2. The changing skill needs of the German labour market

In the decade leading up to 2020, Germany experienced robust economic growth, having recovered faster than other OECD and European countries from the global financial and economic crisis. Unemployment rates fell to their lowest level since reunification, and high employment growth brought about critical shortages of skilled labour (Bundesagentur für Arbeit, 2019[1]; OECD, 2020[2]). Individuals benefited from high standards of living and high levels of well-being by OECD standards, with many people seeing real wage gains (OECD, 2018[3]; OECD, 2020[2]).

As the COVID-19 pandemic spread and global output collapsed, Germany’s GDP contracted sharply, with an estimated decrease by more than 5% in 2020 (OECD, 2021[4]), although this drop is less pronounced compared to other OECD countries. To date, employment and unemployment rates have also been less affected by the COVID-19 crisis than in many other OECD countries. This is due to the government’s strong action in extending the well-established short-time work scheme (Kurzarbeit), the provision of liquidity support for enterprises and other measures to stimulate domestic demand.

Now more than ever, one of the main concerns for policy-makers in Germany is to ensure that there is a sustained recovery back to the strong social and economic outcomes experienced prior to the pandemic. Even before the crisis, the nature of work was changing through digitalisation, population ageing and the transition to a low-carbon economy. According to OECD analysis, 18% of jobs in Germany are at high risk of automation in the next 15 years, while a further 36% are at risk of significant change, adding up to one of the highest shares of jobs at risk across OECD countries (Nedelkoska and Quintini, 2018[5])

The economic fallout from the COVID-19 pandemic may well accelerate these trends. In this context, it becomes ever more important to ensure well-functioning continuing education and training systems that enable individuals to upskill and reskill to stay abreast of these developments. The National Skills Strategy (Nationale Weiterbildungsstrategie), adopted in 2019, recognises the ongoing structural changes in the German labour market. It aims to address this challenge by strengthening upskilling and reskilling opportunities and by reaffirming a culture of continuing education and training in Germany (BMAS et al., 2019[6]).

Reviewing the outcomes of the current CET landscape demonstrates where improvements are necessary. Just over half of all adults in Germany participate in CET in a given year and there are large inequalities in access to CET. There is limited evidence of the impact of CET participation on employment and social outcomes. Evaluation evidence from the area of Active Labour Market Policy (ALMP) in Germany suggests that some types of CET, in particular longer CET opportunities, lead to positive labour market outcomes for certain groups and occupations in the medium to long term (Bläsche et al., 2017[7]; Bernhard, 2016[8]; Bernhard, Lang and Kruppe, 2017[9]; Doerr et al., 2014[10]; Ehlert, 2017[11]; Kruppe and Lang, 2015[12]).

This chapter discusses issues of skill demand and supply in Germany, highlights the resulting skill imbalances and the increasing need for upskilling and reskilling throughout the life-course. Following this analysis, it discusses current patterns of participation in continuing education and training in Germany, as well as the economic and social outcomes associated with participation.

The world of work is changing, as digitalisation, technological progress, population ageing and the transition to a low-carbon economy are transforming the type of jobs that are available in Germany and how they are carried out. These changes had led to critical skills shortages prior to the COVID-19 crisis. The economic fallout from the crisis may now accelerate pre-existing trends that are changing the skills that are in demand in the labour market.

Prior to 2020, Germany had experienced steady and robust economic growth for a decade. The strong economic performance built on healthy domestic demand, a robust trade performance and good social outcomes (OECD, 2018[3]; OECD, 2020[2]). As the COVID-19 pandemic took hold, containment and mitigation policies, as well as the great uncertainty of the global outlook led to a sharp contraction in economic activity in Germany (OECD, 2020[2]; OECD, 2021[4]). According to OECD estimates, real GDP decreased by more than 5% in Germany in 2020, although less than in many neighbouring economies (Figure 2.1).

The German Government provided strong support to protect jobs and firms in the crisis, notably through fiscal and employment measures, which cushioned the economic downturn. Nevertheless, uncertainty and a drop in demand has had important effects on business investment and exports in key sectors, particularly manufacturing (OECD, 2020[2]). German GDP is forecasted to grow in 2021 and 2022, although further containment measures may bring prolonged uncertainty (OECD, 2020[13]).

Germany entered the COVID-19 pandemic with high employment rates, record low levels of unemployment and important labour shortages (Figure 2.2). The crisis has impacted labour markets across OECD economies differently, and, similar to the global financial crisis, Germany has avoided rapid job losses in the initial stages of the COVID-19 pandemic. An important part of this success can be attributed to the use and extension of the established short-term work scheme (Kurzarbeit), through which the government subsidises the wages of employees in companies who are in temporary economic distress (OECD, 2020[2]).

Notwithstanding, an excess of 550 000 individuals were unemployed at the end of 2020, compared to the same time in the previous year, according to data by the Federal Employment Agency (Bundesagentur für Arbeit, BA) (Figure 2.3). Additionally, more than 2 million individuals, 4.5% of the working population, remained in short-time work in September 2020 (Bundesagentur für Arbeit, 2020[16]). Short-time work schemes aim at job retention, they moderate the increase of unemployment and act as an economic stabiliser by supporting aggregate demand. Data from the global financial crisis suggest that short-term work can help to avoid immediate income losses and well-being costs for individuals as well as to maintain viable job matches (Hijzen and Martin, 2013[17]; Balleer et al., 2016[18]).

