EUR-Lex Access to European Union law

Back to EUR-Lex homepage

This document is an excerpt from the EUR-Lex website

Document 52018DC0761

DRAFT JOINT EMPLOYMENT REPORT FROM THE COMMISSION AND THE COUNCIL accompanying the Communication from the Commission on the Annual Growth Survey 2019

COM/2018/761 final

Brussels, 21.11.2018

COM(2018) 761 final

DRAFT JOINT EMPLOYMENT REPORT
FROM THE COMMISSION AND THE COUNCIL

accompanying the Communication from the Commission













on the Annual Growth Survey 2019


TABLE OF CONTENTS

FOREWORD    

KEY MESSAGES    

1.    OVERVIEW OF LABOUR MARKET AND SOCIAL TRENDS AND CHALLENGES IN THE EUROPEAN UNION    

1.1    Labour market trends    

1.2    Social trends    

2.    SNAPSHOTS FROM THE SOCIAL SCOREBOARD    

2.1    The scoreboard explained    

2.2    Evidence from the social scoreboard    

3.    EMPLOYMENT AND SOCIAL REFORMS – MEMBER STATES PERFORMANCE AND ACTION    

3.1    Guideline 5: Boosting demand for labour    

3.1.1    Key indicators    

3.1.2    Policy response    

3.2.    Guideline 6: Enhancing labour supply and improving access to employment, skills and competences    

3.2.1    Key indicators    

3.2.2    Policy response    

3.3.    Guideline 7: Enhancing the functioning of labour markets and the effectiveness of social dialogue    

3.3.1    Key indicators    

3.3.2    Policy response    116

3.4.    Guideline 8: Promoting equal opportunities for all, fostering social inclusion and combatting poverty    130

3.4.1    Key indicators    130

3.4.2    Policy response    151

ANNEXES    158



FOREWORD

The Joint Employment Report (JER) by the European Commission and the Council is mandated by Article 148 of the Treaty on the Functioning of the European Union (TFEU). The initial proposal for this report by the European Commission is part of the Autumn package, which includes the Annual Growth Survey launching the European Semester cycle. The Joint Employment Report provides an annual overview of key employment and social developments in Europe as well as Member States' reform actions, in line with the Guidelines for the Employment Policies of the Member States 1 . The reporting on these reforms follows the structure of the Guidelines: boosting demand for labour (Guideline 5), enhancing labour supply and improving access to employment, skills and competences (Guideline 6), enhancing the functioning of labour markets and the effectiveness of social dialogue (Guideline 7), and promoting equal opportunities for all, fostering social inclusion and combatting poverty (Guideline 8).

In addition, the Joint Employment Report monitors Member States' performance in relation to the Social Scoreboard set up in the context of the European Pillar of Social Rights. The Pillar was established as an inter-institutional Proclamation by the European Parliament, the Council and the Commission on 17 November 2017. It identifies principles and rights in three areas: i) equal opportunities and access to the labour market, ii) fair working conditions, and iii) social protection and inclusion. Monitoring of progress in these areas is underpinned by a detailed analysis of the Social Scoreboard accompanying the Pillar.

The Joint Employment Report is structured as follows: an introductory chapter (Chapter 1) reports on main labour market and social trends in the European Union, to set the scene. Chapter 2 presents the main results from the analysis of the social scoreboard associated to the European Pillar of Social Rights. Chapter 3 provides a detailed cross-country description of key indicators (including from the social scoreboard) and policies implemented by Member States to address the Guidelines for Employment Policies.



KEY MESSAGES

Europe is making progress regarding the Social Scoreboard accompanying the European Pillar of Social Rights. In a context of improving labour markets and declining poverty, 13 out of the 14 headline indicators of the Social Scoreboard recorded an improvement over the last year, on average. Still, the economic recovery is not yet benefitting all citizens and countries in the same manner. Challenges with regard to specific principles of the Pillar are identified for a majority of Member States. The current economic upswing provides an opportunity for stepping up reforms aiming at improving the inclusiveness, resilience and fairness of labour markets and social protection systems, thus fostering convergence towards better living and working conditions in the EU. However, there are also downside risks to the recovery, which make it urgent for Member States to seize this opportunity.

Strong job creation continues, with employment reaching a record level in the EU. In the second quarter of 2018, 239 million persons were in employment in the EU, 14 million more than the lowest level reached in mid-2013, at the peak of the crisis. The employment rate of people aged 20-64 rose to 73.2% in the same period: with the current trend, the EU is well on track to reach the Europe 2020 target of a 75% employment rate in 2020. Employment growth in 2017 and the first two quarters of 2018 was spread among all main demographic groups, with the largest increases recorded by older workers (55-64) as in previous years. Yet, substantial disparities in employment rates across the EU suggest that there is room for further improvement, notably for those Member States that are still far from attaining their national Europe 2020 targets.

Unemployment has returned to its pre-crisis level, but remains high in a number of Member States. Thanks to the steady labour market recovery, the unemployment rate kept declining in 2017, to reach 6.9% in the second quarter of 2018. It now stands at the lowest level in ten years, more than 4 pps below the 2013 peak. In the euro area, the unemployment rate at 8.3% in Q2-2018 remains one percentage point higher than the lowest level registered in 2008. Unemployment rates are still particularly high in Greece, Spain, Italy, Croatia and Cyprus.

Employment gains continue to be more prominent in terms of employed persons than of hours worked. The volume of total hours worked in the EU kept increasing in 2017 – though more slowly than total employment – and is not yet back to the 2008 level. A high number of involuntary part-time workers (i.e. workers who are in part-time but would like to work more), still 1.3 million above 2008, also suggests that there is remaining slack in the labour market. The decreasing number of hours worked per person is nonetheless part of a structural trend that started in the early 2000s.

Household incomes continue rising in almost all Member States. Supporting the process of upward convergence, real per capita disposable household income grew more strongly in the Member States that most recently joined the Union. While these developments were mostly driven by favourable economic conditions, they benefitted also from reforms leading to improvements in the adequacy of social benefits, including minimum income schemes. Yet, in a number of countries the real gross disposable income per capita remains significantly below the pre-crisis levels. Household incomes grew more slowly than GDP, highlighting that income gains from the recovery have reached households only to some extent – thus raising questions about the inclusiveness of recent growth.

On the back of robust economic and labour market recovery, the share of people at-risk-of poverty or social exclusion decreased markedly in 2017. More than 5 million people exited from poverty or social exclusion, the largest decline since the recovery started. This change was driven mainly by decreasing numbers of people experiencing severe material deprivation and/or living in very-low work intensity households (since their respective peaks, these indicators dropped by respectively 15 and 8 million). The total number of people at risk of poverty or social exclusion, at 113 million people or 22.5% of total population in 2017, is now below pre-crisis levels. Estimates indicate that this trend is set to continue into the next year. Still, there remains a long way to reach the Europe 2020 poverty and social exclusion reduction target. The risk of poverty or social exclusion remains a challenge particularly for children, people with disabilities and people with a migrant background.

