A highly proficient workforce does not automatically lead to greater use of skills. Understanding the different patterns of skills use at work is essential to design effective measures to raise productivity and improve job matching. Drawing on data from the 2023 Survey of Adult Skills, this chapter examines the relationship between workers’ proficiency levels and their actual use of skills at work. It highlights significant cross-country differences in the extent to which different skills are used in the workplace. The chapter also explores how skills use varies across different groups of workers and types of jobs, identifying patterns linked to factors such as gender, age, education, occupation and industry.
How Workers Use, or Don’t Use, their Skills in the Workplace
Abstract
High skills do not always lead to high-skill work
Across OECD countries, there is a substantial pool of untapped talent: individuals with strong skills who are not given adequate opportunities to apply them in their daily work.
The skills most commonly used in the labour market are self-organisation, task discretion, and co‑operative skills, reflecting a growing premium on autonomy, collaboration, and adaptability. Workers increasingly rely on personal initiative and teamwork rather than on information-processing skills such as numeracy, reading, and writing, which are used less often.
On average, women use their skills less often than men in the workplace, even when they have similar proficiency levels and hold comparable roles. This suggests that workplace task allocation may be biased by gender norms.
Differences in skills use between age groups are not solely the result of ageing, but vary considerably with task type. Numeracy use typically peaks in mid-career before declining, learning-at-work skills decrease steadily with age, and self-organisation skills rise and then stabilise as workers gain experience and autonomy.
Having a degree does not always translate into greater skills use, even after controlling for occupation. For example, graduates use numeracy, task discretion, and learning-at-work skills at similar levels to those with only upper secondary education. However, without at least upper secondary qualifications, access to skill-intensive work remains limited.
Skills use varies systematically across occupations and industries. High-skilled roles such as managers and professionals rely heavily on cognitive and social skills, while lower-skilled and manual occupations depend more on physical and dexterity skills. Co‑operation and self-organisation skills are widely used across all industries, underscoring the universal value of teamwork and autonomy, whereas ICT skills are concentrated in digitally intensive sectors like finance and communications.
Across the OECD, skill policies have historically emphasised the supply side: investments in lifelong learning, training participation, and qualifications. This focus reflects the relative ease of targeting supply levers through public policy. Expanding access to adult learning, improving the quality of training, and increasing overall skills proficiency are all well-established goals. Yet, this emphasis has often come at the expense of serious attention to how skills are actually deployed in workplaces. A growing body of research suggests that improving the stock of human capital is only part of the equation; how employers use those skills – i.e. the demand side – is equally critical for translating capabilities into economic performance (Cappelli, 2015[1]; Russo, 2017[2]).
The distinction between skills supply and skills use has important implications. Even when workers acquire high levels of proficiency through education or training, their skills may not be used regularly or effectively on the job. This underutilisation of skills has received some attention shortly after the release of the first cycle of the OECD Survey of Adult Skills (PIAAC), which highlights a persistent gap between what people are capable of and what they actually do at work (Quintini, 2014[3]; OECD, 2016[4]; OECD, 2016[5]; Jonas, 2018[6]). Workers in many OECD countries report lower levels of engagement with problem solving, numeracy, and literacy tasks than would be expected given their actual skills proficiency. The result is a form of latent human capital: skills that exist but are not activated, thereby reducing the potential returns to investment in education and training.
This gap has both microeconomic and macroeconomic consequences. At the firm level, poor skills utilisation can constrain productivity, reduce innovation potential, and lead to suboptimal work organisation. At the individual level, it can dampen job satisfaction, limit career progression, and contribute to disengagement or turnover. At the national level, this translates into lower aggregate productivity growth, reduced competitiveness, and weaker returns on public and private investments in education and training.1
Despite this, policy frameworks have been slow to address the demand side of skills use in a systematic way. Notable exceptions exist; for example, Singapore’s SkillsFuture Enterprise Credit encourages employers to support projects aimed at improving skills utilisation (OECD, 2019[7]). Nevertheless, most national strategies still prioritise skills supply. The challenge going forward is to develop policies that do more than simply expand the skill base. Governments, employers, and social partners need to focus on the conditions that enable the productive use of skills. If skills are to function as genuine levers of productivity and inclusion, then greater attention must be paid to how they are deployed and not just acquired.
A decade after the last major analysis of skills use based on PIAAC data, the most recent results from 2023 in Figure 1.1 continue to show no strong positive correlation between the average proficiency of adults in literacy or numeracy and the frequency with which these skills are used in the workplace (see Box 1.1 for details on how PIAAC measures skills use).2 In other words, countries with higher average skill levels do not necessarily report more frequent use of those skills at work. This weak association suggests that factors beyond skills proficiency play a critical role in shaping how skills are utilised.3 The mismatch may reflect structural issues in the labour market, including limited availability of high-skill jobs, underemployment, or inefficient allocation of talent. They can also arise from organisational practices that do not fully leverage employees’ capabilities, or from cultural and institutional factors that influence the nature of tasks performed at work. At the same time, the opposite pattern can be found as well. For instance, New Zealand ranks 15th in terms of average literacy proficiency but leads all OECD countries in the frequency of reading skills use at work. This indicates that, despite more moderate skill levels, New Zealand’s labour market and workplace environment more effectively exploit literacy skills into daily work tasks.