The downside of short-term work schemes is that they can impede labour reallocation, reduce the probability that jobseekers find work and slow job creation during the recovery (Hijzen and Martin, 2013[17]; Cahuc, 2019[19]). The COVID-19 crisis is particular in that it hit the economy more broadly and very rapidly, with many firms across different sectors having to reduce their activity or shut down temporarily, irrespective of their pre-crisis performance. As Germany moves in and out of confinement, policy makers are challenged to strike a balance between ensuring job retention and preventing income losses, on the one hand, and avoiding that the scheme subsidises jobs that would either be preserved anyway or that are not viable in the long term, on the other hand (OECD, 2020[2]). In the aftermath of the COVID-19 pandemic, more individuals may become unemployed and need retraining to cope with a changed labour market situation.

Already prior to the COVID-19 pandemic, the German labour market was undergoing structural changes driven by digitalisation, population ageing and the transition to a low-carbon economy. The economic fallout from the COVID pandemic may well accelerate these trends in sectors such as manufacturing, and speed up structural changes in the labour market. As some client-oriented service sectors face longer-term disruption, people currently covered by the short-time work scheme may lose their jobs.

According to the latest projections of the German Ministry of Employment and Social Affairs (BMAS, 2021[20]), approximately 5.3 million jobs will disappear in the next 20 years (until 2040), while 3.6 million new jobs will be created. This takes into account the impact of the COVID-19 pandemic, recent policies on the transition to a low-carbon economy, increasing obstacles to trade and accelerated digitalisation. These estimates signify accelerating structural change compared to previous projections.

Machines are increasingly able to perform tasks previously done by humans, due to technological innovations such as artificial intelligence (AI) or industrial robotics. These innovations change the type of jobs available and how they are performed. While workers may more and more be able to focus on more creative, productive and safe tasks at work, automation will also make some jobs redundant, requiring workers and companies to adjust. According to OECD analysis, 36% of jobs in Germany are at risk of significant change and an additional 18% at a high risk of automation in the next 15 years (Nedelkoska and Quintini, 2018[5]). This is one of the highest overall shares across OECD countries, exceeded only by Japan, Greece, Lithuania, the Slovak Republic and Turkey (Figure 2.4). Across OECD economies, adults with lower skills and education levels will be the most affected by job automation.

National research complements this picture. An analysis by the German Institute for Employment Research (IAB) referring to 2016, for example, suggests that 25% of all jobs are at high risk of automation, that is more than 70% of the tasks in these jobs could already be done by computers or computer-controlled machines (Dengler and Matthes, 2018[21]). This is higher than the OECD estimates on risk of automation and would be equivalent to 8 million existing jobs disappearing through attrition or displacement. The study also finds a higher risk of automation for adults with lower qualification levels, in line with OECD findings. A number of similar studies exist, all estimating the share of jobs at risk of automation to be between 12% and 25% in Germany (Arntz, Gregory and Zierahn, 2019[22]; Effenberger, Garloff and Würzburg, 2018[23]).

While the estimates of the shares of jobs at risk of automation differ, the policy-implications are the same: The key challenge for German policy makers lies in supporting workers who hold these jobs in the transition to the new employment opportunities that emerge in a changing world of work. Continuing education and training is crucial in this endeavour and enables individuals, enterprises and the economy to harness the benefits of digitalisation and automation.

In addition to automation, demographic trends are a key driver of change in the German labour market, with a manifold impact: As smaller cohorts of workers are replacing retiring cohorts, skill shortages are expected to increase, constraining the potential for economic growth in the medium term (BMWi, 2020[24]). Further, individuals will need to maintain and update their skills over longer working lives. At the same time, demographic change will lead to a change in skill needs due to the associated shifts in demand for goods, services and qualified labour – notably health care professionals and personnel in elderly care (OECD, 2019[25]).

In Germany, the working age population is shrinking, despite an increase in birth rates and net migration in recent years. In the coming ten years, the population aged 15-64 is expected to diminish by 4 million people, which is equivalent to 7% of the working age population (Figure 2.5). This decrease in the working age population will not be equal across regions, with Eastern Germany experiencing greater population ageing, Berlin being the exception, and some regions in the South maintaining a stable working population or even experiencing an increase (Zika et al., 2020[26]).

The trend in employment growth until 2020 has not been uniform across all skill levels. Similar to other OECD countries, Germany has experienced a pattern of increasing job polarisation, although less pronounced in comparison (Figure 2.6). New employment opportunities over the past two decades have increasingly required high-skills, while many middle-skill jobs have been replaced and growth in low-skilled occupations was more modest.

