This chapter examines the key aspects of two active labour market policies (ALMPs) evaluated in this report: training programmes and employment incentives provided to unemployed individuals by the Slovenian public employment service (ESS). It offers an overview of the main features of these programmes and the characteristics of their participants. The chapter also details the extensive individual-level administrative data underpinning the empirical analysis, along with the econometric approach employed in the counterfactual impact evaluation of these measures in subsequent chapters. Beyond the commonly assessed outcomes in ALMP evaluations, such as employment probabilities, this chapter highlights additional outcomes of interest, most notably career progression. To analyse career progression, the chapter introduces the construction of an occupational index, derived from the observed wages of individuals based on detailed occupational codes.
Impact Evaluation of Wage Subsidies and Training for the Unemployed in Slovenia
Abstract
The preceding chapter noted that Slovenia’s spending on active labour market policies (ALMPs) is relatively modest: its ALMP spending is less than half of the OECD average. At the same time, while the overall unemployment rate is at record lows, the share of long-term unemployed out of total unemployment remains considerably above the OECD average and older workers, in particular, have disproportionately low employment rates. ALMPs are a key labour market policy instrument that can help address these challenges.
To what extent do Slovenia’s ALMPs succeed in placing jobseekers into high-quality, sustained employment – and which programmes are most effective for specific groups? Which aspects of these policies perform well, and which require improvement? To address such questions, policymakers often rely on key performance indicators – such as job placement rates and participant satisfaction – or feedback from PES staff and jobseekers. Both sources of information are valuable for evaluating the effectiveness of a policy. For instance, performance indicators can reveal which ALMPs are most likely to lead to employment after participation, track improvements over time, or monitor the real-time performance of specific training providers. Likewise, feedback from PES staff and clients provides a nuanced understanding of a programme’s strengths and weaknesses, as well as practical recommendations for enhancements. At the same time, however, such approaches cannot provide a rigorous answer to the crucial question of the precise impact of a policy – this requires accounting for what would have happened to individuals in the absence of the policy. This is the motivation for conducting counterfactual impact evaluations (CIEs), such as the one outlined in this chapter.
The CIE presented in this chapter focuses on two of Slovenia’s key ALMPs: training and employment incentives for the unemployed.1 The training programmes aim to address skill gaps among jobseekers through courses lasting several months, while the employment incentives provide subsidies for one year to help offset employers’ wage costs when hiring workers from specific target groups of jobseekers. The evaluation tracks outcomes continuously for up to four years from the start of the programmes, offering a long-term perspective on their effectiveness. The empirical analysis leverages rich and comprehensive data, enabling the examination of a wide range of outcomes while accounting for diverse characteristics of jobseekers. Several types of data underpin this evaluation, including unemployment register records, ALMP participation data, and employment and earnings information, ensuring a robust and detailed assessment of the policies’ impacts.
The time period during which programme participation is examined spans ALMP participants from January 2015 through December 2018 (unless specified otherwise). This period was selected to balance competing considerations: ensuring the programmes are recent enough to have parameters comparable to current ones, while also providing sufficiently long time horizons to evaluate longer-term outcomes. Data on outcomes such as employment is available up to December 2022, enabling the tracking of results for a minimum of four years after individuals begin a programme and at least three years after they complete it (the ALMPs examined have a maximum duration of one year).
The chapter opens with an overview of the ALMPs under analysis and the characteristics of the individuals and employers participating in them. It then provides a detailed account of the rich, individual-level administrative data that underpin the empirical analysis, as well as the econometric approach used in the CIE of these two measures in subsequent chapters of this review. The final sections outline the labour market outcomes assessed in the impact evaluation. While standard outcomes such as employment probabilities are included, the chapter also introduces additional measures of interest, particularly career progression. To evaluate career progression, it outlines the construction of an occupational index calculated on the basis of the observed wages of individuals by detailed occupational codes.
Training for the unemployed and employment incentives together account for roughly one half of Slovenia’s ALMP expenditures (excluding PES and administration costs) during the period when participation is examined in this evaluation, 2015‑18. Only one variant of the employment incentive programme, Employ.me, was implemented during this period, while training was implemented as several distinct programmes, each with their specific features in terms of their content, target groups, duration, and objectives. Nevertheless, they share enough key similarities to allow them to sensibly be grouped as one set of programmes for the purposes of the evaluation (although some of the different implementations of specific programmes are analysed separately as well).
From the perspective of the CIE, it is worth emphasising that broad sets of the unemployed are eligible for participating in training and employment incentives. Some programmes do have strict participant eligibility: for example, the employment incentives programme imposes minimum unemployment duration criteria, but these depend on several other participant characteristics. Furthermore, some training courses require individuals to have secondary education. However, taken as a whole, the programmes are not limited to specific groups. From the perspective of choosing an approach for identifying the programme effects, another important consideration is that counsellors from the public employment service (ESS) have considerable discretion in deciding whether to refer an individual to a specific measure. This fact informs the choice of the econometric procedure used, with comparisons of similar individuals made based on detailed information on their observed characteristics (for details on the methodology, see Section 3.5).
A key strength of Slovenia’s framework for delivering ALMPs lies in its responsiveness, particularly its ability to adapt target groups to reflect the evolving labour market landscape and the changing composition of jobseekers registered with the ESS. This reflects the institutional flexibility of the system. Both the ESS and the Ministry of Labour, Family, Social Affairs and Equal Opportunities (MoLFSA) possess considerable discretion in revising target groups, as these are defined in the ALMP catalogue, which is updated regularly, as well as in specific public calls for tenders. This flexibility enables the adjustment of essential programme parameters – such as participant eligibility criteria and subsidy amounts – to align with prevailing labour market conditions.2 For example, between 2017 and 2022, a total of 47 iterations of the ALMP catalogue were issued, averaging 8 versions per year. While some of these updates were relatively minor, they illustrate the system’s capacity for regular and targeted refinements.
3.2.1. Five training programmes are analysed in the impact evaluation
The system of training available to jobseekers in Slovenia includes various programmes tailored to their needs. MoLFSA plays a central role in designing these programmes, while the ESS oversees their practical implementation through its network of 58 local offices and 12 Career Centres across the country. The training offer is primarily funded through a combination of EU (via the European Social Fund, ESF, from 2015 until 2018 and the European Social Fund Plus, ESF+, from 2021 until 2027), national funding, and, to a lesser extent, local contributions. During the period analysed by the impact evaluation (January 2015 to December 2018), five types of training programmes were available to jobseekers, each characterised by specific features:
Preparation for national vocational qualifications (NVQ): In Slovenia, NVQ certificates attest an individual’s ability to perform the tasks required for a specific occupation or specific tasks within that occupation. These certificates can be obtained in various fields, such as bartending, dental assistance, and bricklaying. Typically, candidates compile a personal portfolio containing work certificates, letters of reference and records of participation in training programmes providing evidence of their work experience in the relevant occupation. A commission then reviews and assesses the portfolio to determine whether the candidate meets all the requirements for obtaining an NVQ certificate (Institute of RS for VET, 2024[1]). These programmes support the validation of prior learning, allowing professionals who have acquired skills and experience outside formal education to have their competences recognised and obtain a nationally recognised certificate. During the period analysed in the impact evaluation, NVQ preparation programmes were available to jobseekers from January to August 2015, aiming to support them throughout the validation process.
