1. Who are the low-educated in Belgium?

Abstract

This chapter paints a picture of who the low-educated in Belgium are, how the demographics of this group have changed over time, and how they compare to those with more education. The analysis also assesses to what extent employment outcomes for the low-educated in Belgium can be linked to observable differences in their demographic characteristics compared to the low-educated in neighbouring countries. As in many other countries, the share of low-educated in Belgium has fallen over time, but it remains higher than in the Netherlands, France and especially Germany. The share of low-educated is particularly high among non-EU migrants and older individuals. However, the analysis presented in this chapter shows that these differences in demographics between Belgium and its neighbours are small, and that they cannot explain the differences in employment rates for the low-educated.

This chapter paints a picture of who the low-educated in Belgium are, how the demographics of this group have changed over time, and how they compare to other education groups. The analysis also assesses to what extent employment outcomes for the low-educated in Belgium can be linked to observable differences in their demographic characteristics compared to the low-educated in neighbouring countries.

As in many other countries, the share of low-educated in Belgium has fallen over time, from 41% in 1998 to 22% in 2018. It remains higher than in the Netherlands (20%), France (20%) and especially Germany (14%). The share of low-educated is also much higher in Wallonia (25%) and the Brussels-Capital Region (28%) than in Flanders (19%). The share of low-educated is particularly high among non-EU migrants (43%) and older individuals (35%). Compared to its neighbouring countries, Belgium has a slightly higher share of migrants, and of low-educated migrants in particular. However, the analysis presented in this chapter shows that these differences in demographics between Belgium and its neighbouring countries cannot explain the differences in employment outcomes among the low-educated across countries. Differences in policies and institutions are likely to play a more important role (see Chapter 4).

1.1. Belgium has a high share of the population that is low-educated compared to neighbouring countries

The share of the population without an upper-secondary degree in Belgium is slightly above the average for European OECD countries. In 2018, 22% of the population aged 20-64 in Belgium held less than an upper-secondary degree as their highest qualification (Figure 1.1). The European OECD average in 2018 was 19%. France (20%), the Netherlands (20%) and Germany (14%) all had lower shares of low-educated workers than Belgium. In terms of skill levels, the low-educated in Belgium are comparable to their peers in neighbouring countries (Box 1.1).

Figure 1.1. The share of the population with a low education has fallen across OECD countries, but it still represents a significant share of the working-age population

Across the OECD, the share of the population without an upper-secondary degree fell dramatically over the past two decades. On average, across the European OECD members in Figure 1.1, the share of the working-age population with less than an upper-secondary degree fell from 34% to 19% between 1998 and 2018. Belgium is no different, with the share falling from 41% to 22%. Although the decline in the share of the working-age population without an upper-secondary degree is a trend shared by almost all European OECD countries, the low-educated still make up 19% of the population on average. In Belgium, this is equivalent to over a million individuals of working age.

The share of the population without an upper-secondary degree is not uniform across regions within Belgium. In 2018, the share of the population without an upper-secondary degree stood at 19% in Flanders, while for the Brussels-Capital Region and Wallonia, the shares were 28% and 25%, respectively. To put this in perspective, shares in the Brussels-Capital Region and Wallonia were similar to those of Greece and Iceland, while Flanders is more comparable to Denmark, Norway and France.

Box 1.1. Are the low-educated in Belgium less skilled than those in neighbouring countries?

One possibility is that the low-educated in Belgium are less skilled than the low-educated in neighbouring countries which, if true, could also partly explain their poorer labour market outcomes. This short box explores this possibility and concludes that the skills of the low-educated in Belgium are about average compared to those in neighbouring countries, and so their poorer labour market outcomes cannot be attributed to lower skills.

The first evidence comes from the Survey of Adult Skills (PIAAC), which tested adults on both numeracy and literacy skills in 2012. The survey only covered Flanders, but it is clear that the skills of the low-educated in Flanders are on a par with those in Germany, slightly lower than those in the Netherlands, and significantly higher than those in France (Figure 1.2).

