1. Assessment and recommendations

Kristine Langenbucher
Marius Lüske
Judd Ormsby
Theodora Xenogiani

Bulgaria’s labour market performed relatively well prior to the COVID-19 pandemic. In 2019, Bulgaria’s unemployment rate was among the lowest in the EU, at 4% among 15-64 year-olds, against 7% on average in the EU, and labour force participation had risen strongly within just a few years. From a rate of just 66% in 2011, it rose to its highest level in decades in 2019, at 73%. Employment gains had been strongest in urban areas and among older people, while rural areas and young people had benefitted less.

The outbreak of the pandemic abruptly interrupted these labour market improvements, with particularly strong effects on young people and sectors most susceptible to activity losses because of social distancing, such as tourism and other parts of the hospitality industry. In May 2020, only a few weeks after the state of emergency had been declared, the number of people holding an employment contract had shrunk by 5% compared to one year earlier, and by up to 40% in the sectors that were most severely affected. In the second quarter of 2020, the number of 15-24 year-olds who had a job was down by 17% year-over-year. As a consequence, the number of jobseekers registered with the National Employment Agency (NEA) escalated, rising from 205 000 in February 2020 to 295 000 in May 2020.

Bulgaria’s labour market stabilised relatively quickly after the initial shock in spring 2020, and has improved since then, supported by rapidly implemented policy responses, including the new “60/40”wage subsidy scheme. The unemployment rate stabilised at about 5% in the second half of 2020, and the number of registered job seekers fell back to its 2019 level in summer 2021. Nevertheless, employment challenges persist. In the second quarter of 2021, the employment rate was still three percentages points lower than at the same period in 2019, at 68% against 71%. Especially in the accommodation and food service industry, the number of employees remains low, in spring 2021 it was still about one-quarter lower than in 2019. These challenges are likely to persist until the sanitary situation fully stabilises and, as of today, the long-run effects of the crisis on employment and activity remain uncertain.

Bulgaria faces structural challenges weighing on the long-term prospects of its labour market. Unfavourable demographic dynamics, combining rapid population ageing and a stark population decline due to low fertility rates and emigration, are one of the most serious challenges confronting the country. According to recent projections, Bulgaria will lose close to one-third of its working-age population by 2050, making it the fastest shrinking country in the world. By then, one-third of Bulgaria’s adults will be aged 65 or older, which will most likely result in labour market shortages and put stress on Bulgaria’s social security system. Therefore, efforts to design a policy mix to increase employment rates as much as possible will be crucial to alleviate the consequences of a quickly declining labour force. Such a policy mix requires sound and effective active labour market policies which should be designed based on evidence and take account of the experiences of other countries in order to be efficient.

A second structural challenge the Bulgarian labour market faces is its high degree of inequality. Employment rates of highly educated people are among the highest in the EU, at 89% of 15-64 year-olds in 2019, while rates among the low-educated are considerably lower, at only 38%. Similarly, both employment rates and wage levels are substantially higher in economically prosperous regions than in poor and remote parts of the country and employment outcomes vary significantly across ethnic groups. For example, 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.

Despite the labour market improvements Bulgaria has achieved over the last years, there are about 900 000 working-age adults who are unemployed or inactive (excluding students). Among this out-of-work population, 15% are unemployed (i.e. available for employment and actively looking for a job) and the remaining 85% are inactive (i.e. outside the labour force). While inactivity and unemployment span across many different groups, some population groups are particularly exposed and face a high risk of becoming or remaining inactive or unemployed. These groups, which are overlapping,1 require specific attention from the NEA, not only because they account for the bulk of the out-of-work population, but also because many of them face substantial barriers to employment, and cannot overcome them without support.

  • Youth not in employment education or training – NEETs (~170 000 in 2019): Bulgaria has one of the highest NEET rates in the EU, at 17%, against 13% in the EU, highlighting that the need for further efforts in this area. Analysis of SILC data shows about 42% of youth NEETs are Roma. Activating NEETs is particularly important as failing to acquire human capital or skills can lead to long lasting scaring effects.