Real wage growth slowed down in 2017, but has picked up in 2018. Overall, wage growth remains below what could be expected given the positive labour market and economic performance. The modest wage dynamics over the past years can be explained by weak productivity growth, still low inflation expectations and remaining reserves in the labour market. In real terms, average wages still lag behind pre-crisis levels in many Member States and their growth remained, in 2017, below productivity growth. This is in line with a long-term trend: in the EU, between 2000 and 2017, real value added per person employed grew by 15.6%, while real compensation per employee grew by only 11.2%. Despite these developments, there is evidence that convergence in labour income levels is taking place although large differences remain between and within Member States.

Wage setting frameworks (including minimum wages) are starting to react to the improvement in labour market conditions. In particular, statutory minimum wages were increased in several countries, with the involvement of social partners. This development is important in view of the persisting high rates of in-work poverty in a number of Member States, which also require action in the areas of tax design and benefits adequacy. Against this background, it is important that the adjustment in minimum wage levels follows transparent and predictable rules, taking into account their impact on competitiveness, job creation and in-work poverty.

In a context of sustained job creation, some groups still face difficulties in reaping the benefits of the recovery. Job growth in 2017 has been mainly driven by women, older workers and high-skilled people. On the other hand, the employment rate of low-skilled workers is still below pre-crisis levels and remains almost 30 pps lower than that of high-skilled workers. Though increasing, the employment rate of young people is lower than in 2008 (by 2.7 pps). On a positive note, the share of young people neither in employment, education or training at 10.9% is now back to pre-crisis levels. People with a migrant background face employability challenges: the employment rate gap between native and non-EU born workers stood at 10 pps in 2017 (up from 4.5 pps in 2008). This gap is especially pronounced among migrant women. Finally, people with disabilities tend to participate less in the labour market; the potential to use their talents remains largely untapped.

Participation of women in the labour market continues to grow at a fast pace. The employment rate of women stood at 66.5% in 2017, almost 5 pps higher than in 2008. Still, the employment rate gender gap remains substantial, with considerable disparities across Member States. Though women are generally better qualified than men in terms of educational attainment level, the gender pay gap is high and only gradually declining. Women are over-represented in lower-paid sectors and occupations and work more frequently in underqualified jobs compared to their skills level. The impact of parenthood and caring responsibilities remains the main driver of lower employment rates, with underdeveloped services being a major impediment to stay in or return to employment. Furthermore, informal carers, the majority of whom are women, are more at risk of experiencing poverty and financial dependency, with career interruptions often translating into lower pension entitlements. A number of Member States are taking action to provide affordable and equal access to quality childcare and long-term care services, but important challenges remain. A more balanced distribution of paid family-related leaves between women and men would be beneficial in some cases. A few Member States are adapting their tax and benefits systems to remove disincentives to work for second earners. Concrete actions to tackle the gender pay gap are in place only in a limited number of countries.

The employment rate of older workers increased substantially over the last decade. For the age group 55-64 it rose from 45.5% in 2008 to 57.1% in 2017. Older workers were relatively more shielded from the recession and their employment rate kept increasing during the crisis, to then become an important driver of job recovery. The rising duration of working lives is explained by a number of factors, including increases in statutory retirement age, better access to care services, availability of flexible working arrangements and active ageing strategies. Several Member States are further promoting labour market participation of older workers, including by supporting flexible transitions into retirement and providing financial incentives both for employers and employees.

The labour market situation of young people continues to improve, but youth unemployment remains high in a number of Member States. While steadily declining and back to the 2008 level, the youth unemployment rate (age group 15-24) presents large differences across countries and very high rates in some of them. A still considerable part of the young people is economically inactive. Overall in the EU, almost 6 million people aged 15-24 were neither in employment, education or training (NEET) in 2017. If prolonged over time, detachment of young people from the labour market can have negative consequences for potential growth as well as negative effects for the individuals concerned, such as the depreciation of skills and a higher risk of poverty and social exclusion later in life. Actions taken by Member States in line with the Council Recommendation establishing the Youth Guarantee are key drivers for improvement.

Ensuring access to quality and inclusive education and training enables younger generations to become engaged and active citizens, helping them integrate into the labour market and society. Europe is progressing towards reaching the 2020 headline target on early school leaving rate of 10%, and has almost reached the target of 40% on tertiary education attainment. Still, large differences persist across Member States and among population groups (for instance, between men and women and between native- and non-EU born people). The high shares of low achievers in basic skills, and the strong correlation of educational outcomes with socio-economic status and labour market outcomes are a matter of concern. Member States are undertaking steps to improve their education systems, notably to further reduce drop-out rates, foster equal access and improve educational outcomes among disadvantaged learners. More (and more efficient) investment in education and training systems is a priority in some Member States. Improving the quality of higher education and its labour market relevance is also high on the agendas of Member States, especially in the context of rising tertiary education attainment rates.

Technological changes and related transformations in labour markets crucially require upskilling and reskilling of the working age population. Having a labour market relevant qualification is increasingly important for workers to adapt to a rapidly changing environment. In the EU, the ratio between low-qualified adults and the number of jobs requiring a low qualification level is, on average, three to one. Yet, low skilled and older people are significantly less likely to participate in adult learning programmes than the average. Significant gaps remain in terms of digital skills: more than 40% of adults in the EU do not have basic digital skills, with peaks of 70% in some Member States. This implies that a significant part of the population cannot access a large variety of services, with negative impacts on inclusion and productivity.

Member States are adapting their skills development systems and are developing strategies to improve labour market relevance of training, with a view to facilitating learners’ transitions into and within the labour market. This includes making skills and qualifications easier to be understood and recognised throughout Europe and taking into account learning outside institutional contexts. Vocational education and training systems are being reviewed and updated, with the goal of improving their labour market relevance and promoting access, but challenges remain in these domains. The provision of incentives to disadvantaged groups to participate in adult learning, together with proper guidance and the provision of financial support to companies for training their staff are important policy levers to achieve better outcomes. Member States continue to increase the offer of learning and qualification opportunities to low-skilled adults, in line with the Upskilling Pathways initiative. Promoting development of digital skills occupies an important place among Member States’ education and skills priorities.