Institutional Training: This programme is designed to enhance the employability of registered jobseekers by providing targeted re‑skilling and up-skilling opportunities aligned with the needs of the labour market. Accredited external providers deliver the training, which can last up to 12 months. Extensions of up to 18 months are available for individuals with disabilities or reduced employability.
During the period analysed in this report, institutional training programmes were available from January to August 2015 and varied in both content and duration. One example is teacher education training, which covered subjects such as educational psychology and pedagogy, with a minimum of 375 hours of training. Another example is a basic computer literacy course, which introduced participants to the fundamentals of Windows, Word, and email, with a total duration of 30 hours.
Non-Formal Education and Training: Starting from July 2016, Institutional training programmes were replaced by Non-Formal Education and Training. Similarly to institutional training, these programmes offered participants access to targeted re‑skilling and up-skilling opportunities through non-formal education and training, with the goal of enhancing their employability and improving their prospects of re‑entering the labour market. Accredited external providers provided training across a range of fields, such as business, IT, construction, and production technologies. The programme included three subprogrammes, each tailored to specific target groups and needs:
Regular Non-Formal Education and Training: This programme targets registered jobseekers aged 50 and above, as well as unemployed individuals aged 30 and above who are either long-term unemployed (over 12 months) or with an educational level below ISCED Level 3.
Non-Formal Education and Training Programmes for Youth: This programme targets young jobseekers under 30.
Local Non-Formal Education and Training Programmes (starting from August 2017): These programmes deliver training tailored to the specific needs of local labour markets.
Note: Includes information on individuals entering training from January 2015 and December 2018.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia and the Employment Service of Slovenia.
Although the training programmes were offered at different times, they share several similarities. Both the Non-Formal Education and Training programme and its predecessor, the Institutional Training programme, provide a diverse range of courses, including language, entrepreneurship, web application programming and welding. To ensure alignment with labour market needs, the course catalogue is updated regularly. Moreover, all five programmes have comparable duration, ranging from 44 to 57 days, equivalent to just under two months.
However, the costs of the programmes show greater variation. The NVQ preparation programme has the highest median cost with EUR 780, followed by the Local Non-Formal Education and Training programmes with EUR 726. The regular Non-Formal Education and Training programme and its variant targeted to young jobseekers have slightly lower median costs, with EUR 547 and EUR 560, respectively. In contrast, the Institutional Training programme has significantly lower costs, a median cost of just EUR 290. While the main analysis of the effects of training programmes in Slovenia (Chapter 4) considers the joint effect of all five training programmes, Section 4.3.1 examines the impact of each programme individually.
The provision of training in Slovenia operates through a two‑stage tendering process administered by the ESS. Stage 1 comprises an application to enter the training registry, establishes their eligibility to offer specific programmes in a given location. The application process operates on a rolling basis, enabling potential providers to apply at any time. Stage 2 comprises of competitive bidding for specific courses. Once a training need arises, providers in the registry are invited to submit bids. Contracts are awarded based solely on the lowest bid price, a practice that ensures cost efficiency but may inadvertently affect training quality.
This setup has resulted in a wide array of training programmes that can be implemented quickly. For instance, by October 2024, the ESS catalogue featured 140 distinct training programmes (ESS, 2024[2]). Programmes are added and removed regularly based on skills anticipation exercises (e.g. ESS (2024[3])), employer surveys (e.g. ESS (2024[4])), the ESS occupational barometer, and long-term forecasts of skills needs (MoLFSA, 2023[5]). For example, 32 programmes were removed from the registry in October 2023 based on perceived future irrelevance (ESS, 2024[2]).
An illustrative example of the system’s flexibility is the addition of a ChatGPT training programme in May 2024. Recognising the rapid emergence of artificial intelligence and its relevance to jobseekers, the ESS adopted innovative eligibility criteria for instructors, requiring only prior experience in conducting workshops on the subject, rather than formal qualifications. This approach underscores the adaptability of the training provision system in addressing cutting-edge skills demands.
3.2.2. Employ.me is the employment incentive programme analysed
Zaposli.me (Employ.me) is the longest-running employment incentive programme in Slovenia. The precise parameters of the programme vary depending on labour market conditions and the amount of funds available, with eligibility conditions for jobseekers tightened during times of high unemployment. In addition, as with most other ALMPs in Slovenia, it is subject to volatile financing, leading to periods where the programme does not accept new participants. During the time horizon during which new participants were analysed, 2015‑18, the programme did not have any individuals entering the programme until June 2016 (thereafter, there were new participants throughout the period studied). The subsidy duration is for one year for all jobseekers. However, the subsidy amounts vary depending on the participants, with larger subsidies for jobseekers with more barriers to becoming employed, as defined by their education level, age and duration of unemployment (Figure 3.1). The median and mean subsidy amounts paid during the periods examined were EUR 6 000 and EUR 5 529, respectively (the latter being lower because 8% of participants prematurely finished the programme, before the one‑year period envisioned). The subsidy amount thus covered 38% of participants’ gross annual wages on average.
Subsidy amounts for the “Zaposli.me 2017/19” programme
Note: Amounts relate to entrants for 2017‑19 programme. An additional category of eligible individuals consists of social inclusion or social activation programme participants who are at least 30 years old. They are eligible for the subsidy immediately upon that programme’s conclusion. Subsidy amounts listed are for full-time employment; in the case of part-time employment, they are correspondingly lower.
Source: Employment Service of Slovenia (ESS) (2017[6]), Javno povabilo delodajalcem za izvedbo projekta v okviru programa spodbujanje zaposlovanja – Zaposli.me 2017/19 [Public invitation to employers for the implementation of a project within the programme promotion of employment – Employ.me 2017/19].
As in other countries, the employment incentive programme imposes a number of additional conditions designed to minimise strategic behaviour by employers and to minimise deadweight costs, i.e. the hiring of workers through the subsidy programme who would have been hired in the absence of the subsidy. They restrict eligibility to employers who have not reduced employment (or workers’ hours) in the previous three months, with specific exceptions allowed in cases such as retirements. They exclude the hiring of workers who have worked for the employer in the previous 12 months and those with ownership or management rights, although they do not exclude individuals who have other personal ties to the employer.