  • People out of work from ethnic minorities2 (~360 000 in 2019): Roma suffer much higher rates of joblessness than other ethnic groups and face many barriers and challenges to labour market participation. The Turkish community, too, is confronted with comparatively high levels of inactivity, in particular among women, although to a lesser extent than the Roma community. In total, there could be around 240 000 working-age Roma who neither work nor study, and less than one-fifth of them is in contact with the NEA. For the Turkish minority, estimates point to about 123 000 working-age adults who neither work nor study.

  • People out of work due to care and family commitments (~290 000 in 2019): Care and family responsibilities are the most common reason for inactivity in Bulgaria, except for studying. This barrier particularly effects women who account for essentially all of the people who report care and family commitments as their primary barrier for not seeking employment.

  • People out of work for illness and disability (~190 000 in 2019): Disability is a common reason for inactivity and employment rates for people with health problems are among the lowest in the EU.

  • Older working age people 55-64 who are out-of-work (~340 000 in 2019): While employment rates for older people are above the EU average and have strongly increased over the last years, they remain far below the level of prime-age adults. Given the large number of 55-64 year-olds who do not work, and the growing share of the elderly in the Bulgarian population, activating older people would have the potential to help mitigate the effects of a shrinking and ageing population and address labour shortages.

The most common barriers to labour market participation for inactive and unemployed people in Bulgaria are skills barriers (e.g. limited education or work experience), family-related barriers (e.g. care responsibilities), health impediments, and geographic barriers (e.g. living in remote settlements without a vehicle). Some labour market obstacles concern certain groups of inactive and unemployed much more often than others. For example, 42% of out-of-work Roma face a geographic barrier compared to 21% of the overall inactive population. Close to half of inactive or unemployed 55-64 year-olds have a significant and long-lasting health impediment, compared to one-third of all inactive and one-tenth of all unemployed. In many cases, inactive and unemployed people are confronted with several employment barriers simultaneously. In total, 75% of the inactive and 61% of the unemployed face at least two significant barriers to employment, against 18% of the employed.

In order to help different out-of-work groups overcome their barriers, activation solutions have to be joined up to deal with multiple barriers and tailored to address individual needs. In particular, in some cases, support by the NEA has to be combined with additional services provided by other institutions addressing e.g. health obstacles or social difficulties, in order to be effective.

Prolonged periods spent by young people out of the labour market can have scarring effects making labour market integration more difficult the longer such periods last. In addition, young people were hit hardest by the labour market repercussions of the COVID-19 pandemic. In the second quarter of 2020, the number of employed 15-24 year-olds had plummeted by 17% compared to one year earlier, while the effect on workers aged 25 and older had been much smaller, dropping by only 5%. Consequently NEET rates increased again and stood at 18% in 2020, against a European average of 14%.

To reduce high NEET rates, both the NEA and municipalities employ activators/mediators who specifically focus on reaching out and activating NEETs. In 2019, close to 100 youth mediators were employed throughout Bulgaria. In addition, the NEA set up Family Labour Consultations, aiming to provide comprehensive services to all members of families confronted with unemployment and inactivity, with the potential to prevent teenagers and young adults from these families becoming NEETs. However, less than 15% of NEETs aged under 25 register with the NEA, limiting the support the NEA can provide. Therefore, the NEA should step up efforts to reach more young people who neither work nor study. For example, additional and more systematic partnerships between the NEA, schools and NGOs could contribute to identifying and supporting NEETs and school-drop outs. Such co-operation should be reinforced both in rural and urban areas, and combine approaches to prevent young people from becoming inactive (e.g. information campaigns in schools) and reengage young people who are inactive and not connected with public services. In addition, the NEA should carry out a rigorous evaluation and cost-benefit analysis to assess if increasing the number of youth activators and mediators could support better outreach to youth and their re-integration into employment or education and training.