Figure 1.2. The skills of low-educated in Belgium (Flanders) are relatively high

One drawback of the Survey of Adult Skills is that only Flanders participated. One may only speculate, but it is possible that the skills of the low-educated in Wallonia and Brussels-Capital Region may be lower than those in Flanders. Another source of information on skills is the OECD’s Programme for International Student Assessment, which measures the reading, mathematics and science knowledge of 15-year-olds across a wide range of countries. These tests are of students, so they will not reflect the skills of adults in the labour force. Moreover, the latest PISA tests are for 2018, and so the majority in this cohort will not have entered the labour market yet. However, anyone tested in 2015 and before would, by now, be aged 20 or above. Averaging the scores for all the years that PISA was run prior to 2018 (i.e. 2000, 2003, 2006, 2009, 2012 and 2015) shows that students in Belgium perform better than those in France and Germany in both reading and mathematics, but less well than those in the Netherlands. In Science, Belgian students do slightly better than those in France, but less well than those in Germany and the Netherlands. These PISA scores reflect the performance of all students, not just the low-educated. However, further evidence suggests that the share of low-achieving students in Belgium is similar to that in neighbouring countries (OECD, 2019[1]).

1.1.1. Most of the low-educated are located in Flanders, but they are over-represented in Wallonia and the Brussels-Capital Region

The distribution of the low-educated is not equal across regions in Belgium. In 2018, 49% of the population without an upper-secondary degree resided in Flanders compared to 36% in Wallonia and 11% in the Brussels-Capital Region (Figure 1.3). This partly reflects the distribution of the overall population across regions. In 2018, 57% of the general population resided in Flanders while 32% lived in Wallonia and 11% in the Brussels-Capital Region. In fact, the low-educated are slightly under-represented in Flanders and over-represented in Wallonia.

Figure 1.3. Just under half of the low-educated in Belgium are located in Flanders

The distribution of the low-educated has slowly shifted away from Flanders and towards Wallonia and the Brussels-Capital Region over the last 20 years. In 1998, 57% of the low-educated resided in Flanders, 34% in Wallonia and 9% in the Brussels-Capital Region. This shift occurred even though the population distribution remained largely stable. The overall share of the population without an upper-secondary qualification has declined over the past 20 years, but the decline has been swifter in Flanders.

1.2. Each generation is getting more education, driving down the share without an upper-secondary degree

The majority of the low-educated in Belgium are prime age. The population aged 25-54 made up nearly 50% of the low-educated in 2018 (Figure 1.4). Another 14% were young (20 to 25) and 36% were older (55 to 64).

The low-educated in Belgium are older compared to those in other European OECD countries. On average, across European OECD countries, the share of the low-educated who are prime working-age is 53%. Belgium’s share (50%) is lower than in the Netherlands (51%), France (51%) and Germany (53%). In contrast, the share that is old (55-64) in Belgium is slightly above the European OECD average and the share that is young (20-24) is below average. The distribution of the low-educated across age groups in Belgium suggests that younger cohorts are more likely than older cohorts to have completed an upper-secondary qualification.

Belgium’s higher share of older low-educated is due to older cohorts in Belgium being less likely to complete an upper-secondary qualification compared to neighbouring countries. In 2018, among OECD countries in the sample, 27% of the population aged 55-64 lacked an upper-secondary degree, compared to 35% in Belgium (Figure 1.5). Shares in neighbouring countries were 14% in Germany, 32% in France and 30% in the Netherlands. In Flanders and Wallonia, 33% and 37% of 55-64 year-olds, respectively, did not have an upper-secondary degree. In the Brussels-Capital Region, the equivalent was 43%.

Figure 1.4. Belgium has a relatively low share of prime-age low-educated

Figure 1.5. The low-educated population is ageing across European OECD countries

The ageing of the low-educated population suggests that some of the decline in the employment prospects of the low-educated is mechanical. Across European OECD countries, younger workers are now more likely to achieve an upper-secondary qualification. Those who still fail to graduate with an upper-secondary qualification likely have different ability or family circumstances compared to those who did not gain an upper-secondary degree a few decades earlier. Thus, it is not unreasonable to suspect some degradation in labour market prospects for the low-educated simply through this selection effect.

