2. The labour market outcomes of the low-educated

This chapter reviews the labour market for low-educated workers in Belgium. It focuses on two dimensions: job quantity and job quality. Along the job quantity dimension, Belgium has low employment rates for the low-educated, and unemployment rates which place it around the average compared to other European OECD countries. The dissonance between the two indicators can be explained by high rates of the population who leave the labour force due to disability or illness, and an elevated share of early retirement.

The overall picture of job quantity masks divergent labour market outcomes for the low-educated at the regional level. When restricting to just prime-age workers, the employment rate for the low-educated in Flanders is one of the highest among European OECD countries and its unemployment is one of the lowest. The corresponding employment and unemployment rates in Wallonia and the Brussels-Capital Region are among the worst in Europe.

Job quality for the low-educated in Belgium is relatively high compared to neighbouring countries. The low-educated in Belgium who are employed enjoy relatively high wages and low levels of non-standard work (temporary contracts, part-time work, and self-employment). There is much less regional variation in job quality than there is in job quantity. Earnings are relatively equal across regions, though Flanders enjoys somewhat lower levels of non-standard work and involuntary part-time employment.

Employment rates in Belgium are low, especially compared to neighbouring countries. In 2018, the employment rate1 for those aged 20-64 was 70.0%. For comparison, France had a slightly higher employment rate of 71.1%, while the rates in the Netherlands (79.3%) and Germany (79.8%) were significantly higher (Figure 2.1).

Although Belgium’s employment rate is low in comparison to neighbouring countries, it has risen substantially over the past two decades. In 1998, the employment rate for Belgium stood at 62.6% increasing to 68.0% in 2008 and then to 70.0% in 2018. The increase of 7.4 percentage points over the period 1998 to 2008 exceeded the 6.2 percentage point average increase for the European OECD countries in the sample. The 2 percentage point increase in Belgium between 2008 and 2018 was equal to the increase across European OECD countries in the sample.

Looking deeper at the three main regions in Belgium, Flanders (74.9%) has a significantly higher employment rate than Wallonia (63.9%) and the Brussels-Capital Region (61.7%). The employment rates of Wallonia and the Brussels-Capital Region are comparable to those of the poorest performing European OECD countries, while the employment rate in Flanders is similar to the average for European OECD countries (75.4%).

The gap between the regions in Belgium has increased over time. In 1998, the employment rate in Flanders stood at 65.5% which was 6.9 percentage points higher than in Wallonia and 7.5 percentage points higher than in the Brussels-Capital Region. By 2008, that gap had widened to 9.5 percentage points between Flanders and Wallonia and 12.1 percentage points between Flanders and the Brussels-Capital Region. The gap widened further between 2008 and 2018 although more modestly, to 11.0 (Wallonia) and 13.3 (the Brussels-Capital Region) percentage points, respectively.

Employment rates tend to be lower among the low-educated. For European OECD countries in 2018, the employment rate among the low-educated was 57.0% compared to 74.9% for workers with at least an upper-secondary degree, and 85.2% for those with a tertiary degree (Figure 2.2).

Belgium has one of the lowest employment rates among the low-educated. At 46.6%, the employment rate for the low-educated in Belgium is higher only than that of the Slovak Republic and Poland. By contrast, Iceland and Switzerland achieve employment rates for the low-educated of 76.6% and 69.8%, respectively. Neighbouring countries Germany, the Netherlands and France have rates of 60.8%, 62.6% and 51.6%, respectively – all significantly higher than in Belgium.

The employment rate of the low-educated also varies significantly within Belgium. Wallonia (38.9%) and the Brussels-Capital Region (41.9%) have some of the lowest employment rates compared to OECD countries in the sample. Flanders has higher employment rates among the low-educated (53.6%), though it too lags neighbouring countries.

Perhaps most concerning is the drop in employment rates among the low-educated in the Brussels-Capital Region and Wallonia over the past 10 years. Since 2008, employment rates overall have grown in Belgium, but that growth has not filtered down to the low-educated. Employment rates for the low-educated have fallen by 2.5 percentage points in Belgium. They remained mostly stable in Flanders, but fell by 1.5 percentage points in the Brussels-Capital Region and by 5.5 percentage points in Wallonia.