Average proficiency scores and average skills use at work among the working 16‑ to 65‑year‑old population
Note: R-squared (R²) indicates how well the average literacy proficiency explains the variation in the use of reading skills at work across countries. For instance, R² =0.37 means that only 37% of the differences in reading use can be attributed to differences in proficiency scores. For consistency, the country name is reported for Belgium and the United Kingdom, even if the Survey of Adult Skills is conducted only at subnational level – namely in the Flemish region and in England, respectively.
Source: 2023 Survey of Adult Skills.
The Survey of Adult Skills is widely recognised for its unique assessment of adult proficiency in literacy, numeracy, and adaptive problem solving. This self-administered assessment (which takes about two hours to complete) is conducted through a series of carefully designed items that directly evaluate individuals’ abilities to understand, interpret, and apply information in practical contexts. Less well known, however, is that, in its background questionnaire, PIAAC also gathers standardised, cross-country data on how often these skills are used in both work and everyday settings.
Specifically, PIAAC captures the frequency with which adults carry out tasks involving core information-processing skills – such as reading, writing, numeracy, ICT use, and problem solving – as well as seven additional generic workplace skills – task discretion, learning at work, influencing, co‑operation, self-organisation, dexterity, and physical skills.1 Respondents report how often they perform specific tasks associated with each skill.2 Task frequency is scored on a five‑point scale: 1 (never), 2 (less than once a month), 3 (less than once a week but at least once a month), 4 (at least once a week but not every day) 5 (every day).3
Skills use indicators are derived by aggregating multiple task indicators to enhance the robustness of the composite measures (Table 1.1). More specifically, composite indices for each skill type are generated by summing responses to the relevant task items, producing a semi-continuous scale between 1 and 5. A higher score indicates more frequent application of the corresponding skill. Internal consistency of these scales is verified using Cronbach’s Alpha to ensure that the selected tasks appropriately reflect the underlying skill construct (see Quintini (2014[3]) and Marcolin, Miroudot and Squicciarini (2019[8]) for a similar approach).4
Note: For the actual identifier code of the selected PIAAC variable, refer to the Annex.
1. The report refers to these indicators as “skills”, including dimensions such as task discretion and self-organisation. This choice follows established practice in the skills measurement and labour-market literature, where these constructs are commonly conceptualised and labelled as skills rather than solely as job characteristics or working conditions. Major international skills taxonomies, such as O*NET and ESCO, explicitly classify comparable constructs (e.g. autonomy, self-management, planning, influencing others, and co‑operation) as skills or skill-related competencies. Using the term “skills” therefore ensures consistency with these widely used frameworks and facilitates comparability with other empirical and policy-oriented analyses, while recognising that some of these skills are expressed and developed through the organisation of work and task allocation.
2. From a conceptual perspective, work tasks can be described in a number of ways (OECD, 2013[9]): by looking at their frequency (the frequency with which a given task is performed), complexity (the level of difficulty required to perform the task successfully) or criticality (the importance of the task to the perform the job). The Survey of Adult Skills adopts frequency as the primary indicator for assessing skill requirements, since it is generally more objective and easier for respondents to recall, which reduces measurement bias. By contrast, the perceived complexity of a task can depend on multiple situational factors and may be influenced by subjective judgement. For example, respondents are usually able to report how often they read work related documents, but the complexity of that reading depends, among the others, on the nature of the documents, their technical content and length. It is, however, important to recognise that, while measures of skills use based on frequency (such as those collected in PIAAC) provide a valid indication of workplace skill demands, for certain occupations the complexity or criticality of specific tasks may depend on contextual conditions that frequency measures alone cannot fully capture.
3. As questions measuring ICT skills use are only posed to respondents who indicate prior computer use, very few respondents report never using ICT skills in the workplace. This leads to a distribution that differs from other skills use scales, and therefore the ICT-at-work scale should be interpreted accordingly.
4. Another strand of the literature computing skills use indices with PIAAC data relies on item response theory (IRT) (see, for instance, OECD (2019[10])). This technique estimates the probability with which the respondent gives a certain answer to the underlying questions (“items”). A set of skills use indices based on IRT is available in the PIAAC Public Use Files (PUFs) for researchers to explore.
Another way to demonstrate the issue of skill underutilisation is to examine the share of workers who demonstrate high levels of literacy proficiency – specifically, those scoring at Level 4 or above – but who make limited use of these skills in their jobs. These individuals fall into the bottom third (tercile) of workplace reading use, despite having the capacity for much more demanding tasks. This mismatch highlights a significant pool of untapped potential: workers who possess strong skills but are not given opportunities to apply them in their day-to-day work.
