This chapter sets out a proposal for procuring intensive job placement support for vulnerable jobseekers in Greece through contracted-out services. It reviews the groups most at risk of long-term unemployment, drawing on legal provisions and the data analysis in Chapter 3. It then examines the income support, ALMPs, digital tools and specialised services DYPA already deploys for its vulnerable jobseekers, highlighting where current provision falls short of their needs. Discussion turns to how intensive support can be contracted out in practice. It highlights elements of procurement and payment models, from outcome‑based contracts to grey-box approaches. Safeguards such as minimum service standards, risk-adjusted payments and robust data-sharing are also discussed. The chapter concludes by assessing institutional and legal preconditions in Greece and outlining a pilot model for contracted-out intensive support that draws on international best practice and could be scaled up over time.
Strengthening Individualised Support for Jobseekers Furthest from the Labour Market in Greece
Abstract
This chapter outlines a proposal for procuring intensive job placement support from external providers to support vulnerable jobseekers who face complex barriers to employment. The model aims to incorporate international best practices, which include outcome‑based contracting, employer engagement strategies, and multi-professional support to improve labour market integration outcomes.
The proposal draws on lessons across OECD countries that have implemented similar contracting models for employment services for vulnerable jobseekers. The key focus is on proposing a procurement framework that ensures quality service delivery while being realistic given Greece’s labour market, institutional and legal contexts. A pilot implementation of the framework is outlined, to allow Greece to test the implementation of new contracted services before scaling up a new model.
DYPA, the Greek public employment service (PES), provides services to a diverse set of clients, with differing needs for employment support. Greek legislation provides a legal provision for the identification of specific groups of individuals with discrete vulnerabilities or special needs (Government of the Hellenic Republic, 2011[1]). But alongside this definition of vulnerability, the data analysis in Chapter 3 has identified the characteristics that make individuals more prone to experiencing long-term unemployment.
The data analysis identifies the main characteristics that are associated with becoming long-term unemployed. These are grouped along three main categories (Table 4.1). Those individuals with complex labour market histories, women and older jobseekers, and those with skills barriers are more likely to become long-term unemployed. These individuals therefore face a set of circumstances where additional intensive employment support to unlock barriers has the potential to reduce the chances of suffering from prolonged spells of unemployment.
Note: Analysis from Chapter 3 of the project. The data include all registrations from 1 January 2017 and starting before 31 December 2023
Source: OECD analysis of data from the Greek public employment service (DYPA) and ERGANI.
DYPA delivers a range of passive and active labour market policies, as well as social policy programmes tailored to the needs of vulnerable groups. For example, vulnerable jobseekers co-design an Individual Action Plan with specialised job counsellors to provide personalised re‑integration support towards employment. Additional psychosocial support aims to improve resilience and reduce additional barriers that can prevent engagement with job search activities. DYPA also operates interventions for this group, for example it has previously run subsidised employment programmes specific for vulnerable claimants (DYPA, 2022[2]). Unemployment allowances and a flat-rate benefit (EUR 718.5 every two years), available for those ineligible for regular unemployment insurance, support the income of vulnerable groups (Government of the Hellenic Republic, 1989[3]). In addition, DYPA provides social policy support such as subsidised tourism, childcare camps and offers book/entertainment vouchers. Digital services, such as MyDYPAlive facilitate access and include, for example, sign language interpretation. DYPA has also invested in the capacity of its staff to support vulnerable groups, including training programmes to upskill counsellors focused on vulnerable groups, and dedicated employment support in two model EKO offices for vulnerable individuals.
However, as the data analysis in Chapter 3 demonstrates, transitions into employment for vulnerable groups remain relatively low, and support to them could be enhanced. Previous research has also shown that some vulnerable groups, such as the long-term unemployed, are also less likely to participate in active labour market policies (ALMPs) which could help them into work (OECD, 2024[4]). In addition to this, public spending on ALMPs in Greece is lower than average in the OECD. This provides significant scope for further support to build a pathway towards employment.
A suite of extra support, to help vulnerable jobseekers at risk of long-term unemployment could help to overcome some of the barriers they face and help reintegration into the labour market. This note looks at some of the best practices seen across the OECD supporting vulnerable groups and how contracting out employment services can support broader PES efforts.
Vulnerable jobseekers often face complex and overlapping barriers to labour market integration. These may include long-term detachment from the labour market, low skills, care responsibilities, health conditions, limited work experience, or migration-related challenges. In such cases, standard employment services may be insufficient. Evidence from across OECD countries highlights the importance of tailored, multi-professional, and sustained support pathways that go beyond traditional ALMPs.
Existing practices in OECD countries help to shine a light on the types of targeted support that is offered to individuals with additional barriers to employment. An important step in the process of providing support for vulnerable individuals is to identify needs early, so that people can receive services tailored to their individual needs. This was discussed extensively in Chapter 2 of this project, and it will be critical that the recommendations there are applied to provide DYPA and job counsellors with the tools needed to identify individuals with support needs. The focus in this note is on the different elements of support that can be provided to vulnerable individuals, and how they are provided to them.
4.3.1. Helping those furthest from the labour market requires co‑ordinated steps towards employment
Recognising that vulnerable jobseekers often have multiple and complex barriers to employment means that employment might not be an immediately achievable objective for these jobseekers. The combination of needs means that individualised and tailored approaches are important to provide effective support to different individuals. Addressing different barriers also necessitates a gradual progression to the labour market, reliant on co‑operations between institutions to meet diverse needs, which may involve social, health, education, childcare, housing and others (OECD, 2021[5]).
Some OECD countries have made this recognition explicit and have designed services that build in a progressive approach to support. For example, in the Netherlands the PES has developed a specific concept of employment services for individuals with considerable barriers to employment. These services are designed in a step-wise manner, so the individual climbs a “participation ladder”. Job counsellors determine whether an individual has made sufficient progress to move towards the next step. There are six discrete steps which start when the participant has little physical contact outside of the home, and end in paid employment. Steps in between include being able to participate in structured activities, such as volunteering, and reaching the position to be ready to start work or training (Vodopivec, 2023[6]). Building progressively in this manner allows individuals to build confidence reaching milestones and gradually remove their barriers to work.
Given the high rates of long-term unemployment in Greece, programmes that recognise the challenges for individuals with complex barriers to find employment in the labour market and design progressive approaches to re‑engage individuals offer a promising avenue to explore.
4.3.2. Pro‑active outreach efforts are also needed to engage vulnerable individuals
To tailor services and offer a pathway back to employment, PES first need to be able to engage with jobseekers. These include both jobseekers actively registered with the PES, and those individuals further from the labour market that are not currently seeking employment but who might be able to with the right support to help them. An additional problem with providing support to vulnerable jobseekers is that they are more likely to fall into the second category and may not yet be registered with the PES. Research has shown that there are often sizeable proportions of the vulnerable population that have no contact with the PES (OECD, 2022[7]). Different OECD countries approach outreach in different ways.
In Estonia, a dual pronged approach to helping vulnerable individuals with disabilities has been implemented. Firstly, the payment of the work ability allowance was tied to registration with the Estonian PES, which incentivised individuals with a health issue to register with the PES. This led to a fivefold increase in the proportion of the registered unemployment with a reduced ability for work. This has been coupled with targeted information provision to inform individuals on what support is available to them. Initiatives such as mobile counselling also aim to reduce barriers to uptake of support among these groups (OECD, 2021[8]).
Bulgaria’s experience highlights the importance of proactive, tailored, and integrated approaches to engage and support vulnerable jobseekers. Services are particularly targeted to the inactive, those that live in remote areas, or face complex barriers to employment. Mobile Labour Offices (MLO) and Mobile Labour Office Workplaces are deployed to reach individuals in small or isolated settlements who might otherwise be excluded from support due to lack of access or awareness. These units provide jobseeker registration, counselling, and, in the case of MLOs, the full suite of labour office services (including training referrals and engagement with local employers) to help overcome supply- and demand-side barriers. Family Labour Consultants (FLCs) offer a family-centred model of support, recognising that labour market exclusion often affects entire households. These consultants tailor services to each family member, helping identify and activate individuals who may not seek support on their own initiative, such as parents with caring responsibilities or young people at risk of early school leaving.
A core feature of Bulgaria’s outreach strategy is the use of Roma and youth mediators and youth activators, employed by either the PES or municipalities. These community-based intermediaries conduct direct outreach and fieldwork, often in co‑operation with schools, local NGOs, and municipal authorities. They play a critical role in identifying hard-to-reach individuals (e.g. those not in employment, education or training (NEET), Roma, long-term inactive), motivating them to register, and offering initial counselling and referral. In some cases, they also use digital tools, such as social media, to increase visibility and engagement (OECD, 2022[7]).
In Greece, high numbers of inactive individuals outside of the labour market create strong outreach needs by DYPA to bring these individuals back to the labour market (OECD, 2024[4]). DYPA has made some outreach efforts digitally, including via its MyDYPAlive service, but it is important to recognise that vulnerable groups often have lower capacity to engage digitally (Lyons et al., 2019[9]), and Greece in particular suffers from poor digital outreach generally to people with disabilities and older individuals (Paidousi and Efstratoglou, 2020[10]). Contracted-out employment services provide scope to better leverage local networks and partnerships to reach out to vulnerable populations. Similarly, greater efforts to utilise the network of Community Centres in Greece to proactively refer more job-ready individuals to DYPA could help to enhance participation in ALMPs and reduce the coverage gap of employment services.
4.3.3. Integrated service delivery can help to provide comprehensive and holistic support
The integration of employment and social services through one‑stop shops or joint case management can facilitate access to a wider range of support and ensure that individuals with complex needs are given wraparound support that addresses multiple barriers. Several OECD countries have implemented programmes that integrate support to provide more holistic services to individuals.
Finland’s youth one‑stop-shops combine employment counsellors, social workers, psychologists, and healthcare professionals. These centres gained momentum as part of Finland’s response to the COVID‑19 crisis, especially in strengthening mental health support for young people. The design of the centres puts young people at the heart and includes young people working directly at the centre, to encourage and re‑assure centre visitors (Dromundo, Lüske and Tuccio, 2023[11]). The centres are locally tailored and offer differentiated services dependent on needs. For example, some have strong links with local chambers of commerce, NGOs and occupational health services. There is central funding for the centres, alongside local funding dependent on municipalities. Joint service provision enables staff to address multiple barriers in a single‑setting, and the design is intended to lower barriers for young people to register for important employment and social services that they may otherwise not engage with (OECD, 2024[12]). An evaluation of the effectiveness of three youth one‑stop-shops estimated annual benefits of the centres’ operation to both government finances (EUR 340 000) and the broader national economy (EUR 34 000 000), though noting its uncertainty due to numerous assumptions in the calculations. These positive effects result from better referrals to ALMPs, and faster job entries of service users (Valtakari et al., 2020[13]).
France has also demonstrated the efficacy of integrated support in its implementation policies initiatives to support vulnerable people. L’Accompagnement global is a French ALMP co-financed by the European Social Fund (ESF), designed to support people furthest from employment through intensive, personalised support. The initiative pairs a PES employment counsellor with a social worker for jobseekers facing multiple barriers (France Travail, 2025[14]). The programme targets long-term unemployed individuals, minimum income recipients, the low-skilled, single parents, young people, and residents of disadvantaged neighbourhoods. It delivers integrated pathways that combine social support (such as housing, childcare, financial management and healthcare access) with employment-related services, including job counselling, training, and employer engagement. The programme aims to reduce the time taken to access support; ensure jobseekers have information about and access to all the social support that is available from other services; improve co‑ordination between France Travail and social workers; and establish shared operational leadership to drive forward innovation (Pôle Emploi, 2023[15]) Delivered by a diverse network- including departments, local integration and employment plans (Les Plans locaux pluriannuels pour l’insertion et l’emploi), Pôle Emploi, and non-profit associations- l’accompagnement global uses a case management model, often with a dedicated advisor co‑ordinating cross-sector interventions. An impact evaluation found the programme was successful in reaching priority groups and generating additional employment outcomes. Counterfactual analysis estimated an additional 27 percentage point increase in employment probability six months after participation, relative to similar individuals not supported by the programme. These effects were particularly pronounced for those facing multiple employment barriers (France Travail, 2018[16]). The average cost per additional job was approximately EUR 3 000, suggesting good cost-effectiveness given the vulnerability of the target population. The evaluation also noted the value of co‑ordinated delivery and intensive case management, though highlighted underuse of employer engagement and limited innovation in delivery models (Amnyos/Edater, 2019[17]).
