EUROPEAN COMMISSION

Brussels, 17.9.2015

SWD(2015) 176 final

COMMISSION STAFF WORKING DOCUMENT

Analytical Supporting Document

Accompanying the document

Proposal for a Council Recommendation

on the integration of the long-term unemployed into the labour market

{COM(2015) 462 final}

Contents

1.Introduction

2.Long-term unemployment trends

3.Problem Definition

3.1.Transitions from long-term unemployment to employment

3.2.Focus of the initiative

3.3.Low coverage of activation support

3.3.1.Registration with employment services is declining

3.3.2.Participation in active labour market measures is limited

3.3.3.Conditionality of benefits is not always enforced

3.4.Discontinuity in service delivery

3.5.Activation design – Individualisation of service and labour market relevance of support

3.5.1.An heterogeneous group

3.5.2.Services are not sufficiently tailored to individual needs

3.5.3.Low employer engagement

3.5.4.Support is not always leading to increased employability

4.The Envisaged initiative

4.1.Objectives

4.2.Job integration agreements

5.articulation with existing measures

5.1.Links with other policy instruments

5.1.1.The EU Semester process

5.1.2.Support to the long-term unemployed through the European Structural and Investment Funds

5.1.3.Mutual learning at EU level

5.1.4.Active inclusion initiatives and fundamental rights

5.1.5.Reforms undertaken at national level

5.1.6.Additional action needed

5.2.Lessons learned from the Youth Guarantee

6.Effectiveness of the envisaged recommendation

6.1.Increasing coverage of activation support (specific objective 1)

6.2.Continuity and coordination in support delivery (specific objective 2)

6.3.Increased effectiveness of interventions: individualised support (specific objective 3)

6.4.Increased effectiveness of interventions: employer support (specific objective 3)

7.Expected impacts and costs

7.1.Expected impact on transitions to employment

7.2.Expected impact on poverty and social inclusion

7.3.Overall economic impacts

7.4.Implementation costs

7.4.1.The role of the European Structural and Investment Funds

7.4.2.Monitoring costs

7.5.Budgetary impact

7.6.Timely impact and implementation risks

8.Evaluation and monitoring

9.Stakeholder consultations and expertise used

9.1.Stakeholder consultations

9.1.1.Labour Market Observatory (LMO), European Economic and Social Committee (EESC)

9.1.2.European Social Fund Committee

9.1.3.European Network of Public Employment Services

9.1.4.Employment Committee

9.1.5.Social Partners

9.1.6.Social Protection Committee

9.2.Expertise used

Annex 1 Abbreviations and glossary of technical terms used

Annex 2 - Supporting data and analysis

1.Introduction

The strengthened economic activity in the European Union has a positive impact on employment, but progress is subdued and the situation is very different across the Member States. In 2014 the level of unemployment varied from 5.0% in Germany and 5.6% in Austria to 24.5% in Spain and 26.5% in Greece 1 . Since mid-2013, employment has risen and unemployment has fallen in the EU. As economic growth strengthens, labour market conditions should improve further. Long-term unemployment rates also increased and remained at high levels in 2013-14, most markedly in the Member States (MS) most hit by the crisis.

Action is already undertaken to support MS in securing a rapid return to work of those who are long-term unemployed in order to prevent the permanent loss of their skills and capabilities. But more is needed to complement the mechanisms already in place and to reinforce their effectiveness, notably by a more coordinated approach

This report presents the problems addressed by the initiative, the effectiveness of the solutions proposed and provides an image of the estimated impact of a Council recommendation to support the integration of long-term unemployed (LTU). The initiative (2015/EMPL/005) is one of three elements contained in the Commission Work Programme 2015 2 item on Promoting Integration and Employability.

Stakeholders were consulted in the preparation of the report in the period February-June 2015. Main results and positions expressed have been taken into account throughout the process. A summary of the stakeholder consultations is included in the last chapter of this document.

A public consultation was carried out on the ‘Your-Voice-in-Europe’ web portal between 19/02 and 15/05/2015. Twelve weeks were allocated for responses. The consultation was supported by a summary of context and scope of the envisaged initiative 3 . A summary of the consultation responses is available online 4 .

2.Long-term unemployment trends

After six years of subdued growth and low labour demand, LTU numbers doubled in the EU 5 . Half of the unemployed are now long-term unemployed 6 (more than 12 million people or 5% of the active EU population), 60% of whom have been jobless for at least two consecutive years 7 .

The economic crisis led to a surge of LTU, reversing the progress made before 2009 in re-integrating the long-term unemployed in the labour market. Unlike overall unemployment, LTU has had an increasing trend of growth since the beginning of the crisis and it is now at historically high levels (Figure 1).

This is partly due to positive developments in terms of ensuring the long-term unemployed remain on the labour market rather than fall into inactivity by narrowing pathways to early retirement and linking social benefits with activation requirements 8 .

Figure 1: Long-term unemployment rates and share of LTU among unemployed, EU28

Source: Eurostat, LFS, data seasonally adjusted (unemployment rate) [une_rt_q] and non-seasonally adjusted (long-term unemployment rates) [une_ltu_q]

Left axis: Unemployment rates (% of labour force). Right axis: unemployment share (% of unemployed)

Structural dimension

The increasing share of the LTU in total unemployment and decreasing one of the short-term unemployment STU indicate that the job recovery risks benefiting mainly the STU leaving the existing LTU even longer in unemployment and risking further deterioration of their employability and making the LTU structural (Figure 2).

Figure 2: Evolution of short term versus long-term unemployment, 2008-2014

Source: Eurostat LFS [lfsa_upgan], reference population aged 25-74 years

Different subgroups of the unemployed return to employment at different rates, with LTU being less responsive to economic recovery 9 . The slow decline reflects cyclical factors such as an insufficiently strong and broad recovery but also the persistence of high structural unemployment 10 . The ECB warns that a significant share of employment losses during the crisis could become permanent, with structural unemployment stabilising around 10% 11 . Pushing down unemployment therefore requires reinforced efforts to tackle its structural component 12 .

EU Dimension

The 12 million stock of the long-term unemployed at EU level masks a significant amount of diversity at MS level. Most recently, the LTU rate ranged from 1.5% in Austria and Sweden to almost a fifth of the total labour force in Greece (19.5%). Still, LTU rates increased in all MS except for Germany since 2008, as shown in Figure 3 below.

Figure 3. Change in LTU rates EU-28, 2008-14

Source: Eurostat Labour Force Survey (LFS) [une_ltu_a]

3.Problem Definition

3.1.Transitions from long-term unemployment to employment

The on-going economic recovery already started to decrease short-term unemployment, limiting inflows. However, job finding rates among the LTU decreased and continue to be low, as shown by the lines for 2009 and 2013 in Figure 4.

Job finding chances diminish quickly with unemployment duration (Figure 4). If 29% of unemployed succeed in returning to employment by the end of the first unemployment year, less than 17% do so beyond 2 years.

Figure 4. Transition rates from unemployment to employment by unemployment duration

Country differences in transition rates from unemployment to employment are large. There is therefore huge potential for improvement: Over 40% of LTU find a job before the second year of unemployment in DK, EE, SI, compared to less than 20% in EL, SK, RO, IT, BG (Figure 5).

Figure 5. Transition rates from unemployment to employment by unemployment duration, 2013-2014

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates. Note: 'U' refers to unemployment duration; only countries with available data included; lower reliability of data for Austria regarding unemployment duration above 2 years; reference population 15-74 years old.

In spite of lower chances to find a job, fewer unemployed are demotivated and become inactive. Transitions to inactivity dropped during the crisis, more so in the MS most affected, e.g. EL, HR, SK 13 . This shows that transitions to employment are a more acute problem than transitions to inactivity.

3.2.Focus of the initiative

Structural factors on the labour and product markets play an important role in explaining the high long-term unemployment rates and the unemployment duration. Several barriers prevent the LTU to return to employment:

Investment and regulatory constraints limiting job creation or job reallocation between sectors and firms,

Lack of responsiveness of the labour market 14 :

oHigh taxes on labour that do not reward the work effort enough, wages not in line with productivity, rigidity in employment protection and lack of responsiveness of working arrangements, lack of a constructive social dialogue may act as deterrents to hiring;

oBenefits trap: Too high or too long income replacement levels can diminish job search efforts;

oLow educational achievement and attainment: low skills and levels of qualifications make it difficult to (re-)enter employment.

Insufficient support to the LTU: Low public expenditure on employment services and active labour market policies (ALMPs) and low LTU participation in ALMP measures and lifelong-learning correlate strongly with lower transitions to employment.

Important as they are, the first two challenges fall outside the remit of the proposal. The European Semester continues to address structural barriers on the labour and product markets. The low skills of the workforce will be tackled as part of the same CWP initiative "Promoting integration and employability" under a forthcoming skills initiative, while the Commission Investment Plan addresses investment constraints.

These actions will have to be considered in a more comprehensive set of ambitious reforms needed to facilitate the adjustment of labour markets and foster economic growth. As underlined in the Annual Growth Survey (AGS) 2015, structural reforms, fiscal responsibility and investment are priorities. Simultaneous action in these three priority economic policy areas is critical to restore confidence, reduce the uncertainty that is impeding investment and maximise their strong mutually reinforcing effects.

In this respect, there is no one size fits all solution at the national level. In particular, labour market reforms have to take into consideration country specific institutions, which can help explain divergent experiences across countries in terms of their long-term unemployment rates. Apart from effective active labour market policies targeting the long-term unemployed, including the role of public employment services, other important institutional factors relate to the design of unemployment benefit and social security systems, employment protection legislation, and taxation.

The actions proposed under this recommendation are in line with the country specific recommendations issued in the context of the European Semester and they should be implemented in full compliance with the rules of the Stability and Growth Pact.

The initiative would tackle the quality of support to LTU and its root problems: low coverage of activation support, ineffective design of activation models, both in terms of individualised support to jobseekers and of employer involvement; and the discontinuities in support to the LTU.

Figure 6 below summarises the "problem tree" for the initiative, outlining the problems identified, their drivers and consequences, further detailed in the next section.

Figure 6. Problem tree for the initiative

3.3.Low coverage of activation support

3.3.1.Registration with employment services is declining

Registration with the PES of the long-term unemployed decreased during the crisis and the rates vary widely among MS. Around 73% of the long-term unemployed declare to be registered with the PES, compared to 77% in the pre-crisis period (average 2004-2007). Registration rates are high (above 85%) in countries as diverse as DE, ES, SK, much lower (around 50% or less) in IT, UK, NL, BG and very low in RO (Figure 7) 15 .

Figure 7. Long-term unemployed by Member State, stock and share registered with the PES, 2013

Source: Eurostat LFS 2013, missing data on registration in IE.

Countries with higher levels of LTU participation in ALMPs achieve higher transition rates to employment 16 . Low registration rates limit the access to support and limit the ability to tackle the real stock of LTU.

3.3.2.Participation in active labour market measures is limited

In 2012 available data indicate that LTU over 25 had a participation rate of 13.5% in ALMPs 17 . Involvement in ALMP measures is often linked to the receipt of unemployment benefits and less often to other forms of social assistance. In 17 MS, unemployment benefits end by the first year of unemployment. On average, 24% of the LTU received unemployment benefits in 2013, down from 36% in 2007 18 .

A first driver of low coverage is the insufficient priority for LTU interventions compared to interventions for the short-term unemployed. In spite of making up half of the unemployed, the LTU receive one fifth of support 19 . Much under-investment occurs by design: 22.1% of ALMP expenditure is explicitly (and not exclusively) targeting LTU 20 .

According to the MLP database, only eight countries target more than 30% of ALMP measures explicitly (but not exclusively) to the long term unemployed. Ireland (64.6%) and Finland (72%) are the only MS to target more than half of expenditure on LMP measures to the long-term unemployed. Indeed, more than half of the countries for which the proportion is known (14/25) target less than a fifth of expenditure at the long-term unemployed.

Table 1 – Groups of countries by expenditure on LMP measures targeted at long-term unemployed

Source: DG-EMPL, LMP database (own calculations)

While the share of LTU increased, dedicated expenditure decreased from 28.4% of ALMPs in 2008. Only DE, DK, SE and UK offer a right of placement on LTU support services or measures, achieving above average transition rates, while seven others are using performance targets on LTU activation 21 .

3.3.3.Conditionality of benefits is not always enforced

A second driver of low coverage is the lack of enforcement of benefits conditionality 22 . Current national legislation in 21 MS links social assistance benefits to registration with the PES and in 26 MS to mandatory participation to activation measures. The impact on coverage depends on the degree to which conditionality applies to registration with PES, to participation in ALMP measures and on whether graduated sanctions are applied. In 11 MS, sanctions are not gradual: the unemployed are simply erased from the PES registry and lose the associated benefits for a period of time 23 . This can create perverse incentives for leaving out the long-term unemployed from activation services.

Table 2: Conditionality for social assistance recipients

Sources: ESPN (2015) Integrated support for the long-term unemployed – A study of national policies ; Vern (2012);*Regional variation

Some countries impose unpaid community service activities (BE, BG, CY, HU, NL, PT, RO, SK) or a mandatory participation to activation measures (all MS except EE, EL, ES, FR, SK), sometimes along with sanctions such as cuts in benefits, in cases where the participation is not satisfactory.

PES from 12 MS report data on the share of the registered unemployed receiving social assistance benefits 24 ; they range from a high share in IE (76%) to very low shares in EL (only 1%).

When enforced, activation conditions for the receipt of social assistance have a strong effect on participation in active support and on transitions to employment 25 . However, given the low coverage of activation offers, existing conditionality is rarely enforced.

3.4.Discontinuity in service delivery

The PES primarily focus on unemployed eligible for unemployment benefits. When eligibility ends (on average after 14 months), the LTU may be eligible for non-contributory minimum income or other social assistance benefits 26 . In 10 MS the unemployed may still receive contributory-based benefits after 12 months of unemployment. In general, the average duration of unemployment benefit is much below the maximum duration in most MS.

The main driver behind fragmentation is the lack of coordination of service delivery. The shift from the responsibility of the PES to other authorities leads to potential discontinuity in service provision 27 . In 11 MS, the PES hands primary responsibility to social assistance authorities. In 21 MS both PES and social authorities are involved, creating a potential for ambiguity and grey zones.  

Table 3: Inter-agency coordination of offers for long-term unemployed

Source: ESPN (2015) Integrated support for the long-term unemployed – A study of national policies

Among MS in which several agencies are involved, only seven provide for formalised agreements that regulate the cooperation between agencies (BE, CY, DE, ES, LU, NL, SK). Their scope (data sharing, cross-referrals), depth (degree of management coordination, shared resources, common performance monitoring) and coverage (only in some municipalities or for subgroups of LTU) vary considerably 28 .

Nine MS promote an institutionally coordinated offer of service delivery, in the form of a single point of contact or a one-stop-shop 29 . A single point of contact refers to coordination of front-line services and client intake, through a system of cross-organisational referrals and data sharing. In DK, they integrate client intake, follow-up, employment services and sanctions. Broader one stop shops entail joint provision of several functions in one organisation. In UK, NL (local Werkpleinen within the UWV Company) and DE they also disburse benefits. In FI, municipal LAFOS centres integrate supporting social services, but not benefit disbursement. In NO all functions are integrated 30 .

Blurred responsibilities between PES and social authorities affect the coverage of activation. They could result in "parking" difficult cases or, when agents have incentives to reduce assistance rolls to over-use of ALMPs and "revolving door" effects, with loss of efficiency 31 . Unclear pathways to support lead numerous LTU to de-registration once their unemployment benefit ends 32 .

Limits in coordination also hinder coverage and individualised service delivery. They diminish the capacity to enforce benefit conditionality and hinder access to adult learning, debt counselling, migrant integration support (language courses, recognition of qualifications) and to supporting social and family services (elderly or child care, addiction, health or rehabilitation support), thus limiting activation impact on the personal barriers to work 33 .

3.5.Activation design – Individualisation of service and labour market relevance of support

Effective activation design combines individualised services to jobseekers and to employers and a tailored ALMP offer to increase the involvement of the employers and reinforcing the relevant skills for employability 34 .

3.5.1.An heterogeneous group

As emphasised by the Employment Committee (EMCO) in the stakeholder consultation, the group affected by LTU is heterogeneous. Specific groups such as persons with disabilities and the Roma are more affected. LTU rates are double as high among the low qualified and third country migrants (Figure 8 and 9).

LTU rates increased faster for the low qualified (Figure 8). By 2025 employment for the low qualified is expected to decrease by 10% 35 and that low qualified workers are increasingly out-competed by medium-qualified workers in elementary occupations 36 . Unemployment among the low qualified is three times higher than among high-qualified persons 37 , while LTU increased faster among the low qualified during the crisis.

Workers with low basic skills are 1.8 times more likely to be unemployed 38 and run a high risk of inactivity: only 55% of the low skilled surveyed in the PIAAC study were in employment, showing their fragile link to the labour market. Re-entering the labour market often requires acquiring a formal qualification, but low qualified persons remain more than 4 times less likely to take up lifelong learning opportunities than high qualified persons 39 .

