2. Why is adult learning important in Latin America?

Globalisation and rapid technological change, together with demographic developments have considerably changed skill demands and supply in many OECD countries and partner countries and economies in the last decade. These trends are expected to continue in the coming years at an increasing pace. While in Latin America some of these structural megatrends are not yet fully realised – the pace of digitalisation, for instance, has been slower in the region than in the most advanced OECD economies - other structural factors such as demographic dynamics and informality are having an impact on skills imbalances. This chapter addresses these issues providing evidence on the main challenges facing Latin America and highlighting the importance of developing an adequately skilled workforce.

    

Summary of the main results

Technological progress is profoundly transforming the world of work, posing threats and opportunities that Latin American and the Caribbean (LAC) countries will face in the near future

• Recent OECD estimates suggest that on average across Chile, Ecuador, Mexico and Peru 24% of jobs will face a high risk of automation. This figure is around 9 percentage points higher than across OECD countries. An additional 35% of jobs in the region are likely to experience significant changes in the tasks that workers carry out daily, a figure that is approximately 5 percentage points higher than the OECD average.

• Employment in potentially automatable sectors is very high in Latin America. In Brazil, Chile, Colombia, Costa Rica, Mexico and Peru, the share of workers employed in highly automatable sectors such as manufacturing and agriculture ranges in between 30 and 40% of total employment.

• While many jobs could be “technically automatable”, automation may not be yet economically attractive or viable for many firms in LAC economies as costly investments in advanced technology are usually out of reach to most entrepreneurs, especially for small and medium-sized enterprises (SMEs) in the region. This leaves room for policy makers in LAC countries to anticipate the potential change and react accordingly by developing today the skills that the workforce will need in the future.

Demographic dynamics are also crucial for skills development

• Population ageing is likely to put considerable pressure on education and training systems as it increases the need for individuals to maintain and update their skills over the life-course in the context of longer working lives. In Latin America, a large proportion of the region’s population is still young (aged between 15 and 29), creating a demographic dividend. This window of opportunity, however, may be closing soon as the absolute number of young people is expected to fall after 2020.

• Brazil and Chile are projected to have a higher old-age dependency ratio than most OECD and G20 countries by 2075. This puts considerable pressure on education and training systems and increasing the need for individuals to maintain and update their skills over the life-course. The retirement of large cohorts can lead to significant shortages of qualified labour in some countries; a gap that can only be filled through the constant training of the existing workforce and the creation of effective adult learning systems.

• In Latin America, the weak participation of older workers in adult training along with the foreseen introduction of new technologies requiring the re-skilling and up-skilling of the workforce are going to put increasing pressure on labour markets and, potentially, magnify economy-wide skill and social divides.

Job quality and informality are major concerns in Latin America

• Workers in Latin America also tend to be more vulnerable to labour market risks than their counterparts in more advanced economies. High and persistent informal employment represents also a major policy concern, this latter ranging in between 9% in Uruguay and 65% in Guatemala.

• Workers in the informal sector have limited access to training opportunities, a situation that further exacerbates inequalities and penalise them when acquiring skills relative to workers in formal jobs.

• In all LAC countries, the probability of being employed in the informal sector decreases dramatically with the level of education of a worker. In Colombia, for instance, recent evidence points to skill upgrading as a major driver of the reduction in informality from 70% in 2007 to 62% in 2017.

• The low skills of many LAC employers and, in particular, the shortage of educated entrepreneurs in the informal sector, play a fundamental role in driving informality, much more important than the lack of demand in the economy. Much needs to be done to strengthen the skills of employers in the informal economy in Latin America to ensure that they play a key role in developing an effective adult learning response to the challenges of the future.

• In Guatemala and Brazil respectively only 8% and 13% of managers in the informal sector have a college degree. Boosting the number of well-educated entrepreneurs could contribute significantly to reduce the number of informal firms.

Despite substantial progress in recent years, the educational attainment and the skill levels of Latin America’s population remain low by international standards

• Improving the coverage and the quality of LAC education systems is key but challenges lie ahead: only 64.2% of individuals aged 25-34 years old in the region has completed secondary education. This figure is about 20 percentage points lower than the OECD average (84%).

• Around 24% of 25-34 year-olds holds a tertiary degree (43% on average across OECD countries), suggesting that both enrolment rates of secondary and tertiary education need to be boosted to promote the further upskilling of the workforce.

The relatively poor quality of LAC’s education and training systems also represents a key challenge in many LAC countries

• The OECD's Programme for International Student Assessment (PISA) measures 15-year-olds' proficiency in reading, mathematics and science to meet real-life challenges. Across many LAC countries, PISA scores are substantially below the OECD average and that of other emerging economies. The socio-economic background of students plays a prominent role in students’ scores in LAC, much more than on average across OECD countries.

• In Peru, the socio-economic status of students explains more than 20% of the variability in PISA scores in science. Socio-economic status plays a great role also in Chile, Costa Rica and Colombia where its impact is well above that observed on average across OECD countries.

• Progress has been made in the last decade in many LAC countries to ensure that education and training systems are more inclusive and to boost their quality for students of all backgrounds. Some countries, however, are still struggling. In Chile and Colombia, for instance, the importance of socio-economic background in explaining students’ performance has increased considerably in between 2006 and 2015.

Evidence highlights how LAC countries need to boost participation in high-quality training and, especially, in adult learning where LAC are lagging particularly behind

• The Survey of Adult Skills (PIAAC) shows that in many countries in the region (Chile, Ecuador, Mexico and Peru), on average, up to 60% of adults have low levels of skills – both in terms of literacy and numeracy proficiency. This calls for immediate action to boost the participation in up-skilling and learning activities for many adults in the region.