1.3. Belgium has a relatively high share of migrants and they are less likely to have an upper secondary qualification

The share of migrants living in Belgium is slightly higher than in other European OECD countries. Across all European OECD countries, migrants comprise a little over 15% of the population aged 20-64, compared to 20% in Belgium (Figure 1.6). Among neighbouring countries, Germany (21%) has a slightly higher share of migrants, while both France (14%) and the Netherlands (15%) have lower shares.

In OECD countries, but particularly in European OECD countries, migrants have less education than their native-born peers. Migrants face unique challenges in the labour market: they may not perfectly master the local language, and the qualifications they have obtained may receive less (or no) recognition. However, the employment rate of low-educated migrants is not always lower than that of their native-born peers. Only about 50% of OECD countries have employment rates for low-educated natives which exceed those of the low-educated migrants (OECD/European Union, 2018[2]). In Belgium, employment rates for low-educated natives (47%) only slightly exceed those of migrants (45%).

Figure 1.6. Belgium has a high share of migrants compared to other European OECD countries

Among Belgian regions, the Brussels-Capital Region in particular has a large share of migration (59% of its population aged 20-64). In comparison, Flanders and Wallonia straddle the average of European OECD countries. Slightly less than 15% of the population of Flanders are migrants and 17% in Wallonia.

Migration in Belgium is relatively balanced between migrants from within the EU and those from outside the EU. The Brussels-Capital Region is again the exception, with almost 25% of its population coming from other countries within the EU and about 33% coming from outside the EU. This compares to a little over 6% for EU migration and slightly less than 10% for migration from outside the EU on average among OECD European countries. Flanders and Wallonia have shares of EU and non-EU migrants roughly in line with the European OECD average. A key focus of employment policy in the Brussels-Capital Region should therefore be on the better integration of migrants in the labour market (Chapter 2).

1.3.1. The low-educated are disproportionately likely to be migrants

In 2018, 33% of the low-educated in Belgium were migrants, with 23% coming from outside of the European Union. In Germany, 48% of the low-educated were migrants compared to 26% in France and 20% in the Netherlands (Figure 1.7). The share of migrants from outside the European Union was 35% in Germany, 21% in France and 16% in the Netherlands. Belgium has a higher share of migrants among the low-educated than the Netherlands and slightly higher than France, but lower than Germany.1

Figure 1.7. A high share of the low-educated in Belgium are migrants

Within Belgium, the Brussels-Capital Region stands out with the highest shares of the low-educated who are migrants. In the Brussels-Capital Region, 70% of the low-educated are migrants compared to 29% in Flanders and 23% in Wallonia. In fact, just over 50% of the low-educated in the Brussels-Capital Region were born outside of the European Union compared to 21% in Flanders and 14% in Wallonia. Policy issues surrounding the low-educated in the Brussels-Capital Region are inextricably linked to migration policy and the integration of migrants, whereas this appears to be relatively less important in Flanders or Wallonia.

1.4. The majority of the low-educated in Belgium are now men

A majority (53%) of the low-educated population in Belgium are men. This share is the same as the average among European OECD countries. Across regions in Belgium, there is little variance in the share of the low-educated who are men. In Flanders the share of the low-educated who are men stood at 54% in 2018, compared to 52% in Wallonia and 54% in the Brussels-Capital Region. Compared to neighbouring countries, Belgium has a higher share of the low-educated who are men. In France, the Netherlands and Germany, the share of the low-educated who are men stood at 49%, 50% and 49%, respectively (Figure 1.8). As with the other demographic characteristics, the differences between Belgium and its neighbouring countries are small, casting some doubt on how much of the difference in low-educated employment rates they can explain (Box 1.2).

The past 20 years has seen a small and gradual shift in the composition of the low-educated from women to men. In 1998, 50% of the low-educated in Belgium were men, which rose to 51% in 2008. The same broad trend – though at a faster pace – held across European OECD countries, with the average share of men among the low-educated rising from 46% in 1998 to 52% in 2018. Even among countries neighbouring Belgium (France, Germany) where women still make up a majority of the low-educated, men comprise a slowly increasing share of the low-educated.