Korea has also made significant strides in delivering integrated and person-centred employment support through its Employment Welfare Plus Centres (EWPCs). These one‑stop shops bring together employment counsellors and social workers from multiple institutions, enabling comprehensive support for individuals facing multiple and complex barriers to employment. EWPCs serve as hubs where clients can access a broad range of services in a co‑ordinated manner – combining job search assistance, career counselling, and placement support with social services such as housing support, healthcare access, and financial guidance. Staff from different organisations- including Korea’s Employment Centres (under the Ministry of Employment and Labour), local governments, NGOs, and other ministries- work under one roof to provide streamlined services tailored to individual needs. The model is particularly notable for its focus on vulnerable groups, including youth, older jobseekers, individuals with disabilities, and mothers returning to the labour market. Specialised counsellors are assigned to these groups, and EWPCs aim to offer in-depth, personalised support that addresses both employment-related and social challenges. The integration of services improves co‑ordination, reduces fragmentation, and simplifies access for jobseekers who might otherwise struggle to navigate complex support systems.
Despite regional variation in implementation quality, evidence suggests that EWPCs have had positive impacts on employment outcomes. Research shows that the model has increased the reach and effectiveness of active labour market policies in Korea, especially by providing equitable access and building synergies between employment and welfare services (OECD, 2024[18]). Ongoing improvements, such as clearer co‑operation frameworks and dedicated inter-institutional teams, are expected to further enhance the impact of this approach.
Korea’s integrated service delivery approach illustrates how employment and social services can be jointly mobilised to support labour market inclusion. The model is grounded in central co‑ordination with local adaptability, and Korea is continuing to invest in strengthening collaboration practices and evaluation systems to inform future refinements.
While integrated service delivery can improve outcomes for vulnerable jobseekers, international experience demonstrates that a wide variety of implementation models exist. These range from formal one‑stop shops with co-located multiprofessional teams (e.g. Finland) to more informal or virtual co‑ordination platforms (such as in Spain’s Basque region) (OECD, 2021[5]). Each approach has trade‑offs.
Fully integrated centres can reduce access barriers and promote holistic support but require significant structural reform and investment. By contrast, models built around joint case management and voluntary collaboration can offer a more fluid and pragmatic path to integrate services, such as in France’s l’accompagnement global. These rely on close co‑ordination between PES counsellors and social workers to support individuals with complex needs, without the need to overhaul institutional frameworks.
DYPA has the potential to make relatively small adjustments to existing operating model which would allow it to offer similar services. This would require some additional funding to support operations. DYPA is actively collaborating with social enterprises and civil society organisations through memoranda of co‑operation to support the labour market integration of vulnerable groups and currently has nine such partnerships. These partnerships focus on skills development, vocational training, counselling, and awareness-raising. With organisations like the Margarita special education laboratory, ELEPAP (a society in aid of people with disabilities), and the Hellenic Federation of the Deaf, DYPA supports people with disabilities by providing tailored training and employment services, exchanging expertise, and promoting inclusive hiring practices. Through its partnership with EPANODOS, DYPA helps prisoners and ex-prisoners reintegrate into the labour market via vocational workshops, counselling, and employer engagement. In collaboration with KETHI (the Research Centre for Gender Equality), DYPA works to address gender inequalities in employment by empowering women through training, entrepreneurship support, and gender-sensitive counselling.
DYPA has already made some progress to increase outreach for some groups. Its two specialised employment centres for vulnerable and special groups in Athens and Thessaloniki aim to provide tailored services to vulnerable claimants, in collaboration with specialised third sector partners (DYPA, 2024[19]). This already provides a small footprint from which to build further specialised, collaborative service centres for vulnerable groups.
The collaborations with social enterprises demonstrate DYPA’s existing commitment to inclusive employment policies and its use of social partnerships to reach disadvantaged populations, but they lack comprehensive integrated employment support. Meanwhile, the two specialised vulnerable group centres offer the promise of better integration of employment support alongside more holistic social and health support, but are currently limited in scale.
Given DYPA’s existing service infrastructure and evolving partnerships with social services, a model closer to the French example could be easier to implement and build upon its existing strengths. This would enable targeted, intensive support for clients facing multiple barriers while building capacity incrementally through stronger referral systems, shared case management protocols, and clearer inter-agency leadership.
Giving greater consideration to how more intensive employment support can be woven directly into the partnerships with the social enterprises could allow DYPA to provide a greater support net and development framework to support these vulnerable individuals to progress towards work. For example, closer co‑operation with the Community Centres could enable greater cross-referral between the two the organisations. Where DYPA identify clients with social service needs, they could refer them to the Community Centres. Greater provision for information sharing and joint working may also allow for informal joint case management, so that the diverse needs of clients could be met through joint action in the two organisations. Enabling these practices would require further investment in staff resources and operational join-up (for example on access to shared case information) to put in place the support required, but would then leverage the best of both organisations in providing a holistic service to individuals.
4.3.4. Involving social economy actors can provide a bridge to unsupported employment
Involving broader social economy actors can help to diversify support, and offer vulnerable jobseekers a stepping stone towards unsubsidised employment. Several OECD countries have implemented models where vulnerable jobseekers are matched with varying levels of support based on their distance from the labour market and the social economy is directly integrated into service delivery.
Austria operates a tiered model through partnerships with social enterprises. Jobseekers furthest from the labour market are offered subsidised employment within the social economy, accompanied by training and mentoring (Hiebl, 2020[20]). Those closer to employment may receive temporary job placements via NGOs or direct employment subsidies. Across all tiers, integrated counselling and support services are provided. Evaluations show this approach significantly improves employment outcomes, especially for women, older workers, and people with disabilities (OECD, 2021[5]).
France’s “Territories of Zero Long-Term Unemployment” (TZCLD) is an innovative programme that puts local stakeholders at the centre of a suite of support to reintegrate the long-term unemployed. At a national level, the social partners lead an association to oversee operational delivery, whilst government, local authorities and the social partners control overall project financing and the approval of territorial participation. Local employment committees in each of the participating territories bring together all the voluntary stakeholders and are chaired by the local elected representatives.
TZCLD provision allows for the formation of social enterprises aimed at providing jobs for individuals to engage in socially beneficial activities. The programme aims to create “additional” jobs rather than replacing or substituting existing ones, and a suite of evaluation is planned using linked administrative data to conduct counterfactual impact evaluation to determine the programme’s success (DARES, 2024[21]).
Job carving is the process of adapting jobs to ensure that specific job tasks are suitable for employees to carry them out. Working with employers to adapt jobs can hold promise in opening up employment for people with disabilities, who may have specific needs or capabilities. It is a relatively resource intensive process, as it involves counsellors or labour experts working jointly with employees, to review how specific roles may be adapted to specific jobseekers. Malta has introduced job carving through a partnership between the PES and the Lino Spiteri Foundation (LSF), a third sector organisation dedicated to helping to improve the employability of those people facing long-term unemployment (European Commission, 2018[22]). The LSF approaches the issue holistically where a programme of job coaching for disabled people at the workplace supports on-the‑job learning, integration and independence, and is inter-twined with job carving via constant feedback loops, ensuring that as individuals progress in their jobs, the job tasks also evolve adapted to the individual’s needs (Lino Spiteri Foundation, 2024[23]).
Through the use of tender specifications in potential contracted-out employment programmes, DYPA can encourage bidders to make use of links with social actors to build re‑integration pathways with their participants.
4.3.5. Targeted mental health support can help individuals to re‑connect with the labour market
Mental health remains a significant factor limiting the employment prospects of many jobseekers. Across the OECD, people with mental health condition face a 20% employment gap and a 17% wage gap (on top of the employment gap), and at any point in time, about 20% of the population experiences a mental health condition (OECD, 2021[24]). Effective ALMPs that integrate mental health support help individuals to overcome barriers that limit their labour market engagement.
Vulnerable groups identified in Greece’s legal provisions include those with mental health conditions, meaning that there is already recognition of their increased levels of need. Despite this, it is difficult with existing administrative data to precisely identify whether an individual has a mental health condition. A survey of jobseekers in the city of Piraeus showed that two‑thirds of jobseekers (63.8%) presented elevated levels of depression, and the risk of presenting these symptoms was greater for those with longer durations of unemployment and greater levels of economic distress (Latsou and Geitona, 2018[25]). Whilst it is not possible to extrapolate directly from the survey results in Piraeus to Greece as a whole, the particularly high levels of mental health challenges in Piraeus suggest there is potentially a large demand for services to address these issues at a national level.
Targeted mental health support can help individuals overcome health barriers before they begin to build confidence towards (re‑)entering the labour market. A promising specialised approach is the Individual Placement and Support (IPS) model, designed for people with serious mental health conditions. IPS integrates mental health and employment support services, prioritises rapid job placement, and has been validated in 26 randomised controlled trials (IPS Employment Centre, 2025[26]). It is being scaled up in Australia (through headspace centres) and Flanders, where the NGO GTB is adapting IPS for clients with mental health conditions referred by the PES.
Innovations to the IPS model are also applied to further test how its efficacy can be even more strengthened. In Australia, a pilot has recently been announced in six headspace centres to introduce peer workers to provide work and study support to young people. These peer workers have previously encountered their own mental health challenges, and offer pastoral advice and support. They work with headspace staff, other health workers, and members of the local community to support young people in the centres (DSS, 2025[27]) A meta‑analysis of IPS has shown that it works across a variety of settings and economic conditions, and participants are more than twice as likely to enter competitive employment (Modini et al., 2016[28]).
Another programme integrating mental health and employment support is the JOBS programme. Originally developed in the United States, it combines structured job-search training with a strong emphasis on psychosocial resilience‑building to protect unemployed individuals from the adverse mental health effects of job loss. The standard intervention is typically delivered as a five‑day, 25‑hour group workshop (delivered in parts), led by trained facilitators. They are usually two facilitators comprising a professional trainer (for example, from employment services or psychology) and a peer trainer with lived experience of unemployment. The programme covers topics such as CV writing, interview techniques, goal setting, and networking, but also integrates cognitive‑behavioural elements. These include sessions to explore resilience against setbacks in job search, enhancing self-efficacy, and building social support within the group. A programme manual has been compiled that aids implementation of the JOBS programme in different settings (Curran, Wishart and Gingrich, 1998[29]).
The JOBS programme has been implemented across several OECD countries and is generally associated with results that improve both labour market and health outcomes ( (Hollederer and Jahn, 2023[30]; Barry et al., 2006[31]; Vuori and Silvonen, 2005[32]). Though the impacts are not universally positive and some studies find no evidence of improved labour market attachment (Carlier, Schuring and Burdorf, 2017[33]), and in the United Kingdom even where positive labour market outcomes were present the costs of delivering the programmes can offset the benefits (Rayment et al., 2021[34]). An interesting recent feasibility study in Romania looked at implementing the JOBS programme online and was positive in its assessment, though with the caveat of higher drop-out rates for participations (36.1%) relative to face‑to-face delivery (Bodnaru et al., 2023[35]).
The evidence demonstrates the potential for programmes which aim to address underlying mental health conditions at preventing mental ill-health and at aiding individuals to reconnect with the labour market. Given the potential demand for such services for jobseekers in Greece, a programme of support that offered similar services could improve jobseekers’ engagement and success in the labour market.