There are wide disparities in labour market outcomes between different ethnic groups in Bulgaria. While ethnicity is self-identified in labour force surveys and there are difficulties in accurately capturing ethnicity, the available data show that Roma suffer rates of joblessness more than double that of ethnic Bulgarians. Among the working-age population, about 49% of Roma men and 69% of Roma women were not in employment in 2019 compared to around 24% and 29% for male and female ethnic Bulgarians. Roma out-of-work often face complex and multiple barriers to employment, including skills, health, family-related, and geographic distance barriers. Beyond these barriers, Roma also face further obstacles hindering the labour market participation, including discrimination based on their ethnicity, high rates of poverty, and low quality and overcrowded housing in segregated Roma neighbourhoods. Barriers to labour market participation already manifest themselves at an early age: Roma children benefit less from early childhood education and have lower attendance rates at all levels of school and often attend segregated schools. Consequently, Roma children lag significantly behind children of Bulgarian ethnicity in educational attainment.

Against this challenging background, the NEA’s use of Roma mediators to support outreach to Roma is important to enrol more Roma with employment services. However, the data available suggests that only one-fifth of out-of-work Roma are registered with the NEA. In 2019, 78 Roma mediators worked for the NEA, which is likely to be insufficient considering that there could be about 200 000 unemployed or inactive Roma of working age who have no contact with the NEA. The NEA should carry out an in-depth assessment of the impact of the work of Roma mediators and evaluate to what extent the number of Roma mediators should be increased.

Beyond outreach, active labour market policies (ALMPs) can and should be used to help Roma in overcoming the barriers they face. However, registered unemployed of Roma origin are referred to ALMPs less frequently than jobseekers of other ethnicities. Whereas 20% of non-Roma jobseekers were referred to labour market programmes in 2020, this share was only 8% among Roma jobseekers. What is more, Roma are almost always referred to direct job creation programmes, which international research shows are, when deployed as stand-alone measures, less effective in securing sustained employment in the open market than other types of ALMPs. Seventy-six percent of ALMP places for Roma in 2020 were in direct job creation, while only 56% of non-Roma were referred to direct job creations programmes. This high proportion could be linked to the significant employment barriers Roma people face, including skills-related and family-related obstacles.

One of the largest groups of Bulgaria’s inactive population, are individuals who cite care, family or personal responsibilities as their primary reason for not currently seeking work. It is the most common reason for not seeking work among the non-student-inactive population, representing 289 000 people or 37% of the non-student-inactive-working-age population. People citing care responsibilities for not working are almost entirely women which suggests that men could play a greater role in sharing care burdens and raising children. Bulgaria provides some of the longest leave available to mothers in the EU (more than two years in total including both maternity leave and the less generous paid parental leave). However, like most other EU countries, Bulgaria provides much less leave for fathers (about two weeks). To encourage fathers to take on a greater role in caring for children, some countries, such as Korea, Sweden, and Iceland have increased the amount of father-specific leave, sometimes on a use-it-or-lose-it basis. Other policies which can also bring cultural changes may help parents too. For example, wider access to affordable day care and the option to work part-time can help parents balance their care and work responsibilities. However, the number of day care nurseries is limited, especially in rural settlements, and Bulgaria has a full-time work culture with less than 2% of the employed in part-time roles, the lowest share in the EU. In addition, salaries of part-time workers in Bulgaria are often too low to make a decent living.

Many women out-of-work for family-related reasons often also face other barriers to labour market participation – particularly skills and experience barriers including a lack of recent work experience (while raising children), low levels of education or a history of working in low-skilled occupations.

As a first step in supporting jobseekers with children and other care barriers, Family Labour Consultants have been recently introduced at the Public Employment Service (PES) to help provide comprehensive support for families. These counsellors are regular NEA staff (counsellors, mediators, psychologists and caseworkers) that spend some of their time specialising in the work with jobseekers and their families (630 experts provided such services in 2019). Family Labour Consultants not only support family members who are jobseekers in addressing their barriers, but provide support and counselling to the entire family, being mindful of the wider family context. Family Labour Consultants thereby can also identify and activate family members who are not yet registered with the NEA. As Family Labour Consultants have just recently been introduced, it is too early to evaluate their impact. Other initiatives to foster employment among parents with care obligations include the programme “Parents in Employment”, which provides job mediation specifically to jobseekers with children and grants access to childcare, and the project “Children’s corners”, which trains jobseekers to acquire childcare skills.

Other countries use a range of measures that can support labour market participation of jobseekers with care barriers on a wider basis, and good results have been achieved with programmes offering counselling support. Beyond counselling, the NEA may need to intensify its work with employers to seek family-friendly work places and possible part-time opportunities. International evidence also highlights the importance of wider access to childcare and as well as targeted initiatives to specific groups such as increased employment service support for example for lone parents.