These structural changes came with critical skill shortages in the German labour market, as skill supply struggled to keep pace with these changes (see also sub-chapter ‘the qualification and skills of the German adult population’ below). According to a survey conducted by the Association of German Chambers of Industry and Commerce, close to half of German organisations had difficulties recruiting the staff they needed prior to the COVID crisis (DIHK, 2020[28]). This development is also reflected in vacancy data from the Federal Employment Agency (Bundesagentur für Arbeit, 2019[1]). While in 2009, it took 61 days on average to fill a job vacancy, recruitment time was more than double (124 days) in 2019. Recruitment difficulties were most pronounced in the social and health care sector, where 80% of organisations reported a shortage of skilled workers, according to data by the Association of German Chambers of Industry and Commerce.

According to OECD Skills for Jobs data, shortages in Germany are primarily in high-skilled occupations (Figure 2.7). More than 7 in 10 shortage occupations in Germany were high-skilled, one of the highest shares of shortages in high-skilled occupations across all countries analysed. OECD data suggest that there were no significant shortages in low-skilled occupations. In comparison, only 5 in 10 jobs in shortage on average across countries were high-skilled, 4 in 10 were medium-skilled and 1 in 10 were low-skilled.

Looking at specific skills, shortages were particularly severe in computer and electronics, engineering and mathematics (all typical STEM skills), and customer and personal services, according to OECD data (OECD, 2020[2]). German data on occupational shortages complement this picture: Referring to data from 2019, i.e. prior to the COVID-19 crisis, the Skilled Labour Shortage Analysis (Fachkräfteengpassanalyse) of the Federal Public Employment Agency suggests that occupational bottlenecks were largest in medical and care professions, information technology, construction and skilled trades occupations (Bundesagentur für Arbeit, 2019[1]). Along the same lines, the BMAS 2020 skilled labour monitor (Fachkräftemonitoring) finds that bottlenecks will continue to be most pronounced in occupations that require a high degree of ICT-skills, health care professions, skilled trades (such as plumbing, sanitation, heating, air conditioning) and in occupations related to mechatronics and automation technology (BMAS, 2021[20]).

To ensure that past strong social and economic outcomes persist in the future, German policy makers need to pay particular attention to addressing existing skill imbalances, that is, the matching of skill supply and demand. In terms of cognitive skills, Germany’s adult population scores slightly above average in the international skills assessment test PIAAC, which assesses literacy and numeracy skills. At the same time, the proficiency in digital information processing skills is low compared to the OECD average.

The overall qualification levels of the adult population have changed very little in the past 15 years, although most recently, an increasing share of young people has obtained tertiary degrees, instead of vocational upper secondary and post-secondary non-tertiary degrees.

According to data collected by the Survey of Adult Skills (PIAAC), most adults in Germany have medium levels of proficiency in both literacy and numeracy (70% and 66% respectively) (OECD, 2019[30]). Close to 11% display a high proficiency in literacy skills and 14% a high proficiency in numeracy skills. By contrast, 18% of adults have low proficiency in either literacy or numeracy skills in Germany. As in most European OECD countries, a significant minority of Germans have very low proficiency in literacy (3%) and numeracy (5%). This is lower than the average share of adults with low literacy or numeracy in the OECD, however, Germany lags behind countries with comparable education and training systems such as Austria, Denmark or the Netherlands, as well as high-performing countries such as Japan, Finland or the Slovak Republic (Figure 2.8).

The average proficiency in literacy in Germany (score of 270, not displayed in graph) is slightly but significantly above the OECD average (score of 266). The same is true for the average score of German adults on the numeracy dimension (272), compared to 261 across the OECD. In addition to examining differences in average proficiency between countries, it is also useful to explore differences in the distribution of scores within each country. On average among OECD countries, 61 score points separate the 25% of adults who attained the highest and lowest scores in literacy, and 68 score point separate these quartiles on the numeracy dimension.1 In Germany this gap is slightly larger, at 65 score points of difference between the highest and the lowest quartile for literacy scores, and 71 score points for numeracy scores. This means that the skill levels in the German population are more dispersed and that larger differences exist between the highest and the lowest performers compared to the OECD average.

This data is complemented by national-level data on literacy skills, collected by the University of Hamburg. The LEO Survey2 assessed the reading and writing skills of the German-speaking adult population age 18-64 in 2010 and 2018. It measures literacy levels of the adult population, with a particularly differentiated scale for the lower levels of reading and writing proficiency called Alpha-Levels (Grotlüschen et al., 2019[32]). Individuals with low literacy, according to the LEO 2018 survey, comprise the first three Alpha Levels out of a total of four Alpha Levels. They may not be able to read simple written instructions at work and have difficulties to be completely autonomous in various areas of their life.

In 2018, 12% of the German-speaking adult population were found to have a low level of proficiency in reading and writing (Alpha Levels 1-3), which is equivalent to around 6.2 million adults. An additional 21% of the population made frequent spelling errors (Alpha 4). Compared to 2010, the share of adults with the lowest literacy skills (Alpha Level 1) has remained stable, the share scoring at Alpha Level 2 decreased slightly and the one scoring at Alpha Level 3 dropped more strongly by 2 percentage points. The share of adults making frequent spelling errors (Alpha Level 4) decreased more significantly from 26% in 2010 to less than 21% in 2018.

The overall findings of the LEO study are in line with the PIAAC results described above. The LEO study also confirms OECD analysis that individual and parental background play an important role in the development of information-processing skills (OECD, 2013[33]; Grotlüschen et al., 2019[32]).