The incidence of atypical forms of work is roughly stable at EU level, but high labour market segmentation remains an issue for a number of Member States. The share of total employees on a temporary contract did not change significantly over recent years, hovering around 14% on average. Globally, more than half of temporary employees are "involuntary", reaching 70% or more in 12 Member States. In several countries, the combination of high shares of temporary contracts and low transition rates towards permanent contracts is symptomatic of labour market duality. This is a matter of concern as atypical workers experience lower job quality and higher in-work poverty risks. In addition, around one quarter of all self-employed workers in the EU can be classified as "vulnerable" or "concealed". Also, survey results seem to suggest that the proportion of people earning more than half of their income from platform work may now have reached around 2% (2017) and is expected to increase, which underlines the importance of increased policy focus on this development. Reforms in the area of employment protection legislation are taking place in some Member States, with the aim to achieve a better balance between flexibility and security and avoid segmentation. These include, in some cases, stricter conditions for using temporary contracts, or larger scope for collective bargaining to define their framework. Regulation on new forms of work, including platform workers and own-account workers, has been initiated in a few Member States.

Unemployment benefits of adequate amount, reasonable duration, accessible to all workers and accompanied by effective activation measures are key to support jobseekers during transitions. The design of these systems varies to a considerable extent from one Member States to another, across all dimensions. Recent reforms in this domain have mostly focused on strengthening activation requirements for jobseekers receiving benefits, for instance by reinforcing job search obligations and conditions to accept a new job. Concerns remain about the coverage of atypical workers, who often do not have full access to the system, and the absence of coverage for the self-employed – issues highlighted in the Commission's proposal for a Council recommendation on access to social protection for workers and the self-employed.

Effective active labour market policies and Public Employment Services are crucial to ensure well-functioning and inclusive labour markets. Active labour market policies improve labour market matching and increase the chances of job seekers to find a new job. Their role is especially important to foster the integration of the long-term unemployed. Public Employment Services are the main institutions in charge of supporting job search efforts of the unemployed and referring them to activation measures. Yet, participation and investment in active labour market policies differs significantly across the EU. Similarly, the effectiveness of Public Employment Services in providing job search support is heterogeneous across and, sometimes, within Member States. While important measures are being taken in a majority of Member States, with a focus on the provision of individualised services, several Member States have scope to reinforce their active labour market policies systems. Member States made progress on implementing the Council recommendation on the integration of long-term unemployed, with further action needed to foster cooperation across different actors and on improving outreach towards inactive people. Public Employment Services are continuing their reform agenda, in the framework of the European PES Network. While some Member States took further measures to promote the integration of migrants into the labour market (with a focus in particular on refugees), systematic approaches are lacking and there is a need to invest more into upskilling and recognition of skill and qualifications, efficient labour market policies and support by Public Employment Services. 

For the first time since the crisis income inequality in the EU decreased slightly in 2017, driven by faster increases in the incomes of lower income households. This suggests that the recovery begins to reach the most vulnerable. Available estimates indicate that this positive trend will continue in a majority of Member States. Still, in 2017 the richest 20% of the population had a disposable income that was 5.1 times higher than that of the poorest 20% in the EU (from 5.2 in 2016), with large variation across Member States. Income inequality is above pre-crisis levels in some countries, often linked to unequal opportunities in access to education, training and social protection, and correlated to poor labour market outcomes. Some Member States are undertaking inequality-reducing policies, notably in the design of minimum wage setting and tax and benefit systems. To break the transmission of inequalities across generations, Member States can take further action in different areas, such as by fostering equal opportunities in education and training, ensuring access to quality healthcare and other services, promoting gender equality and addressing regional disparities.

The measured impact of social transfers (excluding pensions) on poverty reduction continued to decline in 2017. This is in line with reduced automatic stabiliser effects in a phase of economic expansion, though large variations remain among Member States. The poverty-reducing impact of social transfers fell mostly in long-standing members of the EU, while it increased in countries that joined more recently (suggesting that convergence is taking place). The overall outcome depends on improved labour market conditions (and related changes in the characteristics of those at risk of poverty) as well as changes in benefits adequacy and coverage, including the fact that benefits sometimes lag behind generally increasing incomes. The adequacy of minimum income benefits varies significantly among Member States, as shown by the results of the related benchmarking exercise.

Member States continue taking measures to modernise social protection systems, strengthening coverage and adequacy of benefits and services. Steps are being taken to improve access to social protection, notably for non-standard workers and the self-employed who continue to face significant gaps. Some innovations are also being introduced with regard to new forms of work. Work on improving adequacy of benefits continues, though in some cases there have been delays. A number of Member States are improving their minimum income schemes by combining adequate levels of support with access to enabling goods and services, and with incentives to (re)integrate into the labour market, in an active inclusion approach. Some Member States are enhancing integrated delivery of services (such as social assistance, employment and other social services). In a context where housing-related expenditures amount to a significant share of many households’ incomes and the evolution in the number of homeless is not improving, some Member States have undertaken reforms to improve access to housing, either through the provision of incentives or via preventative measures.

Demographic change and rising life expectancy present the pension, healthcare and long-term care systems with a clear need to adapt. The demand for long-term care and healthcare is growing and the needs are changing as populations age. Over the next five decades, the number of Europeans aged 80+ is set to double. By 2050, there will be only two persons in active age (15-64) for one person over 65 year old, compared to three today; the ratio is already on the decrease. Therefore, while measures to improve financial sustainability are still high on many Member States’ agendas, ensuring pension adequacy gains in importance. The income replacement and poverty prevention capacity of pensions varies significantly among Member States. Member States are increasingly taking steps to safeguard pension adequacy through minimum guarantees and indexation of benefits, promote flexible retirement, tailor pension accrual conditions to diverse categories of workers and enhance the role of supplementary pensions. 

Improving access to quality healthcare and long-term care, together with increasing their effectiveness, is a guiding principle for reforms in Member States. For some Europeans, costs and waiting times remain important barriers for the accessibility of healthcare. Thus, in a number of Member States reforms of healthcare systems focus on improving their effectiveness by better coordination and a stronger role assigned to primary care and prevention. Measures are taken to improve training and working conditions of health workers. In the area of long-term care, most care is still provided by family members, given the lack of comprehensive schemes to cover the need for old age care in the majority of Member States. Current reforms intend to combine a system of support for informal and family carers by public institutions and a network of community and institutional services, with attention to sustainability in the face of demographic challenges.

