Chapter 10. Tackling deep and persistent inequalities in the labour market

This chapter discusses the role of policies and institutions for tackling deep and persistent inequalities in the labour market. To this end, it discusses the role of policies and institutions for ensuring that workers are not at a disadvantage in the labour market because of their socio-economic background and, once in the labour market, all workers have good prospects for accessing quality employment and moving up the earnings ladder. The chapter also considers the role of the taxes-and-benefits system for ensuring that everybody benefits from increased prosperity.


The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.


Not only have wages tended to decouple from productivity and inequalities in earnings and incomes tended to increase in many countries (Chapter 2), there are also growing concerns that people do not get a fair chance in the labour market due to a lack of social mobility (OECD, 2018[1]). In many countries, socio-economic background is an important determinant for labour market success and the possibility of improving one’s socio-economic position over the life-course (OECD, 2018[2]). In other words, concerns about the depth and persistence of inequality have gone hand-in-hand.

While there is no consensus on the acceptable – or desirable – level of inequality, very large and persistent inequalities in the labour market can have significant adverse social and economic consequences. First, they tend to reduce well-being as people do not only care about their own success in the labour market but also that of others (Clark, Frijters and Shields, 2008[3]; OECD, 2014[4]). Second, high inequality can reduce human capital accumulation by limiting the ability of low-income people to invest in their skills (Cingano, 2014[5]), with adverse consequences for social mobility and long-term growth. Third, inequality can reduce mutual trust as well as trust in economic and institutional processes , with potentially important consequences for social cohesion, economic growth and mainstream politics (Gould and Hijzen, 2016[6]; Alesina and Giuliano, 2015[7]).1

Yet, deep and persistent inequalities in the labour market are not inevitable as illustrated by the large differences observed across countries. Governments have a range of instruments at their disposal that can be used to tackle inequality or promote equal opportunities. The way they choose to address these challenges depends to an important extent on societal values regarding the importance of solidarity, redistribution and equality, as well as many other factors related to the national context (e.g. demographic structure, macroeconomic situation, the fiscal and administrative capacity of the state).

After a short section to set the scene (Section 10.1), this chapter discusses three broad and complementary policy approaches that can be used to tackle deep and persistent inequalities in the labour market: i) policies and institutions to promote equality of opportunity through greater social mobility across generations (Section 10.2); ii) policies and institutions to promote better career opportunities for those already in the labour market (Section 10.3); iii) tax-and-benefit policies to contain excessive inequalities and ensure that nobody is left behind (Section 10.4). The last section concludes.

10.1. Setting the scene

Inequality of earnings is not only high in many countries, but also tends to persist over the life-course and across generations (OECD, 2015[8]; OECD, 2018[2]; Garnero, Hijzen and Martin, 2019[9]). Figure 10.1 provides an indication of the degree of earnings inequality in a given year (short-term inequality) as well as over the entire working life (life-time inequality). Life-time inequality tends to be lower than inequality in a given year. This reflects the role of earnings mobility during the life-course that results from movements into and out of employment or up and down the earnings ladder. However, the difference between life-time and short-term inequality is modest in most countries, suggesting that the equalising effect of earnings mobility over the life-course is fairly small and most inequality is permanent: on average across countries, life-time inequality is only about 20% lower than short-term inequality. Moreover, earnings mobility is not sufficiently different across countries to bring about significant changes in the cross-country ranking of inequality. Life-time inequality tends to be high in countries where wage inequality is high (e.g. Japan and the United States), earnings mobility is low (e.g. Japan and Portugal) or unemployment is high and unemployment insurance limited (e.g. Greece and Italy).

Figure 10.1. Inequality is not only high by historical standards but also very persistent
Gini coefficients of earnings in 2010 (short-term inequality) and cumulative earnings over the working life (life-time inequality)

Note: Earnings inequality is measured across all active persons of working age and reflects hourly wages, working time and unemployment benefits when unemployed. Since data that allow following persons over their working lives are not available on a cross-country basis, life-time inequality is estimated using statistical simulation methods rather than directly observed in the data.

Source: Garnero, A., A. Hijzen and S. Martin (2019[9]), “More unequal, but more mobile? Earnings inequality and mobility in OECD countries”, Labour Economics,


Tackling deep and persistent inequalities requires addressing three important policy priorities:

  • Reducing the impact of socio-economic background on labour market outcomes. This requires making sure that all children have access to quality education and health services irrespective of their parents’ income. It also requires ensuring that socio-economic background, related to gender, ethnicity, religion, age or LGBTI status, does do not become a source of discrimination in the labour market. Box 10.1 provides a discussion of the gender gap in the labour market and the role of policies.

  • Promoting access to quality employment and prospects for career advancement. Promoting the outcomes of those already in the labour market requires measures that enhance workers’ employability and earnings potential by investing in adult learning and making adult learning systems more inclusive by promoting participation among vulnerable workers with limited basic skills. However, it also requires ensuring that sufficient opportunities are available and workers have good opportunities for career progression. This requires policies that facilitate job transitions towards better-paid and more stable employment.

  • Providing effective redistribution. This requires designing tax-and-benefit systems that contain excessive inequalities, protect individuals and households from negative income shocks and prevent financial hardship, without undermining work incentives and, as a result, the long-term career prospects of those involved. Taxes-and-benefits policies can also contribute to more equal opportunities by alleviating financial constraints that limit access to quality education and health services of children in poor households.

Box 10.1. Bridging the gender gap in labour market outcomes

Women continue to have lower labour market incomes than men

Despite major improvements in the education of girls, rising female labour force participation and widespread laws against gender discrimination, women still lag behind men in the labour market along a number of key dimensions. The gender gap in labour income (GGLI) – defined as the average gap in labour income between men and women – remains a global phenomenon (see Figure 10.2). It averages 40% across the OECD, with considerable variation across countries, ranging from 20% in Slovenia to over 60% in Turkey. On average in the OECD, gender differences in employment explain 40%, differences in hours worked 20% and differences in hourly wages 40% (which is consistent with an hourly wage gap of 16%).

Gender disparities tend to widen over the life course

The GGLI tends to widen over the working life, with most of it arising in the first half of the career. On average across the OECD, the GGLI amounts to around 25% for people in their 20s, widens to 40% for those in their 30s and early 40s, but increases only little after that. The evolution of the GGLI over the life-cycle reflects different labour market mobility patterns for men and women. Most importantly, childbirth often has long-lasting effects on women’s careers, in terms of time spent out of employment, limited hours of work, earnings and career progression. Part-time work can help to avoid that women withdraw from the labour market after childbirth, but also may limit opportunities for career progression within their firm. Women also experience fewer job-to-job transitions, partly reflecting the role of family commitments. Weaker job mobility, within and between firms, harms career progression and earnings growth (Barth and Kerr, 2017[10]).

Targeted measures can reduce gender inequalities

Effective strategies to promote gender equality consist of a number of key elements: i) family policies that improve access to childcare, correct disincentives to work for second earners and move towards gender-neutral taxes-and-benefit systems; ii) measures to encourage behavioural changes, including combating long hours, getting fathers more involved in caring, and promoting more equal forms of paid leave; and iii) inducing changes in the workplace, including increased take-up of part time and flexible working-time arrangements. The relative importance of these different elements varies from country to country. In countries where women are less likely to enter the labour market at any stage (e.g. Greece, Italy and Spain), policies should focus on ensuring high female labour participation at young ages. By contrast, in countries where women tend to withdraw from the labour market following childbirth (e.g. Australia, Austria and several Eastern European countries) and those where women tend to spend large parts of their careers in part-time work (e.g. Germany, the Netherlands and Switzerland) attention should focus on policies promoting the reconciliation of parental care responsibilities with work and an equal sharing of care responsibilities in the household.

Figure 10.2. The gender gap in labour income
Decomposition of the gender gap in labour income (20-64 years old), 2013-15, percentages

Note: For Canada and Turkey, data on earnings refer to wage and salary only. For Norway, the breakdown of hourly wage gap and hours gap is not available.