While it is not directly relevant for the results in the current evaluation, it is worth noting some recent changes that have been made to the parameters of the Employ.me programme since the period evaluated in the study (ESS, 2024[7]). First, the subsidy amounts have been increased by 8% relative to their amounts in 2018, to a maximum of EUR 8 640 (increases which amount to less than a half of the general rate of inflation during this period). Second, the target groups have been expanded to include all jobseekers considered at risk of long-term unemployment, such as social assistance recipients and members of the Roma population. In addition, since October 2024, subsidies have been made available to jobseekers regardless of age. This marks a departure from the original tender issued in April 2024, which had retained the 30‑year age threshold from previous programmes, and reflects the ability of the ALMP system in Slovenia to adapt to evolving labour market conditions.
One notable feature of the new Employ.me programme is that an employer’s eligibility is also based on whether they are listed in a dedicated registry of employers with negative references. This registry enables the ESS to apply more comprehensive eligibility criteria for employment incentives than is commonly practised in other countries. This registry, maintained by the ESS, is a publicly-available database which lists employers based on violations of labour legislation or outstanding obligations (ESS, 2024[8]). Violations include violations of labour laws such as illegal hiring or staffing policies, paying sub-minimum wages, received fines from the Financial Administration for undeclared work, failure to submit monthly tax forms or pay mandatory social security contributions. Employers remain in the database for six months following the finalisation of penalties, unless new violations occur, or until their financial obligations are settled. In addition to being ineligible for receiving employment incentive payments, such employers are not eligible for job vacancy referrals from the ESS.
The precise procedure for hiring someone via an employment incentive programme has varied slightly from 2016 to present. In the period examined in the impact evaluation, 2015‑18, the procedure was as follows. After posting a vacancy with the ESS and filing an Employ.me application with the necessary documentation, an employer would receive a short list of potential candidates compiled by a PES counsellor, taking into account the eligibility criteria for the subsidy, as well as the skills and experience required for a particular vacancy. The employer could then hire someone from this list of candidates through the subsidy. For the 2024 implementation, the application process differs in one important aspect: employers may indicate that they already have a suitable candidate in mind. If they do, they can indicate whether they would also like additional job candidates to be referred to them from the ESS. This reflects the fact that the employment incentives have become less appealing to employers in recent years. This is due partly to the fact that, given the record low unemployment rate, the remaining unemployed often face multiple barriers to employment, and also to the fact that the value of the subsidies has decreased in real terms.
Employ.me had a total of 9 523 participants during the period analysed (Figure 3.2). At the beginning of the period, there was a spike in the number of participants, as the programme had not been offered for more than a year. Thereafter, the programme had roughly 300 new participants every month. As participation in the programme is one year, the last participants examined finished the programme in December 2019. A vast majority of participants (92%) participated in the programme for its full duration, with 5% of participants terminating the programme early for justified reasons and 3% finishing it for unjustified ones.
Monthly inflows into Employ.me programme
Source: OECD calculations based on data provided by the Employment Service of Slovenia (ESS).
This section explores the outcomes and characteristics of individuals participating in the training and employment incentives programmes evaluated in this report. The section begins by presenting key insights into the employability of jobseekers across a range of demographic and socio‑economic factors, including gender, age, duration of unemployment, education level and nationality. This serves as a basis for evaluating the extent to which the studied ALMPs are directed towards individuals who are either closer to or further from the labour market. The section then explores differences in the take‑up rates among these different groups.
3.3.1. Average job-finding rates confirm that groups such as older jobseekers have a greater distance to the labour market
The patterns of exit from unemployment show that certain groups of jobseekers, such as older jobseekers or those with lower levels of education, face greater challenges in finding employment. Figure 3.3 illustrates the subsequent employment outcomes of individuals registered as unemployed during the period analysed in the impact evaluation (January 2015 through December 2018), based on monthly unemployment data.
Among certain groups such as women and men over 50, individuals with primary education or lower and those unemployed for more than 23 months, only a small proportion transition into employment within two years. However, the probability of exiting unemployment is higher for younger jobseekers, those with at least secondary education, and individuals unemployed for shorter periods of time. For example, 70.9% of individuals newly unemployed (fewer than 2 months) find employment within the next two years, compared to just 22.2% of those who have been unemployed for at least two years. This pattern is broadly consistent with stylised statistics from other countries (Eurostat, 2024[9]).
Shares of registered unemployed becoming employed within 12, 24 and 36 months, Slovenia
Note: Statistics are calculated based on monthly statistics of individuals unemployed from January 2015‑December 2018 for employment outcomes experienced through December 2022.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia and the Employment Service of Slovenia.
3.3.2. Participation in the ALMPs examined is generally higher amongst certain demographic groups
To illustrate the extent to which specific categories of jobseekers are likely to enter ALMPs, this section compares the characteristics of the participants in the selected training and employment incentive programmes with those of all individuals who are registered as unemployed with the PES. These comparisons – presented in Figure 3.4 – are based on the average monthly unemployment stocks from January 2015 through December 2018.
Structure of ALMP participants and registered unemployed within each broad category, Slovenia
Note: ALMPs stand for the training and employment incentive programmes evaluated in this report. Shares are calculated within each of the five broad categories in the figure: if a demographic category of ALMP participants were represented in proportion to their share amongst all registered unemployed, the length of the bars would coincide with the red squares. Statistics for stocks of all unemployed aged 18‑65 are calculated based on averages of monthly statistics during the 2015‑18 period. Participant numbers refer to totals during the 2015‑18 period for individuals aged 18‑65 entering either training or employment incentives.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia and the Employment Service of Slovenia.
Relative to their share of the unemployed, women are slightly more likely than men to participate in both training and employment incentive programmes. Between 2015 and 2018, women accounted for 54.5% of training participants and 52.5% of participants in employment incentives, despite making up only 49.7% of the unemployed during the same period. In contrast, men accounted for 45.7% of training participants and 47.5% of participants in employment incentives, while representing 50.3% of all unemployed. However, the overall differences between women and men remain relatively small.
The age profile of participants in training and employment incentive programmes shows some differences. Younger jobseekers are disproportionately more likely to enrol in (or being referred to) training programmes, while they do not participate in the employment incentives programmes examined, reflecting the eligibility criteria for the Employ.me programme at the time (other programmes in place at the time were targeted towards those under 30 but are not the subject of the current evaluation). Indeed, female jobseekers under 30 make up 17.4% of training participants, despite representing only 12% of the unemployed population. Similarly, male jobseekers under 30 account for 15.4% of training participants, compared to their 12% share in the unemployed population.