Figure 8: EU28 Long-term unemployment rate (aged 20-64), total and by level of education, 2004 to 2014

Some groups in the labour force are particularly affected by long-term unemployment. In some EU countries, the share of the third-country nationals among the long-term unemployed is particularly high and on the rise, e.g. in Belgium 12%, in Spain 15%, in Sweden 16% of LTU 40 . In comparison, the share of third-country nationals among the economically active population of the latter countries is at about 3% in Sweden and Belgium, and at about 9% in Spain. 41  

Figure 9. Third-country nationals as a share of the long-term unemployed, 2013 and 2008

Source: Eurostat, EU-LFS ad-hoc calculations. Note: low reliability of data for CY, CZ, LU, SI; country selection based on data availability.

The diversity of the group points to a lack of a one-size-fits-all solution. Support should start from each person's specific needs. This requires a differentiated approach and a broad tool box matching the individual barriers to work.

3.5.2.Services are not sufficiently tailored to individual needs

Limits in individualised service are a first driver of ineffective activation design taking the form of insufficient profiling, lack of generalised use of individual action plans (IAPs) and limited use of integration contracts or targeted action plans in social assistance 42 . As illustrated below, many MS lack a coherent system for referral to the support they need and for follow-up.

A pre-requisite for individualised support is an effective profiling of clients, able to identify employment barriers. Profiling allows in turn the differentiation of support, through "client segmentation" within employment services. This aims to better target services at high risk groups, to match individuals to instruments of public action and  enable tailor-made services suiting the individual needs of jobseekers. 43  Services are more efficient when individualised support is based on profiling. This is particularly the case in countries with limited resources, where caseworkers have high workloads 44 .

14 MS currently have a system of profiling clearly linked to client segmentation in the PES, while in 9 other MS the profiling is either under development or partially applied 45

Table 4: Use of profiling and segmentation in the PES, 2014

Sources:  EC (2015) - Identification of latest trends and current developments in methods to profile jobseekers in European Public Employment Services ; Country fiches prepared for the Peer Review Peer Review ‘PES approaches for sustainable activation of long-term unemployed’, April 2014; PES Business Model questionnaire 2014.

While jobseeker capabilities are affected by the unemployment duration, profiling is only used in the first months of unemployment and no re-assessment occurs systematically 46 . When updates are done, they are more likely to be a review later in the process. Moreover, it is unclear to which degree profiling is systematically used when clients are cross-referred between organisations, for instance for long-term unemployed on social assistance.

IAPs are key administrative tools linking the needs and objectives of jobseekers to the offer of support services and ALMP measures. In 10 MS, PES do not offer IAPs to all registered unemployed and 17 lack targets for their completion 47 .

The extent of use of individual action plans differs: while some MS use it for all registered clients, others only use it only for sub-segments or on a programme basis (Table 5).

Table 5: Use of individual action plans within the PES, 2014

Source: European Network of Public Employment Services, 2014 questionnaire on business models.

*To be extended to all clients according to new legislation

Administrative targets for completion of action plans range from 3 days in the UK to 5 months in the Czech Republic. Some PES use action plans only for certain categories e.g. workers younger than 25 and older than 50 unemployed for more than 6 months in Poland or workers below 50 in Belgium (Brussels region), or recipients of unemployment benefits in Portugal. For some other PES, individual action plans are not formalised and come close to job-search counselling sessions (e.g. in Greece, Malta or Romania). Evidence from PES peer learning indicates that in some countries (e.g. in SK, LT) profiling and individual action plans systems within the PES were abandoned under the pressure of high caseloads during the crisis.

Social integration contracts

Social integration contracts offered by social assistance authorities target a segment usually more remote from the labour market 48 . Integration contracts differ from action plans insofar as they focus on addressing the social conditions underlying the unemployment spell and focus on enabling social services – psycho-social counselling, housing support or debt counselling for instance. Integration contracts target beneficiaries of unemployment/social assistance and are drawn up by social assistance agencies , while IAPs are managed by the PES.

12 MS provide integration contracts that target beneficiaries of social assistance, guaranteed minimum income scheme or recipients of specific programme schemes. Except for DE these countries have both IAPs within the PES and integration contracts in the social assistance agencies. Nevertheless, apart from France and Portugal, there is no formalized transition foreseen between action plans drawn up by the PES and integration contracts. Their incidence is hard to assess, as they are rather localised initiatives. This creates a duality of support between PES and social authorities in the 12 MS where they are offered. Two countries developed a seamless approach, integrating them with IAPs in a common instrument (DE, UK).

Most integration contracts provide individualized support for the unemployed. In 7 MS, signing an integration contract is an obligatory requirement in order to receive social assistance benefits. Integration contracts may include a plan defining which steps have to be taken in order to reintegrate the unemployed into the labour market (in BE, DE, ES, FR, LU, PT, SI) or can be linked with activation measures (in BE, DE, ES, FR, LU, PT).

Higher intensity interventions with reduced caseloads, in the range 70-100 clients/ employment counsellor impact positively employment outcomes and client satisfaction 49 . Individual follow-up offers further gains in tackling lack of motivation and complex barriers to work. However, intensified individual follow-up occurs sporadically and generally is programme-based: 19 MS lack dedicated LTU counsellor roles 50 .

3.5.3.Low employer engagement

A second driver of ineffective activation design is the low employer engagement. Effectiveness of activation measures depends on the degree of employer buy-in. 58% of employers surveyed in 5 MS find government support ineffective for dealing with LTU 51 .

Firstly, developing an employer service offer which builds a long term direct relationship to employers leads to improved LTU placement results 52 . The PES allocate between 4% (HR or EL) to 20% (LV or LT) of staff to employer services 53 .

Secondly, employment incentives generally increase LTU transitions to employment but can have high displacement effects 54 . Even if 48% of ALMP LTU participants benefit from employment incentives 55 , such schemes are grossly under-funded. In actual spending on LTU, 26% of the ALMP budgets are spent on employment incentives. This can further the risk for placement in low quality, unsustainable jobs, a concern expressed by the European Trade Union Congress (ETUC) in the consultation 56 .

Often, the MS' nominal commitment to private employer support is not followed through. 31% of ALMP expenditure for LTU goes to public works, more than the expenditure for employment incentives 57 . There is however ample evidence on the ineffectiveness and inefficiency of such programmes on returns to employment 58 .  Public work schemes show however a high degree of resilience. In a survey from 2013, their average duration was no less than 17 years 59 . Public work schemes were expanded during the crisis in HU, LV, LT and CZ and new schemes were adopted in 2014 in SI and EL.

Figure 10: Distribution of expenditure by category of LMP measure and by LTU status, EU, 2012 (%)

Source: DG-EMPL, LMP database (own calculations)

Note: Data not available for EL, CY and UK in 2012. These are therefore excluded from the EU figures

3.5.4.Support is not always leading to increased employability

A third driver of ineffective activation design is the lack of employability relevance of support. Skills mismatches are a root cause of LTU, increasing with duration of unemployment 60 . Broader unemployed participation in training and life-long learning is associated to higher transitions to employment 61 . In spite of this, training is one of the least frequently offered ALMPs, making up 15.7% of LTU participations in ALMPs, corresponding to 20.8% of ALMP expenditure 62 . The joint employers' contribution emphasises the need to align training to the employer's needs 63 .

Figure 11. Distribution of participants in LMP measures by category and LTU status EU, 2012 (%)

Source: DG-EMPL, LMP database (own calculations).

Note: Data not available for EL, CY and UK in 2012, which are therefore excluded from the EU figure.

Workers lacking basic skills or formal qualifications are more affected by LTU. Results of OECD’s Adult Skills Survey show that 20% of Europeans have inadequate basic literacy and number skills and in the case of digital skills the figure increases to 25% 64 Low skilled unemployed are four times less likely to participate in life-long learning 65 . In order to benefit from suitable training, boosting their basic skills is required.

 Support for basic skills is sparingly integrated in support, with 7 MS offering LTU some support for addressing basic skills 66 . Upskilling the low-skilled long-term unemployed is a complex long-term project for which their period of unemployment can be put to good use. The process builds on the assessment of each individual’s skills and skills needs, can include validation and accreditation of existing skills and follows up with referral to a training course/ adult education, preferably including on-the-job training, supported by counselling and guidance throughout the path to stable employment. An essential ingredient for such a complex process is cooperation between employment services and private or public adult education services.

Start-up support increases employability of the unemployed with lower skills or part of vulnerable groups. In few MS, start-up subsidies are targeting explicitly LTU. Even if not leading to higher overall job creation, their re-employment effects are generally positive 67 . Start-up support programmes however represent 8.4% of participations and 4.7% of expenditure of ALMP measures for LTU 68 .

4.The Envisaged initiative

4.1.Objectives

The general objective is to support MS to increase the rate of transitions from LTU to employment.

Within this general framework, the specific objectives of this initiative are to support MS to:

(1)Increase coverage with higher registration and active support for the long-term unemployed

(2)Ensure continuity and coordination between relevant services

(3)Increase the effectiveness of interventions aimed at both long term unemployed and employers;

4.2.Job integration agreements

The proposed initiative combines the introduction of the Job integration agreements (JIAs) for jobseekers at the latest when they reached 18 months of unemployment with general policy guidance on individual assessment and on strengthened links with employers. The threshold of 18 months is chosen due to the large drop in transitions to employment between 12 and 24 months (Figure 4) and to match existing dates for intervention. Most MS proposing similar interventions act within the period (SE and FI at 10 months; UK and DE at 12 months; DK (pilot) at 16 months).

The design of the JIAs builds on three steps which open a pathway to job integration: (1) encourage registration with an employment service; (2) assess individual needs and potential; (3) propose a plan for return to work under a JIA approach. These three elements can be implemented in a flexible way corresponding to the specific situation in MS.

JIAs would consist in a written agreement covering the service offer and the obligations, between the jobseeker and a single point of contact within an employment or social service. To allow a lower caseload and intensified support, the JIA would be targeted to beneficiaries at the latest when they reach 18 months of unemployment, irrespective of age or qualification levels.

The creation of the single point of contact will not require full integration of employment services and (social) benefits disbursement. It could be set up through inter-institutional cooperation, focused on the relation with the client.

JIAs would also include intensified individual follow-up (or 'case management') entailing intensified provision of employment services (counselling, guidance) and step-by-step monitoring of job search efforts.

The policy guidance would serve to place JIA into a minimum frame of action common for all MS:

The (re-)assessment of the individual skills, needs and prospects of all registered LTU before reaching 18 months of unemployment, based on the individual employment and support history.

The re-orientation of ALMPs away from public works towards employer incentives, services for employers and reinforced up-skilling and start-up support for LTU.

The target group would avoid any overlap with the Youth Guarantee. MS which do not have integration agreements along the lines outlined above would be required to set them up for the target group.

Implementation can be expected to be gradual. All MS offer currently support to the LTU, using both national and ESIF financing (primarily ESF). Such support can be re-structured and integrated in JIAs. Countries with consolidated systems will be encouraged through the recommendation to institute a "last call" point of intensification of effort.

A Council recommendation can provide MS specific guidance on the policy pathway, whilst accepting MS discretion in adapting implementing measures to national institutions and delivery models. This is in line with the principle of proportionality in Article 5 TFEU and meets the concerns raised by several MSs on the degree of prescription of the initiative.

The recommendation introduces the approach of Job integration agreements, which will act as a focal point of integration support for the long-term unemployed. Under this approach, the following elements of flexibility for Member State implementation are included:

- deciding the content of the offer included in the Job integration agreement to best tailor the individual circumstances of the beneficiaries and the labour market conditions; it can be focused on employment services, ALMP measures or social support services or a mix thereof;

- choice of organisations providing the Job integration agreements and of the coordinating organisation (governments, PES, regional or local authorities);

- choice of public or private provision of support included in Job integration agreements;

- the organisation of the single point of contact – whether this would be done through cross-referrals, data exchange, or wider organisational restructuring such integration of front-office functions, co-location or transfer of organisational functions between agencies;

- the organisation of personalised services, whether this would take the form of a personal counsellor providing case management or would include on-line channels for follow-up

- the means to reach higher registration, more employer involvement and more relevant support for employability.

5.articulation with existing measures

5.1.Links with other policy instruments

LTU has been identified by the Council (ECOFIN and EPSCO) and the ECB 69 as an impediment to growth and as a key challenge in the 2015 Annual Growth Survey. 

A number of initiatives at EU level are already in place to support employment, notably the European Semester recommendations and Employment Guidelines, the European Structural and Investment Funds – in particular the European Social Fund (ESF) and the European Regional Development Fund (ERDF), and mutual learning.

5.1.1.The EU Semester process

The Council recommendation will support implementation at national level of the European Employment Strategy and of Europe 2020 goals on increasing employment rate and reducing poverty.

The overall objectives of the European Semester process of economic policy coordination are supported by the integrated guidelines, which include both the Broad Economic Policy Guidelines and the Employment Guidelines. The former guidelines cover priorities such as boosting investment, implementing structural reforms, and improving the coordination, sustainability and growth-friendliness of public finances in order to contribute to growth and jobs at EU level. The latter ones cover priorities such as facilitating job creation, enhancing labour and skills supply, enhancing the functioning of labour markets, ensuring social protection and fighting against poverty.

Within the European Employment Strategy, the current Employment Guidelines aim to address structural barriers linked to rigidities on the labour market and inadequate skills or insufficient investment in human capital. Elements in these areas which feature prominently in the country specific recommendations include reducing financial disincentives to work, adjusting wage-setting mechanisms and the benefits, shifting the tax burden away from labour, strengthening active labour market policies and modernising vocational education and training and apprenticeships.

In addition, the newly proposed Employment Guidelines propose to reduce LTU "by means of comprehensive and mutually reinforcing strategies, including the provision of specific active support to long-term unemployed to return to the labour market" 70 .

The 2015 country specific recommendations 71 acknowledge that LTU needs to be tackled by combining effective social assistance with employability. In this respect, they stress that efficient and effective vocational education and training programmes, including targeted adult learning play a key role in improving employability. They also note that the role of the PES needs to be strengthened in some MS and that active labour market tools could be more efficiently targeted to support the inclusion of those furthest from the labour market.

Structural reforms of labour markets will have an impact on the LTU level in the medium to long term, as policy transmission channels take time to impact the labour market. By adding a focus on LTU, the current proposal can support that this group benefits from positive developments on labour markets quicker than otherwise, in combination with structural labour and product market reforms. Moreover, in the context of the streamlined European Semester, country specific recommendations (CSRs) focus on issues where reform is macro-economic relevant and requires urgent attention 72 . Thus, LTU might not be mentioned in the CSR even if LTU is high in a particular MS. By providing specific guidance on the process of addressing LTU, the recommendation complements the analysis and guidance provided in the European Semester through country reports and CSRs.

A common action framework aimed at increasing the performance of all MS would offer a basis for subsequent analysis of reforms and progress. For countries addressing CSRs related to LTU or related areas (activation or active measures), the recommendation provides further guidance supporting their implementation effort. For countries with high LTU levels who do not have CSRs in the area, it provides an EU level impetus for improvement of support to long-term unemployed.

While the European Semester follows the structural reform priorities for each MS, the initiative would complement this through additional guidance, joining up different reform streams in the employment and social field with PES and social services capacity building. By reducing divergence in LTU rates, the initiative would strengthen the social dimension of the Economic and Monetary Union (EMU).

5.1.2.Support to the long-term unemployed through the European Structural and Investment Funds

ESF-supported activation measures were equivalent to around 20% of the total ALMP expenditure in the EU between 2007 and 2013 73 . The LTU account for 29% of the unemployment allocations supported by the ESF, slightly higher than in corresponding national ALMPs 74 . Relative to the high LTU rates of the period, the same pattern of under-investment applies, particularly in MS with high LTU 75 .

Table 6. Share of ESF allocations to unemployed and long-term unemployed

Source: DG Employment estimates based on past and current ESF programming

On the basis of current programming for 2014-2020, the unemployed will constitute at least 41% of the allocations supported by the ESF; at least 25% of these will be for the long-term unemployed 76 .

The 2014-2020 ESF allocations allow room to increase the financing of LTU support and to enhance its efficiency. Under investment priority 8i on active labour market measures, 11.08 billion Euro were programmed for 2014-2020, with particularly high shares in SK, FI, and IE. Under the investment priority 8vii on increasing the capacity of labour market actors, 980 million were programmed, with higher allocations in IT and RO. 12,93 billion Euro were programmed under the social inclusion priority (9i), with higher shares in NL, FR, IE, BE.

Early results from ex-post ESF evaluation identify on the basis of interviews with programme managers the following success factors for programmes addressing disadvantaged groups: (1) effective referral, engagement and outreach approaches; (2) provision of intensive support; (3) ensuring access to a relevant range of services; (4) ensuring continuity in support to participants 77 . The recommendation tackles through the approach chosen (Job integration agreements) the entire spectrum of improvements identified in the evaluation of the ESF. It reinforces the referral system for long-term unemployed to the support needed, it encourages intensive, personalised support, and opens access to a broader range of services, under a contractual framework which entails continuity of support and follow-up. This goes beyond the usual reach of ESF programme-based interventions, as it tackles the basic set-up of interventions, on which further ESF interventions can be built.

