3. How does migration shape industry structure?

Abstract

Migrants are not equally distributed across the economy. They are strongly concentrated in a number of sectors, especially in low-skilled services but also in information technology and manufacturing. This raises the question of the impact of migration on sectoral development. With the asymmetric employment impact of the COVID-19 crisis across sectors, and the essential role of migrants in certain key sectors, there is a renewed interest in the links between sectors and migration. This chapter provides a first comprehensive comparative overview of the presence of migrants across sectors in OECD countries. It also assesses how migration policy instruments are associated with specific sectors, with a focus on agriculture, hospitality, and IT. Migration has sectoral implications, whether these are intentional or not. Being aware of the sectoral dimension of migration is a precondition for reaping potential benefits for specific sectors where intended, as well as for avoiding unintended adverse impact.

In Brief

Key findings

Migrants are not equally distributed across the economy. There are large differences in the sectoral presence of migrants, with a strong overrepresentation in services sectors with high shares of lower-skilled workers. In OECD Europe, Israel and the United States, while the overall share of migrants in the workforce is 16%, 24% and 18%, respectively, the share of migrants across sectors varies from 53% (United States: 50%, Israel: 66%) in domestic services and 27% (United States: 24%) in accommodation and food services, to 6% (United States: 9%) in electricity, gas and air conditioning supply and 6% (United States: 9%, Israel: 11%) in public administration and defence.
In the settlement countries, such as Canada, Australia and New Zealand, the overall share of migrants is respectively 25%, 30% and 32%. While migrants are also over-represented in hospitality and support service activities in these countries, sectors with higher shares of workers with highly skilled occupations also concentrate many migrants. These include in particular information and communication (33%) and real estate (32%) in Canada, finance in Australia (almost 37%) and health and social care in New Zealand. Migrants are under-represented in public service sectors, although to a lesser extent than in Europe or the United States, and not in New Zealand.
Between 2005 and 2018, migrants have contributed to more than half of the overall increase of the employment in five sectors in the United States (finance, transport and storage, agriculture, information and communication, and other services). In Europe, this is also true for the hospitality sector. Migrants were also the only contributors to the employment growth in domestic services in Europe, Israel and the United States.
Concentration across sectors differs between countries. While in most longstanding European destination countries, Australia, and the United States, migrants are overrepresented in services sectors with a large percentage of low-skilled employment, this is not the case in Canada, Central and Eastern European countries, where migrants tend to be more concentrated in some higher-skilled manufacturing sectors and Information Technology (IT). Foreign nationals in Japan and migrants in the United States and Israel also tend to be concentrated in many manufacturing industries.
Concentration has increased over time. In Europe, sectors in which migrants were already strongly overrepresented in 2005 have also seen the strongest growth in migrant employment by 2018, while the reverse was the case in sectors where migrants were under-represented. The picture was more uneven in Canada, Australia, Israel and the United States, however.
Concentration declines with duration of stay. Recent arrivals are a lot more concentrated in certain sectors than longstanding migrants. Recent migrants are disproportionately often found in accommodation and food services. They are also over-represented in the information technology sector, which has become an important entry sector for newcomers since the early 2000s. In contrast, they are much less likely to be employed in the health and social services sector, compared with their settled peers, because they must satisfy certain employment requirements – notably for foreign credential recognition – that may take years to complete.
In most European OECD countries, migrants from EU countries tend to be concentrated in similar sectors as migrants from non-EU countries, but usually in a higher position. Indeed, in European countries, the share of EU migrants in low-skilled occupations is significantly lower than that of their non-EU peers in all sectors (bar real estate). There are significant differences in the sectoral concentration between the EU and non-EU migrants only in Luxembourg, the Netherlands and the United Kingdom.
While migrants are most often found in the sectors with a large share of workers in low- skilled occupations, they are also overrepresented among the highly skilled occupations in these sectors, with agriculture being a notable exception.
Relative to their native-born peers, migrant men and women are both over-represented in domestic services, hospitality and services to buildings. This pattern is the same in the United States and Europe. The one notable exception is construction, where migrant men are concentrated but migrant women are not.
The strong differences of migrants’ presence across sectors and the observed differences across countries and migrant groups raise the question to what degree migration policies may affect employment distribution by sector.
Migration policy instruments that favour specific sectors are quite limited notably in OECD settlement countries and Europe. Australia has however implemented since 2019 a programme targeting very highly skilled workers in seven top sectors. From 2021 onwards, New Zealand will also implement sector agreements to facilitate the recruitment of temporary migrant workers. Italy is one example that has had a sector-based approach for some of its regularisation programmes implemented over the last decades. To the contrary, Japan, Korea and Israel have traditionally given more weight to sector-based migration policies.
The agriculture sector appears as one exception. Almost half of all OECD countries have a specific programme for this sector, which is often viewed as essential not only in ensuring food supply, but also with respect to regional and rural development. Most programmes support seasonal peaks in labour demand. It is not uncommon in OECD countries that the majority of seasonal agricultural workers are migrants. They account for 90% of such workers in Germany, for example. All migration programmes for this sector are temporary, often of the circular kind with the same migrants coming back every year. In Europe, the importance of managed seasonal work programmes has declined strongly since the EU enlargements of 2004 and 2007, as most seasonal agricultural workers came from these previous accession countries and now benefit from free mobility.
Working Holiday Maker (WHM) and similar youth mobility schemes are important non-economic temporary migration categories, in which the visa holders are able to take a job without undergoing the usual requirements for admission as a labour migrant. While the hospitality sector tends to be the largest employer of WHM, in some countries many of them are also working in agriculture. In Australia, for example, an estimated 81 000 WHM worked in agriculture in 2018, almost ten times the number of migrants admitted as foreign seasonal workers.
While no OECD country has specific migration policy instruments in place to support the hospitality sector, still about one fourth of total employment in Europe, Australia, New Zealand, Canada and the United States in this sector are migrants. The non-standard working times and working opportunities concentrated during the holiday season may explain this result.
While migrants are dramatically over-represented in domestic work virtually everywhere, few OECD countries have specific migration policies for this sector. One of the exceptions are Israel and Canada. The latter has a large and longstanding programme for domestic caregivers. In Canada in 2016, 60% of the domestic personnel were migrants, most of whom were recent arrivals.
While there are no specific programmes directly targeted at the IT sector with the exception of Japan, it strongly benefits from a broad range of policies favouring migration of IT professionals. This profession is regularly included in shortage occupation lists where they exist, and some countries such as Germany have specific waivers for qualification or skills requirements that otherwise apply. In the United States, half of all foreign-born workers in the IT sector were from India in 2018. In Europe, the region of origin of foreign-born workers in the sector is much more diverse.
Consideration of sectoral elements in migration policy can support specific sectors where intended. Being aware of this dimension is also necessary to avoid unintended adverse impact. This is particularly important in a context where OECD countries are facing a strong economic crisis, which tends to have an uneven economic impact across sectors.

Introduction

There is extensive literature on the economic impact of migration in OECD countries, notably on the labour market, with a wealth of studies on how migration affects employment and wages of native-born. For instance, extensive research focused on how migration affects the job polarisation (i.e. the growth at the bottom and at the top of the skills distribution). Migrants tend to specialise in manual intensive tasks and push local low-skilled workers towards non-manual jobs, and highly skilled ones to better wages (Murphy and Oesch, 2018[1]; Foged and Peri, 2015[2]). Trade economists have also long shown the importance of factor mobility on the distribution of employment by sectors notably according to comparative advantages. However, migration and labour economists have paid little attention so far to the importance of migration on sectoral development and to the extent to which easier access to foreign labour force may influence the localisation choice of firm and economic growth in specific sectors.

Many factors affect industry structure, including technology, labour and capital – both regarding their nature and regarding intensity. Though migration can affect all factors, its predominant effects are on the price, composition and availability of labour. To what degree the sectoral 1 structure of an economy is shaped by the supply of labour – whether migrant 2 or native-born – in turn depends on a range of factors. These include sector competitiveness, its labour intensity, the tradability of the goods and services it produces 3, and whether the good or service is a final or intermediate product. These factors can change over time and are interlinked. Technological change, may also change the composition of labour demand as it tends to be skills biased (Acemoglu, 2002[3]). Skill-biased technological change has actually affected the job polarisation, which has led to a growth of some sectors and shrinking of others. However, investment in technologies, saving labour – notably of the low-skilled kind – depends on the price and scarcity of labour. For example, after the “Bracero” agreement termination in 1964 in the United States, employers appear to have adjusted to a more stringent access to foreign-worker by changing production techniques where that was possible (Clemens, Lewis and Postel, 2018[4]). Some countries with high entry wage and limited low skilled workforce (e.g. Nordic countries and Japan) have experienced higher automation of low skilled services. In other contexts, some authors (Basso, Peri and Rahman, 2017[5]) have indeed shown for the United States that unskilled immigration may attenuate the drop in routine employment proceeding from technological change.

The question of the impact of migration on sectoral structure is complex because it is associated with a plethora of different factors, including behaviours of economic actors (such as domestic and foreign investors, employers regarding technological choices, education and training decisions for young people, etc.). It is also linked with country endowment (potential for tourism and agriculture, etc.).

Isolating the impact of labour migration policy rather than migration is even more challenging as in virtually all OECD countries, only a minority of migrant workers are admitted specifically for purposes of employment, although settlement countries (Australia, Canada and New Zealand), put an emphasis on high human capital in their migration policy, through Expression of Interest or Point-Based systems. What is more, labour migrants who have been admitted for a specific job can change sectors thereafter – at least once they have obtained permanent residence or after some years of stay. Nevertheless, there is the important question to what degree labour migration policies may affect offshoring (Ottaviano, Peri and Wright, 2010[6]), foreign and domestic investment in specific sectors, as well as employment and productivity growth.