Employment rates are a robust guide to the underlying health of the labour market but they have some drawbacks. The employment rate makes no distinction between people who are looking for work and those who are not. The downside to this approach is that there are groups who quite clearly would not accept a job in any short time-span. Retired persons, people whose primary occupation is caregiving, and students are groups who are unlikely to consistently and actively seek employment. Countries with particularly generous retirement and education systems, for example, will tend to report lower employment rates even though their underlying labour markets may be performing well. Looking at unemployment rates, particularly side-by-side with employment rates, helps to remedy this problem.2

In 2018, the unemployment rate in Belgium stood at 5.8%, which was 3.3 percentage points lower than in 1998 and 0.9 percentage points lower than in 2008 (Figure 2.3). While Belgium’s employment rate puts it towards the lower end of European OECD countries in the sample, the unemployment rate puts Belgium solidly in the middle of the pack. The unemployment rate in Belgium is slightly lower than in France, but it significantly lags the Netherlands and Germany who have some of the lowest unemployment rates of OECD countries in the sample (3.4% and 3.3%, respectively).

The unemployment rate in Belgium masks even greater heterogeneity at the regional level than the employment rate. Unemployment rates of 8.4% in Wallonia and 13.2% in the Brussels-Capital Region place them near the bottom of European OECD countries, although with lower unemployment rates than France in the case of Wallonia. By contrast, at 3.3% in 2018, Flanders has one of the lowest unemployment rates among European OECD countries, on par with Germany and the Netherlands.

Breaking the unemployment rate down by education leaves this story unchanged. For low-educated workers, Belgium’s unemployment rate (12.8% in 2018) places it towards the middle of European OECD countries (Figure 2.4). For those with an upper-secondary qualification, an unemployment rate of just under 6.0% again places Belgium somewhere in the middle of European OECD countries. The same holds for the population with a tertiary degree with an unemployment rate of 3.2%, although in this case Belgium’s performance is slightly better than some peer countries (e.g. France, Sweden and Denmark). However, in just about every country in the sample, the unemployment rate for the population with a tertiary degree is uniformly low, and with less variance than the populations examined without a tertiary degree.

In contrast to employment rates where the performance of Flanders placed it in the middle of the pack, its unemployment is low among the low-educated. In 2018, Flanders had one of the lowest unemployment rates for low-educated workers among European OECD countries. With an unemployment rate of 6.2%, Flanders has a lower unemployment rate than Denmark, Norway and Germany. Only the United Kingdom, the Netherlands and Iceland had lower unemployment rates for low-educated workers. The difference between the relative performance of Flanders when using the employment rate and unemployment rate is mostly due to labour supply decisions (Section 2.3).

By contrast, Wallonia and the Brussels-Capital Region have some of the highest unemployment rates in the sample. Wallonia’s unemployment rate stood at 19.6% and the Brussels-Capital Region’s at 21.7% in 2018. Only Greece, Spain and the Slovak Republic had higher unemployment rates among the low-educated. Perhaps more alarmingly, while the unemployment in Flanders remained mostly stable over the last 20 years, the unemployment rate in Wallonia has actually risen over the past 10 years. The Brussels-Capital Region also experienced a decline in unemployment rates for low-educated workers over the last 10 years. The regional divergence in unemployment rates is therefore greater than the regional divergence in employment rates.

Long-term unemployment for the low-educated is high in Belgium compared to neighbouring countries. The share of the low-educated unemployed who have been unemployed for longer than one year stood at 62% in 2018. This rate is slightly higher than a decade earlier, but has fallen significantly compared to 20 years ago (Figure 2.5). The share of long-term unemployment is significantly lower in France (52%), Germany (50%) and the Netherlands (48%).

There are also regional disparities in the share of long-term unemployment, with the share being lower in Flanders (51%) than in the Brussels-Capital Region (68%) or Wallonia (65%).

Unemployment rates for the low-educated in Belgium, and especially in Flanders, look quite a bit better than one might expect by examining the employment rate. One problem with the unemployment rate is its inability to deal with issues around labour supply. The example of education and early retirement was previously discussed, but this also extends to countries with generous disability or unemployment insurance (Chapter 4), or countries with inadequate resources for child or elder care. In these cases, many potential workers may drop out of the labour force. They would therefore not be counted in the unemployment rate (possibly leading to a low unemployment rate), even though the employment rate might be low. This is the case of the labour market for the low-educated in Flanders, and to some extent the low-educated in Belgium overall.