This also shows that with additional changes in the support delivery through the implementation of the recommendation, existing ESF budgets could be spent more efficiently, in particular in addressing issues linked to continuity of support and individualised referrals to support.

While the European Social Fund (ESF) is the EU's main financial instrument for tackling long-term unemployment, other funds like the ERDF may also support the objectives of the initiative. In addition to the measures financed by ESF, ERDF measures within the existing operational programmes for 2014-2020 can support investment in infrastructure for employment services (investment priority 8d), vocational education and training and skills and lifelong learning (investment priority 10), as well as financing for self-employment and start-ups (investment priorities 8a and 3a).

Ex-Ante conditionalities can also play a role in making sure that jobseekers (including LTU) can be better supported 78 . A number of MS will have to fulfil them by the end of 2016 at the latest and report on fulfilment in 2017. Both ex ante conditionalities 8.1 and 8.3 79 focus on the capacity of employment services to deliver: (1) personalised services and active and preventive labour market measures at an early stage, which are open to all jobseekers while focusing on people at highest risk of social exclusion, including people from marginalised communities; (2) comprehensive and transparent information on new job vacancies and employment opportunities taking into account the changing needs of the labour market; (3) the setting up of formal or informal cooperation arrangements with relevant stakeholders.

The recommendation provides guidance which supports MS who are engaged in strategic changes to meet the ex-ante conditionalities. The first and second ex-ante conditionalities are closely related to the first and second specific objectives of the Council recommendation. The establishment of JIAs will help implementation of both principles embedded in the conditionalities. However, while ex-ante conditionalities can play a role in improving the policy framework for effective implementation, they do not focus on long-term unemployed as a specific group and do not cover all elements of service delivery.

The envisaged initiative could provide further guidance on how to best achieve results with LTUs when designing the framework and when implementing the operations. The ex-ante conditionalities and the initiative are reinforcing each other and do not duplicate.

5.1.3.Mutual learning at EU level

In terms of sharing of good practices, actions supporting Europe 2020 reform implementation through policy learning are taking place and will continue (for example, regular peer reviews on LTU and active labour market policies as part of mutual monitoring within EMCO, as well as best practice exchange under the Mutual Learning Programme 80 ).

The PES Network is currently developing a "benchlearning" process to provide an overview of the performance of PES in all MS, in order to encourage mutual learning (peer-to-peer or group activities, analysis, cooperation, and exchanges). This will contribute to capacity building and increased awareness within the PES on effective ways of developing support for LTU. However, mutual policy learning takes place at a technical level and needs support at political level to ensure implementation of the agreed solutions. In addition, capacity building in the PES cannot influence the composition of ALMPs and has limited influence on social assistance authorities. Therefore, the above mutual learning processes are not sufficient to incentivise structural and process changes at the required level in the MS.

5.1.4.Active inclusion initiatives and fundamental rights

Previous Commission initiatives focused on the active inclusion approach, encompassing outreach to groups such as the inactive or non-registered long-term unemployed that are the furthest from the labour market and face multiple barriers 81 .

The initiative reinforces protection of fundamental rights. Improving the access to support services and establishing an individualised services approach for long-term unemployed guarantees and further enhances the right to access to free placement services laid down in Article 29 of the Charter of Fundamental Rights of the EU. Furthermore, providing better protection of long-term unemployed duly observes the principles enshrined in Article 34 of the Charter of Fundamental Rights of the EU, since it improves the entitlement to social security benefits and social services protection in case of loss of employment, helps combating social exclusion and ultimately reinforces human dignity.

Moreover, better individualised support, coordination of delivery, assistance for return to employment, training and for work experience would also contribute to a better implementation of the UN Convention on the Rights of Persons with Disabilities.

5.1.5.Reforms undertaken at national level

LTU affects MS to a different extent, notably as the impact of the crisis has been uneven and countries differ in terms of macroeconomic situation as well as economic structure and labour market functioning. The policy response has also been uneven.

Economic recovery and lower inflows to LTU and the improved budgetary situation open up opportunities for additional interventions. However, LTU is addressed by current reforms in less than half of the MS. Many well performing MS (e.g. EE, FI, FR, NL, SE) continue to improve their support, while other MS with high LTU rates have not undertaken initiatives (e.g. RO, EL, HR, SK), or the initiatives do not address the main problems (e.g. the launch of new public works programmes in EL, HR, SK). The momentum of reform in the MS most affected by unemployment is not likely to lead to a substantial change either in terms of priority given to LTU or in enhancing service delivery. This risks to increase the divergence in the impact of support and consequently in LTU transitions.

Several MS launched additional ALMP programmes, combining services and counselling, provision of programme income, and additional subsidies for employers 82 . While such initiatives can have a positive impact, they are limited, addressing the coverage problem with less focus on the problems of discontinuity and activation design.

After dropping for the last 5 years by a total of 10%, expenditure on public employment services began stabilising during the last two years. In 2014, a small increase of expenditure on public employment services occurred in most MS 83 .

Creating a coherent referral and support system for LTU requires structural holistic change. Such changes occurred only in a handful of MS during the last decade (CZ, DE, EE, IE, FI) 84 and none of the countries with above average LTU rates undertook the process.

5.1.6.Additional action needed

Addressing the specific challenges linked to long duration of unemployment requires additional policy guidance and a more coordinated approach at EU level in addition to and supporting the existing frameworks.

Existing national and EU initiatives can be expected to address the problem of insufficient coverage only partly. Smaller inflows to LTU during recovery and the launch of additional programmes in some MS can allow a partial re-direction of ALMPs towards LTU. Additional ESF support explicitly targets the LTU in the MS which received CSRs in the area in 2013, further reinforcing provision. However such interventions are mostly programme-based, and cannot be expected on their own to systematically improve the coherence of the support system and the enforcement of benefit conditionality.

The problem of lack of continuity is likely to be least addressed by current initiatives. Recent national reforms led to better coordination only in a few MS, e.g. in ES through joint evaluation of ALMP programmes and the introduction of performance criteria for the allocation of funds on in IT through the start of the reinforcement of a central PES.

The problem of ineffective design of interventions can only be partly alleviated through existing mutual learning. The PES Network can influence performance management in the PES and models for delivery of both individualised services and employer services. However, the establishment of partnerships and lasting cooperation between different authorities, as well as the required shifts in active measures, go beyond the PES remit.

Divergent labour market national performance in terms of LTU rates and transition from LTU to employment may have both negative economic impacts and undermine social cohesion across the EU.

The persistence of LTU can have a dampening effect on economic growth and is detrimental to potential output growth in the medium term. The longer the active population remains unemployed the more the risk that skills will be lost and the LTU may leave the labour force. This would imply both a waste of resources and an increase in social costs. A long period of unemployment can lead to an increase of NAIRU (non-inflationary rate of unemployment) and a reduction in the growth potential of the most affected MS and of the EU as a whole 85 . A lower growth potential renders it more difficult to reduce public deficit and debt levels which exploded during the crisis. High LTU is thus a source of rigidity and further vulnerability in the economic recovery.

Social cohesion is one of the Treaty's general objectives. However, the problem definition has demonstrated that the current national divergences regarding LTU support can grow further. Persistent long-term unemployment leads to skill erosion and obsolescence and lower utilisation of human capital in a context of demographic decline. It is a strong source of social exclusion and a serious obstacle to the achievement of Europe 2020 goals.

In conclusion, additional action is required: "The situation of young people and the long-term unemployed in particular requires determined action. Benefit systems should combine adequate income replacement with activating and enabling services targeted at individual needs and delivered through one-stop shops (AGS 2015)".

The recommendation would complement and support action at national level since:

a) It will create a Political Momentum

The Youth Guarantee has shown that a recommendation adopted by the Council increases ownership and gives a stronger political signal than cooperation at technical level. A politically visible commitment centred on one delivery approach with citizen visibility deepens mutual learning and creates stronger impact than purely technical exchange on good practice.

Recognising the relevance of an EU level policy response, the Council invited the Commission "to develop proposals to help support the long-term unemployed, taking lessons from the introduction of the Youth Guarantees across the EU, while fully integrating them into the European Employment Strategy".

b) The issue concerns all the MS

The performance of MS in supporting LTU towards re-employment varies significantly, partly because MS with low performance have not put in place adequate and effective integration policies, but also due to other factors such as overall macroeconomic performance and rigidities in product and labour markets. There is scope for performance improvement in most of the MS to address in an integrated way the three problems identified (coverage, discontinuity, activation design), thus allowing the economic recovery to translate more swiftly into a return to work for more of the long-term unemployed.

The initiative will cover all MS with the aim to support a general improvement in the efficiency and modernisation of the integration process. It translates the lessons from mutual learning into an action framework aiming to raise service delivery performance in all MS. This should contribute in aspects related to LTU to the overall EU efforts towards upward social convergence.

c) Need to reinforce and complement existing EU level interventions

The recommendation will complement and reinforce the effectiveness of the existing tools. Providing a foundation for prioritising support aimed at job finding for the long-term unemployed at national level, it encourages the allocation of necessary resources, including support from the European Structural and Investment Funds. It also provides an impetus for greater administrative capacity, better cooperation between service providers and efficiency in public spending.

5.2.Lessons learned from the Youth Guarantee

The Youth Guarantee (YG) was established by the Council in 2013 to address high levels of youth unemployment and inactivity resulting from the crisis and to improve school to work transitions. Similar to the current initiative, the YG proposes an intervention model based on an individual, coordinated offer. The current initiative also includes a three–step approach: registration, assessment and an offer, similar to the YG model.

A full evaluation of the early impacts of the YG will be available in 2016. However, the evolution of youth employment following adoption and implementation reports already suggests a positive impact.

The adoption of the Council recommendation and the political momentum that it created enabled the adoption of Youth Guarantee Implementation Plans by each Member State. Already the drawing up of these plans and the setting up of the YG schemes created new structures (for example YG coordinators) and ways of working together in the MS.

The recommendation acted as basis for many of the CSRs on youth employment in the first European Semester in which the implementation of the Council recommendation was monitored. In 2014 there were 18 CSRs on youth employment (of which 8 explicitly mentioned the Youth Guarantee).

The Employment Committee (EMCO) – in its preparation of Council deliberations - pursues multilateral surveillance on Youth Guarantee implementation. To underpin monitoring and multi-lateral surveillance, an Indicator Framework for Monitoring the Youth Guarantee was developed. This form of multilateral surveillance within the Council and across MS not only provides for thorough evaluation and continuous improvement of schemes, but also enables peer learning, creates peer pressure and strengthens ownership of the MS.

The implementation of the YG acted as a driver for structural reforms in MS and has already lead to first successes on the ground. Reforms, including for example the stepping-up of early activation and intervention measures, the strengthening of Public Employment Services for the delivery of the YG, a renewed focus on the quality of education and training systems and better school to work transitions could be observed in many MS.

Since the inception of the YG there has been a noticeable decline in youth unemployment. In April 2015, compared with April 2013, youth unemployment decreased by 935 000 in the EU28. The youth unemployment rate was 20.7% in the EU28, compared with 23.9% two years ago.

The monitoring and regular reporting on developments concerning the design, implementation and results of Youth Guarantee schemes are part of the annual programme of work of the European Network of Public Employment Services. Qualitative reports on the impact of the YG on service provision in the PES demonstrates that a Council recommendation can produce significant changes to support delivery models, leading to increased effectiveness of delivery. Most PES recognise a significant impact of the initiative on their delivery capacity 86 .

Within the PES, implementation lead to a mainstreaming of a partnership approach (in 25 PES), to resourcing specialist youth caseworkers (in 18 PES) and specific management targets (in 18 PES). Two thirds of PES have engaged in outreach work aiming for higher registration rates, including through proactive partnerships and awareness campaigns and increased use of e-services 87 .

Almost half of PES have also developed pilot projects on youth employment in 2014. These include a strong focus on disadvantaged and vulnerable youth groups. Over half of PES also provided staff with training on specific aspects of working with young people in 2014.

Since April 2013 (date of adoption of the Youth Guarantee recommendation), the youth unemployment rate in the EU started to decrease: from 23.9% or about 5.7 million people (April 2013) to 20.7% or about 4.75 million people in April 2015. The youth unemployment rate fell more than one would expect on the basis of the relationship between youth unemployment and GDP growth.

Based on the experience of the Youth Guarantee, the design of the initiative addresses a number of points of concern raised in the European Court of Auditor's Special report on the EU Youth Guarantee . By stating clear requirements for a JIA and by defining as a clear outcome re-employment, the initiative gives orientation on a "quality offer" to the jobseekers 88 .

Secondly, by proposing a monitoring framework from the outset (included in Section 8 below), the initiative would allow tracking progress in implementation from the very beginning.

It is crucial for ensuring a genuine individual approach that the choice of the offer done within JIAs is not pre-defined. The type of support offered (provision of employment services, of ALMP measures such as employment incentives, training, start-up support, or referral to social services or further education) should be defined based on the individualised assessment and on local labour market needs. While allowing autonomy in the choice of offer, it is important to ensure accountability and value for money through a sound monitoring framework. The quality of employment outcomes in terms of sustainability of re-employment and of the type of labour contract are captured by the indicators proposed.

6.Effectiveness of the envisaged recommendation

The analysis of the overall effects is based on experience from MS that have already introduced similar measures. In the analysis of coordination of support, individualisation and employer relevance (Objectives 2 and 3), both the mechanism of impact and the effectiveness in increasing transition rates are presented, privileging among the sources used published evaluations with control groups and systematic reviews. While for each sub-section the analysis integrates evidence from evaluations in 5-8 MS, generalisations to EU level impact should be read with caution, as the fit of each scheme and the effectiveness may vary across MS.

6.1.Increasing coverage of activation support (specific objective 1)

Reinforced support in a single point of contact has an outreach effect. The experience of countries that established one-stop-shops confirms its positive effect on extension of both registration and coverage of employment services 89 .

The table below presents the target group, based on up the stock of registered LTU over 30 unemployed for at least 18 months, as well as the effective target group (excludes countries which offer JIAs or quasi-universal coverage of LTU (DE, DK, SE, UK). Data for IE is missing.

Table 7: Coverage of LTU support

Source: Own calculations based on LFS 2013 data, ad-hoc transitions calculations and the LMP Database.

It is not possible to provide an ex-ante assessment of the expected increase in coverage of services and measures, based on the introduction of the offers within the JIAs. Experience in the countries which have already introduced JIAs shows that more precise targeting of individual needs would lead to a significant shift of resources from ALMPs to employment services. As the cost ratio of employment services to ALMPs can be roughly estimated at 3:1, this would entail an increase of coverage of long-term unemployed through employment services. MS retain however flexibility in the establishing the balance between ALMP measures and employment services in their provision mix, as well as control of resources allocated to LTU relative to other groups.

JIAs targeted by duration (at the latest when having reached 18 months of unemployment) offer a balanced impact geographically, with relatively high impact in weak support countries and thus lowest deadweight loss. Given that on average the low skilled and the older workers have higher registration rates with the PES, increased coverage by duration will also benefit these groups.

6.2.Continuity and coordination in support delivery (specific objective 2)

Two mechanisms ensure continuity in support delivery: 1) A systematic referral system to the support needed and 2) Data sharing mechanisms across organisations.

Degree of coordination already in place:

Based on Table 3 above.

Table 8. Coordination of support delivery

6.3.Increased effectiveness of interventions: individualised support (specific objective 3)

Three mechanisms ensure a stronger individualised support: 1) A minimal standard of service ensuring that administrative processes achieve tailored support, matching the needs and goals of jobseeker, while allowing case handler discretion and choice; 2) Re-orienting ALMPs towards employability support (up-skilling/basic skills/start-up support); 3) Re-assessment of the needs and employment prospects of the LTU.

Impact will be higher on MS without individual approaches in place:

Base on tables 4 and 5 above.

Table 9. Individualised support

6.4.Increased effectiveness of interventions: employer support (specific objective 3)

Contact to employers impacts transitions to work. Among employers who did recruit from the LTU in the five countries covered by the IPPR survey, 56% recalled contacts with the PES, compared to 21% who did not recruit from the LTU 100 .

Two mechanisms ensure stronger employment support: (1) enhancing services to employers and (2) focusing ALMPs on the competitive labour market, reducing to a minimum public works. There is consistent evidence on the lack of effectiveness of public works on promoting transitions to employment 101 .

Impact is expected to be higher for countries with large public work schemes:

Source: Own estimates, based on LMP Database, 2012.

Table 10. Employer involvement

7.Expected impacts and costs

7.1.Expected impact on transitions to employment

Transition rates depend not only on the factors tackled by the initiative, but also on a range of exogenous factors including the broader macroeconomic environment, the degree of flexibility of the labour markets, the level and relevance of skills in the workforce. All these factors are highly country specific and therefore are not taken into account in the estimates presented below.