This chapter tries to shed some light on the links between migration and industry structure by first assessing the presence of migrants across sectors and countries, and then taking a closer look at the immigration policy parameters that affect sectoral distribution of employment. This issue is particularly pertinent now as with the strongly asymmetric impact of the COVID-19 crisis across sectors, and the essential role of migrants in key sectors, there is a renewed interest in the links between sectors and migration.

The first section provides a general overview of the presence of the foreign-born across sectors, and how their composition has evolved over the past two decades. This is followed by an analysis of the concentration of migrants in relation to sector growth and an assessment of how migrant characteristics are associated with sectoral distribution. The next section takes a closer look at the links between migration policy and sectors, starting with an overview of sectoral impact of migration policy design, followed by a closer look at the examples of seasonal agriculture, the hospitality industry (often also referred as Hotels, Restaurants and Catering – Horeca), and the Information and Technology (IT) sector. The final section concludes with the implications of the analysis for the links between migration policy and industrial policy-making, including in the context of the COVID-19 crisis.

Overview of the presence of migrants across sectors in OECD countries

The presence of migrant workers in the labour markets of OECD countries has increased over the last 15 years virtually everywhere. While 9% of the population in employment were foreign-born in European OECD countries in 2005, that share reached 14% in 2018 (Figure 3.2). The corresponding shares were 16% and 18%, respectively, in the United States, 26% and 30% in Australia, 22% and 32% in New Zealand. In 2016, the share of foreign-born among the employed in Canada was 25% and 1% in Japan in 2015. In 2019, it was 24% in Israel, a notable exception among OECD countries, since that share was 33% in 2005. However, migrants are not equally distributed across sectors. This section provides an overview of the presence of migrants across sectors and how it has evolved over time. Data by sector used two sets of industry classification – one aggregated and one disaggregated – to avoid sample size issues as much as possible (see Annex 3.A). The disaggregation level has obvious impact on the analysis, as breakdowns that are more detailed tend to reveal larger differences between migrants and the native-born. The most detailed classification has been used wherever possible, to strengthen the conclusions drawn.

The presence of migrants across sectors and their evolution

The contribution of migrants to the population in employment varies widely across sectors in all OECD countries. Services sectors with the highest shares of workers in lower-skilled occupations are also the sectors in which the share of migrant workers is the highest. In contrast, migrants are under-represented 4 in public administration and in sectors with high shares of highly skilled employment, with the notable exception of Information Technology (IT). This holds in Europe, Israel and in the United States (Figure 3.1). The sector in which migrants are most strongly present are the so-called “activities of households as employers”, which includes domestic personnel employed by households. A large and growing number of migrants provides services to households in virtually all European countries, Israel, Canada (see Box 3.2), and the United States. In the latter, the share of migrants in employment by private households has increased from 39% in 2005 to almost 50% in 2018 (Figure 3.2). The trend was even stronger in European countries, where the share increased from 36% to 53%. In three quarters of OECD countries, the share of migrants in domestic personnel is at least twice as high than in the economy overall. While both the number and the share of migrants employed in this sector has grown, not only the share but also the number of native-born providing these services has dropped virtually everywhere.

Figure 3.1. Share of migrants in OECD countries, by sector, 2018

In Europe, the sector with the second-largest shares of migrants is hospitality with 27%, followed by administrative and support service activities (22%), which include security and cleaning services to buildings. In the United States, and Oceanian OECD countries (Figure 3.1), these two sectors are also in the top five (with 24% and 28%, respectively in the United States, 37% for both sectors in Australia, and 44% and 35% in New Zealand), though construction in the United States (27%) and Finance in Australia (almost 37%) actually comes third. In New Zealand, behind hospitality, the highest shares of migrants are in information, media and telecommunication and health care and social assistance, with 36% each. In Israel, while migrants are actually under-represented in hospitality (20%), they are over-represented in administrative and support services (35%) and in manufacturing (36%). In Canada, by contrast, while migrants are also over-represented in hospitality and support service activities, the sectors with the largest shares of migrants are those with higher shares of workers with highly skilled occupations. These include in particular information and communication (33%) as well as real estate (32%). Concentration of foreign-born in sectors with a large share of highly skilled employment is linked with the country’s focus on higher capital economic migration (see OECD (2019[7]) for a discussion of the key features of the Canadian migration system). In Japan, given the overall small number of migrants, the share of migrants remain low in all sectors, but it is almost three times the average in manufacturing (3%) and twice in hospitality.

In contrast, migrants account for only a small share of the employed in energy supply (6% in Europe, 9% in the United States and 16% in Canada), mining and quarrying (9% in Europe, 13% in Canada, 15% in Israel and New Zealand, and 24% in Australia), arts and entertainment in the United States, Australia and New Zealand (12%, 21% and 24%) and water supply and sewerage in Canada (14%). In all OECD countries, however, with the exception of New Zealand, migrants are under-represented in public service sectors 6, although to a lesser extent in Australia and Canada. In public administration, the share of migrants is only 0.3% in Japan, 6% in Europe, 9% in the United States and 11% in Israel (Figure 3.1). In the human health and social work sector, which includes both public and private services, 14% of the population in employment was foreign-born in 2018 in Europe, 17% in the United States, 25% in Canada, 33% in Israel, 34% in Australia and 36% in New Zealand.

Beyond these general trends, there are a few country specificities. In the Baltic countries, Canada, Iceland, Japan, Israel, the Netherlands and the United States, migrants are concentrated in manufacturing, especially in textile industry (see Annex Table 3.B.1). In Central European countries, there is no clear pattern, and foreign-born tend to be present in quite heterogeneous sectors. In the Southern European countries, migrants are particularly under-represented in both public services and in the sectors where shares of highly skilled workers are the highest, such as finance, information-communication, and professional, scientific and technical activities 7. Compared to the average of OECD-Europe, the importance of migrants in domestic services is more limited in Sweden and Finland. A number of countries also have a high share of migrants in IT, such as the United Kingdom, Canada, the United States, France, Ireland, Finland and Luxembourg.

Looking at the evolution since 2005, sectoral disparities in the presence of migrants have grown. The share of migrants in the sectors, which already had a high share of migrants in 2005, has also grown disproportionately since, while the reverse was the case in sectors where migrants were under-represented.

Figure 3.2. Share of foreign-born in OECD countries, by sector, 2005 to 2018

While this picture is rather clear-cut in Europe, there are many exceptions in Israel, where the disparities between migrants and natives have dropped in half of the sectors, although neither in domestic work, where the presence of migrants has increased, nor in the public administration, where they were even more under-represented than in 2019. There are also some notable exceptions in Australia and the United States (Figure 3.2) – notably a decline in manufacturing and, in the United States, in the Horeca sector. Most European countries and Australia, in contrast, experienced an increase in migrants’ presence in this sector. It was particularly pronounced in Austria, Germany, Ireland, Spain and Nordic countries.

Other services sectors involving many low-skilled workers have also seen migrants becoming a larger part of their labour force. In the transportation and storage sector, the share of migrants has increased by almost 6 percentage points over the last ten years 8, in both Europe and the United States (Figure 3.2), with even higher increases in the German- speaking countries, Australia and the Nordic countries.

The share of migrants has not only increased in sectors with large shares of low-skilled jobs. Both in European countries and the United States, migrant workers are playing an increasing role in high-value-added sectors such as finance, real estate or information and communication. In the latter, the share of migrants reaches almost 17% in European countries and 29% in the United States, where the share of migrants was already 21% in 2005.

Concentration of foreign-born across sectors and their evolution

On average in the OECD, one fifth of migrants would need to change sectors to reach the same sectoral distribution as their native peers (Figure 3.3). Sectoral polarisation is particularly high in Luxembourg, Japan and the Southern European countries (bar Portugal). In these countries, more than a quarter of migrants would need to change sectors to reach the same distribution as the native-born. By contrast, it is relatively low in non-European settlement countries, the United Kingdom, Switzerland, Portugal, Baltic countries (bar Estonia), Hungary and the United States where less than 15% of migrants would need to change sector to equal native distribution. Although differences in sectors between foreign- and native-born are important, they are, however, only half of those observed between men and women, both in Europe and in the United States.

Compared with 2005, the concentration of migrants in certain sectors has grown virtually everywhere. There are only a few exceptions such as the Southern European countries (bar Italy) and Switzerland, where migrants are less sectoral-concentrated than before.

Recent migrants tend to be even more concentrated in specific sectors. Newcomers tend to have fewer choices at arrival, because either their work options are constrained by their permit, or because access to certain jobs takes more time (learning the host-country language, recognition of skills, etc.). Many new arrivals initially take a job below their formal qualifications and rely on their networks. In contrast with the overall increase in the concentration of migrants, the evolution for recent migrants is more mixed. It has actually decreased since 2005 in about half of the countries, especially in countries with previous high levels of dissimilarity such as in Central and Eastern Europe (Estonia, Lithuania, Poland and Slovenia).

Figure 3.3. Sectoral concentration of migrants has increased in most countries

Differences by migrant characteristics

Migrant presence across sectors by occupational skills levels

As seen above, migrants are disproportionately present in sectors with high shares of lower-skilled employment 9. This raises the question whether this could be due to the fact that migrant workers are in general concentrated in low-skilled jobs. Figure 3.4 sheds some light on this issue by looking at the share of migrants by sector and skills level for European countries. In sectors where migrants are overrepresented, this overrepresentation generally holds for both high- and low-skilled employment. In other words, migrant workers, in both Europe and Israel, are concentrated regardless of their skills level.