A large share of employment in LAC economies is found in low to medium-tech sectors, which is reflected in the demand for skills in the labour market

• In countries such as Argentina, Brazil, Chile, Mexico or Peru the shortages of highly skilled professionals is significantly lower than the OECD average, with less than 2 out of 10 jobs in shortage being “high-skilled” and the majority of jobs in demand being found, instead, in medium to low-skilled occupations.

• Evidence suggests that occupational shortages in LAC countries are mostly concentrated in low to medium-skill jobs in the agriculture and manufacturing sector and that. Certain shortages of skilled workers also emerge, but the demand is relatively weak.

Occupational and skill shortages are rather heterogeneous across countries and call for tailored policy intervention, boosting adult learning participation in the region

• In Chile, labourers in the mining and construction sector have experienced robust wage growth signalling a strong demand for low-medium workers in the country. At the same time, business and administration professionals have also seen a sharp increase in both their hours worked and wages, implying a sustained demand for these medium-skilled professionals.

• In Brazil, where the largest demand is in middle-skill professionals, health associate professional and personal care workers have been on the rise in recent years, while wages have been declining in traditionally high-skilled occupations such as science and engineering professionals, signalling a decrease in demand.

• Technical skills are in strong demand in both Mexico and Chile while Argentina’s shortages are in social as well as in basic and complex problem solving skills. Surpluses of complex problem solving skills are instead found in relatively less developed economies such as Peru where demand for high-level skills is weak.

Factors affecting and reshaping skill demands in Latin America

Technology and demographics are shaping skill demands and mismatches in Latin America

The risk of automation in Latin America is potentially higher than across other OECD countries

Two major trends will affect the future of work globally and in Latin America: technological change and population ageing. Technological progress is already profoundly transforming the world of work and, as such, the skills demanded by employers around the world. While, on the one hand, new technologies have the potential to free up workers’ time to do carry out productive and less routine-intensive tasks, on the other hand, they will likely change the nature of many jobs and the skills demanded to workers to perform them.

One particular aspect of technological change that has recently captured the attention of policy makers around the globe is that of automation. While certain degree of uncertainty in the estimation of the risk of automation remains, recent estimates (OECD, 2019[1]) report that 14% of jobs across OECD countries participating in the Survey of Adult Skills (PIAAC) are potentially highly automatable (i.e. probability of automation of over 70%). This is equivalent to over 66 million workers in the 32 countries covered by the study. Moreover, another 32% of jobs have a risk of between 50% and 70% pointing to the possibility of significant change in the way these jobs will be carried out as a result of automation.

Several dimensions are likely to affect the penetration of technology in a country and, as such, the risk of job automation. These factors include regulations on workers dismissal, unit labour costs or social preferences with regard to automation. Productive structure is likely to play an important role in the potential impact of technology and automation on job creation or destruction. Recent estimates confirms, in fact, that about 30% of the cross-country variance in the risk of automation is explained by cross-country differences in the structure of economic sectors (OECD, 2019[1]) and several analysts have found positive relationships between automation and job growth in the service sector but, in parallel, job destruction primarily in manufacturing.

While the literature on the impact of automation on jobs has largely focused on advanced economies, it is important to notice that emerging economies, such as LAC countries, start from very different initial conditions than more developed OECD countries. These include a different occupational mix, higher costs of information and communication technologies (ICT) capital, and greater skills shortages (Maloney and Molina, 2016[2]).

Based on their current stage of development, a large share of employment in emerging economies is found in sectors such as manufacturing and agriculture that could potentially face high risk of automation when new technologies are introduced in production activities.

In several LAC countries, for instance Brazil, Chile, Colombia, Costa Rica, Mexico and Peru, the share of workers employed in highly automatable sectors such as manufacturing and agriculture ranges in between 30% and 40% of total employment (Figure 2.1).

Figure 2.1. Agriculture and industry sectors make up a large share of employment in LAC

Employment in industry and agriculture (% of total employment), 2017

Source: The World Bank (2018[3]), World Development Indicators, http://datatopics.worldbank.org/world-development-indicators/.

Recent OECD estimates based on Nedelkoska and Quintini (2018[4]) suggests, in fact, that in Chile, Ecuador, Mexico and Peru,1 an average of 23.6% of jobs face a high risk of automation. This is around 9 percentage points higher than across OECD countries, where more than 15% of jobs are, on average, facing a high risk of automation. Furthermore, in LAC countries for which information is available, an additional 35.1% of jobs are likely to experience significant changes in the tasks that workers carry out daily, a figure that is approximately 5 percentage points higher than the OECD average (Figure 2.2).

Similar results are found in other recent studies (The World Bank, 2016[5]; Weller, Gontero and Campbell, 2019[6]; McKinsey Global Institute, 2017[7]).2

Figure 2.2. A large share of jobs are at high risk of automation or significant change

Percentage of jobs at significant or high risk of automation

Note: Jobs are at high risk of automation if their likelihood to be automated is at least 70%. Jobs at risk of significant change are those with the likelihood of being automated estimated at between 50% and 70%. The values for “OECD” are weighted averages.

Source: OECD calculations based on estimates from Nedelkoska, L. and G. Quintini (2018[4]), “Automation, skills use and training”, OECD Social, Employment and Migration Working Papers, No. 202, https://doi.org/10.1787/2e2f4eea-en.