JIAs offer a framework for upward convergence of support to the long-term unemployed in the MS. The possible convergence effect of the initiative can be presented under two scenarios: (1) the gap between the current transition rates and the top 3 MS in transition rates (DK, EE, SI) is closed by half; and (2) the gap to 3 countries with support very similar to the proposed JIAs (DK, DE, NL) is closed by half. 108 The scenarios do not control for other intervening factors (increased labour demand, labour market flexibility, etc.) and are not based on a modelling of impact in the MS. They illustrate numerically the potential gains from a process of convergence in the quality of support 109 .

Table 11. Estimated additional LTU transitions to employment

Numerical annual estimates based on the assumption that the gap between the current transition rates and the top 3 MS in transition rates (DK, EE, SI) is closed by half (scenario 1 in first row) or that the gap to MS with support systems similar to that proposed by the recommendation (scenario 2 in second row).

Own estimates, based on Eurostat transition experimental transition data and LMP Database 110  

Based on the more conservative scenario (2) above, the potential annual gains in transitions to employment are illustrated below for the MS most concerned for which both transition data and ALMP data are available.

Table 12. Estimated additional LTU transitions to employment, by country

Numerical annual estimates based on the assumption that the gap between the current transition rates is half closed relative to MS with support systems similar to that proposed by the recommendation.

Own estimates, based on Eurostat transition experimental transition data (2013/14) 111 .

While transition rates vary also due to specific labour market conditions, a better focus of integration policies would have impact independent from the macroeconomic situation and labour market structures.

Figure 12 presents both the positive impact of the good structured support on the transitions to employment, as well as the impact of reforms on increasing transitions. Countries which have already increased individualisation of service work systematically with IAPs and have ensured data sharing for service coordination, thus offering more structured support to LTU achieve higher rates of transitions to employment (e.g. DK, EE, SE, CZ, FI, DE IE in Figure 12) 112 . Moreover, countries which reinforced support structures show considerably higher transition rates following reforms, as illustrated by CZ, EE and IE 113 . This suggests the potential for improvement, although the figure below should be interpreted with care, as other factors such as macroeconomic trends or wider labour market reforms may affect performance.

Figure 12. Support to LTU and transitions to employment

Source: Eurostat, EU-LFS, experimental ad-hoc transitions calculations. For CZ, EE and IE, data for both 2008/09 and 2013/14.

In light of the figure above, the scenarios presented at the beginning of this section appear realistic. This is also in line with research outcomes on support for highly disadvantaged groups of unemployed where it is not unusual to find that the most successful programmes increase by 10% their employment rates 114 .

7.2.Expected impact on poverty and social inclusion

Long-term unemployment is one of the most significant determinants of poverty. Half of those who take a job also escape the risk of poverty, with a share varying from 20% in BG and RO to 65% in PT or SE 115 .

Chances to exit poverty are linked to the chances of successful transitions to employment. Figure 13 shows that on average half of those taking up a job exit from poverty. Transitions out of poverty depend amongst others on the type of contracts (permanent/ temporary, part time/full time), on wage levels and on the availability of in-work benefits 116 .

Figure 13: Transitions to employment and exits from poverty

Source: EMPL calculations based on Eurostat, EU-SILC 2010 longitudinal data; data for DE and IE missing. ( ESDE 2013 ).

While the initiative would not target specific groups (such as persons with disabilities or partial work impairments, persons from migrant background, Roma, single mothers), strengthening individualised approach leads to more targeted solutions and improved ascertaining of support for return to employment. Support would be embedded in a simpler institutional environment improving the identification of specific barriers to labour market inclusion and providing a gateway for access to supporting social services, often needed to address multiple disadvantages. Given that the initiative is living flexibility to MS and focusing to the support infrastructure, the risk of diverting support away from groups more remote to the labour market emphasised in the consultation by the Social Protection Committee, is minimised.

The impact on the overall level of social protection would be positive. Increased transition rates would relieve pressure on the social benefits budgets 117 , but on the other hand more integrated provision can increase the number of claimants registering with the integrated services. Simplified service delivery leads to higher registration not only with employment services, but also with social authorities 118 .

Successful transitions to employment will mitigate the negative social consequences of LTU: lower educational achievement both for workers and for their children, family instability, deterioration of mental health, increased mortality and lower access to healthcare 119 . At societal level, LTU is closely correlated with increased inequalities, poverty rates and social exclusion, decreased trust in public policy and social cohesion 120 .

7.3.Overall economic impacts

The initiative is likely to have a positive impact on EU competitiveness. Decreasing LTU would increase disposable income, increase aggregate demand and lower fiscal imbalances. The overall budgetary impact of the initiative can be expected to be positive by reducing benefit recipiency and increase of tax receipts, provided that other structural reforms are implemented to boost job creation. Offering more individualised support would improve the human resources matching of labour to productive opportunities.

Ensuring more transitions to employment and bringing the LTU closer to the labour market would facilitate labour market transitions and re-allocation of labour across sectors, contributing to a more flexible labour market.

Addressing LTU would also increase the contribution of labour to growth – which has been rather low during the last 6 years - and alleviate some of the demographic pressure currently building up 121 .

An indirect effect of the JIAs is the opening of employment services market to a plurality of delivery agents. By creating or further reinforcing the individual referral system, they lay the groundwork for performance based outsourcing of employment services 122 . Research results show that past introduction of single points of contact led to an increase in private employment service provision. 123  

By emphasising simplified support, enhanced services and partnerships with employers, the initiative will impact positively on the business sector. Advocating a shift away from public works will help re-focus ALMP resources towards the competitive labour market and the needs of private employers.

In a setting of low labour demand and limited fiscal space, employment incentives can be attractive instruments for supporting employment while also improving the employability of specific vulnerable groups. However, if not well designed, employment incentives may result in wasteful spending of public money, particularly if the subsidised jobs would have been created/maintained anyway (deadweight loss), or if the subsidies simply induce employers to hire/keep certain workers instead of others (displacement effects). They may also reduce productivity to the extent that they create incentives for companies to substitute more productive workers with less productive ones. If jobs are created only for the duration of the subsidy, this still improves employability of the involved workers, but the net social benefit is much more limited, or even negative 124 .

For these reasons, the foreseen initiative emphasises more employer services (e.g. pre-recruitment selection of candidates, pre- or post-recruitment workplace training, post-placement counselling and mentoring) rather than financial incentives for employers. While such services can still carry risks of deadweight loss, they are less likely to produce distortions on the labour market.

Reinforced and personalised employer support is also likely to be more relevant for the small and medium enterprises (SMEs). Particularly small enterprises without HR departments can be expected to benefit more from employer services and direct contact with employer counsellors. Currently, .medium and large enterprises (100-1000 employees) were found to have more contact with the PES than small or very large ones 125 .

7.4.Implementation costs

MS need to continue to finance the IT infrastructure for benefit and ALMP eligibility and monitoring, and the administrative routines for handling services, monitoring and sanctions. Current costs for delivery of LTU support within the PES, social services or by outsourced provision are not available 126 . The costs for LTU ALMPs average at 8.900 Euro/participant-year, varying from over 1000 € in RO, SK to over 16.000 € in SE, NL, DK. Such costs of basic service provision are likely to continue to vary strongly across Member States, with more investment needs in countries with lower administrative capacity.

The initial costs for establishing the JIA infrastructure are the costs related to the organisational changes to achieve better coordination in order to offer an integration agreement through a single point of contact. They relate to IT, new administrative protocols, equipping or co-locating one-stop shops and staff training and are dependent on the extent of prior collaboration between the agencies in charge, which varies as shown in section 3.4 above 127 . Available evidence suggests that the initial additional financing does not vary strongly with the level of capacity of national administrations. In Germany it amounted to 175 € per household in 2005, in Hungary to 300 € per person and for the case of Jobcentre Plus in the UK the roll-out costs were around 275 € per benefit recipient in 2008 (excluding infrastructure investment costs which were around the same amount). The running costs of offering the JIA offers are however more dependent on the administrative capacity of the administrations in charge.

The ICON study shows that running costs per participant vary considerably and are lower in countries with lower capacity 128 . They are mainly related to staff re-assignment to ensure lower caseloads, while the ensuing administrative procedures of provision of job offers, referral to ALMP measures or monitoring entail only small additional costs, as in many cases they only need to be adapted based on current practice and integrated in the JIAs.

For both initial and running costs, there are strong counter-financing effects through better integration rates into the labour market (shorter unemployment duration) and savings on benefits through better monitoring under JIAs 129 . The single point of contact leads to savings through joint direct customer contact in employment and social services.

Based on the ICON study, estimated additional costs are presented in more detail for 5 MS which differ in the extent of service provision to LTU, in resources and administrative capacity and in institutional setting (Table 13) 130 . The estimated costs are already incurred in Germany and Denmark, who would not need additional initial investment to fulfil the requirements of the recommendation. The running costs are built on a scenario assuming an individual qualified contact person for each beneficiary, the assessment of employment potential, the agreement of a JIA opening access to the full spectrum of ALMP measures/ provision of job offers and regular follow-up assuming a contact intensity of 3-6 months and caseloads between 1:150-1:250. They can be considered a minimum package of individualised support.

In addition, the table estimates in the last row the additional costs of intensified intervention should Member States opt for an ambitious implementation of the recommendation, by considerably reducing caseloads for counselling the long term unemployed to 1:70-1:100. Such option would have the consequence of increasing the staffing of the implementing agencies.

Table 13: Estimated additional costs in five EU MS

* UB2-benefit recipient in Germany, LTU in Lithuania and registered unemployed in all other countries ** Depending on the data source used for calculating costs *** Additional costs for intensified (70 LTU jobseekers/caseworker vs. 250 LTU jobseekers/caseworker in the row above on direct costs). Source: ICON (2015) .

While the table above shows estimates of both initial and direct costs per capita of provision of JIAs, the estimates should be interpreted with caution.

The recommendation asks MS to put in place Job integration agreements which should fulfil certain standards, but does not prompt MS to provide a certain type of offer (e.g. counselling, training, work placement, referral to other services). The re-allocation of resources from public works could create additional budgetary space which could be used in interventions targeting employers on the competitive labour market. The cost of implementation in terms of expenditure on ALMP measures is therefore not included in the cost estimates in Table 13 above. The costs of additional counselling, translating into more intense intervention, are presented in the last row. This is fully in line with the objective of increasing the individualisation and effectiveness of support, which means that support should be adapted to the needs of beneficiaries and not pre-determined through administrative rules or targets. The estimates of running costs of provision should therefore be taken with caution, as they are subject to subsequent choices at MS level on the implementation tools.

In light of the flexibility given to MS on the type of offers, a price estimate with regard to their content is not appropriate. It can be generally assumed that many of the offers included in Job integration agreements would be focused on employment services, rather than placement in ALMP measures (such as training, employment incentives, etc.). The data in LMP database shows that there is very wide variation in the cost of such services (whether provided internally in the PES or by other providers). Costs per capita average for instance between 31 and 1183 Euro/ participant in SI, or between 31 and 2655 Euro/participant in DE, depending on the complexity and intensity of interventions. Similarly, the average expenditure ranges from below 150 euro in PL or HR to over 1000 Euro in BE or DE. The optimal choice of offer should be therefore chosen by the implementing organisations in the MS, given individual circumstances and conditions on local and regional labour markets and cannot be adequately priced at EU level.

Overall, implementation would entail additional costs related mainly to the initial set up of coordinated provision of support offers through JIA and the single point of contact. The running costs of provision, including both the initial assessment and the follow-up of JIA recipients, as well as the costs of offers (whether taking the form of employment services or placement on ALMP measures) can be expected to be covered mainly through re-allocations from existing expenditure on labour market policies and a stronger prioritisation of the LTU in the service offer. A precise estimate of these running costs is difficult given the high variation in the costs of such offers across MS and because of numerous offsets and possibilities to re-allocate within existing budgets. It is fair to assume that the initial investment would translate into gains ensuing from reduction of long-term unemployment at a later stage, as discussed in section 7.5 below.

7.4.1.The role of the European Structural and Investment Funds

Part of the costs associated with implementing the foreseen changes could be mitigated by directing existing ESF budgets towards the actions envisaged in the initiative. According to the ESF Committee opinion, substantial amounts of the ESF are available for putting in place integrated LTU support structures and for the groups furthest away from the labour market 131 .

All MS are in a position to act. MS with more limited fiscal space and in need of deleveraging can make best use of ESF allocations. ESF funding dedicated to improving the capacity of labour market institutions is already budgeted in the Operational Programmes in BG, CY, CZ, DK, ES, FR, EL, HR, IT, PL, PT, RO, SK, covering all countries which received CSRs on employment service capacity during the last three years.

JIA will impact differently provision of support in the MS. While countries which already have strong support systems in place will need to invest less to deliver JIAs, countries which are more exposed to LTU and also have a weaker fiscal position will need to invest more. However, the ESF allocations already follow the strength of the support system, as shown in the table below.

Table 14: ESF allocations by expected impact on administrative costs

The one-off administrative costs of full implementation of initiative 132 in one year would amount to 11% of average ESF annual allocation under the three relevant thematic investment priorities (8i, 8vii, 9i) in HU, 17% in PL, 27% in LT, 28% in IT, respectively 43% in FR 133 . The initial costs are likely to be much lower in all MS.

In addition to the ESF measures, MS may also make use of ERDF under their existing allocations for the relevant thematic investment priorities to support the implementation of the initiative, namely by supporting job creation, and modernisation of public employment services and vocational education.

7.4.2.Monitoring costs

No additional reporting envisaged either for businesses (including private employment service providers) or the voluntary sector.

For the public administration, the introduction and the implementation of JIAs involve substantive costs affecting the essence of their activity, and they were analysed in the section above. While the initiative imposes administrative costs primarily on public administration, it will impose a very limited additional administrative burden related to reporting costs.

The additional administrative burden for public administration would relate to the monitoring and evaluation of the initiative. 134 Existing reporting and monitoring mechanisms will be privileged.

National monitoring of the implementation of JIAs would become an integral part of the existing mechanisms for ALMPs, employment agencies and social authorities. EU level monitoring would partly use existing EMCO indicators and the existing LMP database.

7.5.Budgetary impact

The medium-term budgetary impact could be limited or positive provided that other structural reforms are implemented, as the costs for running JIAs could be outweighed by increased tax receipts and benefit savings. Most of the estimated initial costs for setting up the JIA infrastructure would be recovered in one year through increased transitions to employment, assuming that these occur in a recovering labour market characterised by net job creation. There is scope to use the ESIF, in particular the ESF and the ERDF 2014-2020 allocations for supporting the implementation of the initiative.

To assess the gains of the initiative, estimates have been adjusted from increased exits from unemployment in terms of savings on benefits and additional tax receipts, assuming a return to work with a 75% of average wage 135 . For each LTU finding a job, associated gains in tax receipts and reduced benefit expenditure range from 440 €/year in RO to over 11.880 €/year 136 .

These estimates assume that there are no substitution effects, i.e. that more transitions from long term unemployment to employment do not hamper the chances of the short-term unemployed to find a job. Evidence on such substitution effects is mixed and seems highly dependent on the types of offers, with higher effects for demand-side offers such as employment incentives and much lower effects for supply-side offers such as counselling or training 137 . Moreover, substitution effects diminish over time in an expanding economy, as capital adapts to higher levels of human capital and as the lower costs of matching for employers translate in more hirings at the same wage level 138 .

Even when strong substitution effects occur and even when the return to work is not sustainable, there is a net increase in human capital when the long-term unemployed are hired. Short-term unemployed workers have higher levels of employability and higher probabilities of re-employment. In contrast, for the long-term unemployed even a limited period of work gives a considerable boost to their employability 139 .

If combined with other structural reforms to boost job creation, the channels through which this recommendation can overall support the increase in employment rate are multiple (i.e. higher growth potential through lower hysteresis effects, decrease in structural unemployment through decrease of the hysteresis effects and higher job turnover, decrease in mismatch between labour demand and labour supply through the personalised services to long term unemployed, decrease in social exclusion, etc.).

Costs and benefits presented below are based on an EU-wide average, which builds on the five case studies presented in the previous section. They are additional to the current costs and gains from ALMP provision: current ALMP expenditure on LTU can be estimated at 6.5 bn € per year (2012). The cost estimates are based on the assumption of an initial investment of 300 €r for all LTU targeted. Under these assumptions, 79% the one-off costs of establishing JIAs and their infrastructure are expected to be recovered in one year.

Average annual benefit savings per unemployed finding a job amount to 1805€, while tax receipts assuming low incomes amount to 1714€. While the savings on benefits are immediate with transitions to employment, the additional tax receipts should be interpreted with more caution, given the possible substitution effects of labour market interventions.

The estimate relies on the anticipated additional transitions to employment presented in section 7.1. Such positive impacts are subject to caveats on the likelihood of MS with weaker support structures achieve gains in transitions to employment comparable with countries which benefit both from stronger support structure and better functioning of labour market and positive investment patterns.