Panel A of Figure 1.2 illustrates that this is a widespread phenomenon across all countries participating in the second cycle of the Survey of Adult Skills. In every country, at least 10% of highly proficient workers hold jobs that require very little use of their proficiency. In certain countries, such as Croatia and Singapore, this figure rises dramatically – up to one‑third of high-skilled workers are not using their literacy skills to their full extent on the job. Panel B examines high-skilled occupations, which may be more likely to entail frequent literacy and reading tasks compared to elementary jobs. The evidence demonstrates that even within these occupations, underutilisation persists. For example, almost one‑quarter of health associate professionals with high literacy proficiency scores fall in the lowest tercile for reading use at work. Whatever the cause of such underutilisation of talent, the result is a missed opportunity for both individuals and economies: skills that are developed but not deployed represent lost productivity, lower job satisfaction, and diminished returns on investment in education and training.
Percentage of the working 16‑ to 65‑year‑old population who scores at or above Level 4 in literacy but is in the bottom tercile in terms of using reading skills at work
Note: Panel B covers only high-skilled occupations, i.e. those with ISCO 1‑digit occupation codes 1 to 3, and presents them at 2‑digit level for more granularity.
Source: 2023 Survey of Adult Skills.
According to the most recent cycle of the Survey of Adult Skills of 2023, self-organising skills are the most commonly used in the workplace in nearly all countries surveyed (Figure 1.3). In most cases, the reported frequency of using self-organising skills exceeds 4 out of 5, indicating they are used at least weekly, though not necessarily every day. Task discretion – i.e. the ability to make independent decisions about the order and speed of one’s tasks – is also reported very frequently in a majority of countries, underscoring the importance of personal initiative and self-management in modern labour markets. Taken together, self-organising skills and task discretion reflect a worker’s capacity to plan tasks and adapt to changing circumstances without constant supervision. As job roles become more autonomous and less routine, employers increasingly value individuals who can take proactive responsibility for their own performance and development (Autor, Levy and Murnane, 2003[11]).
Co‑operative skills are also among the most frequently used in workplaces across the countries surveyed by PIAAC. This aligns with recent labour market research (e.g. Deming (2017[12])), which shows that social skills – such as co‑operation – are becoming increasingly valuable. By enabling more efficient collaboration, these skills help reduce co‑ordination costs and support greater adaptability in response to changing workplace conditions. Contrary to the assumption that modern jobs routinely demand high levels of analytical reasoning, the deployment of problem-solving skills at work remains less frequent than that of other skills. On average, workers across the PIAAC sample report being confronted with complex problems less than once per week, with only 13% of them using problem-solving skills daily.
By contrast, numeracy is the least frequently used skill in the workplace in the majority of countries participating in the second cycle of the Survey of Adult Skills, with 24 out of 31 countries reporting the lowest average frequency of use for this skill. In no country does the average reported use of numeracy at work exceed a score of 3 on the survey scale, which corresponds to less than once per week. This finding suggests that, despite its foundational nature, numeracy plays a relatively limited role in the day-to-day tasks of most workers.
More broadly, the data indicate that information-processing skills – such as reading, writing, and numeracy – are employed less frequently on the job than generic skills.4 In today’s labour market, which is increasingly characterised by collaborative, service‑oriented, and dynamic work environments, workers more commonly engage in activities that emphasise teamwork and autonomy. Academic research supports this shift in work practices. Studies such as Autor (2015[13]), who documents the declining relevance of routine cognitive tasks (such as calculation and record-keeping) in favour of non-routine interpersonal and analytical work, highlight a structural transformation in skill utilisation patterns.
Note: The heatmap uses a blue‑white‑red colour scale to illustrate the relative use of skills in the workplace across countries. Blue indicates above‑average use, white represents average use, and red signifies below-average use. The intensity of the colour corresponds to the degree of deviation from the overall average.
Source: 2023 Survey of Adult Skills.
Figure 1.4 and Figure 1.5 present the average use of information-processing skills and generic skills at work, by country (while this analysis focusses on workplace practices, Box 1.2 discusses the relationship between skills use at work and in everyday life). Interesting patterns emerge from the comparison between countries. For example, while reading and writing are both integral components of literacy, the data reveal that countries often exhibit different patterns in the use of these skills. Notably, high use of one skill does not necessarily imply high use of the other. This divergence highlights the complex and context-dependent nature of workplace literacy.
For instance, workers in Belgium report, on average, using writing skills almost every week, ranking first overall in terms of writing use in the workplace. In contrast, they use reading skills at least once a month but less than once a week, placing Belgium 10th out of 31 countries for reading use at work. These disparities can arise from differences in job structures, sectoral compositions, and workplace practices across countries. For example, a workforce heavily concentrated in administrative or managerial roles may require more writing – such as drafting emails, reports, or documentation – than reading, particularly if much of the information is communicated verbally. Conversely, jobs in fields such as education, healthcare, or technical services might require more reading of complex texts, regulations, or technical manuals than original writing. Furthermore, cultural norms around communication can influence how literacy skills are deployed. In some countries, concise oral communication may be preferred over written documentation, or vice versa (Richardson and Smith, 2007[14]; Ortiz, Region-Sebest and MacDermott, 2016[15]).
