On average across the OECD, full-time full-year workers without upper secondary attainment earn 31% less than the overall average for all full-time full-year workers. For those with vocational upper secondary or post-secondary non-tertiary attainment the figure is 17% less while those with general upper secondary attainment earn 13% less.
Using the earnings of adults with upper secondary attainment as a benchmark, across OECD countries, adults with short-cycle tertiary attainment enjoy an earnings advantage of 18% more on average. This earnings advantage reaches 39% among those with a bachelor’s degree and 80% among those with a master’s, doctoral or equivalent qualification.
Based on the OECD Survey of Adult Skills (PIAAC), 41% of adults with medium literacy, numeracy and adaptive problem-solving proficiency levels are in the top 40% of earners across the OECD. The probability of being among the top earners rises to 65%, of adults with high skills in all domains, while it falls to 17% among those with low skills in all domains.
Education at a Glance 2026
Context
Higher levels of educational attainment are closely linked to stronger employment prospects (see Chapter A3) and higher earnings (OECD, 2025[1]). The prospect of greater financial returns, together with wider social benefits, encourages individuals to invest in education and training throughout their lives.
However, the earnings premium associated with higher education is far from uniform. Within countries, it differs according to factors such as age, gender, programme orientation and field of study. Labour-market participation patterns also matter; individuals working part time generally earn less overall, and often less per hour worked, than those employed full time. Similarly, those who have gained more experience typically earn more. Despite rising levels of educational attainment, gender pay gaps continue to persist across all attainment levels and programme types.
More young adults than ever are attaining tertiary qualifications (see Chapter A1), and tertiary education systems continue to expand (OECD, 2024[2]). Labour markets in most countries have largely absorbed this increasing supply of highly educated workers, although substantial earnings differences remain, depending on the field of study. These differences may reflect differing demand for particular skills across sectors, as well as wider structural and cultural influences. As economies and labour markets evolve, education systems face growing pressure to ensure that they are equipping their graduates with the competencies needed not just by labour markets, but by society more widely.
Earning inequalities are also shaped by broader economic and institutional contexts. In countries where the share of tertiary-educated adults remains comparatively small, higher earnings may be more concentrated among this group, contributing to wider income disparities and raising concerns about social mobility. Wage outcomes are also shaped by the interaction between the supply and demand for skills, minimum wage legislation, labour-market regulations, and institutional features such as trade union coverage, collective bargaining systems and the overall quality of working conditions (ILO, 2024[3]).
25-64 year-olds; full-time, full-year workers; compared to average earnings of workers across all attainment levels
Note: Upper secondary and post-secondary non-tertiary combined are plotted when no breakdown into general and vocational programmes was available.
1. Year of reference differs from 2024.
2. Includes part-time and part-year workers.
For data, see Table A4.1 and Table A.A4.5. in Sources, Methodology and Technical Notes. The data for this figure can be accessed via https://stat.link/kxu07h.
Other findings
Young women (25-34 year-olds) earn on average 15-19% less than their male peers, depending on the level of educational attainment, while 45-54 year-old women earn between 20% and 23% less.
On average across OECD countries, 29% of workers with below upper secondary attainment earn at or below half the median earnings, compared to 18% of those with upper secondary or post-secondary non-tertiary attainment and just 10% of tertiary-educated workers.
The impact of social and emotional skills on wage levels is more modest than traditional factors. Personality traits explain a smaller share of wage variation (1.5%) compared to educational attainment (8.8%) and literacy skills (4.3%).
Assertiveness (a facet of extraversion) is the strongest economic predictor for wages among the social and emotional skills. The PIAAC survey reports a correlation of 3.2% after adjusting for years of education and literacy skills.
Note
The analysis uses three different baselines for comparing relative earnings: 1) all workers; 2) workers with upper secondary attainment; and 3) male workers. In all cases, given the focus on relative earnings, any increase or decrease in the results could reflect a change in the interest group (numerator) or in the baseline group (denominator). Readers are advised to consider actual earnings in Tables A.A4.4 and A.A4.5 from Education at a Glance 2026 Sources, Methodologies and Technical Notes when interpreting relative earnings ([link to be added]).
