This policy brief summarises the results of a comprehensive counterfactual impact evaluation and cost-benefit analysis of Slovenia’s four main active labour market policies (ALMPs) based on rich individual-level data from multiple administrative databases. The scope of the evaluation extends beyond the labour market outcomes typically examined in such evaluations: in addition to earnings and employment probabilities, it also evaluates the effects of the ALMPs on health and social outcomes, and the extent to which ALMPs are facilitating the green and digital transitions.
All four of Slovenia’s main ALMPs examined – classroom-based training, on-the‑job training, private sector wage subsidies and public works – have generally positive effects on participants’ labour market outcomes, including employment and earnings.
The cost-benefit analysis shows that participants and society benefit from all four of the programmes analysed – but not all are cost-effective from the perspective of the taxpayer (total government expenditures minus revenues). The two training programmes are relatively inexpensive and yield net benefits to the taxpayer within 14 months; the private sector wage subsidies also yield net benefits but lower cost-effectiveness due to higher associated costs, while the public works has a net negative effect from the taxpayer perspective.
While not cost-effective from the taxpayer perspective, Slovenia’s public works programme has beneficial effects on health, mortality and social outcomes for a subset of participants. These include a significant decrease in the mortality risk for older men, a positive effect on the general health of younger men and a positive effect on the likelihood of living with a partner for several subgroups.
Wage subsidies and classroom-based training are positively contributing to Slovenia’s transition to a more digitally-intensive and green economy.
Similar types of comprehensive evaluations have the potential to shed light on the potentially wide‑ranging effects of ALMPs in other countries as well – including on social outcomes, health, and the green and digital transitions. Such analyses require linking detailed, individual-level administrative data, which exist in many OECD countries, and whose full potential for research purposes is increasingly being realised (albeit with uneven progress).
The wide‑ranging effects of active labour market policies in Slovenia
Active labour market policies (ALMPs) typically have the explicit aim to improve the employment outcomes of participants, but they can also potentially have more wide‑ranging effects that are important in determining whether policies are cost-effective. OECD countries are increasingly linking different sources of administrative data to examine the effects of ALMPs on a broad array of labour market outcomes, such as earnings and occupational mobility, as well as other outcomes such as social integration and health (OECD/European Commission, 2025[1]). This is illustrated by a recent set of counterfactual impact evaluations of Slovenia’s ALMPs conducted by the OECD, which examine a comprehensive set of outcomes.
The analysis of Slovenia’s ALMPs also highlights how cost-benefit analyses can help guide policy by pinpointing which programmes are cost-effective and for whom. Slovenia’s relatively low expenditure on ALMPs, which is also skewed towards public works, bolsters the case for ensuring that the scarce funds are allocated in a cost-effective way. More broadly, similar types of analyses in other countries can help inform their decisions to expand, modify or scale back programmes based on their effectiveness.
The analysis is based on counterfactual impact evaluations that identify the precise effects of specific ALMPs by accounting for what would have happened to participants had they not participated in the ALMPs examined. It is based on detailed, individual-level administrative data that tracks participants and otherwise similar non-participants over several years before and after their participation in the ALMPs.
All four major ALMP programmes boost employment and earnings for participants (OECD, 2025[2]; OECD/European Commission, 2025[3]). Evaluations of Slovenia’s classroom training, on-the‑job training, private‑sector wage subsidies, and public-sector wage subsidies (public works) programmes show that participants’ employment prospects and earnings improve significantly relative to similar non-participants (see Box 1 and (OECD, 2025[4]) for descriptions of these programmes). These gains are also durable – positive impacts on employment persist for at least four years after individuals first enter a programme, well beyond the intervention period. Importantly, the benefits extend beyond just getting a job: participants have lower inactivity rates and are less likely to leave the labour force or emigrate abroad in the years following participation.