In contrast, registered jobseekers over 50 display the opposite pattern: they are disproportionately more likely to enrol in employment incentive programmes and less likely to participate in training. Women over 50 accounted for just 6% of training participants, even though they make up 14.2% of the unemployed population. Similarly, only 7% of training participants were men over 50, despite this group constituting 20.3% of the unemployed population. Despite their lower participation in training, older jobseekers demonstrate higher enrolment in employment incentives: 25.9% of employment incentive participants were women over 50, despite making up only 14.2% of jobseekers. Similarly, men over 50 accounted for 29% of employment incentive participants, compared to their 20.3% share in the population of unemployed.
In the middle of the age distribution, participation patterns differ by gender. Women aged 30 to 50 are overrepresented in both training and employment incentive programmes, making up 31.1% of training participants and 26.6% of employment incentive participants, while comprising only 23.5% of the unemployed population. In contrast, men aged 30 to 50 are overrepresented in training programmes (23.1%) but underrepresented in employment incentives (18.5%) relative to their 20.2% share in the population of unemployed. Overall, these statistics suggest that age plays a more significant role than gender in determining the participation of unemployed individuals in training programmes or employment incentives.
ALMP participation also varies by unemployment duration. For short-term unemployed (less than 2 months), participation in the two ALMPs analysed is very low. For unemployment durations between 2 and 6 months, registered jobseekers are disproportionately more likely to enter training, while participation in employment incentives remains close to zero. This pattern might also correlate with the age patterns described above: younger jobseekers who experience, on average, shorter unemployment spells are more likely to enter training programmes. Participation in employment incentive programmes increases substantially after 7 months of unemployment, with individuals who have been unemployed for more than 12 months being disproportionately overrepresented among employment incentive programmes. Figure 3.5 illustrates these patterns over the full distribution of unemployment duration (up to 71 months): short- to medium-term unemployed (between 2 and 12 months) are disproportionately more likely to enter training programmes while long-term unemployed (more than 12 months) are more likely to enter employment incentives.
Taken together, Figure 3.3 and Figure 3.4 indicate that individuals with longer unemployment durations, and thus further from the labour market, are more likely to participate in both ALMPs. This relationship is clearly illustrated in Annex Figure 3.A.1. , which illustrates that jobseekers with longer unemployment spells not only have lower probabilities of finding employment within 12 months but are also more likely to enrol in ALMPs.
In relation to education, most registered jobseekers have secondary-level education. Compared to their proportion in the population of unemployed (53.1%), they are disproportionately more likely to participate in employment incentive programmes (56.5%) and less likely to enrol in training programmes (49.2%). At both ends of the educational distribution, participation patterns in ALMPs differ: Tertiary-educated jobseekers are disproportionately more likely to participate in both training and employment incentives programmes, while those with only primary-level education are less likely to participate in either.
Participation patterns also vary by nationality. Slovenian nationals are more likely to enter employment incentive programmes and less likely to enrol in training, while the opposite trend is observed for non-EEA nationals. However, it is important to note that these patterns do not account for other characteristics of the unemployed population.
Stock of registered unemployed and ALMP entrants by unemployment duration
Note: ALMPs stand for the training and employment incentive programmes evaluated in this report. Statistics for stocks of all unemployed aged 18‑65 are calculated based on averages of monthly statistics during the 2015‑18 period. Participant numbers refer to totals during the 2015‑18 period for individuals entering either training or employment incentives.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia and the Employment Service of Slovenia.
3.3.3. Small employers are more likely to hire training participants, while wage subsidies are more commonly used by larger employers
Thus far, the chapter has described the attributes of jobseekers engaging in ALMPs. Another relevant question concerns the characteristics of firms participating in wage subsidies or hiring former training participants after they complete their training. Figure 3.6 contrasts the distribution of these firms across size categories with that of firms hiring unemployed individuals in general. In the case of wage subsidies, the figure refers to the employment during the subsidy period, while for training programmes it refers to the first job obtained by individuals who become employed after completing the training. The alignment or non-alignment of the bars with the red squares indicates whether ALMP participants are overrepresented or underrepresented in an employer size category compared to their overall distribution among the registered unemployed.
Larger firms, starting from 50 employees, are disproportionately more likely to participate in wage subsidy programmes compared to their share in the population of firms hiring unemployed individuals in general. Firms with a size of 150 to 499 employees are particularly active in wage subsidy programmes: while they account for 11% of the hirings among unemployed individuals, their share among wage subsidies amounts to almost 16%. This finding – that larger firms are more likely than smaller ones to participate in wage subsidy programmes – contrasts with the experience of e.g. Greece, where larger firms appear less likely to engage in wage subsidies (OECD, 2024[10]). The difference may be attributable to the eligibility conditions in Slovenia, where the number of participants an employer can hire is closely tied to employer size. Furthermore, in contrast to Greece, Slovenia does not require that employers retain workers past the subsidy receipt period; rather, the ESS gives preferential treatment for subsequent subsidies to employers who have retained subsidised workers in the past. Cross-country evidence shows that larger employers are more able to offer more job stability to their workers (Haltiwanger, Scarpetta and Schweiger, 2014[11]), which could make it more likely that they qualify for subsequent subsidies.
In contrast to wage subsidy programmes, small firms are disproportionately more likely to hire jobseekers who have completed training programmes compared to large firms. This is particularly evident for firms with 1 to 10 employees, which account for over 60% of the hirings of jobseekers after training, despite hiring only 30% of registered unemployed in general. This may be tied to the fact that international evidence shows that smaller firms are disproportionally likely to suffer shortages of skills (OECD, 2021[12]): they are less successful in attracting and retaining skilled workers; and face higher direct and indirect costs of training the workforce. As a result, they may rely more heavily on hiring workers who have already acquired these skills through external training programmes, like the ones provided by the ESS.
Share of ALMP participants and registered unemployed across firm size category, Slovenia
Note: ALMPs stand for the training and wage subsidy programmes evaluated in this report. Shares are calculated as the respective shares within the total distribution of size categories: if firm size category were represented in proportion to their share amongst all registered unemployed, the length of the bars would coincide with the red squares. Size categories are based on the total number of workers at a given employer in the SRDAP employment database based on the month in which an individual became employed at that employer. Statistics for wage subsidy participants refer to the employers receiving the subsidies; statistics for the training participants refer to the first employer after completion of the theoretical and practical training. Statistics for stocks of all unemployed are calculated based on averages of monthly statistics during the 2015‑18 period. Participant numbers refer to totals during the 2015‑18 period for individuals entering either training or wage subsidies.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia and the Employment Service of Slovenia.
In terms of economic sectors (Annex Figure 3.A.2.), ALMP participants – whether in wage subsidy or training programmes – are disproportionately overrepresented in sectors such as “Water supply, sewerage, waste management and remediation,” “Transportation and storage,” “Financial and insurance activities,” “Real estate activities,” “Professional, scientific and technical activities” and “Other activities.” Notably, the sector “Administrative support and service activities” has a particularly high share of wage subsidies relative to the share of all registered unemployed who find employment in this sector. Similarly, former training participants are more likely to find employment in the sectors of “Wholesale and retail trade, repair of motor vehicles and motorcycles” and “Professional, scientific and technical activities” after their training.