Table 15: Estimated budgetary impact of the initiative for the first years following the implementation of the initiative (million Euro)

No additional impacts are envisaged for the EU budget. Implementation could be supported by the ESF under the 'Access to Employment' and the 'Active Inclusion' priorities. Proactive programme management through existing mechanisms foreseen by the Common Provisions and the ESF Regulations (e.g. reallocating funds within a priority axis, aligning programming and project selection criteria) could support the efforts of MS. Additionally, the ERDF may also complement the measures financed by the ESF in accordance with the allocations for the relevant investment priorities for 2014-2020 (not subject to reallocation), namely by supporting job creation, the modernisation of employment services and vocational education.

Further to that, any MS wishing to provide additional funding to implement a new CSR on LTU could address a re-programming request to the Commission. Conversely, under measures linking the ESIF funds to sound economic governance, the Commission may make such a request to a MS where it is deemed necessary to support the implementation of relevant CSRs.

7.6.Timely impact and implementation risks

A timely impact is important for the budgetary impact and for the effectiveness of the initiative in tackling the current LTU stock. Shorter lead time to implementation enables quicker transitions to employment, which pay off much more in terms of savings on benefits and in tax receipts.

Experience of previous institutional integration measures in DE, UK, EE, AT show that setting up a one-stop-shop takes 1-2 years. The introduction of the German reformed support produced observable effects on transitions to employment within one year of its introduction. Setting up the lighter coordination mechanism (single point of contact) can be expected to take at most 1 year, depending on the MS' starting points. Further to the implementation lags, newly introduced JIAs would take time to produce effects on the possibilities of transitions from LTU to jobs. Such impacts can be quicker in expanding labour markets, with higher levels of job creation. The implementation of structural labour and product market reforms can thus support a quicker impact of the initiative.

In cases with regionalised provision structures (e.g. BE, ES, IT, PL, FI) an additional layer of political arbitrage is needed due to the vertical differentiation of the agencies involved, as they are placed under municipal, regional or national authority, with different budgetary authorities.

Main implementation risks refer to low levels of overall administrative capacity – mainly staff competence and resources for an individualised approach to jobseekers. CSRs on PES capacity were addressed to BE, BG, CZ, ES, FR, IT, PT, RO, SK during the last 3 years.

Success factors include the clarity and harmonisation of the goals across the different bodies, a clear division of responsibilities (clear procedural rules), careful design of financial and administrative incentives for cooperation as well as the adjustment of the performance indicators to avoid dysfunctional competition between the different bodies. The reforms should be supported by appropriate training and capacity building of the administrations involved, especially of the front line staff.

The main coordination mechanism is data sharing across organisations. Data protection legislation might limit or slow down the development of a system of coherent case management and cross-referrals across organisations. The barriers outlined above can lead to uneven implementation of the initiative, with some MS registering delays in implementation.

8.Evaluation and monitoring

A monitoring framework based on existing data collection would be rolled out together with the initiative. The Commission's proposal invites the Member States to adopt within EMCO a list of monitoring indicators and guidelines for national evaluations within 6 months of the adoption, allowing an early start of regular data collection.

Monitoring would be partly based on existing indicators in the Joint Assessment Framework of the European Employment Strategy and in the PES Benchlearning process. Additional data would be collected through the delivery organisations in line with current MS practice, on an annual basis. It would refer to the JIAs, and be delivered through the coordinating organisations (employment services or other responsible organisations identified by the Member States): number of agreements concluded, share of those linked to a service to an employer and average numbers of referrals in JIAs. Referral data is regularly collected where PES use IAPs. Information could be aggregated annually in the existing LMP database published by Eurostat/DG Employment. Data would be split by gender, age and level of qualification.

Table 16 below details the proposed monitoring indicators and their link to the operational objectives. The monitoring objectives are each related to the specific objectives mentioned in section 4.1.

Table 16: Monitoring framework

Member States are encouraged to carry out national evaluations of the implementation of the initiative. A common set of guiding documents would lead to national results to be reported in a harmonised way and would increase comparability of results. Such guidelines for the national evaluations of JIAs, to be adopted by EMCO, would include:

Comparing outcomes (e.g. transitions to employment) for JIA beneficiaries with a control group;

Implementation information on the design and functioning of the single point of contact; types of offers included in JIAs (by LMP category);

Information on costs of delivery and marginal costs of provision (per additional beneficiary).

The re-assessment of the LTU would provide information on characteristics of both JIA recipients and non-recipients. A gradual roll-out of JIAs and discontinuities in design provide a possibility to define a control group (e.g. LTU in regions/ offices where JIAs are first introduced; unemployed younger than 30 or for less than 18 months). Counterfactual methods of evaluation can thus be used at no additional cost. Further to that, MS could opt for rolling out JIAs through randomised assignments within the target group, using the PES registry.

In designing their evaluations, MS would be encouraged to monitor transitions to employment for longer periods, e.g. 2 years after JIA, e.g. by crossing the unemployment registry with fiscal data. This would allow an evaluation of both the impact and efficiency of JIAs.

Based on national monitoring efforts and evaluations, the Commission will conduct at EU level an evaluation within five years of the adoption. The evaluation will provide assessment of effectiveness, efficiency, relevance, coherence and EU added value of the initiative.

9.Stakeholder consultations and expertise used

The Commission services organised also targeted consultations of stakeholders. The European Network of Public Employment Services (PES Network) provided data and adopted a report on the consultation. The European Social Fund Committee issued an opinion following the discussions on 26/03. The Employment Committee (EMCO) adopted an opinion on 08/05, following three discussions on the topic, including a joint EMCO-Economic Policy Committee (EPC) meeting on 1/04. On 24/04, DG EMPL organised a dedicated consultation workshop with representatives of sectorial and cross-industry social partners. The Social Protection Committee was consulted on 6/05.

9.1.Stakeholder consultations

The stakeholders consulted acknowledged LTU as a significant challenge and welcomed a Commission initiative. The following issues were raised and have been taken into account in the preparation of the initiative:

-The ESF Committee stressed the importance of aligning the initiative with already existing policy instruments and resources. Synergies with the current use of ESF in addressing long-term unemployment have been highlighted in this report.

-The PES Network endorsed the principle of individualised service provision, of increased cooperation among service providers, and spelling out mutual obligations.

-The social partners advised that an option based on general policy guidance and one on targeted action are not mutually exclusive and should be combined.

-EMCO highlighted the heterogeneity of the group requiring an individualised approach to activation and the need to allow flexibility for MS to tailor their policies.

-The Social Protection Committee endorsed the general approach of the Commission, but pointed out to the risk of crowding out resources from support of vulnerable groups most remote from the labour market.

The outcome of the consultation pointed to a need to better balance a specific instrument with general policy guidance. The initiative was therefore re-oriented towards combining an instrument focused approach with general guidance, thus offering flexibility in implementation to the MS. JIAs address at individual level the heterogeneity of the group, while the embedded cooperation mechanisms aim to improve access to support for those with complex needs and more remote from the labour market.

9.1.1.Labour Market Observatory (LMO), European Economic and Social Committee (EESC)

The observatory was consulted in the occasion of a Conference on "Taking action on long-term unemployment: national and European perspectives" which took place on 11 March 2015 in Zagreb, Croatia. EESC Members welcomed the Commission's initiative and participants to the conference agreed to the challenges identified.

9.1.2.European Social Fund Committee

The European Social Fund Committee was consulted on 26th March 2015. Following the consultation, the Committee issued an opinion in which it welcomed the initiative of the Commission and also made the following comments:

It stressed the importance of aligning the initiative with already existing policy instruments and resources in order to build synergies of and ensure consistency, including with initiatives supporting job creation and self-employment.

It underlined that in many MS a substantial amount of the ESF is already dedicated to putting in place integrated support structures for long-term unemployed. The ESF therefore has an important role in supporting the ALMPs on offer in the MS and in increasing the effectiveness of interventions.

It considered that the Commission's initiative can add value to the way national and ESF funding is used in MS to integrate the long-term unemployed into the labour market.

It considered that the ESF can play a significant role in the implementation of the initiative by supporting MS interventions for addressing long-term unemployment both under the employment and the social inclusion thematic objectives. Implementation of the initiative should increase the effectiveness of ESF actions and embed them further in mainstream ALMPs and social interventions.

9.1.3.European Network of Public Employment Services

The Network of European Public Employment Services drafted a contribution on the proposed initiative 140 . It supported the principle of a Council recommendation as a positive initiative to improve LTU re-integration rates and stressed that the proposal for the Council recommendation should take into account the different starting points in the MS.

In its contribution the Network stressed the following points:

Single Point of Contact: Where there is no one-stop shop established for LTU, a single point of contact - or a less costly option of establishing digital/automated single contact points based on linking databases of benefits and employment authorities - are needed.

Coordinated action: An integrated approach is needed to overcome multiple barriers to re-integration including organisations from the public and private sectors, providing social, family, health, and employability support services.

Introducing Personalised Services: Development of a personalised, tailored, client centred approach will be an essential component of a re-integration package.

Spelling out Mutual Obligations: Mutual obligations should detail possible support available through counselling, financial assistance and social services, and describe the commitment required from the client in return.

9.1.4.Employment Committee

The ad-hoc working group of the Employment Committee was consulted on 30 March 2015. The initiative was discussed also at a joint EMCO-EPC meeting on 1 April 2015. The Committee welcomed the initiative of the Commission and recognised the relevance of the problems identified in the public consultation. The Committee stressed the following:

EU-level guidance might reinforce the MS' country-specific approach to supporting the long-term unemployed in line with their individual needs.

The initiative should be aligned with already existing policy instruments and resources in order to build synergies and ensure consistency. EMCO emphasised the need to link future initiatives to the Employment Strategy and the Employment Guidelines.

The situations differ considerably across MS, notably regarding the levels and the characteristics of the long-term unemployed.

Employer engagement and combining supply and demand-side measures are important elements.

An individualised approach to activation should be adopted. A mutual obligation approach is essential.

9.1.5.Social Partners

On 24th April 2015, DG EMPL organised a consultation with the social partners, including representatives of cross-industry organisations at European and national levels and sector federations.

The social partners agreed on the relevance of the problems identified and the necessity of an intervention to support LTU. The social partners suggested considering a combination of the policy options proposed, while taking into account the EU labour market needs, synergies with existing initiatives and instruments and the differences among MS. Well-functioning of public employment services and well-targeted active inclusion initiatives, including the provision of support to employers were found to be crucial for preventing and overcoming LTU.

In addition to the above:    

Trade Unions highlighted the importance to offer incentives to those employers who hire long-term unemployed and to equip LTU with labour market relevant skills and qualifications.

Employer representatives stressed the need for better cooperation between public and private actors and the importance of better using funds available for supporting LTU rather than having additional funding. They supported a mutual obligations approach

The social partners also submitted written comments to the consultation:

Employers (a joint position of BUSINESSEUROPE, CEEP and UEAPME) proposed in their contribution a two stage approach. As a first step they see the necessity to focus on the overall systemic factors and the roles of and interaction between different actors and services. An initiative at European level to support re-integration of LTU should be part of a systemic approach to job creation and higher levels of employment participation, by focusing on:

-more efficient, effective and integrated services to the unemployed reflecting available best practices;

-more cooperation and mutual learning between public employment services and private employment agencies, municipalities and social services;

-Appropriate engagement of employers to strengthen the demand side approach. This includes closely aligning re-training and up-skilling opportunities offered as part of activation support measures with enterprises’ needs.

As a second step, they stress the importance of a targeted individualised approach to achieve better matching of available workforce with available jobs. However, they do not consider an individual action offer that targets specific groups as the right approach.

Trade Unions (ETUC) stressed that initiatives aiming at the re-integration of LTU into the labour market should lead to quality and sustainable jobs and that are challenges connected with the financing of LTU support in the context of budgetary constraints of those MS most concerned by the problem, which could reduce the impact of any proposed initiative if not addressed appropriately

Trade Unions stressed that the options identified by the Commission should not be treated as being mutually exclusive. There is the need for better coordination among services and for comprehensive and mutually reinforcing strategies, taking account of specific circumstances in the MS. The development of ‘individual activation offer’ is considered as an important element.

Trade Unions proposed that the Commission investigate and develop the idea of a ‘European job guarantee for the long-term unemployed’, similar to the Youth guarantee, with European financial support. Finally, trade unions stressed that social partners have a role to play, particularly in developing a framework to help prevent workers from falling into LTU.

In addition, Eurocadres (Council of European professional and managerial staff), while not present at the consultation meeting, submitted written comments to the consultation. Eurocadres supported ETUC position and added that development of skills and provision of lifelong learning are among the key factors improving the chances of employment, emphasising also the role of voluntary mobility.

9.1.6.Social Protection Committee

The Social Protection Committee was consulted on 6 May 2015. The Committee stressed the following points:

The situation of long-term unemployment differs across MS therefore any future initiative should to be flexible to adapt to the national context.

The initiative should take into account that LTU with more complex barriers to the labour market require extra support in order for activation to be successful; it is important not to divert resources from this group.

The initiative should be coherent with previous initiatives.

9.2.Expertise used

Both internal and external expertise was used in the development of the initiative:

Mapping of inclusion contracts and co-ordination of income support with activation conducted by the Network of Independent Experts on Social Inclusion between February and May 2015

Analytical work was carried out using available surveys (Labour Force Survey, the EU statistics on income and living conditions, and the Labour Market Policy Database).

A collection of ESF good practice initiatives and a mapping of MS initiatives were conducted internally by DG EMPL.

A review of existing cost-benefit studies on LTU interventions and a detailed assessment for five MS were conducted as part of the European Network of PES bench-learning process 141 .

Reports of mutual learning events on addressing LTU: Thematic Review Seminar on “Tackling long-term unemployment - effective strategies and tools to address long-term unemployment”, 2012. Report of peer review of the public employment services' Network on good practices to address LTU, 2014.

The European Network of public employment services organised a working group between January and April 2015, provided data on the share of registered long-term unemployed, their participation in active measures and on existing public employment services practices for long-term unemployed and produced a report published in May 2015.

Annex 1
Abbreviations and glossary of technical terms used

A. List of abbreviations

B. Glossary

Annex 2 - Supporting data and analysis

Contents

1.Who are the long-term unemployed?

The LTU in the EU are on average more often of medium and low level of education (87%) and are slightly more commonly male than female (54%) (Figure 1). Almost a fifth (19%) has never been employed.

Figure 1. LTU characteristics before and after the crisis (2008 and 2013)

Source: Eurostat LFS 2008-2013, own calculations

The age distribution of the long-term unemployed is quite even, with slightly higher rates before 30 and after 55 (Table 1) and remained stable throughout the crisis.

Table 1. Distribution of LTU by age and duration, 2013

Older workers have more stable employment and are overall less affected by unemployment. However, when they become unemployed, they have higher risk to remain so. This leads to an increase with age in the share of LTU among the unemployed, by steps which are highly country specific (Figure 2). The share of LTU among unemployed aged 40 to 49 ranges from 27% in AT or FI to 72% in EL and 75% in SK.

Figure 2. Long-term unemployed as share of unemployed, by country 2013

Source: Eurostat, LFS

Men are more often in LTU than women, especially in some countries (MT, IE, FI, UK) and if compared to short-term unemployed (STU) (Figure 3 below).

Figure 3. Share of LTU by gender and share of men in short-term unemployment (STU) - 2014

Source: Eurostat, LFS

2.The regional dimension of long-term unemployment

The regional NUTS2 dispersion of LTU increases at higher rates of LTU. Targeting the initiative at regional level would not add considerably to its precision. For example, assuming that only regions with very high LTU rates (50 % higher than EU average of 7.5%) should be engaged would exclude entire countries as Romania, Germany, France, Hungary, Estonia or would cover only arbitrarily selected regions regarding other socio-economic performances. For example, under these conditions only a capital region for Belgium would qualify as a target group. The added value of targeting the initiative at regional level is thus limited. Moreover, the inter-institutional cooperation aspects cannot only be sustainably addressed only at regional level, as in most MS they depend on relations and arrangements between centralised public bodies.

Figure 4. LTU rate by NUTS2 regions and national average, 2014

Only a weak relationship exists however between the level of NUTS2 GDP and the LTU rate, and the degree to which higher LTU rates attribute to less developed regions varies considerably across countries. Within countries, highly varied relationships are observed between regional (i.e. NUTS2) GDP and LTU rates. Figure 5 displays that a clearer and as expected relationship – i.e. a higher GDP is associated with lower LTU levels – is only noted in a handful in countries, such as ES, IT or BG. In other countries, factors other than GDP are of higher importance. For example, a U shaped relationship between LTU and GDP is observed in BE. In the NL, very low dispersion in LTU rates is noted for highly varied GDP levels. The same holds for SE. In FR, distinctive groups by GDP and LTU rates occur. In a number of countries, as PT or RO, higher LTU rates also seem if anything associated with higher rather than lower GDP levels. In addition, capital regions stand out as the biggest outliers in most countries –often with the largest GDP levels and large if not the largest (i.e. BE) LTU rates.