The four broad groups of ALMPs examined in the impact evaluations are:
Wage subsidies: Temporary incentives for private‑sector employers to hire jobseekers by subsidising a portion of their wages (for 12 months).
Public works: Direct job creation programmes offering temporary jobs (around 12 months) in the public or non-profit sector for jobseekers, targeted towards the long-term unemployed.
Institutional training: Training courses provided by external training institutions to equip jobseekers with new skills or certifications to improve their employability, lasting on average 2 months.
On-the‑job training: Work-based learning placements with employers (usually lasting a few months) where jobseekers gain practical experience under a mentor’s guidance.
Source: OECD (2025[4]), Evaluating active labour market policies in Slovenia – Module 2: Reviewing and assessing the existing system of designing and implementing ALMPs, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module2.pdf.
The impacts are broad-based but vary by programme and target group. Positive effects on employment and earnings are observed for jobseekers of all ages, education levels and other characteristics, though the magnitude differs by programme and participant profile. On-the‑job training tends to yield larger and more sustained improvements – participants are about 15 percentage points (p.p.) more likely to be in unsubsidised employment within six months of starting on-the‑job training – whereas the public works scheme achieves a more modest boost of around 6‑9 p.p. by one year after participation (with the gains sustained thereafter through the four‑year period during which outcomes are tracked). Interestingly, older workers and those with only primary education experience smaller employment gains from ALMP participation than younger or better-educated peers (who already have higher re‑employment chances even without participating in ALMPs). These patterns highlight the need to tailor programmes so that all groups can fully reap the benefits of effective ALMPs while ensuring that the most vulnerable are also supported.
Cost – benefit analyses (CBAs) of ALMPs can support evidence‑based policymaking by allowing policymakers to comprehensively compare the benefits and costs of a specific programme or policy (Pregaldini, forthcoming[5]). They go beyond individual outcomes to assess overall value for money from different perspectives: individuals, taxpayers, and society (see Pregaldini (forthcoming[5]) and OECD (2025[6]) for details and definitions). CBAs can account for a wide range of effects, including changes in earnings, tax revenues, benefit payments, programme costs, and economic output. To ensure a fair comparison over time and across groups, CBAs discount the value of future benefits and costs. This comprehensive approach helps policymakers understand who gains from ALMPs and whether programmes deliver sufficient returns for the resources invested.1
The evaluations of Slovenia’s programmes show that all programmes deliver net gains for participants and society. From both the individual and societal perspectives, each of the four ALMPs generates a positive net benefit within four years of the intervention (OECD, 2025[6]). Participants in every broad category of programme enjoy cumulative net income gains (after taxes and transfers, but including earnings from subsidised employment), meaning they are financially better off than comparable non-participants. For instance, wage subsidies provide the largest average boost to participants’ earnings, followed (in order) by public works, on-the‑job training, and institutional (classroom-based) training. This partly reflects the fact that wage subsidies and public works directly subsidise employment and therefore immediately boost participants’ earnings. Similarly, when accounting for the additional economic output from higher employment, all four programmes yield a net social gain by the end of the four‑year follow-up period.
From an individual’s perspective, the gains are mostly due to increased gross earnings (Figure 1). This is tied to the positive employment effects of the programmes, which are present across all four programmes examined (OECD/European Commission, 2025[3]; OECD, 2025[2]). These positive employment effects also result in increased payments of income taxes and social contributions which slightly attenuate the gains from participation from the perspective of the individual. Interestingly, government benefit transfers slightly increase from participation in the two programmes subsidising employment because participants are more likely to become eligible for unemployment and parental leave benefits (both tied to prior employment history). The rich administrative data enable the detailed disaggregation, which the analysis produces in a relatively straightforward manner. In contrast to alternative approaches (such as simulations based on theoretical modelling of different types of households), the disaggregation relies on few additional assumptions and is comparatively precise.