4.3.6. Post-placement follow-up support can help to cement labour market attachment
Many vulnerable jobseekers require continued support even after securing employment. Evidence shows that early job exits can occur when counselling promotes early initial employment which then subsequently dissipates as transitions out of employment occur (Cottier et al., 2018[36]). Post-placement support can help to support individuals to remain in employment by offering advice and guidance on job retention and development.
In many OECD countries, post-placement follow-up support is being used to sustain and encourage the labour market integration of jobseekers that have recently secured employment. In Germany, nationwide post-placement counselling has been in place since 2016, following successful pilots. The services focus on the critical first six months of employment, helping reduce contract terminations among vulnerable groups. Post-placement support is tailored to individual circumstances and can include continued counselling on training opportunities, information provision on wider services (e.g. childcare facilities), help with mobility (e.g. transportation costs), or advice on job contract extensions (Staible, 2017[37]). In Estonia, follow-up support is provided specifically to vulnerable groups. Individuals who are at risk of cycling back into unemployment, including those with health issues, limited work experience, or previous spells of long-term unemployment, are given follow-up support from case managers for six months after entry into employment (OECD, 2021[8]).
Post-placement follow-up support may also be integrated into contracted-out employment services. For example, the United Kingdom explicitly built post-placement follow-up support into its Work Choice programme. The programme aimed to help jobseekers with severe and complex disability-related barriers to employment enter unsubsidised employment. Two phases in its three‑phase design were explicitly designed to implement post-placement support. The first step towards employment was via supported employment (if a participant entered regular employment, they left the programme). When an individual entered supported employment, they left the first phase (work entry support) and moved into phase two. In this phase, providers kept in contact with both participants and employers to review development plans, progress in the job, and check adequate support was in place for participants. Providers worked with participants to ensure at least eight hours of development activity towards unsupported employment took place each month. Phase two could continue for two years. After this (and dependent on the participant not entering regular employment), participants entered phase three. Here providers continued engagement for at least four hours per month, and this module had no durational limit. The programme was operated as a “black box” and outside of the broad phase constraints, providers were free to work with participants with content and structure as they saw fit. The programme was a success. Evaluation showed that eight years after referral, participants had a payrolled employment rate 10.9 percentage points higher than non-participants and had spent all a whole year extra (337 days) in payroll employment. For every euro invested, the government gained EUR 1.67, and society gained EUR 2.98 (DWP, 2025[38]).
Even in contracted-out programmes that do not explicitly build follow-up into programme design, contracts can be designed to encourage follow-up with participants. For example, the Work Programme in the United Kingdom encouraged providers to offer post-placement follow-up support via the use of contracts which paid out bonuses for sustained employment of participants (Finn and Peromingo, 2019[39]). Careful contract design can encourage supportive behaviours from employment service providers, even where there are no obligations in the contracts to provide specific activities like post-placement support.
As the analysis in Chapter 3 demonstrated, long-term unemployment in Greece is both prevalent and persistent and individuals that experience prolonged periods of unemployment are both immediately detached from the labour market and more likely to experience detachment in the future. Designing services to offer prolonged periods of follow-up support for vulnerable individuals in these circumstances could be a promising mechanism to encourage the growth of stronger links to the labour market. If DYPA were to offer such support internally it would be necessary to invest in additional capacity for additional staff, specialised training on the specific barriers for vulnerable claimants, and to build networks between these staff and the broader ecosystems of support within DYPA and externally.
4.3.7. The good practices in OECD countries provide fertile ground for DYPA to integrate new services within its existing delivery model
These elements offer a blueprint for the design of new contracted-out employment programmes in Greece. By embedding features such as integrated wraparound support, mental health and resilience‑building interventions, and follow-up support after job placement, DYPA can ensure that contracted services are responsive to the needs of those furthest from the labour market.
Contracting out works differs in important aspects in the implementation of some of the support mentioned in the preceding sections, which are delivered as part of in-house services. For contracted programmes, providers are often given freedom to work with participants in the way they feel most appropriate to achieve agreed objectives. DYPA will need to design the tender specification of its contracted-out programmes to ensure that the advantage of the features mentioned here are built into its externally delivered programmes.
For example, to leverage partnerships with other third sector partners DYPA can write tender specifications asking prospective providers to detail how they would work with other support services in the local area to deliver services for programme participants. By giving appropriate weight to this specification in the tender document, it provides scope for service providers to build in community and partnership to their delivery model.
Similarly separate service streams, or separate programmes, designed to support individuals with physical or mental health needs can ensure programmes are designed to integrate specialised and tailored support to address the challenges that these individuals face.
Designing contracts with outcome‑based payments and performance monitoring that encourages providers to engage with participants to ensure employment outcomes are sustained (through payments triggered after employment lasts a specified minimum period of time) can help to achieve the same goals that are sought in Germany, without specifying the precise means (and resource deployment) that are necessary to achieve these goals. Instead, providers determine how best to support their clients to sustain employment.
Utilising contract design in this way, well-designed contracted-out models can complement and extend DYPA’s existing service offer, while mobilising additional capacity and specialist expertise to improve employment prospects for vulnerable jobseekers.
This section of the note provides more information on how contracted-out services can be used by PES to supplement their own service offering. Contracting-out employment services is a well-established practice in the OECD countries, particularly for providing intensive job placement support to vulnerable jobseekers. Countries have adopted diverse models, shaped by institutional frameworks, labour market characteristics, and policy priorities. While specific designs vary, several core principles have emerged that underpin effective procurement and delivery.
4.4.1. Contracting out can facilitate innovation and flexibility to support diverse needs
Contracting-out employment services offer advantages in both service flexibility and innovation. In contrast to scaling-up internal service delivery, the use of external partners can allow PES to quickly scale‑up capacity to meet demand, allowing for efficient delivery of services for a set period of time or contract value.
For example, during the COVID‑19 pandemic, the Department for Work and Pensions (DWP) in the United Kingdom implemented a new contracted-out programme – the Job Entry Targeted Support (JETS) – specifically as a short-term measure to cope with the influx of claimants to unemployment benefits at the onset of confinement measures. This allowed it to rapidly scale up services without having to onboard significant new estate and personnel itself, instead using the capacity of existing private sector employment services. This delivery mechanism is of note to DYPA, which has its own capacity constraints related to caps in headcount. In this instance, contracting out could offer DYPA scope to increase capacity without having to circumvent protocol on headcount restrictions.
The use of external providers also confers important potential advantages to service innovation. By allowing external providers to shape their own services for jobseekers, it permits a greater diversity in approaches to meeting jobseekers’ needs. This is particularly the case for employment programmes with many different private providers, or for programmes with few specific requirements made on providers.
But innovation brings risks as well as opportunities. Providers that innovate and fail, risk leaving clients without jobs. Without appropriate incentive structures, providers may not be encouraged to take risks or work with certain clients. Therefore, the design of contracts for providers is key to mitigate risks whilst promoting innovation. When programmes are designed to encourage innovation by providers, there are typically two broad typologies for contracts, “black box” and “grey box”, which differ in the level of freedom providers have to autonomously set programme content.
In black box models, providers are granted significant autonomy to design and deliver services as they see fit, with minimal prescription from the contracting authority. While this approach encourages innovation and allows providers to tailor services to individual client needs, it also carries risks, particularly around delivery for harder-to-place clients. Evidence from Australia’s early employment services indicates that black box models can lead to “parking”, where providers focus on clients who are easier to place in employment and neglect those who are harder to serve. The corollary of parking is “creaming”, whereby greater efforts are placed on the more job-ready, so that providers can guarantee any contract incentives for quick placements. Both parking and creaming lead to inequitable service delivery for participants. For example, a study examining Australian unemployed workers’ experiences found that some clients received minimal support, with providers prioritising those more likely to achieve employment outcomes quickly. This led to some job-ready participants being placed quickly into sometimes unsuitable employment, with fewer hours, whilst other clients with greater barriers were left with little support (O’Halloran, Farnworth and Thomacos, 2020[40]).
To mitigate some of the concerns associated with black box models, some countries have introduced “grey box” models which seek to strike a balance between provider autonomy and prescribed service elements. Providers have flexibility in service delivery but must adhere to certain minimum standards or service components. (Langenbucher and Vodopivec, 2022[41]) For example, Sweden’s Rusta och Matcha programme exemplifies a grey box approach. The Swedish Public Employment Service (Arbetsförmedlingen) sets certain standards and guidelines, while allowing providers the flexibility to innovate and tailor services to individual jobseekers (OECD, 2023[42]). Similarly, in the United Kingdom, the DWP used the experience it has gained over years of operating contracted-out employment schemes to implement minimum service standards into its recent Restart scheme. This was designed to mitigate concerns over parking of difficult to place clients. The scheme is designed to encourage local flexibility and tailoring, whilst at the same time introducing customer service standards which stipulate minimum services requirements (such as one‑to‑one support a minimum of every 10 days and diagnostic assessment of the individual’s progress completed at least every four months) (DWP, 2025[43]).
When looking broadly across the implementation of contracted-out employment services, the large majority of contracted-out programmes stipulate minimum service requirements in one form or another (Langenbucher and Vodopivec, 2022[41]). Most often, these requirements are for elements such as individual action plans, frequent one‑to‑one meetings, in-work support, and client activity tracking (e.g. training or job applications). However, some programmes tie service expectations to performance indicators like placement rates or skills development.
Due to the relative novelty of offering contracted-out employment programmes in Greece and the persistence of long-term unemployment among many individuals it is suggested that any prospective programme design includes key elements of minimum service standards. This will help to avoid some of the potential pitfalls that can occur in outcome‑based performance contracts and ensure equity in service delivery among different individuals.
4.4.2. Designing contracting models to support differing individual and institutional needs
OECD countries employ a range of models for contracting out employment services, depending on the degree of risk transferred to providers, the outcome targets, and the mechanisms for innovation and accountability (Langenbucher and Vodopivec, 2022[41]). However, the common theme among these models is that they try to incentivise providers to support as many individuals as possible into work, at the same time ensuring that certain individuals are not under-served by providers.
Contracts can vary dependent on the degree of risk borne by the contracting authority and the provider. In outcome‑based models, providers are typically paid a small amount of their total possible fees upfront, with the remaining payments dependent on how successful participants are in the labour market. In other settings, where outcomes are uncertain and specific services are required, “cost-plus” contracts can be employed. In these contracts, payments are fixed and are determined based on the underlying costs of delivering the services in the programme.
Outcome‑based payment models are designed to incentivise performance
Outcome‑based contracting ties provider payments to specific results, most commonly sustained employment. This model is designed to incentivise providers to tailor services toward achieving long-term labour market integration. But alongside the “black-box” and “grey-box” considerations outlined above, good contract design is vital to promote support for all participants.
In many contracted-out employment programmes, contracts reward provider performance for better jobseeker attachment to jobs, and recognise that some individuals are harder to place than others via differential payments. Typically, jobseeker attachment is promoted via payments which increase as the participants’ subsequent employment tenure increases.
One way that countries encourage more support for vulnerable jobseekers is by using higher payment rates for individuals with greater needs. Australia has been a pioneer of outcome‑based contracting since replacing its public employment service with the Job Network in 1998. It pays providers based on sustained labour market outcomes and the degree of individual need. Workforce Australia is Australia’s latest contracted-out employment service, introduced in 2022. Providers are paid for employment outcomes at 4, 12 and 26 weeks. Payment levels for each of these outcomes are dependent on the individual’s characteristics. This is determined using the results from Australia’s JCSI profiling tool. Individuals profiled to have a high risk of long-term unemployment are paid at a higher rate for employment outcomes. Additional bonus payments are made if the employed individual was previously very-long term unemployed (over two years unemployment at point of job placement). Upfront “engagement” payments, and progress payments paid every two years provide minimum funding independent of job outcomes (DEWR, 2025[44]). Providers can also access a flexible Employment Fund to reimburse costs for client-specific support like transport or training (Langenbucher and Vodopivec, 2022[41]). Structuring payments in this manner attempts to incentivise providers to work more with more vulnerable jobseekers, reward better connection to jobs, whilst also protecting providers’ immediate ability to finance services for jobseekers.