A well-functioning social dialogue is a key element of the European social market economy. It helps strengthening social cohesion and reducing conflicts in the society, to the mutual benefit of workers, employers and governments. The involvement of social partners in the preparation of reforms can improve their design and implementation, increase ownership among citizens and eventually lead to better socio-economic outcomes. However, the degree and impact of social partners' involvement varies considerably among Member States and is weak in several cases. While there is no one-size-fits-all model for social dialogue practices, in some Member States there is clear scope for improving the capacity of social partners and providing them with an adequate framework for predictable and timely consultation including in all key stages of the European semester. Similarly, using the experience of civil society organisations can play and plays an important role to ensure that reforms are designed and implemented effectively. Yet, the degree of engagement with societal stakeholders varies significantly among Member States, with insufficient capacity to actively participate in the policy debate being an issue in some of them.



1.    OVERVIEW OF LABOUR MARKET AND SOCIAL TRENDS AND CHALLENGES IN THE EUROPEAN UNION

This section presents an overview of labour market and social trends and challenges in the European Union, providing a detailed analytical account of major employment and social policy areas.

1.1    Labour market trends

Labour market conditions continue to improve, with employment reaching a record level in the EU. On the back of strong economic growth (2.4%), the number of people in employment in the EU increased by 1.6% in 2017, the largest annual increase rate since the start of the recovery. Total employment kept increasing in the first two quarters of 2018, to reach 238.9 million 2 – about 3.2 million jobs more than one year ago, and the highest level ever reached in the EU. Since employment started recovering in mid-2013, more than 14 million additional jobs were created.

The employment rate (of people aged 20-64) is on a steady rise, getting closer to the Europe 2020 target. It increased slightly faster than in 2016, by 1.1 percentage point to 72.2% in 2017 and continued rising in the first two quarters of 2018, up to 73.2% (Figure 1). If the positive trend continues at the current pace, the EU would be well placed to reach its Europe 2020 target of a 75% employment rate. The situation continues to improve also in the euro area, where the employment rate reached its highest level at 71.9% in Q2-2018. These positive developments are supported by a continuing upward trend in labour market participation. In Q2-2018, the activity rate (15-64) achieved a record high at 73.8% (73.5% in the euro area). The activity rate in the EU has been increasing at a constant pace, even during the crisis, closing the gap with the United States. In 2017, older workers and women continued to drive the increase in labour force participation.

The unemployment rate is now back to its pre-crisis level. Thanks to the steady job creation trend, the unemployment rate kept falling, down to 6.9% in Q2-2018 – a level not recorded in the EU since Q2-2008 and more than 4 pps below the 2013 peak. The improvement is less substantial in the euro area, where the unemployment rate, at 8.3% in Q2-2018, remains one percentage point higher than the lowest level recorded in 2008. These positive trends are associated with a continuing decrease in the long-term unemployment rate (i.e. share of the unemployed for at least one year within the active population), which dropped in the EU by 0.5 pps year-on-year to 3% in Q2-2018 (3.9% in the euro area). Though unemployment (including long-term) decreased across all Member States over the last year, a significant dispersion of unemployment rates persists (as shown in Section 3.1.1), with some countries still far from their pre-crisis minimum levels.

Figure 1: Employment and unemployment rates in the EU and euro area

Source: Eurostat, LFS. Note: seasonally adjusted figures for Q2-2018.

Youth unemployment also continues to decrease quickly, dropping by 1.8 pps year-on-year to 15.2% in Q2-2018 (16.9% in the euro area). This level corresponds to the minimum reached just before the crisis (Q2-2008) and is now almost 9 pps below the 2013 peak. Still, youth unemployment remains high in some Member States, with rates above 30% in Spain, Italy and Greece (see Section 3.2.1). Continuing improvements are recorded in terms of people aged 15-24 who are neither in employment, nor in education or training (NEET), whose rate declined by 0.6 pps to 10.9% in 2017, a similar level as in 2008.

The decrease in youth unemployment is mirrored by a parallel increase in the educational attainment of youth: the rate of early leavers from education and training (aged 18-24), which has been steadily declining over the last decade, with the latest decrease (by 0.1 pps) reached 10.6% in 2017. This is very close to the Europe 2020 target of 10%, though there is room for further reduction 3 . Tertiary educational attainment for those aged 30 to 34 continued increasing steadily, reaching 39.9% in 2017 – almost reaching the Europe 2020 target of 40%.

The recovery continues to be more prominent in terms of employed persons than of hours worked. The volume of total hours worked in the EU increased by 1.2% in 2017, continuing the positive trend started in 2015. Nonetheless, this increase remains below that of total employment (which rose by 1.6%, see above), and implies that the number of hours worked per capita decreased. Total hours worked are not yet back to their 2008 peak level. This evidence points to remaining slack in the labour market, as also signalled by other indicators. In 2017, there were almost 9 million involuntary part-time workers in the EU (i.e. workers who are in part-time but would like to work more), declining from a peak of 10.3 million in 2013, but still 1.3 million above the level of 2008. In a longer-term perspective, the moderate dynamics of working hours is part of a structural shift, linked to a growing incidence of part-time work over the last 15 years and changing preferences of workers as concerns working time arrangements. As a result, the number of hours worked per person is on a gradually declining trend since 2000 4 .

Employment gains were spread among all the main demographic groups in 2017. As in previous years, older workers (55-64) recorded the largest increase (Figure 2): the number of employed persons in this group rose by 4.3% in 2017, pushing the employment rate up to 57.1%, the highest ever (almost 12 pps higher than in 2008). The number of young people (15-24) in employment increased slightly faster than in 2016 (1.6% vs. 1.3%). Yet, the youth employment rate has not fully recovered from the crisis, as at 34.7% in 2017 it remains 2.7 pps below the 2008 level. Nevertheless, in view of a stable activity rate (at 41.7% in 2017 vs. 41.6% in 2016) the employment gains translated into a continuing decrease in unemployment for this age group. Employment increased slightly faster among women than among men in 2017 (1.5% vs. 1.3%). Yet the employment gender gap remains almost unchanged at 11.5 pps, 0.1 pps below 2016 (though significantly lower than the pre-crisis value of 15 pps in 2008). Although increasing by 1.7 pps at 63% in 2017, the employment rate of non-EU born people aged 20-64 remains 10 pps lower than that of native-born. The gap is higher for women (around 14 pps).

Figure 2: Employment rates and employment growth across different groups in the EU

Source: Eurostat, LFS.

The recovery keeps increasing the number of high skilled workers in the economy. The number of people with higher education in employment increased by 2.9% in 2017 (age group 25-64), while a moderate increase of 0.8% was recorded among medium-skilled workers (i.e. those with upper secondary education). On the contrary, the number of low-skilled workers (i.e. with lower secondary education or below) dropped by 0.4%. Since the overall population of low-skilled workers aged 25-64 is on the decline (by 2.7% compared to 2016) – as part of a trend reflecting population ageing and higher educational attainment among younger generations – this group's employment rate actually increased from 54.3% in 2016 to 55.6% in 2017. The gap between employment rates of low- and high-skilled workers slightly decreased from 30.5 pps in 2016 to 29.7 pps in 2017, but remains very high and indicative of the room for action to increase the employability of people with low education level. Detailed trends by Member States are provided in the following sections of the report.