Due to the difference in survey methods used to gather data from countries, the analysis of relative earnings is based on full-time full-year workers to ensure better comparability across countries. Refer to Education at a Glance 2026 Sources, Methodologies and Technical Notes ([link to be added]) for more information on the survey methods. Data on relative earnings for all workers (full- and part-time) are available for consultation online (http://data-explorer.oecd.org/s/4s).
Earnings relative to all workers
Higher levels of educational attainment are generally associated with higher earnings. This section compares the earnings of workers who have not attained tertiary education, relative to the whole adult workforce, regardless of educational attainment. This means that the average is partly shaped by the comparatively high earnings of tertiary-educated workers, who represent a growing share of the workforce across OECD countries. As a result, workers with lower levels of educational attainment may appear further behind than when benchmarked against those with upper secondary attainment, or when considering median earnings (see below).
The analysis is restricted to full-time full-year workers aged 25-64 to improve comparability across countries and educational groups by limiting the influence of differences in working hours or labour-market attachment. Including part-time and part-year workers would lower average earnings in many countries with increased representation of groups that are more likely to work reduced hours or experience intermittent employment, such as younger adults, women or adults with lower levels of educational attainment.
The skills, knowledge and competencies acquired through upper secondary education provide an essential foundation for participation in the labour market and help ensure that individuals attain minimum levels of literacy and numeracy that are required in most occupations (OECD, 2024[2]). Without these core skills, individuals can struggle to find a job or be more likely to be confined to lower-paying jobs.
On average across the OECD, full-time full-year workers who have not completed upper secondary education earn 31% less than the average earnings across all levels of educational attainment. The earnings disadvantage ranges from 18% in Denmark, Finland and Romania to over 50% in Colombia and South Africa. On average, workers whose highest attainment is vocational upper secondary or post-secondary non-tertiary education earn 17% less than the average earnings, with the gap ranging from 1% more than the average earnings in Mexico to 34% less in Luxembourg. Workers who have a general upper secondary or post-secondary non-tertiary qualification as their highest educational level earn 13% less than the average earnings, ranging from 1% in Finland to 25% in Israel (Figure A4.1).
Earnings relative to workers with upper secondary attainment
Tertiary education plays an important role in promoting upward economic mobility, as a key enabling factor for strong labour-market outcomes. The advanced knowledge and specialised skills developed through tertiary programmes are in high demand by employers and lead to low rates of unemployment (see Chapter A3). Holding a tertiary qualification also opens up a wider range of occupations, including professional and managerial positions that are generally associated with higher levels of pay. In addition, tertiary institutions provide opportunities for students to build networks with peers, academic staff and industry professionals, which may contribute to stronger employment prospects and higher earnings over time.
The average earnings of tertiary-educated adults working full time and throughout the year are considerably higher than those of workers whose highest level of attainment is upper secondary education. On average across OECD countries, completing a tertiary qualification is associated with an earnings premium of 54%, although the magnitude of this advantage varies substantially across countries. The earnings premium linked to tertiary attainment is 25% or less in Denmark, Norway and Sweden, while it exceeds 100% in Chile and Colombia among OECD countries, as well as in Brazil and South Africa among OECD partner countries (Table A4.1).
The earnings advantage of tertiary education generally increases with higher attainment. In most OECD and partner countries, adults working full time and throughout the year who hold a master’s, doctoral or equivalent qualification earn more on average than those with a bachelor’s degree, who in turn earn more than individuals with a short-cycle tertiary qualification. Across OECD countries, adults with a master’s, doctoral or equivalent qualification earn on average 80% more than those with upper secondary education as their highest qualification. This advantage drops to 39% among those with a bachelor’s degree and to 18% among those with short-cycle tertiary attainment (Table A4.1).
Earnings advantages by educational attainment tend to increase among older workers. On average across OECD countries, tertiary-educated 25–34 year-olds earn 38% more than their peers with upper secondary attainment, while 45-54 year-olds earn 65% more. Within the levels of tertiary attainment, the earnings advantage of a short-cycle tertiary qualification is 11% among 25-34 year-olds and 18% among 45-54 year-olds on average across OECD countries. Among those with bachelor’s attainment, 25-34 year-olds earn 31% more and 45-54 year-olds earn 47% more than their peers with an upper secondary attainment. For master’s or higher attainment, the advantage is 52% for the younger age group and 92% more for the older one (Table A4.1). Comparing countries, the earnings advantage for tertiary-educated 25-34 year-olds ranges from 20% or less in Italy, Korea and the Nordic countries to 100% or more in Brazil, Colombia and South Africa, while for 45-54 year-olds it ranges from 30% or less in Denmark, Norway and the United Kingdom to around 200% more in Colombia and South Africa (Table A4.1).