The case of numeracy is particularly noteworthy. Not only, as previously discussed, does numeracy emerge as the least frequently used skill at work, but it also exhibits a remarkable consistency across countries: the average frequency of numeracy use hovers around 2.5 on the 5‑point scale, with minimal variation across countries. One explanation for this uniformity may lie in the occupational distribution of numeracy-intensive tasks. Numeracy skills – such as performing arithmetic calculations, interpreting tables and graphs, or estimating quantities – are not universally embedded in the day-to-day activities of most workers. Rather, these skills are concentrated in specific technical roles in sectors like engineering, finance, accounting. In contrast, large portions of the workforce in industries such as retail, hospitality, healthcare, personal services, and administrative support may rarely need to engage in explicit numerical tasks on a regular basis.5
Dexterity use exhibits the highest standard deviation of the 12 skills assessed in the Survey of Adult Skills. This indicates that the frequency with which dexterity is used in the workplace varies more widely across countries than any other skill. For instance, in countries like Chile and Latvia tasks requiring manual precision are performed at least weekly, if not daily, with average use ratings close to 4 on a 5‑point scale. Conversely, in countries such as France and Singapore, dexterity ranks as the least used skill, with an average score of approximately 2.5. This level of use implies that tasks involving dexterity occur infrequently – perhaps only once a month or less in many occupations.
This stark contrast highlights the diverse economic structures among OECD Member states. Countries with a significant share of employment in manufacturing tend to emphasise dexterity more heavily, whereas those with economies oriented around information technology and professional sectors – commonly referred to as knowledge economies – place less emphasis on manual skills (Kunst, 2020[16]). These patterns are consistent with findings from the economic literature, which underscore how occupational structures and technological advancement shape skill demand across national labour markets (Sasso and Ritzen, 2019[17]).
More broadly, examining how countries rank in the use of various skills in the workplace reveals several interesting patterns. New Zealand stands out as the country with the most frequent use of a wide range of skills. Among the 31 participating countries, it ranks first in the use of reading, numeracy, ICT, problem solving, and influencing skills, and it is among the top five for use of learning at work, co‑operative skills, and physical skills. The United Kingdom, instead, performs strongly in the use of information-processing skills (consistently ranking among the top three), but it displays significantly lower use of generic skills, such as task discretion, learning at work, and manual dexterity. Some countries demonstrate a more mixed profile. For instance, Finland ranks at the top in the use of task discretion, self-organisation, and influencing skills, yet falls below the average for ICT and co‑operative skills use. Similarly, in the United States, workers report high use of reading, numeracy, problem solving, and dexterity, but rank among the lowest for self-organising skills.
Source: 2023 Survey of Adult Skills.
Source: Survey of Adult Skills (Cycle 2).
While this report primarily examines the use of skills in the workplace, the Survey of Adult Skills also gathers information on how adults apply reading, writing, numeracy, and ICT skills in their everyday lives. To explore how skills are used across different contexts, Figure 1.6 shows the representative case of the relationship between the use of numeracy skills at work and in everyday life. The data reveal a positive correlation between these two domains at the country level (R-squared = 0.62). In general, countries where adults frequently use numeracy in their jobs also report higher levels of numeracy use at home.
However, the two are not perfectly aligned. Across the PIAAC sample, adults often report greater use of numeracy in everyday life than in the workplace. This gap is particularly pronounced in the Nordic countries – such as Finland, Norway, and Sweden – where the average adult reports using numeracy skills at home approximately 0.50 points more (on a 1‑to‑5 scale) than at work. Several factors may help explain this pattern. First, in many advanced economies, individuals are increasingly responsible for managing complex aspects of their personal and household affairs – including budgeting, financial planning, and interpreting utility bills. These activities routinely require the application of numeracy skills such as estimation, calculation, and data interpretation. Second, the digitalisation of public services and commerce has shifted more numeracy-intensive tasks to individuals. Online banking, e‑government portals, and e‑commerce platforms often demand a basic proficiency in numeracy for tasks like verifying transactions, comparing prices, and assessing service terms. Third, in many work environments, especially in sectors characterised by high levels of automation, opportunities for employees to regularly use numeracy skills may be limited (Hodgen and Marks, 2013[18]). Lower-skilled jobs, in particular, may involve tasks that are either repetitive or supported by technology, reducing the need for active numerical reasoning. While these interpretations help to frame the observed differences, the magnitude of the gap in some countries suggests that further analysis would be needed to fully understand the underlying drivers.
