70,000 effort that should be continued by implementing jobseekers have been de-registered from the PES rigorous impact evaluations for evidence-based (unemployment registry) because they found a decisions. In line with the findings of evaluations and job. Both PES job portals have been upgraded to the changing labor market conditions, ALMP menus become job platforms where vacancies and job- have become better targeted toward hard-to-serve seekers can be matched and to raise the level of groups (which constitute 80 percent of all regis- available data on labor markets. The adoption rate tered unemployed), including less skilled workers, of automated matching remains low in FBiH, and women, self-employed and youth, and away from more outreach to cantonal PES staff is planned. higher-skilled jobseekers. In both entities, ALMPs for jobseekers interested in self-employment have The menu of ALMPs is narrow, mostly focused on been created. The support for start-ups in RS was wage subsidies to employers to support job place- modified following an evaluation to include manda- ment. The menu of ALMPs remains essentially tory training and mentorship. In 2019, the Federal focused on job placements with formal employers Employment Institute introduced ALMPs specif- expressing needs during specific calls for proposals ically targeted at women (Employment of Women Back to table of contents 22 60 World Bank. 2018a. Bosnia and Herzegovina Functional Review: Efficiency Analysis of the Public Employment Services 2016, Summary Note. Washington DC. Note that the effectiveness analysis is restricted to FBiH (data from 2016) due to data limitation in RS. The review estimated that higher effectiveness could already increase total job place- ment by 25 percent while keeping staff costs and expenditures constant. However, these estimates need to be taken with caution since the exercise conducted for Bosnia and Herzegovina is more illustrative than reliable given the limited and low-quality indicators provided by the employment offices. 61 There are six branches in RS and ten cantonal PES in FBiH. The ratio of unemployed to the number of staff ranged from just over 200 (Gorazde) to over 1,300 (Doboj) in 2016. In comparison, the staff/client ratio in the European Union is 1:150 and ILO’s recommendation is 1:100. If it is taken into account that half of the registered unemployed do not search for a job, the ratio still is 1:650, which continues to be high. 62 In the past, firms would contact the PES to take advantage of wage subsidies rather than requiring intermediation services. 63 In RS, all participants take part in a general entrepreneurship training, and the introduction of coaching is being considered. In FBiH, while coaching was considered before the pandemic, it could not be implemented. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis and Entrepreneurship for Women) and others spe- free health insurance for unemployed workers from cifically targeted at hard-to-serve groups (such as, unemployment registration increases the number ALMP Employment of Roma that has been active of administratively registered unemployed who are since 2017, “Second Chance”, “I Am Looking for not actively searching for a job (some working infor- an Employer” (a voucher program) and “Training mally). and Work”). For selected ALMPs with the largest number of beneficiaries, the employment rate 12 months after the program completion is close to 60 percent, depending on the programs and entity, which attests to the relative success of the pro- grams in supporting employment.64 However, there are no ALMPs specifically adapted to and targeting women in RS. In general, beyond the broad eligi- bility criteria, no further targeting is implemented. There is no initial profiling of the jobseekers that would lead to placements into specific ALMPs. The menu of ALMPs has improved but there is room for further innovation, expansion of coverage, and targeting of resources to those most in need for support. The coverage of programs remains low at about 5 percent of the total registered unemployed, while there is oversubscription for such programs.65 The targeting of ALMPs could be further improved by ensuring the wide and consistent use of the profiling system introduced as part of reforms in 2019 to assess clients’ likelihood of finding employ- ment. There is a lack of coordination and referrals as well as an absence of data sharing between social assistance systems (Centers for Social Work) and employment services (PES). No specific effort is made to identify those with the highest potential to be activated, in large part due to weak or non-ex- istent linkages between CSWs, who deal with social assistance beneficiaries, and PES, who implement labor programs where some could be referred. The combination