Investing in education has a significant impact on earning potential and employment outcomes (see Chapter A3), but it is not the only factor affecting success in the labour market; employers and the labour market can also reward specific or fundamental skills or other factors. Possessing certain personality traits (or social and emotional skills) is also positively associated with labour-market rewards (Box A4.1).
The recent publication from the 2023 Survey of Adult Skills (Cycle 2) (OECD, 2025[4]) represents a paradigm shift in the OECD's approach to adult skills. For decades, the Programme for the International Assessment of Adult Competencies (PIAAC) focused primarily on the key information-processing skills (namely, literacy, numeracy and problem solving) deemed essential for workforce participation. However, as the global landscape is being radically reshaped by the rapid advancement of artificial intelligence (AI), population ageing and international migration, the definition of a skilled adult has necessarily evolved. While cognitive skills such as literacy and numeracy remain essential, social and emotional skills — including the ability to collaborate, adapt, manage emotions and persevere — are increasingly recognised as important drivers of success and well-being throughout life. In an era where AI surpasses human performance in routine analytical tasks, social and emotional skills (SES) constitute a distinct human advantage in performing tasks that cannot be performed by machines. The survey findings underscore how social and emotional skills complement cognitive ones and independently contribute to adults’ success at work and beyond.
Theoretical framework: the Big Five model and its measurement
The 2023 survey adopted the Big Five framework (or Five Factor model) (OECD, 2025[4]), the most robust taxonomy in psychological research. This model is descriptive rather than normative; it does not categorise traits as inherently good or bad but organises human behaviour into five higher-order domains (Figure A4.2).
Source: OECD (2025[4]), Skills that Matter for Success and Well-being in Adulthood, https://doi.org/10.1787/6e318286-en, Chapter 1.
Social and emotional skills as drivers of educational attainment
The Survey of Adult Skills provides empirical evidence that social and emotional skills are powerful predictors of how long individuals stay in formal education (proxy for attainment). Openness to experience and emotional stability emerge as the most consistent predictors of educational attainment across nearly all participating OECD countries. On average, a one-standard-deviation increase in openness correlates to an additional 0.4 years of education, ranging from 0.2 years in Canada to 0.8 years in Italy. However, analysis of individual facets reveals important nuances. Intellectual curiosity and aesthetic sensitivity are much stronger predictors of education length than creative imagination, which actually showed a negative association in Canada, Czechia and Norway. This suggests that the desire for abstract ideas and self-exploration is more important for navigating formal academic structures than pure divergent thinking.
The role of conscientiousness is markedly context dependent. It shows a positive link with attainment in 16 countries, but exhibits a negative correlation in Austria, suggesting that national education systems reward different behavioural profiles. Furthermore, the association between openness and education is significantly stronger for older adults and those from less-educated families, indicating that these traits may serve as critical tools for individuals navigating structural barriers and unequal educational opportunities.
Labour market outcomes: employment status and economic returns
Extraversion and emotional stability are the traits most significantly associated with the likelihood of being employed in modern labour markets. Individuals who are socially proactive and resilient to stress navigate recruitment and workplace challenges more effectively. However, although social and emotional skills are essential for securing work, their direct impact on wage levels is more modest than traditional factors. Personality traits explain only 1.5% of wage variation, compared to 8.8% for educational attainment and 4.3% for literacy proficiency. This reinforces the primary importance of formal qualifications in securing high earnings.
Employed adults aged 25-65, in per cent
Note: Aggregated results across the OECD countries that used the BFI-2-S measure [see OECD (2025[4]), Chapter 1]. Estimates are obtained by controlling for gender, age, parental education, immigrant background, whether one lives with a partner and whether one has children – and then, in a second step, for years of education attained and literacy proficiency. Wages are gross hourly earnings for employed and self-employed individuals, including bonuses, in PPP-adjusted 2022 USD. (r) denotes reverse-coded sub-domains, where a positive score corresponds to a low tendency for anxiety, depression or emotional volatility (and thus higher emotional stability). Darker colours denote differences that are statistically significant at the 5% level. Results for Korea are not presented.