There are a number of exceptions to this pattern, however. In hospitality, migrants in Europe are even more over-represented among highly skilled workers (mainly managers) than among the low skilled. In agriculture, migrants in both Europe and Israel are strongly overrepresented among the low-skilled workers, but under-represented among those in high-skilled occupations (mainly farmers and other specialised workers). In a number of services sectors with high shares of highly skilled workers, especially in the information and communication sector, the reverse is the case – migrants being overrepresented in highly skilled occupations but not in low-skilled ones.

Figure 3.4. Share of migrants in OECD European countries among low- and highly skilled workers, by sector, 2016-18

Box 3.1. Gender differences in the sectoral distribution of migrants

Migrant men and women tend to be overrepresented in the same sectors. At the same time, sectors where migrant men or women are over-represented are the same where native-born men or women are also present. There is one notable exception, however. In accommodation and food services, migrant men are over-represented compared with migrant women, while there is a higher share of women than men in that sector among the native-born (Figure 3.5).

Unskilled female migrants tend to work in less-regulated and less visible sectors than male migrants (O’neil, Fleury and Foresti, 2016[8]). In both the United States and Europe (according to national sources described in Annex Table 3.B.1), around one fifth of migrant women work in the health and social care sector, one tenth in hospitality and one tenth in retail trade. In Israel, one migrant woman in four work in health and social care and one tenth in education. Compared with native-born women, foreign-born women are especially over-represented in domestic work and in services to buildings, such as security and cleaning. They are also dramatically over-represented in hospitality in both Europe and the United States, and in manufacturing and health and social care in Israel. Foreign-born men are over-represented in the same sectors as women, although to a lesser extent in services to buildings and in domestic work. In both the United States and Europe, transportation-storage and construction are the only sectors where foreign-born men are concentrated but foreign-born women are not (i.e. they are relatively more native women in these sectors than foreign-born women).

Figure 3.5. Presence of migrants across sectors in Europe, by gender

Differences between EU and non-EU migrants in Europe

Given the different channels through which EU and non-EU migrants enter European OECD countries, it is of particular interest to see to what degree the sectoral presence of these two groups varies. In most European countries, there are few differences between these two groups. In OECD Europe as a whole, the sectors with the largest overrepresentation of migrants are essentially the same for both groups (Annex Table 3.B.2 and Annex Table 3.B.3). EU migrants, however, usually have higher skilled occupations. Indeed, in OECD European countries, the share of EU migrants who are in low-skilled occupations is significantly lower than that of their non-EU peers in all sectors (bar real estate).

There are sectors, however, in which EU migrants are much more concentrated than their non-EU peers are. The first is construction, where posted workers within Europe account for a significant proportion (about 1 million intra EU/EFTA postings in the sector per year). The other exception is textile, where non-EU migrants are overrepresented in some countries, such as Germany and Italy, while EU migrants are overrepresented in this sector in the Netherlands and Sweden.

Only in few countries, there are strong differences between the two groups, including Luxembourg, the Netherlands and the United Kingdom. In most OECD European countries, domestic services is the sector with the strongest concentration of both EU and non-EU migrants. In Luxembourg, this is only the case for migrants from other EU countries. The reverse is true in the Netherlands. In that country, indeed only non-EU migrants follow the global pattern observed for migrants across Europe – that is, a strong concentration in domestic services, hospitality, and services to buildings (security, cleaning) – while EU migrants are concentrated in manufacturing industries.

The United Kingdom is the country where the sectoral presence of EU migrants differs most from that of their non-EU peers. While EU migrants tend to be in similar sectors as elsewhere, non-EU migrants – unlike most other European OECD countries – are largely under-represented in accommodation and warehousing but strongly concentrated in IT, human health and land transport.

Differences between recent arrivals and settled migrants

Recent migrants (with less than five years of residence) enter the EU and the United States labour market through basically the same main sectors in which their settled peers (with more than ten years of residence) are already strongly represented: domestic services, hospitality, warehousing, security/cleaning (mainly industries with a high share of lower skilled workers, see Annex Table 3.B.4 and Annex Table 3.B.5).

By contrast, the sectoral presence of recent migrants in Canada and Israel differs most from that of their settled peers. In Canada, recent migrant shares are the highest in domestic services, hospitality and food manufacturing, while settled migrants are especially over-represented in real estate, and other manufacturing industries (textile, electrical equipment and electronics). IT is the only sector where both recent and settled migrants are largely concentrated. In Israel, migrants are the most over-represented in domestic services and accommodation, while settled migrants are more concentrated in various manufacturing sectors. Only in electrical equipment manufacturing are both recent and settled migrants concentrated.

In Europe, if food services are one of the top sectors for both recent and settled migrants virtually everywhere, this is not true for accommodation. Indeed, in the Netherlands, Sweden and the United Kingdom, accommodation is one of the top entry sectors for recent migrants, but settled migrants are somewhat less represented in that sector.

IT sector has been an important entry sector for highly skilled migrants since 2005. In the United States, for instance, the share of recent migrants working in that sector was only 2% of the total recent migrant workforce (arrived after 1999) in 2005, while 7% of recent migrants arrived after 2013 worked there in 2018. In European countries, almost 4% of recent migrants work in IT, compared with only 1.6% of their settled peers (Annex Table 3.B.5).

Figure 3.6. Recent migrants by sector in OECD countries, 2016-19

Settled migrants are much more likely than recent arrivals to work in manufacturing virtually everywhere, with the exception of the United States (Figure 3.6). This sector is less likely to be an entry sector for recent migrants than 13 years ago. This is also true for domestic services in Europe. Unlike Canada and Israel, where recent migrants are respectively 7 times and 15 times more likely than settled migrants to be employed in domestic work, settled migrants in Europe are much more concentrated than recent arrivals in that sector. This is partly due to changes in the labour market conditions in Southern Europe. Due to high demand for domestic workers in Southern European countries before the global financial crisis, a quarter of all recent migrants worked in that sector in 2005, compared with only 7% in 2018.

Recent migrants are more likely to enter the European labour markets through finance or scientific and professional activities than migrants living in their host-country for at least ten years (Figure 3.6). This is not true in the United States, New Zealand and Canada. In the former two, recent migrants are as represented as settled migrants and in the latter, settled migrants are even more represented. Recent migrants are also significantly overrepresented in the education sector in the United States (Figure 3.6). However, this high share is linked with the large number of enrolled international students that have a teaching occupation during their studies. Excluding migrants currently in education, the share of recent migrants in the education sector is only 11%, below the one for settled migrants.

Although at least 9% of recent migrants in the United States, Canada, New Zealand, Israel and Europe are working in health and social work (a relatively high share), it is the sector in which recent migrants are most under-represented compared with their settled peers (Figure 3.6). While it is a growing sector with labour needs, there are strict eligibility requirements – especially for the highly skilled. As a result, it is a sector where professional entry may be delayed due to foreign credential recognition requirements.

While the health/social sector has become a larger entry sector for new comers over the last 13 years, a closer look by occupational skills show different patterns. The sector is actually less an entry sector now than in the past. In 2005, according to the Current Population Survey, 15% of migrant workers in a highly skilled occupation entered the United States labour market through the health sector, while that share was only 9% in 2018. A decline was also observed in European countries. In contrast to the highly skilled, health has become a growing entry sector for migrants in low- and medium-skilled occupations in both the United States and Europe. The share of that group entered through the health sector has doubled over the last 13 years in the United States. Medium skilled health workers include health care support occupations (health care aides and assistants).

The association between the presence of migrants and sectoral growth

Contribution of foreign-born to sectoral employment growth

Between 2005 and 2018, migrants have made up almost half of the 10% employment growth in the United States and two-thirds of the 9% growth in European countries (Figure 3.7). This contribution was more than half in five sectors where the employment has increased in the United States: finance, transport and storage, agriculture, information and communication, and other services. This is also true in hospitality in Europe. In both areas, the number of migrants in employment has increased in all sectors over the last 13 years, even in sectors where native-born employment has declined, such as in construction, activities of households as employers (domestic work), trade and manufacturing. However, with the exception of domestic work and trade in Europe, the increase in the foreign-born workforce did not compensate for the drop in the labour force in these sectors.

In Israel, which population in employment has increased by two thirds between 2005 and 2019, the contribution of foreign-born was very small, in contrast. Migrants contribute to less than 10% of this increase in total employment and it was even lower in many sectors. One notable exception is activities of households as employers, where migrants have made up four-fifths of the increase (Figure 3.7).

Figure 3.7. Contribution of the foreign-born to the employment growth in OECD countries, by sectors, 2005-18

Figure 3.8. Migrant presence and sectoral growth

Migrants and sectoral economic growth

Given the multiple factors influencing sectoral growth, one would not expect a clear association between the presence of migrants and growth. As expected, a first glance at the association between the presence of migrants and sectoral economic growth (measured by Gross Value Added (GVA), see Annex 3.A) reveals no clear pattern (Figure 3.8). Indeed, migrants tend to be overrepresented in the same sectors across countries, even though the growth pattern differs widely. For example, while migrants are overrepresented in Europe, Canada and the United States in security and cleaning, that sector is growing in Europe, stable in Canada and declining in the United States. However, with the notable exception of construction, some declining sectors in Europe and the United States in terms of the sector’s share in the economy, such as communication, manufacturing and the public administration, tend to have a higher concentration of native-born virtually everywhere.

In contrast, migrants are overrepresented in a number of growing sectors (hospitality, warehousing, administrative support). Yet, while these sectors have been growing, their overall labour productivity is rather low. The pattern of migrants being concentrated in growing sectors with low productivity is particularly pronounced in Austria, Belgium, France and Italy. The same pattern holds in Norway and Sweden as well as in the Netherlands, although to a lesser extent.