Figure 5. Relationship between regional GDP and LTU rates - selected countries

Source: DG EMPL calculations based on EUROSTAT data

Note: here and further on, only countries with larger number of NUTS2 regions are included; for scale reasons, the GDP level of London is capped.

3.Transition rates from unemployment to employment and Inactivity

Long-term unemployed have about two times lower chances of finding employment than short-term unemployed, with the situation worsened due to the financial crisis (see Figure 6). If in 2006 almost 25% of long-term unemployed (i.e. those in 2005) were able to find jobs, only 16% of them had this chance by 2014 (see Panel A in Figure 6 below). These results pertain to 12 EU member states 142 , though shorter (available) series for 24 EU countries 143 reflect a comparable trend (see Panels B and D). In contrast to LTU, the last few years have already brought employment gains to STU – their transition rates to employment increased from 29% in the period of 2012-2013 to 33% in the period of 2013-2014 (see Panel C). LTU transition rates to employment do not yet reveal signs of improvement.

Some long-term unemployed are able to break the LTU spell by, for example, finding temporary jobs after which they return to unemployment (LTU to STU). This share of LTU has remained relatively stable – of about 6% to 8% of LTU – since 2005 (see Panel A). Persistent LTU status accounts for major part of transitions from LTU and has increased from 40% in the period of 2005-2006 to 49% in the period of 2013-2014, with a peak of 52% a year before. As Figure 7 shows, the slight drop in a persistent LTU status in the last few years is mainly attributable to once-again starting to increase rates into inactivity rather than jobs. LTU transition rates into inactivity were on a continuous decreasing trend from a peak time in 2008-2009 until 2012-2013. These combined trends lead to a continued increase of the share of LTU among unemployed in most MS 144 .

Figure 6. Outflows from long-term and from short-term unemployment

3.1.Transitions from long-term unemployment to inactivity

The longer the unemployment duration, the more frequent are also drop-outs to inactivity. Based on experimental transition rates from unemployment to inactivity, the average EU24 transition rate from unemployment to inactivity is about 20% among unemployed from 1 to 2 years and about 5 percentage points higher - for those in unemployment for more than 2 years. The chance of being inactive becomes larger the longer the unemployment duration is (see Figure 7). A similar pattern can be observed in the US.

 Figure 7. Transition rates from unemployment (2008-2012) to inactivity (2009-2013), by unemployment duration in the previous year, EU24 (without BE, LU, NL, PT)

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates.

As shown in Figure 13, variation in transition rates from unemployment to inactivity across countries is also large, from as low as 3% (Greece) to 42% (Italy) among unemployed from 1 to 2 years. In a few countries transition rates to inactivity are higher for those in unemployment from 1 to 2 years than for those unemployed for more than 2 years – e.g. Ireland and Germany (Figure 13). Also, if in comparison to transition rates from unemployment to employment, transitions to inactivity have much more diverse pattern across unemployment durations. In some countries little difference is noted by unemployment duration – e.g. Ro, BG, EL, pointing to other factors than unemployment duration as strong determinants of exits to inactivity.

Figure 8. Transition rates from unemployment to inactivity (2013-14) by duration of unemployment

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates. 145

Analysing the impact of age on transition rates to employment, this appears to be particularly stable for the segment 25- 55 years old. Beyond 55, transition rates decrease while labour market detachment becomes high. After 12 months of unemployment, changes in transitions to inactivity between 2009 and 2013 were lower for all age groups (Figure 9).

Figure 9. Transition rates from unemployment to employment and inactivity, by age groups EU24

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates.

3.2.Determinants of transition rates to employment

Figure 15 we show that in some countries transition rates to employment started to improve (see, group “increase”), following generally improving employment situation in these countries. There is substantial progress in transition rates of LTU in Denmark and Estonia – against otherwise rather “flat” growth of total employment rates. The worrying situation is however noted in the group of countries with recently decreasing transition rates of LTU, despite generally steady (i.e. Austria, Finland) or improving overall employment situation (i.e. Lithuania, Latvia). In a number of countries transition rates to employment are mainly unchanged and in line with similar development in total employment rates. In Germany, on the other hand, stable transition rates are observed whereas total employment rate are already on the rise. In sum, the developments observed in transition to employment rates of LTU and total employment rates show that numerous other factors than general economic situation influence job finding chances of this population group.

Figure 10. LTU (in t-1) to E (in t) rates and total employment rates (in t)

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates and EUROSTAT series on employment.

A full analysis of causal relationships between institutional, structural variables and the transitions to employment of the long-term unemployed is beyond the scope of this paper. The sections below show the correlation of variables associated with activation support with transitions to employment of the long-term unemployed, as well as those of structural variables related to labour market flexibility.

3.2.1.Strength of support to the long-term unemployed

Figure 12 shows a strong correlation between ALMP expenditure (inclusive on public employment services) and transitions to employment, pooling the data for 2005-2013. Countries with highest investment on ALMP services register highest transitions from LTU to employment. That said, one should note that dispersion is very high among countries with low ALMP expenditures.

Figure 11. Expenditure on ALMP and transitions from LTU to employment (2005-2013)

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates and LMP database.

Figure 12 below shows the correlation between transition rates of the long-term unemployed and a composite index of ALMP strength, detailed in the report on Economic and Social Developments in Europe 2014.The Figure shows that the lower the strength of ALMPs, the lower the transition rates to employment are. Higher levels of investments in ALMPs have on the other hand a strong positive correlation with transitions to employment for the long-term unemployed.

Figure 12. Investment in activation and transitions from LTU to employment

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates and ALMP index presented in ESDE 2014 .

The same applies to the levels of lifelong-learning participation of the unemployed, which is a close proxy for overall activation participation. Figure 13 below shows that low levels of participation of the LTU in life-long learning translate into lower transitions to employment.

Figure 13. Unemployed participation in lifelong learning and transitions from LTU to employment

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates.

Figure 14 indicates there is a strong and positive relationship between higher net replacement rates of the LTU (i.e. the rate measures income before and after the loss of job) and LTU finding jobs. It actually shows that countries which have high net replacement rates also can achieve very high exit to employment rates. The variation is large across countries, though none of countries with low replacement rates registers high transitions to employment for the long-term unemployed.

Figure 14. OECD Net replacement rate of LTU and transitions from LTU to employment (2005-2013)

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates.

The remainder of this section will analyse the impact of structural barriers (investment constraints, unreformed labour markets and low educational achievement and attainment) on transitions to employment.

3.2.2.Investment constraints

Reduced labour demand accounts for a large part of the rise in LTU. Unemployment rates are largely driven by the different growth trajectories in the MS. 146 Weak cyclical labour demand has depressed job finding rates, leading to lengthened unemployment duration. Policies aimed at enhancing macroeconomic demand – such as expansion of fiscal and monetary policy – can reduce this component. 147 However, this is not enough. Some of the increase of unemployment is already structural and needs to be addressed by strengthened support or structural reforms.

Figure 15 below illustrates the relation between cyclical and structural unemployment, including a forecast to 2016. While cyclical unemployment (mostly affecting the short term unemployed) will be re-absorbed by the end of 2016, the structural component (mostly affecting the long term unemployed) would persist.

Figure 15: Structural and cyclical components of unemployment in the EU-28, with forecast till 2016

Source: Eurostat and Ameco. Note: cyclical unemployment is the difference between total and structural unemployment, shown by up and down bars.

The bigger the delay in action and the longer unemployment spells last, the higher the chances that the increase in LTU will entrench the hard-to-remedy structural unemployment, with long lasting consequences. Figure 16 illustrates the strong correlation between increases in LTU rates and increases in structural unemployment between 2008 and 2013.

Figure 16: Link between long-term and structural unemployment developments (percentage point)

Sources: Eurostat, European Commission estimates and ESCB calculations, from: Anderton, R., Jarvis, V. et al. (2015), 'Comparisons and Contrasts of the Impact of the Crisis on Euro Area Labour Markets.' ECB Occasional Paper No. 159.

3.2.3.Unreformed labour markets

Weak labour market institutions contribute to a protracted labour market adjustment following shocks, and to a structurally high LTU.

Employment protection legislation creates incentives for workers and firms to invest in existing relations, but by making dismissals more costly, it can deter hiring in the first place, with potentially detrimental effects on LTU. 148 Empirically its impact on employment and transitions out of unemployment, especially during times of low labour demand is unclear. On the one hand, high and uniform levels of employment protection with threshold effects lead to insider-outsider dynamics on the labour market, making it more difficult to re-integrate the long-term unemployed. Higher EPL can also mean lower numbers of new hirings, especially the longer the average unemployment spell, and a higher incidence of LTU. A high EPL can lower employment expansion, and thus keeps LTU high in the aftermath of a crisis 149 . The effect of employment protection on the probability of being hired is particularly strong for those workers whose productivity is a priori uncertain, such as the long-term unemployed. 150

Figure 17. Employment protection legislation and transitions from LTU to employment (2005-2013)

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates

On the other hand, some countries with high EPL have both high employment rates and good transitions from LTU into employment (e.g. DE and NL). Moreover, during times of low labour demand EPL reform has not been found to have an impact on transitions in the short- to medium-term, 151 with some studies signalling that in such times reducing EPL may result in more dismissals rather than hiring. 152  

Figure 17 shows the correlation between transition rates of the long-term unemployed and OECD employment protection legislation index. Though the strength of relationship is weak, if anything higher EPRC is associated with lower transitions rates. Both observations (i.e. weak and/or negative relationship) are actually in line with existing literature findings.

EPL alone cannot explain labour market outcomes but in combination with other labour market institutions it can be utilised to combat LTU and a dual labour market. One such example are short-time working schemes accompanied by partial unemployment benefits that were extensively used during the early phase of the recession and were successful in maintaining employment and containing unemployment. 153  

A high tax wedge increases the cost of labour to firms, potentially increasing unemployment and LTU levels. Specific schemes, such as recruitment subsidies or tax credits can mitigate the negative effects of the tax wedge. Similarly, high minimum wages relative to average wages can make employing lower-productivity individuals, such as low-skilled workers, too costly, and increase inactivity and LTU. Examining the relationship between exits from long-term unemployment and the tax wedge 154 , a weak relationship appears, and the correlation seems to be reversed compared to theoretical expectations – the higher the tax wedge, the higher the LTU exits to employment.

A functioning social dialogue can help reducing unemployment in general, and also limit LTU. The LTU rate is traditionally smallest in countries with a very strong social dialogue tradition, with trust among the social partners. A culture of cooperation helps adjustment in the day-to-day business, while national agreements between social partners can also help the adjustment of wages and prices in response to major macroeconomic shocks, avoiding sudden fluctuations in employment. 

Figure 18. Tax wedge for low income earners and transitions from LTU to employment, 2005-2013

Source: Eurostat, EU-LFS, ad-hoc experimental calculations of transition rates and Tax & benefits indicators database

Barriers to geographical labour mobility cater for persistent unemployment differences across regions, 155 and hence to persistence of LTU. Policies that make mobility easier, including various policies of the European Union, help labour market adjustment and act against long-term labour market exclusion.

4.Current support for the long-term unemployed

4.1.Coverage: Income support for the long-term unemployed

Most MS provide income support to long-term unemployed. As shown in Figure 19 below, in 10 MS the unemployed may still receive contributory-based benefits after 12 months of unemployment. In general, the average duration of unemployment benefit is much below the maximum duration in most MS.

In all MS except IT, workers who have exhausted their unemployment benefits may be eligible for non-contributory social assistance benefits. Non-contributory benefit schemes distinct from the social assistance (such as unemployment assistance) exist in 10 Members States (AT, DE, EE, ES, FI, FR, IE, MT, PT, UK) 156 .

Figure 19: Maximum duration of unemployment benefit across MS 2007/2014

Source: MISSOC

Generally, as shown by Figure 20 below, the share of long-term unemployed covered by unemployment benefits has decreased between 2007 and 2013.

Figure 20: Unemployment benefit coverage of long-term unemployed

Source: Eurostat and EMPL LFS 157 .

4.2.Coverage: Active support for the long-term unemployed

Different approaches can be observed when it comes to defining the focus of activation support measures during the unemployment spell. Whereas some countries tend to prioritize people closer to finding work (MT, NL, UK), others choose to focus on people further from the labour market (FR, LT, SE).

Based on information available to the European Network of PES and to ESPN, only three MS make statistics on the share of LTU who receive activation support publicly available. In DE and FI, the most represented category is composed of long-term unemployed receiving only benefits and no activation support (89% of LTU in DE 158 and 78% of LTU in FI), while in LV those being granted activation support and no benefit prevail (73% of LTU).

The questionnaires of the European Network of PES identify 6 cases of PES internal performance targets linked to LTU activation:

Table 2: National targets on activation of LTU, 2015

Source: European Network of Public Employment Services

Overall, PES in 19 MS neither target explicitly the activation of the long-term unemployed nor publish the share of those receiving activation support.

Only 12 MS have data on the share of the registered unemployed receiving social assistance benefits (Table 3); they range from a high share in IE (76%) to very low shares in EL (only 1%).

Table 3: Share of registered unemployed receiving social assistance benefits

Source: European Network of Public Employment Services

4.2.1.Intensified support

While several PES now use specialist youth counsellors, fewer PES have counsellors specifically working with the long term unemployed. PESES in 9 MS are using LTU counsellors or have staff directly servicing the LTU client group. 159 Table 4 below provides an overview of the PES that have dedicated support for the LTU, and the numbers of frontline staff dealing directly with the LTU 160 .

Table 4: PES staff directly servicing the long-term unemployed (2014)

*Includes all employees in the fields market and integration in the joint facilities as well as some 50 % of the employee capacity of the organisational area “market and integration and granting of benefits without functional separation"

** Percentage figure provided on the basis of total staff directly servicing LTU

Source: European Network of PESA

4.3.Activation design –labour market relevance of support

There is increasing evidence showing that employers actively sort out the long-term unemployed when reviewing job applications 161 . The relevance of interventions to the demand on the labour market and the direct involvement of employers in shaping the offers to the long-term unemployed are important factors determining their effectiveness.

Beyond adjusting the parameters of the incentives for employers (wage- and non-wage subsidies), some MS are introducing novel ways of ensuring employer buy-in. France uses clauses in public procurement contracts to stimulate employers to hire long-term unemployed. In Bulgaria, there are legal provisions which can require tenders for public contracts to include recruitment of long-term unemployed. In Denmark or the Netherlands similar requirements can be imposed at the municipal level, including as a part of infrastructure development initiatives.

Additional bespoke support to employers is being adopted in preference to more generalised approaches. In Belgium the PES (VDAB) has recognised this with an increasing emphasis on competency based matching to assist the re-integration of LTU, including company based individualised tailored training programmes. The private companies contracted to provide re-integration services for LTU in the UK also use personalised trajectories. In the Netherlands, employers are counselled for finding "hidden vacancies" and adjust the way they draft their vacancies; several social enterprises also use "job carving" techniques for indicating to employers the potential to hire persons from disadvantaged groups (e.g. low skilled or workers with disabilities).

Several PES develop post-placement support for the long-term unemployed who are re-employed, supporting enterprises with additional coaching and mentoring. France developed after-care programmes especially targeted at SMEs providing post recruitment support. This reflects an emphasis on capability and emphasising what an LTU client can do, rather than focussing on barriers. The Netherlands relies on self-regulation efforts on the part of employers, with the support of social partners: A tripartite agreement proposed a specific target to increase the recruitment of vulnerable jobseekers with the prospect of government quotas and sanctions if this is not achieved.

An agreement between the Employment Ministry, the PES and the employer federation in Luxembourg signed in 2015 foresees an increase by 5000 people of the number of recruitments from PES clients by 2018, particularly for youth, older workers and long-term unemployed. The PES committed to increasing the quality of recruitment services, to create new training schemes targeting specific employer needs and to an increase in the index of client satisfaction, while the employers' federation committed to an awareness campaign on the benefits of recruitment through the PES.

Estimates of the numbers of case handlers providing support to employers, or as staff directly servicing employers, are summarised in the table below.

Table 5: Employer case handlers and staff directly servicing employers 2014/2015

Note: IT, PT and SE do not use specialised employer counsellors, while in Slovakia the position of employer counsellor is currently being redesigned, so only an estimate of the number of such counsellors is available.

Source: European Network of Public Employment Services

5.Costs and benefits of transitions to employment

Estimates of budgetary gains from transitions to employment were re-calculated based on in EC (2014) – The Cost of Poverty . The estimates assume that if someone unemployed moves into employment, the household in which they live loses any social exclusion benefits and housing allowances that they were receiving. An indication is given in the last column of the tables of the relative importance of these benefits and in some cases they amount to over half of the overall reduction in benefits.

As emphasised in the analysis in section 7.5. of the Staff Working Document (SWD), the estimated gains are subject to a number of provisos, as they do not take into account any substitution effects between long-term and short-term unemployed. The different incidence of channels of job creation in the MS, which are a critical factor for achieving the budgetary gains calculated in the SWD, are not factored into the calculations either.