Source: OECD (2025[4]), Skills that Matter for Success and Well-being in Adulthood, https://doi.org/10.1787/6e318286-en, Figure 3.6.
Disaggregating the domains reveals that, among the Big Five facets, assertiveness (a facet of extraversion) shows the largest positive association with wages (Figure A4.3). A one-standard-deviation increase in assertiveness is associated with 4.8% higher hourly wages before accounting for years of education and literacy proficiency, and with 3.2% higher hourly wages after additionally accounting for them. Accounting for these factors is crucial; it demonstrates that assertiveness is economically rewarded independently of someone's educational attainment or literacy proficiency. Factors related to emotional stability are also significant: a lower tendency for depression is consistently linked to higher earnings, likely due to its impact on productivity and professional stamina.
A pivotal finding of the Skills that Matter report is the compensatory role of social and emotional skills, particularly for vulnerable populations. Emotional stability, extraversion and conscientiousness are markedly more important for employment among individuals with low literacy proficiency. This suggests that strong social and emotional competencies act as a vital buffer for adults who struggle with complex information processing. In service-oriented, care-based or manual work environments, the "human advantage", i.e. the ability to regulate one's emotions, communicate effectively with colleagues and persist with tasks, can mitigate the labour-market disadvantages typically associated with lower cognitive scores. For these individuals, social and emotional skills are not just soft skills but survival mechanisms that provide a pathway to employment stability that technical training alone might not achieve.
Source: OECD (2025[4]), Skills that Matter for Success and Well-being in Adulthood, https://doi.org/10.1787/6e318286-en,
Gender disparities in earnings
Although increasing educational attainment narrows gender differences in employment rates (see Chapter A3), the gender gap in earnings does not vary much across educational attainment levels. On average across OECD countries, tertiary-educated women and those with vocational upper secondary or post-secondary non-tertiary attainment working full time and for the full year earn 22% less than their male peers, while women with general attainment at the same level and those with below upper secondary attainment earn 20% less (Figure A4.4 and Table A4.3). As women are more likely to work part time or only for part of the year than men, the gender differences in earnings are even wider among all workers than among full-time full-year workers (OECD, 2025[1]).
For all education levels, the gender gap in earnings is wider among 45-54 year-olds than it is for younger age groups. Young women (25-34 year-olds) in full-time full-year employment earn between 15% and 19% less than their male peers, depending on their level of educational attainment, while 45-54 year-old women earn between 20% and 23% less. On average, the gender gap is between 2 and 8 percentage points wider for 45-54 year-old women than for 25-34 year-old ones. However, differences across educational attainment levels vary by country and are relatively small on average (Table A4.3).
There is no single explanation for the persistence of gender pay gaps, despite women now surpassing men in educational attainment levels in many countries (see Chapter A1). These disparities reflect a combination of interconnected factors, including occupational segregation, sectors, firms and job ladders, and differences within firms in tasks, responsibilities, promotion and wage setting (D., Blau and Kahn, 2017[5]; OECD, 2022[6]). Women are generally less likely than men to receive promotions or substantial wage increases when changing employers. Childbirth and caregiving responsibilities may reinforce this pattern by affecting labour-market attachment, hours worked, job mobility and wage growth (Rabaté et al., 2021[7]; Goldin et al., 2017[8]; Kleven, Landais and Søgaard, 2019[9]). Women are therefore more likely to seek careers offering greater flexibility and work-life balance in order to accommodate family responsibilities, which may reduce their earnings relative to men with similar levels of educational attainment. Cross-country differences in childcare policies, parental leave, working-time riles, second-earner incentives, collective bargaining, pay transparency and equal-pay enforcement can therefore affect how strongly family responsibilities and workplace inequalities translate into earning gaps (OECD, 2026[10]; OECD, 2022[6]). Consequently, although progress has been made towards greater gender pay equality, substantial disparities persist, with women in many countries continuing to earn less than men for comparable work due to enduring structural inequalities and discrimination (ILO, 2024[3])
Percentage difference between women’s and men’s earnings among 25-64 year-old full-time full-year workers
1. Year of reference differs from 2024.
2. Low reliability of the value for workers with below upper secondary educational attainment.
For data, see Table A4.3. The data for this figure can be accessed via https://stat.link/kxu07h.