Employment rates of older people (aged 55-64) have increased very substantially over the last decade, rising from 45% in 2011 to 64% in 2020. This increase is closely linked to the gradual rise of the official retirement age, which currently stands at just over 64 years for men and just under 62 years for women, up from 63 years and 60 years in 2015, respectively. Nevertheless, the Bulgarian labour force is predicted to shrink swiftly, and further increases in the employment rate of older workers will be necessary to alleviate labour shortages. While the employment rate of 55-64 year-olds is now higher than on average in the EU, it is still at least 10 percentage points lower than in Europe’s leading countries,.

There is scope for the NEA to play a bigger role in the activation of older people, instead of just providing a bridge to retirement as some ALMPs do. While many inactive and unemployed who are in their early 50s register with the NEA as jobseekers, NEA registration gets substantially less common as people approach the official retirement age. Among 60-64 year-old men who do not work, less than one in every six registers with the NEA, although they have not reached the official retirement age yet. This pattern is partly linked to financial disincentives to work in the retirement system, which reduce the willingness of older people to take up employment. Besides removing such disincentives, reaching out to more older inactive or unemployed people and introducing ALMPs that are specifically designed for older jobseekers, in addition to existing wage subsidies for older workers, can be one part of a wider strategy to encourage longer working lives. For example, networking groups of older jobseekers and media campaigns to improve the image of working at older ages could prove successful, including to increase the motivation of older people to take up employment. As a large share of older inactive and unemployed people face health barriers to employment, providing comprehensive services, in particular employment services in combination with health-related services, would encourage more older people to extend their working lives or take up a new job. In addition, employers are crucial in strengthening the role of older workers and should be supported to e.g. hire older jobseekers and provide retrain opportunities to older staff members permitting them to keep their competences up-to-date.

Unemployment insurance in Bulgaria is a contributory benefit that provides income support to those who have paid contributions for at least 12 of the last 18 months, with benefit duration and amounts tied to previous employment history. Unemployed people who are not eligible or who have exhausted their unemployment insurance can apply for the means-tested social assistance.

Unemployment benefit replacement rates (full rates) are among the highest in the EU and provide good protection against a sudden loss of work income for those who can claim non-reduced rates. However, not all unemployed are eligible or are only eligible at a reduced rate (for example because their contribution records are too short, because they are returning claimants or because they quit their work on their own initiative). In these cases, the level of protection is much lower. The share of unemployed claiming unemployment benefits is lower than on average in the EU, and close to 30% of unemployment recipients only receive the minimum rate, amounting to about half of the median unemployment benefit. Among some groups of unemployed beneficiaries, the share receiving only the minimum rate is higher: 39% among the low-educated and 47% among Roma.

Jobseekers who are not eligible for unemployment benefits or have exhausted their eligibility rely on social assistance as a last resort. This is in general the case for people with no recent formal employment histories or scattered employment records. However, social assistance benefits are very low and entitlement criteria are strict, contributing to a low take-up of social assistance. Bulgaria’s minimum income benefits are among the lowest in the EU, at less than one-fifth of the median disposable income. In order to qualify, recipient’s income and assets, assessed at the family level, must be very low and beneficiaries are required to perform regular community service work. Furthermore, recipients have to register with the NEA as unemployed for at least six months before they are eligible to receive social assistance. This requirement means that low-income individuals face a long waiting period before they can claim social assistance.

The low level of social assistance and its strict entitlement criteria contribute to low take-up and risk pushing more people into the informal economy, which accounts for about 16% of total employment. Furthermore, the low take-up of social assistance limits its ability to prevent poverty. In 2019, about one-third of the Bulgarian population was at risk of poverty or social exclusion, which was more than in any other EU country.