Certainly, while many jobs may be “technically automatable”, automation may not be yet economically attractive or viable for many firms in LAC economies as costly investments in sophisticated technology are usually out of reach to most entrepreneurs, especially in small and medium-sized enterprises (SMEs). In Latin America, SMEs account for more than 80% of employment and more than 90% of firms (OECD/CAF/UN ECLAC, 2016[8]) and such massive presence of SMEs has, until now, limited the penetration of high-end automating technologies and slowed down the potential risks for low skilled workers stemming from technological change. As of now, the incentives to adopt automation-intensive technologies are hampered by the relative abundance of cheap unskilled labour and by the lack of skilled workforce to benefit from digital technologies. The situation, however, is likely to change in the near future as automation becomes relatively cheaper and more efficient (Box 2.1).

Box 2.1. When will automation invest Latin America and other emerging economies?

It could take considerable time for the productivity gains from new technologies to be realised in emerging economies such as LAC countries. The past has seen unrealistic enthusiasm regarding timescales for the delivery of some industrial technologies. In some cases, as with nanotechnology, this reflects miscalculation of the technical challenges. And many technologies, such as big data and the Internet of Things, have developed in a wave-like pattern, with periods of rapid inventive activity coming after periods of slower activity and vice versa (OECD, 2015[9]). In terms of adoption, advanced ICTs remain below potential across OECD countries. By one estimate “the full shift to Industry 4.0 could take 20 years” (Lorenz et al., 2015[10]). The mere availability of a technology is not sufficient for its uptake and successful use and realising the benefits of a technology often requires that it be bundled with investments in complementary intangible assets, such as new skills and organisational forms, and that better adapted business models are invented that channel income to innovators.

Successful absorption of new technologies in emerging and developing countries could help to achieve productivity, structural transformation and environmental goals. Indeed, some new production technologies are well suited to economic conditions in many developing countries. For example, certain state-of-the-art robots are relatively inexpensive and do not require highly skilled operators and low-cost drones could make some agricultural processes more efficient. With improved channels of knowledge diffusion, such as the Internet, opportunities for technological “leapfrogging” could emerge, particularly in large developing economies but the learning curve needed to use new technologies is clearly a challenge for companies in many developing economies.

Source: OECD (2017[11]), The Next Production Revolution: Implications for Governments and Business, https://doi.org/10.1787/9789264271036-en.

Demographic changes and population ageing are also affecting skills needs

The world’s population is ageing. In Latin America, a large proportion of the region’s population is, however, still young (aged between 15 and 29), creating a demographic dividend which has the potential for dramatic increases in productivity, savings, and economic growth. These windows of opportunity, however, may be closing soon as the absolute number of young people is expected to fall after 2020 and the relative share of young people over population aged 30 and more will continue the decline that has already started a few decades ago.

In Latin America, demographic conditions are therefore expected to shift towards a less favorable scenario entailing mounting pressure on public budgets and pension systems (OECD/CAF/UN ECLAC, 2016[8]).3 In particular, the low dependency ratios4 that LAC countries are experiencing nowadays are set to increase strongly in the next decades putting pressure on younger cohorts. To give an example, countries such as Brazil or Chile are projected to have a higher dependency rate than most OECD and G20 countries by 2075 (Table 2.1), posing questions about the sustainability of pension systems, productivity and economic growth.

Table 2.1. Old-age dependency ratios: Historical and projected values, 1950-2075
	1950	1975	2000	2015
OECD	13.9	19.5	22.5	27.9
Argentina	7.5	14.1	18.6	19.5
Brazil	6.5	8.0	9.3	13.0
Chile	8.6	11.3	13.1	17.0
Mexico	7.9	9.6	10.0	11.4
China (People’s Republic of)	8.5	8.8	11.4	14.5
India	6.4	7.6	8.7	10.0
Indonesia	8.6	7.9	8.7	8.7
Russian Federation	8.7	15.5	20.4	20.7
Saudi Arabia	7.5	7.6	6.1	4.8
South Africa	8.5	8.1	7.8	9.0
EU28	14.7	21.2	24.3	29.9
Note: The demographic old-age dependency ratio is defined as the number of individuals aged 65 and over per 100 people of working age defined as those aged between 20 and 64.
Source: Adapted from United Nations, Department of Economic and Social Affairs, Population Division (2019[12]), World Population Prospects 2019: Highlights (ST/ESA/SER.A/423), https://population.un.org/wpp/Publications/Files/WPP2019_Highlights.pdf.

From a skills development point of view, an ageing population is an often-overlooked driver of changing skill demand and supply (OECD, 2019[13]). An ageing population can have substantial impacts on a country’s skills supply and training needs in a number of important ways. To start with, population ageing is likely to put considerable pressure on education and training systems as it increases the need for individuals to maintain and update their skills over the life-course in the context of longer working lives. In Latin America, the weak participation of older workers in adult training, coupled with large margins for the introduction of new technologies requiring re-skilling and up-skilling is deemed to put increasing pressure on labour markets and magnify economy-wide skill gaps. In addition, in the context of an increase old-dependency ratio, the retirement of large cohorts can lead to significant shortages of qualified labour in some countries; a gap that can be filled through training of the existing workforce amongst other measures (OECD, 2019[13]). Finally, population ageing is likely to contribute to additional changes in the structure of the economy, leading for instance to increase in the consumption of specific goods and services, an example being an increased demand for health and elderly care services which will require the development of specific skills to fill gaps in the labour market.

Labour market conditions: Job quality and informality are barriers to developing effective labour markets and so an obstacle to skill development

Job quality in Latin America tends to be low

Job quality, in the form of earnings, labour market security and the quality of the working environment can raise well-being, foster productivity while reducing labour market inequalities (Cazes, Hijzen and Saint-Martin, 2015[14]). Nonetheless, job quality is a major concern in Latin America (Figure 2.3). Lower earnings quality compared to the OECD average reflect both substantially lower average earnings and higher levels of earnings inequality. Workers in Latin America also tend to be more vulnerable to labour market risks than their counterparts in more advanced economies. In most emerging economies, this primarily reflects the risk of falling into extreme low pay. The quality of the working environment is also generally lower in Latin America compared with OECD countries. One indication of this is the higher incidence of working very long hours (OECD, 2015[15]).