In sectors with higher labour productivity such as health, real estate and professional-engineering activities (legal, accounting, management, architecture, engineering activities), which GVA shares have grown in the vast majority of OECD countries, migrants tend to be under-represented in both Europe and the United States, but are over-represented in Canada. The only growing high value-added sector where they are over-represented everywhere is IT, with the highest concentrations in Luxembourg, Poland, Ireland, the United States, Canada and the Czech and Slovak Republics.

The links between migration policy and sectoral composition of the economy

Sectoral elements in migration policy

Just like in the study of the impact of migration, much less attention tends to be paid in migration policy to sectoral considerations than to occupational ones. Most OECD countries favour specific occupations in their migration policy, though the degree to which specific occupations are targeted varies. Virtually all countries favour highly skilled over low-skilled immigration, either by requiring minimum salary levels (e.g. Denmark and the Netherlands) or by favouring migrants with higher qualifications (such as the points systems in Australia, Canada and New Zealand). Many countries have specific lists for admission to occupations (OECD, 2014[9]), which are either considered skilled or considered to be in shortage. Skilled occupation lists exist for example in Australia, Canada, New Zealand, and the United Kingdom. Shortage occupation lists are found in Austria, Australia, Belgium, Denmark, France, Latvia, Lithuania, the Netherlands, New Zealand, Poland, Spain, the United Kingdom and the United States, as well as in Sweden for within-country status changes. The EU Blue Card combines salary, occupation and qualification requirements.

The implicit focus of migration policy on occupations rather than sectors has not always been the case. During the economic boom period after the Second World War, several large-scale temporary migration programmes were in force across the OECD. Many of these had a sectoral focus. These included the “Bracero” programme for seasonal farm workers in the United States and the “guest worker”-type programmes for Europe’s growing labour needs. Most of these programmes were based on bilateral agreements with origin countries, with select businesses being able to recruit directly from abroad. The sectoral focus varied across countries. Virtually all of these agreements were ended prior to or with the first oil shock of 1973. Since then, labour migration has been more restrictive in both the United States and Europe. Since the early 2000s, with a renewed interest in labour migration, there has been a clear preference towards favouring highly skilled occupations, at least in European countries.

There are several reasons behind the preference of occupations rather than sectors in the migration policy in most countries. First, and perhaps most importantly, in many OECD countries, economic policy in general has shifted away from specific industrial policies. Second, and closely related, rather than supporting a specific industry, broader occupational profiles can support different sectors and allow for more adaptation if the economic landscape changes. Third, occupation-based migration regimes also allow better targeting of migrants with higher skills, whose integration prospects and longer-term employability tend to be better than those of low-skilled migrants are. What is more, occupation-based migration regimes are also often seen to be less prone to rent seeking or lobbying by business than industry-specific policies and thus generally less distortive, although this is not necessarily the case. In any case, occupation-based regimes tend to be more acceptable to the public. Finally, in Europe, a migration policy favouring specific sectors in which large companies are operating could also potentially be considered an illicit subsidy, thereby conflicting with EU competition law.

Sector-based approaches to migration

In contrast, apart perhaps from the agricultural sector that will be discussed further below, the list of countries favouring specific sectors is much shorter. Among OECD countries, Japan has a strong and explicit sectoral policy. Its Technical Intern Training Programme, the largest internship programme in the OECD with 144 000 new visas issued in 2017, is only available to certain occupations in specified sectors. Technical Intern Training visas are issued for a maximum of three years to foreigners from developing countries for work in the following sectors: agriculture, fishing, construction, food processing, textiles and metal manufacturing.

Moreover, Japan’s Specified Skilled Worker visa, created in 2019, is available to foreigners who work in jobs that require considerable knowledge of, or experience in, specified industries. The programme covers different occupations at all skill levels. It is restricted to 14 designated approved industries: nursing care, building cleaning services, casting, industrial machine manufacturing, electric and electronic information, construction, shipbuilding and marine equipment, automobile repair, aviation, hotel, agriculture, fishery, food and drink manufacturing, and food service. The prospective number of foreign nationals by industry is capped, with the total cap applying over a period of five years. Applicants for the Specified Skilled Worker scheme must prove their skills on the listed job by passing an industry-specific evaluation test. They also need to pass a Japanese language proficiency test to prove at least conversational level. Foreign workers who have already completed a technical intern training are exempted from both tests.

A largely sector-based migration policy has also been in place in Korea since 2004 with the introduction of the Employment Permit System (EPS). The EPS is a large programme which allows SME employers in manufacturing, agriculture, livestock, fishery and construction sectors to request temporary foreign workers, who would be selected, trained and brought through a government-to-government bilateral programme (OECD, 2019[10]).

The EPS matches employers in Korea with workers in origin countries. The Foreign Workforce Policy Committee, an inter-ministerial working group, determines the annual ceiling for inflows of low-skilled foreign workers, including by industry and country of origin. EPS especially supports the so-called “root industries”, which provide basic manufacturing processes as part of the supply chain to larger firms. In Korea, larger firms rely on such subcontractors for important parts of the production process. Competition among SMEs for contracts is intense and margins are low. At the same time, wages are low and working conditions often unfavourable, making employment unattractive for the domestic workforce. The programme thus provides workers to companies that would otherwise not be competitive.

The admission of low-skilled foreign workers in Israel is also managed through a sector-based migration policy. Selected sectors can recruit migrant workers through the Foreign Workers permit (B-1): agriculture, construction, care, tourism and manufacturing. Employers must pay a levy for each foreign worker recruited, fixed at 15% of the monthly pay since 2019. Employers in the care and agriculture sectors are exempted from the levy. B-1 permits are issued for up to one year, renewable for a maximum duration of 63 months. They are subject to quotas by sector (except care), which are generally fixed in consultation with representatives of the respective sectors.

While New Zealand does not have a specific sectoral policy, from 2021 onwards, sectors that hire large numbers of temporary migrant workers will be able to negotiate an agreement with the government to facilitate recruitment of migrant workers. Industries that have been identified as candidates for these agreements include residential care, meat processing, dairy, forestry, road freight transport, tourism and hospitality, construction, and horticulture and viticulture. Sector agreements will include special rules for the access to foreign temporary workers in the specified occupations over the duration of the agreement (e.g. exemptions of labour market test, more generous visa terms). Agreements will also include sector-specific plans to employ more New Zealanders and to reduce its reliance on foreign temporary workers (Immigration New Zealand, 2019[11]). Sector agreements will have a duration of up to three years and will be compulsory (employers will not be allowed to opt out and use other schemes). The introduction of sector agreements in the New Zealand immigration policy is a further step towards sector-based policies for lower skilled migration in that country. These started with the “construction and infrastructure skill shortage list” introduced after the 2010 Canterbury earthquake.

Australia launched in November 2019 a Global Talent Independent programme (GTI), designed to attract highly skilled migrants for employment in one of the seven top “future-focused” sectors in the Australian’s economy. These include AGTech, Space and Advanced Manufacturing, FinTech, Energy and Mining Technology, MedTech, Cyber Security and Quantum Information, Advanced Digital, Data Science and IT. Applicants must receive to a salary of at least AUD 148 700 (around USD 100 000), either through an actual contract or job offer, or through a degree in the target sectors giving access to such a salary.

Labour migration in Italy is managed through quotas (with some exemptions for certain skilled occupations). If this system currently does not have a sector-based approach, a specific sector quota within the annual quotas was in place in the mid-2000s for domestic work (2005-10) fisheries (2006-07) and construction (2007). Also a specific quota for seasonal agricultural has been in place since 2000. In addition, its large-scale regularisation programmes attend to have a strong sectoral focus. The 2002 regularisation, by which around 650 000 migrants were regularised, was opened to every sectors, but the compensation sum for forgone fiscal and social welfare contributions were lower for domestic workers. In the end, half of regularised workers were domestic workers and caregivers (Baldwin-Edwards and Zampagni, 2014[12]). The 2009 regularisation, by which around 250 000 migrants were regularised, was accessible only to foreign workers working illegally in the personal and homecare services sector. Although the 2012 regularisation programme was not dedicated to specific sectors, it also had a strong sectoral impact, as 86% of all applications were domestic workers. In 2020, in reaction to the COVID-19 crisis, the government introduced from June to August a specific regularisation programme for those working in agriculture, livestock, animal husbandry, fisheries and fish-farming, long-term care and domestic work. As of 15 July, around 125 000 applications have been received, of which 86% were in long-term care or domestic work.

Migration policy instruments with direct links to specific sectors

Even in the absence of specific sectoral policies, many migration policy parameters do favour some specific sectors. Indeed, in many occupations, there is an evident link with sectors (e.g. IT professionals – IT sector; medical doctors – health sector; cooks and chefs – hospitality sector)10, while this is less evident for others (e.g. clerks). For the former, schemes that favour specific occupations also have a strong impact on favouring the sectors in which they are employed.11 What is more, in countries that rely on shortage occupation lists, determining which occupations are considered to be in shortage is often done in consultation with representatives of specific sectors, which – depending on their bargaining power – can have a strong impact on the introduction of a specific occupation in the list. Indeed, lobbying takes place at the sectoral level, and not all sectors are on an equal footing in this respect.

Other migration policy instruments with indirect sectoral impact

It is also common that the design of a migration programme, though not necessarily intended to favour specific sectors, tends to disproportionately benefit some of these. Illustrative examples are travel and work programmes – or Working Holiday Maker Schemes – and work possibilities for international students. The flexible working times, high presence in cities and touristic places, and working possibilities during holidays that are the nature of these programmes favour the hospitality sectors.

