Kristine Langenbucher
OECD
Marius Lüske
OECD
Judd Ormsby
OECD
Todor Todorov
(University of National and World Economy)
Reaching Out and Activating Inactive and Unemployed Persons in Bulgaria
Connecting People with Jobs
3. Characterisation of the out-of-work population and their barriers to employment
Abstract
This chapter describes the out-of-work population in Bulgaria and the barriers to labour market integration they face with the aim of informing activation policy. In 2019, prior to the impact of COVID‑19, there were about 1.3 million unemployed and inactive people of working age in Bulgaria, around 900 000 of whom were not studying and might benefit from support with finding employment. Despite overall strong employment levels for the working age population, rates of youth not in employment education or training (NEET) are some of the highest in the EU, while Roma people face extremely high levels of joblessness and barriers to employment. In addition, Bulgaria has some of the lowest shares of employment for people with a disability and one of the largest shares of the inactive out-of-work for reasons of family and care responsibilities. Finally older workers aged 55‑64 years also make up such a large share of the out-of-work population that they must not be ignored. Appropriate activation responses will need to be tailored to the different needs these groups face.
3.1. Introduction
Integrating people who are out-of-work into employment contributes to individual well-being through raising the living standards of activated individuals and their families, saves the government money from reduced benefit spending and increased tax revenues, and provides a wider pool of workers for employers. The benefits to the government and employers are especially important in the context of a rapidly shrinking working-age population in Bulgaria. Integrating workers displaced by the COVID‑19 crisis presents a further challenge.
However, to effectively support people into sustainable employment, governments first need to understand who the out-of-work are as well as the barriers to work that they face. This chapter provides a profile of the out-of-work population in Bulgaria, including a detailed analysis of their barriers to employment. This work helps inform who the National Employment Agency (NEA) – Bulgaria’s public employment service (PES) – should reach out to and where active labour market policies (ALMPs) can best support people’s needs. These topics are then taken up in depth in Chapter 5 on outreach to the out-of-work population and in Chapter 6 on job brokerage and activation strategies.
The next section, Section 3.2, provides an overview of the inactive and unemployed population and identifies key groups whose labour market barriers need to be better understood. Subsequently Section 3.3, examines the barriers these groups face. Both Section 3.2 and 3.3 use survey data to provide a picture of the entire out-of-work population. A final section, Section 3.4, uses NEA administrative micro-data to better understand the characteristics of the NEA’s clients and gain insights on common employment barriers they are facing. While the discussion in Section 3.4 cannot be extrapolated to all unemployed and inactive people in Bulgaria, it describes the situation of the jobseekers whose labour market outcomes can be most directly improved through NEA support.
3.2. The inactive population in Bulgaria
The out-of-work population is not one homogenous group, but rather is made up of people from many varied backgrounds who may face quite different challenges to employment. This section, building on the overview from Chapter 2, provides an in-depth understanding of the different groups of people who are out-of-work in Bulgaria. The inactive and unemployed populations are analysed by standard demographic characteristics including, age, gender, ethnicity, and region within Bulgaria, as well as their self-reported primary barrier to employment. From these largely demographic characteristics five important groups are identified, bearing in mind their size, their labour market outcomes, and their potential to face different barriers to employment. Many characteristics – perhaps most notably people’s education – are not analysed in the current section or used to form these groups. Rather some characteristics, including low education and skills, are instead seen as barriers to labour market participation and analysed in Section 3.3.
3.2.1. Methodology and related literature
This chapter complements and builds on previous work in the literature, providing a more up-to-date analysis and with more extensive data than used in prior work. The three past studies that relate most closely to this work are Sundaram et al. (2014[1]), Dimitrov and Duell (2014[2]), and the Institute for Market Economics (2019[3]). Of these, Sundaram et al. (2014[1]) is perhaps the closest to the work in this chapter with respect to analysis based on survey data. Their work considers many variables from the EU-SILC as used in this chapter, but the study’s data are now a decade or more old (2008‑11) and the study lacks SILC data on ethnicity. Dimitrov and Duell (2014[2]) provide a high-level picture of the out-of-work population but go into less depth on barriers facing the out-of-work population, as they focus more on the institutional set up supporting vulnerable groups. Finally, the Institute for Market Economics (2019[3]) examines youth not in employment education or training (NEETs) aged 15‑34 in 2017 using a combination of SILC, LFS, and NEA admin data. Their analysis covers much detail on the situation facing NEETs, but by design does not address other groups. What this chapter offers, in addition to the previous three studies, is a detailed micro-data analysis of the NEA’s registered unemployed for the entire working age population. Nevertheless, despite these differences, this chapter too identifies similar groups to the above three studies as top priorities to connect to the labour market.
In order to gain a detailed understanding of the barriers that out-of-work people face, it is necessary to have information covering a variety of different domains at the household and individual level in addition to labour force status. The Survey of Income and Living Condition’s (SILC) data achieves this goal and is used in this chapter. It provides rich information on multiple domains including health, education, previous work‑experience, household characteristics, and income, among other variables. The SILC survey also includes information on self-declared ethnicity, which is not available in the Labour Force Survey (LFS) – though see Box 3.1 for some important caveats related to the ethnicity data. While SILC data is better suited to understanding barriers to employment than the LFS, a drawback is that the SILC’s measure of labour force status does not conform as easily to the International Labour Organization’s (ILO) definition that the LFS uses. Box 3.1 further compares the LFS and SILC data and their measurement of labour force status, in addition to the discussion of SILC’s measure of ethnicity noted above
Due to the different strengths and limitations of the LFS and SILC, this section (Section 3.2) primarily uses the LFS to provide a high level profile of the out-of-work population. As the LFS is the preferred survey for classifying labour market status it is best placed to understand the number of unemployed and inactive people in Bulgaria and their distribution over demographic groups (excepting ethnicity, which requires SILC data). Following the high level profile of the out-of-work in Section 3.2, Section 3.3 primarily uses SILC data to study in detail the barriers to employment the out-of-work population face, with the SILC survey chosen as it offers richer data on this topic. The SILC and the LFS provide information that covers all out-of-work people in Bulgaria including those who are not registered as with the NEA. To zoom in on the NEA’s clients, Section 3.4 performs analysis using detailed NEA micro-data.
Box 3.1. Strengths and limitations of the LFS and SILC surveys for analysing people out-of-work
The Survey of Income and Living Conditions (SILC) includes detailed information on health, ethnicity, education, previous work-experience, household characteristics, and household income, among other variables. This allows for a detailed analysis of the barriers to work faced by the inactive and unemployed population making it the survey of choice for the type of work in this chapter.
However, the common classification of activity status in SILC does not follow the International Labour Organization (ILO) definition used in the Labour Force Survey (LFS) (which uses a set of questions on hours worked, job search and availability rather than the single question answer used for SILC). Table 3.1 shows that the standard SILC self-defined definition reasonably captures employment (Column 1), but compared to the LFS (Column 3) SILC overestimates the unemployed relative to the inactive. It is possible to partially replicate the ILO definition of labour force status in SILC through using additional variables. This bespoke estimate is used in this chapter for SILC estimates of labour force status, with headline figures shown in Column 2. Nevertheless, the LFS is used throughout the chapter to show the overall numbers of people by activity status while SILC data is used to provide a detailed analysis of the barriers the out-of-work population face.
Other more minor differences between SILC and the LFS include that the SILC combines the first two quarters of the year whereas LFS is collected in each quarter. Finally, survey respondents must be at least 16 years old in SILC rather than at least 15 years in the LFS.
Table 3.1. Comparisons between SILC and LFS labour force status
Note: Working age 15‑64 population for LFS and 16‑64 for SILC.
Source: OECD calculations based on EU-SILC 2019 and the European Union Labour Force Survey (EULFS) https://ec.europa.eu/eurostat/databrowser/bookmark/07078448-2673-4666-8cc3-3f584f66967a?lang=en.
Limitations with ethnicity data
Ethnicity is an important variable for understanding the out-of-work population as employment rates differ by ethnic affiliation in many countries including in Bulgaria. More than 98% of SILC respondents in Bulgaria self-identify with one of three ethnic groups: Bulgarian, Turkish, and Roma. However, care should be taken when interpreting analysis on ethnicity. Capturing accurate ethnic information has proved challenging in Bulgaria. The SILC survey weights are based (in part) on the 2011 census. However, the 2011 census had, on average, a 9% non-response rate to the ethnicity question. Non-response was even higher in regions with high Roma populations (Haralampiev and Blagoev, 2014[4]). The 2011 Census reported around 5% of the Bulgarian population identifying as Roma. In contrast, the 2012 estimates by the Council of Europe (2012[5]) put the figure closer to 10% or about 750 000 people. Another issue could be that ethnicity information is captured as single category rather than multiple response. In a Bulgarian context it is important to recognise that some people prefer to self-identify as Turkish but are seen by others as Muslim Roma (Council of Europe, 2016[6]). Nonetheless, while the absolute number of people who are Roma, Turkish, or Bulgarian is challenging to estimate, differences among groups (such as differences in barriers faced) are qualitatively indicative even if uncertainty remains over exact quantitative figures.
Source: Council for Europe (2012[5]), Estimates and official numbers of Roma in Europe – Document prepared by the Support Team of the Special Representative of the Secretary General of the Council of Europe for Roma Issues, https://rm.coe.int/1680088ea9); Council of Europe (2016[6]), Ad Hoc Committee of Experts on Roma and Traveller Issues (CAHROM): Thematic Report of the Group of Experts on Roma Health Mediators.; Haralampiev, K. and D. Blagoev (2014[4]), Ethnicity non-identification in the 2011 census in Bulgaria.
3.2.2. Profile of out-of-work people by demographics and reason for not working
To understand who the unemployed and inactive people are, this section provides a high level profile of the out-of-work population in Bulgaria. It discusses the distribution of the inactive and unemployed across different demographic groups as well as inactive people’s stated reasons for not seeking employment. The difference between unemployment and inactivity can be especially relevant for the PES as unemployed persons are by definition actively seeking work and hence may be more likely to register themselves with the PES whereas, inactive persons are by definition not seeking work and hence may require more active outreach from the PES.