Source: OECD (2018[11]), OECD Employment Outlook 2018,


10.2. Promoting social mobility across generations

Most people agree that democratic societies should promote equality of opportunities – i.e. try to ensure that all people have similar chances to succeed in the labour market regardless of their socio-economic background. Inequalities in outcomes and opportunities, however, are inherently linked: in the absence of equal opportunities, today’s inequalities transmit into inequalities for the next generation. High inequality in outcomes in turn undermines equality of opportunity – or “social mobility” across generations – as children from well-off backgrounds get a substantial head start in life. Indeed, empirical evidence shows that social mobility across generations tends to be higher in societies where outcomes are less unequal (a relationship described by the so-called “Great Gatsby Curve” (Corak, 2006[12])).

Intergenerational mobility is low in many OECD countries

People’s socio-economic outcomes tend to be strongly related to those of their parents, i.e. intergenerational mobility tends to be low (OECD, 2018[2]). About 40% of the differences in fathers’ earnings carry over to the next generation in the OECD on average. The degree of earnings persistence between generations varies substantially across countries, from below 20% in the Nordic countries to 60% in Hungary, Luxembourg and some emerging economies.2 These figures imply that, in an “average OECD country”, it could take about five generations for the offspring of a family in the bottom decile of the income distribution to reach the average income (Figure 10.3). In the Nordic countries, this process could only take two generations, while in some emerging economies, it could take around ten.

Figure 10.3. Intergenerational mobility tends to be low
Expected number of generations for the offspring from a family at the bottom 10% of the income distribution to reach the mean income

Note: These estimates are simulation-based and intended to be illustrative. They are based on estimates of earnings persistence (elasticities) between fathers and sons and the current level of household incomes of the bottom decile and the mean, assuming constant elasticities.

Source: OECD (2018[2]), A Broken Social Elevator? How to Promote Social Mobility,


The strong persistence in earnings across generations reflects various factors, including low educational mobility, the inter-generational transmission of health outcomes and the role of parental social networks. Educational attainment is highly persistent across generations: 42% of people with low-educated parents – i.e. those without an upper-secondary qualification – do not finish high school, while only 12% complete tertiary education. By contrast, people whose parents have completed tertiary education nearly always obtain at least a high school degree (only 7% do not) and 63% obtain a tertiary degree. Health outcomes also tend to be transmitted from one generation to the next. Having grown up in a family with little or no wealth and having ill parents are two main predictors of poor health, and young people with poor health are four times more likely to be NEET than their peers (OECD, 2016[13]). Parental social networks also play a role in sustaining earnings inequalities across generations, particularly for those from high-income backgrounds, e.g. Narayan et al. (2018[14]), Corak and Piraino (2011[15]).

Improving social mobility

There is nothing inevitable about socio-economic privilege or disadvantage being passed on from one generation to another. Large differences in social mobility across countries indicate that there is room for policies to make societies more mobile. Improving social mobility requires investments in education and health as well as effective family and regional policies.

Public investment in quality education, from early childhood education and care through tertiary education

For the youngest, good-quality childcare and pre-primary school attendance can have large beneficial effects for all, but in particular for children from more disadvantaged backgrounds with long-term consequences for educational attainment and labour market outcomes.3 In the United States, for instance, the Infant Health and Development Program (IHDP), which provided free full-day early education with a focus on language development, was found to substantially boost cognitive ability, particularly for children from low-income households.4 Similarly, children attending a French pre-school (école maternelle) from the age of two showed significantly improved cognitive and non-cognitive skills at age six and better literacy and numeracy from grades three to nine (Filatriau, Fougère and Tô, 2013[16]). Good-quality childcare and pre-primary school attendance can also have beneficial effects in the long-run, particularly for disadvantaged children (Van Huizen and Plantenga, 2015[17]; Ruhm and Waldfogel, 2012[18]). The introduction of universal childcare in Norway, for instance, was found to have strong positive effects on labour market participation and to reduce welfare dependency later in life (Havnes and Mogstad, 2011[19]). Many countries offer pre-primary education as a statutory right to all children from the age of three, and attendance is frequently subsidised or free. In most OECD countries, 80% or more of 3-5 year-olds are enrolled in pre-primary education or primary schools.

At compulsory-schooling level, policies to support social mobility need to address the often large performance gap for students from disadvantaged backgrounds and fight early school leaving. PISA mathematics scores are about 20% lower for 15-year-olds with parents of lower socio-economic status than for those with parents of a high status. Young people with less-educated parents and those from workless households are also much more likely to leave school without an upper-secondary degree. These early school leavers are strongly overrepresented among young people who are not in employment, education or training (NEETs), accounting for about one third of all NEETs across OECD countries (OECD, 2016[20]). Effective strategies to support students from disadvantaged backgrounds include targeted investments in low-performing schools and those located in marginalised communities (e.g. for smaller classes, better educational material and physical infrastructure and the recruitment and training of teachers and support staff),5 eliminating grade repetition, avoiding early tracking and deferring student selection to upper-secondary education (OECD, 2012[21]). The OECD Action Plan for Youth (OECD, 2013[22]) recommends a set of educational measures to tackle high youth unemployment. It places particular emphasis on second-chance programmes for young people who quit school without an upper-secondary degree and quality apprenticeships and internship programmes that help smooth the school-to-work transition.

Students from disadvantaged socio-economic backgrounds also tend to be under-represented in higher education, especially at prestigious institutions. At the most competitive U.S. colleges, for instance, the large majority of students come from households in the top income quartile, while the bottom half accounts for only a tiny portion e.g. Carnevale and Strohl (2010[23]), Chetty et al. (2017[24]). Targeted outreach in upper-secondary schools and counselling/tutoring services during tertiary education can be effective at increasing tertiary-level educational enrolment and completion rates of young people from underprivileged backgrounds. Differential admission policies, such as class-based affirmative action, can help high-potential candidates from disadvantaged backgrounds pass the initial admissions screening.

Early health support, notably for children and young people from lower socio-economic backgrounds

Targeted investments in health support for children and young people from low socio-economic backgrounds can help to address low social mobility related to health. Pre- and post-natal homecare for low-income families has been shown to increase child well-being (Greenberg and Shroder, 2004[25]). Pre-natal and childhood health in turn are significant predictors of adult health and employment status (Case, Fertig and Paxson, 2005[26]). Moreover, many governments are taking steps to promote a culture of healthy eating and regular physical activity in schools (OECD, 2017[27]). Direct health support for young people should be part of an effective effort to reduce early school leaving and promote a successful school-to-work transition. To tackle the rapidly growing share of young people diagnosed with mental health problems easily, accessible mental health services are needed.6 More generally, the absence of universal health coverage, high privately-financed out-of-pocket payments and the lack of public screening programmes are associated with greater income-related health inequalities (Devaux and de Looper, 2012[28]).

Family policies that address early disadvantage

Policies that provide financial support to low-income parents and help them invest in their children’s human capital can also help to promote social mobility. There is ample evidence of a strong causal impact of parental income on children’s health, school performance and future success in the labour market. Financial support that effectively targets low-income parents, such as child benefits or earned income tax credits, can therefore contribute to reducing the achievement gap of children from disadvantaged backgrounds.7 Similarly, policies that promote the labour market participation of both parents – such as generous parental leave entitlements, especially for fathers, the availability of affordable childcare, flexible working-time arrangements and fiscal incentives for second-earners to participate in the labour market – can help improve the financial situation of families and hence promote intergenerational mobility. In emerging economies, conditional cash transfer programmes that provide income support to poor households conditional on children’s school attendance or health check-ups have had positive effects on educational attainment, health and future labour market outcomes (OECD, 2015[29]).

Tackling spatial segregation and strengthening lagging regions

The geographical concentration of disadvantaged families in low-income neighbourhoods and large economic differences across regions tend to reinforce some of the mechanisms responsible for the transmission of disadvantage across generations. Low-income urban neighbourhoods and less developed regions tend to suffer from poorer-quality public services (notably in education and healthcare), a weaker transport infrastructure and fewer employment opportunities. This creates additional hurdles for children and young people from disadvantaged backgrounds. Urban planning and regional development policies that support equal access can help address these. This could include housing policies that promote the geographical mobility of families to help them move to regions with better economic prospects (see Chapter 14). It could also involve measures that limit school choice – to avoid that large differences in school quality in urban areas induce social segregation – and facilitate a more equitable allocation of children to schools in terms of their socio-economic background.8 It may further involve measures to improve the quality of education and health services in low-income communities, for example by offering more attractive pay packages for education and health professionals.