At times, the impact is even more indirect. For example, the fact that Australia and New Zealand used to strongly favour domestic qualifications in their skilled migration programme has contributed to fuelling international students into their education sectors. Both countries have among the largest shares of international students in tertiary education in the OECD, accounting for a significant part of the countries’ services exports. Indeed, a growing number of OECD countries target international students to help develop their tertiary education sector. As a result, the impact of migration policy changes on the education sector is increasingly debated, as witnessed for example in the United Kingdom in the discussions around its new migration system.

Where sectoral labour needs are strong but labour migration is highly restrictive, irregular migration may result (OECD, 2018[13]). Some migration policy instruments that focus on sectors with a track record of illegal employment of foreign workers, for example in domestic services and in the construction sector thus do not necessarily intend to “favour” these. Instead, they rather intend to better channel and control employment of migrants. This is evident in cases where employers from certain sectors are barred from using the general schemes, such as under the planned sectoral agreements for temporary workers in New Zealand, where employers will not be allowed to opt out and use other schemes. Trusted employer schemes, which have been implemented in the OECD countries settled by migration and in the United Kingdom, can fulfil the dual role of favouring companies and sectors that follow the rules, thereby providing incentives to do so. Such schemes are particularly pertinent for employers strongly dependent on migrant labour (see below on the importance of such schemes for seasonal agriculture). More generally, as employers are operating in specific sectors, employer-based policy instruments are, by their very nature, also sector-based.

Overall policy restrictiveness and the implications of free mobility

A priori, the issue of sectoral elements in migration policy tends to be more pertinent in restrictive migration systems than in ones that are more liberal. Sweden, for example, has a very liberal labour migration system, with the sole basic requirement that the job must have been nominally advertised beforehand, and that wage and conditions in prevailing collective contracts are respected (OECD, 2011[14]). In such a context, where all sectors have full access to migrant labour, there is no need for special sectoral schemes. Likewise, in Europe, as seen above, the access to foreign labour through the intra-EU free mobility covers a large part of needs, notably in sectors heavily reliant on lesser-skilled labour for which most countries have very limited labour migration pathways. Within the free mobility area in Europe, the posted worker scheme has a special status as the social security provisions are those of the origin country. In the EU/EFTA area, almost 3 million intra-EU/EFTA postings were recorded in 2018, and these have a strong sectoral bias. Among these, one third were posted in the construction sector, one fifth in manufacturing and one fifth in freight transport by road.

It is thus not surprising that the examples of strong sectoral components in migration policy are now all outside of the EU free mobility zone, with the exception of seasonal agriculture. In this context, it is also important to stress that managed labour migration is in many countries only a small part of all migration flows, even leaving free mobility aside. The actual impact of labour migration policy on sectors is also limited by the fact that labour migrants are generally allowed to change employers and sectors, although those on temporary visas often require approval for such changes. However, this is an issue in which there are important differences across countries.

The example of Sweden, which went from a rather restrictive to a very open labour migration framework overnight with little increase in labour migration to most sectors, is also illustrative of a further issue: even sudden access to foreign labour does not automatically mean that there is a large impact on this sector. For example, employers may not be willing or capable to hire more people, especially in a situation where wages are sticky and where there is uncertainty about skills obtained abroad or required training – even when migrants’ reservation wages are lower than those of the native-born. Indeed, in Sweden as in other Nordic countries, wages and employment conditions are generally set by collective agreements, and these are done at the sectoral – not occupational – level.

Even where wages are not the issue, and employers would be happy to hire migrants, other institutional obstacles may prevent labour migration. The health sector is an example. In spite of the fact that medical doctors and other health care professionals are in shortage virtually everywhere and actively sought by many OECD countries, actual migration into the sector is quite limited. Part of the reason is that it is a heavily regulated sector, and the recognition of foreign degrees and work experience for doctors and nurses is a burdensome process. As a result, and as seen above, the health sector is not as much an entry sector for migrants as the labour shortages in that sector seem to indicate.

In summary, sectors are not a focus of migration policies. However, many elements of migration policy directly or indirectly affect employment in sectors. The remainder of this section therefore assesses more in depth the role that labour migration has played in shaping three key sectors with high migrant presence across the OECD: i) seasonal agriculture; ii) hotels, restaurants and catering; and iii) information technology.

Seasonal agriculture

Why is migration particularly important for this sector?

Migration is particularly important in seasonal agriculture because of the nature of the work. Strong labour needs arise only for a few weeks or months, often in rural areas with low population density. In these areas, there are often few job opportunities outside the peak season.

Seasonal agriculture is thus a highly labour-intensive activity with peak season involving large number of workers for a short period. For example, 95% of agriculture labourers in Sweden (mostly recruited for berry picking) have permits for a duration of fewer than three months (OECD, 2011[14]). At the same time, work in seasonal agriculture requires little training and little if any language skills. In European countries, for instance, more than three quarters of migrants working in agriculture in 2014 had only a beginner or intermediate proficiency in the host-country language, the lowest language proficiency of all sectors.

Seasonal work in agriculture is often low-paid and physically demanding. In Spain, for instance, the share of workplace accidents per worker (5%) in agriculture is the highest among all sectors (Bellés-Obrero, Martin Bassols and Castello, 2020[15]). Therefore, seasonal agriculture jobs tend to be disregarded by the resident population – including by the unemployed – and face high turnover. For example, in Australia, more than 80% of employers in the sector stated in a survey that they found it to be ‘very difficult’ (49%) or ‘somewhat difficult’ (32%) to find Australian workers in the local labour market (OECD, 2018[16]). The primary reason was that there are “not enough local workers” (38%), although many also stated that ‘local workers do not have the right skills’ (26%). In New Zealand, according to the Recognised Seasonal Employer (RSE) Survey 2019, 98% rated migrant seasonal workers who arrived through the Pacific RSE scheme positively for their enthusiasm, 96% for their dependability and 94% for their productivity. These compare with respectively 10%, 8% and 9% of new workers who were recruited through the intermediation of Work and Income, the agency in charge of bringing welfare recipients into employment (Research New Zealand, 2019[17]).

The short-term nature of the seasonal agriculture work makes it particularly attractive for people from neighbouring countries, especially where wage disparities are high. Indeed, in Europe, the EU enlargements in 2004 and 2007 offered to employers in seasonal agriculture a quick and simple way to recruit seasonal workers. There are important measurement issues with respect to such short and often multiple stays, especially in the context of free mobility, which render the assessment of the contribution of migrants difficult (see Annex 3.A for a discussion).

How do general and specific migration policy tools (and parameters) influence migration to this sector?

Specific schemes supporting immigration in agriculture

a. Seasonal worker programmes through bilateral agreements

Agriculture is by far the sector that benefits most from specific migration policy instruments. At least 15 OECD countries currently have a seasonal worker programme. This seems to be due to the essential nature of agriculture on the one hand and on the other hand, to the nature of the work, which is often not popular among the native-born (see above). In many countries, supporting agriculture is also an important part of regional and rural policy making.

Seasonal worker programmes are often based on a co-operation agreement between Public Employment Services (PES) in origin and destination countries. Large programmes of this kind are in place for example in Canada (with Mexico and Caribbean countries), Greece (with Albania) and Spain (with Morocco). Germany does not have an intergovernmental bilateral agreement, though its PES maintains a small bilateral programme with PES in some non-EU eastern European countries.

The main advantage of such co-operation is ensure some selection and compliance – otherwise, the programme may be discontinued – with a limited administrative burden for host-country administration and employers. In the Canadian seasonal agricultural workers programme (SAWP), for example, Mexico’s Department of Labour maintains lists of workers who would like to be selected to work in Canada, and employers can select from these lists if they do not have enough named workers. Farmers evaluate each SAWP worker at the end of the season. SAWP workers are required to present their employer’s evaluation to a government agency at home to be selected for the next season (see Martin (2016[18])).

In Spain, the Collective Management of Hiring in Origin (‘GECCO’) involves the posting of thousands of predominantly female workers from Morocco to Spain for the strawberries and red fruit harvest, with the cooperation of the Moroccan employment service (ANAPEC). The latter selects workers directly in Morocco, first from a list of previous participants who express the intention to return in the next harvest season, and then among a list of new applicants. The programme also includes explicit goals to promote development in the origin country upon return. The programme supports the harvest season (February to May) in Southern Spain. In Huelva, one of the provinces at the core of this programme, 45% of the seasonal work is done by nationals: 25% by EU mobile citizens, 10% by third-country nationals already in the country, and 20% by GECCO temporary migrant workers. Seasonal workers who have participated in GECCO more than twice are exempt from the labour market test, which gives them more chances to be employed in Spain under other migration schemes.

Some seasonal worker programmes are also subject to numerical limits, such as in Italy and Austria.

b. Trusted employer schemes

There are also seasonal worker programmes where the responsibility of the recruitment is directly on employers, through trusted employer schemes in agriculture. Under such schemes, countries allow certified employers who fulfil a number of criteria to hire workers under certain conditions. Depending on the scheme, recruitment can be either directly by the employer or with the obligation to use dedicated intermediaries. For instance, two thirds of certified agricultural employers in New Zealand under the Recognised Seasonal Employer (RSE) scheme recruit Pacific Island workers directly (Research New Zealand, 2019[17]). Lack of seasonal labour as well as compliance concerns led to the formation of the Horticulture and Viticulture Seasonal Working Group, in which the key stakeholders from industry, government, and unions are represented. That group developed a specific strategy, from which the RSE Policy eventually emerged. Likewise, under Australia’s Seasonal Worker Programme, selected companies are pre-authorised to hire seasonal workers in agriculture.