Figure 1.8. The majority of the low-educated in Belgium are now men

1.5. Household composition and childcare may affect employment rates among the low-educated

One issue known to suppress labour supply and employment rates, particularly for women, is childcare. If individuals find it difficult to find adequate, reasonably priced child care, it may make sense for them to stay home and care for children rather than enter the labour market (OECD, 2012[3]). There are at least two indicators for childcare as a detriment to labour market participation. The first is the design and adequacy of the childcare system in any given country. The second is the underlying fertility rate, and whether the share of households with children has changed in any meaningful way.

Among the low-educated, the Brussels-Capital Region has the highest share of households with children. For the Brussels-Capital Region it is 56% while Wallonia finds itself around average among European OECD countries in the sample with 41%. This compares to an average of just 35% in Flanders, which is among the lowest of countries in Figure 1.9.

Figure 1.9. The share of households with children present has not changed appreciably over the past 10 years in Belgium

Across countries, but within education groups, there is not a lot of variation in the share of households with children. On average, the higher the education of the household head, the more likely the household is to have children present. Thus, households headed by someone without an upper-secondary degree are less likely to have children present than households headed by someone with a higher level of education. Finally, there is little variation over time and within countries. On average the share of households with a child has moved very little over the past ten years falling from 42.2% to 41.6% between 2008 and 2018. Most of the drop is accounted for by countries in Eastern Europe.

Box 1.2. The lower employment rate among the low-educated in Belgium compared to neighbouring countries cannot be explained by differences in demographic composition

This chapter is concerned with building a demographic picture of the low-educated in order to get a broad sense of what groups will be most affected by any policy changes. A second, and equally important question is how the demographic composition of the low-educated in Belgium might contribute to the lower employment rate for the low-educated compared to neighbouring countries. The subject of employment rates and labour market outcomes for the low-educated are addressed in Chapter 2. Using a shift-share framework, this box pulls together the various demographic dimensions discussed in this chapter to determine to what extent they help explain lower employment rates for the low-educated in Belgium compared to neighbouring countries.

To do this, the analysis focuses on the following decomposition of the difference in employment rates between Belgium and neighbouring countries:

$y_{b} - y_{g} = \sum_{i} y_{i, b} (w_{i, b} - w_{i, g}) + \sum_{i} w_{i, g} (y_{i, b} - y_{i, g})$ ,

where $y_{b}$ is the employment rate for the low-educated in Belgium in 2018, and $y_{g}$ is the corresponding employment rate for each of Germany, France and the Netherlands. The shift-share proceeds by partitioning the low-educated population into one of 48 cells (indexed by i) according to the demographic profile of the low-educated population for each country. The cells are defined by age group (prime-age, older working age), migration status, gender, presence of children in the household, and urban/rural location of the household. For each cell in each country, a corresponding employment rate ( $y_{i, b}$ , $y_{i, g}$ ) and population share ( $w_{i, b}$ , $w_{i, g}$ ) of that cell is calculated. The shift-share decomposes the difference in employment rates into the part that is attributable to the different demographic composition ( $\sum_{i} y_{i, b} (w_{i, b} - w_{i, g})$ ) of the low-educated across countries, as well as the within-cell difference in employment rates ( $\sum_{i} w_{i, g} (y_{i, b} - y_{i, g})$ ).

The results indicate that Belgium’s demographics do not contribute to lower employment rates among the low-educated compared to neighbouring countries. Compared to Germany, differences in composition account for a little more than 10% in the difference in employment rates. Compared to France, Belgium’s demographic differences play no role, and compared to the Netherlands, the Belgium’s demographics actually act to increase employment rates among the low-educated compared to the counterfactual case where Belgium had the same demographic profile of the Netherlands among the low-educated.

The intuition for the result, which is clear from the rest of this chapter, is Belgium’s demographics do not deviate appreciably from its neighbours. One area in which it does, and which is important for employment rates, is the gender composition of the low-educated population. Men are much more likely to be employed than women, and Belgium has a high relative share of the low-educated who are male, which helps to increase predicted employment rates.