10.3. Bridging the labour market divide

Deep and persistent inequalities in the labour market reflect the growing imbalance between relative skill supplies and demands, partly driven by globalisation and technological change. However, segmentation in product and labour markets – as reflected by significant differences in firm and contract-specific pay – can also play an important role when labour markets are imperfectly competitive. This section first discusses the role of adult learning policies for vulnerable workers (“the supply side of the labour market divide”) and then turns to the discussion of segmentation in product and labour markets (“the demand side of the labour market divide”).9

Adult learning systems need to become more inclusive

Adult learning is key for addressing labour market inequality, particularly for persons with low skills. Taking part in training provides adults with the opportunity to up-skill and re-skill in line with changing labour market demands. It also supports positive labour market transitions and promotes upward mobility to better-paid jobs.

Participation in adult learning tends to be weak, particularly among those who need it most

According to data from the OECD Survey of Adult Skills (PIAAC), only 41% of adults in the surveyed countries participate in adult learning in any given year. This varies markedly across countries, ranging from over one half in countries such as New Zealand, the Netherlands and the Nordic countries to less than a quarter in countries such as Greece, Italy and Turkey. Importantly, participation in adult learning and training is much lower among low-skilled workers, i.e. those most in need of upskilling and reskilling. On average across countries, the participation gap in job-related training between low-skilled and more skilled persons is 23 percentage points.

Engaging adults who currently do not participate in training is a major task for all stakeholders involved, as most of them (82%) have no interest or motivation to do so, according to data from the Survey of Adult Skills (PIAAC). This implies that across OECD countries participating in the PIAAC survey, about half of adults neither participate nor want to participate in adult learning. With adult learning being a key lever to improve labour market outcomes of low-skilled workers and prepare workers for changing skill needs, it will be crucial to find effective ways to motivate this part of the population to take part in education and training.

Promoting adult learning among disadvantaged groups

A comprehensive strategy for improving the inclusiveness of adult learning systems focuses on those most in need of up- and re-skilling, but least likely to participate in adult training and consists of the following features: i) providing good quality information and guidance; ii) tackling participation barriers; and iii) engaging employers and unions in the provision and financing of adult learning.10

Providing information and guidance

Policy should enable adults to make good education and training choices. Yet, there is evidence to suggest that adults, in particular those with low skills, are not particularly good at recognising the need to develop their skills further (Windisch, 2015[30]). Hence, engaging adults in learning must go beyond providing training opportunities and involve raising public awareness, providing high-quality individualised advice and guidance services and developing on-line databases to inform workers of re-training opportunities and labour market needs.

Public awareness campaigns are a good way to promote the benefits of adult learning and reach out to underrepresented groups (Kowalczyk et al., 2016[31]). The Institute for Adult Education in Slovenia, for example, has been organising an annual life-long learning week since 1996, which today includes more than 1 500 events implemented in cooperation with partner organisations across the country. Portugal launched its adult learning program Qualifica in 2016/2017 with a large-scale public awareness campaign titled “Get yourself together and invest in your future!”. In 2017, the city of Brussels, Belgium, launched a mobile information centre Formtruck to engage job-seekers, the low-qualified and young people not in employment, education and training in adult learning.

Career guidance helps individuals understand their skill set and development needs and navigate available learning opportunities. To be effective, it must take into account timely labour market and skill needs information. Career guidance is typically delivered through a range of channels, including public employment services (PES), specialised guidance services, education providers and social partners. Dedicated services tend to be the most effective as they are specifically set up to provide in-depth and timely skill-needs information drawn from effective skills anticipation systems. Career New Zealand provides a wealth of services, increasingly available online, ranging from tools that allow users to explore careers and find jobs that match their skills and qualifications. Also in New Zealand, the Occupation Outlook is a mobile phone app that allows exploring study and career options, with extensive information on labour supply and demand in over 100 occupations. Some countries have developed one-stop shops to ensure individuals get all the information they need to make informed decisions. The House of Orientation (Maison de l’Orientation) in Luxembourg provides a one-stop shop for education and labour market orientation.11

Online databases can also be useful for end-users (workers, prospective students) as well as trainers, counsellors and other adult learning experts. Australia’s national directory of vocational education and training providers and courses ( allows users to search VET qualifications by industry and to access information about average course fees, course duration, available subsidies and average employment outcomes. While employment outcomes are currently available by qualification, a plan exists to make them available at provider level.

Reducing barriers to participation

Policy efforts must also focus on those individuals who want to take up or continue adult learning opportunities, but face a variety of obstacles to doing so. 18% of people who do not take part in job-related adult learning would like to do so. Equally, 34% of people who take part in job-related adult learning would like to continue, but do not do so. Barriers to participation are diverse and include a lack of time due to work (29% of those indicating they would like to participate or continue in adult learning but do not do so), a lack of time due to family commitments, a lack of financial resources (both 16%), inconvenient time or location of the learning opportunity (12%) and a lack of employer support (7%), according to PIAAC data. For low-skilled adults, an additional barrier to participation is the lack of basic skills, which are often set as pre-conditions for participation in adult learning.

Most adult learning opportunities require individuals to have a basic level of literacy and numeracy and some have specific skills or qualifications as a pre-condition for entry. As some skills are acquired through labour market experience, validating and certifying existing skills can help to re-engage individuals in formal learning (OECD, 2019[32]). Effective recognition of prior learning must be transparent, streamlined and ensure the buy-in of all relevant stakeholders, including employers and education and training providers. In Portugal, the launch of Qualifica programme in 2017 included the creation of a credit-based system for professional training in line with European frameworks, an online tool for the recording of qualification and competences (‘Passaporte Qualifica’), and the establishment of a network of 300 Qualifica centres which provide free services related to information and guidance, as well as the recognition, validation and certification of skills (OECD, 2018[33]).12

In many countries, several forms of flexible learning provision exist that facilitate combining training with work and family commitments, including on a part-time basis, in the evenings, weekends, distance learning, or in a modular and/or credit-based format. Modular approaches are especially helpful in allowing adult learners to focus on developing the skills they currently lack, complete self-contained learning modules on these skills and combine these modules to eventually gain a full (formal) qualification. Research suggests that such provisions can broaden access to formal qualifications, in particular for disadvantaged groups (Kis and Windisch, 2018[34]). The Danish adult learning system allows students a high degree of flexibility by enabling them to combine modules from different providers and across different subjects. For example, individuals working towards a vocational qualification in Labour Market Training Centres (Arbejdsmarkedsuddannelse) can select from a wide range of vocational training courses but also tap into subjects provided by the general education system.

Statutory education and training leave – i.e. the right to take time off during working hours to participate in education and training – is typically regulated in national legislation or set out in collective agreements and can be universal or targeted at specific groups such as the low-skilled or small and medium-sized enterprises (OECD, 2017[35]). In order to ensure its uptake, many countries provide financial incentives for learners and employers alongside statutory leave. In Belgium, full-time private-sector employees participating in recognised training and education programmes are entitled to training leave for up to 180 hours per year. During training leave, workers receive full-pay (up to a ceiling), while employers can be compensated for the wages paid during training leave by the Federal Public Service for Employment, Labour and Social Dialogue.

Financial incentives are used widely to encourage adults’ participation in education and training and they often include specific provisions targeted at the low-skilled. An equitable system of financial incentives should focus on addressing cases of under-investment in adult learning. A range of tools can be mobilised including wage or training subsidies (also voucher-based), tax incentives, subsidised loans or training/time account schemes (OECD, 2017[35]). For example, the Austrian city of Vienna supports its employed and unemployed inhabitants with below tertiary education through education accounts (Bildungskonto), which co-finance the costs of recognised education and training programmes or that of recognition procedures for prior learning. To promote equity and avoid deadweight loss – financial support to persons who would have participated in training anyway – financial incentives are often targeted at under-represented groups, such as the low-skilled, by limiting entitlements to such groups or allowing for more generous support (as in the Vienna initiative described above). In France, the Compte Personnel de Formation is a personal account that provides individuals with training credits based on the time spent in employment during the year. Entitlements are transferable between employers. Low-skilled individuals receive double the credits of their higher-skilled counterparts (48 hours per year as opposed to 24 hours) and the ceiling for the accumulated credits is also significantly higher (400 hours compared with 150 hours).

Engaging employers and unions

With much learning taking place in the workplace, the engagement of employers and unions in the design, implementation and financing of skill development opportunities is critical to raising participation in job-related training.