Temporary employment, as a proportion of total employment, remained almost stable in 2017, while part-time employment decreased marginally. Following a similar trend as in recent years, the recovery is fostering job creation in both permanent and temporary contracts, which rose respectively by around 2.7 million and 0.8 million in 2017 (corresponding to percentage increases by 1.7% and 2.9%). Nonetheless, as a proportion of total employment, the share of temporary employees remained almost stable, slightly increasing by 0.1 pps to 14.3% (age group 15-64). For a second year in a row, the proportion of part-time workers (age group 15-64) decreased marginally (by 0.1% to 19.4% in 2017), remaining almost 2 pps above the 2008 level. On the upside, as also mentioned above, the share of involuntary part-time workers is decreasing significantly (from 21.1% in 2016 to 19.8% in 2017) though remaining substantial. Self-employment (15-64) over total employment continued a slow decline, down to 13.7% in 2017 (from 14.0% in 2016 and 14.4% in 2013).

In terms of sectoral developments, employment continued to shift towards services. In line with the trend of recent years, the largest number of jobs was created in services 5 (2.8 million additional persons in employment in 2017, or +1.6% compared to 2016; based on national accounts). From 2008 to 2017, the share of employment in services in the EU increased continuously from 70.1% to 73.9%. With an increase by 2%, construction recorded the largest increase in employment since the recession, consolidating the recovery started in 2015; however, the number of employed persons is still almost 15% lower than in 2008. Industry also showed a solid expansion (by 1.5%, the highest rate since 2007). Finally, after a long series of declines, employment in agriculture slightly increased by 0.3% in 2017.

1.2    Social trends

The number of persons at risk of poverty or social exclusion (AROPE 6 ) 7 has substantially declined in 2017 and is now below pre-crisis levels. The declining trend in this indicator continued for a fifth consecutive year in 2017, falling to 113 million persons (or 22.5% of the total population) in line with the recovery in employment and the increase in disposable income. As a result, in 2017 there were 5 million fewer persons at risk of poverty or social exclusion in the EU than before the crisis (in 2008), while the decline from the peak in 2012 amounts to nearly 11 million. However, given the setback of the crisis, the Europe 2020 headline target (20 million fewer people at risk of poverty or social exclusion compared to 2008) remains far from reach. The current overall decline in the indicator is driven by all its three sub-components, though to a different extent (see below and Figure 3).

Figure 3: Percentage of population at-risk-of-poverty or social exclusion (AROPE) and subcomponents (2005-2017).

Source: Eurostat, SILC. Note: the legend is explained in footnote 6.

A sharp fall in the number of persons suffering from severe material deprivation (SMD) brings the figures to its lowest level in recent history, reflecting rising standards of living. Over 3 million people were relieved of severe material deprivation in the year to 2017, bringing the overall number of persons affected down to 34.8 million or 6.9% of the EU population (0.6 pps less than in 2017 and below 2008). This decline represents a significant improvement for a fifth year in a row, reflecting an improving material situation of households. In spite of these positive developments, a large variation between Member States persists (see section 3.4).

The record increase in employment rates has contributed to reducing the number of persons living in quasi-jobless households by 3.7 million in 2017. As a share of the population aged 0-59, this corresponds to a decrease from 10.5% in 2016 to 9.3% in 2017 – below 10% for the first time since 2009. Yet, both the rate and the number of people affected remain above pre-crisis levels.

The percentage of the population at risk of poverty has shown its first decline since the crisis. Following an increase in the number of people at risk of poverty until 2014, this indicator had stabilised in the following two years. In 2017 the rate fell by 0.4 pps to 16.9%, or nearly 2 million people – marking the first year after the crisis in which the incomes of poor households have risen faster than median incomes. The latest data from Eurostat flash estimates 8 suggest that this decline in poverty rates is set to continue. However, the at risk-of-poverty rate for persons living in households with very low work intensity has increased for a fourth consecutive year, and is now at a record-high 62.3%. This points to persistent gaps in the adequacy of social benefits in several countries and has been identified as a trend to watch by the Social Protection Committee 9 .

In spite of the overall decline in the share of people at risk of poverty, in-work poverty remains high. In 2017, 9.6% of the working population had household income below 60% of the national median, a figure that was unchanged from 2016, and which remains well above the 2008 figure of 8.5%. The increases to date have affected both part-time and full-time workers, although the former remain at substantially higher risk of poverty (15.8% as against 8.0%). The trend has affected in particular younger workers (aged under 30 years) who face a higher and increasing risk vis-à-vis workers aged 30 years and over. In-work poverty has been identified as a trend to watch by the Social Protection Committee both from a short- and long-term perspective 10 .

Poverty figures for the most vulnerable people indicate modest improvements, reversing previous trends. The poverty gap, which measures how far away from the poverty line those at risk of poverty are, fell in 2017. It decreased modestly from 25% to 24.7%, yet remaining well ahead of pre-crisis levels. This suggests that the relative income position of the most vulnerable is slightly improving. For the unemployed, the poverty risk declined for the first time since the crisis, yet at 48% it remains near record-high levels.

Despite overall improvements, the risk of poverty or social exclusion facing children remains high. The AROPE rate for children (aged 0-17) continues to decline, falling from 26.4% to 24.5% in 2017, well below the pre-crisis level. However, 62.9% of the children of low-skilled parents remain at risk of poverty or social exclusion, against only 9% for the children of high-skilled parents. For the children of non-EU born people, the proportion of those at risk of poverty or social exclusion has been rising throughout the post-crisis period, reaching 34.5% in 2017, more than twice the rate facing the children of native-born parents. A key driver of child poverty is the growth in the number of single parent households, whose share has increased from 4.4% of the EU population in 2011 to 4.8% in 2016 (but fell to 4.3% in 2017). The at-risk-of-poverty rate for children from single parent households is twice the average for children overall, a gap that continues to widen.

Household incomes continue to grow, but at a slower rate than the economy as a whole. While the gross disposable income of households (GDHI) rose for a fourth year in a row to 2017, and is now well ahead of the 2008 level, the annual increase still lags behind GDP growth (GDHI grew under 1% for the year to 2017, while real GDP per capita rose by 2.2% in the same period). This highlights that households' income gains are lagging behind overall income growth in the economy. As shown in Section 3.4, gross disposable household income per capita, in real terms, remains below pre-crisis levels in a number of Member States.