The inactive population has shrunk but remains sizeable
Estimates from the LFS put the working age out-of-work population at 1.3 million in 2019 (see Chapter 2). 140 000 of these people are unemployed and a further 1.2 million inactive. Of these 1.2 million inactive people, about one‑third (426 000 people) are estimated to be out-of-work because they are studying. Students are not usually a target of ALMPs. This leaves about 770 000 inactive people who are not studying and might benefit from ALMPs in addition to the 140 000 unemployed people. Since 2009, the number of inactive non-students has substantially declined by nearly a million people, reflecting both a shrinking working-age population and improved labour market outcomes (see Annex Figure 3.A.1). The decline in working-age retirees has been particularly strong.
Care responsibilities are the most cited reason for inactivity
Among the inactive (non-student) population the most frequent reason for inactivity is care, family or personal responsibilities, with 289 000 people in Bulgaria stating this as their reason for not seeking work. This is 37% of the inactive (non-student) population, one of the largest shares in the EU (Figure 3.1). The next most common reason given for inactivity is illness or disability with 192 000 people in this group. This represents a quarter of the inactive population (excluding students), which is slightly below the EU average of 29%. A further 146 000 or 19% of the inactive cite retirement as their main reason for inactivity. Annex Figure 3.A.1 shows how the number of people in these groups have changed over time, with working age retirees having fallen dramatically since 2011, potentially related to the strong overall labour market performance over this period and rising retirement ages.
Figure 3.1. Care responsibilities in Bulgaria are a frequent barrier to labour force participation
Note: The European Union (EU) is a weighted average of the 27 member countries shown. Excludes those not seeking work due to study.
Source: OECD calculations based on the European Union Labour Force Survey (EULFS).
Youth NEETs and older working-age people are both important groups to activate
Figure 3.2 examines inactivity by age and gender within Bulgaria and internationally. When looking at youth, it is important to recognise that many who are not working are students. Hence in Panels A and B non-working inactive students are separated out into a unique category so that the unemployed and inactive (non-student) categories sum to the NEET population. NEET rates in Bulgaria are among the highest in Europe, highlighting the urgent need to connect more youth to the labour market or, where suitable, to education and training. Male NEET rates are lower than female NEET rates as elsewhere in Europe, and female NEETs are somewhat more likely to be inactive than unemployed with 89% of female NEETs inactive compare to 72% for male NEETs.
Those aged 55‑64 are about twice as likely to be out-of-work as prime‑age workers, driven by much higher rates of inactivity rather than unemployment. Excluding students, older workers are twice as likely to be out-of-work than youth. Part of the reason for older working age people dropping out of the labour market, is likely, especially for women, related to a younger retirement age of 61 years and six months for women in 2020 compared to 64 years and three months for men. However, ongoing reforms are set to continue increasing these retirement ages until they reach 65 for both men and women (National Social Security Institute, 2021[7]). Nevertheless, despite some potential to further improve, employment rates for older working age people compare favourably to other countries, with those aged 55‑65 employed at slightly higher rates than the EU average for this group.
Figure 3.2. Inactivity is more common for older workers and women but NEET rates are high internationally
Note: Panel A refers to Bulgaria. Data refer to 2019. Panels A and B show inactivity, unemployment and non-working students as a percentage of people in the group in the European Union (EU). The groups are defined so as to be mutually exclusive, so that unemployed and inactive non-students sum to the total NEET population. Students are included in the figures for older age groups but make up less than 2% of the population over 30. The EU is a weighted average of the 27 member countries shown.
Source: OECD calculations based on the European Union Labour Force Survey (EULFS).
Ethnic minorities have much lower levels of employment than ethnic Bulgarians
Measuring ethnicity can be challenging. This is can be due to a combination of reasons, such as reaching hard-to-survey populations and the self-defined nature of ethnicity. Indeed, estimating the size of different ethnic groups, especially for Roma, has proved difficult in Bulgaria (see Box 3.1), though a commonly cited estimate puts the number of Roma at around 750 000 (Council for Europe, 2012[5]) and the Bulgarian census puts the Turkish minority at around 588 000 (NSI, 2011[8]). The 2021 census will soon provide a further and more recent estimate of the number of people in each ethnic group (which will also be single response).
Bulgarian law does not identify ethnic minorities and the Employment Promotion Act does not specify ethnic minorities as a disadvantaged group, however the Employment Promotion Act under Article 2 does forbid discrimination or privileges on ethnic grounds. Despite these difficulties in measuring ethnicity, and noting Bulgarian legislation, ethnicity remains an important dimension for the analysis presented here, as there are large differences in outcomes across ethnic groups in Bulgaria.
Figure 3.3 shows that the Roma population records extremely high rates of joblessness – more than double that of ethnic Bulgarians. Turkish minorities too have lower rates of employment than Bulgarians, though the differences are not as stark. Employment rates for working-age Roma men are 51%, compared to 65% for Turkish males and 76% for ethnic Bulgarian males. For women, the differences are even larger with employment rates of 31%, 48% and 71% for ethnic Roma, Turkish and Bulgarians respectively.
Figure 3.3. Roma experience very high rates of joblessness
Note: Unemployed and inactive exclude students so that all three categories are distinct. Students shown refers to students not working.
Source: OECD calculations based on the European Union Statistics on Income and Living Conditions (EU-SILC) 2019.
The data suggest that many Roma men out-of-work are actively seeking work, with 42% of them being unemployed rather than inactive (excluding students). For Roma women this is not the case with only 13.4% of the jobless seeking work. This may be related in part to cultural differences with, for example, Roma women having an earlier age at their first birth than Turkish and Bulgarian women (Koytcheva and Philipov, 2008[9]). Care barriers are looked at in Section 3.3 and further information on Roma is provided in Annex 3.B. The exact unemployed/inactive split given here should be interpreted with some care given the issues discussed in Box 3.1 measuring both ethnicity and labour force status in SILC.
Roma spend much longer periods out-of-work compared to ethnic Bulgarians too. Of those Roma out-of-work at the time of the 2019 SILC survey, 85% had not worked at all in 2018 compared to 75% for ethnic Bulgarians. Similarly, of Roma registered with the NEA in December 2019, the median Roma person had been registered with the NEA for 286 days compared to 117 days for the median ethnic Bulgarian and indeed the NEA report that Roma transition to employment at below average rates.
There are wide regional disparities in employment outcomes
As discussed in Chapter 2 there are wide regional labour market disparities across Bulgaria which are greater than in most OECD and EU countries (OECD, 2021[10]; OECD, 2021[11]; Hermansen, 2021[12]). Comparing joblessness across Bulgaria’s regions, shows that the North Western and Southern Central regions have the lowest employment rates and the highest unemployment rates in Bulgaria. The North Western region has the lowest employment and highest unemployment rates. However, since the North Western region is small, in absolute terms the largest number of unemployed people are found in the Southern Central region.
In addition to regional disparities, there are large differences in employment outcomes between urban and rural environments. The 1.3 million Bulgarians living in rural areas face unemployment rates more than double those in densely populated places (around 11.7% compared to about 4.3% from SILC 2019, Table 3.2). This difference is greater than the largest difference among planning regions (4.6% in the South Western compared to 9.6% in the North Western area). The issue of living in a rural area, particularly without a car, is picked up in the Section 3.3.
Table 3.2. Labour market statistics by region and urban/rural environment
Working age population (16‑64), 2019
Note: “Working age” 16‑64, based on SILC data. Unlike some other tables students are counted among the inactive and unemployed in these figures so as to be more comparable with other sources.
Source: OECD calculations based on the European Union Statistics on Income and Living Conditions (EU-SILC) 2019.
3.2.3. Five key groups to activate
The analysis above, in line with the existing literature, suggests five key groups that would benefit from activation policies. These groups overlap and are not mutually exclusive, Annex Table 3.A.1 shows the degree of overlap across these five groups. The groups are:
1. Youth NEETs aged 15‑29, ~170 000 people (2019 LFS estimate): Bulgaria faces some of the highest NEET rates in the EU and Bulgaria’s NEETs have previously been identified as a group of significant concern (e.g. the Institute for Market Economics (2019[3])).Activating Youth NEETs is particularly important as failing to acquiring human capital or skills can lead to long lasting harm to labour market opportunities. Annex Table 3.A.1 shows about 42% of Youth NEETs are Roma.
2. Ethnic minorities, ~360 000 people (2019 SILC estimate): Roma suffer rates of joblessness more than double that of ethnic Bulgarians. This makes them a highly relevant group for policy makers to consider. The European Commission have previously identified Roma as a group that could benefit from more support (European Commission, 2019[13]; European Commission, 2020[14]; European Commission, 2021[15]). Turkish minorities, albeit to a lesser extent, also have lower levels of employment than ethnic Bulgarians. With perhaps around 750 000 Roma living in Bulgaria and around 588 000 ethnic Turkish people, a special focus of activation strategies on ethnic minorities is therefore important.
3. People out of work due to care and family commitments, ~290 000 people (2019 LFS estimate): This is the most commonly cited reason for non-student inactivity in Bulgaria and it is cited at one of the highest rates among inactive populations in Europe. Strikingly, more than 99% of people in this category are women (SILC 2019 estimate).
4. People out of work for reason of illness or disability, ~190 000 people (2019 LFS estimate): A significant share (25%, LFS 2019; 27% SILC 2019) of non-student inactive people say they are out-of-work due to illness or disability. The next section shows people living in Bulgaria with a disability have low levels of employment compared to people living with a disability in other EU countries. This suggests that there is scope to improve the labour market outcomes of this population in Bulgaria.
5. Older people (55‑64) out-of-work ~340 000 people (2019 LFS estimate): While the number of retirees among 55‑64 year‑olds has been falling in recent years and employment for this group is above the EU average, older working age people are twice as likely to be out-of-work as prime aged individuals 30 to 54 years old.
When considering who activation policies can be most effective for, it is important to identify groups of people who are far from the labour market as well as to recognise that different people face different barriers to labour market participation. Hence the approach used to select the groups above, chose groups that might be expected to face different needs and (all else equal) prioritises groups that have weaker labour market outcomes and groups that are larger. Nevertheless there is no one way to identify groups with activation potential. In another study, Sundaram et al. (2014[1]), also use a judgement based approach to rank the importance of similar groups although the authors first used the “Faces of Joblessness” methodology to define these groups. This “Faces of Joblessness” method uses an algorithmic approach described in Fernandez et al. (2016[16]).
As the five groups are defined in such a way that they might be expected to face different barriers to labour market participation, the categorisation given here helps to separate out people with different needs. The needs of these five groups are then analysed in Section 3.3 which looks at barriers to employment. This means that some factors, notably education, are analysed as barriers to employment below rather than being used above as demographics defining key groups to activate.