Distribution of earnings among workers
Relative earnings by level of educational attainment not only indicate the extent to which labour markets reward additional education but also reflect broader patterns of income distribution and social inequality (OECD, 2025[1]). Higher relative earnings among tertiary-educated adults may provide strong incentives for individuals to pursue further education, but they can also point to wider wage dispersion and income inequality, particularly where earnings among less-educated workers remain comparatively low or stagnant. Although education has the potential to promote greater social mobility and reduce inequality, disparities in access to education and in educational outcomes may also contribute to the persistence of existing socio-economic inequalities (UNESCO, 2020[11]).
This trade-off is evident in countries where high earnings premiums coexist with greater income inequality. For example, Chile, Colombia and Costa Rica are among the OECD countries with the highest earnings premiums for tertiary-educated adults, as well as the highest levels of wage dispersion. Conversely, in countries with more compressed wage structures, such as the Nordic countries, the earnings advantage of tertiary education is smaller, but overall income inequality is also lower (Table A4.1).
25-64 year-old workers, in per cent
1. Year of reference differs from 2024.
For data, see Table A4.2. The data for this figure can be accessed via https://stat.link/kxu07h. .
A key indicator of education-related labour-market inequality is the proportion of individuals at each attainment level who earn significantly more or less than the median earnings of all full- and part-time workers. On average across OECD countries, 29% of workers with below upper secondary attainment earn at or below half the median wage, compared to 18% of those with upper secondary or post-secondary non-tertiary education and just 10% of tertiary-educated workers (Figure A4.5). Conversely, only 26% of workers with below upper secondary attainment earn more than the median, compared to 42% of those with upper secondary or post-secondary non-tertiary attainment and 68% of tertiary-educated workers (Table A4.2). However, these statistics should be treated with some caution, as they include part-time and part-year workers, who are mostly concentrated in the low earners’ categories.
These disparities are even more pronounced at the top of the earnings distribution. On average across OECD countries, just 3% of workers with below upper secondary attainment earn more than twice the median wage, compared to 6% of those with upper secondary or post-secondary non-tertiary attainment and 22% of tertiary-educated workers. Among OECD and partner countries, more than 40% of tertiary-educated 25-64 year-olds earn more than twice the median in Brazil, Colombia, Costa Rica and South Africa (Table A4.2).
Income distribution can also be analysed in relation to skill levels, to analyse the extent to which low skills are preventing adults from accessing high-paying jobs (Box A4.2).
Traditional single-domain assessments offer only a partial view of human capital because literacy, numeracy and problem-solving skills are deeply interconnected and tend to cluster as strengths or weaknesses. Data from the OECD Survey of Adult Skills (PIAAC) confirm that multidimensional skill profiles are a powerful predictor of economic success independent of formal education, acting as a sorting mechanism within labour markets. Analysing outcomes through income quintiles illustrates this compound effect: adults with consistently high proficiency across all three domains benefit from a substantial earnings premium and are heavily concentrated among higher earners, while those with multiple weaknesses face a severe earnings penalty and significant barriers to upward mobility.
Analysis of PIAAC data across OECD countries shows a systematic relationship between combined skill profiles and the probability of being a high earner, defined here as belonging to the top two earnings quintiles or top 40% of earners. Approximately 64% of high-skilled individuals (those with high skills in at least two skills domains) are considered high earners, compared to 42% of medium-skilled individuals (those with medium skills in at least two domains) and 20% of low-skilled ones (those with low skills in at least two domains) (Table A4.4, available online).
Breaking down these groups further concedes some precision, as the greater number of combinations produces many small sub-populations with smaller sample sizes for less robust cross-country comparisons. However, it does allow the effect of having low, medium or high skills in individual domains to be analysed. Taking those adults who have medium proficiency levels in all three domains as a useful benchmark, on average across the OECD, approximately 41% of these individuals are in the top 40% of earners. There is a substantial earnings premium associated with higher proficiency levels: adults with high skills in at least one domain and medium skills in the others have a 58% probability of being among the high earners. Adults with high proficiency levels across all three domains reach a probability of approximately 65% (Figure A4.6).
These findings demonstrate that multiple strengths across literacy, numeracy and adaptive problem solving provide a compound economic advantage. Labour markets reward not only isolated competencies but also the consistency of high-level performance across domains.