While employers play a key role in providing and financing job-related adult learning, underinvestment in training and education is common, particularly among small and medium enterprises. This often reflects a lack of information, capacity and/or resources. Many governments therefore actively seek to engage employers in adult learning activities through awareness campaigns, tailored support and financial incentives. For example, the Japanese System to Support Human Resource Development (Jinzai Kaihatsu Shien Joseikin) co-funds business expenses for training, with higher subsidies available for small and medium-sized enterprises.

Trade unions can also play a key role, building on their involvement in job design, work organisation, and the application of new technologies at work. Indeed, in many countries, unions engage in the promotion and management of training provisions. In the United Kingdom, Unionlearn – the skills and learning branch of Trade Union Confederation – assists its members in the delivery of learning opportunities and the management of the Union Learning fund. This initiative has been particularly successful in recruiting low-skilled workers into training courses.

In some countries with sector-level bargaining, unions and employer organisations collaborate to invest in the skills of the workforce. In the Netherlands, the sectoral training and development funds (Opleidings- en Ontwikkelingsfondsen) are social partner initiatives that are financed primarily through payroll levies, fixed in collective agreements. They provide learning possibilities to workers to maintain their skills and anticipate future skills needs. Similarly, the Finish social partners administer an education and training fund (Koulutusrahasto), which provides education and training grants to employed and self-employed adults, who meet certain eligibility criteria.13

Tackling labour market segmentation

When labour markets are imperfectly competitive, inequalities do not just depend on the skill structure of labour demand and supply, but also on the composition of firms and contracts. Moreover, the persistence of inequalities depends on the extent to which workers can take advantage of the different opportunities offered by firms and progress from fixed-term to open-ended contracts. The role of firms for labour market inequality is discussed in Box 10.2.

Contractual segmentation has tended to increase before the global financial crisis

Temporary contracts tend to be associated with lower wages, even after controlling for the characteristics of workers and their jobs (OECD, 2015[36]). Moreover, people on temporary contracts are less likely to receive bonuses (Venn, 2011[37]) and typically experience lower levels of labour market security and lower quality working environments (OECD, 2014[4]). The wage penalty associated with temporary contracts most likely reflects the weaker bargaining position of people employed on such contracts. People employed on temporary contracts also tend to experience weaker earnings growth over time, possibly due to more limited in human capital accumulation, further reinforcing pre-existing wage gaps (OECD, 2015[36]). Given the weaker outcomes associated with temporary contracts, it is particularly important that they enhance access to permanent jobs in the future. Indeed, to the extent that temporary contracts help workers gain access to stable jobs, a short-lived period of weaker employment conditions may well be worth the investment in the long-term.

In the late 1990s and early 2000s, many OECD countries engaged in reforms to tackle persistent unemployment and promote labour market flexibility. These reforms in many cases involved easing the use of flexible contracts, while maintaining a relatively strict employment protection for open-ended contracts (and are often referred to as “partial labour market reforms”). Today’s evidence, however, suggests that such reforms did little to promote overall employment, but increased the incidence of temporary contracts (Kahn, 2010[38]), with adverse consequences for job quality and inclusiveness.14 The positive effect of partial labour market reforms on the probability of finding work was probably offset by an increase in the risk of losing one’s job, as a result of the substitution of permanent contracts by temporary ones (Cahuc, Charlot and Malherbet, 2016[39]). As suggested by Panel A of Figure 10.4, this substitution effect tends to be stronger and the use of temporary contracts more important, the stricter the employment protection of permanent contracts. This is consistent with econometric findings in Bassanini and Garnero (2013[40]) for a large number of OECD countries as well as Hijzen, Mondauto and Scarpetta (2017[41]) for Italy and Centeno and Novo (2012[42]) for Portugal.

Moreover, an increase in temporary employment at the expense of permanent employment tends to slow the transition from temporary to permanent work. Panel B of Figure 10.4 shows that a higher use of temporary work is associated with a significantly lower probability that a person with a fixed-term contract will have an open-ended contract three years later. In other words, a higher incidence of temporary contracts increases labour market segmentation. This may seem to contradict findings from a literature showing that temporary jobs provide a stepping stone to permanent work, see e.g. Booth et al. (2002[43]), Ichino et al. (2008[44]), Jahn and Rosholm (2014[45]) and OECD (2015[36]). However, these studies adopt an individual perspective in which the structure of employment is taken as given and hence do not shed light on the consequences of an increase in the number of temporary jobs on the career opportunities of workers.15 In fact, there is growing evidence that an excessive use of temporary work can have adverse consequences for individual pathways. For example, the liberalisation of the use of temporary contracts in Spain in 1984 reduced life-time earnings among the low-skilled as a result of weaker job stability and increased non-employment spells (García-Pérez, Castelló and Marinescu, 2018[46]).

In sum, reforms that promote the use of temporary work in a context of strict protection employment for workers on open-ended contracts risk being counterproductive by inducing an excessive use of fixed-term work. A widespread use of temporary work tends to have only a limited impact on improving employment opportunities for disadvantaged groups, while at the same time, tends to come at the expense of permanent employment, reducing job quality, slowing the transition of temporary to permanent work and reinforcing long-term inequalities in the labour market. Reforms in recent years have therefore tended to follow a different pattern, by focusing on ways to reduce the regulatory gap in employment protection between permanent and temporary contracts (see Box 7.3 in Chapter 7 for details).

Figure 10.4. Employment protection, temporary work and contractual segmentation

Note: *** Statistically significant at the 1% level.

Source: OECD estimates based on the European Union Statistics on Income and Living Conditions (EU-SILC), the German Socio-Economic Panel (SOEP) for Germany and OECD Indicators of employment protection,


Tackling contractual segmentation

One possible strategy to reduce contractual segmentation is to make hiring on temporary contracts more difficult and costly, while leaving unchanged or simultaneously reducing dismissal costs for permanent workers. Reforms along these lines typically take the form of restricting renewals of temporary contract, their cumulative duration or their scope. One concern with such reforms is that restricting the use of temporary contracts tends to be difficult to enforce in practice. The enforcement of employment protection is mainly dependent on legal complaints of employees who consider themselves wrongfully treated. In the context of temporary contracts, employees often benefit from non-enforcement, at least in the short-run, and therefore have little incentive to lodge a case (Muñoz-Bullón, 2004[47]).16

Another strategy is to ease the employment protection of permanent workers in countries where this is overly restrictive. Reforms of this type have been relatively common in recent years and typically focus on extending or clarifying the conditions for economic dismissal and limiting the costs of unfair dismissal. An early assessment of such a reform in Spain suggested that this reduced the reliance of temporary work by increasing the share of permanent contracts among new hires (OECD, 2013[48]). However, by making it easier to dismiss workers for economic reasons, such reforms also increase the risk of job displacement and financial hardship for incumbents. They should, therefore, be accompanied by the provision of adequate unemployment benefits as part of a comprehensive activation strategy (see Chapter 9). Recent labour market reforms in Italy and Portugal that increased the ease with workers on open-ended contracts can be dismissed also involved measures to strengthen the effectiveness of their unemployment benefit and activation systems.17

A third strategy is to focus on the convergence in termination costs across contract types. Full convergence would effectively imply moving to a single or unified contract. Under a single contract all existing contracts are replaced by a single open-ended contract, with the level of protection increasing in tenure.18 This option therefore does not provide an explicit instrument for time-limited activities beyond dismissal. This is problematic when compensation costs for unfair dismissal costs are high and their grounds wide-ranging or unclear.19 Under a unified contract, all contracts types are preserved, but tenure-dependent termination costs are harmonised across contracts. To the extent that the costs for unfair dismissal are high and uncertain, this option may still leave strong incentives for substituting permanent by temporary contracts. The effective operation of a single or unified contract, therefore, requires that the grounds for unfair dismissal are limited to discrimination and prohibitive grounds (see Chapter 7).

To evaluate these different reform options, it is useful to recall why temporary contracts exist in the first place. Temporary contracts provide employment flexibility to employers for dealing with product demand shocks, overcoming information asymmetries about worker productivity, and organising time-limited activities. Introducing employment protection entitlements (e.g. severance pay) from the start of a job spell for everyone reduces the risk of labour market segmentation, but also increases the cost of uncertainty and engaging in truly time-limited activities, with potentially important implications for job creation (or the use of alternative work arrangements such as self-employment). The challenge for policy-makers is to limit the statutory divide between contract types in terms of employment protection, while allowing for sufficient employment flexibility for firms, particularly at the start of the job spell.