A comprehensive assessment of the impact of ALMPs on subsequent labour market outcomes requires detailed data on jobseekers’ characteristics, their participation in ALMPs, and their employment outcomes. The data used for the evaluation in this report were obtained from multiple sources, as summarised in Table 3.2, and cover various periods from January 2012 to December 2022.3 Employment and earnings data from January 2012 onwards are used in order to establish earnings and unemployment histories of jobseekers in order to aid in the process of identifying comparable pairs of participants and non-participants (as described in Section 3.5 below). To ensure a comprehensive analysis while safeguarding individuals’ privacy, pseudonymised unique identifiers were used to integrate data from these sources. The resulting database provides a rich understanding of individuals’ participation in ALMPs, their background characteristics – including prior labour market history – and their labour market outcomes, such as employment and wages.
The resulting database includes detailed information on the 298 584 unique individuals who were registered as unemployed at any point during the 2015‑18 period. These individuals experienced 1.2 million distinct unemployment spells in total, with some individuals experiencing only one spell and others experiencing multiple spells. The final database used in the analysis contains 13 901 entries into training and 9 498 entries in employment incentives for people who registered as unemployed from January 2015 through December 2018. Individuals are observed to enter into training and employment incentives in 8.0% and 3.2% of employment spells, respectively, during this period. A small percentage of participants in the raw data are not included in the final analysis (3.3% and 1.7% of training and employment incentives participants, respectively). More information on the data used and how they were processed are available in the technical report accompanying this publication (OECD, 2025[13]).
One relevant note relates to the quality of the registry data on migration. The source of the information on migration is the population registry maintained by the Statistical Office, which compiles this database from several sources, most prominently the Ministry of the Interior (Official Gazette of the Republic of Slovenia, 2023[14]). Residents of Slovenia have strong financial reasons to have any migration from Slovenia accurately recorded in this database due to its use by the tax authorities, who use it (in combination with other data sources) to establish whether someone is still considered a resident for tax purposes. Conversely, when they move to Slovenia and if they are not employed, they also have financial incentives to register with relevant authorities for health insurance purposes.
One potential problem often encountered in impact evaluations of ALMPs concerns the question of how to deal with multiple, sequential entries into ALMPs. In the presence of multiple interventions and possible overlap between different ALMPs, identifying the precise effects of one specific ALMP presents an important challenge. In the case of Slovenia, this is a somewhat relevant concern: for those who were observed to enter an ALMP, the large majority (71.6%) entered into only one ALMP during their entire unemployment spell.4 In practice, this turns out to not have a meaningful impact on the results – as presented in the accompanying technical report (OECD, 2025[13]), the results are robust to whether or not the analysis focused only on the first ALMP entered during the spell.
While the data underpinning this evaluation are exceptionally detailed, a few limitations are worth noting. First, the employment data pertain exclusively to formal employment relationships defined by labour law – those registered with the National Health Insurance Institute for health insurance purposes. This excludes certain forms of work that would typically be captured by survey-based measures, such as civil employment contracts, student work, and undeclared work. In addition to dependent employment, the data also include information on the self-employment. Second, monthly earnings are derived from annual data, extrapolated to the monthly level for periods when an individual is recorded as employed in the employment data. Unlike the employment data, the earnings data include income from civil employment contracts and student work. However, this methodological difference may result in slight inconsistencies between the employment and earnings measures, introducing potential measurement error. From an analytical perspective, such measurement error can affect the precision of the estimates, leading to wider confidence intervals. Additionally, the point estimates may be biased towards zero, meaning that any positive or negative programme effects could be underestimated (Levi, 1973[15]).
Additional questions related to data are discussed in the accompanying technical report (OECD, 2025[13]). This report discusses the data in more detail, identifying how the analysis could be enriched with additional data and discussing ways to make better use of data in the future.
Assessing the impact of an ALMP requires comparing the labour market outcomes of participants – such as their employment status or earnings – with the outcomes they would have experienced had they not participated in the ALMPs. Because these hypothetical, or “counterfactual,” outcomes cannot be directly observed, it is necessary to construct them using the available data. A straightforward approach might involve comparing the outcomes of those who participated in the ALMP, such as training, with those who did not. However, as detailed further below, this method is problematic in the absence of random assignment to the training. Participants and non-participants are likely to differ in ways that affect their labour market outcomes, making these groups inherently non-comparable. Such simple comparisons risk introducing selection bias, which could lead to inaccurate estimates of the programme’s true effects.
As with other evaluations employing similar estimation techniques, this impact evaluation must account for several potential sources of selection bias. For instance, certain individuals – such as those who are more motivated – may be more likely to participate in training programmes and subsequently achieve better employment outcomes for reasons unrelated to their participation in the programme. Conversely, individuals facing additional barriers to employment, and who therefore experience poorer employment outcomes, are often more likely to be referred to ALMPs by PES counsellors. Additionally, many individuals who do not participate in an ALMP may be excluded simply because they find employment quickly and exit unemployment without requiring support from the ESS. This group may, in fact, achieve better future employment outcomes than ALMP participants. Having secured a job rapidly, they are likely to sustain that employment and are more likely to be employed in the following months or years compared to individuals who remain unemployed. Moreover, ESS counsellors may be less inclined to recommend ALMPs for these individuals, as participation could unintentionally prolong their unemployment spell by the duration of the programme. This dynamic further complicates the evaluation, as it highlights the need to carefully address selection bias to ensure accurate estimation of programme effects.
To mitigate these sources of bias, the approach adopted in this report accounts for differences in demographic characteristics (e.g. gender, education, age) as well as observed skills and employment barriers between ALMP participants and non-participants. This method allows for a more accurate estimation of the “treatment effect” by comparing individuals who are similar in terms of their observable attributes. The evaluation compares the outcomes of ALMP participants (the “treatment” or “intervention” group) with those of a closely matched group of non-participants (the “control” or “comparison” group). By ensuring that the two groups are comparable in terms of key characteristics, this approach aims to isolate the effect of the ALMP from other factors influencing employment outcomes.
The econometric approach used in this evaluation incorporates several features to ensure the comparability of treatment and control groups and to provide unbiased results:
Matching based on unemployment histories: Individuals are compared only if they have similar registered unemployment histories. Specifically, the labour market outcomes of those entering an ALMP in a given month are compared with those of individuals who have not (yet) entered an ALMP but have exhibited comparable patterns of registered unemployment over the preceding three years. For individuals in the control group, the time from which outcomes are measured begins at the point in the unemployment spell when they are matched with a similar individual who does enter an ALMP (with both individuals having similar unemployment duration). This methodology, initially adopted by Sianesi (2004[16]), is explained in the accompanying technical report (OECD, 2025[13]), which also examines results with an alternative set of assumptions.