Figure 2.3. Job quality in LAC countries is low compared to other OECD countries

Note: Data are for 2010, except for Brazil (2009), Chile (2011 for labour market insecurity and 2013 for earnings quality) and Mexico (2013). The OECD average is a simple cross-country average in 2013. OECD calculations based on national household and labour force surveys (EPH: Argentina, PNAD: Brazil, CASEN: Chile, GEIH: Colombia, ENHAO: Costa Rica, ENIGH: Mexico). Earnings quality refers to the extent to which the earnings received by workers in their jobs contribute to their well-being by taking account both the average level as well as the distribution, and can be interpreted as the hourly earnings in USD adjusted for inequality. Labour market insecurity measures the risk of unemployment (the risk of becoming unemployed and the expected duration of unemployment) and the degree of public unemployment insurance (coverage of benefits and their generosity), and can interpreted as the expected monetary loss associated with becoming and staying unemployed as a share of previous earnings. For Argentina, Brazil, Colombia and Costa Rica unemployment insurance is measured as the ratio of the average net income of the unemployed relative to the median net earnings among the employed, and the risk of becoming unemployed is approximated by the unemployment rate, due to limited data availability.

Source: OECD (2014[16]), OECD Employment Outlook 2014, https://doi.org/10.1787/empl_outlook-2014-en; OECD (2016[17]), “Job quality” OECD.Stat (database), http://stats.oecd.org.

Training and adult learning can play a fundamental role in lifting workers out of informality

High and persistent informal employment represents a major policy concern and greatly complicates the challenge of promoting strong productivity growth and more inclusive labour markets in Latin America. Informality, the share of wage earners and self-employed without contributions to the pension system, slightly decreased in the region over the last decade but it remains pervasive across many LAC countries ranging from about 9% in Uruguay to around 65% in Guatemala (Figure 2.4).

Informality is widespread across very different groups of workers, including own-account workers, family workers and self-employed, but also unregistered wage employees in formal or informal firms (OECD, 2009[18]). Moreover, disadvantaged individuals such as low-skilled youth, women or older workers are much more likely to work informally with important repercussions on their earnings and options for skill development.

Evidence shows that workers in informal jobs can earn less than workers in formal jobs, regardless of their socio-economic background and or individual characteristics such as qualification level (ILO, 2016[19]).

From a skill development point of view, informality represents an important challenge. Workers in the informal sector have, in fact, reduced access to training opportunities, a situation that further exacerbates inequalities and penalise them relative to workers in formal jobs when acquiring skills (Carpio et al., 2011[20]).

Figure 2.4. Informality is still pervasive in the LAC region

Informally employed persons as a % of the working-age employment (latest year available)

Source: Adapted from CEDLAS and The World Bank (2019[21]), Socio-Economic Database for Latin America and the Caribbean, http://www.cedlas.econo.unlp.edu.ar/wp/en/estadisticas/sedlac/.

Skills development and informality are linked in a two-way relationship where informal firms tend to be less productive as they generally hire less skilled workers and skilled workers have fewer incentives to be employed in informal firms. Lower participation in training among informal workers is due, in large part to the reduced incentives and capacity that their employers have to provide training (OECD et al., 2019[22]).

If informality hinders skills development, boosting the skills and training opportunities for both employees and employers can play a fundamental role in reducing informality. Recent evidence (IMF, 2018[23]) highlights, in fact, that in all countries in the LAC region, the probability of being employed in the informal sector decreases dramatically with the level of education of a worker.

For instance, in Colombia, recent evidence points to skill upgrading explaining two-thirds of the reduction in informality from 70% in 2007 to 62% in 2017 (IMF, 2018[23]).5 In particular, a worker with a postgraduate degree is nine times as likely to be formal than a worker without any education, and twice as likely as a worker with a high school degree, but no tertiary education.

Equally important are the skills of employers as evidence shows that they play a fundamental role in driving informality. La Porta and Shleifer (2014[24]) argue, for instance, that the lack of human capital, and in particular, the shortage of educated entrepreneurs in the informal sector might be the most important driver of informality, much more important than lack of demand in the economy. World Bank data on a set of developing countries report that, on average, only 7% of the managers of informal firms have a college degree against 76% of employers in formal firms. In Guatemala and Brazil only 8% and 13%, respectively, of managers in the informal sector have a college degree but boosting this number could contribute significantly to reduce the number of informal firms.

Educational attainment and outcomes remain low in the LAC region for both young and adult individuals

Developing strong basic skills in initial education is key to foster participation in further education and training. LAC countries have made substantial progress in improving the coverage and the quality of their education systems, particularly at the primary level (OECD/CAF/UN ECLAC, 2016[8]). Nonetheless, important challenges remain ahead. The low quantity of students enrolled in education and the poor quality of courses are two interrelated challenges in Latin America.

The educational attainment and the skills level of Latin America’s population remain low by international standards. Only 64.2% of individuals aged 25-34 years old have completed secondary education, a figure that is about 20 percentage points lower than the OECD average (84%). In addition, only 23.9% of 25-34 year-olds hold a tertiary degree (43% on average across OECD countries) (Figure 2.5), suggesting that both enrolment rates of secondary and tertiary education need to be boosted to promote the further upskilling of the workforce.