Box 10.2. Segmentation across firms

While rising wage inequality reflects to an important extent growing returns to skills, recent evidence suggests firms are increasingly contributing to rising inequality as well. Indeed, a significant share of the rise in wage inequality has been linked to increased wage dispersion between firms, e.g. Song et al. (forthcoming[49]) for the United States and references therein. Rising wage dispersion across firms may reflect growing differences in firms’ compensation policies, the tendency for good workers to sort into good firms as well as that of similar workers to cluster in the same firm.

Firm performance. The growing dispersion in firm’s compensation policies in many countries is likely to reflect a growing variation in the ability of firms to offer more generous compensation packages due to growing productivity dispersion (Berlingieri, Blanchenay and Criscuolo, 2017[50]), or growing product market concentration (Diez, Leigh and Tambunlertchai, 2018[51]). Productivity dispersion has tended to rise as frontier firms have pulled away from the rest, possibly because they were better placed to take advantage of the opportunities presented by globalisation and technological change, while technology diffusion from the frontier to the rest of the economy has stalled. Globalisation and technological change are also likely to have contributed to growing market concentration, resulting in rising market shares and mark-ups at the technological frontier. To tackle high and persistent inequalities in the labour market it is therefore not enough to invest in skills development. It is equally important to invest in policies that can revive productivity growth in lagging firms or promote the reallocation of resources to more efficient firms (OECD, 2016[52]).

Sorting. A growing variation in the ability of firms to provide generous compensation packages does not necessarily explain why compensation packages have grown more diverse; firms also need to have an interest in doing so. One possibility is that firms use more generous compensation packages to attract better workers in the presence of complementarities in production between technologically advanced firms and skilled workers. High-performance work and management practices are one example since they tend to be more effective in the presence of skilled workers. The increased importance of information and communication technology may be another. The greater role of such production complementarities may have increased the dispersion in average pay across firms and reinforced the sorting of better workers into better firms. Sorting may also have increased as a result of reduced “frictions” in the labour market, i.e. the ease with which firms can create or terminate jobs and workers can move between jobs. While increased sorting may be good for economic efficiency, it also contributes to the depth and persistence of labour market inequality (OECD, 2015[8]).1

Segregation. Rising wage dispersion across firms also reflects the clustering of workers with similar workers in the same firm, so-called “segregation”.2 Increased segregation in terms of skills or occupations may reflect the growing importance of outsourcing. Human-resource practices may be most effective when tailored to the specific nature of work activities, but at the same time, there may be limited scope for differentiating HR practices within a single firm. This provides incentives for firms to concentrate on “core” activities and outsource non-core activities. Consistent with this argument, Goldschmidt and Schmieder (2017[53]) find, using data for Germany, that outsourcing is associated with lower wages for workers in outsourced activities. The resulting segregation of the workforce has raised concerns about increased competition on the basis of labour standards and reduced opportunities for career advancement (Weil, 2004[54]). Possible policy responses involve protecting workers through minimum standards related to for example wages, health and safety (statutory or collectively agreed) or establishing social norms that increase the responsibility of firms for labour standards in their supply chain.


1 As mentioned in Box 10.1, the increased importance of sorting may have reinforced existing gender inequalities as men are typically better positioned to take advantage of evolving job opportunities in the labour market than women who, in many cases, place a greater emphasis on family considerations when making career choices.

2 As opposed to the sorting described above, this is not driven by firm-specific wage premia and hence does not contribute to efficiency. Segregation also does not contribute to inequality in a given year since an increase in between-firm wage dispersion as a result of segregation is exactly offset by a reduction in inequality within firms. However, segregation is likely to reduce mobility over the life-course, and hence increase the persistence of inequality.

10.4. Fair and efficient redistribution

Redistribution through the taxes-and-benefits system can play a crucial role in making labour markets more inclusive by ensuring that the gains from economic growth are broadly shared in the population, including among families with low incomes.

Redistribution has an important role to play but has weakened over time

The redistributive effects of taxes and transfers among the working-age population can be quantified by comparing market income inequality among working-age households before taxes and transfers with disposable income inequality after taxes and transfers.20 On average across OECD countries, taxes and benefits reduce market income inequality among the working-age population by about one quarter (Figure 10.5). However, there is considerable variation across countries. The inequality-reducing effect of taxes and transfers amounts to about 40% in countries such as Finland, Ireland and Slovenia and less than 10% in countries such as Chile and Korea.21

Redistribution mainly reflects the role of cash transfers, which account for about two- thirds to three-quarters of the total inequality-reducing effect of taxes and transfers on average across OECD countries (Immervoll and Richardson, 2011[55]; Causa and Hermansen, 2017[56]). When measuring inequality in terms of the Gini, the redistributive impact of cash transfers is strongly associated with their size and less so with their targeting (Causa and Hermansen, 2017[56]). This explains why social insurance benefits – which account for a much larger share of public social expenditures than social assistance benefits – play a much larger role for redistribution than social assistance benefits, despite being less targeted.22

Personal income taxes account for the remainder of redistribution. In countries with relatively little redistribution, the role of personal income taxes tends to be relatively more important for redistribution (Japan, Korea, Israel and the United States). Social security contributions tend to be distribution-neutral and even are regressive in some countries. However, they are used to finance highly progressive benefits, implying that the social security system as a whole tends to be strongly redistributive.23

Figure 10.5. The equalising effect of taxes and transfers
Difference between the Gini coefficient of market income and disposable income as a share of the Gini of market income, working-age population (18-65), 2014 or latest available year

Note: The Gini index measures the extent to which the distribution of incomes among households deviates from perfect equal distribution. A value of zero represents perfect equality and a value of 100 extreme inequality.

Source: Causa and Hermansen (2017[56]), “Income redistribution through taxes and transfers across OECD countries”, OECD Economics Department Working Papers, No. 1453,


Over the last two decades, redistribution through taxes and benefits has fallen in the majority of OECD countries (Causa and Hermansen, 2017[56]).24 This reflects a fairly pervasive decline in the redistributive effect of cash transfers, driven by a reduction in their overall importance as well as their reduced targeting on the poor. The latter results from a shift in emphasis away from social insurance benefits for workless households to less generous social assistance benefits and a greater use of in-work benefits among working households to strengthen work incentives among poor families (Causa, Vindics and Akgun, 2018[57]). By contrast, the redistributive effect of taxes has changed only modestly on average across countries: a shift towards more progressivity in personal income taxes in the bottom end of the distribution in some countries was compensated by a slight decline in progressivity at the top end.

Making redistribution fairer and more efficient

Efforts to promote labour market inclusiveness by strengthening the importance of redistribution policies should bear in mind the possible adverse effects on work incentives, particularly among low-income families. Indeed, the design of such policies is key for their effectiveness in reducing inequality and overcoming short-term trade-offs between inequality, on the one hand, and employment and economic growth, on the other. Taking a long-term perspective can help overcome such short-term trade-offs since this requires taking account not only of the impact of taxes and transfers on financial hardship at a given time, but also of the extent to which they help or hinder worker careers. Consequently, the impacts of well-designed redistribution policies on different economic outcomes are likely to be more aligned in the long-term.

Government transfers are needed to address financial hardship

At the lower end of the income spectrum, government transfers have an important role to play in lifting low-income households from financial hardship, but they need to be sufficiently large to make a significant difference and be paired with measures to re-establish self-sufficiency and prevent long-term benefit dependency. In most advanced economies, social transfers consist of multiple components, including: i) primary out-of-work benefits (e.g. unemployment insurance benefits) for those who meet certain minimum contribution requirements; ii) secondary out-of-work benefits (e.g. unemployment assistance or minimum-income benefits such as social and housing assistance) that are available to all non-employed persons subject to a means test; and iii) means-tested in-work benefits (e.g. partial unemployment insurance or social assistance for individuals who are working) for those working but with low incomes.