These groups taken together cover more than four‑fifths of the working-age jobless population. The residual covers people not from ethnic minorities, aged 30‑54, who do not report that they are inactive due to illness, disability, or family commitments (though may report being inactive for other reasons).
3.3. Employment barriers facing inactive and unemployed people
ALMPs can be an effective tool for improving labour market outcomes for the inactive. However, the optimal mix of ALMPs needed to effectively and efficiently activate them depends on what barriers people face. This section investigates the barriers to employment for the out-of-work, including for the five groups defined above, to gain a better understanding of where ALMPs may be most helpful.
ALMPs can be broadly classified into three types: those that strengthen people’s motivation to work; those that improve people’s labour-supply capabilities; and those that expand people’s opportunities through intermediation and improved labour demand (Immervoll and Scarpetta, 2012[17]; OECD, 2015[18]). This chapter follows this framework (as is also applied in Sundaram et al. (2014[1])) and groups labour market barriers under these categories.
Five barrier types from 13 individual obstacles are defined in Box 3.2. Three types of barriers relate to workers capacity (experience and skills; health; and family/care commitments); one barrier type relates to opportunities (those geographically distant from the labour market); and one final barrier type related to motivation (household income). People often face multiple types of employment barriers, therefore the analysis in this section also observes the percentage of people facing at least one, two, or three types of barriers.
The rest of this section is laid out as follows: Box 3.2 provides definitions of various barriers to labour market participation. Then, the different barriers to labour market participation facing the five groups identified as a priority for activation above are discussed in Section 3.3.1. Finally, some of the barriers that cut across groups are discussed in Section 3.3.2.
Box 3.2. Capturing labour market barriers with SILC data
This box defines five different types of barriers to employment (measured using 13 specific barriers). The barrier groupings defined here closely follow OECD (2021[11]) differing only slightly and primarily for reasons related to data availability. The first three barrier types broadly relate to people’s capacity for work. The geographic distance barrier relates to people’s work opportunities and finally the high household income barrier relates to people’s motivation to work. The barriers to labour market participation defined here are not exhaustive, however they jointly provide a rich picture on the obstacles to employment for the out-of-work population.
Skills and experience barriers. Lack of skills and experience reduce opportunities to find a good job. In this analysis, skills and experience barriers are measured with four variables: 1) Low education defined as ISCED 2011 level 0‑2 (lower secondary education or below); 2) whether the most recent (or current) role was in an “elementary occupation” (ISCO Code 08 classification 91‑96, those who have never worked are also counted as having this barrier); 3) absence of recent work experience, defined as not having worked at all in 2018, the calendar year prior to the survey; and 4) whether a person has never worked.
Health related barriers: Poor health can reduce people’s capability to work. Two variables relate to health challenges: 1) whether a person reports a chronic health problem; and 2) whether they report that a health problem causes limitations or sever limitations in their usual activities.
Family related barriers: Care responsibilities at home can reduce the time people have available for paid-work. Three measures seek to capture care these barriers: 1) Whether there is a child three or younger in the household; 2) whether there is a person aged 80 or over in the household; 3) whether there is a person in the household with a severe health limitation and who is inactive due to disability (this question is only asked of primary respondents 16 years or over; hence this measure excludes those who care for children with disabilities). Finally, living in a household where the entire family is far from the labour market may make forming links to the labour market including acquiring information on job search and job readiness from family members harder and could sometimes involve overcoming inter-generational joblessness. So one further family related non-care barrier is included: 4) everyone in the household is out-of-work.
Geographic distance barrier: Being far from a local labour market hub limits the opportunities people have for work. Indeed, this relates to a labour demand barrier in the sense that there may be little demand for labour within this person’s effective commute area. A variable that reflects a geographic distance barrier is defined as living in a rural area in a household without a car.
High household income barrier: Some of those out-of-work choose not to work because they have sufficient funds (from benefits, non-labour income, or other household members) that they have low monetary incentives to work. This barrier attempts to capture this by looking at whether the person’s equivalised disposable household income is in the top quintile of the distribution.
Care should be taken when interpreting these barriers to labour market participation. In particular, the direction of causality is complex. The barriers defined above can decrease labour market attachment but conversely, in some cases, being out of work can worsen the above labour market barriers themselves. For example, without income from a job it is harder to afford a car, potentially creating a geographic distance barrier.
Source: OECD (2021), Improving the Provision of Active Labour Market Policies in Estonia, https://dx.doi.org/10.1787/31f72c5b-en.
3.3.1. Different groups face different barriers to labour market participation
Table 3.3 shows the percentage of people facing different barriers to labour market participation. Motivated by the previous analysis, this section examines the prevalence of these barriers across the five key sub‑groups defined in Section 3.2.3: namely youth NEETs, older jobless people aged 55‑64, those out of work for illness/disability, those out of work due to family/care obligations, and Roma. In addition, for working age non-students, Table 3.3 also shows the prevalence of barriers faced by the inactive, the unemployed, and, for comparison purpose, the employed – irrespective of which of the five key groups (if any) these people fall into.
Table 3.3. Barriers to employment in Bulgaria vary across different populations
Working age population (16‑64), 2019
Note: “Working age” population 16‑64. Students under 30 are excluded. “Youth NEET” refers to NEETs age 15‑29. “Older out-of-work” refers to people out-of-work aged 55‑64 (regardless of inactivity/unemployment classification). “Inactive with a disability” and “Inactive family/care” refers to people who self-define as inactive as they state they do not search for work due to disability/care obligations. Unemployed rather than inactive people with disabilities/care responsibilities can be identified under the “Unemployed” column and by looking up the “health”/”family” barriers. The three labour force status columns are defined using a bespoke combination of variables that more closely replicates the ILO definitions than the standard SILC definitions (see Box 3.2).
Source: OECD calculations based on the European Union Statistics on Income and Living Conditions (EU-SILC) 2019.
Youth NEETs often lack skills and experience and many also face care barriers
Engaging youth in education, society, and the labour market is important both for their individual well-being and economic futures as well as for those of the country as a whole (OECD, 2021[19]). In Bulgaria, youth NEETs often struggle with low qualifications and frequently face care barriers related to younger children. Just over half of youth NEETs have lower secondary education or less and about half of youth NEETs live in households with children under three, although these may not necessarily be their own children.
Other common barriers for youth also reflect, in part, their age. For example, many youth NEETs have not yet made the transition from education into a first job – with nearly 60% having never worked and three‑quarters having not worked in the past year. Again, reflecting their age, compared to other groups, a relatively large share of youth NEETs, about one‑quarter, live in rural areas without a car. Over time, as people in this group age, these barriers may lessen, however, they do represent a contemporaneous challenge in managing the move into work.
Less than 8% of youth NEETs live in high income households. If this were the only factor affecting motivation, it would suggest low motivation barriers to employment for youth. However, this data does not measure all motivational issues. Indeed, one study has argued that low motivation – in particular too high reservation wages and reliance on other family members income including remittances from abroad – is an issue affecting youth in Bulgaria (Institute for Market Economics, 2019[3]). The study however has limitations and is based not on measuring youth motivation directly, but rather on a small focus group of 39 “local experts” from among the NEA, Roma and youth mediators, municipal officials, regional education management bodies of the Ministry of Educational and Science, industrial associations and NGOs working with youth NEETs. Unfortunately, the data used in this chapter are not able to offer more quantitative insights into remittances from abroad or reservation wages.
Ethnic minorities face very high barriers to labour market inclusion
As discussed above, ethnic minorities, particularly Roma, and to a lesser extent the Turkish ethnic minority, have weaker employment outcomes than ethnic Bulgarians.
Table 3.3 shows Roma people without jobs face some of the highest barriers to employment of any group. Compared to out-of-work Bulgarian and Turkish ethnic groups, out-of-work Roma are the most likely to face a skill or experience barrier, the most likely to face a family related barrier and the most likely to live in rural areas without a car in the household.1 About 82% of out-of-work Roma face multiple types of barriers to employment compared to only around 69% for ethnic Bulgarian’s. Taken together the high prevalence of so many barriers shows the need to prioritise supporting Roma.
A key challenge in understanding the issues facing Roma is limited data availability, both in Bulgaria and across the EU. Indeed, even the size of the Roma population is difficult to estimate precisely (see Box 3.1). However, despite these data limitations, there are known further challenges facing Roma beyond those analysed in Table 3.3. These include discrimination in the labour market and beyond, higher risks of poverty, lower use of early childhood education and childcare, and many Roma living in low quality and overcrowded housing within segregated Roma neighbourhoods. Recognising the importance of lifting Roma labour market participation, Annex 3.B contains a review of the labour market literature on Roma and further analysis of Roma using the NEA data is also shown in Section 3.4. Chapter 5 includes a discussion of the NEA’s use of Roma mediators for outreach to Roma.
Out-of-work Turkish people also face high education and skills barriers. However, out-of-work Turkish people face comparable rates of health and family barriers as out-of-work ethnic Bulgarians. Among the out-of-work, both Roma and Turkish people are less likely to live in high income households than ethnic Bulgarians.
Women citing care related reasons for not working often also lack skills and experience
Among mothers with very young children (less than two years old), those in Bulgaria use less child care and have lower employment rates than those across many OECD and EU countries (Figure 3.4). This may in part reflect Bulgaria’s relatively long duration of paid maternity leave which is discussed in Chapter 4. Out-of-work Roma in particular are more than twice as likely as out-of-work Turkish and Bulgarian ethnic groups to live with children under three (Table 3.3).
Figure 3.4. Mothers with very young children in Bulgaria have low rates of employment and use of formal care
Note: Shown are EU and OECD countries with sufficient data available in the OECD Families Database. The EU and OECD are unweighted averages of the member countries shown in each panel. Due to data availability there are some differences in the methodology applied across countries. Additional details are provided in the source below.
Source: OECD Family Database https://www.oecd.org/els/family/database.htm.
While not all mothers with very young children will want to work, and indeed should not all be expected to work, greater availability of child care could make it easier for mothers to return to work for those who want to. Indeed, universal access to early childhood education for those aged four has previously been recommended for Bulgaria (OECD, 2021[11]). Cultural changes too could help women return to work. A full-time work culture potentially makes it harder to balance parenting with work and very few men (less than 1% in SILC 2019) cite care related reasons as their main reason for not working which potentially highlights a greater role for men to play in balancing care burdens within families.
As a mother’s youngest child grows older, however, many mothers do return to employment. Employment rates in Bulgaria rise to about 80% for mothers whose youngest child is aged 6 to 14 – similar to the OECD average of 79% for this group (Figure 3.4).