Note: The dashed line represents the 45‑degrees line (and not the trend line). R-squared (R²) indicates how well the average numeracy use at work explains the variation in the use of numeracy skills in everyday life across countries. For instance, R² = 0.62 means that 62% of the differences in numeracy use everyday can be attributed to differences in numeracy use at work.
Source: 2023 Survey of Adult Skills.
To better understand the factors influencing skills use in the workplace, it is essential to examine how individual and job characteristics shape the way workers translate their proficiency into on-the‑job skill deployment.
Figure 1.7 presents the estimated average difference in skills use between women and men, controlling for age, education, literacy proficiency, occupation, industry, firm size, full-time status, contract type, and public versus private sector employment. On average, women use their skills less frequently than men in the workplace, even when they possess similar levels of proficiency and hold similar occupations. For instance, women score 0.24 points lower than men in the use of numeracy skills, a difference that remains significant after accounting for all covariates. This could reflect persistent gender segregation within occupations – e.g. within the same occupational titles, men may disproportionately occupy roles requiring quantitative tasks (such as budgeting or data analysis). Dexterity is the only skill that women report using slightly more frequently than men, though the difference is modest. Indeed, women may be more represented in roles requiring fine motor skills (e.g. precision work, manual tasks in healthcare or services), even within similar occupational categories.
Overall, these results point at the fact that task assignments in the workplace may be gender-biased, even after controlling for observable characteristics (Pető and Reizer, 2021[19]; Hauret et al., 2023[20]). This task-level segregation can reinforce gender disparities in skill development, productivity, and pay. The fact that men and women in the same occupation are routinely assigned different types of work limits opportunities for women to accumulate experience in high-value or high-visibility tasks (such as data analysis, decision making, or problem-solving) that are more likely to lead to promotion and wage increases (De Pater, Van Vianen and Bechtoldt, 2010[21]; Babcock et al., 2017[22]; Bizopoulou, 2019[23]).
OLS coefficients
Note: The Figure represents the coefficient of OLS regressions of skills use on gender. Coefficients are adjusted for age, age squared, educational attainment, literacy proficiency, occupation, industry, firm size, full-time work, permanent contract, public employment, and country fixed effects. The Figures also includes standard error bars.
Source: 2023 Survey of Adult Skills.
Studies on adult skills consistently finds that older adults (often defined as individuals aged 55 to 65) tend to exhibit weaker proficiency in information-processing skills such as literacy, numeracy, and adaptive problem-solving, relative to younger age groups (Picchio, 2015[24]; Paccagnella, 2016[25]; OECD, 2025[26]). These age‑related differences persist even after accounting for observed differences in education level and field of study. However, while the decline in cognitive performance with age is well documented, the relationship between age and the use of skills in the workplace is less straightforward. One might expect an ageing effect – that is, the natural decline in certain cognitive functions, including memory and information-processing speed – to result in lower skills use at older ages. Yet, empirical evidence does not suggest a uniform pattern of decline. Instead, skills use across age groups appears to be heterogeneous and task dependent.
Figure 1.8 illustrates this complexity by showing varied trajectories of skills use with age. For some skills, such as numeracy, use follows an inverted U-shaped pattern: it tends to increase during mid-career as individuals accumulate experience and confidence, before declining in later years, potentially due to cognitive ageing or changes in job roles.6 For other skills, such as learning at work – i.e. the extent to which workers acquire new knowledge or update their skills – use tends to decline steadily with age. This may reflect reduced training opportunities, including informal learning from peers (as found by OECD (2025[26])), or lower motivation to engage in upskilling. Conversely, the use of self-organisation skills tends to increase with age and then stabilise. This pattern likely reflects the accumulation of job experience, seniority, and autonomy over time, which can compensate for declining cognitive skills by enabling older workers to draw on their expertise.
These heterogeneous patterns highlight that age‑related differences in skills use cannot be solely attributed to ageing. Workplace dynamics, job design, and institutional factors – such as access to training and career progression pathways – also play critical roles in shaping how skills are used throughout the life course (Hanushek et al., 2025[27]).
Note: The Figure displays the predicted skills use by age, based on a weighted regression model. Confidence intervals indicate the precision of each estimate.
Source: 2023 Survey of Adult Skills.
Figure 1.9 presents the relationship between educational attainment and the predicted use of various workplace skills, controlling for a set of individual and job-level characteristics.7 The data reveal that skills use does not increase systematically with higher levels of educational attainment. Instead, the pattern varies markedly across domains. For several skills, such as task discretion, learning at work and numeracy, individuals with post-secondary or tertiary education report frequency of use comparable to those with upper secondary qualifications once occupation, proficiency and other characteristics are taken into account. In contrast, substantial differences by educational attainment emerge for skills such as self-organising, problem solving and influencing skills, where tertiary-educated workers demonstrate notably higher levels of use. While this finding suggests that tertiary education does not always translate into higher on-the‑job skills use, it still underscores the critical importance of achieving at least upper secondary education as a minimum threshold for participation in more skill-intensive work. It also highlights the challenge of addressing low skills use among workers with lower levels of formal education – an issue with direct implications for productivity, social mobility, and lifelong learning policy.