High-coverage unemployment insurance benefits tend to play an important role in absorbing the financial costs of unemployment as well as for the importance of redistribution. If the risk of unemployment were to be evenly distributed across the workforce and unemployment spells relatively short-lived, the role of unemployment risk and unemployment insurance for life-time inequality would be limited. However, in practice, the risk of unemployment tends to be highly concentrated among workers with low-paid jobs and, hence, tends to contribute significantly to life-time inequality (OECD, 2015[8]). In countries, where informal employment is pervasive or eligibility criteria for receiving benefits are very tight, unemployment-insurance benefits tend to be limited to workers with relatively well-paid stable jobs and may even be regressive. Ensuring high coverage of unemployment benefits therefore is not only important for making labour markets more secure (Chapter 9), but also for making them more inclusive, provided that systems are designed to preserve work incentives through the enforcement of a “mutual-obligations” framework.

Last-resort social safety nets play a crucial role for preventing financial hardship, especially for those who are not eligible for unemployment insurance or have exhausted their benefit entitlements. Such benefits are particularly important for out-of-work youth or persons with a history of non-standard work, since they often do not meet minimum contribution requirements for unemployment insurance (OECD, 2016[58]). They are also key in many countries for alleviating the social costs of economic downturns. Due to the decline in hiring during such periods, the typical duration of unemployment tends to increase, increasing the number of unemployed persons that exhaust their unemployment benefit entitlements (OECD, 2011[59]). One concern with last-resort benefits is that, even when combined with housing benefits, they are often not sufficiently generous to lift recipient households out of poverty. Moreover, social assistance benefits are typically associated with considerable duration dependence, i.e. the extent to which benefit receipt in one period increases the probability of benefit receipt in the future. This “state dependence” mainly reflects persistence in the structural drivers of low income (e.g. being a lone parent, having weak socio-economic skills) rather than the role of benefit receipt on work incentives (Immervoll, Jenkins and Königs, 2015[60]).

In some countries, it may be possible to enhance the efficiency of social assistance by making greater use of in-work benefits. This encourages people to take up paid work and gives additional income support to poor working households.25 The effectiveness of in-work benefits in reducing in-work poverty depends importantly on the design and the context in which they are used (Immervoll and Pearson, 2009[61]). For example, they tend to be more effective when they are simple, transparent and well understood by potential recipients. This is more likely when they are permanent rather than temporary or when they are means-tested on family incomes rather than individual earnings, given the importance of household composition for the risk of poverty. Moreover, they tend to be more effective in reducing in-work poverty when combined with a binding minimum wage at a moderate level – by law or collective agreement – since this reduces the extent to which in-work benefits can be appropriated by employers (Chapter 8). However, in-work benefits work less well in the context of compressed wage structures in the bottom of the distribution since this prevents a proper targeting, making them either ineffective or very expensive.

The design of tax systems should ensure that everyone contributes fairly

In addition to government transfers, there is also scope for strengthening the role of taxation, so that everyone contributes a fair share, including those at the top end of the income distribution. More specifically, tax systems can be made more efficient and more inclusive by broadening the tax base, increasing their fiscal progressivity, while taking account of the impact of possible reform options on incentives for working, effort at work and skills development (Brys et al., 2016[62]).

A broadening of the tax base levels the playing field and allows charging lower tax rates. The tax base can be broadened by improving tax compliance and removing or capping tax expenditures that are not well targeted at redistributive goals such as tax exemptions that disproportionately benefit high income groups related to items such as mortgage interests, pension savings or capital gains from secondary residences. Moreover, and as discussed in Chapter 8, when the link between social contributions and expenditures is not strong, levying contributions through progressive personal incomes taxes or other taxes that do not bear exclusively on labour can help to reduce the cost of labour, particularly for low-skilled workers, and increase their employment.

The overall progressivity of the tax system could also be strengthened through a better alignment of capital and labour income for tax purposes and reassessing the role of property and inheritance taxes. The presence of significant tax differentials across income and asset classes distorts compensation, investment and saving decisions, with adverse implications for efficiency and equity (Aghion et al., 2017[63]). Progressive personal income rates do not have to be limited to labour income but can also be applied to capital income, although possibly at a lower and less progressive rate.26 Taxes on immovable property, such as real estate taxes, can promote both efficiency and equity since poor households are less likely to own property (Akgun, Cournède and Fournier, 2017[64]). Reforming inheritance taxes, by raising rates, reducing exemptions and fighting avoidance, can also contribute to greater overall tax progressivity, labour market inclusiveness as well as intergenerational mobility.27


Countries differ importantly in the depth and persistence of inequalities in the labour market. This variation suggests that there is nothing inevitable about deep and persistent inequalities and policy has an important role to play in limiting excessive inequalities in the labour market while fostering equality of opportunities.

Tackling deep and persistent inequalities in the labour market requires a comprehensive policy strategy that simultaneously seeks to promote: i) equal opportunities to avoid that socio-economic background determines success in the labour market; ii) access to quality jobs for vulnerable workers by encouraging their participation in adult learning and tackling labour market segmentation; and iii) an appropriate sharing in prosperity through fair and inclusive tax and benefits policies.

The presence of important policy complementarities across these three elements means that the best strategy for promoting labour market inclusiveness is to consider all three at the same time. In large part, this is due to the interdependence between inequality in outcomes and inequality in opportunities. Excessive inequalities in outcomes typically reflect inequalities in opportunities, but also contribute to their persistence due to the role of family income for investments in the education and health of children.

Yet, strategies for labour market inclusiveness may differ in their emphasis across policy priorities. For example, some countries may opt for placing education and adult learning at the centre of their inclusiveness strategy, with only a limited role for redistribution through the taxes and benefits system. Others may choose to place more emphasis on solidarity and redistribution, recognising the limits of a strongly merit-based model in a world where not everyone is born alike and luck remains an important factor for success.

Finally, taking a long-term perspective can help focussing the public debate on inclusiveness and the development of an effective policy strategy. It allows focusing on persistent differences in inequality, which should be the prime focus of redistribution, while abstracting from public insurance issues related to temporary income shocks. Moreover, it can help overcome possible short-term trade-offs between different policy objectives since long-term inequalities do not just take account of wages at a point in time, but also employment and wage growth during the working life.


[63] Aghion, P. et al. (2017), Tax Simplicity and Heterogeneous Learning *, (accessed on 02 November 2018).

[64] Akgun, O., B. Cournède and J. Fournier (2017), The Effects of the Tax Mix on Inequality and Growth, (accessed on 02 November 2018).

[7] Alesina, A. and P. Giuliano (2015), “Culture and Institutions”, Journal of Economic Literature, Vol. 53/4, pp. 898-944,

[75] Autor, D. and S. Houseman (2010), “Do Temporary-Help Jobs Improve Labor Market Outcomes for Low-Skilled Workers? Evidence from “Work First””, American Economic Journal: Applied Economics, Vol. 2/3, pp. 96-128,

[66] Balestra, C. and R. Tonkin (2018), “Inequalities in household wealth across OECD countries: Evidence from the OECD Wealth Distribution Database”, Statistics and Data Working Paper, No. 2018/01, OECD, Paris.

[10] Barth, E. and S. Kerr (2017), “The Dynamics of Gender Earnings Differentials: Evidence from Establishment Data*”,

[40] Bassanini, A. and A. Garnero (2013), “Dismissal protection and worker flows in OECD countries: Evidence from cross-country/cross-industry data”, Labour Economics, Vol. 21,

[50] Berlingieri, G., P. Blanchenay and C. Criscuolo (2017), “The great divergence(s)”, OECD Science, Technology and Industry Policy Papers, No. 39, OECD Publishing, Paris,

[65] Boeri, T., P. Garibaldi and E. Moen (2017), “Inside severance pay”, Journal of Public Economics, Vol. 145, pp. 211-225,

[43] Booth, A., M. Francesconi and J. Frank (2002), “Temporary Jobs: Stepping Stones Or Dead Ends?”, Economic Journal, Vol. 112/480, pp. 189-213, (accessed on 04 November 2018).