Both unemployment benefits and social assistance benefits are only paid out to people who are registered as jobseekers, thereby increasing the NEA’s ability to establish contact with jobseekers. However, these incentives only concern people who are eligible for benefits and choose to claim them. Therefore, Bulgaria should keep registration with the NEA as a pre-condition to receive unemployment or social assistance benefits to favour NEA registrations, but reduce barriers to claim social assistance for eligible groups. For example, the 6-month waiting period before benefits are granted, should be reduced or eliminated. The benefits of higher social assistance take-up would be two-fold, contributing to mitigating poverty risks and facilitating NEA outreach to inactive people. While higher social assistance payments could, on their own, reduce incentives to find work, such risks are low in Bulgaria relative to other countries as current levels of social assistance are already very low in an international comparison. Such disincentive effects can also be offset through requiring and monitoring job-search behaviours and the increased incentive a higher social assistance level could provide to register with the employment agency and comply with such requirements.

Bulgaria imposes eligibility requirements both on unemployment benefit recipients and social assistance beneficiaries, aiming to favour active job search. These include requirements to be available for suitable work, participate in ALMPs and provide evidence of job-search. Failure to fulfil these obligations can lead to sanctions on benefit receipt.

In an international comparison, Bulgaria is relatively lenient on availability and job-search requirements. Jobseekers are allowed to decline jobs that do not match their education, qualifications, or profession and experience during the first 12 months of registration. After this period, requirements are stricter, but jobseekers can still refuse a job if it is too far from their home or incompatible with their health. In contrast to many other countries, which require social assistance beneficiaries to be available for a wider range of jobs, Bulgaria affords clients receiving social assistance the same protections and ability to refuse jobs that do not match their education and qualifications as those clients on unemployment insurance.

About 11% of the December 2019 stock of participants left the register via a sanction by February 2021. While Bulgaria is relatively lenient on availability and job-search requirements, the sanctions it imposes on failing to follow these rules are severe. Rejecting a suitable job or ALMP results typically in a six-month sanction on benefit receipt as well as loss of access to the NEA’s services for this period. Some countries, e.g. Luxembourg, Latvia, and Greece, still allow access to PES services for sanctioned clients. This option potentially preserves the effect sanctions have on incentives to search for work while helping those sanctioned to find employment faster.

The NEA is the central organisation reaching out to unemployed and inactive people. In addition, further institutions also carry out or contribute to outreach activities, including the Social Assistance Agency (SAA), municipalities, NGOs and schools. The NEA has set up a number of initiatives aiming to broaden its contact with people in need of support. For instance, it created mobile labour offices for inactive people in remote areas, set up centres for employment and social assistance (CESA) in co-operation with the SSA and hired activators and mediators focusing on inactive youth and Roma.

Despite the NEA’s efforts, the share of unemployed or inactive who register with the NEA is low compared to other countries. Only about 22% of inactive or unemployed 25-64 year-olds were in contact with the NEA, against an EU average of 35% of jobseekers who were in contact with the PES in 2019. In total, there could be around 700 000 working-age adults who neither work nor study, but have no contact with the NEA. Not all, but many of them would benefit from support to find employment.

While the NEA has stepped up efforts to intensify its outreach to groups further from the labour market, the share of inactive or unemployed registering with the NEA is particularly low among those groups, in particular young NEETs and Roma. For example, only about 12% of 15-24 year-old NEETs registered with the NEA in 2019. This compares to an EU average of 47% for youth registering with the PES. The share is similarly low among inactive and unemployed Roma, at 13%.

Further efforts are needed to identify and establish a contact with people in need of support. Good co-ordination and co-operation between the NEA and other stakeholders is key to strengthening outreach to people who are currently off the radar. For instance, the NEA has established an automated data exchange via the Inter-Institutional Registry Information Exchange System (RegiX) with the SAA in order to identify inactive people. Similar automated data exchanges could be set up between the NEA and other public institutions to facilitate outreach. In addition, stronger and more regular co-operation with organisations close to people who are inactive or risk becoming inactive, in particular certain NGOs and schools, would contribute to getting hold of people in need of support.

An important element in the NEA’s activation strategy, is supporting jobseekers to find employment through individualised job-assistance services and counselling. However, the caseload of caseworkers is high and limits the time they can spend with each client, making an intensive provision of services more challenging.