Note: The Figure displays the predicted skills use by educational attainment, based on a weighted regression model. Coefficients are adjusted for gender, age, age squared, literacy proficiency, occupation, industry, firm size, full-time work, permanent contract, public employment, and country fixed effects.
Source: 2023 Survey of Adult Skills.
The average use of skills across occupations generally aligns with expectations (Figure 1.10). Workers in high-skilled occupations – such as managers, professionals, and technicians – tend to make more frequent use of a broad range of skills. These include cognitive (e.g. writing) and social (e.g. influencing) skills, which are integral to complex decision-making tasks. In contrast, lower-skilled occupations, including plant and machine operators and workers in elementary occupations, rely more heavily on physical and manual dexterity skills. Interestingly, co‑operation skills are frequently used across all occupational groups. This underscores the pervasive importance of teamwork and interpersonal collaboration in today’s labour market, regardless of skill level or occupation.
Note: The Figure displays the predicted skills use by occupation, based on a weighted regression model. Coefficients are adjusted for gender, age, age squared, literacy proficiency, educational attainment, industry, firm size, full-time work, permanent contract, public employment, and country fixed effects. The heatmap uses a blue‑white‑red colour scale to illustrate the relative use of skills in the workplace across occupations. Blue indicates above‑average use, white represents average use, and red signifies below-average use. The intensity of the colour corresponds to the degree of deviation from the overall average (i.e. the average of all skills and occupations together).
Source: 2023 Survey of Adult Skills.
An analysis of skills use across industries reveals significant variation in the demand for different skills depending on the nature of work. Table 1.2 presents the relative ranking of skill intensity in 20 sectors (with a rank of 1 referring to the most used skill and a rank of 12 pointing to the least used one). In nearly all sectors, self-organising, task discretion, and co‑operative skills rank among the most used (commonly ranked in the top 3). In line with previous findings from this Chapter, reading, writing, and particularly numeracy are infrequently used at work across most sectors. For example, numeracy is the least used skill (rank 12) in 13 of the 20 sectors, including manufacturing, retail, and education. Reading and writing also see low usage outside of a few high-skill services like real estate and public administration. ICT skills are highly ranked (1‑3) in financial services, information and communication, and administration and support services, reflecting digitisation in these areas. Conversely, ICT is less used (ranks 7‑8) in accommodation and food services and in health. These two sectors use more frequently dexterity and physical skills (ranks 2‑4).
Ranking of the frequency of use of skills at work in each industry
Note: The Figure displays the ranking of use of skills at work. To construct this ranking, the predicted skills use by industry, based on a weighted regression model, is exploited. Underlying coefficients are adjusted for gender, age, age squared, literacy proficiency, educational attainment, occupation, firm size, full-time work, permanent contract, public employment, and country fixed effects.
Source: 2023 Survey of Adult Skills.
To conclude, the chapter assesses the relationship between a range of job characteristics and the frequency of skills use at work (Table 1.3). Job stability, proxied by the presence of an open-ended (indefinite) contract compared to a temporary or fixed-term arrangement, is positively associated with the use of information-processing skills – particularly writing and problem-solving. This aligns with evidence from OECD (2016[5]), which suggests that stable employment relationships provide more opportunities for skill development and utilisation, likely due to greater investments in training and in tailoring job content by employers. However, no statistically significant differences are observed between permanent and temporary contracts regarding the use of dexterity or physical skills. This may reflect the more routinised nature of these tasks, which are less sensitive to employment duration.
Work schedule arrangements have a substantial effect on skills use. In particular, part-time workers report markedly lower use of virtually all skill categories at work.8 While employment in the public sector is associated with increased use of reading, dexterity, learning at work, and influencing skills, these differences are relatively modest when compared to the stark contrast between part-time and full-time workers. This suggests that job structure and time commitment may play a more critical role than sectoral affiliation in determining skill utilisation.
The size of the company also plays a significant role. Larger firms tend to promote greater use of information-processing skills such as reading, writing, and problem solving. This may be due to the more complex, hierarchical, and bureaucratic structures of large firms, which often require higher levels of documentation, co‑ordination, and analytical capacity. In contrast, smaller firms tend to rely more heavily on dexterity and physical skills, as well as on task discretion and self-organising competencies. The flatter organisational structures and broader role expectations typical of smaller enterprises often necessitate greater autonomy and physical versatility from employees (OECD, 2019[28]).
OLS coefficients
Note: Coefficients are adjusted for gender, age, age squared, educational attainment, literacy proficiency, occupation, industry, and country fixed effects.
Source: 2023 Survey of Adult Skills.