of a relatively high tax wedge and no progressivity generates incentives for informal- ity and disincentives for formal job creation. The relatively high tax wedge results from high pension and health contributions (as high as 42 percent of the total labor cost in FBiH).66, 67, 68 This, com- bined with a tax system that penalizes low-income workers (with relatively low nontaxable minimum threshold and a flat personal income tax), helps to keep unemployment and inactivity high. Introduc- ing the phased withdrawal of social assistance ben- efits combined with social services (e.g., in form of income disregards, in-work benefits and vouchers to facilitate access to care services) would encour- age employment among poorer groups and help to boost the prospects of the poorest 40 percent of households. Further, in FBiH, the lack of de-linking of Back to table of contents 23 64 PES Monitoring Reports dated 2019 and 2020. 65 Given the linkages between health care and unemployment registration in FBiH, it can be assumed that a large share of registered unemployed are not actively searching for a job and not interested in ALMP participation. However, coverage remains low considering the estimated number of active jobseekers. 66 World Bank. 2020a. Bosnia and Herzegovina Systematic Country Diagnostic Update. World Bank, Washington, DC. 67 In RS, pension contributions represent 18.5 percent of gross salary. Mandatory social security contributions are only employee contributions (no employer part) and are with- held by the employer. Starting in 2022, following amendments of the Labor Act, RS reduces the tax wedge, reducing social security contributions from 32.8 percent to 30.8 percent (health insurance contributions are reduced from 12 percent to 10 percent) and the salary tax rate from 10 percent to 8 percent. 68 In FBiH, the mandatory social contributions of employees represent 31 percent (including 17 percent for pensions) and employers have mandatory contributions of 10.5 percent, both calculated based on gross salary. In addition, the employer pays 0.5 percent of the contribution for protection from natural and other disasters, as well as 0.5 percent of the water protection charge, calculated based on net salary. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Learning from the COVID-19 pandemic response The challenges of the social assistance system were magnified during the COVID-19 crisis. Last-resort income support was able to slightly expand vertically but less so horizontally. Both entities ensured continuous payment of benefits to all permanent social assistance and home care support beneficiaries, including top-ups in FBiH. However, legislative and budget restrictions limited the ability to respond to the crisis in both entities. With the exception of the use of one-off assistance in RS, no new measures were introduced for the newly poor as a result of increased unemployment during the pandemic. The lack of a comprehensive and updated social registry in each entity also limited the ability to reach newly poor and households at risk. Public Employment Services were quick in responding to crises and provided timely registration and payment of unemployment benefits despite budget constraints. However, strict unem- ployment benefit eligibility rules combined with the legislation on severance pay left many workers who were previously employed for a short period of time without appropriate financial support. Bosnia and Herzegovina took measures to contain the primary legislation; it is categorical and cannot the spread of COVID-19, which have resulted in cover individuals who are able to work but who an economic slowdown and threaten economic have lost their job.70 While the legislation in FBiH prospects. In response to the COVID-19 outbreak, sets the basis for a flexible system, the number of both entities declared a state of emergency, which beneficiaries did not increase. The lack of a compre- included travel restrictions, lockdowns and stay-at- hensive and updated social registry in each entity home orders. These restrictions had a serious impact also limited the ability to reach the newly poor and on economic activity. While the emergency status households at risk. No specific outreach to house- was lifted in late May 2020 in both entities, FBiH holds that might have become eligible for support declared an epidemic again in mid-July 2020 after has been undertaken. No new measures were intro- a renewed spike in infections but did not impose duced for the newly poor, with one exception in RS lockdown or travel restrictions. Welfare also was where around 5,800 additional beneficiaries were affected by a reduction of remittances from abroad, reached through one-off financial assistance deliv- which are a major source of income for the poor. A ered by