The United Kingdom has followed a similar approach to Australia but has used its recent experience to experiment with contract design to better incentivise working with harder-to-place individuals. Its Work Programme (which operated between 2011 and 2017) utilised a similar outcome‑based model to Workforce Australia. Providers were incentivised to place clients into work and received larger payments for sustained employment (up to two years for harder-to-help groups). Over time, attachment fees (paid upon client entry) were phased out to increase focus on performance (DWP, 2020[45]). The introduction of its Restart programme in 2021 maintained an outcomes-based approach but included new features like decreasing monthly service fees over time, payment tiers based on accumulated earnings (rather than simply having a job or not) and an “accelerator” model of high payment fees for helping the harder to place. The appeal of the accelerator model is that payments are based on “cohorts” of participant intake. In this manner, individuals do not have to be individually characterised as high or low risk to generate different payment levels. Rather, the providers work with an intake cohort and once a certain proportion are in work, any participants that find work in addition to this are paid at a higher rate. It does not incentivise providers to work with one type of individual relative to another, but incentives high employment rates at the cohort level (DWP, 2025[43]).
But with every contract design, alongside opportunities, risks are also embedded due to the nature of the contract. For example, Sweden’s Rusta och Matcha programme uses contract design to encourage providers to reconnect individuals quickly with employers. “Speed premiums” incentivise fast placement by transferring unused service fee entitlements into bonuses (so that providers are paid all possible daily payments immediately on entry to employment). Alongside the speed premium, providers also receive a single outcome payment after four months of employment or entry into education. However, the speed premium has also led to concerns about cycling of clients into poor quality jobs with limited tenure (Langenbucher and Vodopivec, 2022[41]). Despite some of the concerns around the speed premiums, the inclusion of educational outcomes into the Rusta och Matcha contracting has helped broaden incentives beyond formal employment and avoid some of the potential short-termism in outcomes.
Cost-plus contracting can be used when it is hard to estimate outcomes or as direct payment for services.
Cost-plus contracts reimburse providers for agreed costs, plus a margin. This model can be used where outcomes are difficult to measure precisely but are not routinely used as part of contracted-out employment programmes. Cost-plus contracts are more commonplace for contracted-out health assessments, where there is no immediate link made to employment outcomes and payments are made for the cost of services rendered (for example in confirming eligibility for health-related benefits). In cost-plus contracts, providers report costs transparently under open-book rules and payments are then determined by these costs. For example, during the COVID‑19 pandemic, the Department for Work and Pensions (DWP) in the United Kingdom switched to cost-plus contracts for its employment programmes under emergency measures. This mitigated potential issues with payment by results contracts, where job outcomes would have been difficult to achieve. Providers praised this response as a means of safeguarding their cash flows, in otherwise testing circumstance, even in instances where upon reflection they could have made higher profits sticking to the original payment by results contracts (DWP, 2024[46]). A similar move to cost-plus contracting was also undertaken for two contracted-out employment programmes in New York during the pandemic (Langenbucher and Vodopivec, 2022[41]).
During the pandemic, DWP introduced JETS to specifically to cope with 1 million additional unemployment benefit claimants. The scheme ran from October 2020 to September 2022. JETS was contracted on a cost-plus basis (costs +5%), due to uncertainty on ongoing volumes and on the expectation for entries into work due to the sanitary measures in place (DWP, 2025[47]). It was designed for individuals unemployed for at least 3 months and offered 6 months of employment support services to claimants who were judged to be capable of moving relatively quickly into work.
These initiatives demonstrate how contract types can be adjusted to ensure contracted programmes are mindful of the labour market conditions, and work to ensure effective and targeted support to jobseekers.
4.4.3. Contracting authorities create market structures and provider competition
The structure of markets, or quasi-markets, in contracted-out employment services varies a lot in scope and is determined by the contracting authority that creates the programmes. These markets can range from a small number of large providers that are assigned via contract tenders, to permitting a large number of providers based on defining licensing or provider accreditation rights. Market structures bring different advantages and disadvantages that help to determine quality levels and variation. For example, a smaller number of large providers might allow the contracting authorities to manage performance and delivery in a more streamlined fashion, but also make it harder to provide true client choice and may limit the innovation that comes from more flexible and nimble independent providers.
An example of a provider market which makes use of fewer, larger organisations to deliver services is the “prime provider” model often used in the United Kingdom. In this model, large organisations are contracted to deliver services across broad geographic areas, often managing a network of subcontractors. The United Kingdom implemented the prime provider model in its Work Programme – a contracted-out employment programme running between 2011 and 2017 for long-term unemployed individuals. Prime providers were responsible for delivering services and managing subcontractors within designated Contract Package Areas (large geographical delivery regions). This model aimed to ensure accountability and efficiency but raised concerns about reduced competition and potential monopolistic behaviour Even within prime contracting models, attempts can be made to introduce choice, by allowing more than one prime provider to operate within a contract area. But evidence shows that even when this process is implemented it does not always lead to higher quality and better choices by participants (Ingold and McGurk, 2023[48]).
An alternative to models where large incumbent providers offer services across contractual delivery periods, is to place much greater power for individuals to choose providers via the provision of vouchers. Voucher systems empower jobseekers to choose among accredited service providers, fostering competition based on service quality and outcomes. In the Netherlands, individuals receive vouchers from the municipalities or the PES office and can choose between thousands of different providers. This enables a vast degree of claimant choice, though it has also generated concerns over transparency of quality and the ability of jobseekers to make informed decisions (Langenbucher and Vodopivec, 2022[41]). Sweden’s recent reform of its PES includes elements of voucher-based competition, where jobseekers can choose among various service providers approved by Arbetsförmedlingen. This approach is designed to increase efficiency and tailor services to individual needs.
Authorities have choices on how to influence price‑setting decisions
Alongside determining the particular contract modalities and payment-by-results incentive structures, contracting authorities may also choose to provide information to tendering organisations on the expected pricing levels of the bids. Where contracting authorities provide price guidelines (a “lead price”), they are used to provide a benchmark for which providers can depart (for example by offering lower prices). The premise is to avoid unrealistic tender bids by providers that may over-estimate potential performance, and therefore lead to financial impairment (and provider market exit) in a performance‑based contract setting (Langenbucher and Vodopivec, 2022[41]).
A pre‑requisite for lead prices is the ability to accurately estimate expected outcomes. In this context, providing bidders with comprehensive information on how expected outcomes and performance levels have been calculated is important to allow to assess how potential programme payments might cover their expected costs and offer a sufficient return on investment for them to proceed with a bid.
An alternative approach is to leave the price‑setting completely to the market. In this instance, bidders include their pricing estimates as part of their tender bid. This approach has been used in JobPath in Ireland and in France’s Private Placement Operators pilot. The advantage of doing this is that it does not rely on estimates of expected outcomes from the contracting authority. In some instances, if the expectations are unrealistic, lead prices can lead to a situation where tenders end without a bid (Langenbucher and Vodopivec, 2022[41]). By contrast, letting the bidders set their own price can help to avoid this, and encourages cost-efficiency in the bid process.
Tender assessments can be determined on a range of factors
The procurement process for contracted-out employment services varies across OECD countries, with the design of tender procedures typically reflecting the complexity, scale, and referral mechanism of the programme. Where providers are guaranteed a minimum quantity of referrals, tendering tends to involve more rigorous and multi-stage procedures. In contrast, voucher-based models, where jobseekers select their provider directly, often rely on simpler, one‑stage accreditation processes (Langenbucher and Vodopivec, 2022[41]).
In voucher-based systems such as Italy’s Assegno di Ricollocazione (reintegration vouchers), providers are accredited through either regional or national processes. Accreditation is based on eligibility checks, such as organisational capacity and minimum service quality standards, rather than competitive bids. Public employment offices are automatically eligible, while private providers must be approved either by regional authorities or nationally by the central employment agency. Similar streamlined models exist in Sweden’s Rusta och Matcha and the Netherlands’ reintegration markets, where client choice determines provider selection, subject to meeting basic criteria.
More complex contracting models, including those used in Australia, Canada, Ireland and the United Kingdom, employ tendering procedures that incorporate scoring on provider capabilities. Potential bidders are screened based on their experience, financial viability, and capacity to serve specific geographic or client segments. Submission of detailed proposals then also permits scoring against pre‑defined award criteria. Consultations with potential bidders are often conducted between stages to refine expectations and improve transparency – particularly around client characteristics and outcome expectations.
Award criteria in these models typically combine quality and price elements, with an emphasis on avoiding low-cost, low-quality bids. The balance between these criteria varies considerably. For instance, Ireland’s JobPath programme used a weighted approach that gave 40% of the score to cost, and the remaining 60% to a mix of implementation, service delivery, and contract governance. The UK’s Restart programme gave even more weight to qualitative factors (65%), encouraging providers to propose partnerships with local actors and robust delivery models. In contrast, Canada’s Employment Services Transformation pilot placed 70% of the total score on qualitative criteria, including client engagement strategies and prior experience, with only 22% based on cost and 8% on interviews. This reflects a broader international trend toward prioritising quality and past performance, especially in services targeting vulnerable jobseekers.
Australia represents a notable outlier in its approach. In its jobactive system, the Australian Government did not consider cost in its selection criteria at all. Instead, 90% of the scoring focussed on provider capability to deliver results, including governance, employer engagement, and expected employment outcomes. A third of that score specifically assessed the provider’s past performance. This model reflects a mature market in which service costs are well understood, enabling a sole focus on quality and impact. However, it has also attracted criticism for contributing to market concentration and raising barriers to entry for smaller or new providers. Its successor programme, Workforce Australia, attempts to mitigate these concerns by embedding demonstrated performance as a sub element within a broader capability criterion.
International experience highlights the importance of carefully designing selection criteria to balance cost control with service quality, particularly when working with vulnerable groups. Countries moving toward contracted-out models should consider using service quality thresholds, structured two‑stage tender procedures, and transparent performance frameworks to ensure that selected providers are both capable and motivated to deliver meaningful employment outcomes.
4.4.4. Strong performance frameworks are vital to manage service delivery
Performance management is a central element of any contracted-out employment service model. It ensures that public funds are used effectively and that providers deliver services aligned with policy goals. In outcome‑based contracting, managing performance should go beyond simple assessment of whether providers achieve employment outcomes for jobseekers to overseeing service quality more generally. The challenge for contracting authorities is to strike a balance between encouraging innovation and flexibility on the part of providers while maintaining sufficient regulation and oversight to guard against gaming or neglect of difficult-to-serve clients.
Effective performance management involves setting clear, demanding yet achievable performance targets, and regularly monitoring providers through both quantitative and qualitative mechanisms. Quantitative targets are intricately tied to hard outcomes which often form part of the payment structure (such as sustained job placements). Qualitative aspects of performance management include site inspections, audits of payment claims, and systems for tracking and responding to client feedback or complaints. These mechanisms help assess service quality and ensure that providers maintain a satisfactory standard of delivery, particularly for vulnerable jobseekers. Performance management frameworks may also include requirements around customer satisfaction, based on surveys or independent monitoring.
Australia’s provider performance rating system exemplifies the implementation of a mixed approach that aims to provide individuals with a demonstrated progress towards employment allied to a high-quality service delivery. It’s latest programme Workforce Australia replaced an earlier system of star ratings to include a broader set of performance measures. Views of participants and employers are combined with metrics on outcomes across five individual modules (sustained employment; progress to employment; quality of service‑ participants; quality of service‑ employers; and licensing standards). High, medium and low ratings are provided for each of the individual modules, and for the individual metrics comprising each module. A final overall rating then aggregates these individual modules, giving most weight to the sustained employment module (which measures 12‑ and 26‑week employment outcomes) (DEWR, 2025[49]). Ratings are published quarterly and used to guide both jobseeker choice and government decisions on contract continuation (DEWR, 2025[50]). This transparent, data-driven system enhances accountability and provides ongoing performance incentives.