However, there is a group of women that are not returning – or initially joining – the labour force but that are citing care barriers as their main reason for inactivity. In Table 3.3 the “Inactive/Family care” column shows the barriers faced by those who state their main reason for not seeking work is related to family or caring. Table 3.3 shows that many citing care as a barrier to work do not live with young children in their household. Indeed, only about half of those in this group actually face a care barrier of the type defined in Table 3.3.
This means that many of those out-of-work for care related reasons are still not working even though their children (if they have any) are no longer very young. Hence, many mothers citing care and family reasons for not working may instead be held back from the labour market by another barrier especially a lack of skills and experience. In fact, more than 90% of those in the “Inactive family/care” group face a skills or experience barrier. Activation efforts that build skills and experience, including ALMP provision, might therefore benefit many in this group.
Employment rates for people with disabilities are low internationally
While not everyone with a disability is able to work, it is important to support into employment those who can, and, indeed, in many cases want to work. In fact, for those with disabilities that can work, getting a job can not only reduce welfare costs to governments, but can also reduce poverty, improve social inclusion and improve the mental health of individuals with disabilities (OECD, 2010[20]). Employment rates for persons with disabilities are low in Bulgaria compared to other European countries (Figure 3.5). This suggests ways need to be found, where possible and appropriate, to help more of this group into work.
Table 3.3 shows that in addition to health problems, people with disabilities also face low levels of education and long breaks from working (both of which may often be caused by the disability itself). Nearly half of people with a disability live in a jobless household. Indeed, people with disabilities often have complex needs with more than 70% facing multiple types of barriers to employment (Table 3.3).
ALMPs have a role in improving employment outcomes for those with disabilities by targeting these barriers. For example, lifting skill levels and training or re‑training for appropriate roles can support people into jobs (for example, re‑training to desk based work may help some with physical disabilities) and subsidies to employers and support for sheltered employment in specialised enterprises can incentivise employers to hire people with disabilities. More information on the ALMPs Bulgaria provides is discussed in Chapter 6.
More generally, achieving better health and employment outcomes for people with disabilities involves having well-functioning, accessible, and well-co‑ordinated health and public employment services that are tailored to individual needs as well as incentives that make work pay both for workers and for employers (OECD, 2010[20]). Making work pay for workers requires that disability benefits (discussed in Chapter 4) need to balance providing income support for people who cannot work with incentivising work for those who can.
Figure 3.5. Bulgaria has low levels of employment for people with disabilities
Note: Living with disability is measured by having a chronic health condition and reporting of limitations in activities due to health. The data excludes a small number of people in some countries whose employment status is unknown. Data for Ireland and Italy refer to 2018.
Source: OECD calculations based on the European Union Statistics on Income and Living Conditions (EU-SILC), 2019.
Many older workers have potential to contribute to Bulgaria’s shrinking labour market
Many 55‑64 year‑old jobless people have good potential to work and contribute to Bulgaria’s economy. Of the five key groups to activate shown in Table 3.3, jobless 55‑64 year‑olds have: the lowest percentage whose last job was in an unskilled sector, the lowest percentage who have never worked, and the lowest percentage with only lower-secondary education (or less).These results suggests many older workers have good skills to contribute with, though in some case ALMPs may support retraining. Given their age, many persons in this group face health problems. Table 3.3 shows a health issue for about half of this age group.
Motivation for older people out-of-work may be understated by the “motivation/high income” barrier. Indeed, many jobless people in this age group are no longer searching for work: with the previous section showing that the vast majority of this group are inactive rather than unemployed. Further, many older jobless individuals live in entirely jobless households and 83% have not worked in the previous calendar year. Indeed, the fact that education and other skill barriers (except recent work experience) are low relative to other groups in Table 3.4 in part be due to some well qualified individuals preferring to live off savings rather than try to re‑enter the workforce.
Bulgaria has made efforts to increase its retirement age in recent years which may increase labour market participation by increasing motivation to work. ALMPs too can potentially play a role in reconnecting this group to the workforce and targeted trainings to lift the skills of this group have been recommended for Bulgaria before (OECD, 2018[21]).
3.3.2. Many barriers cut across multiple groups
Many of the barriers to labour market integration in Table 3.3 cut across different groups that are a priority for activation. In fact, more than 80% of out-of-work people face a skills or education barrier and more than half face a family or care related barrier. Indeed, even when barriers are uncommon within a group there are typically still some individuals in each group who face them. For example, health barriers are present in only 4.6% of youth NEETs, but for some of these individuals this may be the most important factor preventing them from their active participation education or employment. For this reason, while the analysis in Table 3.3 can inform high-level decisions (for example about where there may be widespread demand for ALMPs), activation services to specific people should be individually tailored. This topic, the process of service provision to PES clients, is discussed in Chapter 6.
Another noticeable, cross-cutting feature, is the large pool of people have not worked for a long time. The SILC data in Table 3.3 shows 82% of the inactive population and 65% of the unemployed population did not work in the last calendar year (i.e. 2018 for the 2019 SILC). These numbers are somewhat higher than the traditional measures of long-term unemployment found in the LFS, where long-term unemployment captures people who have been out-of-work and searching for a job for more than 12 months rather than merely out-of-work. However, the LFS long-term unemployment numbers confirm that many of the unemployed are long term unemployed, with this share fluctuating between 53‑59% during Q1 2019‑Q2 2019 (the period when the 2019 SILC survey was conducted). Similarly, Chapter 2 showed that fewer than 40% of the inactive have any work experience in the last five years with inactivity for most being a longer-term issue.
Living in a rural area without a car, a geographic distance barrier, affects a minority of the out-of-work population: with about 21% of the inactive and 26% of the unemployed population affected by this barrier (Table 3.3). However, in an international context, this is a higher lever level than in all but two countries for which sufficient data is available (Figure 3.6). People who live in a remote area without access to good transport can face few job opportunities. This barrier affects many groups but is particularly prevalent among Roma people (42%).
Further exacerbating geographic distance barriers are Bulgaria’s high levels of regional inequality combined with underdeveloped and under maintained transport infrastructure (OECD, 2021[11]). Investments in better transport infrastructure including on roads and rail, as well as in better digital infrastructure, could better connect Bulgaria’s regions (OECD, 2021[11]).
Beyond good regional development policy, policies that encourage mobility and activation policies that improve the attractiveness of workers may help improve the labour market outcomes of people living in remote areas. Indeed, only 3% of Bulgarian’s moved to a new dwelling between 2007 and 2012, which while potentially underestimated due to large outward migration, is very low compared to the EU average of 16% (Hermansen, 2021[12]). Part of the reason for low mobility may be Bulgaria’s high level of home ownership with several international studies supporting the association between owning a home and reduced residential mobility (Causa and Pichelmann, 2020[22]). The subsidies Bulgaria provides to help support mobility are discussed briefly in Chapter 6.
Finally, reaching out-of-work populations in remote areas is a challenge for the PES. To reach people in distant areas the NEA uses mobile labour offices, which are discussed in Chapter 5.
Figure 3.6. Geographic distance barriers are high for out-of-work people in Bulgaria
Note: Geographic distance barrier is defined as living in a thinly populated area and in a household without a car. Due to data comparability across country the ‘self-defined’ measure of out-of-work is used in this chart (see Box 3.1) and students are included so that the numbers are not directly comparable to Table 3.3. Data for Ireland and Italy refer to 2018.
Source: OECD calculations based on the European Union Statistics on Income and Living Conditions (EU-SILC), 2019.
3.4. Typical profiles of registered jobseekers: Who are the unemployed and inactive the NEA reaches?
This section describes the profile of jobseekers who are registered with the NEA, complementing the discussion on inactive and unemployed people in the previous sections of the chapter. Since registered jobseekers are typically expected to search for employment, the NEA’s clients examined in this section should generally be considered closer to unemployment rather than inactivity even though the ILO definitions of these concepts do not exactly overlap with registration with the PES.2 In particular, the section uses administrative NEA data to describe common characteristics of NEA clients and identify potential barriers to employment they are facing. In contrast to the other parts of the chapter, the discussions in this section only apply to jobseekers who are in contact with the NEA and do not extend to other unemployed and inactive. Therefore, the discussions provide insights on the situation of inactive and unemployed who the NEA can support through well-tailored support, helping them to alleviate their employment barriers. The section relates closely to Chapter 5, which analyses outreach activities to jobseekers in greater detail and uses the same administrative NEA dataset to estimate the number of NEA clients compared to the size of the out-of-work population.
The vast majority of NEA clients are unemployed (96.1%), while pensioners (1.3%), people who are employed (1.2%) and students (0.5%) account for a small share of NEA clients only. Almost 40% of registered jobseekers are 50 years old or older, whereas the share of people under 25 is very low, at only 5%. In no other OECD or EU country for which data is available, the share of young people under 25 among all registered jobseekers is as low as in Bulgaria (see Chapter 5). On average across EU countries, 11% of registered jobseekers are under 25 years, and up to almost one‑fifth in Belgium (European Commission, 2019[23]).
Most NEA clients are women, accounting for 56.2% of NEA clients, with the number of female NEA clients exceeding that of men by about one‑third between ages 30 and 60 (Figure 3.7). At younger ages, the higher share of women among NEA clients mirrors differences in labour market patterns across genders, i.e. a significantly lower employment rate for young women than for young men, including due to maternity. For prime‑aged and older workers, however, the employment gender gap is small, at about 2 percentage points among 45‑59 year‑olds, suggesting that other factors contribute to the larger number of women among NEA clients, e.g. better-performing outreach to women. Only past age 60, the majority of registered jobseekers are men, most notably due to a lower statutory retirement age for women than for men, at 61.3 years for women against 64.2 years for men in 2019 (MISSOC, 2019[24]).
Figure 3.7. Only few young Bulgarians register with the NEA and most NEA clients are women
Source: National Employment Agency micro data.
More than two‑thirds (68%) of registered jobseekers self-identify as ethnic Bulgarians, while 15% are Roma, 12% belong to the Turkish community and 5% of NEA customers belong to another ethnic group or their ethnicity is not reported in the data. Jobseekers from ethnic minorities are highly concentrated in a few parts of the country, in line with general demographic patterns. For instance, close to 60% of registered jobseekers in Razgrad belong to the Turkish community and 40% of jobseekers in Sliven are Roma, while in Sofia city the Turkish and Roma community jointly account for less than 3% of jobseekers.