As studies from other countries show, lower jobseeker-to-caseworker ratios, can speed transitions to work by allowing for more intensive counselling to support job-search, increased monitoring of job-seekers, and increased collection of job-vacancies. Cost-benefit analysis from these studies shows that lower job-seeker to caseworker ratios can not only lead to faster unemployment-to-work transitions but also net savings for the government, i.e. additional expenditures on PES staff are more than offset by reduced benefit payments. Hence, while Bulgaria would need an initial up-front investment to increase caseworker numbers, providing these additional resources are used effectively, the up-front investment could lead to long term net-financial gains for the government, in addition to supporting jobseekers, through an “invest-to-save” approach.

Groups that are furthest from labour market often face very specific and sometimes multiple barriers to employment. Hence those furthest from the labour market need individual tailored support which requires more intensive counselling. Such intensive and early PES counselling can identify specific barriers faced by these groups sooner and offer further services to alleviate these barriers including specialised counselling services (e.g. psychologic counselling), training to improve employability, and potentially employment subsidies to support clients into work. Some clients also need in-work follow up, even after finding employment.

However, with its limited resources it is difficult for the NEA to meet the needs of every jobseeker, especially intensive support for those with the bigger barriers to labour market integration. Indeed, even prior to the pandemic, NEA counsellors met clients furthest from the labour market less frequently than more job-ready clients.

The pandemic has made the situation for jobseekers furthest from the labour market even more challenging. It has increased workloads for counsellors, which makes it even more difficult to provide intensive counselling to those far from the labour market. At the same time, the crisis has often hit groups furthest from the labour market harder, with increased competition for jobs.

Bulgaria should explore different options to ensure it can provide adequate levels of service to all its clients, acknowledging that the intensity and types of services may vary across jobseekers. Options include:

  • Increase the number of PES counsellors: Evidence from a number of other countries suggests that hiring more caseworkers could result in net-savings for government through increasing the speed and number of job-transitions.

  • Contracting-out employment services: Rather than hiring more PES staff, Bulgaria could consider contracting out some services to private providers either through tendering procedures or voucher systems.

  • Reallocate resources towards those who need them most: Irrespective of whether additional resources are available to expand the support for jobseekers, it is important to review whether existing resources are allocated in the most efficient manner. Priority should be given to those who benefit the most and those who are most in need of support. Such allocation decisions could be assisted with a more sophisticated jobseeker profiling tool (e.g. a statistical profiling tool that segments jobseekers based on their distance to the labour market for instance, their predicted unemployment duration). Reallocation of support should ensure an optimal match between the needs of jobseekers and the support they receive.

  • Make greater use of digital tools: Greater use of digital tools could free up resources and enable the NEA to do more with less, as discussed next.

The number of registered unemployed could increase even further if Bulgaria succeeds in reaching out to a higher share of the inactive population. This further underscores the importance of considering the above options to meet demand.

In Bulgaria, registration and employment services via digital channels currently play only a secondary role in engaging with jobseekers. There is scope to make more use of available technology to increase the efficiency of service provision and free up staff time for harder-to-place clients. This could be achieved through introducing a new profiling tool, developing and using other digital tools and services, and modernising the online vacancy database (“e-labour office”) to improve the collection and advertisement of job-vacancies.

While the NEA uses a profiling tool to segment jobseekers into sub-categories depending on their distance to the labour market, the IT tool is about one decade old and does not have the same level of sophistication as profiling tools used in other countries. In addition, many caseworkers do not follow the recommendations of the tool, limiting its utility. The NEA should either update the IT tool or replace it with a more modern statistical profiling tool to improve its performance and streamline the profiling of jobseekers. Counsellors should be trained in the use of this new or updated tool and should be encouraged to use it widely.

Similarly, the NEA has introduced some digital services (e.g. online registration), but has not adopted a “digital first” approach as some other countries have. Some countries operate employment services largely online and apply the results of their profiling tools to target online services and online training to jobseekers. Profiling information gathered before a first personal interaction can make the counselling process leaner and support PES counsellors in tailoring services to jobseekers’ needs. Therefore, the NEA should strengthen the use of digital services, in particular in its interactions with jobseekers who have close ties to the labour market and possess sufficient digital skills, while ensuring that no jobseekers are pressured to use digital channels. The use of digital interaction with groups of jobseekers that can largely “self-manage” their unemployment spell through e-services frees up time for more intensive and costly face-to-face services for jobseekers with larger barriers to employment.