In most programmes, seasonal foreign workers are assigned to one employer and one location exclusively and cannot change employers in the host-country. Likewise, while extension of visas are usually possible, they must often take place with the same employer.

c. Facilitations in the labour market test

In countries where labour market tests continue to apply for seasonal labour, there are often facilitations. In Canada, for example, the required advertising period of the job offer is shorter than usual in agriculture. In other cases, labour market tests are waived. For instance, nationalities from non-EU Eastern Europe (Armenia, Belarus, Georgia, Moldova, Russia and Ukraine) have been exempt from the labour market test in Poland since 2018. Likewise, in Italy, there is no labour market test for the sector. In the Nordic countries, there are specific facilitations within seasonal agriculture for berry pickers. In Sweden, foreign seasonal workers employed by foreign businesses hiring to a Swedish berry company are exempt from the advertisement requirement of the job that otherwise applies. In Finland, berry pickers do not need to apply for a residence permit, unlike other seasonal workers.

d. Other seasonal worker programmes

In other OECD countries that have seasonal work programmes, such as in Norway, Austria and France, the procedure is the same as for the general programme. Therefore, seasonal workers are subject to labour market tests. However, there are often other facilitations, such as the issuance of multi-year entry residence permits, to facilitate circular migration. For instance, France has implemented since 2007 a three-years residence permit for seasonal workers, which allows foreign workers to work up to six months per year and then come back every year.

e. Non-seasonal programmes for agriculture

Although most migration policy programmes targeting the agricultural sector are seasonal, some OECD countries have specific policies for non-seasonal agriculture, such as Japan, with its three-years programme and Korea (five to ten years programme). Migrant temporary workers in Israel can also get a residence permit for up to five years. Since 2019, employers in that sector have been exempted from recruitment fees, which are otherwise set at 15% of the salary for temporary migrant workers. All non-seasonal programmes are subject to quotas and caps, which are usually negotiated between responsible ministries and industry representatives and regularly updated (every two years in Japan, five years in Israel).

Canada launched an Agri-Food Pilot on 15 May 2020. The three-year pilot tests an industry-specific approach to help employers in the meat processing, mushroom and greenhouse crop production, and livestock-raising industries to fill ongoing labour needs for full-time, year-round employees. It provides a pathway to permanent residence for experienced temporary foreign workers. A total of 2 750 applications will be accepted annually, with caps allocated per eligible occupation within each sector.

General schemes supporting immigration in agriculture

Temporary migration programmes that are not targeted at the agricultural sector can also help filling seasonal labour shortages. This is the case of Working Holiday Makers (WHM) schemes, which are intended to encourage travel and cultural exchange between citizens of signatory countries. They are thus considered a non-economic temporary migration category with work.

Such schemes are widespread in the settlement countries, with Australia having the largest such programme. Indeed, the most important policy tool supporting work in agriculture is a provision allowing working holiday visa holders who have performed specified work in an eligible regional Australian area for a minimum of 88 days to apply for a second working holiday visa. Specified work includes work in the agriculture, mining, and construction sectors. According to the Australian taxation office, a quarter of WHM on their first visa worked in agriculture between January 2017 and June 2018. In 2019, 82% of second WHM visa holders worked in agriculture. An estimated 81 000 WHM worked in agriculture in 2018, ten times the number of foreign seasonal workers (8 500).

In Europe, free mobility has helped filling labour shortages in seasonal agriculture. The number of migrant seasonal workers is difficult to assess in a free mobility context (see Annex 3.A for a discussion). However, data on seasonal migration flows before and after the introduction of free mobility for the countries that joined the EU in 2004 and 2007, respectively, provides a glance of the importance of EU free mobility for seasonal agriculture (Figure 3.9).

Figure 3.9. Impact of the EU free mobility agreements on registered seasonal worker inflows in Europe

The most important destination of intra-EU movements to support seasonal agriculture is Germany. After the transitional arrangements have been lifted in 2011, the number of seasonal workers with a work authorisation (mainly Polish citizens) felt from almost 300 000 seasonal workers in 2010 to less than 4 000 in 2012. Other EU countries also witnessed a sharp drop in the number of seasonal worker permit issued at the end of the transitional arrangements. The number has been divided in half between 2006 and 2008 in Italy, between 2011 and 2013 in Austria, and again between 2013 and 2015 (when the transitional arrangements were lifted for Bulgaria and Romania), between 2007 and 2009 in France and between 2008 and 2010 in Belgium. In Spain, the number of seasonal worker permits issued (mainly to Romanians) fell by a factor of seven between 2008 and 2009. Following the EU enlargements, Belgium does not have a seasonal worker programme anymore, and Germany’s programme (based on bilateral agreements of the PES with non-EU European origin countries) is marginal.

In most European OECD countries in which there is no seasonal worker programme, seasonal workers in agriculture are mainly EU mobile citizens. In Italy, there were in 2019 almost 370 000 foreign workers in the agriculture sector from 155 countries (see Centro Studi e Ricerche IDOS (2019[19])), accounting for 25% of the workforce. Almost one third of the total came from Romania (108 000), followed by Morocco (35 000), India (34 000), and Albania (32 000).

In Germany, there are around 200 000 seasonal agriculture helpers in Germany at the seasonal peak (May/June), about 60% of whom are foreigners, of which 95% from EU. The vast majority are not subject to the standard social security regime, in contrast to the German helpers. In Austria, where almost half of seasonal agriculture is done by foreigners, about 80% coming from the EU.

However, some EU countries where seasonal agriculture is important still rely mostly on seasonal foreign workers under managed temporary labour migration programmes. This is the case notably in Poland, where inflows of agricultural workers are predominantly third-country nationals. In 2019, 127 000 work permits have been issued for nationals from third countries who applied for work in agriculture.

Hospitality

Why is migration particularly important for this sector?

The hospitality sector – accommodation and food services – is a services sector dominated by low-skilled and low-paid employment with atypical working hours. For example, in OECD European countries, the share of low-skilled employment in the sector is twice as high as in other sectors (18% vs. 9%). What is more, one quarter of hospitality workers was in the lowest income decile and three fifths were below the third decile in 2018. Hospitality workers had the lowest income across all sectors, with the exception of domestic workers. Almost three quarters of workers in the sector were affected by at least one of the following atypical working hours: shift work, evening work, night work, weekend work. These jobs are deemed unattractive for native-born workers and the share of foreign-born in hospitality is the highest among all sectors in Europe.

Indeed, migrants are overrepresented in this sector in all OECD countries where data are available, with the exception of Portugal where the sector is particularly large, and Israel. This is especially the case for recent arrivals, who in many countries are three to five times more likely to enter the sector than its share in overall employment would suggest (Figure 3.10). In more than two thirds of countries, hospitality is the entry sector for more than 10% of new entrants.

Figure 3.10. Share of employed migrants and native-born who work in the hospitality sector

How do general and specific migration policy tools (and parameters) influence migration to this sector?

Specific schemes for the hospitality sector are rare. As a result, the hospitality sector is a particularly important entry sector for non-labour migrants. Data for OECD Europe for 2014 by category of entry show, for example, that 56% of recent migrants working in that industry came for non-employment reasons, one of the highest share across industries. This is especially true for non-EU migrants: 78% of recent non-EU migrants working in hospitality were non-labour migrants (the highest share across industries), against 60% of all recent non-EU migrants in employment.

Although most seasonal worker programmes are predominantly used by agriculture, there are some exceptions. Among these is the H-2B programme in the United States, which is dedicated to fill non-agricultural positions. H-2B workers can stay for up to nine months before returning to their home country. Migrants with H-2B visas are tied to an employer who serves as a sponsor. The programme is capped to 66 000, with occasional decisions to lift the quotas exceptionally.

In May 2003, the United Kingdom Home Office introduced a specific Sector Based Scheme (SBS) for the hospitality and food processing (OECD/Federal Office of Immigration, Integration and Emigration, 2004[20]). The scheme and quota allocations were monitored by employer and industry representatives in conjunction with the Home Office. This new scheme has been withdrawn after the 2004 EU enlargement.

Working Holiday Maker (WHM) schemes can also be a tool to fill labour shortages in tourism activities. This is again true in Australia, although to a lesser extent than agriculture. According to the Australian taxation office, 22% of WHM holders have worked in accommodation and food services between January 2017 and June 2018. Unlike agriculture, the tourism sector in the United States can benefit from three youth mobility programmes within the J-1 visa. Trainees and interns enrolled in hospitality schools have their own visa programme. The most important scheme, however, is the Summer Work Travel (SWT) programme, with 104 500 visas issued in 2018.

Some occupations in the hospitality sector are classified as skilled occupations (and thus found on skilled occupation lists, such as cooks and restaurant managers), although they do not necessarily require high formal qualification. As a result, the hospitality sector is a sector where under-education is an issue. “Skilled” migrants with no academic qualifications but with skills acquired though vocational trainings or past work experience are also of importance. In 2018, for instance, 59% of migrant workers in hospitality in the United Stare were undereducated (with a lower educational attainment than required by their job skills), the highest share among all sectors (with the exception of agriculture).

Another scheme that can indirectly affects the ability of the hospitality sector to fill its labour shortage is giving an access to OECD labour markets to international students who have not yet completed their degree in the host-country. With jobs with hours outside regular universities courses time, food service is one of the most important sector of employment of student in most OECD countries. In Europe, 15% of foreign-born aged 18 to 35 in tertiary education worked in hospitality in 2016-18, the highest share with the wholesale and retail trade sector. Hospitality is a more important sector of work for international students than for their native-born peers, whose share working in that sector is only 10%.

Free mobility is also a tool for many employers in Europe to fill shortages in hospitality, both seasonal and permanent. While the share of EU migrants in the hospitality sector was only 4% in 2005, compared with 14% for non-EU migrants, almost half of the foreign-born employment growth between 2005 and 2018 was the contribution of EU migrants. The contribution of EU migrants in the foreign-born employment growth has been particularly large in the German-speaking countries and the United Kingdom. In Austria, almost all the foreign-born employment growth between 2005 and 2018 was from the EU, compared with 76% across all sectors. EU countries used the 2004 EU enlargement to fill vacancies in some sectors. In half of OECD European countries though, the contribution of EU migrants to migrant employment growth in hospitality was less than a third.