Socio‑economic and individual characteristics vary strongly across gender, ethnic groups and age, highlighting that registered jobseekers in Bulgaria are far from forming a homogeneous group and that individualised approaches are required to support jobseekers depending on their individual circumstances (Table 3.4).
Many registered jobseekers have a low degree of education (lower secondary education or below, 43%) or a medium level of education (upper-secondary or post-secondary non-tertiary education, 45%) while only a minority of NEA clients has tertiary education (12%). The educational attainment of registered jobseekers is comparatively high among ethnic Bulgarians, with 56% having a medium level and 16% a high level of education, whereas two‑thirds (67%) of NEA clients from the Turkish community and the overwhelming majority of jobseekers from the Roma community (92%) only have low or no completed education.
On the same note, skills and competences are unevenly spread among registered jobseekers. Especially older people and jobseekers belonging to ethnic minorities are less likely to possess specialised skills that could facilitate their job search. For instance, 18% of registered jobseekers have known or certified digital skills and 14% have English skills, but among jobseekers aged 55 or older, they are only 8% and 3%, respectively. Similarly, only 5% of jobseekers from the Turkish ethnic community and 1% of NEA clients from the Roma community have known digital skills, respectively. These numbers highlight the high frequency of skills barriers among the out-of-work population (see Section 3.3), including jobseekers registered with the NEA, in particular among vulnerable groups.
Table 3.4. Main characteristics of jobseekers registered with the NEA
Characteristics of jobseekers who were registered with the Bulgarian NEA on 31.12.2019, by gender, ethnicity and age
Note: People whose ethnicity is unknown or other than ethnic Bulgarian, Turkish or Roma (about 6% of NEA clients) are not included in the statistics by ethnic group. “Pre‑school children” refers to children under seven. Information on children, health problems and skills are self-reported and might not be comprehensive.
Source: National Employment Agency (NEA) micro data.
Caring responsibilities can be a barrier to employment, too. While there is no information on care tasks for elderly family members, the NEA data includes information on whether the jobseeker reports to have children or not. Among jobseeker who are registered with the NEA, 13% report having one child under the age of 16, 11% have two children and 5% report having at least three children (Table 3.4). These numbers may under-estimate the real number of children, as they are self-reported and some job-seekers may fail to indicate that they have children, in particular in cases where family circumstances do not influence benefit entitlements. Among registered jobseekers from the Roma community, the share of people with at least three children under 16 is much higher than among other NEA clients, at 13%, suggesting that care responsibilities could be particularly widespread in this group. Similarly, care responsibilities for children are likely to be a stronger obstacle to employment for young people than for prime‑aged jobseekers, as 30% of 15‑29% registered jobseekers report having pre‑school children under seven, against 13% among all registered jobseekers.
Recognised health problems are not common among registered jobseekers, concerning only 7% of NEA clients, which is less than suggested by the estimates on the prevalence of health impediments among inactive and unemployed people reported in Section 3.3. The low number of health problems in the NEA stems partially from the fact that health issues are only reported in case they are officially diagnosed and affect a person’s work ability. In addition, people with very severe health problems rarely register with the NEA. Many of them are entirely unavailable for the labour market and do not look for employment.
Recognised health problems shown in the data are particularly uncommon among jobseekers belonging to the Roma community, at less than 5%, against more than 8% among ethnic Bulgarians. However, Roma jobseekers who have a recognised health problem tend to have quite severe impediments, with an estimated average remaining work ability of only 57%, against 62% among jobseekers from the Turkish ethnic community with a health problem and 66% among ethnic Bulgarians.
Vulnerable groups tend to remain registered with the NEA for longer periods because they do not find work, further reinforcing their labour market obstacles due to longer periods without work experience. For example, median registration times at the end of 2019 show that jobseekers from the Roma community had been registered for 286 days, against 144 days among all jobseekers (Table 3.4). For job seekers aged 55 and older, the median registration length was 201 days.
3.5. Key findings
In 2019, prior to the impact of COVID‑19, there were about 1.3 million unemployed and inactive people of working age in Bulgaria, around 900 000 of whom were not studying. These 900 000, represent around 20% of the working‑age population, which provides a large pool of people who could benefit from active labour market policies. Naturally, the out-of-work population is a diverse group, with people facing different barriers and having different reasons for not working. This chapter therefore groups the out-of-work population that could benefit from activation support into five key groups based on their employment outcomes, size, and similar labour market barriers. The five key groups identified are:
Youth not in employment education or training (NEET) ~170 000 people: Bulgaria has some of the highest NEET rates in the EU suggesting there is scope for improvement in this area.
Ethnic minorities ~360 000 people: Amongst Bulgaria’s different ethnic groups, especially Roma suffer much higher rates of joblessness than other ethnic groups and face many barriers and challenges to labour market participation.
People out of work due to care and family commitments ~290 000 people: This concerns mainly women and is the most commonly cited reason for (non-student) inactivity in Bulgaria.
People out of work for illness and disability ~190 000 people: This is the second most cited reason for (non-student) inactivity in Bulgaria. Further, employment rates for people with health problems are some of the lowest in the EU.
Older working age people 55 – 64 who are out-of-work ~340 000 people: Older working age people are twice as likely to be out-of-work as prime aged individuals 30 to 54 years old. The many people in this group could contribute much if activated.
The chapter also zooms in on the different barriers to work of these groups, including skills barriers, health related barriers, family related barriers, and geographic distance barriers. While all of these groups should be a priority for labour market policies, the findings in this chapter also highlights the different needs of those groups. This suggests that different activation strategies may need to be found for them, which will be discussed in subsequent chapters.
References
[38] Angelova, R. et al. (2020), Education and Finding Employment in the Roma Community: Main Conclusions from a National Representative Survey of the Roma Community in Bulgaria, Trust for Social Alternative, Global Metrics Ltd., https://us4bg.org/wp-content/uploads/2021/01/Education-and-Employment-survey-report-2020-in-BG.pdf.
[39] Arbex, C. et al. (2013), Guide for Working with Roma Families towards Achieving the Success of their Children in School: A transnational methodological proposal for professionals, Fundación Secretariado Gitano, https://www.schooleducationgateway.eu/files/esl/downloads/08_Guide_for_working_with_Roma_families.pdf.
[47] Bogdanov, L. and G. Angelov (2006), Roma Integration in Bulgaria: Necessary Reforms and Economic Effects, Open Society Institute – Sofia, https://www.economedia.bg/getatt.php?filename=o_280347.pdf.
[59] Card, D., J. Kluve and A. Weber (2017), “What Works? A Meta Analysis of Recent Active Labor Market Program Evaluations”, Journal of the European Economic Association, Vol. 16/3, pp. 894-931, https://doi.org/10.1093/jeea/jvx028.
[22] Causa, O. and J. Pichelmann (2020), “Should I stay or should I go? Housing and residential mobility across OECD countries”, OECD Economics Department Working Papers, No. 1626, OECD Publishing, Paris, https://dx.doi.org/10.1787/d91329c2-en.
[55] Ciaian, P. and D. Kancs (2018), “Marginalisation of Roma: Root Causes and Possible Policy Actions”, European Review, Vol. 27, pp. 115–130, https://www.cambridge.org/core/services/aop-cambridge-core/content/view/33510787C9D12EF6B5B015F67B7C58C9/S106279871800056Xa.pdf/marginalisation_of_roma_root_causes_and_possible_policy_actions.pdf.
[5] Council for Europe (2012), Estimates and official numbers of Roma in Europe - Document prepared by the Support Team of the Special Representative of the Secretary General of the Council of Europe for Roma Issues, https://rm.coe.int/1680088ea9 (accessed on 23 July 2021).
[6] Council of Europe (2016), Ad Hoc Committee of Experts on Roma and Traveller Issues (CAHROM): Thematic Report of the Group of Experts on Roma Health Mediators, Council of Europe, Strasbourg, https://coe-romed.org/sites/default/files/documentation/CAHROM%20(2016)7%20EN%20Thematic%20report%20on%20Roma%20health%20mediators.pdf.
[25] Council of Europe (2012), Estimates and official numbers of Roma in Europe - Document prepared by the Support Team of the Special Representative of the Secretary General of the Council of Europe for Roma Issues, Council of Europe, https://rm.coe.int/1680088ea9.
[57] Council of Europe (2012), Human rights of Roma and Travellers in Europe, Council of Europe Publishing, https://rm.coe.int/the-human-rights-of-roma-and-travellers-in-europe/168079b434.
[44] Council of Ministers (2020), National Roma Equality, Inclusion and Participation Strategy of the Republic of Bulgaria (2021 - 2030) - draft, Council of Ministers, https://www.strategy.bg/FileHandler.ashx?fileId=24803+.
[28] Council of Ministers (2012), National Roma Integration Strategy of the Republic of Bulgaria (2012 - 2020), Council of Ministers, https://ec.europa.eu/info/sites/default/files/roma_bulgaria_strategy_en.pdf.
[43] Dimitrov, D., V. Grigorova and J. Decheva (2013), Civil Society Monitoring Report on the Implementation of the National Roma Integration Strategy and Decade Action Plan in 2012 in Bulgaria, Decade of Roma Inclusion Secretariat Foundation, https://www.rcc.int/romaintegration2020/romadecadefold/decade%20implementation/4.%20civil%20society%20reports/Civil%20Society%20Monitoring%20Report%202012%20(Bulgaria,%20local%20language).pdf.
[2] Dimitrov, Y. and N. Duell (2014), Activating and increasing employability of specific vulnerable groups in Bulgaria: a diagnostic of institutional capacity, The World Bank, Washington, D.C.
[31] EC (2018), Evaluation of the EU Framework for National Roma Integration Strategies up to 2020, Report on the evaluation of the EU Framework for National Roma Integration Strategies up to 2020, European Commission, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018SC0480&from=EN.
[63] EC (2015), Report on the implementation of the EU Framework for National Roma Integration Strategies 2015, European Commission, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52015DC0299&from=EN.
[27] EC (2020a), A Union of Equality: EU Roma strategic framework for equality, inclusion and participation, European Commission, https://eur-lex.europa.eu/resource.html?uri=cellar:9a007e7e-08ad-11eb-a511-01aa75ed71a1.0001.02/DOC_1&format=PDF.
[49] EC (2019c), Roma inclusion measures reported under the EU Framework for NRIS, Report on the implementation of national Roma integration strategies – 2019, Part 2/2, European Commission, https://ec.europa.eu/info/sites/default/files/cswd_roma_inclusion_measures_reported_under_the_eu_framework_for_nris_pt2_en.pdf.