municipalities. Otherwise, top-ups have household survey conducted from November 18 been paid to increase the level of support to former to December 11, 2020, indicates that 15 percent of beneficiaries of permanent social assistance in both households experienced a decline in remittances.69 entities, as well as for the disability benefit in RS and homecare cash assistance in FBiH. More generally, Cash transfers assistance provided to the poor during the pan- demic varied from one municipality to another, so During the COVID-19 crisis, the benefit levels of that many beneficiaries may have been left without permanent social assistance and one-off financial any additional support during the pandemic.71 assistance (existing in both entities) were slightly increased (vertical expansion) but the interventions Pensions were not able to reach additional households at risk of poverty or the newly poor (horizontal expan- The economic consequences of the COVID-19 pan- sion). Both entities ensured continuous payment demic on the pension system are expected to be of benefits to all permanent social assistance and temporary. In 2020, the fall in the number of con- home care support beneficiaries. However, legis- tributors impacted the pension deficit. In 2021, lative and budget restrictions limited the ability to the contribution base strongly recovered. In 2021, respond to the crisis in both entities, in particular, pension expenditures in both entities stood about in terms of reaching additional households. In RS, 1 percentage point above the pre-pandemic level eligibility for permanent social assistance is set in (equivalent to 10.5 percent of GDP in FBiH and 11 Back to table of contents 24 69 Prism Research and Consulting. 2021. “Social Impacts of COVID-19 in Bosnia and Herzegovina: Second Household Survey.” 70 Eligible households have no members able to work. 71 Matkovic, G., and Stubbs, P. 2020. Social Protection in the Western Balkans: Responding to the COVID-19 Crisis. Friedrich-Ebert-Stiftung (FES) Dialogue Southeast Europe. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis percent of GDP in RS) and pension revenues above tributions, thus reducing PES revenues, and created that, resulting in slightly smaller pension deficit in an urgent need to increase funding for unemploy- comparison to the pre-pandemic period. ment benefits, which was not initially budgeted.75 As the first priority of PES was to pay unemployment Measures implemented during the COVID-19 crisis benefits, the funds available for ALMPs declined aimed at supporting formal employment and the sharply. Any new calls for ALMPs were stopped until level of pension revenues by subsidizing contribu- late 2020, and their scope was limited. In 2020, tions. In FBiH, business entities severely hit by the FBiH published only two calls for ALMPs titled “I Am crisis (defined by a 20 percent or larger fall in turno- Looking for Employer,” and RS published no calls. ver, compared to the same month in 2019) received The number of recipients of unemployment benefits subsidies to cover mandatory contributions (pension increased starkly both in RS and FBiH in response to and disability insurance, health insurance and unem- job losses. In RS, the number of recipients peaked in ployment insurance) up to BAM 244.85 (or approx. February 2021 at 4,567 recipients and has declined USD 128) per month per employee.72 Beyond that, since. In FBiH, the number of recipients spiked more in 2021, FBiH paid out a COVID-19 supplement and starkly and peaked in May 2020 at 19,000. While proposed a modification to the pension indexation the number of jobseekers still receiving benefits in mechanism. However, no adjustment was done on RS is higher than on average in 2019, the recipient the level of pensions in 2021.73 In RS, the full amount count in FBiH has since approximately returned to of contributions was covered for all contribution its pre-crisis average (Figure 16). payers who were prevented from doing business.74 In 2020, pension levels were regularly adjusted (at Eligibility rules for unemployment benefits and the rate of 3.11 percent) and in 2021, at 5.08 percent legislation on severance pay left many workers (out of which, there were two extraordinary indexa- without appropriate financial support. While tions: 0.86 percent in March 2021 and 2 percent in unemployment benefits were offered to the reg- September 2021). istered unemployed who have been continuously employed in formal jobs for a certain period of time, Employment and ALMPs no measures were taken to support those who were employed for less than eight months before losing The sharp increase in unemployment following the their job due to the pandemic. In addition, the leg- lockdowns increased the budget need for unem- islation requires severance payments in