A large share of non-employment in Belgium is due to disability or family caregiving responsibilities. Figure 2.6 shows the share of the population without an upper-secondary degree who are not in the labour force due to their own illness or disability, or who have personal or family caregiving responsibilities. The share in Belgium is 24.0% which is one of the highest shares of European OECD countries, behind only Poland and the Slovak Republic. Of the total, 14.9% is due to illness or disability, and 9.1% is due to caregiving.

While disability and caretaking duties explain a relatively large share of non-employment in Belgium overall, the differences across regions are relatively small. The Brussels-Capital Region and Wallonia have the highest rates at 24.7% and 27.2%, respectively, which is just higher than the overall average in Belgium. Flanders with a rate of 22.5% is below the overall average.

The share of the population in Belgium not available to work due to disability or illness has been rising. Figure 2.7 shows the share of the population unavailable to work due to disability or illness in both 2008 and 2018. The share has risen in Belgium from 8.3% in 2008 to 14.9% in 2018. Only France had a higher percentage point increase over this period.

The share of the population unavailable to work due to illness or disability rose at similar rates across regions. From 2008 to 2018, the share rose from 9.1% to 15.3% in Flanders (6.2 percentage points), 7.7% to 14.8% in Wallonia (7.1 percentage points), and 6.7% to 13.8% in the Brussels-Capital Region (7.1 percentage points). The large and proportionate rise in the share of the low-educated population unavailable to work due to illness or disability is holding down employment rates in Belgium, but it is unable to explain the large difference in employment rates across regions.

A relatively large share of the population aged 55-64 is not available to work in Belgium due to early retirement. The share of the population aged 55-64 in 2016 who were retired stood at 22.3% in Belgium. This was substantially higher than in neighbouring countries Germany (14.7%) and the Netherlands (8.9%), but less than in France (30.9%).

Just as with disability and illness, the share of the older working-age population unavailable to work due to early retirement cannot explain the large regional gaps in employment. The share in Flanders in 2016 is 23.6% compared to 24.3% in Wallonia. The Brussels-Capital Region has a significantly lower share unavailable to work due to retirement with 14.8%.

Although the share retired cannot explain the large regional difference in employment rates, it does appear to be holding down employment rates overall. Belgium has one of the lowest ages of effective labour market exit for older workers. Men leave the labour market at 61.6 years of age, on average, and women leave the labour market at 60.5 years of age. For comparison, the average age of effective labour market exit is slightly lower in France, but much higher in Germany and the Netherlands (Figure 2.8).

This helps explain the low employment rates for the low-educated older working-age population in Belgium. In 2018, 31% of low-educated workers aged 55-64 were employed compared to 40% in France, and 54% and 53% in Germany and the Netherlands, respectively.

Belgium’s relatively high share of migrants may also be holding down employment rates. The Brussels-Capital Region in particular has a high share of migrants (Chapter 1), whose employment rates are lower than those of natives. Belgium’s relatively high share of migrants is only a small contributing factor to lower employment rates compared to neighbouring countries.

Employment rates for low-educated migrants are generally lower than their native peers. The employment rate for the low-educated, native population in 2018 was 47% in Belgium compared to 45% for migrants (Figure 2.9). In Germany and France the rates were 63% and 52% for the native population and 59% and 51% for migrants, respectively. The Netherlands has the largest differential among neighbouring countries between native and non-native employment rates for the low-educated, with 66% of the native population employed compared to 49% for migrants.

Across regions within Belgium, the difference in employment rates between natives and migrants varies greatly. In Wallonia and the Brussels-Capital Region, employment rates for low-educated natives exceed those of migrants, while native employment rates remain low. In Wallonia, the employment rate is 34% for migrants, and 40% for natives. In the Brussels-Capital Region, the employment rate for migrants is equal to natives in Wallonia at 40%, but lower than the 46% of natives employed in the Brussels-Capital Region. Finally, the employment rate for low-educated migrants (57%) in Flanders exceeds that of natives (53%).

The preceding section examined gaps in employment rates between various groups within Belgium, as well as employment gaps between Belgium and neighbouring OECD countries. One question that naturally arises is how important the various employment gaps are for closing the overall difference in low-education employment rates between Belgium and neighbouring countries. Answering this question can help policy makers decide which sets of policies will be the most effective in raising employment rates for the low-educated.