[62] Brys, B. et al. (2016), “Tax Design for Inclusive Economic Growth”, OECD Taxation Working Papers, No. 26, OECD Publishing, Paris,

[39] Cahuc, P., O. Charlot and F. Malherbet (2016), “Explaining the spread of temporary jobs and its impact on labor turnover”, International Economic Review, Vol. 57/2, pp. 533-572,

[23] Carnevale, A. and J. Strohl (2010), “How Increasing College Access Is Increasing Inequality, and What to Do About It”, in Kahlenberg, R. (ed.), Rewarding Strivers: Helping Low-Income Students Succeed in College, Century Foundation Books, New York,

[26] Case, A., A. Fertig and C. Paxson (2005), “The lasting impact of childhood health and circumstance”, Journal of Health Economics, Vol. 24/2, pp. 365-389,

[56] Causa, O. and M. Hermansen (2017), “Income redistribution through taxes and transfers across OECD countries”, OECD Economics Department Working Papers, No. 1453, OECD Publishing, Paris,

[57] Causa, O., A. Vindics and O. Akgun (2018), “An empirical investigation on the drivers of income redistribution across OECD countries”, OECD Economics Department Working Papers, No. 1488, OECD Publishing, Paris,

[42] Centeno, M. and Á. Novo (2012), “Excess worker turnover and fixed-term contracts: Causal evidence in a two-tier system”, Labour Economics, Vol. 19/3, pp. 320-328,

[24] Chetty, R. et al. (2017), “Mobility Report Cards: The Role of Colleges in Intergenerational Mobility”, No. 23618, NBER, Cambridge, MA,

[70] Chetty, R. et al. (2015), “The Economic Impact of Tax Expenditures: Evidence from Spatial Variation Across the U.S.”,

[5] Cingano, F. (2014), “Trends in Income Inequality and its Impact on Economic Growth”, OECD Social, Employment and Migration Working Papers, No. 163, OECD Publishing, Paris,

[3] Clark, A., P. Frijters and M. Shields (2008), “Relative Income, Happiness, and Utility: An Explanation for the Easterlin Paradox and Other Puzzles”, Journal of Economic Literature, Vol. 46/1, pp. 95-144,

[12] Corak, M. (2006), “Do Poor Children Become Poor Adults? Lessons from a Cross-Country Comparison of Generational Earnings Mobility”, in Creedy, J. and G. Kalb (eds.), Research on Economic Inequality, Volume 13, Dynamics of Inequality, Emerald Publishing,

[15] Corak, M. and P. Piraino (2011), “The Intergenerational Transmission of Employers”, Journal of Labor Economics, Vol. 29/1, pp. 37-68,

[74] de Graaf-Zijl, M., G. van den Berg and A. Heyma (2011), “Stepping stones for the unemployed: the effect of temporary jobs on the duration until (regular) work”, Journal of Population Economics, Vol. 24/1, pp. 107-139,

[28] Devaux, M. and M. de Looper (2012), “Income-Related Inequalities in Health Service Utilisation in 19 OECD Countries, 2008-2009”, OECD Health Working Papers, No. 58, OECD Publishing, Paris,

[51] Diez, F., D. Leigh and S. Tambunlertchai (2018), “Global Market Power and its Macroeconomic Implications”, IMF Working Paper,

[69] Duncan, G. and A. Sojourner (2013), “Can Intensive Early Childhood Intervention Programs Eliminate Income-Based Cognitive and Achievement Gaps?”, Journal of Human Resources, Vol. 48/4, pp. 945-968,

[16] Filatriau, O., D. Fougère and M. Tô (2013), “Will Sooner Be Better? The Impact of Early Preschool Enrollment on Cognitive and Noncognitive Achievement of Children”, Discussion Paper No. 9480, No. 9480, Centre for Economic Policy Research,

[46] García-Pérez, J., J. Castelló and I. Marinescu (2018), “Can Fixed-Term Contracts Put Low Skilled Youth on a Better Career Path? Evidence from Spain”, Economic Journal, (accessed on 12 June 2018).

[9] Garnero, A., A. Hijzen and S. Martin (2019), “More unequal, but more mobile? Earnings inequality and mobility in OECD countries”, Labour Economics, Vol. 56, pp. 26-35,

[53] Goldschmidt, D. and J. Schmieder (2017), “The Rise of Domestic Outsourcing and the Evolution of the German Wage Structure”, The Quarterly Journal of Economics, Vol. 132/3, pp. 1165–1217,

[6] Gould, E. and A. Hijzen (2016), “WP/16/176 Growing Apart, Losing Trust? The Impact of Inequality on Social Capital Growing Apart, Losing Trust? The Impact of Inequality on Social Capital”, IMF Working Paper 16176, (accessed on 12 June 2018).

[25] Greenberg, D. and M. Shroder (2004), The digest of social experiments, The Urban Institute Press, Washington DC.

[19] Havnes, T. and M. Mogstad (2011), “No Child Left Behind: Subsidized Child Care and Children's Long-Run Outcomes”, American Economic Journal: Economic Policy, Vol. 3/2, pp. 97-129,

[41] Hijzen, A., L. Mondauto and S. Scarpetta (2017), “The impact of employment protection on temporary employment: Evidence from a regression discontinuity design”, Labour Economics, Vol. 46, pp. 64-76,

[44] Ichino, A., F. Mealli and T. Nannicini (2008), “From temporary help jobs to permanent employment: what can we learn from matching estimators and their sensitivity?”, Journal of Applied Econometrics, Vol. 23/3, pp. 305-327,

[60] Immervoll, H., S. Jenkins and S. Königs (2015), “Are Recipients of Social Assistance 'Benefit Dependent'? Concepts, Measurement and Results for Selected Countries”, (accessed on 02 July 2018).

[61] Immervoll, H. and M. Pearson (2009), “A Good Time for Making Work Pay? Taking Stock of In-Work Benefits and Related Measures across the OECD”, (accessed on 29 June 2018).

[55] Immervoll, H. and L. Richardson (2011), “Redistribution Policy and Inequality Reduction in OECD Countries: What Has Changed in Two Decades?”, OECD Social, Employment and Migration Working Papers, No. 122, OECD Publishing, Paris,

[45] Jahn, E. and M. Rosholm (2014), “Looking beyond the bridge: The effect of temporary agency employment on labor market outcomes”, European Economic Review, Vol. 65, pp. 108-125,

[38] Kahn (2010), “Employment protection reforms, employment and the incidence of temporary jobs in Europe: 1996-2001”, Labour Economics, Vol. 17/1, pp. 1-15, (accessed on 14 June 2018).

[34] Kis, V. and H. Windisch (2018), “Making skills transparent: Recognising vocational skills acquired through workbased learning”, OECD Education Working Papers, No. 180, OECD Publishing, Paris,

[31] Kowalczyk, K. et al. (2016), “„Arbeitsplatzorientierte Alphabetisierung und Grundbildung Erwachsener“,”, Evaluation des BMBF-Förderschwerpunkts,

[47] Muñoz-Bullón, F. (2004), “Training Provision and regulaton: An Analysis of the Temporary Help industry”, International Journal of Manpower, Vol. 25/7, pp. 656-682.

[14] Narayan, A. et al. (2018), Fair Progress? Economic Mobility across Generations around the World, World Bank Group, Washington DC, (accessed on 05 July 2018).

[32] OECD (2019), Getting Skills Right: Future-ready adult learning systems, OECD Publishing, Paris.

[2] OECD (2018), A Broken Social Elevator? How to Promote Social Mobility, OECD Publishing, Paris,

[77] OECD (2018), Effective Teacher Policies: Insights from PISA, PISA, OECD Publishing, Paris,

[11] OECD (2018), OECD Employment Outlook 2018, OECD Publishing, Paris,

[33] OECD (2018), Skills Strategy Implementation Guidance for Portugal: Strengthening the Adult-Learning System, OECD Skills Studies, OECD Publishing, Paris,

[1] OECD (2018), The Framework for Policy Action on Inclusive Growth, (accessed on 25 September 2018).

[67] OECD (2018), The Role and Design of Net Wealth Taxes in the OECD, OECD Tax Policy Studies, No. 26, OECD Publishing, Paris,

[35] OECD (2017), Financial Incentives for Steering Education and Training, Getting Skills Right, OECD Publishing, Paris,

[27] OECD (2017), Obesity Update 2017,

[71] OECD (2016), Investing in Youth: Australia, Investing in Youth, OECD Publishing, Paris,

[68] OECD (2016), Making Cities Work for All: Data and Actions for Inclusive Growth, OECD Publishing, Paris,

[20] OECD (2016), OECD Employment Outlook 2016, OECD Publishing, Paris,

[58] OECD (2016), “Society at a Glance 2016 OECD SOCial inDiCatOrS a spotlight on youth”, (accessed on 02 July 2018).