case of dis- ployment benefits in both PESs, while reducing missals only for those cumulating two years of con- their revenues and the provision of ALMPs. The tinuous work in their firm and for indefinite-term three months of lockdown throughout BiH increased contracts, therefore excluding fixed-term contracts unemployment and reduced business activity. The and recent workers. rise in unemployment reduced social security con- Figure 16: Unemployment benefits expanded as a response to the crisis Unemployment benefit recipients 0 11.19 11.21 12.19 12.21 10.19 11.20 01.21 10.21 01.19 07.21 07.19 12.20 02.21 03.21 05.21 02.19 03.19 05.19 09.19 04.21 06.21 08.21 09.21 04.19 06.19 08.19 01.20 10.20 07.20 02.20 03.20 05.20 04.20 06.20 08.20 09.20 FBIH RS Source: Ministry of Labor and Public Employment Agency in both entities (2022). Back to table of contents 25 72 “Law on mitigation of negative economic consequences,” also known as the Corona Law. Out of the total subsidy, BAM 135,70 (approx. US$ 70) is directed for pension and disability insurance. 73 An adjustment of 11 percent has been announced for April 2022, which includes an extraordinary adjustment of 3.5 percent. 74 The official list of RS’ COVID-19 pandemic countermeasures at Владине мjере I Коронавирус у Српскоj (koronavirususrpskoj.com). 75 According to data from FBiH Tax Administration, the collection of unemployment insurance contributions in all public employment services in FBiH compared to 2019 decreased by 0.79 percent. RSEB revenues decreased by 25 percent, due to an increase in the number of unemployed eligible for unemployment benefits, jointly with a reduction in the contribution rates (from 0.8 percent to 0.6 percent) since January 2020. BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Reform priorities and knowledge gaps Bosnia and Herzegovina can prioritize policy along need for improved coordination and data exchange a continuum of short-, medium- and long-term across the entities, cantons and municipalities that choices to develop a more robust social protec- administer social protection programs, as well as tion system. Table 3 summarizes recommendations with private sector providers of services, including of this analysis, divided into short-term, medi- up-to-date social registries. The management of um-term and long-term goals. A key goal must be ALMPs could be strengthened through a greater to increase the effectiveness of resources spent, by focus on monitoring and evaluation, along with expanding the resources devoted to social assis- improvements in profiling, the use of technology in tance, the elderly and childcare, and ALMPs. The supporting job applicants, and various business pro- coverage of the poorest households, in particular, cesses. Finally, disaster preparedness and response should be increased in a cost-effective way. This could be strengthened by putting in place contin- entails improving targeting by shifting social assis- gency financing to fund both social assistance cash tance funds from war-related to poverty-targeted transfers and unemployment benefits and intro- programs, and shifting ALMPs from support for ducing the required legislation and implementation jobseekers (that are “not hard-to-serve”) to pro- arrangements for emergency support programs for grams focused on “at-risk” unemployed and acti- households at risk and newly unemployed. vation. In FBiH, incentives for formal employment could be increased by de-linking health insurance In parallel, there are several areas presented in Table from unemployment benefits and allowing social 4 where further research would be useful to better assistance beneficiaries to keep a larger amount assess the situation and provide further guidance. of benefits as wages rise from low levels. There is a Table 3: Reform priorities Short-term Medium-term Long-term Cash transfers and in-kind benefits to support people with disabilities, poor and vulnerable groups • Make the unique social registry in • Link permanent social assistance • Rationalize spending on war-re- FBiH and RS active, up-to-date and with activation; consider phased lated benefits and allowances comprehensive. withdrawal of social assistance and reallocate spending toward • Analyze options for improved combined with social services the poverty-targeted, mean- poverty-targeting of social assis- (e.g., in form of income disregards, tested benefits such as child tance programs (e.g., last resort, in-work benefits and vouchers for allowance and permanent social child allowances, war veterans) to care services). assistance. capture informal income and better • Expand the coverage and adequacy target the poor. of means-tested benefits, while consolidating benefits to increase efficiency. • Expand child allowance to all cantons and entities in a means-and/or