[33] EC, DG JUST (2020), Roma Civil Monitor, A synthesis of civil society’s reports on the implementation of national Roma integration strategies in the European Union: Assessing the progress in four key policy areas, Publications Office of the European Union, https://cps.ceu.edu/sites/cps.ceu.edu/files/attachment/basicpage/3172/rcm-civil-society-monitoring-report-synthesis27-2019-eprint-fin.pdf.
[35] EC, DG JUST (2019), Civil society monitoring report on implementation of the national Roma integration strategies in Bulgaria: Assessing the progress in four key policy areas of the strategy, Publications Office of the European Union, http://amalipe.com/files/publications/RCM-Y2-C1-Bulgaria-PUBLISHED.pdf.
[36] EC, DG JUST (2018), A synthesis report on implementation of national Roma integration strategies in Bulgaria, Czech Republic, Hungary, Romania and Slovakia: Focusing on structural and horizontal preconditions for successful implementation of the strategy, Publications Office of the European Union, https://op.europa.eu/en/publication-detail/-/publication/446cb4f2-3c82-11e8-b5fe-01aa75ed71a1/language-en/format-PDF.
[34] EU Council (2021), Council Recommendation of 12 March 2021 on Roma equality, inclusion and participation 2021/C 93/01, Official Journal of the European Union, https://op.europa.eu/en/publication-detail/-/publication/05c5aac3-8855-11eb-ac4c-01aa75ed71a1/language-en.
[46] EU-FRA (2020), Coronavirus Pandemic in the EU – Impact on Roma and Travellers, Publications Office of the European Union, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2020-coronavirus-pandemic-eu-bulletin-roma_en.pdf.
[40] EU-FRA (2017), Second European Union Minorities and Discrimination Survey: Main results, Publications Office of the European Union, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2017-eu-midis-ii-main-results_en.pdf.
[50] EU-FRA (2014), Poverty and employment: the situation of Roma in 11 EU Member States; Roma survey – Data in focus, Publications Office of the European Union, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2014-roma-survey-dif-employment-1_en.pdf.
[53] EU-FRA (2016b), Second European Union Minorities and Discrimination Survey: Roma - Selected findings, Publications Office of the European Union, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2016-eu-minorities-survey-roma-selected-findings_en.pdf.
[15] European Commission (2021), “COUNCIL RECOMMENDATION of 12 March 2021 on Roma equality, inclusion and participation”, Official Journal of the European Union, Vol. 64.
[14] European Commission (2020), Recommendation for a COUNCIL RECOMMENDATION on the 2020 National Reform Programme of Bulgaria and delivering a Council opinion on the 2020 Convergence Programme of Bulgaria, European Commission, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52020DC0502&from=EN.
[23] European Commission (2019), Labour Market Policy indicators, https://webgate.ec.europa.eu/empl/redisstat/databrowser/view/LMP_RJRU/default/table?lang=en&category=lmp_indic.
[13] European Commission (2019), Recommendation for a COUNCIL RECOMMENDATION on the 2019 National Reform Programme of Bulgaria and delivering a Council opinion on the 2019 Convergence Programme of Bulgaria, European Commission, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52019DC0502&from=EN.
[30] European Parliament (2015), Evaluation of the EU Framework for National Roma Integration Strategies, European Union, https://www.europarl.europa.eu/RegData/etudes/STUD/2015/536485/IPOL_STU(2015)536485_EN.pdf.
[16] Fernandez, R. et al. (2016), Faces of Joblessness: Characterising Employment Barriers to Inform Policy“, OECD Publishing, Paris, https://doi.org/10.1787/5jlwvz47xptj-en.
[56] Frazer, H. and E. Marlier (2011), Promoting the Social Inclusion of Roma, Synthesis Report, EU Network of Independent Experts on Social Inclusion, https://www.euromanet.eu/upload/03/11/synthesis_report_2011-2_final_3.pdf.
[4] Haralampiev, K. and D. Blagoev (2014), Ethnicity non-identification in the 2011 census in Bulgaria, European Population Conference, Budapest 2014.
[12] Hermansen, M. (2021), “Reducing regional disparities for inclusive growth in Bulgaria”, OECD Economics Department Working Papers, No. 1663, OECD Publishing, Paris, https://dx.doi.org/10.1787/d18ba75b-en.
[52] Ilieva, N. (2019), Spatial Segregation - Theoretical Aspects, Bulgarian Academy of Sciences, http://harman-mahala.com/wp-content/uploads/2019/07/%D0%9F%D1%83%D0%B1%D0%BB%D0%B8%D0%BA%D0%B0%D1%86%D0%B8%D1%8F_1.pdf.
[17] Immervoll, H. and S. Scarpetta (2012), “Activation and employment support policies in OECD countries. An overview of current approaches”, IZA Journal of Labor Policy, Vol. 1/1, p. 9, https://doi.org/10.1186/2193-9004-1-9.
[3] Institute for Market Economics (2019), Assessment of the people not in employment, education and training (NEETs) in Bulgaria and policy measures to effectively their integration, European Commission.
[41] Jones, A. et al. (eds.) (2020), Roma Early Childhood Inclusion+: Special Report on Roma Inclusion in Early Childhood Education and Care - Republic of Bulgaria, Open Society Institute – Sofia, Roma Education Fund, UNICEF, https://reyn.eu/wp-content/uploads/2020/11/RECI_Republic-of-Bulgaria-Report_ENG.pdf.
[26] Kahanec, M. (2014), “Roma integration in European labor markets”, IZA World of Labor 2014: 39, pp. 1-10, https://wol.iza.org/uploads/articles/39/pdfs/roma-integration-in-european-labor-markets.pdf?v=1.
[9] Koytcheva, E. and D. Philipov (2008), “Bulgaria: Ethnic Differentials in Rapidly Declining Fertility”, Demographic Research, Vol. 19, pp. 361-402, http://www.jstor.org/stable/26349254.
[54] Mihailova, D. and A. Kachamov (2017), Roma Evictions and Demolition of Roma Houses: A Sustainable Solution for Roma Integration or a Problem of Roma Discrimination in Bulgaria?, Open Society Foundations, https://www.equalopportunities.eu/docs/REPORT-2017-en.pdf.
[45] Ministry of Finance (2019), Bulgaria 2030, Part 1, Analysis of the socio-economic development of the country after its accession to the EU, Ministry of Finance, https://www.minfin.bg/upload/41549/Bulgaria+2030+analiz.pdf.
[24] MISSOC (2019), Missoc Comparative Tables, https://www.missoc.org/missoc-database/comparative-tables/.
[7] National Social Security Institute (2021), Пенсии за осигурителен стаж и възраст [Pensions for length of service and age], https://www.noi.bg/en/pensions/grantpensions/1854-posv13 (accessed on 23 July 2021).
[62] NEA (2021), Месечни бюлетини - 2021, Април, https://www.az.government.bg/bg/stats/view/2/340/.
[61] NEA (2019), Месечни бюлетини - 2019, Декември, https://www.az.government.bg/bg/stats/view/2/305/.
[8] NSI (2011), 2011 Population Census - main results, https://www.nsi.bg/census2011/PDOCS2/Census2011final_en.pdf.
[10] OECD (2021), Decentralisation and Regionalisation in Bulgaria: Towards Balanced Regional Development, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/b5ab8109-en.
[60] OECD (2021), Improving the Provision of Active Labour Market Policies in Estonia, Connecting People with Jobs, OECD Publishing, Paris, https://dx.doi.org/10.1787/31f72c5b-en.
[11] OECD (2021), OECD Economic Surveys: Bulgaria 2021, OECD Publishing, Paris, https://doi.org/10.1787/1fe2940d-en.
[19] OECD (2021), The Updated OECD Youth Action Plan: Building Blocks for Future Action, OECD, Paris, https://www.oecd.org/employment/youth/The-Updated-OECD-Youth-Action-Plan.pdf.
[21] OECD (2018), Key policies to promote longer working lives: Bulgaria country note 2007 - 2017, OECD, Paris, https://www.oecd.org/els/emp/Bulgaria%20Key%20policies_Final.pdf.
[18] OECD (2015), “Activation policies for more inclusive labour markets”, in OECD Employment Outlook 2015, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2015-en.
[20] OECD (2010), Sickness, Disability and Work: Breaking the Barriers: A Synthesis of Findings across OECD Countries, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264088856-en.
[29] Pamporov, A., E. Markova and G. Yordanova (2020), Comprehensive report for evaluation of Roma integration policies in Bulgaria in the period 2012-2019, Bulgarian Academy of Sciences, Institute of Philosophy and Sociology, http://www.nccedi.government.bg/sites/default/files/2021-01/Final_Assessment_Report_Roma_Strategy_2020_Final.pdf.
[58] Parekh, N. and T. Rose (2011), “Health inequalities of the Roma in Europe: A literature review”, Cent Eur J Public Health, Vol. 19/3, pp. 139-142.
[51] Roy, H. (ed.) (2017), Promoting Roma Inclusion, Policy Paper, Eurodiaconia, https://www.eurodiaconia.org/wordpress/wp-content/uploads/2018/02/Policy-Paper-Roma-Inclusion.pdf.
[32] Soler Penadés, V. and M. Corell Doménech (eds.) (2016), Fighting discrimination and anti-Gypsyism in Education and Employment in EU, Florida Universitaria, http://projectpal.eu/wp-content/uploads/2017/06/PAL_WS1.6.-Analysis-of-data-collected-and-research-finalization-Final-version.pdf.
[1] Sundaram, R. et al. (2014), Portraits of Labor Market Exclusion, The World Bank, Washington DC.
[37] Tomova, I. and L. Stoychev (2017), Analysis and expert assessment of the progress of the Republic of Bulgaria on the main priorities for social inclusion of Roma, Bulgarian Academy of Sciences, Institute for Population and Human Studies, http://www.nccedi.government.bg/sites/default/files/2019-01/BAS%20Report_NCCEII_V2.0.pdf.
[48] UNICEF (2016), Research on the Social Norms which Prevent Roma Girls from Access to Education, United Nations Children’s Fund, https://www.unicef.org/bulgaria/media/1886/file/Social-norms-which-prevent-roma-girls-from-access-to-education-EN.pdf.
[42] World Bank (2015), Bulgaria’s Potential for Sustainable Growth and Shared Prosperity: Systematic Country Diagnostic, World Bank Group, https://documents1.worldbank.org/curated/en/521881467994600656/pdf/99721-CAS-P151391-SecM2015-0217-2-IFC-SecM2015-0085-2-Box393212B-PUBLIC-mark-VERSION-REVISED.pdf.
