As digital, environmental, and demographic transformations reshape labour markets, skills influence who benefits from these changes and who falls behind. Lifelong learning therefore becomes a policy priority to counterbalance such inequalities. Beyond narrower objectives like employability and productivity, lifelong learning can contribute to good working conditions, lasting innovation, inclusive societies, and active citizenship. Yet, recent global evidence shows striking gaps. 

This column summarises insights from a new International Labour Organization report which presents original evidence from worker surveys, online vacancy analysis, institutional data, and a meta-study of training evaluations (ILO 2026). It shows that there is still a long way to go for lifelong learning systems to effectively support diverse learning across the life course.

Access to lifelong learning opportunities remains limited and unequal

Despite its importance, participation in structured learning remains limited. Globally, only 16% of workers engaged in structured learning in the last 12 months. Figure 1 shows that average participation rates are similarly low across country-income groups, at around 15% in lower-middle- and upper-middle-income countries and around 17% in low-income and high-income countries. The smoothed distributions also reveal significant variation within each income group. Among low-income countries, participation ranges from 7% in Niger to 24% in Somalia. Among high-income countries, it varies between 10% in the Republic of Korea and 20% in Luxembourg. Even the highest participation rates remain modest, however, in light of major labour market disruptions.

Figure 1 Distribution and means of country-level participation rates in formal and non-formal education and training in the past 12 months, 2023 or latest available year, by country-income group (%)

Notes: The figure shows smoothed distributions (kernel densities) and group means of participation rates, based on data from 123 countries, 2023 or latest available year. Formal learning is recognised by relevant national authorities and leads to qualifications. Informal learning is more flexible, including training in enterprises and structured online learning. 
Source: ILO (2026) based on UN, SDG Indicators Database. 

Important disparities also exist across worker groups. Full-time, permanent workers in formal manufacturing firms have greater access to structured learning. Of these, 51% receive employer-provided training per year. In contrast, new worker-level evidence from Bangladesh, Fiji, and the United Republic of Tanzania indicates that informal workers, workers with lower literacy, and those with low levels of formal qualifications rely predominantly on informal learning. Oftentimes, these workers learn by doing and on their own, without more established learning options or policy support.

Informal learning, although widespread, is weakly connected to formal skills recognition. Many countries lack adequate mechanisms for recognising prior learning and converting work experience into qualifications. Therefore, workers in vulnerable forms of employment face disadvantages in access to training, qualifications, and labour market mobility. 

Labour markets increasingly reward rounded skills profiles

Inequalities in learning are concerning given changing skills requirements. To investigate skills demand, ILO builds on several empirical and methodological studies analysing online vacancy data from eight countries in regions that have been less investigated so far: Brazil, Egypt, Jordan, Morocco, the Russian Federation, South Africa, the United Arab Emirates, and Uruguay. The analysis relies on new taxonomies and natural language processing to elicit various cognitive, socio-emotional, manual, and green skills in texts (Escudero et al. 2024, Adamczyk et al. 2025, Delaporte et al. 2025). 

Across the online labour markets studied, socio-emotional and cognitive skills are in high demand. In addition, Figure 2 illustrates that employers seek rounded skills profiles. The figure reveals which of 15 skills are often jointly required in online vacancies, with thicker lines indicating stronger co-occurrence between skill pairs. Similar patterns emerge across countries. Different socio-emotional skills – shown in shades of purple – are often jointly demanded and frequently required alongside cognitive and technical skills – shown in shades of green. This is consistent with complementarities between technical competencies and non-routine cognitive and interpersonal skills found for the US (Deming and Kahn 2018, Stephany and Teutloff 2023).

From a policy perspective, linking skills demand to labour market outcomes is important to identify the skills workers need to access quality jobs. ILO finds that vacancies requiring more complex and interconnected skills profiles offer higher posted wages and better non-wage characteristics, including opportunities for career development, teamwork, and positive work environments. 

Figure 2 Skill co-occurrence networks in vacancies, by country

Notes: Thicker lines indicate that two skills are more often jointly demanded in online vacancies. Green colours refer to cognitive skills, purple colours to socio-emotional skills, and blue colours to manual skills. The analysis is based on vacancy data from BuscoJobs in Brazil and Uruguay (2010–23), Adzuna for the Russian Federation and South Africa (2016–21), and UN Economic and Social Commission for Western Asia for Egypt, Jordan, and the United Arab Emirates (2020–24). 
Source: ILO (2026). 

The importance of rounded skills profiles indicates that a narrow focus on technical expertise alone is insufficient. The same pattern emerges in the context of the green and digital transitions. 

The ILO report shows that many occupations require environmentally relevant tasks, ranging from science and engineering professionals to plant and machine operators and refuse workers and other elementary workers. Together, these occupations account for 32% of global employment, with higher shares in wealthier countries. Beyond green tasks alone, these jobs combine cognitive and technical, socio-emotional, and manual skills in varying proportions (see also Delaporte et al. 2025, Adamczyk et al. 2026).

Similarly, in digital occupations, technical capabilities are frequently complemented by cognitive and socio-emotional competencies related to problem-solving, project and process management, and communication (Borgonovi et al. 2023). A shift-share analysis using internationally comparable labour force surveys reveals that subnational regions with historically stronger demand for cognitive, managerial, and communication skills experience better employment outcomes following exposure to digital technologies. Comprehensive skills endowments therefore help mediate labour market responses to digital shocks (ILO 2026, Delaporte et al. forthcoming).