Table 6: Gains in tax receipts and benefit expenditure from transitions to employment

Source: EU-SILC, ESSPROS 2011

Among the estimates presented in Table 6 above, the more conservative scenario of return to full-time employment with 75% of average wage was used in the SWD analysis. The per-capita gains for the public budget of exits from unemployment to full time employment, with 75% of average wage amount on average to 3 520 Euro, composed of 1805 euro of benefit savings and 1714 euro gains in tax receipts. The choice of exits with 75% of average wages takes into account the well-established scarring effect of long periods of unemployment, which diminish overall lifetime earnings. To estimate budgetary gains, we have further multiplied the average budgetary gains with the estimated additional transition rates to employment presented on in Table 12 of the Staff Working Document. Additional transition rates were calculated for each country, and vary from below 3 percentage points in HU, CY, PT, CZ to 11 percentage points in EL, HR, with an average of 6 additional percentage points. Such additional transitions are fully in line with evidence from prior reforms or programme in the MS, which often achieved increases in transition rates above 10-15 percentage points. The additional transition rates were applied to the LTU group targeted to obtain an estimate of the additional stock which would transition to employment.

To calculate the costs for the introduction of the JIAs, we have taken the maximum estimates from the ICON 2015 study (300 euro/beneficiary) and applied it to the total population covered.

It is important to note the costs of additional service and ALMPs provision following JIAs are not included in the analysis Table 15 of the SWD. While the LMP database has reliable estimates of the costs of ALMPs which can be broken down by LTU/non LTU participants, there is no such corresponding data for employment services. However, it should be assumed that most increases in support are channelled through employment services. From existing data in 11 countries in the LMP database, the cost of services appears to be at approximately three times lower than the cost of ALMPs. However, lack of available estimate of the per-capita cost of employment service provision, as well as the difficulty to predict the mix of measures and services allocated to the LTU through the Job integration agreements lead to inclusion of only initial costs of setting up JIAs in Table 15 of the Staff Working Document.

The table below estimates on the basis of available data in the LMP database the per-capita expenditure for ALMP measures for long-term unemployed.

Table 7: Per capita expenditure on ALMPs (Cat 2-7) for adult long-term unemployed

Source: EU-SILC, ESSPROS 2011

In contrast to the richer information available on ALMPs, the available data makes it technically possible to estimate unit costs for LMP services and to compare these to unit costs of LMP measures. However, it should be clear that the data on services are not only incomplete but may also be subject to vagaries of reporting that result in a mismatch of expenditure and participants at intervention level, which makes them potentially unreliable. A detailed examination of the data intervention by intervention would be necessary to develop more reliable cost estimates.

Bearing in mind the above warning about data quality, the available data indicate that the unit cost of LMP services (2,408 Euro/p-y) is less than half that of LMP measures (5,758 Euro/p-y). Services appear to be significantly cheaper than measures in most of the 11 countries for which a comparison is possible. In many ways this is to be expected because in general service costs do not include any form of income support for participants who (since the unemployment spell is not broken) would normally continue to receive their existing unemployment or other benefit. In contrast, participants in LMP measures will often receive some other form of income support (e.g. a training allowance) that is included in the costs. Employment incentive measures are generally an exception here since the participants normally receive a wage from the employer.

(1) EUROSTAT LFS survey data [une_rt_a]

(2) COM(2014) 910 final

(3)

  http://ec.europa.eu/social/main.jsp?catId=333&langId=en&consultId=15&visib=0&furtherConsult=yes

(4) http://ec.europa.eu/social/BlobServlet?docId=14186&langId=en  

(5)  Long-term unemployment refers to unemployment (ILO/Eurostat definition) lasting for more than one year. Here and further on, estimates on number of long-term unemployed are based on EUROSTAT lfsa_ugad database. Unless otherwise notified, current reference period is 2014, reference population is aged 15 to 74 years, EU refers to EU28.

(6) Based on EUROSTAT lfsa_upgan database

(7) Here and further on, long-term unemployment if for 2 or more years is referred to as very long-term unemployment or VLTU.

(8) EC (2014a) - Economic and Social Developments in Europe   

(9) Agnello et al (2014) Krueger et al (2014)  ; . Current national trends in LTU support this delayed responsiveness: while the overall level remained stagnant, it decreased in 15 MS, remained constant in 3 and increased in 10 MS between Q4 2012 - Q4 2014 (LFS).

(10) EC (2014f) - ECFIN’s medium term projections: The risk of secular stagnation EC (2015) - European Economic Forecast Spring 2015: the Euro-area's NAWRU (Non Accelerating Wage Rate of Unemployment, which estimates structural unemployment) is expected to rise from 9.8% in 2014 to 9.9% in 2016, and the EU-28's NAWRU is expected to decline only slightly, from 9.0% to 8.8%. There is a high correlation between structural and long-term unemployment (Annex 2, Fig 16). The period of strong economic growth between the mid-1990s to 2007 led to a modest and unsustainable decrease in structural unemployment (from 9.5% to 8.8%), which is unlikely to re-occur in the medium term  Annex 2, Fig 15 shows the forecast trend in cyclical and structural unemployment.

(11) M. Draghi (2015)– Structural reforms, inflation and monetary policy M. Draghi (2015)– Structural reforms, inflation and monetary policy   – Introductory speech at ECB Forum on Central Banking Sintra, 22 May 2015

(12) EC(2014), The 2015 Ageing Report: Underlying Assumptions and Projection Methodologies. European Economy, no. 8 Absent additional policy interventions, unemployment is expected to decline very gradually. The no-policy change scenario in the Ageing Report provides unemployment estimations of 10% for the Euro-Area and 9.2% in the EU-28 in 2020, with unemployment at 22.1% in Greece and 19.5% in Spain.

(13) Krueger et al (2014) Annex 2 Fig 7, 8. This is consistent with findings from the US: find that LTU exits to inactivity diminish during downturns, but increase during recoveries. Different LTU characteristics (e.g. higher motivation, higher job-readiness) or to reduced possibilities of exit towards retirement or pre-retirement schemes can account for this.

(14) The need to address structural barriers on the labour market was emphasised in ETUC's and employer contributions.

(15)

Different registration rates can be attributed to the level, duration and eligibility of unemployment benefits, to the share of those who never worked among the LTU and to the obligations and sanctioning mechanisms linked to benefits, as well as to the quality and attractiveness of PES services.

(16) Bachmann and Baumgarten (2013) Flieges et al (2015) Martin (2014)  Annex 2 Fig. 13, 14. Registered jobseekers have higher job search intensity than non-registered ones. A meta-analysis shows that participation in ALMPs increases by 9% to employment and by 7% the probability of being employed approximately one year post participation. See also .

(17) When adding youth, LTU make up a fifth of all participants in ALMPs. Calculations based on the LMP database. For adult LTU, data is only available for 19 countries (excluding BG, CZ, DK, EL, IT, CY, LU, MT, PL, UK); for LTU participants data is not available for UK, PL, CY, EL.

(18) EC (2014a) - Economic and Social Developments in Europe Annex 2 Fig. 19,20.   finds a positive relation of receipt of unemployment benefits and transitions to employment.

(19)   This corresponds to 20.8% of expenditure on ALMP measures for all LTU and to 16.2% of expenditure on adult LTU (LMP database). The corresponding share of LTU in unemployment was 44.5% in 2012.

(20)  EU average, excluding EL, CY and UK. Although ALMP measures support a wider group than just the unemployed (for inactive persons who want to work), the registered unemployed are the primary target group.

(21) Annex 2, Table 2

(22) Two thirds of organisations to the public consultation, including EMCO and employers, find the mutual responsibilities approach relevant and efficient, while most individual responses challenge its efficiency and one third its relevance.

(23) Information on graduated sanctions vs de-registration for refusal pertains to 2011 and was built on D. Vern (2012) - Eligibility Criteria For Unemployment Benefits: Quantitative Indicators for OECD And EU Countries, OECD Social, Employment And Migration Working Papers No. 131

(24) Table 3, Annex 2.

(25) Eichhorst and Konle-Seidl (2008 Behncke et al (2010) Cockx and Dejemeppe (2012)  ). Explicit discussions with caseworkers on the unemployed persons’ duties are found to increase employment rates . In Belgium, simply notification of higher requirements for job-search activities and more resolute monitoring lead to 22-28% higher transitions to employment among those notified. () Besides the threat effect, receiving sanctions also increases the re-employment probability.

(26) EC(2015b) – Unemployment benefits with a focus on making work pay Pareliussen (2014) OECD (2011)  ;  

(27) report 88% of respondents to the public consultation identified this as relevant/very relevant. In many countries, a multitude of service providers intervenes, from private employment services, NGOs, municipalities, social enterprises, training provides or the social partners, under the responsibility of the public authorities in charge. The PES Network refers to the "conductor" role of the PES for the efforts of the diverse service providers.

(28) Clegg (2013) COPE LOCALISE Policy Brief no 2 EC (2014e) - PES approaches for sustainable integration of long-term unemployed - Toolkit  , FP7 project ; .Mutual learning in the PES Network identified the coordination and balancing of income support with activation as a major challenge when the two are delivered by different institutions.

(29) Minas (2014) INSPIRES  , FP7 project

(30) COPE (2015) European Policy Brief COPE  See also ,, FP7 project 

(31) Dean (2013) McQuaid and Fuertes (2014) Rees et al (2014)  , ,

(32) Minas R. (2014) INSPIRES   , FP7 project

(33) Champion and Bonoli (2014) LOCALISE   article of FP7 project

(34) Kluve 2010 Card et at 2010  ; .

(35) http://www.cedefop.europa.eu/en/events-and-projects/projects/forecasting-skill-demand-and-supply/skills-forecasts-detailed-data  

(36) European Vacancy and Recruitment Report 2014   

(37) In 2013, unemployment rates among high qualified was 6.4%, while unemployment rate for the low qualified workers was 19.2% (Eurostat – LFS)

(38)  Schleicher (2014) – Developing Educational Policies in a Holistic Skills Framework in S.K. Lee et al. (eds.), Educational Policy Innovations: Levelling Up and Sustaining Educational Achievement, Springer

(39)  In 2014, 4.4% of low qualified workers participated in lifelong learning, compared to 18.8% of the highly educated (Eurostat- LFS)

(40) The reasons why third-country nationals are more affected by long-term unemployment vary across Member States and individual circumstances. Recent in-depth analysis points to issues related to access to general information on the functioning of the labour market;overall level of education and skills as well as mastering the host-country language; recognition of qualifications acquired abroad and lack of evaluation of previous qualifications and skills; access to vocational training including on-the-job training; limited contact with professional networks and potential employers. CEDEFOP (2014) Valuing diversity: guidance for labour market integration of migrants ,; OECD/European Union (2014), Matching Economic Migration with Labour Market Needs

(41) Based on EUROSTAT EU-LFS DG EMPL ad hoc calculations - data for 2013. 

(42) IAPs are documents to which jobseekers and employment service providers subscribe, stating rights and obligations, goals and steps to achieve the goals; the extent to which they include detailed information differs across MS. All organisations and 88% of individuals responding to the public consultation identified individualisation as relevant/very relevant. Most organisations find that provision is already individualised, most individual replies find that this is not the case.

(43) Berthet and Bourgeois, (2014 LOCALISE Whitworth and Carter, 2014 O’Connell et al. (2012)  (, article of FP7 project , ). Services are more efficient when individualised support is based on profiling (.

(44) Loxa and Morgandi (2014)  .

(45) EC (2015) - Identification of latest trends and current developments in methods to profile jobseekers in European Public Employment Services  .

(46) report of the PES N EC(2015) op. cit. This view is confirmed in the etwork.

(47) Activation And Integration: Working With Individual Action Plans: Toolkit For Public Employment Services Bimrose (2012) 26 MS use individual action plans. The two exceptions are NL, where they were restricted to persons with disabilities while some elements were kept as part of self-service (e-service provision) and the Luxembourg PES which is currently piloting them. EC (2012a) ; .

(48) Social integration contracts state the obligations of beneficiaries of unemployment/social assistance and their rights, particularly to social services and benefits.

(49) Martins and Pessoa (2014) Dengler et al. (2013) Georges et al (2008)   Evaluations in DE, PT and NL show higher transitions to employment with the requirement to attend more frequently PES and obligatory ALMP participation (where refusal was sanctioned) , , . In FR, LTU intensified support increased both transitions to employment and client satisfaction.

(50) EC (2014b) - PES approaches for sustainable integration of long-term unemployed Individual follow-up / case management designates a one-to-one stable relationship between an employment counsellor and a jobseeker, while IAPs are formalised agreements with the institution and the jobseeker. Individual follow-up including referral to additional support services was found to be effective in increasing transitions to employment in evaluations in DE, FR, DK, EE. ; Annex 2 Table 4.

(51) European Employers' Perspectives on long-term unemployment, recruitment and public employment services IPPR (2015) – . A total of 2552 employers were surveyed in UK, DE, ES, FR, SE in late 2014.

(52) EC (2012b) - Effective services for employers - Forging partnerships for the future PES Network EASPD Eurocities   Employment services include pre- and post- placement support and counselling provided by employment services. There is wide support for a stronger employer dimension in the actions addressing LTU among the contributions received to the stakeholder and the public consultation. The  , ,  emphasize the need to enhance service provision and to create win-win support structures for both LTU and employers.

(53) Annex 2 Table 5. Few MS (SE, DE, DK) are currently piloting dedicated employer counsellors for LTU integration.

(54) EC (2014c) - Stimulating job demand: the design of effective hiring subsidies in Europe Crepon B. et al (2013)  The scale and scope of hiring subsidies programmes, the type (subsidies of wage or non-wage costs, subsidising social security contributions, and refund mechanisms), their duration and conditionality strongly influence their attractiveness.; .

(55) Correspondingly, 50.6% of ALMP expenditure targeting LTU is allocated to employment incentive measures and 32.9% to public works/ direct job creation measures.

(56) EC 2014c Sjögren and Vikström (2015) Low spending per participant in employment incentive schemes reduces significantly their effectiveness. , . In the public consultation, organisations were aware of the existing support to employers. Opinions were split on its effectiveness, with many organisations considering it both sufficient and effective.

(57) Public works or public job creation designate programmes in which state agencies organise work, which is additional to open market demand for the unemployed, in return to payment. In terms of per capita spending per year, spending on public work LTU participants is on average 9.605 Euro/year, while spending on private employer incentives is on average 3.288 Euro/year.

(58) Card et at 2010 Kluve 2010  Meta-evaluations of private employer support versus public job creation in , . 

(59) EC (2013a) – Public Works: How can Public Employment Services contribute to increasing their value as an activation tool?  

(60) Arpaia A. et al (2014); Orlandi (2012)    finds that sector mismatches have a strong bearing on structural unemployment.

(61) Annex 2 Fig.13.

(62) Relative to the other unemployed, adult LTU (over 25 years old) have particularly limited access to training: only 13.3% of ALMP support takes the form of training for these participants, corresponding to 18.2% of expenditure. The contribution of Eurocadres emphasises increasing the efficiency of training and life-long learning.

(63) 59% of public consultation respondents consider that access to training in not sufficient and that training is not efficient. Unlike most individuals, most organisations replying find training provision efficient.

(64) EC (2013) - The Survey of Adult Skills (PIAAC): Implications for education and training policies in Europe  .

(65) EU High Level Group of Experts on Literacy (2012) – Final Report Eurostat LFS 2014; .

(66) EC (2015) – Upskilling unemployed adults  . Regular referrals to basic skills (second chance) education were only identified for young jobseekers: In DK, starting with 2015 all unemployed below 30 who are early school leavers are referred first to educational programme to complete their qualifications. In FR the planned "New Chance" professionalization contract will target improving basic skills for young LTU remote from the labour market.

(67) Caliendo and Künn (2010) EC (2014d) - Activating jobseekers through entrepreneurship: Start-up incentives in Europe  ; .

(68)  Most respondents to the public consultation (both individuals and organisations) agree that there is insufficient start-up support for the long-term unemployed.

(69) http://www.ecb.europa.eu/press/key/date/2014/html/sp140822.en.html  .

(70) COM(2015) 98 final  .

(71) Council Recommendations on the National Reform Programmes, OJ 215/C 272 of 18.08.2015  .

(72) Commission proposals for CSRs COM(2015) 85 final COM(2015) 85 final The in the 2015 European Semester addressed explicitly LTU in 4 countries (FI, RO, SI, SK). In 2014, CSRs for nine countries explicitly mentioned LTU (BG, DE, ES, FI, IE, HR, LT, PT, SK). Under the 2015 Semester, the Commission identified limited progress in addressing CSRs on active labour market policies ( – Annex 2) and pointed to the "need to strengthen the capacity of their Public Employment Services to fight LTU" ().

(73) The 6th Report on Economic, Social and Territorial Cohesion   EC (2014) , p 219.

(74)  The share of LTU participants in ALMPs was 20% in 2012. The participation rates in Table 6 translate into an estimated overall financial allocation to LTU of approximately 2.7 bn €. The estimated financial allocation in 2007-2013 is obtained by multiplying ESF spending on the priorities that could include LTU (31, 9 bn€) by the share of the LTU out of the total number of participations (8.8%).