1.6. The low-educated in Wallonia are disproportionately located in rural areas

The low-educated in Belgium are more likely to be located in towns and suburbs with medium density. Figure 1.10 shows the distribution of the low-educated across various geographies ranked by their degree of population density.2 Compared to neighbouring countries, Belgium has a relatively high share of the low-educated living in medium-density towns and suburbs. In 2018, 56% of the low-educated in Belgium lived in suburbs and towns compared to 42% in Germany, 23% in France and 35% in the Netherlands. Even compared to all European OECD countries, Belgium has the highest share of the low-educated living in suburbs and towns of medium density.

The concentration of the low-educated in Belgium in suburbs and towns leaves very small shares in cities or rural areas. With 34% of the low-educated located in cities and 10% located in rural areas with low density, Belgium has the lowest share of the low-educated living in these urban geographies compared to neighbouring countries. The Netherlands has the highest share of the low-educated in cities (54%), and France has the highest share in rural areas (34%).

The distribution of the low-educated across urban geographies is not uniform across regions in Belgium. This is obviously the case with the Brussels-Capital Region, which lies entirely within an urban geography, and therefore all low-educated in the Brussels-Capital Region live in a high density urban environment. More interesting is the case of Flanders and Wallonia. Flanders and Wallonia have relatively similar shares of the low-educated living in cities. Where they diverge is the shares in rural areas versus suburbs and towns with medium density. Flanders has 74% of the low-educated living in suburbs and towns compared to 55% in Wallonia. Wallonia has 20% of the low-educated living in rural areas compared to just 6% in Flanders. Compared to other regions in Belgium, the low-educated are much more concentrated in rural areas in Wallonia.

Rural areas present particular challenges for accessing jobs and essential government services. Across OECD countries, cities have been engines of firm creation and job growth over the past 20 years (OECD, 2019[4]). Rural areas, which tend to have less access to public transportation networks and broadband, and may have more difficulty accessing government employment services, have often seen lagging employment rates due partially to their geography (OECD, 2018[5]). Wallonia, with its higher share of the low-educated concentrated in a rural setting, presents a further challenge for the low-educated based simply on the existing composition of urban geography.

Figure 1.10. A high share of the low-educated are located in towns and suburbs in Belgium

References

[4] OECD (2019), In-Depth Productivity Review of Belgium, OECD Publishing, Paris, https://dx.doi.org/10.1787/88aefcd5-en.

[1] OECD (2019), PISA 2018 Results (Volume I): What Students Know and Can Do, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/5f07c754-en.

[5] OECD (2018), OECD Regions and Cities at a Glance 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/reg_cit_glance-2018-en.

[3] OECD (2012), Closing the Gender Gap: Act Now, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264179370-en.

[2] OECD/European Union (2018), Settling In 2018: Indicators of Immigrant Integration, OECD Publishing, Paris/European Union, Brussels, https://dx.doi.org/10.1787/9789264307216-en.

[6] UNIA/FOD (2020), 4th Socio-economic Monitoring - Labour Market and Origin 2019, Federale Overheidsdienst Werkgelegenheid, Arbeid en Sociaal Overleg/UNIA, https://werk.belgie.be/nl/publicaties/4th-socio-economic-monitoring-labour-market-and-origin-2019 (accessed on 13 August 2020).

Notes

← 1. For a comprehensive overview of the employment prospects of people of foreign origin in Belgium see (UNIA/FOD, 2020[6]). The report uses a more comprehensive definition of migrants and the foreign born than this report, which uses place of birth to facilitate cross-country comparability.

← 2. Definitions of urban geography are based on the EuroStat definitions. Cities are defined as administrative units where at least 50% of the population lives in a 1km²grid with at least 1 500 inhabitants, and the city has at least 50 000 inhabitants total. Rural areas are defined as areas where more than 50% of the population living in an area with less than 300 inhabitants per 1km². Towns and suburbs fall in the middle by density and have at least 5 000 in habitants.

╳

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.

https://doi.org/10.1787/0140a728-en

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.

╳

: The Future for Low-Educated Workers in Belgium:

Continue to open the publication page on OECD iLibrary to find:

All available formats

Buy option for print or PDF copies

All publication details?

Cancel