Figure 2.10 shows what would happen to the overall employment rate in Belgium if some of the employment gaps were eliminated/narrowed. In other words, it shows what would happen to the overall employment rate if migrants, persons with disabilities, older workers, as well as the low-educated in Wallonia and the Brussels-Capital Region had employment rates similar to those observed amongst comparison groups. Note that these counterfactual employment rates are not cumulative; they present the effect of boosting the employment rate of just one group at the time. The current employment rates for Belgium and neighbouring countries are also presented for comparison.

Closing the employment gap between native-born Belgians and the foreign-born population would result in the smallest increase in overall employment rates for the low-educated. The employment rate for the low-educated in Belgium would move from 46.6% to 47.7%. The increase is modest because employment gaps between the native- and foreign-born are small in Belgium for the low-educated.

Boosting the employment rate of the population with disabilities is the next most effective at increasing the overall employment rate. The employment rate would move to 49.2% if the employment rate of people with disabilities in Belgium moved to the average of Germany, France and the Netherlands. The biggest changes would be observed among the young and working age population (20-54) where the largest gaps in employment due to disability exist.

Decreasing the rate of retirement among the older (55-64) working-age population would result in the second largest increase in the employment rate in Belgium. The counterfactual employment rate for the low-educated in Belgium would rise to 51.4%. The analysis assumes that the share of the older working-age population who are retired (22%) falls to the lowest level among neighbouring countries (the Netherlands, 9%).3

Bringing the employment rate of the Brussels-Capital Region and Wallonia to the level observed in Flanders would result in the highest counterfactual employment rate. In this scenario, the employment rate for the low-educated in Belgium would move to 54.9%, surpassing that of France. This case assumes that employment rates – by age – in Wallonia and the Brussels-Capital Region are equal to Flanders. The biggest change is among prime-age workers where Flanders has a high employment rate compared to other OECD European countries (while Wallonia and the Brussels-Capital Region have quite low employment rates). For all regions, employment rates for the older working-age population remain low as all Belgian regions have low employment rates among this group compared to other European OECD countries (see Section 2.3).

Employment rates for the low-educated in Belgium are low, although there is substantial regional variability. One possible reason for lower employment rates in Belgium compared to neighbouring countries could be higher job quality. Workers in Belgium may enjoy higher compensation and non-pecuniary benefits compared to neighbouring countries which increases the cost of hiring to employers and could result in lower employment. A similar dynamic could also be playing out across regions within Belgium: are better outcomes in Flanders related to lower job quality? This section reviews job quality for the low-educated in Belgium, and examines whether there is any trade-off between higher job quality and lower employment rates.

Belgium enjoys higher job quality compared to neighbouring countries, which combined with the lower rates of employment examined in the previous section might imply some trade-off between job quality and quantity. Within Belgium, however, Flanders not only performs better from a job quantity perspective, but also from a job quality one.

Workers in Belgium have high earnings compared to peer countries. Figure 2.11 shows the evolution of real gross median monthly earnings for Belgium overall, and by region, compared to neighbouring countries. Only the Netherlands has slightly higher median earnings than Belgium. Gross median earnings in the Netherlands are EUR 3 710 compared to EUR 3 530 in Belgium. In contrast, Germany has significantly lower gross median earnings at EUR 2 980 per month and gross monthly earnings in France are lower still with EUR 2 470 per month.

Despite large differences in employment rates across regions, median monthly earnings show little variation across regions in Belgium. Median monthly earnings in 2018 stood at EUR 3 630 in Flanders compared to EUR 3 420 in Wallonia and EUR 3 300 in the Brussels-Capital Region. Although earnings are slightly higher in Flanders, the difference is small compared to the difference with neighbouring countries.

The same pattern holds for the low-educated. Median real gross monthly earnings have grown steadily in all countries in the sample. Germany has the lowest monthly earnings at EUR 1 670 per month compared to EUR 2050 in France. In Belgium and the Netherlands, the earnings for low-educated workers are considerably higher, at EUR 2 760 and EUR 2 880 per month, respectively (Figure 2.12).