Further, the NEA should modernise its online vacancy database, improving functionality such as search features and allowing for direct communication between jobseekers and employers without requiring NEA counsellors to always act as intermediaries – a process that adds frictions and uses up valuable NEA resources. The PES in Flanders, VDAB, has developed particularly advanced tools to help jobseekers and counsellors quickly search the vacancy database for appropriate roles. Several PES (including Austria and the Netherlands) also use digital tools to automatically collect employer vacancies – potentially increasing the total number of vacancies held by the PES and through automation, reducing the staff time needed to register vacancies.

Evidence shows that spending on ALMPs can help reduce unemployment and long-term unemployment, if such programmes are well designed and targeted. Especially for people with longer spells of unemployment, referrals to ALMPs can help improve employment prospects and keep up work habits through the regular participation in programmes.

EU countries, on average, spent 0.39% of GDP on ALMPs (EU labour market policy categories 2-7), including labour market training, employment incentives, supported employment and rehabilitation, direct job creation and start-up incentives. In Bulgaria, expenditure was just 0.16% of GDP in 2019, which may be too low to support more individuals out of work back into the labour market. Against the background of overall low expenditure on ALMPs, Bulgaria has a surprisingly high number of individual programmes, many of which have a small budgets and few participants. This creates unnecessary administrative costs, reduces transparency for all stakeholders involved – jobseekers, employers, PES staff – and also limits the options for programme evaluation and identifying what works for whom. The programme mix should be consolidated going forward reducing the overall number of programmes through merging programmes on the one hand and ceasing inefficient programmes on the other hand.

Bulgaria’s programme mix consists of training programmes, employment incentives, direct job creation schemes, and mostly small start-up incentive programmes. In 2019, two-thirds of the total ALMP spending (excluding employment services and administration) was invested into direct job creation measures in Bulgaria. In contrast, EU countries only spent an average of 11% of their ALMP expenditure in direct job creation. According to international evidence, the effectiveness of direct job creation programmes in bringing participants back to open market jobs is questionable. A number of OECD countries – including Denmark, Estonia, Israel, Norway, and Switzerland – therefore do not use these programmes at all anymore and many other countries have shifted spending from direct job creation towards more effective ALMPs, such as training and employment incentives over the past decade.

Bulgaria should reconsider the large emphasis placed on direct job creation programmes going forward and reserve places on such programmes strictly to long-term unemployed who have no prospect of integrating in the primary labour market in the short term to medium term. Even in these cases there should be wider support made available (e.g. training and mentoring) to help these jobseekers move to the primary labour market. For young people that are currently placed in such programmes, priority should be given to finding employment (including internships and apprenticeships) in the primary labour market, if necessary, with the support of wage subsidies.

Labour market training programmes may improve the skills of those with lower educational attainment, and reverse human capital depreciation that results from longer unemployment spells. However, just 8% of ALMP expenditure was spent on training in Bulgaria in 2019, against an average of 40% in the EU, suggesting that there are limited opportunities to offer training to jobseekers. Furthermore, some of the expenditure is spent on employed people instead. Yet, international evidence suggests that additional expenditure in training programmes can produce positive outcomes, especially in the medium to long run and should, hence, be part of an activation strategy that supports more inclusive and resilient labour markets. While in Bulgaria there are a limited number of places overall, the range of training programmes cover many different types and levels of training, but with few places for each type. Hence, it is crucial not only to invest more into training programmes, but also to streamline the existing ones. It is also important to ensure that information on available training is clear and easily available, that jobseekers receive guidance to choose suitable training and that training meets employers’ needs and addresses labour market shortages. When large additional investments into training are not possible, the focus should be placed on increasing the number of places in shorter programmes, such as short vocational training, general and remedial training and internships, as well as training programmes tailored jointly with employers to support the high number of jobseekers with low levels of education and basic or obsolete skills in Bulgaria. A stronger focus on training activities in the future could result from commitments of the Bulgarian Government to invest more in education and training, as foreseen in the (provisional) Recovery and Resilience Plan.