Figure 2.5. Latin America needs to continue to raise educational attainment

Share of population (23-34 year-olds) having completed, % 2017 or latest available year

Source: OECD (2019[25]), “Education at a Glance”, OECD.Stat (database), http://stats.oecd.org.

Solely increasing the quantity of individuals enrolled in education is, however, going to be insufficient to address the pressing challenges facing Latin America’s skill system. LAC countries need to step up their effort to strengthen the quality of education courses as well as that of their adult learning systems.

The OECD's Programme for International Student Assessment (PISA) measures 15-year-olds’ ability to use their reading, mathematics and science knowledge and skills to meet real-life challenges. PISA scores provide a good approximation of how education and training systems are effective in developing students’ skills. Across many LAC countries, PISA scores in science, but also in mathematics and reading, are substantially below the OECD average, as well as below the level of other emerging economies.

The share of low performers is also considerably higher than the OECD average: more than half of young Latin Americans has not acquired basic-level proficiency (OECD/CAF/UN ECLAC, 2016[8]). Participation in early childhood education plays an important role in students’ performance later on (Box 2.2).

Students’ socio-economic backgrounds also influence their opportunities to benefit from education and develop their skills. These associations partly reflect the advantages in resources that relatively high socio-economic status confers. For example, at the system level, high socio-economic status is often related to greater wealth and higher spending on education. At the school level, socio-economic status tends to be positively correlated with a range of community characteristics that can boost student performance, such as a safe environment or the availability of public libraries and museums. At the individual level, socio-economic status can be related to parents’ attitudes towards education, in general, and to their involvement in their child’s education, in particular. In Peru, socio-economic status explains more than 20% of the variability in PISA scores in science, but socio-economic status plays a great role also in Chile, Costa Rica and Colombia where its impact is well above the average across OECD countries (Figure 2.6). In contrast to other LAC countries, in Chile and Colombia, the importance of socio-economic background in explaining students’ performance has also increased considerably in between 2006 and 2015.

Box 2.2. Participation in early childhood education plays an important role in students’ performance later on

Participation in early childhood education and care (ECEC) varies significantly across countries, despite the evidence that engagement in such activities has a strong impact on students’ cognitive and non-cognitive skills development. In particular, recent studies show that high-quality ECEC can result in better outcomes in subsequent stages of life and disadvantaged children can benefit the most from attending high-quality early childhood education. Later interventions are less efficient because they take place after children’s “development window”.

Increasing rates of participation in early childhood education and care require, however, addressing the financial costs that act too often as the main barrier preventing parents from enrolling their children. To encourage universal participation in initial education, governments can consider measures that enforce school attendance, target students who are at risk of lagging behind and design targeted measures regarding school dropout. Such measures need to be combined with measures that address inequalities of opportunities. The engagement of parents, the local authorities but also teachers as main stakeholders is crucial.

Source: OECD (2019[13]), OECD Skills Strategy 2019: Skills to Shape a Better Future, https://doi.org/10.1787/9789264313835-en.

Figure 2.6. School results and equity remain a challenge

Note: Panel B displays the percentage of variation in science performance explained by the PISA index of economic, social and cultural.

Source: OECD (2006[26]), PISA 2006 Database, https://www.oecd.org/pisa/pisaproducts/database-pisa2006.htm; OECD (2015[27]), PISA 2015 Database, http://www.oecd.org/pisa/data/2015database/.

Negative education outcomes early in life as students translate into poor skill developments as adults. Analysis of data from the Survey of Adult Skills (PIAAC) shows that in the LAC region (Chile, Ecuador, Mexico and Peru), more than 60% of adults have low levels of skills – both in terms of literacy and numeracy. This figure is three times larger than the OECD average (Figure 2.7).

Evidence highlights how LAC countries need to boost participation in high quality training at all levels and especially in adult learning where LAC are lagging particularly behind.

Providing learning opportunities to adults in Latin America can therefore play a crucial role in developing the necessary skills needed to face the challenges of the future of work and in revamping stagnant productivity growth in Latin America while promoting more inclusive labour markets and societies.

Figure 2.7. The proportion of low performers in literacy and numeracy

Adults

Note: Low-performers are defined as those who score at or below level 1 in in either literacy or numeracy.

According to the Survey of Adult Skills (PIAAC). Hungary, Mexico, the United States, Ecuador and Peru: Year of reference 2017. Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference 2015. All other countries: Year of reference 2012. Data for Belgium refer only to Flanders and data for the United Kingdom refer to England and Northern Ireland jointly.

Source: OECD calculations based on OECD (2017[28]), Survey of Adults Skills (PIAAC) (2012, 2015, 2017) (database), http://www.oecd.org/skills/piaac/.

The economic development of LAC countries shapes their skill demands and imbalances

The productive and sectoral distribution of employment of LAC countries combined with recent pressing demographic developments and the relatively poor skill proficiency of the adult population in the region are reflected in the emerging skills imbalances (shortages, surpluses and mismatches) across labour markets.

Measures of skill imbalances are often issued from employer surveys, which include questions on hiring intentions and recruitment difficulties. According to the Talent Shortage Survey by Manpower group, many LAC countries experience gaps between the available pool of skills and those skills that economies and societies require (Figure 2.8 Panel A).6 According to the same survey, employers cited low numbers of applicants, lack of technical competencies, lack of experience, and lack of soft skills as reasons contributing to the difficulty of filling open positions (ManpowerGroup, 2015[29]).

Moreover, based on the World Bank Enterprise Survey 2009-2017, approximately a third of firms (32%) in Latin America identified an inadequately educated workforce as a constraint to their activity.