To accommodate differing labour market conditions and client needs, some countries adjust performance targets regionally or across client groups. This can be built directly into payment mechanisms, or used as part of a suite of performance monitoring. For example, the Work Programme in the United Kingdom adjusted contractual performance targets for its different payment groups, such that some groups triggered a payment for a job outcome when their employment duration reached 13 weeks, whereas for other, more job-ready groups, this threshold was set at 26 weeks. In its previous contracted-out employment service, jobactive, Australia used regression to adjust performance expectations based on the client characteristics and the state of the local labour market for providers (OECD, 2012[51]). It is critical to ensure that within systems where adjustments are made for performance, variations in expected performance accord well to the differences in individuals’ labour market barriers or local economic conditions that providers face.
When underperformance is identified, contracting authorities must have procedures in place to respond. These may include formal performance improvement plans, reallocation of referrals or market share to higher-performing providers, or, in extreme cases, contract termination. Importantly, such actions must be managed to ensure continuity of service for jobseekers. Australia again provides a good practice example, where Star Ratings were used to trigger the reallocation of business and even exclusion of providers from specific regions. Conversely, evidence has also shown that weak performance management frameworks can undermine contracted provision, leading to poor outcomes and poor value for money (Behaghel, Crépon and Gurgand, 2014[52]).
Enforcing performance targets is essential to ensuring that quasi-markets for employment services remain effective and responsive. For market mechanisms to work, underperforming providers must face credible consequences, as without them, there is limited incentive to improve. This requires clearly defined and objectively measurable performance benchmarks, along with mechanisms to assess results in a timely manner.
When performance falls short, contracting authorities typically have several enforcement options. These include initiating performance improvement procedures (such as temporarily withholding fees), reallocating client referrals or market share to better-performing providers, and, in serious cases, terminating contracts. It is vital that such processes protect the continuity of service for jobseekers, and that underperformance is accurately diagnosed to inform corrective action. In Australia, for example, the Star Ratings system not only informed jobseeker choice but was also actively used by the government to shift market share and exclude low-performing providers from future business.
Overall, strong performance management frameworks are essential to the credibility and success of contracted-out models. For Greece, this means investing in robust data infrastructure, ensuring transparent and evidence‑based target setting, and maintaining active oversight throughout the contract lifecycle. A mix of hard and soft indicators – including job placement rates, client engagement, satisfaction, and in-work progression – can help balance incentives, especially when working with vulnerable jobseekers.
4.4.5. Robust data exchange platforms enable smooth service delivery
A robust and secure data exchange infrastructure is essential for the effective delivery of contracted-out employment services. Data sharing to enable case management, performance monitoring, and payment verification is essential to deliver services effectively. Outcome‑based payment models rely on the ability to track jobseeker employment and earnings over time, often requiring access to administrative records such as payroll, tax, or social security data.
Data systems which encourage and facilitate the efficient and secure connection of data between different delivery enterprises is essential for cost efficient and timely delivery of contracted-out programmes which involve multiple actors. Estonia offers a compelling example in how government can harness digital technology to deliver better services for individuals (see Box 4.1 for more details), with benefits that flow through into its employment services. In the ALMP space, the Estonian Unemployment Insurance Fund (EUIF) – the Estonian PES – utilises X-tee to integrate information from over 30 government registers to manage client case files, verify jobseeker status, and administer employment support. For instance, a client’s employment status can be verified automatically through the national employment register, while education history can be pulled from the national education register. Estonia leverages GDPR as a positive tool to place a strong emphasis on transparency and citizen oversight, and collates information on government use of individual data in an easy-to‑access portal (e-Estonia, 2025[53]).
Government-wide data architecture plan facilitates the use of data across government services
A holistic data strategy helps to ensure that countries are placed to seamlessly integrate services, enabling the link-up of services across policy domain. Estonia is at the vanguard of efforts to leverage all of the benefits that such integration can bring.
Estonia’s digital infrastructure is among the most advanced in the OECD, underpinned by the X-tee data exchange platform. X-tee facilitates secure and decentralised exchange of information between more than 3 500 services offered by over 1 150 public and private institutions.
The system operates on the principle of interoperability between decentralised systems, enabling each institution to maintain its own database while allowing for seamless information flow across public and private actors. Citizens benefit from a once‑only data provision principle, whereby data they submit to one government institution can be securely reused by others, reducing duplication and administrative burden.
Source: X-TEE ( (2025[54]), Factsheet, https://www.x-tee.ee/factsheets/EE/#eng; Republic of Estonia, (2025[55]), Data exchange layer X-tee, https://www.ria.ee/en/state-information-system/data-exchange-platforms/data-exchange-layer-x-tee
In systems where there is not an overarching data platform in place, data sharing and security procedures which are risk-based can help to ensure that data can be shared but are done so in a means which is cognisant of the risks encountered. Australia’s “Right Fit For Risk” (RFFR) framework provides a tailored, proportionate approach to data protection in the country’s highly outsourced employment services environment.
RFFR classifies providers based on the nature and volume of data they handle, setting security requirements accordingly. Providers serving more than 2 000 clients annually are designated as high-risk and are required to obtain independently-verified ISO 27001 certification (an international standard for information security management). Medium-risk providers are further sub-classified into two groups, dependent on a qualitative assessment of further risk factors. These factors include the IT environment; the level of outsourcing; subcontracting arrangements; organisational structure; the level of security maturity, and the extent of sensitive information held and level of access to departmental systems. This will dictate whether they need to follow a self-verified version of ISO 27001, or simply provide a “Management Assertion” of compliance with risk protocols. (DEWR, 2025[56]). This tiered structure ensures that data protection obligations are proportionate, avoiding excessive compliance costs for smaller providers while ensuring robust protections for sensitive client data.
Australia has also made further efforts to ensure its data sharing protocols are proportionate and support providers to deliver services. To reduce barriers for smaller organisations, DEWR reimburses a portion of the costs associated with accreditation and technical upgrades (Vodopivec, 2023[6]). DEWR also offers the possibility for third-party systems to be accredited to manage clients’ data under RFFR, expanding choice for providers while maintaining system integrity (DEWR, 2025[57]). This approach balances data security with inclusivity, particularly benefiting smaller providers who may not have the capacity to develop internal systems.
An important element in modern data platforms is ensuring prompt and efficient data transfer. The United Kingdom has developed a comprehensive, automated data infrastructure for its outsourced employment services through the Provider Referral and Payments (PRaP) system (DWP, 2024[58]). PRaP supports both client referral and outcome verification across key programmes, including the Work and Health Programme and the Restart Scheme (DWP, 2025[43]; 2025[59]). PRaP enables public employment services to refer clients to contracted providers with complete digital case information, including eligibility and prior interventions. Once a provider submits evidence of job placement, the system automatically verifies outcomes using employment and income data from HM Revenue and Customs’ Real Time Information (RTI) system. This significantly reduces the risk of error or fraud and simplifies the claims process for providers.
PRaP also automates provider payments based on verified outcomes, triggering payment disbursements once predefined employment duration or earnings thresholds are met. This automation reduces administrative burden on both providers and government, ensures timely reimbursement, and enhances transparency. The PRaP model demonstrates its use in high-volume programmes and showcases what is possible when linking to real-time, digital tax administrative records.
Several practical lessons on data availability and storage emerge. First, DYPA should explore simplified mechanisms to allow contracted providers secure access to essential data, particularly regarding jobseeker eligibility and service history. Second, the basis for automated data exchange could be extended to incorporate additional information (e.g. on earnings) that could support DYPA in defining payment levels when contracting-out ALMPs. Third, the development of a centralised referral and payment platform, similar to PRaP, could enhance the efficiency of DYPA’s contracting operations and provide a clearer audit trail for monitoring outcomes.
Piloting these approaches could allow DYPA to build technical capacity, test system performance, and refine governance procedures before broader implementation without impacting its day-to-day operations. As the contracting model expands, investing in secure and functional data exchange infrastructure will be a critical enabler of service quality, accountability, and efficiency.
4.4.6. Effective employer engagement strengthens labour market links
Effective employer engagement helps to better connect participants in contracted-out services with employers in their area. This can help to ensure programmes are well-tailored to local needs, and builds credibility of providers and their participants with employers. There are numerous different strategies that can be adopted by providers to engage with employers.
One strategy for engagement with employers is for providers to employ internal teams with a direct responsibility to engage employers. The advantage of this means that specialisation within these teams permits greater resources to be developed. However, this then lends itself to larger providers who have sufficient capacity to finance engagement teams. In the United Kingdom, Work Programme providers developed internal employer engagement teams that liaised directly with large and small employers. However, research found limited ability to influence job quality or recruitment practices. Employers maintained control over selection, and the burden fell on jobseekers to adapt (Ingold and McGurk, 2023[48]).
Even in contracted programmes with large providers, co‑operation between providers can be used to increase the efficacy of resources used to build links to employers. The Work and Health Programme in the United Kingdom introduced a “Prime Provider Partnership” that co‑ordinated engagement across five large providers. A single account manager was appointed to engage with employers, reducing duplication and complexity. This aimed to provide a seamless interface (“hide the wiring”) for national employers with vacancies in multiple regions.
In the Restart Programme a bottom-up provider levy (0.01% of income) was proposed to co‑ordinate employer engagement. The genesis of this started in the contracting process for the programme, where for the first time each bidder was required to specify how they would co‑operate and share evidence with other providers (Ingold and McGurk, 2023[48]). The result was the “ReAct Partnership”, a provider-led initiative to fund shared research and innovation on employer engagement. Six out of the eight prime providers signed up, and it is co‑ordinated by the Institute for Employment Studies and industry associations. Among its broader work programme, it has produced research on how to better improve and align engagement with employers (React Partnership, 2022[60]).
For DYPA, thinking about how contracts and tendering processes can be designed to encourage employer engagement will help to ensure that contracted-out programmes are delivered in a manner which maximises the possibility to match jobseekers with employers.
The existing institutional and legal set-up in Greece means that DYPA is well placed to design and implement contracted-out services for vulnerable jobseekers. Its memoranda of understanding with several social enterprises, alongside its broader role in implementing and co‑ordinating housing services means that links are already in place with external organisations that could potentially facilitate the design and delivery of any contracted-out employment programmes. Whilst some amendments to the legislation may be required to guide new contracted-out service delivery, these would be consistent with those already established for existing contracted-out training.
4.5.1. The “Jobs Again” DYPA legislation already enables partnerships in the delivery of existing employment services
The existing legal baseline in Greece already provides for many of the features needed to establish and run contracted-out employment services, including contracts which have performance‑related elements (see Box 4.2). This establishes a solid foundation for implementing new contracted-out services, though additions may be necessary to enable data sharing and performance management with contracted partners. Any new changes to legislation will need to be performed in co‑ordination with ministries that share responsibility for policy in this area (for example, the Ministries of Labour (that oversees DYPA) and of Social Cohesion and Family). A detailed review of the legislation and operational requirements will need to be performed by DYPA’s legal services to establish the degree of changes needed.
DYPA already works with external partners as part of its activities
In the context of the training programmes that are funded by the RRF, DYPA’s law foresees the implementation of these programmes by private training providers as well as by the lifelong learning centres of the Greek public universities, and the participation of the social partners in their design (Government of the Hellenic Republic, 2022[61]). DYPA and its contracted training providers also work closely with certification providers (for the certification of these programmes upon completion). In addition to the providers, DYPA has set collaboration agreements with a range other partners (see Section 4.3.3 for more details).
To implement new contracted-out employment services, additions to the legislation may be needed to ensure the basis for management of performance with any new contractors is determined. Given DYPA’s existing arrangements for close work with external partners, included those contracted to deliver services, these additions should not be overly problematic to implement.