As the general schemes for labour migration tend to have few pathways for low-skilled migrant workers, and given the high informality of employment in the sector (for both migrant and native-born), hospitality tends to be a sector where irregular situations are not uncommon. As a result, regularisation programmes have often involved a high share of workers in this sector (see OECD (2018[13])), including both mass and case-by-case regularisation programmes. According to the French Elipa survey (OECD, 2017[21]), the largest sector from which third-country nationals have been regularised in 2010 was “trade- accommodation-food services” (one third of the total). Restaurant business was also the third sector of regularisation during the 2005 regularisation in Spain, after activities of households as employers and construction (Baldwin-Edwards and Kraler, 2009[22]). According to 2017 estimates, “accommodation and food service activities” was the second working sector of migrants in irregular situation in Spain, with 28% working there, just behind activities of households as employers (Gálvez-Iniesta, 2020[23]).

Box 3.2. Specific migration policy targeting other service workers: the example of live-in care givers in Canada

While migrants are nowhere as over-represented as in domestic work, few countries have specific policy instruments for domestic or care workers. These include Japan (for institutional care under bilateral agreements with certain origin countries) and Israel, which has a large programme for caregivers. Almost 2% of employed persons in Israel are foreign live-in caregivers on temporary work permits.

The most prominent and longstanding example, however, is Canada, which has had specific programmes for caregivers since 1992. The first of these was the Live-in Caregiver Program (LCP), in place until 2014 (see OECD (2019[7]) for a discussion). It allowed temporary labour migrants working in the caregiving sector to apply for permanent residency as economic class migrants within three (later four) years of arrival in Canada, once they had completed at least two years of full-time caregiving work. These employees usually lived and worked in private households caring for children, seniors or people with disabilities. Since its inception, the programme was replaced twice with pilot programmes aimed at testing new approaches to permanent residence for caregivers and which removed the live-in requirement of the LCP. The most recent iteration of the pilots was launched in 2019, with more restrictive criteria for admission and transition to permanent residence than the LCP, including annual application caps. In exchange to tighter access, caregivers now receive an occupation-restricted open work permit, which lets them work as a caregiver for almost any employer throughout Canada (excluding Quebec). It thus provides caregivers with the possibility to change jobs more easily, as they work to acquire the two years of work experience required to apply for permanent residency. Caregivers can also benefit from open work and/or study permits for their accompanying immediate family members. In addition, employers no longer need a labour market impact assessment before hiring a caregiver.

Since 2005, more than 90 000 caregivers and 97 000 spouses and dependants have obtained permanent residence. In 2016, 60% of workers in domestic services, the sector under which caregivers generally fall, were migrants. Compared with the overall share of migrants in employment, Canada has the fifth largest over-representation of migrants in the domestic work sector in the OECD (+35 percentage points compared with the overall share), after Israel (41), Spain (43), Italy (62) and Greece (66).

Among the workers in domestic services, the majority (56%) were living in the country for less than five years, half as non-permanent residents. This is a remarkable contrast to OECD Europe, where longstanding migrants are much more likely to be employed in the sector than recent arrivals. In the Southern European countries, which have particularly high shares of migrants in domestic services, at most 10% of all migrant domestic workers are recent. In Europe, the highest shares of recent migrants among foreign-born in domestic services are observed in Switzerland and the United Kingdom (29%). The corresponding share in the United States is 14%. Indeed, the vast majority of caregivers in Canada who entered through the various programmes have left the sector shortly after obtaining residence.

Information Technology

Why is migration particularly important for this sector?

In terms of GDP, Information Technology (IT) was among the most three fastest- growing sectors between 2005 and 2018 in half of all OECD countries. Over that period, the number of employed in the IT sector has grown by 42% in European countries and by 69% in the United States. In many countries, the number of new jobs created in the sector is growing faster than the number of graduates with specialised IT skills, even if computer occupations do not necessarily require a bachelor’s degree. The U.S. Bureau of Labor Statistics projected that there would be about 116 000 job openings per year in computer-related occupations in the United States during the period 2014-24 (Wolf, 2016[24]). According to the U.S. Census Bureau’s American Community Survey, 48% of the population in employment aged 25-44 residing in the Silicon Valley 12 was foreign-born in 2016, compared with 20% of the total US population in employment in that age group. Among workers in IT in this area, foreign-born even accounted for two in three. The IT sector builds increasingly on highly specialised knowledge, which are not always easy to find in national labour markets, i.e. big data analytics, cyber-security, coding/programming and cloud computing. The European Commission expects the market for big data to grow by 40% each year in the coming years. In the United Kingdom, the number of big data analysts working in larger firms is expected to increase by more than 240% over the next five years (ICF, 2018[25]). In contrast to other highly skilled professions, IT specialists operate in a predominantly English-speaking environment.

Against this backdrop, it is not surprising that recent highly skilled migrants are disproportionately often employed in this sector virtually everywhere in the OECD (Figure 3.11). The shares are particularly large in the Czech Republic, Ireland, Israel and the United States.

Figure 3.11. Share of employed highly skilled migrants and native-born who work in the information and communication sector

One striking observation is that a large share of IT specialists in many countries are Indians. In the United States, almost half of the foreign-born labour force in IT comes from India. Overall, Indian-born workers made up 14% of the population employed in IT in the United States in 2018. This is not the case, however, in Europe, where the origin of foreign- born IT workers is much more diverse, with the bulk coming from other EU countries in most European countries, with the notable exception of the United Kingdom, as well as France and the Netherlands.

The origin of the IT specialisation of India lies in local conglomerates, which began the software export industry by sending programmers to clients’ sites overseas, starting as early as the mid-1970s (Dossani, 2005[26]). As foreign ownership was restricted to 40% of the company’s shares in India at the time, many foreign firms closed their Indian operations. Since software development could not come to India, local firms decided to go to developed countries to supply programmers for installing software, thereby starting an Indian IT diaspora network (Docquier and Rapoport, 2011[27]).

How do general and specific migration policy tools (and parameters) affect this sector?

Specific sector preferences, occupation lists and other preferential schemes

The fast-growing and high value-added IT sector is one of the sectors that is most often favoured by migration policy. In Japan, the electric and electronic information sector is included among industries in shortages, for which additional labour recruitment channels are available. The IT sector is also included in the Global Talent Independent programme in Australia. Elsewhere, where schemes giving residence to foreign workers are occupation-based, most include IT workers.

These include shortage occupations lists which waive the labour market test or provide additional points in the Point Based system (OECD, 2014[9]). In the United Kingdom, for example, the skill shortage occupation list for Tier-2 visa includes many IT professions. This is also the case in the Danish positive list.

Germany has longstanding specific policy provisions for IT professionals. From 2000 to 2004, it had a special visa for IT specialists, the German Green Card. This was replaced by its 2004 immigration law, which facilitated high-skilled migration more generally, though admission of IT professionals remains to be facilitated through specific provisions such as a lower salary threshold and a qualification recognition waiver (see below).

In most OECD countries, IT workers are admitted through the general labour admission schemes, especially those focusing on highly skilled employment. However, several criteria in highly skilled worker schemes tend to disfavour IT foreign workers. As IT workers are often young, the wage threshold is more difficult for them to reach. For instance, as shown in OECD/European Union (2016[28]), the EU Blue Card salary threshold is above the actual wage of half of the highly educated population in seven EU countries out of ten, and above the wage of three quarters of that population in half of the countries. As a response, eight EU countries apply a lower salary threshold for the admission of certain workers, including those in the IT sector. Qualification requirements may also hamper access to the sector, which tends to require highly specialised skills but not necessarily formal qualifications. In Germany, for example, there is a waiver of foreign qualification for IT workers if they can demonstrate three years work experience in the sector. In spite of such facilitations (lower salary threshold, qualification waiver), the majority of foreign-born workers in the IT sector in Germany has still been coming from other EU countries.

Start-up visas

In recent years, a growing number of OECD countries have implemented start-up visas. Currently, 13 OECD countries have such schemes. These visas, generally issued to a rather small number of migrants, come with many advantages, including financial support, accommodation and rights to come with dependents. While these schemes are not necessarily explicitly focused on the IT sector, by the nature of the scheme, they tend to be disproportionately used by IT start-ups.

In France, for example, start-up creators recognised as innovative business by public authorities through the “French Tech” label can obtain the “Passeport Talent”, the highly skilled French immigration programme. Selected start-ups (around 100 per year) – virtually all of which are either developing IT tools or IT applications – obtain in addition financial support of EUR 45K and benefit from a place of work within an incubator. In addition, their employees themselves can get a “Passeport Talent” if they earn at least twice the minimum wage.

Chile was the first OECD country that developed a start-up programme for migrant entrepreneurs. The scheme includes financial support ranging from USD 15K to 60K and a working space in an incubator. Other countries with start-up visas include Canada (with a permanent status), Estonia, Italy and Korea (with a financial support up to EUR 92K).

Other migration policy instruments

Intra-Company Transfer schemes, which have been implemented in virtually all OECD countries, allow multinational companies to reassign an employee with specialised expertise to work for a temporary period in another branch abroad. In some countries, there is a high proportion of IT workers among the beneficiaries. This is notably the case in the United States, where seven out of the ten top employers using intercorporate transfers are IT companies. In other countries, the use of intercorporate transfers for the IT sector is much more limited, however. For example, in Germany, only 3% of intercorporate transfers went to the IT sector. However, as seen above, Germany has particularly advantageous provisions for the admission of IT workers.