At the regional level, for low-educated workers, monthly gross earnings show little variation. The median monthly earnings in Flanders were EUR 2 740 for low-educated workers compared to EUR 2 820 in Wallonia and EUR 2 610 in the Brussels-Capital Region. The lack of variation in earnings across regions for low-educated workers – combined with diverging levels of productivity – offers one reason for the large divergence in employment rates across regions (Chapter 4).

Finally, it should be noted that, although overall median earnings for low-educated workers mimic the same patterns seen overall, this is not uniformly true by education level. Germany is a particularly interesting case. Although median earnings in Germany are significantly lower than in Belgium and the Netherlands for low-educated workers, for workers with an upper-secondary degree, they are considerably higher and, for workers with a tertiary degree, median earnings in Germany are on par with Belgium and the Netherlands. In fact, earnings show far less inter-country variation for workers with a tertiary degree compared to those without.

In addition to earnings, non-standard work constitutes an important dimension for low-educated workers. This section reviews the main dimensions of non-standard work: temporary employment, part-time employment and self-employment. Belgium has low rates of temporary employment, as well as relatively high rates of part-time employment among the low-educated with only small differences across regions. However, part-time employment is much more likely to be involuntary in the Brussels-Capital Region and Wallonia than in Flanders. Rates of self-employment are moderate and stable in Belgium, with only the Brussels-Capital Region showing high rates of self-employment.

The use of temporary contracts has become more prevalent in many OECD countries over the last 20 years. The use of temporary contracts often offers more flexibility to employers. However, this may come at the expense of job security as workers on fixed-term contracts are generally easier to dismiss in OECD countries compared to workers on indefinite contracts (OECD, 2013[2]).

The share of workers in Belgium on a temporary contract is not particularly high compared to other European countries. In 2008, 8.5% of workers in Belgium held a temporary contract (Figure 2.13). This rate is lower than in Germany, Greece and Sweden, and slightly higher than in Iceland, the Czech Republic and Norway. Among regions within Belgium, Flanders has the lowest incidence of temporary contracts. The share of workers on temporary contracts in Flanders is 9.7% compared to 12.6% in Wallonia and 15.7% in the Brussels-Capital Region. The use of temporary contracts does not appear to be a main explanation for the divergence in employment rates across regions.

The share of temporary contracts is higher among low-educated workers than among workers with a secondary and a tertiary qualification. For low-educated workers in Belgium, the share employed on temporary contracts was 11.3% in 2018 (Figure 2.14), compared to 7.9% and 8.1% among medium- and high-educated workers, respectively. Over the past 10 years, the incidence of temporary contracts has increased the most among the low-educated – by about 22% compared to 13% and 6% for the middle-educated and high-educated, respectively.

Differences across regions in the incidence of temporary contracts among the low-educated are similar to the regional differences observed for all workers. In Wallonia, the share of the low-educated workers on a temporary contract was 12.6% compared to 15.7% in the Brussels-Capital Region and 9.7% in Flanders. Wallonia and especially the Brussels-Capital Region have seen an increase in temporary contracts for low-educated workers over the past 20 years. This includes the last 10 years, during which employment rates for these groups have declined. The use of temporary contracts increased 2.2 percentage points in Wallonia and 4.4 percentage points in the Brussels-Capital Region. Flanders, where employment rates for the low-educated have risen modestly over the past 10 years, has seen a considerably milder increase of 1.5 percentage points.

A greater share of part-time work may be one way to explain Flanders’s decent employment rate and low unemployment rate relative to Wallonia or the Brussels-Capital Region. In this case firms in Flanders would rely to a much greater degree on a part-time work force, which would make the employment and unemployment rates rosier than they would at first appear. A work force of many part-time workers allows firms to tailor their staffing decisions more easily to daily or weekly shifts in labour demand.4 Such production processes are quite common in many retail and service industries which employ many low-wage workers (Tilly, 1991[3]; Valletta, Bengali and van der List, 2020[4]).

At the same time, many people choose to work part-time in order to attend to caregiving, gradually retire or simply enjoy a better work-life balance. The case of part-time work and partial retirement is now an explicit strategy for some OECD countries to keep older workers attached to the labour force (OECD, 2019[5]). The reasons a country might have a high share of part-time work are therefore diverse, and it necessitates caution when interpreting inter-country differences.