Time-limited, well-designed and targeted employment incentives can be a cost-effective way to support jobseekers back into employment and strengthen their employability. As in the case of training programmes, merging different employment incentives programmes targeted at overlapping groups of registered unemployed could be considered for further streamlining Bulgaria’s ALMPs. Bulgaria should also reduce excessive red tape for employers hiring jobseekers with support of a wage subsidy. This concerns both cumbersome application procedures and extensive monthly reporting, as well as delays in the recruitment process when recruiting jobseekers for whom a wage subsidy can be claimed.

An important ALMP for youth is the Youth Employment Scheme, which combines training to support the upskilling of youth with private sector employment incentives supporting their hiring. Results from this programme should be compared with other ALMPs that combine training and public works (Training and Employment of Unemployed, Job Programme, and New Perspective Project). Different to the Youth Employment Scheme, the latter type of programmes do not integrate the unemployed in the primary labour market. Both types of programmes should be thoroughly evaluated and compared to decide whether more unemployed could be supported through wage subsidies into the primary labour market directly, instead of maintaining them in public works schemes.

Beyond subsidies for employers, more emphasis should be placed on mobility incentives for jobseekers. Such incentives are available in Bulgaria, but in 2019, only 134 unemployed benefitted from the measure. Support for commuting and relocation are in place in a number of OECD countries and could play a more important role in Bulgaria were geographic distance acts as a barrier to labour market integration for many jobseekers. Evaluations in other OECD countries have shown that labour mobility incentives tend to increase the regional radius for job-search and, hence, lead to higher employment probabilities and wages.

The review highlights the need for additional investments into ALMPs in Bulgaria to promote employment, prevent labour market exclusion, and prepare for future opportunities and challenges in a rapidly changing labour market. At the same time, Bulgaria has a tight ALMP budget and, as all OECD countries, faces pressures to provide good value for money and improve the efficient and effective use of public finances to ensure that ALMP expenditure is spent on programmes that provide the greatest possible economic and social return. This highlights the need to monitor and evaluate ALMPs regularly and rigorously terminate or adjust inefficient ones, while upscaling efficient ones.

Impact evaluations and client satisfaction surveys of ALMPs in Bulgaria have been carried out for programmes and measures running in 2011, 2015 and 2017, but only partially cover the suite of existing ALMPs. Programme participants rated most programmes positively with respect to improving their employability and increasing the likelihood of finding an employment. An assessment of which types of programmes work and for which types of jobseekers is, however, not possible on the basis of the existing evaluations. For the purpose of the evaluations, 32 different programmes have been lumped together, not enabling a separate assessment of each programme. It would be advisable to analyse strengths and weaknesses of each of the programmes in reaching the expected outcomes and to discuss results individually for each programme. It would also be useful to draw conclusions on what type of programmes (training, employment incentives, direct job creation programmes) work for which group and to include Human Resources Development Operational Programme (HRD OP) financed programmes in these reflections.

Going forward, carrying out thorough counterfactual impact evaluations of programmes will be key for policy makers in Bulgaria to know which programmes have a positive impact and which ones have no impact or even a negative one. Combined with cost-effectiveness or cost-benefit analyses, this can show how the right programme mix can support achieving net benefits for the government through ALMP investments. While existing evaluations for Bulgaria have largely relied on participant surveys, future evaluations could make use of Bulgaria’s rich administrative data. As administrative data is already collected, it often is cheaper to use, can cover a greater population, allows tracking of outcomes over long periods of time, and does not suffer from non-response and sample attrition. Bulgaria’s public authorities have registers that contain the most important information required for the comprehensive evaluation of ALMPs, including programme participation, employment outcomes, and information on barriers to labour market participation. The different registers, however, are currently not linked for research purposes. Bulgaria should therefore continue to drive the process and further invest in the capacity to linking administrative data across different registers owned by different public authorities for research purposes.


← 1. The five groups listed here overlap. Hence, the sum of the estimates across groups exceeds the total number of working-age adults who neither work nor study.

← 2. Estimates on ethnic minorities have to be interpreted with caution. Information on ethnicity is self-declared and can be difficult to capture. In Bulgaria, stakeholders report that some people who self-identify as Turkish are seen by others as Roma.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2022

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.