poverty-tested format. • Invest in efficient government delivery systems: human resources for social workers; management information systems; case manage- ment in CSWs. • Centralize data collection across the entities (as well as Brcko District), cantons and municipalities. Back to table of contents 26 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Short-term Medium-term Long-term Social services • Improve access to childcare and • Reorient CSWs to work in closer elderly care by expanding public collaboration with public employ- supply and subsidies for private care ment services through integrated (e.g., with vouchers). IT systems and services. Pensions • Review pension indexation • Revise early retirement options for • Improve adequacy organically frequency and its effectiveness to hazardous occupations. by extending service periods (by prevent real pension erosion • Revise privileged pensions and the raising retirement age and elimi- disability assessment process. nating early retirement). • Improve administrative capacity and user-friendliness of pension admin- istration. • Analyze and engage in reducing the stock of unpaid pension contribu- tions. • Stimulate expansion of voluntary pension savings. Employment and ALMPs • De-link provision of health insurance • Institute a performance monitoring • Redirect ALMP resources from from unemployment registration (in mechanism in local PES to check wage subsidies that are not spent FBiH). on the quality of counseling and on hard-to-serve jobseekers to • Continue to strengthen monitoring application of the “Rulebooks” and programs that promote “acti- and evaluation of ALMPs, including regulations by counselors. vation” and target vulnerable regular impact evaluations. • Digitalize services for jobseekers, jobseekers; implement more • Strengthen collection and analysis introduce digital learning for soft ALMPs tailored to the needs of of data on occupations and skills on skills development. vulnerable groups of jobseekers. demand and growth sectors. • Change call-based procedure for • Prepare strategies for partnering • Further improve PES’s online job ALMPs to programs with contin- with the private sector to boost portals to increase job matching uous enrollment. skills development and increase (and labor market information). • Improve coordination between efforts to promote work-based federal and cantonal PES (in FBiH) learning and career guidance for to improve the monitoring of students in schools and univer- vacancies and harmonize the use of sities. automated tools for matching. • Increase budget for ALMPs • Refer jobseekers to ALMPs based (rather than a residual budget) to on profiling. reduce caseload per counselor. • Invest in collaboration or contracting of private employ- ment agencies to support activa- tion of hard to employ. Disaster preparedness and response • Put in place contingency financing • Introduce flexibility to existing Laws • Carry out a systematic assess- for cash transfers in case of disas- to be able to introduce emergency ment of social protection ters such as pandemics and natural support programs for newly unem- programs for crisis response, events (floods, earthquakes). ployed and households at risk of focusing on the operational falling into poverty. aspects of outreach, targeting, enrollment and payment, among others. • Identify options for contingency financing to support the scale-up of such programs to respond to shocks. Back to table of contents 27 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Table 4: Knowledge gaps Cash transfers and in-kind benefits to support people with disabilities, the poor and vulnerable groups • Review outreach, intake, eligibility criteria and enrollment and identify gaps and inequalities to ensure hard-to-reach populations receive support. • Review of social assistance exit strategy and its results, including the provision of employment services and social support. • Consolidate data on state, entity and municipal benefits for a comprehensive and up-to-date social registry of available support. • Review the current performance of the social assistance programs with comprehensive data on measures (including across ministries) and recent Household Budget Survey data. Social services • Assess the adequacy of services to elderly (e.g., minimum pension beneficiaries with no family members). • Assess the availability and quality of childcare services to support women labor force participation. Pensions • Analyze tax rates and pension parameters to ensure sustainability of pension system over the long term while improving incentives for formal work. Employment and ALMPs • Understand how productivity can be increased in the PESs. • Understand how benefits could be extended to workers in the informal sector. • Understand the performances of the Technical and Vocational Education Training (TVET) system. • Complement the menu of existing ALMPs where there is demand. Disaster preparedness and response • Assess the flexibility of the legislation to allow the use of the social assistance system in case an emergency transfer is required. Back to table of contents 28 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis Resources Agency for Statistics Bosnia and Herzegovina. Prism Research and Consulting. 