Increases in income inequality in post-crisis years have begun to reverse in 2017, though not compensating yet for past increases. On average, the richest 20% of households in Member States have an income that exceeds that of the poorest 20% of households by over five times. The S80/S20 ratio rose from 5.0 to 5.2 between 2008 and 2016, also driven by poor labour market conditions and stagnating incomes, especially in the lowest part of the distribution. In 2017, this ratio has begun to decline, back to 5.1 on average in the EU. The latest Eurostat flash estimates suggest the decline is set to continue.

 



2.    SNAPSHOTS FROM THE SOCIAL SCOREBOARD

The European Pillar of Social Rights, established as an inter-institutional Proclamation by the European Parliament, the Council and the Commission on 17 November 2017, sets out a number of key principles and rights to support fair and well-functioning labour markets and welfare systems. It is designed as a compass for a renewed process of convergence among Member States towards better socio-economic conditions.

The European Pillar of Social Rights is accompanied by a social scoreboard to monitor performances and track trends across Member States 11 . The scoreboard provides a number of indicators (headline and secondary) to screen the employment and social performance of Member States on selected indicators along three broad dimensions, identified in the context of the Pillar: (i) equal opportunities and access to the labour market, (ii) dynamic labour markets and fair working conditions, and (iii) public support / social protection and inclusion. Since the 2018 edition, the Joint Employment Report integrates the social scoreboard, whose results (as concerns headline indicators) are summarised in this Chapter. The analysis is placed in the broader reform context presented in Chapter 3.

2.1    The scoreboard explained    

The social scoreboard is a central tool for monitoring performance in the employment and social domains, and convergence towards better living and working conditions. In particular, it helps monitoring the situation of Member States on measurable dimensions of the Pillar, complementing the existing monitoring tools, in particular the Employment Performance Monitor and the Social Protection Performance Monitor 12 . It notably includes 14 headline indicators that assess employment and social trends at large 13 :

-Equal opportunities and access to the labour market:

§Share of early leavers from education and training, age 18-24

§Gender gap in employment rate, age 20-64

§Income inequality measured as quintile share ratio - S80/S20

§At-risk-of-poverty or social exclusion rate (AROPE)

§Young people neither in employment nor in education or training (NEET rate), age 15-24

-Dynamic labour markets and fair working conditions:

§Employment rate, age 20-64

§Unemployment rate, age 15-74

§Long-term unemployment rate, age 15-74

§Gross disposable income of households in real terms, per capita 14

§Net earnings of a full-time single worker without children earning an average wage 15

-Public support / Social protection and inclusion:

§Impact of social transfers (other than pensions) on poverty reduction 16

§Children aged less than 3 years in formal childcare

§Self-reported unmet need for medical care

§Share of population with basic overall digital skills or above.

Headline indicators are analysed using a common methodology agreed by the Employment Committee and the Social Protection Committee (see Annex 3 for details). This methodology evaluates the situation and developments in Member States by looking at levels and yearly changes 17 of each of the headline indicators included in the social scoreboard. Levels and changes are classified according to their distance from the respective (unweighted) EU averages. Member States' performances on levels and changes are then combined (by using a predefined matrix) so that each Member State is assigned to one out of seven categories ("best performers", "better than average", "good but to monitor", "on average/neutral", "weak but improving", "to watch" and "critical situations"). On this basis, Table 1 provides a summary of the readings of the scoreboard according to the latest figures available for each indicator.

A careful and non-mechanical reading of the table is warranted. For this purpose, a detailed analysis of the fourteen indicators, including longer-term trends and additional indicators, when relevant, is presented in Chapter 3. In addition, the forthcoming country reports will provide an in-depth analysis of all "critical situations" and additional socio-economic and policy background to better qualify country-specific challenges in the context of the European Semester. Together with further analysis included in the Employment Performance Monitor and the Social Protection Performance Monitor, this will provide an analytical basis for the subsequent Commission proposals for Country Specific Recommendations where appropriate.

2.2    Evidence from the social scoreboard

The analysis of the Scoreboard points to a continuing recovery in the labour market and social situation for the EU as a whole 18 . On average for the EU 19 , 13 out of 14 headline indicators recorded an improvement over the last available year (i.e. 2017 or 2016 depending on data availability), with only one (impact of social transfers on poverty reduction) slightly deteriorating, though in line with a reduced impact of automatic stabilisers in a phase of economic expansion. The most significant progress was recorded in (overall and long-term) unemployment rates, which decreased in all Member States in 2017, with only one "critical situation" highlighted. Developments in the employment rate and the at-risk-of-poverty or social exclusion rate were also positive across the board, as the large majority of Member States recorded an improvement compared to the previous year.

Most Member States face challenges on at least one headline indicator, but the overall count has reduced compared to last year. Considering the three more problematic classifications altogether, i.e. "critical situation", "to watch" and "weak but improving", most Member States are flagged at least once – with the exception of Germany, Finland, France, the Netherlands and Sweden. Considering "critical situations" only (i.e. indicators whose level is much worse than average, and either not improving sufficiently fast or deteriorating further), the number of flagged Member States decreased from 14 in the 2018 Joint Employment Report (with a smaller set of indicators) to 13 in the current exercise (Estonia, Malta and Portugal have left this group, while Hungary and Latvia have joined it). The count of challenges points to an improvement across the board. Across the 14 domains assessed, overall 117 "critical situation", "to watch" or "weak but improving" cases are identified, i.e. about 31% of the total number of assessments (compared to 33% in the 2018 JER); of these, 39 are "critical situations", corresponding to 10% of all assessments (compared to 13% in the 2018 JER) 20 .

Looking at the three broad dimensions covered by the scoreboard, similarly to the 2018 Joint Employment Report, problematic flags appear more frequently in the area of "public support/social protection and inclusion", with an average of 9.3 cases (of which 3.5 "critical situations") per indicator. The impact of social transfers on poverty reduction appears as the most challenging indicator, with flags for 11 Member States (of which 5 in the bottom category).

The dimensions of "equal opportunities and access to the labour market" and "dynamic labour markets and fair working conditions" follow, with an average of 8.6 and 7.4 flagged cases per indicator respectively (3.2 and 1.8 "critical situations" each). In the first domain, the most flagged indicator is early leavers from education and training (10 times). In the latter, net earnings of a full-time single worker without children earning an average wage appears as the indicator with most numerous challenges (12 flags).