Disaggregation of estimated cumulative net benefits at 48 months after programme start from individual perspective, in 2022 EUR
Note: “Transfers” include unemployment benefits, social assistance and parental leave benefits. “Increase” refers to outcomes of ALMP participation that raise benefits from the perspective of individual participants, while “decrease” refers to outcomes that reduce benefits. Estimates depict the net effect of participation after accounting for counterfactual outcomes (see original source for details).
Source: OECD (2025[6]), Evaluating active labour market policies in Slovenia – Module 3: Cost-benefit analysis of ALMPs, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module3.pdf (Figure A A.2.), based on individual-level data accessed at the Statistical Office of the Republic of Slovenia.
Taxpayer net benefits vary greatly by programme. In contrast to the uniformly positive outcomes for participants, the fiscal net benefits for the government differ across programmes. The fiscal cost-benefit analysis takes the perspective of the taxpayer, summing the positive income tax revenues associated with participation in a programme, accounting for any possible net reductions in social transfers associated with participation, and subtracting the increased expenditures associated with participation (including direct ALMP costs and increased eligibility for unemployment benefits), examining the effects four years after entry into a programme.
Institutional training and on-the‑job training are relatively inexpensive and yield meaningful employment impacts, translating into high fiscal returns: about EUR 2.7‑2.8 in additional revenue and reduced expenditures for every euro spent, with a breakeven point roughly 9‑14 months after the start of the programme.
Wage subsidy programmes also pay for themselves from the taxpayer’s perspective, but due to the higher upfront subsidy they have lower cost-effectiveness of about EUR 2.3 in additional revenue and reduced expenditures for every euro spent.
By contrast, the public works job creation programme remains a net cost to the government even four years after participation – the increases in income tax revenue and social contributions associated with participants’ higher employment are not sufficient to offset the programme’s considerable costs within the four‑year observation window.
Cumulative taxpayer net benefits per participant, in 2022 EUR
Note: Estimates depict the net effect of participation after accounting for counterfactual outcomes (see original source for details).
Source: OECD (2025[6]), Evaluating active labour market policies in Slovenia – Module 3: Cost-benefit analysis of ALMPs, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module3.pdf (Figure 4.2), based on individual-level data accessed at the Statistical Office of the Republic of Slovenia.
Although the cost-benefit analysis finds public works are not cost-effective from the taxpayer perspective, additional analyses show that Slovenia’s public works programme has some positive effects on social and health outcomes (OECD, 2025[7]). However, these effects are limited to specific groups and are sometimes too small to be precisely measured despite the relatively large sample sizes and long observation periods. Overall, the participants and “comparable non-participants” have broadly similar health-index values, hospitalisation rates, and medication usage over the eight‑year follow-up period, although the public works participants do have slightly lower mortality rates over some time horizons. The evaluation finds modest improvements on aggregate outcomes for certain social outcomes for public works participants – most notably, in terms of living with a partner (e.g. spouse or cohabiting partner). Participation in public works is also found to decrease the probability of social assistance receipt, although the effects are relatively modest compared to the much larger effects on employment described above and in OECD (2025[2]).
In contrast to the relatively modest effects for the general population, certain subgroups of the population experience more significant effects. Specifically, participation in public works is found to have:
a strong effect on decreasing the mortality rates of men over 50 years old;
a positive effect on the general health of younger men and those deemed directly employable, as proxied by the composite health index, and as reflected, respectively, in lower rates of medication usage and hospitalisations;
a positive effect on the likelihood of living with a partner for several subgroups of individuals, including men under 30 or over 50; and
a moderately-strong effect on decreasing reliance on social assistance benefits for groups such as those with lower levels of education and women under 30.
Slovenia’s green and digital transitions are well underway. Between 2015 and 2022, Slovenia increased the share of employment in green and digitally-intensive occupations, as measured using two task-based taxonomies developed by the OECD that are mapped to the occupation codes (OECD, 2025[8]). The overall stock of green jobs – those contributing to environmental sustainability – grew by about 3% in this period. Digitally intensive occupations also expanded, with particularly strong gains in occupations requiring skills such as digital content creation and computer programming.