2021. Social 2018. Demography 2017. Thematic Bulletin Impacts of COVID-19 in Bosnia and Herze- 02, Sarajevo, pp. 19-22. govina: Second Household Survey. Sarajevo: United Nations Development Programme Braithwaite, J., and Mont, D. 2009. Disability and (UNDP) and United Nations Children’s Fund Poverty: A Survey of World Bank Poverty (UNICEF). Assessments and Implications, Alter 3:3, pp. 219- 232. Scharle, A. 2018. Comparative Study on Western Balkan’s PES Performance. Budapest: Cuberes, D., and Teignier, M. 2016. Aggregate Budapest Institute and the Centre for Develop- Effects of Gender Gaps in the Labor Market: ment Evaluation and Social Science Research. A Quantitative Estimate, Journal of Human Capital 10:1. Sucur-Janjetovic, V., Kurta. A., and Oruc, N. 2017. Social Protection System in Bosnia and Herze- European Commission. 2018. The 2018 Ageing govina: Country Brief. A Western Balkan Report: Economic & Budgetary Projections Regional Initiative, December 2017. Belgrade: for the 28 EU Member States (2016 - 2070). Center for Social Policy. Institutional Paper 079. Brussels: European Commission United Nations Development Programme (UNDP). 2020. Social Inclusion in Bosnia and Herze- European Commission. 2021. The 2021 Ageing govina: 2020 National Human Development Report: Economic and Budgetary Projec- Report – Extended Summary. Sarajevo. tions for the EU Member States (2019-2070). Institutional Paper 148. Brussels: European United Nations Children’s Fund (UNICEF). 2012. Commission Bosnia and Herzegovina: Roma Survey. Multiple Indicator Cluster Survey 2009-2012. European Commission. 2015. Assessment Report Sarajevo. on PES Capacity 2015. United Nations Children’s Fund (UNICEF). 2017. Jusić, M. 2019. Bosnia and Herzegovina’s Long-Term Situation Analysis of Children with Disabilities Care Challenge. ESPN Flash Report 2019/38. in Bosnia and Herzegovina. November 2017. Brussels: European Commission. Sarajevo. Matkovic, G., and Stubbs, P. 2020. Social Protec- United Nations Children’s Fund (UNICEF). 2020. tion in the Western Balkans: Responding to Situation Analysis of Children with Disabili- the COVID-19 Crisis. Sarajevo: Friedrich-Ebert- ties in Bosnia and Herzegovina, March 2020. Stiftung (FES) Dialogue Southeast Europe. Sarajevo. Mete, C. 2008. Economic Implications of Chronic Vienna Institute for International Economic Studies Illness and Disability in Eastern Europe and the (wiiw)and World Bank. 2021. Western Balkans Former Soviet Union. Washington DC: World Labor Market Trends 2020. Vienna. Bank. World Bank. 2016. Formula or No Formula? That Is the Question: Targeting Social Assistance Beneficiaries in the Federation of Bosnia and Herzegovina. Washington DC. Back to table of contents 29 BOSNIA AND HERZEGOVINA Social Protection Situational Analysis World Bank. 2017a. Delinking Provision of Health Insurance to Unemployed from Unemploy- ment Registration in the Federation of Bosnia and Herzegovina. Washington DC. World Bank. 2017b. Promoting Women’s Access to Economic Opportunities: Bosnia and Herze- govina. World Bank. Washington DC. World Bank. 2018a. Bosnia and Herzegovina Func- tional Review: Efficiency Analysis of the Public Employment Services 2016, Summary Note. Washington DC. World Bank. 2018b. Technical Note on Coverage, Targeting and Poverty Impact of Social Assis- tance in Bosnia and Herzegovina. Washington DC. World Bank. 2019. Promoting Growth through Better Private Sector Jobs in Bosnia and Herzegovina (Growth and Rapid Jobs Assess- ment). Washington DC. World Bank. 2020a. Bosnia and Herzegovina Systematic Country Diagnostic Update. Wash- ington DC. World Bank. 2020b. Western Balkans Regular Economic Report No. 17. The Economic and Social Impact of COVID-19. Spring 2020. Washington DC. World Bank. 2020c. Western Balkans Regular Economic Report No. 18: An Uncertain Recovery. Fall 2020. Washington DC. World Bank. 2021a. Western Balkans Regular Economic Report No. 19: Subdued Recovery. Spring 2021. Washington DC. World Bank. 2021b. Western Balkans Regular Economic Report No. 20: Greening the Recovery. Fall 2021. Washington DC. World Bank. 2022. Western Balkans Regular Economic Report No. 21: Steering Through Crises. Spring 2022. Washington DC. Back to table of contents 30 © 2022 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This project is funded by the European Union