Germany is known for its highly developed vocational education and training system. More than one in two adults in Germany hold a vocational qualification at ISCED levels 3-4 as their highest educational qualification, compared to around one in four adults across the OECD. This is the second highest share amongst OECD countries and is exceeded only by the Slovak Republic (Figure 2.9). By contrast, a negligible share of adults in Germany holds a general education qualification at ISCED levels 3-4 – 3%, in comparison to 16% of adults on average across OECD countries.

The share of adults with a tertiary qualification in Germany is well below the OECD average (30% vs. 38%). The largest shares of adults with tertiary qualifications can be found in Canada (59%) and Luxembourg (52%). On the other end of the qualification spectrum, the share of adults with no or low qualifications is relatively low in Germany (13%). However, Germany is still far from the best performing OECD countries, such as the Czech Republic and Lithuania, based on this indicator.

Data from the German statistical office show that individuals in older cohorts more frequently hold a vocational degree at ISCED level 3-4 compared to younger cohorts (Figure 2.10). This is the case for both men and women, but the pattern is even more striking for women. Only 40% of women aged 25-29 hold a vocational degree compared to 56% amongst those aged 60-64. There has been a steady decline in the number of people holding a VET degree as their highest qualification over the past decade in Germany.

By contrast, the share of adults with tertiary degrees (ISCED 5-8) has increased steadily over the past decade, predominately driven by higher shares of younger people and especially young women completing these degrees. More individuals in younger age groups hold tertiary degrees compared to older cohorts (Figure 2.10). An exception are individuals in the age group 25-29, many of whom are still in education.

The share of adults with neither vocational nor higher degree in Germany is roughly the same in all age groups (Figure 2.10). Among the group of 25 to 29-year-olds, 10% are still in education. Younger men less often hold a degree compared to older men (not displayed in graph). For women, in contrast, the share without a vocational or higher degree is almost the same for all age groups, except for the youngest many of whom may still be in education. Over the past decade, the share of people without any vocational or higher qualification decreased, mainly because of higher shares of women graduating with such degrees.

As described previously, STEM skills are one of the key skills in shortage in the German labour market. According to recent OECD analysis, the share of STEM tertiary graduates lags behind other leading OECD countries, in particular among women (Figure 2.11, Panel A). There is less than one female graduate in STEM-related subjects for two male graduates. Equally, VET graduates predominantly choose occupations that are not STEM or ICT-related (Figure 2.11, Panel B, C), although sufficient graduates in these fields of study are essential to tackle shortages. In the meantime, shortages will also need to be filled by upskilling the adult population in the area of STEM and ICT.

Skill imbalances in the German labour market highlight the strong need for continuing education and training. Existing evidence from international surveys (see Box 2.2) suggests that participation rates in CET in Germany are slightly above the OECD average. It also shows that participation is unequal across socio-demographic groups, more so than in most other OECD countries. Adults with low skills, those on low wages and those working in small and medium enterprises (SMEs) display particularly low participation rates. Many OECD countries with high-performing CET systems, such New Zealand, Norway and Sweden, have substantially higher participation levels and lower inequalities.

In Germany, 52% of the population aged 25-64 took part in formal or non-formal learning in 2016, according to the Adult Education Survey (AES) (Figure 2.12). This places Germany slightly above the average of OECD countries for which these data are available. However, it lags behind other OECD countries with similar skill development systems, i.e. Austria (60% learning participation), the Netherlands (64%) and Switzerland (69%).3. Other countries with considerably higher participation rates than Germany include Norway (60%) and Sweden (64%). These AES data refer to participation in all types of formal and non-formal learning, irrespective of whether or not its purpose is job-related.

Data from the 2018 AES indicate that German participation may be increasing, continuing an upward trajectory of the last decade. Making reference to the wider adult population aged 18-64, the German AES suggests that 54% of adults took part in CET in 2018, up from 50% only two years earlier (Statistisches Bundesamt, 2021[38]; BMBF, 2019[39]). This continues, and indeed steepens, the country’s upward trajectory when it comes to CET participation, which previously increased from 44% in 2007. It remains to be seen if this recent increase improves Germany’s performance vis-à-vis other OECD countries. Over the past decade, other countries for which AES data are available have seen similar increases in learning participation than Germany.

It should also be noted that other data sources, e.g. the European Labour Force Survey (LFS), cast doubts on the success story of increasing participation. According to the LFS, which measures participation in education and training in the 4 weeks prior to the survey (Box 2.2), the participation rate in Germany has essentially flat lined since 2007 and hovers around 8% of 25-64 year-olds having participated in education and training in the past 4 weeks. By contrast, the participation rate in European OECD countries increased by 3 percentage points in the same period, from 8% in 2007 to 11% in 2019.4

The majority of learning takes place non-formally in Germany (Figure 2.13); in 2016, 50% of adults took part in courses of short duration and/or not leading to a certification in a given year, according to AES data. This compares to 44% of adults who engaged in informal learning, such as learning from their peers or learning by doing. A small share of adults took part in formal learning (less than 4%), according to the AES. German adults participate less frequently in formal and informal learning and more frequently in non-formal learning than adults in other European OECD countries.