Data sharing across organisations is established but may need to be supplemented for external partners
Article 30 of the DYPA legislation adds a legal basis which provides for the data sharing for the Ministry of Labour and Social Affairs with the Mechanism for Labour Market Needs Diagnosis (MDAAE) for the conduct of its affairs laid out in legislation. However, this provision does not extend to wider stakeholders, only to those that are directly engaged with the MDAAE (Government of the Hellenic Republic, 2022, p. Art 30[61]).
In the context of delivery of the Workforce Skills Strategy, Article 26 provides for linking remuneration of training providers to their performance, in terms of the employability of beneficiaries. The aim is to optimise effective management of resources and efficiency of interventions (Government of the Hellenic Republic, 2022, p. Art 26[61]). As part of these operations, DYPA can access the necessary data from its voucher platform hosted by its IT contractor Diofantos. Employment-related outcomes can be accessed from the ERGANI and EFKA databases.
To co‑ordinate and enable data sharing with any new external partners offering employment services, additional provisions may be required to establish the legal basis, however in practice there is already precedence for this in other cases.
Pilots of programmes are already established in law
Article 24 provides for pilot of measures related to the Recovery and Resilience Fund (RRF) (Government of the Hellenic Republic, 2022[61]). Whilst this article is restricted only to those measures in the RRF, it establishes pilots as a legitimate tool for DYPA to deploy as part of its delivery of labour market services. In addition to this, DYPA has also conducted a number of pilots independent of the RRF, demonstrating its capacity to implement them.
4.5.2. DYPA has the ability to co‑ordinate and deliver employment services at regional level
DYPA co‑ordinates delivery via its Head Office and eight Regional Directorates, which manage in total 120 local delivery points of employment services called Employment Promotion Centres (KPA2) (OECD, 2024[4]). This provides DYPA with an agile structure that has the ability to test out new contracted-out employment services across different localities.
In addition to this, reforms to DYPA’s data infrastructure mean that it has the capacity to implement modern employment services supported by secure and comprehensive data management. The Central Interoperability Centre (KED) of the Ministry of Digital Governance, and DYPA’s web services facilitate the exchange of data across a range of registers (OECD, 2024[4]).
DYPA also plays a key role in integrated services for jobseekers. Programmes such as Renovate and Rent (Ανακαινίζω – Νοικιάζω), municipal Housing and Employment initiatives (Στέγαση και Εργασία), and national social housing schemes provide housing support for jobseekers. DYPA collaborates through structured partnerships with the Ministry of Social Cohesion, municipalities, NGOs and the Hellenic Development Bank. DYPA helps ensure that housing support and labour market reintegration are effectively linked by identifying eligible jobseekers, managing referrals, providing employment counselling and facilitating access to wage subsidies (DYPA, 2025[62]). DYPA has also recently intensified efforts of co‑operation with other bodies, including with the Community Centres of the Ministry of Social Cohesion and Family that manage the administration and implementation of the Greece’s guaranteed minimum income scheme.
4.5.3. DYPA has built institutional capacity to deliver contracted-out training programmes
DYPA’s VET Directorate has recently started to manage the administration and implementation of training programmes for both employed and unemployed persons funded by the RRF. This is part of a significant effort to upskill and reskill 500 000 people via vocational training courses. The delivery of these training programmes is performed by contracted-out training providers (private and lifelong learning centres of the Greek public universities) and certification providers (see Box 4.2 for more details).
The management of these services is DYPA’s first systematic effort to manage contracted-out services and it has built its capacity and expertise to deliver these functions, potentially smoothing the pathway for further development of contracted out services. DYPA coped with a complex delivery landscape with a small team and very tight RRF deadlines (World Bank, 2025[63]). In doing so, DYPA has shown that it has the capacity to flexibly manage programmes delivered jointly with external partners. This experience will be invaluable for DYPA to manage any future contracted-out employment provision.
Training providers should be judged on participant completions and labour market outcomes
Legislation stipulates that vocational training providers are paid different amounts dependent on the degree to which participants have completed the training, and on the performance of the training provider.
The evaluation of providers is based on established evaluation milestones including the:
Percentage of trainees who found (or maintained) employment within three, 6 and 12 months after the end of training
Evaluation of the provider by the trainees and, in the case of employee training, by the employers
Percentage of trainees who completed training and were certified.
If a participant does not complete the programme payment is capped at 40%. If participants complete the training but are not certified, payments are capped at 70%.
The evaluation results in a score for the provider’s performance which is expressed along a scale: (1) poor, (2) insufficient, (3) satisfactory, (4) very good, (5) excellent.
Scores 1‑2 correspond to a negative evaluation, scores 3‑4 to an adequate positive evaluation and score 5 to an excellent positive evaluation.
The performance of a provider cannot be graded with a score of 3 or higher, except in cases of in-company training, unless at least three (3) of the following five (5) conditions are met:
At least 50% of unemployed participants found employment within 12 months
At least 90% of employed participants maintained employment for 12 months
At least 75% of participants positively evaluated the training
At least 75% of employers positively evaluated the training (when it is an employee training program)
At least 80% of the trainees who have completed the programme have been certified
Source: Government of the Hellenic Republic (2022[61]), Law No. 4921. Jobs Again: Reorganisation of the Public Employment Service and digitalisation of its services, upgrading of workforce skills and diagnosis of job needs and other provisions, Articles 36 and 38, https://search.et.gr/en/fek/?fekId=587192
The existing institutional and legal foundations position DYPA well to pilot contracted-out employment services tailored to vulnerable jobseekers. Building on international evidence and aligned with DYPA’s current partnerships and infrastructure, two model interventions are proposed. These programmes build on the vulnerable groups established in Greek legislation and on the factors identified in the analysis for this project. The first programme is to address individuals who have barriers to employment, but which are not explicitly linked to any health conditions. The second programme is specifically designed to address who individuals with specific health needs.
The unique needs of those with health conditions mean that usually these groups are served separately in contracted-out employment programmes (for example, in the Work and Health Programme in the United Kingdom, the Fit-for-work programme in the Netherlands or New Zealand’s Placement or Assistance Initiative contracts). The advantage of a separate programme for those with health conditions is that it enables specialised providers to tender for the contract, who can deploy specific and targeted support for these groups.
For the vulnerable groups that do not include health-related barriers, the literature and experiences of OECD countries in delivering these services mean that a number of features emerge which will be important to account for in the design of the programme. These include leveraging of partnerships to address multiple barriers to employment, and using contract design to ensure individuals are supported to maintain employment.
It would be possible to tender an integrated employment programme that included those with health conditions and other vulnerable groups, and to distinguish between them using different service streams and payment rates (for example, the Work Programme in the United Kingdom separated payments into nine streams, including a separate stream for those with health conditions). However, to give space for DYPA to test what is possible and what might work well work with different (and specialised) providers, it is suggested to keep the two potential programmes separate. In the future, after having reviewed what worked well, it may be possible to envisage a unified programme which separates individuals via service streams, if DYPA decide that such an implementation would be preferable.
For both programmes it is suggested to provide sufficient incentives for providers to perform well using outcome‑based payments. For the programme for individuals with health conditions, it is suggested to use a higher fixed fee, in recognition of the increased upfront costs associated with addressing labour market barriers. However, it is important to note that the proportions of outcome‑based fees are just suggestions. Typically, outcome fees for contracted-out employment programmes range from 30‑100% per client, and in most programmes outcome‑based fees comprise the majority of total potential fees (Langenbucher and Vodopivec, 2022[41]). The proportion of outcome‑fees proposed in the two pilot programmes below sit somewhere in the middle of the observed range across OECD countries. Should DYPA wish to be more or less ambitious in its risk appetite, it could consider increasing or decreasing these proportions. The advantage of a pilot approach is that it will allow DYPA to learn how its contract specifications impact provider performance before making any future adjustments when considering a wider roll-out of the programmes.
Both programmes could initially be implemented in one to two regions (of the eight regional directorates DYPA operates in) to test effectiveness and operational viability before scaling nationally.
4.6.1. Programme 1: Intensive support for the long-term unemployed
The first programme would target jobseekers registered with DYPA for more than nine months. The analysis in Chapter 3 of this project identified that there is a high persistence of unemployment. Those that reach nine months as unemployed are very likely to remain unemployed for one year or more. Having this criterion then reduces the cost of implementing the programme and avoids treating individuals who may have quickly exited from unemployment anyway. Eligibility criteria could also include those with past spells of long-term unemployment, low education, older age, or prior dismissals. This could reduce eligibility and keep costs lower but excludes some jobseekers from eligibility who may also potentially benefit from the programme.
Programme features
The tender document should outline broad expectations for working with this group and allocate scores based on how providers detail their plans for working with participants, but with fewer strict conditions made on service standards, to allow the providers scope to innovate their delivery.
Features to include:
Delivered by non-public providers (NGOs, social enterprises, or private employment firms). Suggested to offer contracts to two delivery partners in each contract area to encourage competition, but also maximise learning during pilot phases.
Participants referred automatically by DYPA based on unemployment duration (those unemployed for longer than 9 months).
Duration: Up to 12 months of personalised employment support.
Services to include: regular face‑to-face guidance, job-search coaching, barrier removal (e.g. transport, childcare), soft skills training, employer outreach, and referrals to subsidised work or training.
Service standards
Minimum service standards are envisaged to ensure that claimants are offered high quality and ongoing support by their providers, but should not be overly restrictive, to allow providers to tailor services accordingly.
First appointment within 10 working days of referral.
Personalised action plan within two weeks.
Minimum monthly engagement with progress tracking.
Payment structure
To ensure good value‑for-money for DYPA performance fees can be strongly weighted towards conditional payments based on labour market attachment.
30% engagement fee upon participant start.
30% job entry payment (after 30 days in employment).
40% outcome payments based on sustained employment (e.g. 6 months).
Higher payments for placing individuals registered as a special or vulnerable group, or those with a higher initial profiling score.
4.6.2. Programme 2: Employment support for people with health conditions or disabilities
This programme would target registered jobseekers with physical or mental health conditions, disabilities, or recovery from substance dependency. International models like Individual Placement and Support (IPS) and Austria’s tiered social economy system demonstrate the value of specialist, long-term, wraparound support. This programme would test a more resource‑intensive, relationship-based model whilst allowing providers the space to innovate in services provision.
Programme features
In the tender document for this programme, it is suggested to give adequate weight on how providers can demonstrate their links with broader local social services and how they intend to utilise these links (including through any partnership agreements) to provide more holistic support. Building on the evidence base for services with mental health support, tenders should also include information on what services will be offered to those with mental health support, including the proposed format, duration and professional (medical) expertise of counsellors/facilitators.
Features to include:
Delivered by specialised third-sector or rehabilitation providers.
Participants referred via DYPA or municipal social services. Eligibility based on a registered health condition and triggered at three months of unemployment.
Duration: Up to 24 months, with additional post-placement support.
Services to include: specialist case management, workplace adaptation, health-employment co‑ordination, peer coaching, and mental health or addiction recovery support.
Fund for providers to pay for support for workplace adaptations.
Service standards
One of the key things in the service standards for this programme is sustained contact after employment, building on the evidence of post-placement support seen for vulnerable groups, and scheduled follows seen in the contracted-out Work Choice programme.
Joint employment/health action plan within 4 weeks.
Action plan to include what specific support and interventions are planned to overcome specific health barriers
Ongoing support at least monthly with participants.
Employment follow-up support with employers.
Minimum three‑monthly employer engagement and support for workplace adjustments, continued for one year
Payment structure
60% fixed fee paid in instalments throughout programme, at three monthly intervals.
40% outcome payments (e.g. 20% job entry/education participation, 20% sustained employment/education completion).
Higher payment levels for those with multiple barriers/higher profiling score.
Given the limited current resources within DYPA and its existing reform agenda, it is suggested that a pilot programme is first implemented rather than an immediate full-scale roll-out of new employment programmes. The advantages of proceeding with pilots are that they are cheaper to implement, allow DYPA to test both market capacity and policy design, and permit initial process and counterfactual evaluation to determine possible policy improvements (and test value‑for-money of expansion).