As for the last year, the situation of Member States and the severity of the respective challenges vary widely. Greece, Romania and Italy still present "critical", "to watch", or "weak but improving" assessments on ten or more indicators – of which "critical situations" flagged respectively for 7, 4 and 6 indicators (see Table 1). For these countries, challenges are spread uniformly across the three domains (still, "better than average" performances are flagged once for Greece and Italy, respectively on the early school leavers rate and self-reported unmet needs for medical care, and twice for Romania, on the unemployment rate and the growth of gross disposable household income per capita). In terms of overall count, Croatia and Spain (9 challenges each), Bulgaria (8 challenges), Cyprus, Latvia and Portugal (6 challenges each) follow. By contrast, Netherlands is a "best performer" or "better than average" on 11 headline indicators, followed by Czech Republic and Sweden (10 indicators each), Austria, Germany and Slovenia (8 indicators each).

When looking at equal opportunities and access to the labour market, the largest improvements were recorded, on average, in terms of at-risk-of-poverty or social exclusion and NEET rates, while progress has been more muted as concerns early school leaving, gender employment gap and income inequality (however, the latter indicator shows the first decrease in the post-crisis period). When looking by indicator:

·Spain, Italy and Romania face a "critical situation" when it comes to early leavers from education and training, compared to Croatia, Ireland, Poland and Slovenia as the "best performers";

·Greece, Italy, and Romania score critical on the gender employment gap, compared to Finland, Lithuania and Sweden as "best performers";

·Bulgaria, Spain, Latvia and Lithuania face a "critical situation" in terms of income inequality compared to the best performance of the Czech Republic, Finland, Slovenia and Slovakia;

·The situation as concerns the at-risk-of-poverty or social exclusion rate is critical in Bulgaria and Greece, compared to the Czech Republic and Finland as "best performers";

·Cyprus, Croatia, Greece and Italy face a "critical situation" when looking at NEETs while Austria, Czech Republic, Germany, the Netherlands, Slovenia and Sweden perform the best.

Turning to dynamic labour markets and fair working conditions in the EU, on average the situation improved over the last year across all indicators, notably employment and unemployment rates (both overall and long-term), gross disposable household income (GDHI) per capita and net earnings of a full-time single worker without children earning an average wage. When looking by indicator:

·Croatia, Greece, Italy and Spain face a "critical situation" when it comes to their employment rate, compared to the Czech Republic, Estonia, Germany, the Netherlands, Sweden and the UK as the "best performers";

·No country scores critical on the unemployment rate (Cyprus, Croatia, Greece and Spain are marked "weak but improving", while Italy is "to watch"); on the other hand, the Czech Republic scores as "best performer";

·Italy scores critical on the long-term unemployment rate (no "best performers" identified through the methodology, while 14 countries are "better than average")

·The growth in per capita GDHI is seen as a "critical situation" in Greece and Cyprus, compared to Bulgaria and Poland as "best performers";

·The situation on net earnings of a full-time single worker without children earning an average wage is assessed as critical for Hungary and Slovakia, while Austria, Germany, Luxembourg, the Netherlands and the United Kingdom are the "best performers".

As regards public support and social protection and inclusion, the situation has improved over the last year in terms of childcare availability, self-reported unmet need for medical care and digital skills but it worsened in terms of the impact of social transfers on poverty reduction, as mentioned above. When looking by indicator:

·Bulgaria, Greece, Italy, Latvia and Romania face a "critical situation" when it comes to the ability of their social transfers to reduce the risk of poverty. This compares to Denmark, Finland, Hungary and Sweden as the "best performers";

·Bulgaria, the Czech Republic, Greece, Poland and Slovakia score critical on the participation of children aged less than 3 to formal childcare, compared to France, Luxembourg, the Netherlands and Portugal as "best performers";

·Latvia faces a "critical situation" in terms of self-reported unmet need for medical care (no "best performers" identified through the methodology, while 12 countries are "better than average");

·Bulgaria, Croatia and Romania face a "critical situation" when looking at levels of digital skills, while Finland, Luxembourg, the Netherlands and Sweden perform the best.

Table 1. Summary of headline indicators of the Social Scoreboard

Note: On 26 October 2018: income quintile ratio, AROPE, impact of social transfers on poverty reduction and self-reported need for medical care not available for IE and UK; GDHI per capita growth not available for HR and MT; net earnings of a full-time single worker without children earning the average wage not available for CY; individuals' level of digital skills not available for IT. Breaks in series and other statistical flags are reported in Annexes 1 and 2.

Box 1. Benchmarking - state of play

The importance of benchmarking as a tool to support structural reforms and foster upward convergence in the employment and social fields was recognised by the Five Presidents' Report of June 2015 21 and further underlined in the Reflection Paper on EMU deepening of May 2017 22 . The Communication of 26 April 2017 establishing a European Pillar of Social Rights 23 identified benchmarking as a key vehicle to support the implementation of the Pillar within the European Semester. Since then, benchmarking exercises have been developed and discussed with Member States in several areas.

In particular, the Employment Committee (EMCO) and the Social Protection Committee (SPC) have agreed on a common approach in three steps: (1) identification of key challenges and a set of high level outcome indicators relevant for the policy area under consideration; (2) performance indicators which allow for benchmarking the performance; (3) the identification of policy levers, which are accompanied by general principles for policy guidance and, when available, by specific indicators. At this stage, reference values for policy levers are not set, as the aim is to allow for comparisons across Member States.

The benchmarking framework on unemployment benefits and active labour market policies was used for the first time in the 2018 European Semester. In this context, a comparative analysis of specific design features and performance of unemployment benefit systems, notably as concerns eligibility and adequacy aspects, was included in the 2018 Joint Employment Report and Country Reports. Following the endorsement by the Employment Committee (EMCO), elements of the framework related to the activation component of unemployment benefit schemes (e.g. availability-to-work conditions attached to receiving unemployment benefits) and to labour market services in support of job seekers are now integrated in the European Semester 2019 and are presented in this report.

Upon endorsement by the Social Protection Committee (SPC), the European Semester 2019 also sees a full integration of the benchmarking framework on minimum income benefits, covering adequacy, coverage and activation components of minimum income schemes, including as concerns their relation with in-kind services (healthcare, education and housing). Finally, this Semester cycle benefits, for the first time, from the results of the benchmarking framework on adult skills and learning, which was agreed with the Employment Committee in October 2018. Work is currently ongoing within relevant Committees on possible additional benchmarking frameworks, such as on pension adequacy, to be utilised in future Semester cycles.