References
[13] Autor, D. (2015), “Why Are There Still So Many Jobs? The History and Future of Workplace Automation”, Journal of Economic Perspectives, Vol. 29/3, pp. 3-30, https://doi.org/10.1257/JEP.29.3.3.
[11] Autor, D., F. Levy and R. Murnane (2003), “The Skill Content of Recent Technological Change: An Empirical Exploration”, The Quarterly Journal of Economics, Vol. 118/4, pp. 1279-1333, https://doi.org/10.1162/003355303322552801.
[22] Babcock, L. et al. (2017), “Gender Differences in Accepting and Receiving Requests for Tasks with Low Promotability”, American Economic Review, Vol. 107/3, pp. 714-47, https://doi.org/10.1257/AER.20141734.
[23] Bizopoulou, A. (2019), “Job Tasks and Gender Wage Gaps within Occupations”, VATT Working Papers, No. 124, VATT Institute for Economic Research, Helsinki.
[1] Cappelli, P. (2015), “Skill gaps, skill shortages, and skill mismatches: Evidence and arguments for the United States”, Industrial and Labor Relations Review, Vol. 68/2, pp. 251-290, https://doi.org/10.1177/0019793914564961.
[21] De Pater, I., A. Van Vianen and M. Bechtoldt (2010), “Gender Differences in Job Challenge: A Matter of Task Allocation”, Gender, Work & Organization, Vol. 17/4, pp. 433-453, https://doi.org/10.1111/J.1468-0432.2009.00477.X.
[12] Deming, D. (2017), “The Growing Importance of Social Skills in the Labor Market”, Quarterly Journal of Economics, Vol. 132/4, pp. 1593-1640, https://doi.org/10.1093/QJE/QJX022.
[33] Goos, M. et al. (2023), “Supporting adults to become numerate citizens: a study of adult numeracy provision in Ireland”, ZDM - Mathematics Education, Vol. 55/5, pp. 995-1008, https://doi.org/10.1007/S11858-023-01480-9/TABLES/5.
[29] Green, F. et al. (2013), Job-related Well-being in Britain: First Findings from the Skills and Employment Survey 2012, Centre for Learning and Life Chances in Knowledge Economies and Societies, Institute of Education.
[27] Hanushek, E. et al. (2025), “Age and cognitive skills: Use it or lose it”, Science Advances, Vol. 11/10, https://doi.org/10.1126/SCIADV.ADS1560/SUPPL_FILE/SCIADV.ADS1560_SM.PDF.
[20] Hauret, L. et al. (2023), “Gender gaps in skills, tasks, and employment outcomes”, https://doi.org/10.5281/ZENODO.7998454.
[18] Hodgen, J. and R. Marks (2013), The Employment Equation: Why our young people need more maths for today’s jobs, The Sutton Trust, London.
[6] Jonas, N. (2018), “Numeracy practices and numeracy skills among adults”, OECD Education Working Papers, No. 177, OECD Publishing, Paris, https://doi.org/10.1787/8f19fc9f-en.
[16] Kunst, D. (2020), “Deskilling among Manufacturing Production Workers”, Tinbergen Institute Discussion Paper, No. TI 2019-050/VI, Tinbergen Institute, https://www.tinbergen.nl (accessed on 4 August 2025).
[8] Marcolin, L., S. Miroudot and M. Squicciarini (2019), “To be (routine) or not to be (routine), that is the question: a cross-country task-based answer”, Industrial and Corporate Change, Vol. 28/3, pp. 477-501, https://doi.org/10.1093/ICC/DTY020.
[26] OECD (2025), OECD Employment Outlook 2025: Can We Get Through the Demographic Crunch?, OECD Publishing, Paris, https://doi.org/10.1787/194a947b-en.
[32] OECD (2024), Survey of Adult Skills – Reader’s Companion: 2023, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/3639d1e2-en.
[7] OECD (2019), OECD Skills Strategy 2019: Skills to Shape a Better Future, OECD Publishing, Paris, https://doi.org/10.1787/9789264313835-en.
[28] OECD (2019), OECD SME and Entrepreneurship Outlook 2019, OECD Publishing, Paris, https://doi.org/10.1787/34907e9c-en.
[10] OECD (2019), Skills Matter: Additional Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/1f029d8f-en.
[4] OECD (2016), OECD Employment Outlook 2016, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2016-en.
[5] OECD (2016), Skills Matter: Further Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264258051-en.
[9] OECD (2013), OECD Skills Outlook 2013: First Results from the Survey of Adult Skills, OECD Publishing, Paris, https://doi.org/10.1787/9789264204256-en.
[15] Ortiz, L., M. Region-Sebest and C. MacDermott (2016), “Employer perceptions of oral communication competencies most valued in new hires as a factor in company success”, Business and Professional Communication Quarterly, Vol. 79/3, pp. 317-330, https://doi.org/10.1177/2329490615624108.
[25] Paccagnella, M. (2016), “Age, Ageing and Skills: Results from the Survey of Adult Skills”, OECD Education Working Papers, No. 132, OECD Publishing, Paris, https://doi.org/10.1787/5jm0q1n38lvc-en.