4.7.1. Indicative timeline
The timeline outlined below provides a broad indication of the overall timescales involved in designing, implementing and evaluating a pilot (Figure 4.1). There is some flexibility around the duration of different steps, dependent on the resources invested and the timelines are therefore sensitive on the speed of implementation. The most resource intensive period is at the beginning when all aspects of the programme’s design, management and implementation need to be considered. For example, if it becomes a complex process to agree any necessary legal amendments or protocols to enable data management and sharing, it is easy to envisage these timelines being extended. However, without more extensive prior knowledge the exact implementation resources and challenges, the timeline presented here represents a broad indication of the rough timelines involved, assuming a relatively smooth initial planning and preparation phase.
4.7.2. Steps to implement a pilot
Identify funding stream
Conduct analysis to complete business case for pilot implementation. This should give an ex-ante assessment of potential impacts of the policy, drawing on any previous evidence from Greece and in the wider contracted-out employment programme evaluation literature.
The business case should set out the strategic problem the programme is designed to solve and should use analysis to determine the economic business case that sits underneath this. The economic case provides an assessment of the costs and benefits of implementing the programme. A commercial argument can also set out the expected benefits that will result in a viable procurement agreement (see HM Treasury (2018[64]) for more details on the different dimensions to a business case that may be included).
In the case of a pilot there is often no prima facie evidence of the expected programme impacts. Here DYPA can make use of existing quantitative impact assessment of other ALMPs (such as the recent impact evaluation of some of its ALMPs (OECD, 2024[4])). It may also utilise similar evaluations of contracted-out ALMPs in other countries (such as those mentioned earlier in this report). The idea is to provide a broad indication, that is not overly precise and properly sets out the uncertainties, but which provides some guidance on expected implementation benefits and risks.
Secure funding to implement pilot (using the business case to inform decisions).
Establish senior responsible officer
Appoint a senior responsible officer (SRO) to take ownership and accountability of pilot objectives and delivery (at DYPA executive level).
Assign DYPA policy team to begin co‑ordination
Invest resources to setting up the process, including making first contact with potential co‑operating regional offices/managers of the eight regions.
Engage with the social partners to gather views on potential policy formation and seek interest in sitting on pilot board.
Review the overall legislative basis for conducting pilots and make changes based on what is needed to perform pilot in this area.
Determine participating KPA2s
If possible, randomly allocating a selection of KPA2s within a region would support more robust analysis of pilot effects.
An alternative can also be to consider randomly allocating eligible participants within a KPA (e.g. based on registration or birth dates), or staggered roll-out of the pilot (where offices start offering the programme at progressive dates, so that offices that have not yet started the pilot act as natural comparators for those that have).
For pilot stage, it could be where existing local links in particular KPA2 to third-sector/private delivery partners are already strong and no attempt is made at selected offices at random.
Establish working level steering group to deliver on pilot implementation
Regional partners in KPA2, central DYPA staff (including finance, policy and operational staff), regional DYPA leads, social partners, delivery organisations – to co‑ordinate progress and track milestone.
Determine steering group meeting frequency.
Establish a project plan
Agree project milestones and deliverables.
Define employment programme contents/policy design.
Agree monitoring / review of services.
Conduct initial market testing
DYPA to begin initial contact with stakeholders, partners and existing third- and private‑ sector support providers to establish existing capacity in the market, to determine appetite for tendering and to establish constraints, blockages and opportunities for any possible tender.
Establish legal basis for information sharing with contracted partners
Review existing legislation and make proposals to ensure that information on referrals, management information and outcomes can be shared with external providers.
Engage with Ministry of Digital Governance to explore how to ensure a robust data transfer mechanism with providers.
Conduct data analysis to establish baseline and set performance related payments
Analysis of DYPA/Ergani data to establish baseline outcomes (what proportion of jobseekers usually find jobs without additional support in the same timeframe) and determine expected price levels for support.
Conduct analysis to determine necessary trial size – using information on desired programme impacts and corresponding number of trial participants needed to identify effects.
Determine referral mechanisms
Determine how to identify and refer eligible participants.
Agreements on information sharing between DYPA and external contractors.
Determine payment mechanisms to providers
How and when to pay providers, based on contract agreements.
Determine data sharing protocols
For provider performance metrics.
Consider what information can be shared from DYPA to providers, to aid delivery.
Determine data collection and storage systems to manage programme data.
Tender for delivery partners
Define provider eligibility criteria.
Specify tender assessment criteria – what scores will be given to different qualitative and quantitative indicators on the bid.
Define minimum service standards.
Outline performance expectations, payment thresholds and levels.
Establish data sharing principles.
Manage pilot delivery
Internal DYPA team to manage implementation and performance with providers, using both performance metrics and qualitative inputs from delivery partners.
Regular reporting between delivery partners, KPA2 and steering committee.
Feedback of bottlenecks/implementation issues to refine delivery.
To provide providers with sufficient incentive for investment in the pilots, it is suggested to run the initial pilots for one year. This balances the cost to DYPA for pilot implementation against investment costs for providers (contracts are still shorter than the typical 2‑3-year contracts seen for contracted-out employment programmes). In addition to the one‑year implementation for new referrals, providers may have to continue working longer with clients to ensure post-placement support is implemented.
4.7.3. Steps to evaluate a pilot
Conduct process evaluation
Analysts working throughout the pilot can draw together qualitative information on how the pilot implementation is being managed, what features are working well, and what needs to be improved.
Indicative questions for the analysis to consider include:
Did the pilot progress as intended? What were the barriers? What were the success?
Did the referral process work?
Did providers have sufficient information? What was the information exchange like?
What did participant pathways look like? What did counsellors/implementers think of the process.
Conduct quantitative evaluation
A quantitative evaluation should take place after the pilot has been implemented, drawing together administrative data to look at the progression of participants during and after the pilot. This evaluation should be planned alongside the pilot design and implementation. Alongside monitoring gross outcomes (e.g. how many individuals find work), counterfactual analysis should be conducted to determine the impact of the programme relative to not doing it (e.g. how many extra individuals find work).
Indicative questions to include:
What were the employment outcomes of participants?
Were there any links to broader (e.g. health, social services) interactions?
How did these compare to similar vulnerable groups in other KPA2?
Did certain groups benefit more/less than others?
To answer questions on programme impacts using counterfactual analysis, it will be necessary to have a suite of administrative data on both trial participants and eligible non-participants. These should include data on past outcomes (for example previous spells of unemployment and previous employment and earnings). These data permit the construction of the counterfactual- that is, they ensure the evaluation only compares to the outcomes of participants to non-participants who are very similar to the participants. An example of the type of data and methodology needed can be found in earlier OECD work (OECD, 2024[4]).
In principle a randomised pilot should make comparisons easy (it should ensure that participants and non-participants are already alike). However, attempts to randomise are not always successful in live operational settings and analytical strategies may have to be adapted (see (OECD, 2023[65]; DWP, 2015[66]) for examples). Therefore, in both experimental and non-experimental settings similar data are required to test the similarity between participants and non-participant and make adjustments where needed.
Conduct qualitative research
Conduct a suite of qualitative research to explore in more detail what aspects of policy design and delivery worked well, and what could be improved. Consider distinct research with:
Participants
Design online surveys of participants to explore user experience.
Consider in-person focus groups to conduct deep dives on specific policy aspects.
Service providers
Conduct semi-structured interviews with providers, in particular to explore aspects of contract design, data sharing mechanisms, performance management, participant behaviour and experience.
Employers
Design online survey of employers (if possible, using administrative information on employers to derive the sample for the survey) to explore their experiences of hiring participants in the scheme, and to identify any further area of skills development for future participants, based on business needs.
References
[17] Amnyos/Edater (2019), Évaluation de l’impact du Programme national FSE sur la lutte contre la pauvreté et la promotion de l’inclusion (Axe 3), Fonds Social Européen, https://fse.gouv.fr/sites/default/files/2022-09/Axe3%20FSE_Rapport%20impact_final_dec%202019.pdf.
[31] Barry, M. et al. (2006), “Implementation of the JOBS programme in Ireland”, Journal of Public Mental Health, Vol. 5/4, pp. 10-25, https://doi.org/10.1108/17465729200600028.
[52] Behaghel, L., B. Crépon and M. Gurgand (2014), “Private and Public Provision of Counseling to Job Seekers: Evidence from a Large Controlled Experiment”, American Economic Journal: Applied Economics, Vol. 6/4, pp. 142-174, https://doi.org/10.1257/app.6.4.142.
[35] Bodnaru, A. et al. (2023), “iJobs – An online implementation of the JOBS II program for fostering reemployment: A feasibility and acceptability study”, Internet Interventions, Vol. 34, p. 100674, https://doi.org/10.1016/j.invent.2023.100674.
[33] Carlier, B., M. Schuring and A. Burdorf (2017), “Influence of an Interdisciplinary Re-employment Programme Among Unemployed Persons with Mental Health Problems on Health, Social Participation and Paid Employment”, Journal of Occupational Rehabilitation, Vol. 28/1, pp. 147-157, https://doi.org/10.1007/s10926-017-9704-3.
[36] Cottier, L. et al. (2018), Does Job Search Assistance Really Raise Employment, IZA – Institute of Labor Economics, https://docs.iza.org/dp11766.pdf.
[29] Curran, J., P. Wishart and J. Gingrich (1998), JOBS A guide for teaching people successful job search strategies, Michigan Prevention Research Center, Institute for Social Research, University ofMichigan.
[21] DARES (2024), Note d’étape du comité scientifique de « l’Expérimentation territoires zéro chômeur de longue durée, Déchiffrer le monde de travail pour éclairer le debat public (DARES), https://dares.travail-emploi.gouv.fr/sites/default/files/4d37880f724dbfc386e66e0587dea674/DARES-FS-2024-TZCLD-note%20%C3%A9tape.pdf.
[57] DEWR (2025), Accredited Third Party Employment and Skills (TPES) Systems, https://www.dewr.gov.au/right-fit-risk-cyber-security-accreditation/accredited-third-party-employment-and-skills-tpes-systems (accessed on 3 July 2025).
[56] DEWR (2025), Provider classification for accreditation, https://www.dewr.gov.au/right-fit-risk-cyber-security-accreditation/provider-classification-accreditation (accessed on 2 July 2025).
[44] DEWR (2025), Workforce Australia Employment Services Provider Payments, Department for Employment and Workforce Relations, https://www.dewr.gov.au/download/11385/workforce-australia-employment-services-provider-payments/28097/workforce-australia-employment-services-provider-payments/pdf (accessed on 1 July 2025).
[50] DEWR (2025), Workforce Australia Services Deed of Standing Offer 2022 - 2028, Department of Employment and Workplace Relations, https://www.dewr.gov.au/download/13970/workforce-australia-services-deed-standing-offer-2022-2028/40012/workforce-australia-services-deed-standing-offer-2022-2028/pdf.
[49] DEWR (2025), Workforce Australia Services Provider Performance Ratings, Department of Employment and Workplace Relations, https://www.dewr.gov.au/workforce-australia/workforce-australia-information-providers/workforce-australia-services-provider-performance-ratings (accessed on 2 July 2025).
[11] Dromundo, S., M. Lüske and M. Tuccio (2023), “Innovative approaches to tackle long-term unemployment”, OECD Social, Employment and Migration Working Papers, No. 300, OECD Publishing, Paris, https://doi.org/10.1787/e1f7e16e-en.
[27] DSS (2025), Mental Health Support, Department of Social Services, https://www.dss.gov.au/mental-health-support (accessed on 3 June 2025).
[59] DWP (2025), Chapter 11: Programme completers and early exits, Department for Work and Pensions, https://www.gov.uk/government/publications/work-and-health-programme-including-jets-provider-guidance/chapter-11-programme-completers-and-early-exits (accessed on 2 July 2025).
[47] DWP (2025), JETS (Job Entry Targeted Support) Impact Evaluation, Department for Work and Pensions, https://assets.publishing.service.gov.uk/media/67b7609af5da8e7f36e3ecae/jets-impact-evaluation.pdf.