Conversely, low proficiency in even a single domain generates a substantial earnings penalty. Adults with low skills in one domain and medium skills in the other two see their probability of belonging to the top earnings groups fall to approximately 31%. This penalty intensifies as weaknesses accumulate: the share of top earners drops to 25% for those with low skills in two domains and medium skills in the third, and to 17% for those with low skills in all domains (Figure A4.6).
The disadvantage is even clearer at the top and bottom of the earnings distribution. On average across OECD countries, approximately 38% of adults with low skills across all domains are concentrated in the lowest earnings quintile, while only around 7% reach the highest quintile. In some countries, including Sweden and Norway, more than 55% of adults with low proficiency levels in all domains are found in the bottom 20% of earners. These patterns illustrate a clear “multiple weakness penalty”: labour-market disadvantage increases sharply when deficiencies accumulate across domains rather than in isolation (Table A4.4, available online).
In per cent of 25-64 year-old earners in the top 4th and 5th quintiles; Survey of Adult Skills (PIAAC)
For data, see Table A4.4 (available online). The data for this figure can be accessed via https://stat.link/kxu07h .
Skills versus educational credentials
Both educational attainment and measured skills are positively associated with labour-market success. However, the evidence suggests that combined skill profiles are often a more effective predictor of high earnings than educational credentials alone. Across the OECD, approximately 65% of adults with high skills across all domains and all levels of educational attainment belong to the top two earnings quintiles, compared to 54% of tertiary-educated adults (across all skill levels). Tertiary-educated adults have more heterogeneous outcomes when considering their skill levels. Among tertiary-educated adults, those with high skills in at least two domains have a probability of around 67% of being top earners, while those with low skills in at least two domains see this probability fall to approximately 32% (Table A4.4, available online).
Adults refer to 25-64 year-olds; young adults refer to 25-34 year-olds.
Earnings include annual money earnings as direct payment for labour services provided, before taxes, plus work-related payments such as annual bonuses, result-related bonuses, extra pay for holidays and sick-leave pay from employer(s). Earnings do not include income from other sources, such as government social transfers, investment income, net increase in value of an owner operated business and any other income not directly related to work.
Educational attainment refers to the highest level of education successfully completed by an individual.
Individuals with zero earnings refer to individuals who have earnings, but the result of their business activities is exactly zero.
Individuals with negative earnings refer to individuals who reported deficits in their business activities.
Levels of education: See the Reader’s Guide at the beginning of this publication for a presentation of all International Standard Classification of Education (ISCED) 2011 levels.
Definitions of Skill domains and Proficiency levels and the distributions of the population related to data from Survey of Adult Skills (PIAAC) are included in Chapter A1.
The analysis of relative earnings of the population with specific educational attainment and of the distribution of earnings does not control for hours worked, although the number of hours worked is likely to influence earnings in general and the distribution in particular. For the definition of full-time earnings, countries were asked whether they had applied a self-designated full-time status or a threshold value for the typical number of hours worked per week.
Earnings data are based on an annual, monthly or weekly reference period, depending on the country. This chapter presents annual data, and earnings data with a reference period shorter than a year are adjusted. Please refer to Table A.A4.1 in Education at a Glance 2026 Sources, Methodologies and Technical Notes, for more information on the adjustment methods ([link to be added]). Data on earnings are before income tax for most countries. Earnings of self-employed people are excluded for many countries and, in general, there is no simple and comparable method to separate earnings from employment and returns to capital invested in a business.
This chapter does not take into consideration the impact of effective income from free government services. Therefore, although incomes could be lower in some countries than in others, the state could be providing both free health care and free schooling, for example. The total average for earnings (men plus women) is not the simple average of the earnings figures for men and women. Instead, it is the average based on earnings of the total population. This overall average weights the average earnings separately for men and women by the share of men and women with different levels of educational attainment.
In the earnings data, individuals with zero and/or negative earnings should be reported as earners. Individuals with negative earnings should also be considered in the calculation of the overall median earnings. However, data on individuals with zero and/or negative earnings are not available for all countries. Individuals with zero earnings are included for Belgium, Brazil, Canada, Germany, Ireland, New Zealand, Norway, Sweden, Switzerland, the Republic of Türkiye and the United States. Individuals with negative earnings are included for Belgium, Canada, Denmark, Italy, New Zealand, Norway, Spain, Sweden and the United States. Refer to the Definitions section for the definition of individuals with zero and negative earnings. Note that the share of both zero and negative earners are very low among full-time full-year workers in countries with available data, and this finding holds true when looking at the breakdown by educational attainment levels. The impact of the inclusion/exclusion of zero and/or negative earners is negligible on the relative earnings and the distribution of earnings.