(75)  The share of LTU participating in ESF interventions targeting the unemployed ranged from 11% in LU to 86% in HR and 100% in SE.

(76)  Minimal estimate on the basis of the MS having chosen to set targets for the long-term unemployed. The share will most likely be higher, as it is not an obligation to programme for all common indicators.

(77) EC (2015) - ESF 2007-2013 Ex-post evaluation: Supporting the integration of disadvantaged groups into the labour market and society, forthcoming.

(78) Ex-ante conditionalities aim to ensure that institutional and strategic policy arrangements are in place at the level of the Member States for effectively achieving the specific objectives of investment priorities for ESIF funds.

(79) Annex XI – Part I of the Common Provisions Regulation.

(80) thematic event  A on tackling long-term unemployment was organised in November 2012 and other events focused on related themes, e.g. job creation incentives and segmentation on the labour market. Peer reviews touched on e.g. integration support for the unemployed, the activation of the elderly, pathways to work for lone parents, integration in the labour market of the Roma community, or public works in activation programmes. 

(81) Social Investment Package COM(2013) 83 final C(2008) 5737 SWD(2013) 39 final The (2013) proposed to combine income support, tailor-made activation and access to enabling services to enhance support for those furthest from the labour market (). The Package builds on the evaluation of progress in the implementation of the 2008 Commission recommendation on active inclusion (), which observed piecemeal implementation of integrated and comprehensive active inclusion strategies, due to the lack of coordination between delivery agents, scattered policy responsibilities and variability in funding streams ().

(82)  For example, the Spanish Tripartite Agreement on an Extraordinary Program for Activation and Employment provides an additional subsidy of 426 Euro for 6 months to 500.000 LTU beneficiaries. In exchange, beneficiaries will participate in additional counselling and ALMP measures, including placements with temporary work agencies. The programme is temporary and will run from 15 January 2015 to 15 April 2016. 

(83) EC (2015) – Assessment report on PES Capacity  .

(84) Scharle (2015) Heidenreich and Aurich-Beerheide (2014) LOCALISE  . With only incremental steps towards cooperation and integration of services, NGOs and ESF projects will continue to act as brokers across the organisations in charge, bridging separate streams of provision. These are the predominant coordination mechanism in countries such as IT and PL identified in , article of FP7 project .

(85) Ball (2014) EC (2014) - ECFIN’s medium term projections: the risk of ‘secular stagnation’  , cf. .

(86) EC (2014) – The Role of PES in the Delivery of the Youth Guarantee PES Network Self-assessment Report on PES capacities to implement the Youth Guarantee  ; .

(87) EC(2015) – Report on PES Implementation of the Youth Guarantee  .

(88) Eichhorst et al (2006) Riley et al(2011) Minas (2014) Koch et al (2009) Introduction of the one-stop-shop, coupled with benefit reform in Germany lead to rise in open unemployment and a decrease of hidden (non-registered unemployment) . A similar effect in registration was observed shortly after the introduction of the Jobcentre Plus in the UK (), and identified as a general trend in reforms. FP7 Project INSPIRES; .

(89) Riley et al (2011). Klinger and Rothe (2012) Fahr and Sunde (2009)  Evaluation of introduction of Jobcentre Plus in Effects of 'Harz reforms' on speed of job finding in and on occupational matches in .

(90) ICON 2015 Fertig (2014) Hainmueller et al. 2009  Benefit savings amounted to around 115 million €. Together with savings from reduced ALMP-utilization and revenues from taxes and social security contributions gains of about 182.5 million € were achieved. Thus, net gains amounted to almost 150 million € in almost two years. This is equivalent to a net payoff per head of around 1,970 €. ,  A previous pilot project in 14 local employment offices in Germany has shown that a better workload (1:70 unemployed per caseworker) reduced the average duration of the UE spell by 10 days ().

(91) Scharle (2015)  The Estonian reform of 2009, which merged the Labour Market Board with the Unemployed Insurance Fund, lead to better enforcement of conditionality and to more effective activation of jobseekers. .

(92) Randomised control group evaluation of ACE (Activation and Control leads to Effect) pilot, UWV Werkbedrijf, NL.

(93) Scharle (2015)  .

(94)  A UK support "trailblazer" scheme for the very long term unemployed reported lower levels of assistance payments after six months of starting the programme compared to a control group of clients not included in the scheme.

(95) In the UK, the District Provision Tool (DPT) provides access to the full range of provision and support for up-skilling for the Jobcentre Plus clients, also facilitating post-recruitment workplace training.

(96) Pathways to Work In IE, alongside the Intreo single points of contact, the 2012 governmental strategy established new programmes for re-skilling targeted at LTU – the Momentum and Springboard programmes, which are tailored to labour market needs as registered through data gathered in the single point of contact and through provision of employer services.

(97) McGuinness et al (2014) Carling and Richardson (2004)   find that training focused on job search techniques and medium to high level skills courses in IE leads to more sustainable re-employment. find that employer subsidies and employer-lead training outperform classroom training in effectiveness.

(98)  Increasing the frequency of client contacts, as well as an earlier start of mandatory full-time programme participation reduced unemployment duration by 3 weeks in DK (Rosholm 2008). In the NL experiments focused on a broad client group including LTU, involving a population of 631,104 individuals reported a cost benefit ratio of 1:6 for early intervention advisor interviews with LTU, 1:10 for competence tests, and 1:12 for workshops for older jobseekers. Similar positive returns on programmes of intensive, twice weekly, interviews with LTU clients, mentoring schemes for young LTU clients were found in DK.

(99) IPPR (2015) - European employers' perspectives on long-term unemployment, recruitment and public employment services  

(100) Kluve 2010 Hohmeyer and Wolff (2012), Koch and Fertig (2012) In a review of 137 ALMP programmes in 19 European countries, finds the lowest effectiveness for public work creation programmes. In DE the so-called “1€-Jobs” have been a widely used ALMP-measure, with average stock of participants amounting to more than 300,000 individuals between 2006 and 2010. Several evaluations find a non-positive causal effect of participation in such jobs on employment prospects in the non-subsidised employment and on leaving benefits. This is corroborated by evidence on large lock-in negative effects in HU especially for the low-skilled.  .

(101) National action plan for addressing LTU  .

(102)  The New Start Jobs programme in SE showed significant effects, particularly when subsidies are longer (matching the unemployment duration) and cover higher share of wage costs. The duration of the subsidy was equal to the months spend in unemployment, but with a maximum of two years (and an effective minimum of 12 months). Similar positive results were observed in a BG scheme (Mihaylov (2011)).

(103) ICON 2015). A large scale employer subsidy scheme in DE tailored to an estimated productivity gap of each beneficiary but without a personalised counselling/ employer service element had negative results on transitions to work, estimated at 12000 Euro/participant ( Experience in DK with employer subsidies was more positive, and they have proven more effective than training programmes.

(104) Eppel and Mahringer (2013) The AT integration subsidy offers a wage subsidy covering up to 66.7% of the total wage cost and a lump sum of 50% for non-wage labour costs for a maximum of 2 years, targeting LTU and older workers. Seven years after entering the programme, non-subsidised employment for the beneficiaries was increased between 330 and 454 days (33.3% - 52%) depending on sex and age group. Even with substantial deadweight losses (around 60%) the authors estimate that the wage subsidy led to a net benefit of 1,800 € for women and 1,360 € for men over a five-year period for prime-age LTU. .

(105) Liljeberg and Lundin (2010)  The Swedish Job Net 2 programme, co-financed by the ESF organised a work trial to very long-term unemployed supported by continued coaching from the PES. The follow-up was very intense, with a caseload of 10-20 jobseekers per counsellor and a requirement to allocate at least 50% of the time for counselling towards employers. The intervention lead to shortening by 6% of the duration of unemployment on the open market or by 15% when including subsidised employment. It was later mainstreamed to all long-term unemployed. .

(106) Kluve 2010 Wolff and Stephan (2013)  . .

(107) The scenarios do not control for other factors or modelling the impacts. For all scenarios, only registered LTU are considered, lowering estimates, which can be considered a minimal threshold. The second scenario underestimates impacts, as the benchmark are transition rates in countries where arrangements similar to the JIAs already achieved a strong reduction in LTU and the remaining LTU are very remote from the labour market.

(108) Closing by half the gap to top performers is realistic. We assume an increase in transition rates of 2-3 % in HU, CY, CZ up to an increase of 11% in EL or SK. Increases witnessed in countries which reinforced their support system in recent years were in the range of 10% increase (Figure 12).

(109) No effects are assumed for DE, DK, NL, who already have provision similar to JIAs, while data is missing for BE, IE, LU, MT, UK, which are therefore not included in the scenario.

(110) Estimated by multiplying the increase in transition rates with the target population.

(111) EC (2014a) - Economic and Social Developments in Europe In Figure 12, MS with weakly structured support do not profile, do not offer IAPs to all, lack one-stop-shop provision or extensive data exchange; the converse applies to MS with strongly structured support. The correlation between the strength of the support and transition rates to employment is corroborated by the strong correlation between LTU transition rates and ALMP spending, unemployed participation in life-long learning and ALMP coverage (Annex 2 Figs 11-13). Other factors which might influence transitions to employment such as job creation rates, labour market rigidities or levels of unemployment are not controlled for. However, both sides of the figure include countries with high (UK, RO) and low GDP growth (FI, EL); high (DK, SE, ES) and low (DE, UK) ALMP expenditure; high (IE, SK) and low (PL, SE) LTU rates in 2013. See also

(112) EE reformed in 2009-2011. IE started introducing in 2012 the INTREO centres creating a one-stop-shop for employment services and benefits. CZ introduced in 2010 a similar reform, integrating employment and social assistance in a single organisation. They are the only countries which implemented large changes during the last 6 years and all succeeded in breaking the EU trend of decreasing transition rates.

(113) OECD Employment Outlook 2005  , p. 224

(114) ESDE 2013,

(115) bid

(116) SPC Committee (2015) - Social Europe - Aiming for inclusive growth  .

(117) Scharle (2015) Mploy (2011)  ; .

(118) Nichols et al (2013 Addabbo et al (2011) Paul and Moser (2009) Herbig et al (2013)  General social impacts of LTU are discussed in ), . Involuntary joblessness can be detrimental to mental health e.g. via feelings of helplessness brought on by a perceived lack of control or failure to obtain recognition, social connections and nonmonetary benefits of work. ; .

(119) SPC Committee (2015)) Martin (2014) Easterly et al (2006)   Unemployment has been constantly identified as one of the top two issues of public concern at both national and European level during the last years. (Standard Eurobarometer issues 2012-2014); On impacts on trust and social cohesion, see and .

(120) Kroft, K et al (2014) Krueger et al (2014) EC (2014) - ECFIN’s medium term projections: the risk of ‘secular stagnation’ EC (2014) - Economic and Social Developments in Europe  ; ; ; .

(121) PARES dialogues EC(2014f) - Delivering a coordinated service offer. Fuertes et al (2014) Zimmermann et al (2014)   between private and public employment services reached the consensus view that individualised service based on profiling and diagnostic assessments, data pooling and a coordinated offer are key enablers for the private actors entry of on the market for integration services, particularly for LTU;  See also FP7 Project LOCALISE and  FP7 Project LOCALISE.

(122) Van Berg et al (2012) Minas (2014) INSPIRES  ;  Report of FP7 project .

(123)

  http://ec.europa.eu/europe2020/pdf/themes/2015/employment_incentives.pdf

(124) IPPR (2015) - European employers' perspectives on long-term unemployment, recruitment and public employment services  .

(125) HU is currently introducing client profiling, segmentation and IAPs. The costs for the introduction of the reinforced support are estimated at a one-off investment 9.4 million Euro, which are already financed under the ESF. The entire IT infrastructure of the PES is currently restructured in RO, including e-services for setting up IAPs and linking the PES registry to social benefits, the tax and population registry at a cost of 5 million €, financed by the ESF.

(126) ICON (2015) Estimates in this section are based on study; estimates on costs of ALMPs for LTU are based on own calculations in the LMP database, shown in Table 7 of Annex 2 (2012 data).

(127) ICON (2015)  .

(128) ICON 2015 In DE, 3% of LTU experienced a benefit cut of 20-25% per year leading to benefit savings of around 200 million € per year (20% of the costs of running a system of individualised integrated support) – .

(129) ICON (2015). Table 13 extrapolates experience from countries applying interventions foreseen and applies it to the 5 MS situations, on the basis of a detailed baseline case studies presented in The five countries are chosen so as to present representatively the situation across the EU 28: MS with strong (DE, DK) and weak (HU, IT) employment services, with high (DK, IT) and low (DE, LT) spending on ALMPs, with municipal (DK), regional (IT), centralised (LT, HU) or mixed (DE) level of decentralisation.

(130) This view is supported in the employer's contribution.

(131) The estimate assumes maximal additional costs for initial investment and service provision (300 € additional investment/ beneficiary – amounting to a one-off doubling of administrative costs of current provision for all unemployed with no counter-financing effects.

(132) Back-of-the envelope estimate by dividing the average yearly allocation for ESF priorities 8i, 8vii and 9i in each country to the stock of registered LTU over 18 months and older than 30 in each country and to the higher cost limit of 300€.

(133) Relative to the Youth Guarantee, the burden for MS would be much reduced. Three main cost elements would be avoided: no monitoring outreach to non-registered beneficiaries is foreseen; no quantification of the type of offers received; no requirement of distinguishing transitions to employment before/ during/ after the offer provision.

(134) EC (2014) – The Cost of Poverty The data and methodology for the estimate are described in . The benefits for public budgets are estimated under the assumption that LTU overlap with the unemployed at risk of poverty; estimates used EU-SILC 2010 results for unemployed 20-59 y.o.

(135) Annex 2 Table 7.

(136) Kluve (2014) – A Review of ALMP effectiveness; presentation at European Commission seminar on income support – April 2014 Crepon B. et al (2013) Behaghel et al. (2013)  ., find large substitution effects for a programme focused on employment incentives for hiring youth ; find such substitution effects for outsourced placement support and lower substitution effects when such support is provided through the PES.

(137) Greenberg et al (2010)   provide a review of substitution effects in cost-benefit analysis or employment programmes.

(138) Brown et al (2011)  .

(139) http://ec.europa.eu/social/BlobServlet?docId=13881&langId=en   

(140) ICON (2015)  .

(141) Selection due to longitudinal EU-LFS data availability: BE, CY, EE, GR, HU, IT, MT, PL, RO, SE, SI, SK.

(142) Selection due to longitudinal EU-LFS data availability: EU24 (without BE, LU, NL, PT).

(143)  Only in three MS (IE, HU, SK) there is a clear downward trend in the share of long-term unemployed.

(144)

Note: Transition to inactivity rate is calculated as the share of unemployed people in year "n-1" - for a specific duration spell - who become inactive in year "n"; "I" refers to inactivity status; countries are ranked by the transition rate for those unemployed from 1 to 2 years; data of limited reliability for SI and AT for rates of "1<=U<2 years", and for "U>=2 years" in SK; country estimates are reported on countries with available data.

(145) Van Ours (2015)  .

(146) Unemployment in the euro area  As also explained by Mario Draghi, in his address "" at Jackson Hole on 22 August 2014.

(147) Young, D. (2003)  .

(148) Berger and Danninger (2006)  .

(149) Blanchard, et al (2014).  

(150) European Commission (2014), Employment and Social Developments in Europe 2014  : The Legacy of the Crisis: Resilience and Challenges, pp. 75-93

(151) OECD (2013) ILO (2014)   and .

(152) European Commission (2014) , pp. 87-89

(153) tax wedge Tax & benefits indicators database The is calculated for low wage earners (as % of the total labour cost) on low wage earners (67 % of the average wage for 2013; Source: Joint European Commission-OECD project,  

(154) Boeri et al (2000  )

(155)  MISSOC/European Commission

(156)

Note: IE, HR and NL: not covered. No data for UK in 2010. STU stands for short-term unemployed (less than 12 months) and LTU stands for long-term unemployed (unemployed 12 months or more). EU-25 =EU-28 minus NL, IE, UK (and AT in 2013). * The coverage rate is the ratio of the unemployed who received unemployment benefits or assistance and those who did not receive them in each category of unemployment duration (STU and LTU).

(157) The German labour market statistics provide only data on the benefit receipt of the 88% of all long-term unemployed, who are in the legal competence of the Social Code Book II with 100% receiving basic income support for job-seekers. No data are available on the benefit receipt of the 12% of LTU in the legal competence of Social Code Book III.

(158)

In Ireland, case officers are not employed by the PES. They are employed to deliver activation and case management services through Intreo centres and through the network of Departmental offices which are currently transitioning to a full Intreo service. Numbers of case workers working specifically with the LTU is not known.

(159)

The percentage calculation is based on staff servicing LTU, divided by total staff, multiplied by 100. The percentage figure for Germany is based on the number of counsellors, as a figure for staff directly servicing the LTU was not provided.

(160) Kroft et al (2014) Eriksson and Rooth 2011 Ghayad 2013  ; ;