Figure 2.8. Finding the right skills can be difficult in Latin America (Employer reported labour market imbalances)

Share of firms identifying hiring difficulty, 2017

Note: Data refer to 2017 for Argentina, Bolivia, Colombia, Ecuador, Guatemala, Peru and Uruguay; to 2016 for El Salvador, the Dominican Republic and Honduras; to 2010 for the Bolivarian Republic of Venezuela, Chile, Costa Rica, Mexico and Panama and to 2009 for Brazil.

Source: ManpowerGroup (2017[30]), Talent Shortage Survey, https://go.manpowergroup.com/talent-shortage.

While these results suggest that shortages in Latin America are significant, they need to be interpreted with caution, as they rely on self-reported subjective perceptions of a very selected sample of employers (i.e. formal firms in manufacturing).

A different approach to the measurement of skills imbalances has been recently proposed in the OECD Skills for Jobs database7 which provides detailed information about the skill needs of the labour markets (see Box 2.3 for more details). The indicator is built by tracking a multidimensional set of quantitative and objective (rather than subjective) indicators for each occupational group of the labour market. Indicators include wage growth, employment growth and unemployment and provide an assessment of the demand by the labour market for specific occupations and their associated skills.

Evidence shows that, on average across OECD countries, more than 5 out of 10 jobs that are hard to fill (i.e. in shortage) are found in high-skilled occupations (Figure 2.9). These jobs range from managerial positions to highly skilled professionals in the health care, teaching or ICT sectors. A relatively large share of OECD occupational shortage (approximately 39% of total jobs that are hard-to-fill across OECD countries) is also found in medium-skilled occupations, such as personal service workers or electrical and electronic trades workers. Less than 1 out 10 jobs in shortage across OECD countries are found in low skilled occupations.

The picture is rather different for LAC countries. In countries such as Argentina, Brazil, Chile, Mexico and Peru the shortages of highly skilled professionals is significantly lower than the OECD average, with less than 2 out of 10 jobs in shortage being “high-skilled” and the majority of jobs in demand being found, instead, in medium to low-skilled occupations.

Results seem to suggest that the demand for employment in LAC countries is still mostly concentrated in low to medium-skill jobs in the agriculture and manufacturing sector and that, despite the overall skill supply of the region being relatively low, this is sufficient to meet an equally rather weak demand for high-skills.

Box 2.3. OECD Skills for Jobs database

The OECD Skills for Jobs database provides timely information about skills shortages – i.e. when skills sought by employers are not available in the pool of potential recruits – and skills surpluses – i.e. when the supply of certain skills is higher than the demand. The database has key innovative features compared to existing measures of skills shortages/surpluses. By looking at skills – i.e. the set of competences mobilised to perform the tasks related to a job – the new indicators go beyond the traditional measures of imbalances. Furthermore, unlike the generally subjective information available from employer surveys, the OECD Skills for Jobs database draws from quantitative data derived from household surveys. Finally, the indicator is constructed using a multidimensional set of quantitative signals on skills pressure (i.e. five sub-indices, including wage growth, employment growth and unemployment), which provides a holistic interpretation of skill imbalances in the labour markets. The skill needs indicator is constructed in two consecutive steps:

• In the first step, sub-indices for hourly wage growth, employment growth, unemployment rate, hours worked and under-qualification are used to provide a quantitative indication of the extent of the labour market pressure on each one of the occupations analysed. The result of this analysis returns a ranking of occupations ordered from the one most in shortage to most in surplus.

• In the second step, occupations that are in shortage/surplus are mapped into the underlying skills requirements associated to those occupations, using the occupation-skills taxonomy developed by O*NET.

Information is provided at the 2-digit ISCO occupation level and is disaggregated into three domains of competence – knowledge, skills, and abilities:

• Knowledge refers to the body of information that makes adequate performance on the job possible (e.g. knowledge of plumbing for a plumber; knowledge of mathematics for an economist).

• Skills refer to the proficient manual, verbal or mental manipulation of data or things (e.g. complex problem solving; social skills).

• Ability refer to the competence to perform an observable activity (e.g. ability to plan and organise work; attentiveness; endurance).

The database covers most OECD countries and some emerging economies. LAC countries included in the database are: Argentina, Brazil, Chile, Mexico and Peru. Indicators are available at the regional and occupational level.

Source: OECD (2017[31]), Getting Skills Right: Skills for Jobs Indicators, https://doi.org/10.1787/9789264277878-en.

Figure 2.9. Share of employment in high demand, by skills level

Source: OECD (2018[32]), OECD Skills for Jobs database, https://www.oecdskillsforjobsdatabase.org.

Results at the occupational level confirm that, in Latin America, shortages can be found in a variety of low-medium skill occupations such as assemblers or labourers in mining, construction, manufacturing and transport, customer service clerks but also in medium to high-level occupations such as ICT technicians and business and administration professionals.

Surpluses (i.e. occupations that have recently experienced a decline in wage growth or in hours worked for instance) can be observed in certain managerial occupations. These can be found in production and specialised services managers or chief executives, senior officials and legislator as well as in some occupational categories that may be more exposed to the effects of automation such as general and keyboard clerks or metal, machinery and related trades workers as well as sales workers.

The picture is, however, rather heterogeneous when one looks at the specifics of each country. For instance, in Chile, labourers in mining and construction,8 have experienced robust wage growth signalling a strong demand for low-medium professionals in the country in those occupations. At the same time, also business and administration professionals have also seen a sharp increase in both hours worked and wages, implying a sustained demand for this medium-skilled professionals. In Brazil, where the largest demand is in middle-skill professionals, health associate professional and personal care workers have been on the rise in recent years, while wages have been declining in traditionally high-skilled occupations such as science and engineering professionals.9

The emergence of skills shortages and surpluses signals that labour markets are struggling to align skills supply to demand. The disconnect between the educational offer and labour market needs can lead to persistent skills and qualification mismatch (Figure 2.10) which, in turn, are likely to contribute to high wage inequality and low productivity in the region (OECD/CAF/UN ECLAC, 2016[8]).