Belgium has a relatively high share of part-time employment compared to other European OECD countries and this share has grown over the past 20 years. In 2018, the share of part-time work in Belgium stood at 24.3%, which is a sharp increase from the share in 1998 at 15.6% (Figure 2.15). The share of part-time employment in Belgium is lower than in Germany (26.7%) and the Netherlands (46.8%), but higher than in France (17.5%).

There is little variation in rates of part-time work between Wallonia and the Brussels-Capital Region, while Flanders has a slightly higher share. In 2018, the share of part-time work in the Brussels-Capital Region stood at 21.5% compared to 22.6% in Wallonia and 25.6% in Flanders.

Low-educated workers are more likely to work part-time. In 2018, the share of low-educated workers working part-time in Belgium was 30.3% (Figure 2.16). In Wallonia it was 30.2% and 31.9% in Flanders, while in the Brussels-Capital Region it was much lower at 23.6%.

All regions have seen a substantial increase in the share of part-time work among the low-educated. Since 1998, the share of low-educated workers working part-time increased by 13.3 percentage points in Wallonia, 15.8 percentage points in Flanders and 8.2 percentage points in the Brussels-Capital Region. While the share of part-time work also increased for both middle- and high-educated workers, the percentage point increase was more modest.

One of the challenges in interpreting part-time employment figures is whether or not it is voluntary. Part-time work may be a sign of affluence (people enjoying higher incomes with more leisure), or people with caregiving responsibilities. It can also be a sign of labour market distress with not enough hours available compared to desired hours. Involuntary part-time, defined here as the share of employed workers who are part-time and would like more hours, or those who would rather work full-time, is a measure of what share of part-time work is involuntary.

Involuntary part-time work in Belgium is relatively low compared to other countries, although it is higher in Wallonia and especially in the Brussels-Capital Region. Figure 2.17 shows the share of employed workers who say they would like more hours than their employers are offering them. In general, the incidence of involuntary part-time in Belgium is fairly low (5.0%). The incidence is lower in Flanders (3.9%) than in Wallonia (5.9%) and the Brussels-Capital Region (9.4%).

In Wallonia, the share of involuntary part-time among low-educated workers is slightly higher than the OECD average. Wallonia’s rate (10.2%) is lower than in France and about equal to the Netherlands and Germany. Flanders’s rate of involuntary part-time is lower with 5.7%. The Brussels-Capital Region performs particularly poorly with as many as 12.9% of low-educated workers working part-time involuntarily.

Figure 2.18 shows the same employment rates for the low-educated as in Figure 2.2, but decomposed into the shares of full-time, part-time and short part-time (defined as usual weekly hours less than 20). Belgium’s share of the low-educated population in full-time employment (32.4%) is one of the lowest among European OECD countries, as well as the Netherlands (32.7%), France (38.5%) and Germany (40.1%).

Germany and the Netherlands in particular use much more part-time work, though they differ in the details. In Belgium, 14.1% of the low-educated population is engaged in part-time employment with it almost equally split between part-time and short-part time. In contrast, the Netherlands has almost as many of their low-educated population in part-time (29.8%) work as full-time (32.7%). Germany achieves higher overall employment rates with a relatively high share of the low-educated population engaged in short part-time work (13.7%).5 Only France has comparable rates of part-time employment among the low-educated (13.1%).

The share self-employed in Belgium (13.3%) is about average compared to other European OECD countries and it has receded slightly over the last 20 years (Figure 2.19. The share of self-employment in Belgium is lower than in the Netherlands, but higher than in France and Germany.

Wallonia and Flanders have almost equal shares of self-employment (13.0% and 12.7%, respectively) while the share in the Brussels-Capital Region (17.1%) is higher. Over the last 20 years, Wallonia and Flanders have seen a small decrease in self-employment while the share in the Brussels-Capital Region has remained largely flat.

The same patterns hold by education, with shares of self-employment in Wallonia and Flanders lower among the low-educated. In 2018, the shares of self-employment among low-educated workers were 8.7% and 9.0% in Flanders and Wallonia, respectively, compared to 9.4% in Belgium overall (Figure 2.20). The share of self-employment has also been falling among low-educated workers over the past 20 years.