For more information see the OECD Handbook for Internationally Comparative Education Statistics (OECD, 2018[12]) and Education at a Glance 2026 Sources, Methodologies and Technical Notes ([link to be added]).
Table A4.4 and Figure A4.6 include data on adults’ earnings based on the quintile distribution of Monthly earnings including bonuses for wage and salary earners and self-employed (EARNMTHALLPPPC2, derived variable).
Distribution of social and emotional skills across adults and critical evaluation of the Big Five model
Distribution of skills
Policy design requires a granular understanding of how these skills are distributed. The Survey of Adult Skills (PIAAC) identifies several systemic demographic patterns:
Gender: A near-universal trend shows women reporting higher average levels of agreeableness and conscientiousness, but lower levels of emotional stability than men.
Ageing: Skills evolve over the lifespan. Younger adults exhibit higher extraversion and openness, whereas older adults exhibit higher conscientiousness and agreeableness, reflecting a gradual shift toward social conformity and task-oriented discipline.
Parental education: Adults with tertiary-educated parents report higher openness (likely due to early exposure to diverse ideas) but lower conscientiousness compared to those from lower-educated backgrounds.
Migration: First-generation immigrants typically report higher agreeableness, conscientiousness and openness than native-born populations. This likely reflects the self-selection effect, where individuals who choose to migrate are often those possessing greater resilience and openness to new environments.
Critical evaluation
Descriptive versus normative: The model describes typical behaviour, not ideal behaviour. High scores are not always optimal; for example, extreme conscientiousness can manifest as rigid perfectionism that hinders adaptation.
Self-reporting and social desirability: Responses are subject to the individual's self-perception and the "social desirability bias," where respondents provide answers which they believe the interviewer wants to hear.
Cross-cultural comparability: Cultural norms dictate how assertive or respectful an individual claims to be. This is why comparisons of cross-country averages are statistically invalid.
Lack of causality: The survey is cross-sectional, representing associations at a single point in time. It is impossible to definitively state whether high openness causes one to pursue more education, or if the experience of higher education develops one’s openness. This bidirectional relationship must be considered when designing interventions.
This chapter is based on the data collection on education and earnings by the OECD Network for data development on labour market, economic and social outcomes of education (LSO Network). The data collection takes account of earnings for individuals working full time and for the full year, as well as part time or part of the year, during the reference period. This database contains data on dispersion of earnings from work and on student earnings versus non-student earnings. The source for most countries is national household surveys such as Labour Force Surveys, the European Union Statistics on Income and Living Conditions (EU-SILC), or other dedicated surveys collecting data on earnings. About one-quarter of countries use data from tax or other registers. See Education at a Glance 2026 Sources, Methodologies and Technical Notes, for country-specific notes on national sources ([link to be added]).
Data on proficiency levels and social and emotional skills are based on the Survey of Adult Skills (PIAAC) (2023). PIAAC is the OECD Programme for the International Assessment of Adult Competencies.
References
[5] D., F., L. Blau and Kahn (2017), “The Gender Wage Gap: Extent, Trends, and Explanations”, Journal of Economic Literature, Vol. 55/3, pp. 789-865, https://doi.org/10.1257/jel.20160995.
[8] Goldin, C. et al. (2017), “The Expanding Gender Earnins Gap: Evidence from the LEHD-2000 Census”, American Economic Review, Vol. 107/5, pp. 110-14, https://doi.org/10.1257/aer.p20171065.
[3] ILO (2024), Global Wage Report 2024-25: Is Wage Inequality Decreasing Globally?, International Labour Organization, Geneva, https://doi.org/10.54394/cjqu6666.
[9] Kleven, H., C. Landais and J. Søgaard (2019), “Children and Gender Inequality: Evidence from Denmark”, American Economic Journal: Applied Economics, Vol. 11/4, pp. 181-209, https://doi.org/10.1257/app.20180010.