Yet, developing multidimensional skill profiles does not necessarily improve labour market outcomes when markets fail to adequately value skills. The growing care economy illustrates this tension. Population ageing will increase the global demand for long-term care workers, with estimated needs rising from 85 million workers in 2023 to 158 million by 2050. Yet, while socio-emotional skills are associated with significant wage premia on average (Deming 2017), this is not true in the care sector. This highlights that skills development needs to be embedded within a broader policy agenda, which fosters employment, adequate pay and good working conditions (Liepmann and Hegewisch 2025, ILO 2026). 

The need for ambitious lifelong learning systems 

Whether workers develop multidimensional skills profiles depends critically on inclusive and effective lifelong learning systems. Conceptually, such systems require strong governance, institutional coordination, adequate financing shaped as a shared responsibility, and social dialogue to develop solutions reflecting all actors’ perspectives. 

Many countries still lack ambitious and cohesive lifelong learning systems. New institutional survey data covering 21 countries from different regions reveal, for example, that adult learning is frequently treated as distinct from basic education. According to UNESCO’s Global Report on Adult Learning and Education data, moreover, 46% of countries allocate less than 1% of their public education budgets to adult learning. This share rises to 63% among low-income countries (ILO 2026).

These institutional constraints are often reinforced by individual-level barriers. Workers frequently report costs, time constraints, and lack of information as obstacles to learning. Such barriers are especially pronounced among workers in vulnerable employment situations and women with care responsibilities, thereby perpetuating inequalities. 

At the same time, expanding access alone is insufficient if training programmes do not improve labour market outcomes (Card et al. 2018, Escudero et al. 2019). New evidence from a meta-analysis of 174 training interventions indicates that design and implementation features shape programme effectiveness (Escudero et al. forthcoming, ILO 2026). Figures 3 (panel a) shows that integrated training programmes combining work-based learning with classroom-based instruction are more consistently associated with positive employment outcomes than stand-alone on-the-job training or internships. Panel b shows that programmes leading to recognised qualifications are, on average, more effective than those leading to certificates that are not quality-assured. More effective programmes additionally strengthen combinations of technical, cognitive and socio-emotional skills, and incorporate complementary measures such as other active labour market policies or social protection mechanisms. 

Figure 3 Effectiveness of skills programmes across labour market outcomes

Notes: Results are based on 174 evaluation studies and are correlational. The results indicate the likelihood that evaluations found positive effects of training on earnings, hours worked, employment, and formal employment. Panel a) compares training types; panel b) compares programmes leading to certification and qualification. 
Source: ILO (2026). 

Main take-away

These findings indicate that the capacity of workers and firms to adapt to changing labour markets increasingly depends on the development of rounded skills profiles. Lifelong learning is therefore a crucial component of labour market policy, sustainable growth, and development strategies. However, most countries still lack ambitious lifelong learning systems. The inclusiveness, quality, and adaptability of these systems will influence how economies and societies navigate structural transformation.

References

Adamczyk, W, S Boehmer, I Delaporte, V Escudero, and H Liepmann (2025), “Developing a new method to uncover skills trends in emerging economies using online data and nlp techniques”, ILO Methodological Brief. 

Adamczyk, W, I Delaporte, V Escudero, and H Liepmann (2026), “Assessing the size of the green economy: Global evidence from harmonized labour force surveys”, ILO Research Brief. 

Borgonovi, F, F Calvino, C Criscuolo, J Nania, J Nitschke, L O’Kane, L Samek, and H Seitz (2023), “Tracking the skills behind the AI boom: Evidence from 14 OECD countries”, VoxEU.org, 20 October. 

Card, D, J Kluve, and A Weber (2018), “What works? A meta analysis of recent active labor market programme evaluations”, Journal of the European Economic Association 16(3): 894–931.

Deming, D, and L Kahn (2018), “Skill requirements across firms and labor markets: Evidence from job postings for professionals”, Journal of Labor Economics 36(S1): S337–69. 

Deming, D (2017), “The growing importance of social skills in the labor market”, Quarterly Journal of Economics 132(4): 1593–640. 

Delaporte, I, V Escudero, and W Adamczyk (2025), “Measuring the greenness of jobs in emerging economies: A big data text analysis approach”, ILO Research Brief. 

Delaporte, I, V Escudero, and F Petit (forthcoming), “The role of skills in mediating the effects of emerging digital technologies on employment”, ILO Working Paper.

Escudero, V, I Delaporte, M Á Malo, and M de la Cruz Rothenfusser (forthcoming), “The effectiveness of skills development policies: Evidence from a meta-analysis”, ILO Working Paper.

Escudero, V, J Kluve, E López Mourelo, and C Pignatti (2019), “Active labour market programmes in Latin America and the Caribbean: Evidence from a meta-analysis”, Journal of Development Studies 55(12): 2644–61.

Escudero, V, H Liepmann, and A Podjanin (2024), “Using online vacancy and job applicants’ data to study skills dynamics”, in Big Data Applications in Labor Economics, Part B, B Elsner and S W Polachek (eds.), Research in Labor Economics, vol. 52B. 

ILO (2026), World of work report: Lifelong learning and skills for the future, Geneva. 

Liepmann, H, and A Hegewisch (2025), “Revisiting occupational segregation and the valuation of women’s work”, ILO Working Paper No. 158. 

Stephany, F, and O Teutloff (2023), “Skills to ‘race with the machines’: The value of complementarity”, VoxEU.org, 15 November.