Annex 3.A. Supplementary statistics
Annex Figure 3.A.1. Working age (15‑64) retirees in Bulgaria have fallen dramatically since 2011
Source: European Union Labour Force Survey (EU-LFS).
Annex Table 3.A.1. Overlap across key groups in Bulgaria
2019
Note: “Working age” is 16‑64. Excludes students under 30. The cells show the number of people who are in both the column and row group. For example, looking at the “Older worker” column and reading down to the “Inactive family/caring” cell the table shows there are 13 000 people who are in both these groups. “Other out of-works” shows jobless people who are not in any of the five categories. The second row “Group as share of all (non-students) out-of-work” shows the size of the group as a percentage of all non-student working-age people out-of-work.
1. SILC estimates of the group sizes differ to the LFS figures cited in text as there are methodological differences between the surveys.
Source: OECD calculations based on the European Union Statistics on Income and Living Conditions (EU-SILC) 2019.
Annex 3.B. Literature review on barriers to labour market integration of Roma in Bulgaria and the EU
Roma are considered Europe’s largest ethnic minority, but across the European Union (EU) Roma face major obstacles to labour market integration and lag behind their non-Roma peers with respect to virtually all indicators of social and economic inclusion. Addressing this gap between Roma and non-Roma population requires efforts in many policy areas. This Annex presents figures on the size of the Roma population cross Europe, discusses the challenge of a generally poor evidence base on Roma, identifies important barriers to labour market integration and examines the challenging issue of persistent discrimination of people of Roma origin, which is a further obstacle to their labour market integration.
Roma as one of the largest ethnic minorities in Europe
Roma are considered Europe’s largest ethnic minority, with an estimated 10‑12 million Roma living in Europe in 2012, based on estimates by the Council of Europe (Council of Europe, 2012[25]) and an estimated 5 million Roma living in European countries outside the EU (Council of Europe, 2012[25]). In several EU countries Roma are estimated to present more than 7% of the population. Other countries have smaller Roma population shares in comparison to the general population (Annex Table 3.B.1). In comparison to the estimates from the Council of Europe, official statistics often show a much smaller number of Roma. Especially in EU countries with large Roma populations, Census data are markedly lower than the estimates of the Council of Europe. The observed discrepancy and the absence of more recent estimates highlights the lack of data on Roma populations in EU member states (Kahanec, 2014[26]).
In EU countries with a larger proportion of Roma, Roma represent a growing share of the school-age population and the future labour force (EC, 2020a[27]). In Bulgaria, the Roma population is fairly young: the Bulgarian National Roma Integration Strategy (NRIS; Annex Box 3.B.1) reports that 72.6% of the Roma population is less than 40 years old (Council of Ministers, 2012[28]). Official data from the last Census in Bulgaria, conducted in 2011, indicate that 325 343 people (or 4.9% of the population) identified themselves as Roma. In comparison to the 2001 Census, this would, however, represent a decline of the Roma population, which was then estimated at around 371 000 people (Pamporov, Markova and Yordanova, 2020[29]). One reason for the lower numbers in Census data in comparison to the Council of Europe may be that many Roma self-identify themselves as Bulgarians, Turks, Romanians, etc. (Council of Ministers, 2012[28]). Census ethnicity information is based on self-identification as a certain ethnicity, while responses to the ethnicity question may also be refused (European Parliament, 2015[30]).
Annex Table 3.B.1. Roma population across the European Union
Official data and Council of Europe Estimates
Source: Council of Europe (2012), https://rm.coe.int/1680088ea9 for official numbers and estimates; World Bank (2010), World Population Estimates, https://datacatalog.worldbank.org/dataset/population-estimates-and-projections for Total Population.
Availability of reliable data on the situation of Roma in Bulgaria and the EU
The development of policies targeted at Roma is complicated by insufficient general statistics and outdated information on the number of Roma (with a data gap of more than 10 years between Censuses). As EC (2018, p. 8[31]) highlights, information about Roma in the EU is “still incomplete, with gaps in most Member States”. This lack of accurate data in European countries on Roma labour market situation in particular is an obstacle to understanding the barriers they are facing (Soler Penadés et al., 2016[32]). For example, short and medium-term labour market outcomes of Roma cannot be monitored in most European countries, as ethnicity information is not consistently collected in all household surveys. While ethnicity information is collected for the Survey of Income and Living Conditions (SILC) in Bulgaria, this information is not collected in labour force survey (LFS) data. Hungary is an exception is this respect, collecting ethnicity information in its quarterly LFS (EC, DG JUST, 2020[33]). Improving the evidence base on the situation of Roma with respect to their education, employment, health care and housing therefore has been an important element in the EU Framework for NRIS (Annex Box 3.B.1). A recent EU Council Recommendation highlights the importance of collecting data as necessary background for the design of measures that effectively improve the situation of the Roma population (EU Council, 2021[34]).
In Bulgaria, the lack of a working system for monitoring, evaluating and controlling the Bulgarian NRIS 2012‑20 is defined as a major drawback for the objective assessment of policies and measures concerning Roma (labour market) integration in Bulgaria (EC, DG JUST, 2019[35]). Different projects in Bulgaria during the last NRIS period focussed on improving the data situation on Roma. This included a project to track the development of the Bulgarian NRIS, funded under the Operational Programme Human Resources Development.3 Moreover, more use is being made of the data collected in EU-wide surveys such as SILC and the European Values Study EVS)4 (EC, DG JUST, 2018[36]). The local labour offices also play an important role and have also started to collect ethnic data from registered unemployed, but they face “frequent responses of refusal to identify by Roma for fear of discrimination” (EC, DG JUST, 2018, p. 19[36]). Finally, the Bulgarian National Statistical Institute (NSI) in co‑operation with EU-FRA are currently developing innovative methods for data collection for the provision of indicators on vulnerable groups, including Roma, as part of the project “Novel Approaches to Generating Data on hard-to-reach populations at risk of violation of their rights”.5
The first EU-wide comprehensive approach to address the evidence gap on Roma and other minorities was the European Union Minorities and Discrimination Survey (EU-MIDIS) conducted by the European Union Agency for Fundamental Rights (EU-FRA) in 2008.6 A second wave was conducted in 2015‑16.7 EU-MIDIS surveys different ethnic minority and immigrant groups’ experiences of discrimination and victimisation in everyday life across the EU. In nine EU Member States8 Roma were interviewed as part of EU-MIDIS II and some results are presented in this Annex.
Annex Box 3.B.1. European strategies for Roma integration
In 2021, the European Commission (EC) called upon member states to develop national strategies for Roma integration, with a focus on education, employment, health care and housing. Most importantly it called on Member States ensuring that Roma are not discriminated against and that Member States actions ensure that the cycle of inter-generational poverty will be broken. Following this, each country produced a National Roma Integration Strategy (NRIS) up to year 2020, which the Council of the European Union agreed upon in a Recommendation on effective Roma integration measures in EU countries in 2013. The Commission produced annual reports (until 2020) assessing the NRIS, using information from each country, as well as from civil society, international organisations and the EU Fundamental Rights Agency.
Following the end of this period, a new framework was developed in 2020 and adopted by the Council of the European Union in March 2021. The new EU Roma strategic framework sets a number of targets up until 2030, again in the areas education, employment, housing and health, as well as the three horizontal objectives of promoting effective equality, socio‑economic inclusion and meaningful participation of Roma. Different to the previous framework, the EC now proposed quantitative headline targets to monitor achievement towards these objectives.
Source: European Commission (2020), “EU Framework for National Roma Integration Strategies up to 2020”, https://ec.europa.eu/info/policies/justice-and-fundamental-rights/combatting-discrimination/roma-eu/roma-equality-inclusion-and-participation-eu/eu-roma-national-integration-strategies-2020_en and EC DG JUST (2020), “EU Roma strategic framework for equality, inclusion and participation for 2020 – 2030”, https://ec.europa.eu/info/policies/justice-and-fundamental-rights/combatting-discrimination/roma-eu/roma-equality-inclusion-and-participation-eu_en.
Barriers to Labour Market Integration of Roma
Roma integration in Bulgaria requires solving the problem of the high unemployment and inactivity rates among Roma people. According to a Bulgarian Academy of Sciences (BAS), despite the economic recovery in Bulgaria and the increased labour demand in the years following the GFC, mass unemployment and the Roma drop-out from the formal labour market (and related poverty) remained unchanged (Tomova and Stoychev, 2017[37]). More recent reports highlight some positive developments though, relating both to better educational outcomes and also higher employment of the Roma population in the period 2011‑19 (Angelova et al., 2020[38]). Nevertheless, in Bulgaria as across the EU, Roma are employed mainly in low-paid jobs with temporary contracts and no health insurance9 and many of them have only insecure or informal employment (EC, DG JUST, 2019[35]) and furthermore face discrimination in the labour market (Arbex et al., 2013[39]). In Bulgaria, 51% of working Roma lack health.
Intra-EU mobile Roma, who left their country of origin to work in other EU member states are particularly vulnerable, as they face an increased risk of becoming victims of exclusion from the labour market and public services, exploitation and discrimination due to tight local labour markets and lack of legal frameworks and support and may work in precarious conditions, for low wages and without insurance (EC, DG JUST, 2020, p. 12[33]). This section therefore discusses education, housing and spatial segregation and health as barriers to Roma labour market integration, while the next subsection considers the issue of discrimination.
Access to education as a key prerequisite for Roma labour market integration
Educational outcomes of Roma in Bulgaria and other EU countries lack behind those of other ethnic groups. The gap already manifests itself in early childhood education and care (ECEC) and continues in subsequent levels of the education system. Civil society organisations point out the lack of access to ECEC, including nurseries, kindergartens and other early childcare services and institutions in Bulgaria and other Central and Eastern European countries (CEEC). Reasons identified are: i) unavailability of ECEC services in rural settlements, ii) poor transportation, iii) lack of vacant places in educational facilities in big cities and iv) financial difficulties experienced by Roma parents (EC, DG JUST, 2018[36]). Results from the EU-MIDIS II survey show that only 66% of Roma children aged 4‑6 in Bulgaria attended kindergarten in 2016 (EU-FRA, 2017[40]). In addition, a factor seriously undermining the quality of ECEC provision to Roma children is the insufficient training of teachers and other staff in early childcare institutions (Pamporov et al., 2020[41]). Hence, the low quality of education and care in the kindergartens attended by Roma children, combined with poor interaction with Roma parents, is barrier to Roma integration in education (EC, DG JUST, 2019[35]).