Indicators of qualification mismatch are useful measures of the alignment of a worker’s qualification level to that required in her/his job. So-called over-qualified workers own higher qualifications than usually required in their jobs while under-qualified workers have lower qualifications than those usually held by workers in similar jobs. Evidence from the OECD Skills for Jobs database reveals that, 44% of workers in Latin America are mismatched by qualification, in contrast to 35% across OECD countries.

Table 2.2. Occupational shortage and surplus in Latin America
Top 10 shortage occupations	Top 10 surplus occupations
Market-oriented skilled forestry, fishery and hunting workers	Production and specialised services managers
Personal care workers	Legal, social and cultural professionals
Customer services clerks	Metal, machinery and related trades workers
Assemblers	Chief executives, senior officials and legislators
Subsistence farmers, fishers, hunters and gatherers	Market-oriented skilled agricultural workers
Refuse workers and other elementary workers	Administrative and commercial managers
Information and communications technicians	Handicraft and printing workers
Labourers in mining, construction, manufacturing and transport	General and keyboard clerks
Protective services workers	Legal, social, cultural and related associate professionals
Business and administration associate professionals	Street and related sales and services workers
Note: Values are obtained by taking the average of the occupational indicators for Argentina, Brazil, Chile, Mexico and Peru, when available. Latest year available for each country.
Source: OECD (2018[32]), OECD Skills for Jobs database, https://www.oecdskillsforjobsdatabase.org.

Figure 2.10. Qualification mismatch is high in Latin America

Percentage of mismatched workers

Note: Mismatch is calculated as the share of individuals with a higher (over-qualification) or lower (under-qualification) level of qualification than the modal level in his/her occupation.

Source: OECD (2018[32]), OECD Skills for Jobs database, https://www.oecdskillsforjobsdatabase.org.

Moreover, results show that LAC countries present a higher share of overqualified workers than across OECD countries (30% in Latin America vs. 17% across OECD countries) highlighting the relatively weak demand for high-level skills in economies that are still in their development phase and where there is ample room to boost the adoption of new technologies into the productive structure.

Further evidence from the OECD Skills for Jobs database show that skill demands are quite heterogeneous across the countries examined, reflecting differences in their productive structure, export pattern and internal demand. For instance, technical skills are in strong demand in both Chile and Mexico while Argentina’s shortages are in social as well as in basic and complex problem solving skills. Surpluses of complex problem solving skills (signalling a weak demand) are instead found in relatively less developed economies such as Peru.

Figure 2.11. Skills shortage indicators for selected LAC and OECD countries

Note: Negative (positive) values indicate that a specific skill is in surplus (shortage). Values for Latin America are obtained by taking the average of the occupational indicators for Argentina, Brazil, Chile, Mexico and Peru. Values for the OECD are obtained by taking the average for all OECD countries included in the 2018 OECD Skills for Jobs database. Latest year available for each country.

Source: OECD (2018[32]), OECD Skills for Jobs database, https://www.oecdskillsforjobsdatabase.org.

References

[20] Carpio, S. et al. (2011), “The effect of temporary contracts on human capital accumulation in Chile”, IDB Working Paper Series, No. IDB-WP-253, Inter-American Development Bank, Washington, D.C., https://publications.iadb.org/publications/english/document/The-Effect-of-Temporary-Contracts-on-Human-Capital-Accumulation-in-Chile.pdf.

[14] Cazes, S., A. Hijzen and A. Saint-Martin (2015), “Measuring and Assessing Job Quality: The OECD Job Quality Framework”, OECD Social, Employment and Migration Working Papers, No. 174, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jrp02kjw1mr-en.

[21] CEDLAS and The World Bank (2019), Socio-Economic Database for Latin America and the Caribbean, http://www.cedlas.econo.unlp.edu.ar/wp/en/estadisticas/sedlac/ (accessed on 11 December 2019).

[33] Frey, C. and M. Osborne (2017), “The future of employment: How susceptible are jobs to computerisation?”, Technological Forecasting and Social Change, Vol. 114, pp. 254-280, https://doi.org/10.1016/j.techfore.2016.08.019.

[19] ILO (2016), Non-standard Employment Around the World: Understanding Challenges, Shaping Prospects, International Labour Office, Geneve, https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/documents/publication/wcms_534326.pdf.

[23] IMF (2018), “Colombia : Selected Issues”, IMF Staff Country Reports, No. 18/129, https://www.imf.org/en/Publications/CR/Issues/2018/05/29/Colombia-Selected-Issues-45899.

[24] La Porta, R. and A. Shleifer (2014), “Informality and development”, Journal of Economic Perspectives, Vol. 28/3, pp. 109-26, https://www.aeaweb.org/articles?id=10.1257/jep.28.3.109.

[10] Lorenz, M. et al. (2015), Man and machine in Industry 4.0: How will technology transform the industrial workforce through 2025?, The Boston Consulting Group, https://www.bcg.com/publications/2015/technology-business-transformation-engineered-products-infrastructure-man-machine-industry-4.aspx.

[2] Maloney, W. and C. Molina (2016), “Are automation and trade polarizing developing country labor markets, too?”, Policy Research Working Paper, No. WPS 7922, World Bank Group, Washington, D.C., http://documents.worldbank.org/curated/en/869281482170996446/Are-automation-and-trade-polarizing-developing-country-labor-markets-too.