Unemployed jobseekers who participate in Slovenia’s private‑sector wage subsidy and training programmes move into greener and more digitally intensive occupations even though these ALMPs were not explicitly designed for that purpose.2 The counterfactual impact evaluation results show that wage subsidies and classroom-based training have a significant positive effect on participants’ likelihood of securing employment in greener occupations (see OECD (2024[9]) for definitions), especially in “skills-enhanced” roles – occupations that have existed for some time but whose tasks and skill sets are evolving as part of the net-zero transition. At the same time, wage subsidies demonstrate the strongest positive effects in helping jobseekers move into occupations requiring digital skills. Somewhat surprisingly, the effects of training are not as robust: training programmes also show some positive and statistically significant effects but only across certain sub-indices. A final point relates to a small-scale ALMP dedicated to temporarily subsidising “green “jobs” for the unemployed: it had too few participants to be evaluated as part of the impact evaluation but the analysis finds it shows promise in directing participants into green jobs.
The evaluations presented in this policy brief highlight the potential for counterfactual impact evaluation to inform policymaking on ALMPs in OECD countries. They are invaluable for identifying the precise effect of a policy after accounting for the multitude of potentially confounding factors that affect individual outcomes. Furthermore, incorporating additional information on costs enables comparisons between policies through cost-benefit analyses, which can be used for setting policy priorities that consider returns on expenditures, helping guide the targeting of specific programmes, and communicating the usefulness of policies and government expenditures to the public.
Broader cross-country work conducted by the OECD and European Commission on the topic of ALMP impact evaluations has highlighted some of the potential factors which can contribute to facilitating access to data, carrying out counterfactual impact evaluations and using them to inform policymaking (OECD/European Commission, 2025[1]):
investing in high-quality, modern data infrastructure that enables data to be linked across registers;
establishing clear legal frameworks for data access, linking data and using them for policy analysis, evaluation and research purposes;
mandating regular counterfactual impact evaluations, including as conditions for funding ALMPs;
embedding monitoring and evaluation frameworks into policy design;
providing formal evaluation guidelines or establishing dedicated government units tasked with policy evaluation; and
communicating evaluation findings by tailoring dissemination of results to different target audiences, including political leaders, policymakers, and the broader public.
Redirect funding to cost-effective ALMPs. Slovenia’s wage subsidies and on-the‑job training show strong net benefits, whereas public works is costly and least cost-effective. Unless ALMP spending increases from its currently relatively low levels of funding, scaling back public works and redirecting funds to high-impact measures would improve outcomes.
Target programmes to jobseeker groups based on effectiveness. In Slovenia, older jobseekers benefit most from the public works programme, which is the costliest intervention of the four groups of ALMPs examined. Accordingly, public works should be reserved as a temporary, last-resort inclusion tool for older long-term unemployed, while expanding training and wage subsidies for youth, low-skilled and other groups requiring additional support.
Recognise and enhance ALMPs’ health and social integration functions. As shown in the evaluations, ALMPs have the potential for positive effects outside narrow labour market goals: for instance, public works participants in Slovenia see reductions in social assistance dependency and are more likely to live with a partner, reinforcing social inclusion. Building on these spillovers, policymakers should consider recognising these additional effects as an explicit goal and monitor their effectiveness along these dimensions. They may also consider integrating support services (e.g. counselling and in-work support) into programme participation to boost participants’ well-being and sustain better labour market outcomes.
Align ALMPs with green and digital transitions. The evaluation for Slovenia shows that increasing ALMP spending would allow more jobseekers to benefit from effective programmes that also advance the green and digital transitions. To ensure that participants gain skills for a sustainable and digital economy, policymakers should consider adapting existing measures – for example, by directing wage subsidies toward green jobs and embedding digital or green skill modules in training – or develop dedicated programmes explicitly supporting these goals. Placements into these types of jobs are also more likely to result in durable, stable jobs.