[43] DWP (2025), Restart Scheme provider guidance (version 5.01), https://www.gov.uk/government/publications/restart-provider-guidance/restart-scheme-provider-guidance (accessed on 30 June 2025).
[38] DWP (2025), Work Choice Impact Evaluation, Department for Work and Pensions, https://www.gov.uk/government/publications/work-choice-impact-evaluation/work-choice-impact-evaluation.
[58] DWP (2024), Chapter 6: The Provider Referral and Payments System (PRaP), https://www.gov.uk/government/publications/access-to-work-dwp-provider-guidance/chapter-6-the-provider-referral-and-payments-system-prap#general-information (accessed on 2 July 2025).
[46] DWP (2024), Effective contracting of employment and health services, DWP Research Report 93, Department for Work and Pensions, https://www.gov.uk/government/publications/effective-contracting-of-employment-and-health-services-evidence-review/effective-contracting-of-employment-and-health-services--2.
[45] DWP (2020), The Work Programme - A quantitative impact assessment, Department for Work and Pensions, https://assets.publishing.service.gov.uk/media/5fbce4b2e90e077ee2eadbf2/work-programme-quantitative-impact-assessment.pdf.
[66] DWP (2015), Jobseeker’s Allowance Signing Trials, Department for Work and Pensions, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/398803/jsa-signing-trials.pdf.
[62] DYPA (2025), Housing Policy (Στεγαστική Πολιτική), https://www.dypa.gov.gr/more-steghastiki-politiki (accessed on 3 July 2025).
[19] DYPA (2024), Νέο πρότυπο γραφείο της ΔΥΠΑ για την εξυπηρέτηση πολιτών από ειδικές κοινωνικές ομάδες στην Αθήνα (New model DYPA office to serve citizens from special social groups in Athens), https://www.dypa.gov.gr/storage/deltia-typoy/neo-protipo-ghrafio-tis-dypa.pdf.
[2] DYPA (2022), Service Guide for Special and Vulnerable Groups (Οδηγοσ Εξυπηρετησησ Για Ειδικεσ Και Ευπαθεισ Κοινωνικεσ Ομαδεσ), DYPA, https://www.dypa.gov.gr/storage/eko/odhgos-eksyphrethshs-gia-idikes-kai-efpathis-kinonikes-omades.pdf.
[53] e-Estonia (2025), e-Estonia guide, Enterpise Estonia, https://e-estonia.com/wp-content/uploads/eestonia_guide_08-04-2025.pdf (accessed on 23 July 2025).
[22] European Commission (2018), Promising Pes Practice- Job-carving for jobseekers with disabilities, European Commission, https://ec.europa.eu/social/BlobServlet?docId=19407&langId=en.
[39] Finn, D. and M. Peromingo (2019), Key developments, role and organization of Public Employment Services in Great Britain, Belgium-Flanders and Germany, International Labour Organisation, https://www.ilo.org/sites/default/files/wcmsp5/groups/public/@ed_emp/@emp_policy/@cepol/documents/publication/wcms_724913.pdf.
[14] France Travail (2025), L’accompagnement global (Comprehensive Support), https://www.francetravail.fr/region/hauts-de-france/candidat/france-travail-vous-accompagne/laccompagnement-global.html (accessed on 2 July 2025).
[16] France Travail (2018), Global Support for Jobseekers: A response adapted for vulnerable individuals (L’accompagnement global des demandeurs d’emploi : une réponse adaptée aux besoins d’un public particulièrement fragilisé), France Travail, https://www.francetravail.org/files/live/sites/peorg/files/documents/Statistiques-et-analyses/E%26S/ES_47_accompagnement_global_des_DE.PDF.
[61] Government of the Hellenic Republic (2022), “Law No. 4921. Jobs Again: Reorganisation of the Public Employment Service and digitalization of its services, upgrading of workforce skills and diagnosis of job needs and other provisions”, Gazette of the Government of the Hellenic Republic, (Νόμος ΥΠ’ ΑΡΙΘΜ. 4921. Δουλειές Ξανά: Αναδιοργάνωση Δημόσιας Υπηρεσίας Απασχόλησης και ψηφιοποίηση των υπηρεσιών της, αναβάθμιση δεξιοτήτων εργατικού δυναμικού και διάγνωσης των αναγκών εργασίας και άλλες διατάξεις), https://search.et.gr/en/fek/?fekId=587192.
[1] Government of the Hellenic Republic (2011), “Social Economy and Social Entrepreneurship”, Gazette of the Government of the Hellenic Republic, (Κοινωνική Οικονομία και Κοινωνική Επιχειρηματικότητα), https://search.et.gr/en/fek/?fekId=587192.
[3] Government of the Hellenic Republic (1989), “Promotion of employment and vocational training and other provisions”, Gazette of the Government of the Hellenic Republic, (Προώθηση της απασχόλησης και της επαγγελματικής κατάρτισης και άλλες διατάξεις), https://search.et.gr/el/fek/?fekId=725394.
[20] Hiebl, J. (2020), Outreach to and ALMPs for people furthest from the labour market, https://www.oecd.org/content/dam/oecd/en/events/2020/12/almps-and-good-practices-to-reaching-out-people/austria.pdf.
[64] HM Treasury (2018), Guide to developing the programme business case, His Majesty’s Treasury, https://assets.publishing.service.gov.uk/media/6644948aae748c43d3793bb9/Programme_Business_Case_2018.pdf.
[30] Hollederer, A. and H. Jahn (2023), “Results from a Nationwide Evaluation Study of Labor Market-Integrative Health Promotion for the Unemployed: Impact of the JOBS Program Germany”, International Journal of Environmental Research and Public Health, Vol. 20/19, p. 6835, https://doi.org/10.3390/ijerph20196835.
[48] Ingold, J. and P. McGurk (2023), Employer Engagement: Making Active Labour Market Policies Work, Bristol University Press.
[26] IPS Employment Centre (2025), Evidence for IPS, https://ipsworks.org/index.php/evidence-for-ips/ (accessed on 1 July 2025).
[41] Langenbucher, K. and M. Vodopivec (2022), “Paying for results: Contracting out employment services through outcome-based payment schemes in OECD countries”, OECD Social, Employment and Migration Working Papers, No. 267, OECD Publishing, Paris, https://doi.org/10.1787/c6392a59-en.
[25] Latsou, D. and M. Geitona (2018), “The Effects of Unemployment and Economic Distress on Depression Symptoms”, Materia Socio Medica, Vol. 30/2, p. 180, https://doi.org/10.5455/msm.2018.30.180-184.
[23] Lino Spiteri Foundation (2024), Executive Brief of Annual Report 2023, Lino Spiteri Foundation, https://linospiterifoundation.org/wp-content/uploads/2024/10/Final-Website-edited-brief-Annual-Report-2023.pdf.
[9] Lyons, A. et al. (2019), “Bridging the Gap Between Digital Skills and Employability for Vulnerable Populations”, T-20 Japan 2019, https://www.researchgate.net/publication/332013345_Bridging_the_Gap_Between_Digital_Skills_and_Employability_for_Vulnerable_Populations.
[28] Modini, M. et al. (2016), “Supported employment for people with severe mental illness: Systematic review and meta-analysis of the international evidence”, British Journal of Psychiatry, Vol. 209/1, pp. 14-22, https://doi.org/10.1192/bjp.bp.115.165092.
[40] O’Halloran, D., L. Farnworth and N. Thomacos (2020), “Australian unemployed workers’ experiences of being parked and creamed by employment providers”, Australian Journal of Social Issues, Vol. 56/4, pp. 596-611, https://doi.org/10.1002/ajs4.131.
[4] 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.
[12] OECD (2024), One-stop guidance centres for young people – Finland, OECD Publishing, Paris, https://www.oecd.org/en/publications/2024/11/oecd-youth-policy-toolkit_7ae28a3d/one-stop-guidance-centres-for-young-people-finland_5f0ace7d.html.
[18] OECD (2024), Strengthening Active Labour Market Policies in Korea, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/44cb97d7-en.
[65] OECD (2023), Impact Evaluation of the Digital Tool for Employment Counsellors in Spain: SEND@: Report on the Design and Implementation of an Impact Evaluation of the Digital Counselling Tool for Spain’s Public Employment Services, OECD Publishing, Paris, https://doi.org/10.1787/fe1ec3c3-en.
[42] OECD (2023), “Organisation of public employment services at the local level in Sweden”, OECD Local Economic and Employment Development (LEED) Papers, No. 2023/03, OECD Publishing, Paris, https://doi.org/10.1787/24edca19-en.
[7] OECD (2022), Reaching Out and Activating Inactive and Unemployed Persons in Bulgaria, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/7b91154a-en.
[5] OECD (2021), “Building inclusive labour markets: Active labour market policies for the most vulnerable groups”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/607662d9-en.
[24] OECD (2021), Fitter Minds, Fitter Jobs: From Awareness to Change in Integrated Mental Health, Skills and Work Policies, Mental Health and Work, OECD Publishing, Paris, https://doi.org/10.1787/a0815d0f-en.
[8] OECD (2021), Improving the Provision of Active Labour Market Policies in Estonia, Connecting People with Jobs, OECD Publishing, Paris, https://doi.org/10.1787/31f72c5b-en.
[51] OECD (2012), Activating Jobseekers: How Australia Does It, OECD Publishing, Paris, https://doi.org/10.1787/9789264185920-en.
[10] Paidousi, C. and A. Efstratoglou (eds.) (2020), Digital Inclusion And Human Capital In Greece, Friedrich Ebert Stiftung, https://library.fes.de/pdf-files/bueros/athen/16897.pdf.
[15] Pôle Emploi (2023), The comprehensive approach to comprehensive support - National Council for Policies to Combat Poverty and Exclusion, https://solidarites.gouv.fr/sites/solidarite/files/2023-03/presentation_de_l_accompagnement_global.pdf (accessed on 2 July 2025).
[34] Rayment, M. et al. (2021), Group Work/JOBS II: Cost Benefit Analysis Technical Report, DWP research report no. 990, Department for Work and Pensions, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1003582/cost-benefit-analysis-technical-report.pdf.
[60] React Partnership (2022), Shared employer engagement models - What Works, ERSA, https://ersa.org.uk/wp-content/uploads/2022/05/ReAct-Research-Shared-Employer-Engagement-paper-FINAL.pdf.
[55] Republic of Estonia (2025), Data exchange layer X-tee, https://www.ria.ee/en/state-information-system/data-exchange-platforms/data-exchange-layer-x-tee (accessed on 2 July 2025).
[37] Staible, A. (2017), INA!- Sustain Integration, European Commission, https://web.archive.org/web/20190602114045/https://ec.europa.eu/social/BlobServlet?docId=17306&langId=en.
[13] Valtakari, M. et al. (2020), Ohjaamot - monialaista yhteistyötä, vaikuttavuutta ja uutta toimintakulttuuria, Valtioneuvoston kanslia, Helsinki, https://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/162134/VNTEAS_2020_13.pdf.
[6] Vodopivec, M. (2023), “Raising the bar: Designing and implementing innovative contracted-out employment services in OECD countries”, OECD Social, Employment and Migration Working Papers, No. 301, OECD Publishing, Paris, https://doi.org/10.1787/c7a819e8-en.
[32] Vuori, J. and J. Silvonen (2005), “The benefits of a preventive job search program on re‐employment and mental health at 2‐year follow‐up”, Journal of Occupational and Organizational Psychology, Vol. 78/1, pp. 43-52, https://doi.org/10.1348/096317904x23790.
[63] World Bank (2025), Component 3: Development Of A Proposed Performance-Based Evaluation Framework For Relevant Non-Formal Learning Providers, World Bank.
[54] X-TEE (2025), Factsheet (As of 1 July 2025), X-TEE, https://www.x-tee.ee/factsheets/EE/#eng (accessed on 3 July 2025).