Exact matching on additional characteristics: To ensure comparability, individuals are matched not only on unemployment histories but also on identical values for key characteristics: calendar month and year of entry into the programme, gender, earnings (several categories) and, for those who have not been employed in the last three years, age group (three categories) and broad level of education (two categories).
A rich set of additional personal characteristics is used to identify individuals with similar probabilities of entering the ALMP under consideration: Within the precise groups mentioned above, individuals are further matched to similar individuals on the basis of an estimate of their probability of entering the ALMP under consideration (based on monthly panel data). Such an approach – based on a so-called propensity score – is commonly used in the literature to address the difficulty of otherwise accounting for a wide range of additional personal characteristics (Card, Kluve and Weber, 2018[17]). The propensity score is a measure of the probability of participating in the programme under analysis. The following factors are taken into account when calculating the propensity score: (i) each individual’s employment history (earnings, occupation), (ii) unemployment information (duration until month of entry into ALMP or assignment into control group; unemployment benefit level and remaining duration), employability rating (based on ESS counsellor judgement), disability status, demographic characteristics (age, education, gender, nationality, marital status, and location.) Details on these characteristics are presented in the accompanying technical report (OECD, 2025[13]).
The choice of the research design for this evaluation is shaped by the relatively broad eligibility criteria of ALMPs in Slovenia and the availability of detailed administrative data. Specifically, the lack of strict eligibility criteria – such as an objectively determined profiling score – precludes the use of a research design that exploits these thresholds.5 For example, in programmes targeted at specific groups, such as young people, individuals just above an age threshold could serve as a natural comparison group to estimate counterfactual outcomes – what would have happened to participants had they not participated.
In the absence of such thresholds, this evaluation relies instead on the rich administrative data available. As described above, individuals are matched along multiple dimensions, including their prior employment history and the exact calendar month of ALMP entry, to construct a comparable control group.6 This methodology, designed to ensure robust and unbiased comparisons, is detailed in the accompanying technical report (OECD, 2025[13]).7
The propensity score matching approach is a widely used method in impact evaluations. In a recent meta‑analysis of 95 impact evaluations of ALMPs (European Commission, 2023[18]), 85% of the studies employed this technique. Canada, for example, regularly and systematically evaluates its ALMPs based on this approach (OECD, 2022[19]). This approach has been applied in several recent OECD evaluations of ALMPs, in Greece, Finland, Ireland, Latvia, Lithuania and Portugal (OECD, 2023[20]; 2022[21]; 2019[22]; 2024[23]; OECD/Department of Social Protection, Ireland/EC-JRC, 2024[24]; OECD, 2024[10]). Nevertheless, it is also worth noting that the reason for such pervasive use of the propensity score matching approach is arguably that alternative approaches – which may be preferable because they require less strict econometric assumptions – are often not viable for researchers.
Counterfactual impact evaluations of ALMPs typically focus on outcomes related to labour force participation, particularly changes in employment probability for participants compared to similar non-participants. Among these, the impact of ALMPs on employment probability has been the most extensively studied. For example, a meta‑analysis by Card, Kluve and Weber (2018[17]) included employment probability estimates from 111 impact evaluations of ALMPs and a newer one focusing on ESF-funded programmes identified 94 impact evaluations (European Commission, 2023[18]). Given that a key aim of ALMPs is to facilitate individuals’ entry into employment, this outcome is undeniably important. However, this focus may partly reflect data availability, as data on other outcomes are often more challenging to collect. However, examining additional outcomes, particularly those related to job quality, can offer valuable insights into how ALMPs contribute not only to employment rates but also provide a more nuanced view of the potential benefits and trade‑offs involved in ALMP participation.
In Slovenia, the availability of rich and comprehensive data enables the evaluation of ALMPs to include a broad range of outcomes over an extended period. Outcomes are tracked continuously for up to four years from the start of the programme, providing valuable insights into both short- and medium-term impacts. These outcomes are calculated on a monthly basis and monitored over time relative to a reference month. For the treatment group, the reference month is defined as the month in which an individual enters an ALMP, while for the control group, it corresponds to the same calendar month as that of the matched treatment group participant. Further details on the calculation of these outcomes are outlined in the accompanying technical report (OECD, 2025[13]).
The following outcomes are examined:
Probability of entering employment. This probability is measured using a binary outcome variable which is equal to 1 if individual is employed at a certain time, and equal to 0 otherwise.
Probability of remaining registered as unemployed. This is measured using a binary outcome variable which is equal to 1 if individual is unemployed at the end of a calendar month, and equal to 0 otherwise.
Probability of receiving unemployment benefits. This is measured using a binary outcome variable which is equal to 1 if individual receives unemployment benefits at any point during the calendar month, and equal to 0 otherwise.
Probability of migrating from Slovenia. This probability is measured using a binary outcome variable which is equal to 1 if individual is no longer registered as living in Slovenia, and equal to 0 otherwise.
Probability of becoming inactive. This a binary outcome variable which is equal to 1 if individual is not in the states mentioned previously (employment, registered unemployment, or migrating abroad), and equal to 0 otherwise.
Cumulative employment duration. This measures the cumulative duration of all jobs held during the observation time, in calendar days, after the reference month.
Cumulative earnings. This measures total earnings, gross of taxes and contributions, in constant prices, in all jobs held during the observation time.
Occupational mobility. The analysis maps the occupation of individuals entering employment onto an occupational index, which can be interpreted as a “job ladder”. The construction of the index is detailed in Section 3.7.
In addition to analysing aggregate effects, the results are disaggregated by sub-groups of individuals and selected programme attributes. Chapters 4 and 5 present findings for sub-groups of workers based on gender, age, education level, nationality, and duration of unemployment. This stratified analysis provides a deeper understanding of how different groups benefit from the programmes.
The OECD’s work on impact evaluations of ALMPs seeks to extend beyond commonly examined outcomes, such as employment probabilities or wage effects, to address broader labour market challenges. As in other countries participating in the OECD-EC project on ALMP impact evaluation, the work with Slovenia aims to examine an important yet underexplored dimension: the impact of ALMP participation on occupational mobility. A substantial body of empirical evidence underscores the “scarring” effects of job loss, with long-lasting adverse impacts on wages that persist even after re‑employment (for example, Lachowska, Mas and Woodbury (2020[25])). Empirical evidence also shows that jobseekers exiting unemployment tend to disproportionately enter (or return to) low-skills occupations compared to the employed population (Bisello, Maccarrone and Fernández-Macías, 2020[26]). ALMPs have the potential to help to counteract these effects by mitigating or possibly even reversing the typically observed negative effects of job loss on individuals’ career trajectories. Training programmes, for instance, can equip individuals with the skills and qualifications required for higher-skilled roles, opening pathways to upward occupational mobility. Similarly, employment incentives may encourage employers to hire jobseekers who face barriers to employment and provide on-the‑job training, further enhancing participants’ prospects of securing higher-quality positions.