3.    EMPLOYMENT AND SOCIAL REFORMS – MEMBER STATES PERFORMANCE AND ACTION

This section presents an overview of recent key employment and social indicators and measures taken by the Member States in priority areas identified by the EU employment guidelines 24 , as adopted by the Council in 2018. For each guideline, recent developments on a selection of key indicators are presented, as well as policy measures taken by Member States. As concerns the latter, the section draws on Member States’ National Reform Programmes 2018 and European Commission sources 25 . If not specified otherwise, only policy measures implemented after June 2017 are presented in the report. An in-depth analysis of recent labour market developments can be found in the Labour Market and Wage Developments 2018 report 26 and the Employment and Social Developments in Europe Review 2018. 27  

3.1    Guideline 5: Boosting demand for labour 

This section looks at the implementation of the employment guideline no. 5, which recommends Member States to create conditions promoting labour demand and job creation. It first presents an overview of unemployment and employment rates by Member State, complementing the analysis at EU level made in Chapter 1, to highlight the relevance of the job creation challenge across countries. It then looks at self-employment dynamics, as a proxy for entrepreneurship and as a source of employment growth per se (aspects of self-employment related to new forms of work are discussed in Chapter 3.3). Finally, it investigates wage and tax wedge developments as key macroeconomic determinants of hiring decisions. Section 3.1.2 reports on policy measures implemented by Member States in these areas to promote labour demand, including hiring subsidies.

3.1.1    Key indicators

The decline in unemployment accelerated in 2017. For the first time after the crisis, the unemployment rate decreased in all EU Member States. A quicker-than-average drop occurred notably in some Member States suffering from very high unemployment rates – Greece, Spain, Croatia, Portugal, all experiencing a reduction by 2 pps or more – pointing to a clear convergence trend 28 towards lower unemployment levels. Nonetheless, a strong decrease occurred also in some low-unemployment countries – by more than 1 pp in the Czech Republic, the Netherlands, and Poland. As evident from Figure 4 (which looks jointly at levels and changes according to the agreed methodology for assessing headline indicators of the social scoreboard 29 ), the dispersion of unemployment rates remained substantial in 2017, with values ranging from around 3% in the Czech Republic to 21.5% in Greece. The unemployment rate also remains high in Italy, with only limited improvements compared to 2016. In spite of the steady decrease since 2013, in many Member States the unemployment rate remains much higher than in 2008 (Figure 5) – especially in Croatia, Greece, Cyprus, Italy and Spain. On the contrary, in Germany, Hungary and Poland the unemployment rate in 2017 was more than 2 pps lower than before the crisis.

Figure 4: Unemployment rate (15-74) and yearly change (Social Scoreboard headline indicator)

Source: Eurostat, LFS. Period: 2017 levels and yearly changes with respect to 2016. Note: Axes are centred on the unweighted EU average. The legend is presented in the Annex.

Figure 5: Unemployment rate (15-74), multiannual comparison

Source: Eurostat, LFS.

While employment rates registered strong increases, significant disparities remain. All Member States recorded increases, with the exception of Denmark 30 (a 0.5 pp decrease, though from a very high level). As shown in Figure 6, employment rates are only partially converging: Member States characterised by lower employment rates tend to experience a somewhat faster growth than average – but the situation remains scattered. Employment rates (age group 20-64) diverge widely, ranging from 57.8% in Greece to 81.8% in Sweden in 2017. At the lower end, the employment rate remains much below the EU average in some of the countries most hardly hit by the crisis, namely Greece, Croatia, Italy and Spain (flagged as "critical situations"). Nine countries are already above the 75% target, while the six best performers (Sweden, Germany, Estonia, the Czech Republic, the United Kingdom and the Netherlands are all close to or above 78%).

Figure 6: Employment rate (20-64) and yearly change (Social Scoreboard headline indicator)

Source: Eurostat, LFS. Period: 2017 levels and yearly changes with respect to 2016. Note: Axes are centred on the unweighted EU average. The legend is presented in the Annex.

Average employment rates in the EU have been constantly rising since 2013 and are now well above the pre-crisis peak. Hungary and Malta, who had the lowest employment rates in the EU in in 2008, have experienced the highest long run increases (+11.8 pps and +13.8 pps respectively in 2017) and are now slightly above the EU average. However, disparities are still high. In particular, similarly as for unemployment rates, employment rates remain below 2008 levels in those countries most affected by the crisis, such as Greece, Italy, Croatia, Spain, Cyprus and Ireland (Figure 7).

Figure 7: Employment rate (20-64), multiannual comparison

Source: Eurostat, LFS.

The sustained job growth did not affect the long-term declining trend in self-employment. In absolute numbers, self-employment fell by 0.5% in 2017 (after a slight increase in 2016 31 ). It continued to decline also in relative terms, representing 13.7% of total employment in 2017 (down from 14.0% in 2016 and 14.4% in 2013). The share of self-employed men over the total men employed, at 17.2%, remains much higher than for women (9.7%). Trends in self-employment are less negative if the primary sector and manufacturing are excluded. In particular, self-employment has steadily increased over the last ten years in most of the sectors related to services.

Overall figures are the result of diverging developments across Member States, sectors and sub-groups. However the trends in self-employed do not seem linked to performance in employment at Member State or sectoral level. In terms of age and education groups, developments reflect to a large extent the changing composition of the workforce: self-employment kept falling for people aged 15-49 while it increased markedly for people aged 50-64 and 65 and over (+1.2% and +4.0%, respectively). The level of education is also an important factor, as in 2017 the number of self-employed decreased by 3.2% for people with up to lower secondary education and increased by 1% for people with tertiary education. Finally, an increase in the number of self-employed born outside the EU is recorded. Aspects related to self-employment as an atypical form of work are discussed in detail in Chapter 3.3.

Wage growth remains moderate in most Member States. Nominal wage growth in the EU has remained subdued during the recovery, but started picking up in 2017, with compensation per employee increasing by 2.1%. Differences across countries are considerable, with wages growing generally faster in Member States with lower levels and in those that are not members of the euro area. The increase in nominal compensation per employee was the highest in Romania (16.0%), Lithuania (9.1%), Hungary (7.9%), Latvia (7.9%) and Bulgaria (7.5%), pointing to wage convergence between Eastern and Western Europe. At the low end, nominal wages declined in Finland (-1.1%) and Croatia (-1.1%) and remained flat in Spain, Italy and Greece.

In real terms, wage growth slowed down in 2017 compared to 2016. In the EU, real consumption wages (i.e. wages adjusted for the change in consumer prices) grew by 0.5% in 2017, down from 1.2% in 2016. In the euro area, real wages almost stagnated (+0.2%). A decline was recorded in eight countries: Spain, Italy, Greece, Croatia, Portugal, the Netherlands, Austria and Finland. The slowdown compared to 2016 was partly due to the increase in consumer prices, which was not matched by the rise in compensations. However, even when looking at real production wages (adjusted for the GDP deflator), the picture is almost unchanged, with real wage growth increasing only slightly in the EU in 2017 (by 0.7%) and a decline registered in ten countries (Figure 8).

Figure 8: Real compensation per employee, HICP and GDP deflator, 2017 annual % change