[10] OECD (2026), Pay Transparency in Progress: Valuing Jobs, Closing Gender Pay Gaps, Gender Equality at Work, OECD Publishing, https://doi.org/10.1787/121f268d-en.
[1] OECD (2025), Education at a Glance 2025: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/1c0d9c79-en.
[4] OECD (2025), Skills that Matter for Success and Well-being in Adulthood: Evidence on Adults’ Social and Emotional Skills from the 2023 Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/6e318286-en.
[2] OECD (2024), Do Adults Have the Skills They Need to Thrive in a Changing World?: Survey of Adult Skills 2023, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/b263dc5d-en.
[6] OECD (2022), Same skills, different pay: Takling gender inequalities at firm level, OECD Publishing, https://doi.org/10.1787/a4d18506-en.
[12] OECD (2018), OECD Handbook for Internationally Comparative Education Statistics 2018: Concepts, Standards, Definitions and Classifications, OECD Publishing, Paris, https://doi.org/10.1787/9789264304444-en.
[7] Rabaté, S. et al. (2021), “The child penalty in the Netherlands and its determinants”, CPB Discussion Paper, No. 424, CPB Netherlands Bureau for Economic Policy Analysis, https://doi.org/10.34932/trkz-qh66.
[11] UNESCO (2020), Global Education Monitoring Report 2020: Inclusion and Education: All Means All, United Nations Educational, Scientific and Cultural Organization, Paris, https://doi.org/10.54676/jjnk6989.
The data for the figures and tables in this chapter can be downloaded via https://stat.link/kxu07h.
To access further data and/or other education indicators, please visit the OECD Data Explorer: http://data-explorer.oecd.org/s/4
Data cut-off for the print publication 17 June 2026. Please note that the Data Explorer contains the most recent data.
a – category not applicable; b – break in series; d – contains data from another column; m – missing data; x – contained in another column (indicated in brackets). For further control codes, see the Reader’s Guide.
For further methodological information, see Education at a Glance 2026: Sources, Methodologies and Technical Notes (https://doi.org/10.1787/dcf64a14-en)
Adults with income from employment (full-time full-year workers); upper secondary attainment for each age group = 100
Note: There are cross-country differences in the inclusion/exclusion of zero and negative earners. Columns showing data on relative earnings for workers with upper secondary attainment and for 45-54 year-olds are available for consultation online.
1. Year of reference differs from 2024: 2025 for Mexico; 2023 for Argentina, Belgium, Bulgaria, Canada, Colombia, Finland, Israel, Italy and the United States; 2022 for France, Ireland and Lithuania.
2. Index 100 refers to the combined levels of upper secondary and post-secondary non-tertiary education (levels 3 and 4 in the ISCED 2011 classification).
3. Includes part-time and part-year workers.
4. Earnings net of income tax for Türkiye and a combination of gross (self-employed) and net (employees) earnings for Argentina.
The data for this Table can be accessed via https://stat.link/kxu07h.
Median earnings from work for 25-64 year-olds with income from employment (full- and part-time workers)
Note: There are cross-country differences in the inclusion/exclusion of zero and negative earners. For a given level of educational attainment, the figures by level of earnings relative to median earnings may not add up to 100% because of missing data. Columns showing data broken down by gender are available for consultation online.
1. Year of reference differs from 2024: 2025 for Mexico; 2023 for Argentina, Bulgaria, Canada, Colombia, Finland, Israel, Italy, South Africa and the United States; 2022 for France, Ireland and Lithuania.
2. Earnings net of income tax for Türkiye and a combination of gross (self-employed) and net (employees) earnings for Argentina.
The data for this Table can be accessed via https://stat.link/kxu07h.
Average earnings of adults with income from employment (full-time full-year workers)
Note: There are cross-country differences in the inclusion/exclusion of zero and negative earners. Columns showing data for other age groups are available for consultation online.
1. Year of reference differs from 2024: 2025 for Mexico; 2023 for Argentina, Belgium, Bulgaria, Canada, Colombia, Finland, Israel, Italy and the United States; 2022 for France, Ireland and Lithuania.
2. Includes part-time and part-year workers.
3. Earnings net of income tax for Türkiye and a combination of gross (self-employed) and net (employees) earnings for Argentina.
The data for this Table can be accessed via https://stat.link/kxu07h,