To provide a practical measure of occupational mobility, this analysis employs an occupational index derived from observed wages. Following the methodology outlined by Laporšek et al. (2021[27]) and used in past ALMP impact evaluations in Greece (OECD, 2024[10]), Lithuania (OECD, 2022[21]), Finland (OECD, 2023[20]) and Spain (OECD, 2021[28]), a wage index is calculated for each detailed occupational code using data on the wages and employment of individuals in Slovenia during the 2011‑22 period.8 The index assigns each of the 383 distinct occupational codes a value that is both intuitive and practical: an occupation with an index value one unit higher than another corresponds to an average real monthly wage that is 1 percentage point greater relative to the overall average wage. Moreover, changes in the index can be interpreted as shifts in an individual’s occupational standing: increases reflect upward mobility, while decreases indicate a downward movement on the occupational ladder.
The occupational index distribution for Slovenia shows relatively small changes in the distribution following unemployment: individuals who become re‑employed in aggregate become employed in similar occupations. By contrast, individuals do enter slightly lower-paying occupations after long-term unemployment spells (Figure 3.7). Following a long-term unemployment spell, a slightly larger share of individuals work in occupations with wages of index values below 83 (meaning these occupations pay, on average, less than 83% of the average wage). On average, individuals becoming re‑employed after training have an occupational index that is 1.2 percentage points lower than before they were unemployed.
It is interesting to compare the aggregate effects of unemployment on the occupational index with the findings of similar analyses in Lithuania, Finland and Greece. In Lithuania, unemployment was found to have a considerable scarring effect, with individuals who become re‑employed disproportionally entering lower-paid occupations – although the effect of training was less clear-cut (OECD, 2022[21]). In Finland, ALMPs served to decrease the dispersion in the occupational distribution and earnings inequality (OECD, 2023[20]). In Greece, there were virtually no differences in distributions of occupational index before and after unemployment (OECD, 2024[10]). The lack of strong observed aggregate effects in Greece may be due to the strong seasonal nature of unemployment fluctuations in Greece: large shares of individuals in the tourism sector cycling in and out of unemployment, often in similar occupations.
Occupational index distribution before and after unemployment, long-term unemployed, Slovenia
Note: The heights of the lines indicate the relative share of individuals in occupations whose average wages are on the horizontal axis, relative to the average real wage. The distributions are calculated for all unemployed individuals during the 2015‑21 period who were registered as unemployed for at least one year before becoming employed. Observations with index values below 54 or above 126 are excluded from the kernel density chart.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia and the Employment Service of Slovenia.
Although a descriptive analysis of the occupational index distributions as shown in Figure 3.7 is instructive for understanding the underlying data, it does not take into account a variety of possible underlying factors that could explain the differences in the distributions. For example, differences in the occupational index distributions before and after unemployment may be subject to composition effects, with a subset of individuals more likely to be re‑employed.
This chapter has provided an overview of selected training and employment incentive programmes in Slovenia. It has also highlighted the extensive administrative data available for evaluating these programmes, including detailed information from the unemployment register, data on ALMP participation, and employment outcome data. These data sources form the foundation for the CIE results presented in the subsequent chapters. The impact evaluation uses a range of observable characteristics of jobseekers, including their prior labour market history, to construct comparable treatment and control groups. The causal effects of the programmes are estimated by comparing the observed outcomes of programme participants with the counterfactual outcomes that would have occurred in the absence of the programmes. The richness of the administrative data facilitates the examination of multiple outcomes. Beyond the most commonly analysed metric – employment probability – the analysis also evaluates the impact of these programmes on employment duration, earnings, occupational mobility, and migration.
References
[26] Bisello, M., V. Maccarrone and E. Fernández-Macías (2020), “Occupational mobility, employment transitions and job quality in Europe: The impact of the Great Recession”, Economic and Industrial Democracy, p. 0143831X2093193, https://doi.org/10.1177/0143831x20931936.
[17] Card, D., J. Kluve and A. Weber (2018), “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.
[2] ESS (2024), Catalogue of training programmes [Katalog programov], https://www.ess.gov.si/fileadmin/user_upload/Partnerji/Dokumenti_Partnerji/Dokumenti_Register/Register_Katalog_programov_A.xls (accessed on 21 October 2024).
[8] ESS (2024), Delodajalci z negativnimi referencami [Employers with negative references], https://www.ess.gov.si/iskalci-zaposlitve/iskanje-zaposlitve/iskanje-delodajalcev/delodajalci-z-negativnimi-referencami/ (accessed on 28 November 2024).
[4] ESS (2024), Napovednik zaposlovanja [Hiring forecasts], https://www.ess.gov.si/partnerji/trg-dela/napovednik-zaposlovanja/ (accessed on 1 October 2024).
[3] ESS (2024), Poklicni barometer [Occupational barometer], https://www.ess.gov.si/partnerji/trg-dela/poklicni-barometer/ (accessed on 2 October 2024).
[7] ESS (2024), Zaposli.me 2024 - Javno povabilo [Employ.me 2024 - Public invitation], https://www.ess.gov.si/delodajalci/financne-spodbude/predstavitev-spodbud-za-zaposlitev/zaposlime-2024/ (accessed on 28 November 2024).
[6] ESS (2017), Javno povabilo delodajalcem za izvedbo projekta v okviru programa spodbujanje zaposlovanja - Zaposli.me 2017/2019 [Public invitation to employers for the implementation of a project within the programme promotion of employment – Employ.me 2017/2019].
[18] European Commission (2023), Meta-analysis of the ESF counterfactual impact evaluations – Final report, Publications Office of the European Union, Directorate-General for Employment, Social Affairs and Inclusion, Pompili, M., Kluve, J., Jessen, J. et al., https://data.europa.eu/doi/10.2767/580759.
[9] Eurostat (2024), Transition from unemployment to employment by sex, age and duration of unemployment - annual averages of quarterly transitions, estimated probabilities, https://doi.org/10.2908/LFSI_LONG_E01 (accessed on 15 December 2024).
[11] Haltiwanger, J., S. Scarpetta and H. Schweiger (2014), “Cross country differences in job reallocation: The role of industry, firm size and regulations”, Labour Economics, Vol. 26, pp. 11-25, https://doi.org/10.1016/j.labeco.2013.10.001.