In Germany, the share of adults who participates in formal education decreased from 5% in 2007 to less than 4% in 2016. The share of adults who engages in informal learning decreased from 52% to 44%. At the same time, participation in non-formal learning increased from 43% to 50%.

Adults who do participate in learning take part in 124 hours of formal and non-formal learning on average in Germany (Figure 2.14). This is equivalent to just over 3 weeks of full-time education or training per year. Among those countries for which data is available, adult learners in Slovenia take part in most learning in a given year: 180 hours or the equivalent of 4.5 weeks of full-time education or training.

However, when considering only formal education and training, adults in Germany display the highest learning intensity of all countries by a vast margin. On average, these adults take part in 872 hours of instructed learning per year, which is equivalent to over 5 months of full-time study. The country with the second highest learning intensity for formal learning is Portugal, where adult learners take part in 653 hours (equivalent to 4 months) on average. This statistic may be explained by the fact that formal vocational CET in Germany, has a long duration by law and that the provision of courses to obtain partial formal qualifications in Germany could be developed further (see also Chapter 4).

The OECD’s Priorities for Adult Learning dashboard looks at differences in CET participation of different socio-economic groups across OECD countries.5 It shows that, on aggregate, Germany has some of the largest inequalities in CET participation in the OECD, exceeded only by Chile, the Netherlands and the Slovak Republic (OECD, 2019[36]). In Germany, adults with low levels of basic skills or qualifications, low wages and the long-term unemployed have particularly low participation rates, compared to their higher skilled, higher earning or employed peers.

According to PIAAC data, only 23% of adults with low levels of basic skills in Germany participate in job-related learning compared with 51% of those with medium to high levels of basic skills. Similarly, only 17% of adults with low qualification levels participate in job-related learning compared to 48% with medium or higher qualifications (see Chapter 6). Participation gaps between the long-term unemployed and the employed (26% vs. 53%) and between low-wage and medium to high-wage earners (34% vs. 59%) are similarly large (Figure 2.15).

The OECD Priorities for Adult Learning dashboard uses PIAAC data to ensure comparability across all OECD countries. Using these data also facilitates the analysis of gaps by actual skill and wage level, as well as risk of automation, which cannot be achieved with AES data. However, the AES allows looking at changes over time in the participation gaps. Analysis of the 2018 AES in Germany shows that overall participation gaps remain stark in Germany (BMBF, 2019[39]):

  • While the participation gap between employed and unemployed individuals stayed relatively stable between 2012 and 2016, it narrowed in 2018. This was due to a sharp increase in education and training participation of the unemployed from 27% in 2016 to 49% in 2018, which is being attributed to the large inflow of migrants in 2015/2016 and the subsequent participation of this group in integration courses. Whether this development has reduced the inequalities of participation vis-à-vis other countries remains to be seen.

  • The gap between those with low and high qualification levels has remained relatively stable since 2012, with some fluctuations between the years and a slight overall increase in participation for both groups (32% to 39% and 64% to 69% respectively).

  • Participation gaps between adults in SMEs and larger enterprises have slightly decreased, primarily driven by sharp increases in participation by adults who work in micro-enterprises with up to 19 employees.

Across OECD countries, there is limited evidence on the impact of CET participation on labour market and social outcomes, with Germany being no exception. There are some international surveys, which collect comparative data on subjective outcomes. For example, PIAAC collects information on whether participation has been useful for the participant’s job or if they make use of the acquired skills. Additionally, the AES collects data on self-reported employment outcomes following participation, such as promotions or wage raises. However, evaluation studies assessing the causal impact of CET participation on individual outcomes are typically limited to CET programmes in the context of Active Labour Market Policy (ALMP).

According to PIAAC data, the majority (73%) of adults in OECD countries participates in CET for job-related reasons. These reasons include increasing their chances of finding a job or changing jobs, getting a promotion or wage increase, or performing better in their present job. AES data show that across European OECD countries, 67% of learners thought that participation in non-formal job-related CET had helped them achieve such outcomes (Figure 2.16). The effect of CET is largest in Italy, Hungary, Portugal and Slovenia, where more than 80% of learners self-reported positive outcomes following participation. By contrast, in Germany only 50% of learners thought that CET had had a positive effect on their employment outcomes.6 Only the Netherlands and Turkey record lower shares.

These self-reported employment outcomes likely capture some differences in the quality and relevance of the training received. However, they also reflect differences in labour market conditions, institutions and other contextual factors, such as cultural attitudes. For example, in economies with tight labour markets, i.e. where job-seekers have limited competition for jobs, training participation typically has a much smaller impact on individual employment outcomes than in loose labour markets, where job competition is fierce (OECD, 2019[36]). Further, the data presented relate to non-formal job-related CET only, and may therefore describe the value that different labour markets place on different types of learning. In Germany’s strongly qualification-focused labour market, formal learning opportunities might be much more important in leading to better employment outcomes. Unfortunately, no data on self-reported employment outcomes are available for participation in formal CET.

While the above reported AES data allow for broad comparisons across countries, better insights into the effectiveness of specific CET provision can be gleaned through evaluations from Germany. The bulk of these evaluations are in regard to ALMPs and relate to the impact of CET on unemployed individuals. Key findings include:

  • Individuals can achieve monetary returns through participation in in-service training, but effects vary widely based on individual characteristics, as well as type of training. These effects are also moderated by the job context, such as the sector or occupation individuals find themselves in (Ehlert, 2017[11]).