[30] ManpowerGroup (2017), Talent Shortage Survey, ManpowerGroup, https://go.manpowergroup.com/talent-shortage.

[29] ManpowerGroup (2015), Talent Shortage Survey, https://go.manpowergroup.com/talent-shortage.

[7] McKinsey Global Institute (2017), A Future that Works: Automation, Employment and Productivity, McKinsey & Company, https://www.mckinsey.com/~/media/mckinsey/featured%20insights/Digital%20Disruption/Harnessing%20automation%20for%20a%20future%20that%20works/MGI-A-future-that-works-Full-report.ashx.

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

[25] OECD (2019), “Education at a Glance”, OECD.Stat (database), http://stats.oecd.org (accessed on 11 December 2019).

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

[13] OECD (2019), OECD Skills Strategy 2019: Skills to Shape a Better Future, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264313835-en.

[32] OECD (2018), OECD Skills for Jobs Database, https://www.oecdskillsforjobsdatabase.org.

[31] OECD (2017), Getting Skills Right: Skills for Jobs Indicators, Getting Skills Right, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264277878-en.

[28] OECD (2017), Survey of Adults Skills (PIAAC) (2012, 2015, 2017), (database), http://www.oecd.org/skills/piaac/.

[11] OECD (2017), The Next Production Revolution: Implications for Governments and Business, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264271036-en.

[17] OECD (2016), “Job quality”, OECD.Stat (database), http://stats.oecd.org (accessed on 11 December 2019).

[9] OECD (2015), Data-Driven Innovation: Big Data for Growth and Well-Being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264229358-en.

[15] OECD (2015), “Enhancing job quality in emerging economies”, in OECD Employment Outlook 2015, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2015-9-en.

[27] OECD (2015), PISA 2015 Database, http://www.oecd.org/pisa/data/2015database/.

[16] OECD (2014), OECD Employment Outlook 2014, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2014-en.

[18] OECD (2009), OECD Employment Outlook 2009: Tackling the Jobs Crisis, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2009-en.

[26] OECD (2006), OECD PISA 2006 Database, https://www.oecd.org/pisa/pisaproducts/database-pisa2006.htm.

[8] OECD/CAF/UN ECLAC (2016), Latin American Economic Outlook 2017: Youth, Skills and Entrepreneurship, OECD Publishing, Paris, https://dx.doi.org/10.1787/leo-2017-en.

[22] OECD et al. (2019), Latin American Economic Outlook 2019: Development in Transition, OECD Publishing, Paris, https://dx.doi.org/10.1787/g2g9ff18-en.

[3] The World Bank (2018), World Development Indicators, http://datatopics.worldbank.org/world-development-indicators/.

[5] The World Bank (2016), World Development Report 2016: Digital Dividends, International Bank for Reconstruction and Development / The World Bank, Washington, DC, https://doi.org/10.1596/978-1-4648-0728-2.

[12] United Nations, Department of Economic and Social Affairs, Population Division (2019), World Population Prospects 2019: Highlights (ST/ESA/SER.A/423), United Nations, New York, https://population.un.org/wpp/Publications/Files/WPP2019_Highlights.pdf.

[6] Weller, J., S. Gontero and S. Campbell (2019), “Cambio tecnológico y empleo: una perspectiva latinoamericana. Riesgos de la sustitución tecnológica del trabajo humano y desafíos de la generación de nuevos puestos de trabajo”, Macroeconomía del Desarollo, No. 201, CEPAL, https://www.cepal.org/es/publicaciones/44637-cambio-tecnologico-empleo-perspectiva-latinoamericana-riesgos-la-sustitucion.

Notes

← 1. These are the Latin American countries participating in the Survey of Adults Skills (PIAAC) on which OECD estimates of the risk of automation are based.

← 2. The World Bank (2016[5]) and Weller, Gontero and Campbell (2019[6]) estimates are constructed using experts’ assessment of the probability that different occupations can be automated and follow the same methodology as Frey and Osborne (2017[33]). Nedelkoska and Quintini (2018[4]), while also departing from Frey and Osborne’s analysis, directly explore the task content of individual jobs instead of the average task content within each occupation. Finally, the McKinsey Global Institute (2017[7]) assesses the technical potential for automation through an analysis of the component activities of each occupation. The authors break down about 800 occupations into more than 2 000 activities, and determine the performance capabilities needed for each activity based on the way humans currently perform them. Finally, they further break down each activity into 18 capabilities and assess the technical potential for automation of those capabilities.

← 3. These developments will be challenging for public budgets and pension systems. Indeed, the falling share of the population at traditionally productive ages means relatively fewer people will pay taxes and social contributions at a time when the rising share of older persons implies that more people will receive pensions and costly health services, etc.

← 4. The old-age dependency ratio is defined as the ratio of those of non-active age (65 and over) to those of active age (20-64) in a given population.

← 5. Also almost half of the regional variation in informality can be attributed to access to good quality higher education (IMF, 2018[23]).

← 6. The Manpower Talent Shortage survey is an annual survey of a subsample of formal firms in the manufacturing sector (https://go.manpowergroup.com/talent-shortage).

← 7. The OECD Skills for Jobs database (OECD, 2017[31]) provides regularly-updated international evidence on skill shortages, surpluses and mismatches, using quantitative information from large-scale datasets (e.g. labour force surveys).

← 8. Chile is the leader producer of copper in the world with a production market share of around 30%. In 2018, this industry represented about 50% of the country’s total exports. Recently, there are however signs of a slowdown in this sector as large infrastructure projects are under development.

← 9. See: https://www.oecdskillsforjobsdatabase.org.