TWA employment refers to situations where a firm (lead or contracting) contracts with an employment agency (a temporary work agency) to provide labour on a temporary basis. TWA employment allows firms to fill temporary employment needs, as well as screen workers for direct hire positions. TWA firms may also provide free training to their workers to help with placement in a lead firm and act as a stepping stone for labour market entrants. Critics of TWA employment claim that it does not allow workers to build human capital and discourages meaningful job search resulting in workers cycling through temporary jobs (Autor, 2007[6]).

TWA employment is high and rising for low-educated workers in Belgium. In 2018, 4% of low-educated workers were employed through a TWA. This increased from 3% in 2008. Low-educated workers have a higher incidence of TWA employment compared to the overall employed population in Belgium, which sits at 2% in 2018. Part of this rise may be due to changes in employment protection legislation, which enlarged the number of reasons for using TWA employment (see Chapter 4, Section 4.7).

Compared to neighbouring countries, the incidence of temporary work agency is low in Belgium. In 2018, the Netherlands had the highest incidence of TWA employment of European OECD countries in Figure 2.21 with 7%. Germany had the second highest share (6%) and France had still a higher incidence than Belgium with 5%. Just as with Belgium, all countries have seen an increase in TWA employment over the past ten years.

At a regional level, the share of temporary work agency employment is higher in Flanders (4%) than in Wallonia (3%) or the Brussels-Capital Region (2%).

References

[6] Autor, D. (2007), “Introduction to “Studies of Labor Market Intermediation””, in Autor, D. (ed.), Studies of Labor Market Intermediation, University of Chicago Press, http://www.nber.org/books/auto07-1 (accessed on 3 July 2020).

[7] ILO (2008), Beyond Unemployment: Measurement of Other Forms of Labour Underutilization 1.

[8] Jacobi, L. and J. Kluve (2007), “Before and After the Hartz Reforms: The Performance of Active Labour Market Policy in Germany”, Journal for Labour Market Research, Vol. 40/1, https://www.iza.org/publications/dp/2100/before-and-after-the-hartz-reforms-the-performance-of-active-labour-market-policy-in-germany (accessed on 10 July 2020).

[1] OECD (2019), “Effective age of labour market exit”, in Pensions at a Glance 2019: OECD and G20 Indicators, OECD Publishing, Paris, http://dx.doi.org/10.1787/888934042029.

[5] OECD (2019), Working Better with Age, Ageing and Employment Policies, OECD Publishing, Paris, https://dx.doi.org/10.1787/c4d4f66a-en.

[2] OECD (2013), OECD Employment Outlook 2013, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2013-en.

[3] Tilly, C. (1991), “Reasons for the continuing growth of part-time employment”, Monthly Labor Review, Vol. 114/3, pp. 10-18, http://dx.doi.org/10.2307/41843559.

[4] Valletta, R., L. Bengali and C. van der List (2020), “Cyclical and market determinants of involuntary part-time employment”, Journal of Labor Economics, Vol. 38/1, pp. 67-93, http://dx.doi.org/10.1086/704496.

Notes

← 1. The employment rate is the share of the entire population employed at a given point in time. It includes in the denominator both people who are unemployed and searching for a job, as well as people who are retired, in school, or otherwise not searching for employment. The employment rate is a robust measure of employment in that it does not depend on whether people claim to be actively looking for employment, which can be fickle and unreliable even over short time spans (ILO, 2008[7]).

← 2. Unemployment rates ask people if they are actively looking for work. They therefore give a good indication of a person’s likelihood of finding a job if they want one. They also exclude the population who says they are not actively looking for work which helps with populations who are not in the labour market. If there are many individuals who are willing and able to work, but for whom no work exists, this will show up in high unemployment rates, and it likely means labour markets are not working properly.

← 3. Moving to the average of France, Germany and the Netherlands (18%) results in a more modest increase in overall employment rates (48.1%).

← 4. In addition to managing fluctuations in labour demand, there is some evidence that full-time workers earn a pay premium relative to part-time workers even after accounting for worker characteristics. Part-time employment may therefore directly reduce labour costs for employers.

← 5. Germany’s high share of short part-time work for the low-educated is likely due to the Hartz labour market reforms in the early 2000s which greatly expanded “mini jobs”, which are usually part-time jobs paying less than EUR 450 a month (Jacobi and Kluve, 2007[8]).

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 2020

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.