[19] Pető, R. and B. Reizer (2021), “Gender differences in the skill content of jobs”, Journal of Population Economics, Vol. 34/3, pp. 825-864, https://doi.org/10.1007/S00148-021-00825-6/FIGURES/4.
[24] Picchio, M. (2015), “Is training effective for older workers?”, IZA World of Labor, Vol. 121, https://doi.org/10.15185/IZAWOL.121.
[3] Quintini, G. (2014), “Skills at Work: How Skills and their Use Matter in the Labour Market”, OECD Social, Employment and Migration Working Papers, No. 158, OECD Publishing, Paris, https://doi.org/10.1787/5jz44fdfjm7j-en.
[14] Richardson, R. and S. Smith (2007), “The influence of high/low-context culture and power distance on choice of communication media: Students’ media choice to communicate with Professors in Japan and America”, International Journal of Intercultural Relations, Vol. 31/4, pp. 479-501, https://doi.org/10.1016/J.IJINTREL.2007.01.002.
[2] Russo, G. (2017), “Skill utilization at work: Opportunity and motivation”, IZA World of Labor, https://doi.org/10.15185/IZAWOL.409.
[17] Sasso, S. and J. Ritzen (2019), “Sectoral cognitive skills, R&D, and productivity: a cross-country cross-sector analysis”, Education Economics, Vol. 27/1, pp. 35-51, https://doi.org/10.1080/09645292.2018.1515309.
[31] UKCES (2014), The Labour Market Story: Skills Use at Work, UK Commission for Employment and Skills.
[30] Wright, J. and P. Sissons (2012), The skills dilemma: skills under-utilisation and low-wage work, The Work Foundation, London.
← 1. While these effects are not merely anecdotal, economic evidence is limited and dated. Research has linked better use of skills in the workplace to higher job satisfaction and improved well-being, which has led some analysts to associate skills use with the concept of job quality (Green et al., 2013[29]). There is also evidence that more effective skill deployment contributes to increased productivity in firms (UKCES, 2014[31]), as well as greater innovation, engagement, and investment (Wright and Sissons, 2012[30]). However, more recent evidence is needed on the impact of skills use at work on economic and social outcomes, and this report aims at filling part of this gap.
← 2. Throughout this report, caution is required in interpreting results for Poland due to the high share of respondents with unusual response patterns in the second cycle of the Survey of Adult Skills – see note for Poland in the Reader’s Companion (OECD, 2024[32]).
← 3. In addition, the mismatch may partially reflect the fact that skill proficiency as measured by the Survey of Adult Skills covers a broader range of literacy domains compared to the use of reading skills at work.
← 4. It is important to note that not every skill needs to be used intensively in every job. Certain highly skilled workers may be satisfied and highly productive without regularly using certain skills that they possess. This is why this report does not interpret low use of a specific skill in isolation as a problem per se, nor it suggests that individuals should change jobs simply to increase the frequency with which they use a particular skill. The skills-use measures derived from PIAAC are descriptive: they document how skills are used across jobs and workers, not how they should be used. As such, the analysis is intended to inform policy discussions at the aggregate level – for example, about how education systems, training policies, or workplace practices shape opportunities to use skills.
← 5. Measuring skills use through self-reporting (like in the Survey of Adult Skills) might also underestimate numeracy if some workers cannot recognise their tasks as requiring numerical skills (e.g. interpreting inventory trends or estimating time and cost). This phenomenon of “invisible numeracy” has been well documented in the literature (Goos et al., 2023[33]). Numeracy is also often embedded in other tasks, such as using software (Jonas, 2018[6]). As a result, respondents may be less likely to report these tasks as involving numeracy, compared to more conscious skills like reading and writing.
← 6. Differences in skills use and proficiency among age groups may be also partially driven by cohort effects (OECD, 2025[26]). Cohort (generational) effects stem from the unique conditions each age group experienced growing up – such as the level of access to technology, education systems, labour market dynamics, and public policies. For instance, younger individuals often demonstrate stronger information-processing skills, also because they may have had greater exposure to modern digital technologies during their schooling years.
← 7. The baseline specification controls for a detailed set of individual characteristics as well as job attributes, including 1‑digit ISCO occupation dummies (10 groups) and 1‑digit ISIC industry dummies (21 sections). Given that educational attainment influences occupational sorting, alternative specifications excluding occupation and industry controls were also estimated. The relationship between educational attainment and the predicted use of workplace skills remains qualitatively comparable.
← 8. The PIAAC skill-use variables were designed to be applicable to both full-time and part-time workers, as they measure the frequency of engaging in specific tasks rather than total time spent on them. However, it is possible that part-time workers may under-report their use of skills because they have fewer working hours within a typical week or month, which can reduce the perceived frequency of task engagement even when the nature of their work is similar to that of full-time employees.