[1] Institute of RS for VET (2024), Pridobitev certifikata [How to acquire an NVQ], https://npk.si/kako-do-npk/pridobitev-certifikata/ (accessed on 11 December 2024).
[25] Lachowska, M., A. Mas and S. Woodbury (2020), “Sources of Displaced Workers’ Long-Term Earnings Losses”, American Economic Review, Vol. 110/10, pp. 3231-3266, https://doi.org/10.1257/aer.20180652.
[27] Laporšek, S. et al. (2021), “Winners and losers after 25 years of transition: Decreasing wage inequality in Slovenia”, Economic Systems, Vol. 45/2, p. 100856, https://doi.org/10.1016/j.ecosys.2021.100856.
[15] Levi, M. (1973), “Errors in the Variables Bias in the Presence of Correctly Measured Variables”, Econometrica, Vol. 41/5, p. 985, https://doi.org/10.2307/1913819.
[5] MoLFSA (2023), Prihodnost dela – rezultati srednje- in dolgoročnih napovedi potreb trga dela do leta 2037 [The Future of Work – Results of Medium- and Long-Term Labour Market Demand Forecasts up to Year 2037], https://www.gov.si/assets/ministrstva/MDDSZ/Potrebe-trga-dela-do-2037.pdf.
[13] OECD (2025), “Technical report: Impact Evaluation of Employment Incentives and Training for the Unemployed in Slovenia”, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Technical-Report-Evaluation-of-active-labour-market-policies-in-Slovenia.pdf.
[23] OECD (2024), Impact Evaluation of Active Labour Market Policies in Portugal, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/c4b2ca21-en.
[10] OECD (2024), Impact Evaluation of Training and Wage Subsidies for the Unemployed in Greece, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/4b908517-en.
[20] OECD (2023), Evaluation of Active Labour Market Policies in Finland, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/115b186e-en.
[19] OECD (2022), Assessing Canada’s System of Impact Evaluation of Active Labour Market Policies, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/27dfbd5f-en.
[21] OECD (2022), Impact Evaluation of Vocational Training and Employment Subsidies for the Unemployed in Lithuania, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/c22d68b3-en.
[29] OECD (2022), The scope and comparability of data on labour market programmes, https://www.oecd.org/content/dam/oecd/en/topics/policy-issues/employment-services/almpdata-scope-and-comparability.pdf.
[28] OECD (2021), Impact evaluation of Send@ - a digital tool for PES counsellors in Spain, https://www.oecd.org/els/emp/dg-reform-spain-digital-tool-for-pes-counsellors.htm.
[12] OECD (2021), Incentives for SMEs to Invest in Skills: Lessons from European Good Practices, Getting Skills Right, OECD Publishing, Paris, https://doi.org/10.1787/1eb16dc7-en.
[22] OECD (2019), Evaluating Latvia’s Active Labour Market Policies, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/6037200a-en.
[24] OECD/Department of Social Protection, Ireland/EC-JRC (2024), Impact Evaluation of Ireland’s Active Labour Market Policies, Connecting People with Jobs, OECD Publishing, Paris/Department of Social Protection, Ireland, Dublin/European Commission, Joint Research Centre, Brussels, https://doi.org/10.1787/ec67dff2-en.
[14] Official Gazette of the Republic of Slovenia (2023), Letni program statističnih raziskovanj za 2024 [Annual programme of statistical surveys for 2024], https://pisrs.si/pregledPredpisa?id=DRUG5249.
[16] Sianesi, B. (2004), “An Evaluation of the Swedish System of Active Labor Market Programs in the 1990s”, Review of Economics and Statistics, Vol. 86/1, pp. 133-155.
Active labour market policy (ALMP) participation rates and rates of exit into employment, Slovenia
Note: Statistics are calculated based on monthly statistics of all individuals unemployed from January 2015 to December 2018 for employment outcomes experienced through December 2022. The sizes of the circles are proportional to the number of registered unemployed in each category. ALMP participation refers to participating in any of the four main ALMPs in Slovenia: training, wage subsidies (employment incentives), On-the-job training, and Public works.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia and the Employment Service of Slovenia).
Share of ALMP participants and registered unemployed across sector of economic activity, Slovenia
Note: ALMPs stand for the training and wage subsidy programmes evaluated in this report. Shares are calculated within each group separately: if a sector category of ALMP participants were represented in proportion to their share amongst all registered unemployed, the length of the bars would coincide with the red squares. Statistics for stocks of all unemployed are calculated based on averages of monthly statistics during the 2015‑18 period. Participant numbers refer to totals during the 2015‑18 period for individuals entering either training or wage subsidies.
Source: OECD calculations based on data from the Statistical Office of the Republic of Slovenia, the Employment Service of Slovenia and the Agency of the Republic of Slovenia for Public Legal Records and Related Services (AJPES).
← 1. Under the OECD classification of ALMPs (OECD, 2022[29]), Slovenia’s Employ.me (Zaposli.me) wage subsidy programme falls under Category 4 (Employment incentives), specifically, Subcategory 4.1 (Recruitment incentives includes measures which are payable for a limited period only). In this report, the terms wage subsidies and employment incentives are used interchangeably.
← 2. From the perspective of the evaluation, however, these changes were not large enough that they could be expected to significantly affect the evaluation results or be subject to evaluation themselves.
← 3. Additional data, not outlined here, were also obtained for the purposes of conducting additional evaluations of the Slovene ALMP data, which are outside the scope of the evaluation described in this report.
← 4. This statistic includes not only the two types of ALMPs examined in this report but also the main other ALMPs in place in Slovenia including, most prominently, direct job creation.
← 5. The employment incentives programme imposed strict minimum age criteria during the evaluation period. However, individuals just below the age threshold may have been incentivised to remain unemployed to qualify for the programme once eligible. This, combined with eligibility criteria linked to the duration of unemployment, complicates the use of a discontinuity-based approach to identify the programme’s effects. Such dynamics could distort the natural comparison between eligible and ineligible individuals, making it difficult to isolate the true impact of the programme.
← 6. To give a hypothetical example, an individual entering a training programme in September 2018 after 12‑24 months of unemployment is matched to an otherwise similar individual who, in September 2018, has also been unemployed for 12‑24 months. Outcomes for both individuals from September 2018 are subsequently compared in the analysis: for these individuals, the reference point from which outcomes are measured in subsequent months is September 2018.
← 7. The Technical Report also includes tests which show that outcomes such are earnings and employment probability did not differ between the treatment and control groups in the three years preceding treatment (or entry into the control group), and that the observed characteristics do not differ systematically between the two groups.
← 8. The analysis uses 4‑digit codes and is calculated from real monthly wages at constant 2022 prices. Further restrictions are made in calculating the index, such as excluding individuals who are not employed full-time, those with multiple concurrent employment spells, and outliers with extremely high or low reported wages.