Institutionalise regular counterfactual impact evaluations and cost-benefit analyses. The evaluations conducted for Slovenia demonstrate the value of linking administrative data to measure programme effectiveness across labour market, social, and fiscal outcomes. Policymakers should make such evaluations a routine part of ALMP design and monitoring, particularly as programmes evolve to meet green, digital, and demographic challenges. Embedding a culture of evidence‑based policymaking will help ensure resources are directed toward what works and for whom.
Establish a clear legal framework for accessing administrative data for research purposes. Harnessing the potential of administrative data to inform policymaking requires enabling researchers and policymakers to access administrative data, while ensuring that appropriate safeguards relating to confidentiality and the intended use of the data are in place. Slovenia’s current legal framework prevents public institutions, such as the Ministry of Labour, from linking important sources of administrative data for research purposes. Enabling such access would be in the public interest to facilitate evidence‑informed policymaking.
OECD-EC project on policy impact evaluation through the use of linked administrative and survey data project web page: https://www.oecd.org/en/about/projects/oecd-ec-project-on-policy-impact-evaluation-through-the-use-of-linked-administrative-and-survey-data.html.
References
[2] OECD (2025), Evaluating active labour market policies in Slovenia - Module 1: Impact evaluation of On-the-job training and Public Works on key labour market outcomes, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module1.pdf.
[4] OECD (2025), Evaluating active labour market policies in Slovenia - Module 2: Reviewing and assessing the existing system of designing and implementing ALMPs, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module2.pdf.
[6] OECD (2025), Evaluating active labour market policies in Slovenia - Module 3: Cost-benefit analysis of ALMPs, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module3.pdf.
[7] OECD (2025), Evaluating active labour market policies in Slovenia - Module 4: Analysis of social outcomes (those relating to social benefit receipt and health), https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module4.pdf.
[8] OECD (2025), Evaluating active labour market policies in Slovenia - Module 5: Analysis relating to the green and digital transitions, https://www.oecd.org/content/dam/oecd/en/about/projects/technical-reports-and-presentations-dg-reform/slovenia/Module5.pdf.
[9] OECD (2024), OECD Employment Outlook 2024: The Net-Zero Transition and the Labour Market, OECD Publishing, Paris, https://doi.org/10.1787/ac8b3538-en.
[1] OECD/European Commission (2025), “Counterfactual impact evaluations of active labour market policies: Lessons from using linked administrative data”, OECD Publishing, Paris, https://doi.org/10.1787/bea7860e-en.
[3] OECD/European Commission (2025), Impact Evaluation of Wage Subsidies and Training for the Unemployed in Slovenia, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/47098a5e-en.
[5] Pregaldini, D. (forthcoming), The OECD framework for cost-benefit analysis of active labour market policies, OECD Social, Employment and Migration Working Papers.
Contact
Theodora Xenogiani (✉ Theodora.Xenogiani@oecd.org)
Anne Lauringson (✉ Anne.Lauringson@oecd.org)
Matija Vodopivec (✉ Matija.Vodopivec@oecd.org)
Damiano Pregaldini (✉ Damiano.Pregaldini@oecd.org)
Ioanna Pantelaiou (✉ Ioanna.Pantelaiou@oecd.org)
← 1. Several caveats regarding the cost-benefit analyses should be kept in mind (see OECD (2025[6]) for details). Outcomes are tracked only for four years, so longer-term effects (costs or benefits) may fall outside the window. The calculations do not take into account broader social outcomes such as social inclusion and health (these are examined separately in the wider project, including a dedicated module on Public Works). Finally, the baseline estimates assume the absence of deadweight and substitution effects – that is, they assume that all employment gains generated by these policies would not have occurred in the absence of the intervention.
← 2. The effect of Slovenia’s Public works on the green and digital transitions was not examined in the evaluation.