[13] OECD (2016), Society at a Glance 2016: OECD Social Indicators, OECD Publishing, Paris,

[52] OECD (2016), The Future of Productivity,

[29] OECD (2015), Health at a Glance 2015: OECD Indicators., OECD Publishing, Paris,

[36] OECD (2015), In It Together: Why Less Inequality Benefits All, OECD Publishing, Paris,

[8] OECD (2015), “The quality of working lives: Earnings mobility, labour market risk and long-term inequality”, in OECD Employment Outlook 2015, OECD Publishing, Paris,

[4] OECD (2014), “How good is your job? Measuring and assessing job quality”, in OECD Employment Outlook 2014, OECD Publishing, Paris,

[76] OECD (2014), Non-regular employment, job security and the labour market divide, (accessed on 04 November 2018).

[48] OECD (2013), The 2012 Labour Market Reform in Spain: A Preliminary Assessment, (accessed on 04 November 2018).

[22] OECD (2013), The OECD Action Plan for Youth: Giving youth a better start in the labour market,

[21] OECD (2012), Equity and Quality in Education: Supporting Disadvantaged Students and Schools, OECD Publishing, Paris,

[73] OECD (2011), Divided We Stand: Why Inequality Keeps Rising, OECD Publishing, Paris,

[59] OECD (2011), OECD Employment Outlook 2011 Income Support for the Unemployed: How Well Has the Safety-Net Held Up During the "Great Recession"?, (accessed on 02 November 2018).

[72] OECD (2009), “Is Work the Best Antidote to Poverty?”, Employment Outlook, (accessed on 29 June 2018).

[18] Ruhm, C. and J. Waldfogel (2012), “Long-term effects of early childhood care and education”, Nordic Economic Policy Review, Vol. 1/2012, pp. 23-51,

[49] Song, J. et al. (forthcoming), “Firming Up Inequality”, The Quarterly Journal of Economics, (accessed on 04 November 2018).

[17] Van Huizen, T. and J. Plantenga (2015), “Universal Child Care and Children's Outcomes: A Meta-Analysis of Evidence from Natural Experiments”, U.S.E. Discussion Paper Series, No. 15-13, Utrecht University School of Economics, (accessed on 26 June 2018).

[37] Venn, D. (2011), “Earnings Volatility and its Consequences for Households”, OECD Social, Employment and Migration Working Papers, No. 125, OECD Publishing, Paris,

[54] Weil, D. (2004), The fissured workplace : why work became so bad for so many and what can be done to improve it.

[30] Windisch, H. (2015), “Adults with low literacy and numeracy skills: A literature review on policy intervention”, OECD Education Working Papers, No. 123, OECD Publishing, Paris,


← 1. Alternatively, high inequality, low mobility and weak economic growth may be reflections of deeper underlying problems related to, for example, labour market segmentation, informality or corruption.

← 2. These figures give the earnings persistence from fathers to sons. Earnings elasticities for daughters are more difficult to estimate because women have lower employment rates. Daughters’ decisions to participate in the labour market, moreover, depend on factors that also determine earnings, such as age or educational attainment.

← 3. In addition to formal early childhood education and care, home education programmes can help to improve parenting skills and children’s socio-emotional skills. Effective interventions include support for maternal health during the perinatal period and parenting support programmes targeted at high-risk groups.

← 4. Indeed, an impact evaluation projected that – if systematically implemented – intensive early childhood education programmes like the IHDP could essentially eliminate income-based IQ gaps among three-year-olds (Duncan and Sojourner, 2013[78]).

← 5. OECD evidence based on PISA further suggests that school autonomy in relation to the recruitment of teaching staff can be help to improve the quality of teachers, notably in disadvantaged areas (OECD, 2018[84]).

← 6. The Australian National Youth Mental Health Foundation headspace, for instance, operates centres in which young people can confidentially seek mental health support and treatment outside their immediate social and educational environment (OECD, 2016[80]).

← 7. Evidence for the United States further suggests that both the level of social expenditures and the progressivity of taxation are correlated with greater levels of intergenerational mobility across localities (Chetty et al., 2015[79]).

← 8. For instance, the system of parallel waiting lists in Rotterdam permits oversubscribed schools to give preference to children from a minority background (OECD, 2016[77]). Some countries, such as Chile and the Netherlands, moreover, offer financial incentives to schools to enrol disadvantaged students in order to offset the costs of additional teaching and support staff.

← 9. These are not independent issues. Investing in worker skills is more challenging in an environment in which work is organised based on fixed-term contractual arrangements and firms are poorly managed. Similarly, incentives for providing stable contractual arrangements may be weaker when worker skills are poor and firm productivity is low.

← 10. High and broad participation are not the only features of an effective adult learning system. Other important aspects include the use of good quality labour market information on skill needs, the responsiveness to change, a sound quality assurance system and mechanisms for the recognition of prior learning. These aspects are discussed in Chapter 14.

← 11. While initially targeted at younger persons, there has been an increasing focus on adult learners since 2017.

← 12. In 2017, almost 30 000 adults enrolled in recognition processes and over 10 000 gained recognition.

← 13. In a somewhat similar manner, Austria’s Outplacement Labour Foundations and Sweden’s Job Security Councils provide a range of services targeted at displaced workers to foster re-employment, including in the form of career guidance and training services.

← 14. Consequently, in recent years the tendency has been to reverse these partial labour market reforms by reducing differences in regulatory protection across contract types (see Box 7.3 in Chapter 7).

← 15. These studies ask whether from the perspective of an unemployed person accepting a temporary job offer increases the chance of obtaining a stable job or whether it is better to stay unemployed, keep searching for a permanent job and avoid the risk of getting locked in a “temporary work trap”. Note that not all studies find evidence of stepping-stone effects. Zijl et al. (2011[83]) do not find evidence for this except in the case of immigrants, while Autor and Housemann (2010[84]) find that temporary-agency work harms future employment and earnings outcomes.

← 16. Another option would be to increase social security contributions for the use of temporary contracts as in, for example, France. This would go in the direction of experience-rating unemployment benefits to the extent that temporary work is associated with higher unemployment.

← 17. This could also make it more difficult to implement such reforms since it would impose significant extra costs on public budgets and requires having the capacity to administer such systems effectively. This may be particularly an issue in emerging economies (see Chapter 16).

← 18. Tenure-dependent employment protection can be justified on efficiency grounds in the presence of job-specific investments by workers (Boeri, Garibaldi and Moen, 2017[74]).

← 19. There is also a risk that it provides too little employment flexibility when a prolonged period of relative stability is followed by a severe economic downturn (OECD, 2014[49]).

← 20. Moreover, redistribution is measured in a given year and not over the life-course. This means that the measured extent of redistribution reflects the role of taxes and benefits for stabilising incomes over time (insurance) as well as that for redistribution in terms of life-time incomes.

← 21. This does not take account of in-kind social transfers. While the primary objective of in-kind social transfers is to provide access to good-quality education and healthcare, they are likely to have significant implications for redistribution given their size and declining importance along the income distribution (OECD, 2011[82]).

← 22. However, the relative importance of size and targeting depends crucially on the measure on inequality used, with the role of targeting being more important the more weight is placed on the bottom of the distribution.

← 23. Consumption and environmental taxes and excise duties are not taken into account for the purposes of this decomposition.

← 24. This is all the more notable since in a context of rising market inequality measured redistribution typically increases, everything else equal.

← 25. Countries differ substantially in the relative importance of in-work relative to out-of-work benefits, with many English-speaking countries placing more emphasis on in-work benefits and many Southern European countries on jobless households (OECD, 2009[81]).

← 26. A more equal treatment of capital and labour income also would reduce the role of labour market status (e.g. self-employed versus dependent employee) for tax purposes and hence reduce incentives for employers to replace dependent employees with independent contractors.

← 27. Wealth is distributed much more unequally than income (OECD, 2015[37]; Balestra and Tonkin, 2018[75]). Policies that limit wealth concentration and help low-income households accumulate wealth can therefore promote inclusiveness and contribute to social mobility across generations. However, revenues from taxes on wealth transfers have been very low and declining, from 1.1% of total tax revenues in 1965 to 0.4% today across the OECD on average (OECD, 2018[76]).

End of the section – Back to iLibrary publication page