  • Positive effects of participation in CET are larger for some groups, including women, low-skilled adults, those with a migrant background, and those retraining for specific occupations such as health and care professions (Doerr et al., 2014[10]; Kruppe and Lang, 2015[12]; Autorengruppe Bildungsberichterstattung, 2020[34]).

  • While short-term effects of training participation can be negative due to lock-in effects, positive effects materialise in the long term (Bläsche et al., 2017[7]). Participants in vocational retraining (Umschulung), for example, are significantly more likely than a comparison group to be employed 2.5 to 3 years after the starting training (Bernhard, Lang and Kruppe, 2017[9]).

  • Participation in CET programmes of longer duration, typically vocational retraining, has bigger effects on labour market outcomes than participation in shorter courses. One study finds that participants in CET programmes of one-year duration or more experience positive employment and wage effects up to twice as large as those in shorter CET programmes (Bernhard, 2016[8]).

  • It should not be concluded that training has a causal impact on promotion or monetary returns, as training is often obligatory for those already selected for promotion (Autorengruppe Bildungsberichterstattung, 2020[34]). In fact, one study finds that participation in in-company training has a stabilising – rather than accelerating – effect on careers, preventing downgrading but also promotions (Ebner and Ehlert, 2018[40]).

The benefits of participation in CET for individuals and societies are not limited to the labour market. A vast body of international research supports its positive effects on health and well-being, social and political attitudes and engagement, and social capital (Balatti and Falk, 2002[41]; Dee, 2004[42]; OECD, 2020[31]; Preston and Feinstein, 2004[43]; Schuller and Desjardins, 2010[44]). A recent study for Germany finds that participation in CET leads to increased engagement in politics, civil society and cultural activities (Ruhose, Thomsen and Weilage, 2019[45]).


[22] Arntz, M., T. Gregory and U. Zierahn (2019), Digitalization and the Future of Work: Macroeconomic Consequences, IZA DP No. 12428, http://ftp.iza.org/dp12428.pdf.

[34] Autorengruppe Bildungsberichterstattung (2020), Bildung in Deutschland 2020. Ein indikatorengestützter Bericht mit einer Analyse zu Bildung in einer digitalisierten Welt., Autorengruppe Bildungsberichterstattung, Bielefeld, https://www.bildungsbericht.de/de/bildungsberichte-seit-2006/bildungsbericht-2020/pdf-dateien-2020/bildungsbericht-2020-barrierefrei.pdf.

[41] Balatti, J. and I. Falk (2002), Socioeconomic contributions of adult learning to community: A social capital perspective, http://dx.doi.org/10.1177/074171302400448618.

[18] Balleer, A. et al. (2016), “Does short-time work save jobs? A business cycle analysis”, European Economic Review, Vol. 84, pp. 99-122, http://dx.doi.org/10.1016/j.euroecorev.2015.05.007.

[8] Bernhard, S. (2016), “Long-term Effects of Vocational Training for Unemployment Benefit Recipients in Germany”, Sozialer Fortschritt, Vol. 65/7, pp. 153-161, http://dx.doi.org/10.3790/sfo.65.7.153.

[9] Bernhard, S., J. Lang and T. Kruppe (2017), “Langfristige Wirkungen von geförderter beruflicher Weiterbildung”, in Möller, J. and U. Walwei (eds.), Arbeitsmarkt Kompakt. Analysen, Daten, Fakten, Institut für Arbeitsmarkt- und Berufsforschung, Nürnberg.

[7] Bläsche, A. et al. (2017), “Qualitätsoffensive strukturierte Weiterbildung in Deutschland”, No. 025, Hans-Böckler-Stiftung, Düsseldorf, https://www.boeckler.de/pdf/p_fofoe_WP_025_2017.pdf.

[20] BMAS (2021), Fachkräftemonitoring, Institut für Arbeitsmarkt- und Berufsforschung (IAB), Bundesinstitut für Berufsbildung (BIBB) und Gesellschaft für Wirtschaftliche Strukturforschung (GWS), https://www.bmas.de/DE/Arbeit/Fachkraeftesicherung-und-Integration/Fachkraeftemonitoring/fachkraeftemonitoring-art.html.

[6] BMAS et al. (2019), Nationale Weiterbildungsstrategie, http://doku.iab.de/kurzber/2019/kb0819.pdf.

[39] BMBF (2019), Weiterbildungsverhalten in Deutschland 2018. Ergebnisse des Adult Education Survey-AES-Trendbericht, Bundesministerium für Bildung und Forschung (BMBF), Berlin, https://www.bmbf.de/upload_filestore/pub/Weiterbildungsverhalten_in_Deutschland_2018.pdf.

[24] BMWi (2020), Herbstprojektion 2020. Nach schneller Erholung setzt sich der Aufholprozess langsam fort, Bundesministerium für Wirtschaft und Energie, Berlin, https://www.bmwi.de/Redaktion/DE/Publikationen/Schlaglichter-der-Wirtschaftspolitik/schlaglichter-der-wirtschaftspolitik-11-2020.pdf?__blob=publicationFile&v=36 (accessed on 16 December 2020).