While school attendance for children of compulsory school age has improved over the past decade, segregated education remains an issue in Bulgaria. Data from the last national census of 2011 show that one in four Roma children aged 7‑15 have never attended school, compared to 5.6% of children from the Bulgarian ethnic group (World Bank, 2015[42]). Findings from the EU-MIDIS I and II survey, however, suggest that school attendance at compulsory schooling age improved more recently, increasing from 86% to 90% over the period 2011‑16, while early leaving from education and training dropped from 87% to 68% (EC, 2018[31]). Nevertheless, about half of Roma students in Bulgaria are enrolled in schools located in neighbourhoods with a predominant Roma population, which results in deepening their educational segregation (Dimitrov, Grigorova and Decheva, 2013[43]). What is more, children in smaller settlements grow up with an extreme lack of access to health, education and social services such as speech therapy, rehabilitation, medical treatment, etc. (Council of Ministers, 2020[44]). Consequently, Roma children lag significantly behind children of Bulgarian ethnicity in the educational attainment for the respective age group, the lag equalling three school years in the field of reading and two school years in the field of mathematics and science (Ministry of Finance, 2019[45]). The COVID‑19 crisis is likely to have further increased the educational gap between Roma and non-Roma children, “disproportionately affect[ing] marginalised and socially excluded Roma” (EU-FRA, 2020, p. 7[46]), also because they are more likely to be without access to the internet or IT equipment and cannot benefit from online distance‑learning measures during the COVID‑19 school closures (EU-FRA, 2020[46]).
According to some commentators, cultural and family environment may contribute to those outcomes, as education is not a “virtue” for marginalised Roma (Bogdanov and Angelov, 2006[47]). According to the same authors, poverty is another driver for early school leaving among Roma, as education imposes costs on the household, on the one hand (for transport, textbooks, etc.), and deprives it of income (from child labour), on the other. Marriage at a very young age is also still a common reason for Roma girls to drop out from school (UNICEF, 2016[48]). Furthermore, migration to other EU countries driven by unemployment and poverty in Bulgaria, are considered a reason for school drop-outs among the children of migrants who leave with their parents or experience reduced control over their attendance at school by the relatives with whom they are left to stay (Tomova and Stoychev, 2017[37]).
Consequently, educational barriers with respect to no education or only lower level of education are observed more often for Roma than other ethnicities. The EU-MIDIS II survey results show low education levels among adult Roma population. On average, for the nine surveyed Member States, barely 18% of adult Roma have completed upper secondary, vocational or post-secondary education, while they also tend to have low proficiency in the national language, mainly in reading and writing (EC, 2019c[49]). Data from the 2011 Census in Bulgaria suggests that 93% of Roma do not complete their secondary education, compared to about 30% for ethnic Bulgarians (Council of Ministers, 2020[44]). This is also reflected in the analysis carried out in this Chapter, showing that skills barriers for people of Roma ethnicity are more frequent than for other ethnicities (see Table 3.3 and Table 3.4) in Bulgaria, which is often seen as the main reasons limiting their chances for sustainable employment participation (Pamporov, Markova and Yordanova, 2020[29]).
Housing and spatial segregation as obstacles to Roma labour market integration
Poor housing conditions and spatial segregation are key factors that further aggravate exclusion and inactivity among Roma. EU-FRA identifies spatial segregation as a structural barrier for Roma labour market integration and refers to it as “a severe impediment for access to employment” (EU-FRA, 2014, p. 27[50]). Limited mobility creates additional barriers to employment among Roma, as many of them live in areas with little employment opportunities and limited public transport facilities (Wislock, 2017[51]). Willingness to move, however, is often low. Many Roma communities have a strong sense of spatial belonging to their areas, which are inhabited by many successive generations (Ilieva, 2019[52]).
Beyond spatial segregation, Roma also often live in extremely poor housing. While results from the EU‑MIDIS II survey suggest that nearly all Roma households have access to electricity, the situation with respect to tap water and a toilet or bathroom inside the house is much worse. Twenty-three percent of Roma in Bulgaria live without tap water and 44% – without a toilet inside their dwellings (EU-FRA, 2016b[53]). Regarding housing quality and surrounding environment, 33% of Roma in Bulgaria live in dwellings with a leaking roof, damp walls or other problems, while 27% feel that pollution and other environmental issues are a problem in the places where they live (EU-FRA, 2016b[53]). Moreover, Roma often face issues related to the legal ownership of their homes, which sometimes explains the lack of access to public utilities, such as water supply, sewage, etc., but may also pose a risk of demolition of their dwellings and eviction of the residents (Mihailova and Kachamov, 2017[54]).
Health barriers to Roma participation in employment
Health barriers are another obstacle to labour market integration by Roma in Bulgaria and other European countries. Results from the EU-MIDIS II survey show that in Bulgaria, the share of Roma facing long-term activity limitations is higher than the share of the general population. While the difference for men is relatively smaller (19% of Roma men reported long-term activity limitations compared to 16.4% of the general population), the gap is larger for women and a higher proportion of women report long-term health problems (25% of Roma women 19.7% of the general population), a pattern observed also in other EU countries (EU-FRA, 2016b[53]). These patterns may also be explained by a lack of access to health care for Roma, which is extremely pronounced in Bulgaria in comparison to many other EU countries. Across the EU, health care coverage through public and primary private health insurance is relatively universal, reaching 93% (Slovak Republic) up to full coverage.10 In response to the EU-MIDIS II survey, however, only 45% of Roma in Bulgaria stated that they were covered either by the national basic health insurance or additional health insurance schemes. This is the lowest Roma health insurance coverage rate among the nine surveyed countries, where – on average – three‑quarters of Roma are covered by health insurance, reaching 98% in Spain and 96% in Portugal (EU-FRA, 2016b[53]).
Discrimination as an obstacle to Roma integration
Across the EU, limited access to decent work for the Roma population is driven by both entry barriers to the mainstream society, as well as exit barriers from the traditional Roma community (Ciaian and Kancs, 2018[55]). Exit barriers are driven by a strong sense of community, as well as the issue of spatial segregation discussed before. Entry barriers determine to what extent the “mainstream society” is willing to accept Roma within its socio‑economic structures. In this context, discrimination against Roma, which prevails across all EU countries (Frazer and Marlier, 2011[56]), poses a major problem for social mobility and creates constraints for Roma in their integration into and interactions with the mainstream society (Ciaian and Kancs, 2018[55]). Discrimination occurs both in society more generally and, more specifically, employment, education, health and housing. While Roma are being denied employment on discriminatory grounds, they furthermore often face discrimination in the workplace once employed and are constrained from progressing upwards (Council of Europe, 2012[57]).
A main focus of the EU-MIDIS survey is discrimination individuals experience on the grounds of skin colour, ethnic origin, and religion or religious belief, both within the past five years and past 12 months. Across the nine member states were Roma were surveyed in 2015‑16, 41% of Roma felt discriminated against because of their Roma background at least once in the past five years and 26% indicated that the last incident of discrimination based took place in the past 12 months. While discrimination also prevails in Bulgaria, the level of discrimination in Bulgaria has dropped between the two MIDIS surveys (2005 and 2015‑16) and is lower than in any other of the eight countries surveyed in 2015‑16 (22% of Roma in Bulgaria felt discriminated in the past five years and 14% in the past 12 months). While the results are relatively better in Bulgaria, in comparison to those in the other countries, discrimination prevails and is difficult to counter, as it is often concealed (EC, DG JUST, 2019[35]).
Notes
← 1. Only for health type barriers do Roma appear to do better than other ethnic groups in Table 3.3. However the self-perceived fewer barriers to health should be interpreted with care, as it is known that health inequalities are often worse for Roma, with Roma people suffering lower life expectancy and higher rates of communicable and non-communicable disease than other groups (Parekh and Rose, 2011[58]).
← 2. For example, some unemployed persons may search for a job without registering with the PES, while some persons registered with the PES may state in the LFS that they are not seeking a job (and hence be classified as inactive).
← 3. Project 2014BG05M9OP001‑3.2015.001 „Development and introduction of a system for monitoring, evaluation and control for implementation of the National Roma Integration Strategy of the Republic of Bulgaria 2012‑20“. The NRIS implementation monitoring system is available at the following: web address: https://nrcpsystem.government.bg/SitePages/%D0%9D%D0%B0%D1%87%D0%B0%D0%BB%D0%BD%D0%B0%20%D1%81%D1%82%D1%80%D0%B0%D0%BD%D0%B8%D1%86%D0%B0.aspx.
← 4. EVS is a large‑scale, cross-national, repeated cross-sectional survey research programme on basic human values. It provides insights into the ideas, beliefs, preferences, attitudes, values and opinions of citizens all over Europe.
← 5. The project started in August 2018 and is funded by the European Economic Area (EEA) and Norway Grants. The project goal is to find approaches for drawing a more comprehensive picture on the situation of vulnerable groups in Bulgaria, as this will support the development of more effective policies for those groups of the population at national and local level. The key project activities include conducting: 1) a study of the existing good practices and methods for identification of vulnerable groups of the population and 2) a survey among 15 000 households across Bulgaria, the results of which will be used for the purposes of policy making in the field of social inclusion, development of target indicators for the EU Operational Programmes and monitoring the implementation of the UN Sustainable Development Goals in Bulgaria.
← 6. Using a random sampling approach, the survey interviewed 23 500 respondents across the 27 EU Member States in 2009 – including 3 500 Roma respondents in seven EU Member States and, for the purpose of comparison, additional 5 000 people from the majority population in 10 EU Member States.
← 7. The EU-MIDIS II survey aimed to assess progress made since the first survey in 2008 and was based on face‑to-face interviews across all 28 EU Member States. The survey was carried out between October 2015 and July 2016 and contained questions on perceived discrimination in different settings, such as employment, education, housing and health when using public or private services. More information about the survey is available at: https://fra.europa.eu/en/project/2015/second-european-union-minorities-and-discrimination-survey.
← 8. Bulgaria, the Czech Republic, Greece, Spain, Hungary, Portugal, Romania and the Slovak Republic.
← 9. In Bulgaria, 51% of working Roma lack health insurance, far higher than in other EU countries. In the 11 EU countries covered in EU FRA (2014[50]), on average only 19% of working Roma stated that they did not have health insurance.