[15] Bundesagentur für Arbeit (2020), Auswirkungen der Corona-Krise auf den Arbeits-und Ausbildungsmarkt, Bundesagentur für Arbeit, https://statistik.arbeitsagentur.de/Statistikdaten/Detail/202011/arbeitsmarktberichte/am-kompakt-corona/am-kompakt-corona-d-0-202011-pdf.pdf?__blob=publicationFile&v=1 (accessed on 15 December 2020).

[16] Bundesagentur für Arbeit (2020), Realisierte Kurzarbeit (hochgerechnet) Monatszahlen, Bundesagentur für Arbeit;, https://statistik.arbeitsagentur.de/SiteGlobals/Forms/Suche/Einzelheftsuche_Formular.html?nn=1524090&topic_f=kurzarbeit-hr (accessed on 15 December 2020).

[1] Bundesagentur für Arbeit (2019), Fachkräfteengpassanalyse, Bundesagentur für Arbeit, Nürnberg, https://statistik.arbeitsagentur.de/Statistikdaten/Detail/201912/arbeitsmarktberichte/fk-engpassanalyse/fk-engpassanalyse-d-0-201912-pdf.pdf?__blob=publicationFile (accessed on 4 March 2021).

[27] Bundesamt, S. (2019), Bevölkerung im Wandel. Annahmen und Ergebnisse der 14. koordinierten Bevölkerungsvorausberechnung, https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Bevoelkerung/Bevoelkerungsvorausberechnung/_inhalt.html#sprg233474 (accessed on 15 December 2020).

[19] Cahuc, P. (2019), “Short-time work compensations and employment”, IZA World of Labor, http://dx.doi.org/10.15185/izawol.11.

[42] Dee, T. (2004), “Are there civic returns to education?”, Journal of Public Economics, Vol. 88/9-10, pp. 1697-1720, http://dx.doi.org/10.1016/j.jpubeco.2003.11.002.

[21] Dengler, K. and B. Matthes (2018), Substituierbarkeitspotenziale von Berufen: Wenige Berufsbilder halten mit der Digitalisierung Schritt, IAB, http://doku.iab.de/kurzber/2018/kb0418.pdf (accessed on 15 December 2020).

[28] DIHK (2020), Fachkräftesuche bleibt Herausforderung. DIHK-Report Fachkräfte 2020, Deutscher Industrie- und Handelskammertag e.V. (DIHK), Berlin, https://www.dihk.de/resource/blob/17812/f1dc195354b02c9dab098fee4fbc137a/dihk-report-fachkraefte-2020-data.pdf (accessed on 16 December 2020).

[10] Doerr, A. et al. (2014), “Employment and Earnings Effects of Awarding Training Vouchers in Germany”, Discussion Paper, No. 14-065, IZA, http://ftp.zew.de/pub/zew-docs/dp/dp14065.pdf.

[40] Ebner, C. and M. Ehlert (2018), “Does Further Education lead to Career Advancement? Non-Formal Further Training and Labour Market Mobility in Germany”, Kolner Zeitschrift fur Soziologie und Sozialpsychologie, Vol. 70/2, pp. 213-235, http://dx.doi.org/10.1007/s11577-018-0518-x.

[23] Effenberger, A., A. Garloff and H. Würzburg (2018), Beschäftigungseffekte der Digitalisierung-Forschungsansätze und Ergebnisse. Diskussionspapier Nr. 7, Bundesministerium für Wirtschaft und Energie, https://www.bmwi.de/Redaktion/DE/Downloads/Diskussionspapiere/20180621-diskussionspapier-beschaeftigungseffekte-der-digitalisierung.pdf?__blob=publicationFile&v=8 (accessed on 4 March 2021).

[11] Ehlert, M. (2017), “Who Benefits from Training Courses in Germany? Monetary Returns to Non-formal Further Education on a Segmented Labour Market”, European Sociological Review, Vol. 33/3, pp. 436-448, http://dx.doi.org/10.1093/esr/jcx042.

[35] Goglio, V. and E. Meroni (2014), Adult participation in lifelong learning. The impact of using a 12-months or 4-weeks reference period, Joint Research Center of the European Commission, http://dx.doi.org/10.2788/43117.

[32] Grotlüschen, A. et al. (2019), LEO 2018 Living with low literacy, https://leo.blogs.uni-hamburg.de/wp-content/uploads/2019/07/LEO_2018_Living_with_Low_Literacy.pdf.

[17] Hijzen, A. and S. Martin (2013), “The Role of Short-Time Work Schemes during the Global Financial Crisis and Early Recovery: A Cross-Country Analysis”, Discussion Paper, No. 7291, IZA, http://ftp.iza.org/dp7291.pdf.

[12] Kruppe, T. and J. Lang (2015), “Weiterbildungen mit Berufsabschluss Arbeitslose profitieren von Qualifizierungen”, IAB Kurzbericht, No. 22/2015, IAB, http://doku.iab.de/kurzber/2015/kb2215.pdf.

[5] Nedelkoska, L. and G. Quintini (2018), “Automation, skills use and training”, OECD Social, Employment and Migration Working Papers, No. 202, OECD Publishing, Paris, https://dx.doi.org/10.1787/2e2f4eea-en.

[4] OECD (2021), OECD Economic Outlook, Interim Report. Strengthening the recovery. The need for speed, OECD Publishing, Paris, https://doi.org/10.1787/34bfd999-en.

[14] OECD (2021), Real GDP forecast, https://dx.doi.org/10.1787/1f84150b-en (accessed on 8 March 2021).

[31] OECD (2020), Education at a Glance 2020: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/69096873-en.

[13] OECD (2020), OECD Economic Outlook, Volume 2020 Issue 2, OECD Publishing, Paris, https://dx.doi.org/10.1787/39a88ab1-en.

[2] OECD (2020), OECD Economic Surveys: Germany 2020, OECD Publishing, Paris, https://dx.doi.org/10.1787/91973c69-en.

[36] OECD (2019), Getting Skills Right: Future-Ready Adult Learning Systems, OECD Publishing, Paris, https://doi.org/10.1787/9789264311756-en.

[25] OECD (2019), OECD Employment Outlook 2019: The Future of Work, OECD Publishing, Paris, https://doi.org/10.1787/9ee00155-en.

[30] OECD (2019), Skills Matter: Additional Results from the Survey of Adult Skills, OECD Publishing, Paris, https://doi.org/10.1787/1f029d8f-en.

[3] OECD (2018), OECD Economic Surveys: Germany 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/eco_surveys-deu-2018-en.

[29] OECD (2017), How technology and globalisation are transforming the labour market, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2017-7-en.

[33] OECD (2013), Germany. Country Note-Survey of Adult Skills, OECD Publishing, Paris, https://www.oecd.org/skills/piaac/Country%20note%20-%20Germany.pdf.

[43] Preston, J. and L. Feinstein (2004), Adult Education and Attitude Change, Centre for Research on the Wider Benefits of Learning, https://discovery.ucl.ac.uk/id/eprint/10015019/1/WBLResRep11.pdf.

[45] Ruhose, J., S. Thomsen and I. Weilage (2019), “The benefits of adult learning: Work-related training, social capital, and earnings”, Economics of Education Review, Vol. 72, pp. 166-186, http://dx.doi.org/10.1016/j.econedurev.2019.05.010.

[44] Schuller, T. and R. Desjardins (2010), “Wider benefits of adult education”, in International Encyclopedia of Education, Elsevier Ltd, http://dx.doi.org/10.1016/B978-0-08-044894-7.00045-2.

[38] Statistisches Bundesamt (2021), Weiterbildung - 2020, Statistisches Bundesamt, https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Bildung-Forschung-Kultur/Weiterbildung/Publikationen/Downloads-Weiterbildung/berufliche-weiterbildung-5215001207004.pdf?__blob=publicationFile (accessed on 3 April 2021).

[37] Widany, S. et al. (2019), “The Quality of Data on Participation in Adult Education and Training. An Analysis of Varying Participation Rates and Patterns Under Consideration of Survey Design and Measurement Effects”, Frontiers in Sociology, Vol. 4, p. 71, https://doi.org/10.3389/fsoc.2019.00071.

[26] Zika, G. et al. (2020), Digitaler und demografischer Wandel wirken sich regional sehr unterschiedlich auf den künftigen Arbeitskräftebedarf aus, IAB, https://www.iab-forum.de/digitaler-und-demografischer-wandel-wirken-sich-regional-sehr-unterschiedlich-auf-den-kuenftigen-arbeitskraeftebedarf-aus/?pdf=15850 (accessed on 16 December 2020).


← 1. This measure is known as the interquartile range.

← 2. LEO 2018 is based on a random sample of adults living in private households in Germany, aged between 18 and 64. The net sample size comprised 6 681 people. It was supplemented with an additional random sample of 511 people from the lower levels of education. People were only included in the survey if their command of German was sufficient to follow an approximately one-hour-long interview. The sample was weighted based on key socio-demographic data taken from the German Microcensus. After answering a set of standardised questions about various aspects of their background, the interviewees then completed a skills test comprising reading and writing exercises. All 7 192 subjects were given an initial assessment test. The interviews were carried out by the polling institution Kantar Public as computer-assisted personal interviews (CAPIs) (Grotlüschen et al., 2019[32]).

← 3. Germany may be catching up with these countries, but only the next round of international data collection can confirm this. Data collection for the Adult Education Survey (AES) and the Survey of Adult Skills (PIAAC) is due to take place in 2022/2023.

← 4. Eurostat, trng_lfse_01 indicator, accessed 11 June 2020.

← 5. The analysis included participation differences between older (55 years and above) and prime age (25-54) individuals; women and men; low-skilled and medium/high skilled adults (according to PIAAC); low-wage and medium/high wage workers; unemployed and employed individuals; long-term unemployed and employed individuals; temporary and permanent workers; workers in SMEs and workers in larger firms.

← 6. Positive employment outcomes included: i) getting a (new) job, ii) higher salary/wages, iii) promotion in the job, iv) new tasks, and/or v) better performance in present job.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2021

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.