In the United States, the H-1B visa, which allows employers to recruit specialised workers for a temporary duration (up to three years, renewable once) in “specialty” occupations and sectors, is predominantly used by IT companies (USCIS, 2019[29]). To recruit an H-1B temporary foreign worker, employers must provide proof that the job complies with the prevailing working conditions and wages of other domestic workers with the same job. New H-1B visas are currently capped at 85 000, though there is no cap for H-1B holders working at universities, non-profit and governmental research facilities. As the place of work is taken into account, and not the employer, start-ups located in such facilities are outside the cap. Many of these are in the IT sector. Employers may sponsor H-1B visa holders for permanent residence immediately and indeed, these visas are the main source for employment-based admission for permanent residency. Indeed, around 32 000 H-1B per year over the 2017-19 period have changed status, of whom at least 80% to employment categories (USCIS, 2020[30]). In 2019, 66% of H-1B visas were issued to computer-related workers.

In most countries without specific parameters for the IT sector, specialised foreign workers in that area can usually get another permit. Indeed, IT workers are often overrepresented among these general schemes in countries where no specific provisions exist. For example, IT workers account for about 20% of issued work permits in Sweden.

Conclusion

The above analysis has revealed that migrants are concentrated in certain sectors, and this concentration has grown over time virtually everywhere. Is this sectoral dimension associated with migration policy?

It is not clear to which degree this is actually the case. Indeed, at first sight, the answer seems to be no. Even intra-European migration, where free mobility prevails, is strongly concentrated in certain sectors. Indeed, these tend to be the same sectors where other migrants are found. In addition, in most OECD countries, there are few explicit sectoral elements in migration policy.

Digging deeper, however, one does find that quite a few parameters in migration policy design have a sectoral dimension. In particular, the focus on specific occupations in many countries also implies a focus on specific sectors, as there are strong links between sectors and occupations. In addition, several general migration programmes – such as work and travel programmes – have a strong sectoral dimension. To be aware of this dimension is a first step not only in harnessing potential benefits, but also in avoiding adverse impact on specific sectors.

Like on many other policy fields, the COVID-19-related health and economic crisis is likely to have a profound impact on migration and migrant integration. The impact across sectors is not uniform, and some sectors with large shares of migrants, such as the hospitality sector, have been disproportionately hard hit thus far. Indeed, first labour market trends suggest a disproportionate impact on migrant unemployment – not unlike the pattern observed in previous economic crises (OECD, 2009[31]). The economic crisis may also put a break on demand-driven labour migration, and the higher unemployment will render the recourse to labour migration in general more difficult for policy makers even in those parts of the labour market where needs persist. However, the impact may be less pronounced in countries like Canada where immigration also responds to demographic objectives. Given the sectoral structure of the economy varies from one region to another; the crisis will also have an asymmetric impact on regions, highlighting the needs to better take into account the regionalisation dimension of migration policy, i.e. how migration can be used as a lever for supporting regional economic development.

At the same time, certain essential sectors which rely heavily on lower-skilled migrant labour such as meat processing – where, for example in Canada, foreign-born consist of 48% of total employment – have exposed specific health vulnerabilities, and lead to calls for tighter regulation of the sector, in particular with respect to employment conditions. Migrants have also been at the frontline in the health sector, though many foreign-trained health workers are not working in their profession due to the strong requirements for the recognition of foreign credentials in many health care professions. However, there will likely be a push for more facilitations, and to ease migrants’ employment in this crucial sector where shortages were already apparent prior to the COVID crisis. The closure of borders, in contrast, has put pressure on the agricultural sector in many countries, highlighting the crucial role of temporary migration for seasonal agriculture. OECD countries responses to these seasonal needs varied and some implemented exemption measures allowing foreign seasonal workers to enter. In Greece, migrants in an irregular situation were authorised to work during the harvest peak season.

COVID-19 has also led to calls for de-globalisation of certain key sectors, such as pharmaceutical production. This may give rise to sector-specific labour needs, not all of which may be able to be filled with the domestic labour force – in spite of the likely much higher unemployment. At the same time, teleworking has expanded massively, including across borders. This may have a lasting impact on labour migration in sectors where teleworking is easily possible. In addition, there are some indications that COVID-19 may lead to an acceleration of automation in some sectors. This will impact on the demand for migrant labour.

In summary, migration has sectoral implications – whether these are intentional or not. To be aware of this dimension is a precondition for reaping intended sectoral benefits as well as for avoiding unintended adverse impact. The question is, however, whether countries will want to proactively use migration policy to favour certain sectors. This question is particularly topical in the context of an economic crisis that has a strong uneven sectoral impact (OECD, 2020[32]). Moreover, as sectoral policies tend to be more important in a context where migration is otherwise limited, sectoral implications may become more pertinent if the overall conditions for labour migration were to become more restrictive.

References

[3] Acemoglu, D. (2002), “Technical Change, Inequality, and the Labor Market”, Journal of Economic Literature, Vol. XL, pp. 7-72.

[35] Acemoglu, D. (2000), Technical Change, Inequality, and the Labor Market, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w7800.

[22] Baldwin-Edwards, M. and A. Kraler (eds.) (2009), Regine - Regularisations in Europe, Pallas Publication, https://www.abebooks.fr/9789085550082/REGINE-Regularisations-Europe-9085550084/plp (accessed on 7 May 2020).

[12] Baldwin-Edwards, M. and F. Zampagni (2014), Regularisations and employment in Italy, REGANE Assessment Report, MPRA.

[5] Basso, G., G. Peri and A. Rahman (2017), “Computerization and Immigration: Theory and Evidence from the United States”, No. 23935, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w23935.

[15] Bellés-Obrero, C., N. Martin Bassols and J. Castello (2020), “Safety at Work and Immigration”, GLO Discussion Paper 565, http://www.econstor.eu (accessed on 7 July 2020).

[19] Centreo Studi e Ricerche IDOS (2019), Dossier Statistico Immigrazione 2019 | - Dossier Statistico Immigrazione, https://www.dossierimmigrazione.it/prodotto/dossier-statistico-immigrazione-2019/ (accessed on 7 May 2020).

[4] Clemens, M., E. Lewis and H. Postel (2018), “Immigration restrictions as active labor market policy: Evidence from the Mexican Bracero exclusion”, American Economic Review, Vol. 108/6, pp. 1468-1487, https://doi.org/10.1257/aer.20170765.

[27] Docquier, F. and H. Rapoport (2011), “Globalization, Brain Drain and Development”, IZA Discussion Paper No. 5590.

[26] Dossani, R. (2005), “Origins and Growth of the Software Industry in India”, Shorenstein APARC.

[33] Eurostat (2008), “NACE Rev.2 Statistical classification of economic activities in the European Community”, Eurostat working paper.

[2] Foged, M. and G. Peri (2015), “Immigrants’ Effect on Native Workers: New Analysis on Longitudinal Data”, IZA Discussion Paper 8961.

[23] Gálvez-Iniesta, I. (2020), “The size, socio-economic composition, fiscal implications of the irregular immigration in Spain”, Working papers. Economics., http://economia.uc3m.es/ (accessed on 21 July 2020).

[25] ICF (2018), Study on the movement of skilled labour, European Commission.

[11] Immigration New Zealand (2019), Changes to employer-assisted temporary work visas-Sector Agreements.

[18] Martin, P. (2016), Migrant Workers in Commercial Agriculture, ILO, Geneva.

[1] Murphy, E. and D. Oesch (2018), “Is employment polarisation inevitable? Occupational change in Ireland and Switzerland, 1970-2010”, Work, Employment and Society, Vol. 32/6, https://doi.org/10.1177/0950017017738944.

[32] OECD (2020), OECD Employment Outlook 2020: Worker Security and the COVID-19 Crisis, OECD Publishing, Paris, https://dx.doi.org/10.1787/1686c758-en.

[7] OECD (2019), Recruiting Immigrant Workers: Canada 2019, Recruiting Immigrant Workers, OECD Publishing, Paris, https://dx.doi.org/10.1787/4abab00d-en.

[10] OECD (2019), Recruiting Immigrant Workers: Korea 2019, Recruiting Immigrant Workers, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264307872-en.

[13] OECD (2018), International Migration Outlook 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/migr_outlook-2018-en.

[16] OECD (2018), Recruiting Immigrant Workers: Australia 2018, Recruiting Immigrant Workers, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264288287-en.

[21] OECD (2017), Le recrutement des travailleurs immigrés: France 2017, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264276741-fr.

[9] OECD (2014), International Migration Outlook 2014, OECD Publishing, Paris, https://dx.doi.org/10.1787/migr_outlook-2014-en.

[14] OECD (2011), Recruiting Immigrant Workers: Sweden 2011, Recruiting Immigrant Workers, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264167216-en.

[31] OECD (2009), International Migration Outlook 2009, OECD Publishing, Paris, https://dx.doi.org/10.1787/migr_outlook-2009-en.

[28] OECD/European Union (2016), Recruiting Immigrant Workers: Europe 2016, Recruiting Immigrant Workers, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264257290-en.

[20] OECD/Federal Office of Immigration, Integration and Emigration (2004), Migration for Employment: Bilateral Agreements at a Crossroads, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264108684-en.

[8] O’neil, T., A. Fleury and M. Foresti (2016), Women on the move Migration, gender equality and the 2030 Agenda for Sustainable Development, ODI.

[6] Ottaviano, G., G. Peri and G. Wright (2010), “Immigration, Offshoring and American Jobs”, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w16439.

[17] Research New Zealand (2019), 2019 RSE Survey.

[34] Shin, A. et al. (2019), “Primary Resources, Secondary Labor: Natural Resources and Immigration Policy”, International Studies Quarterly, Vol. 63/4, pp. 805-818, https://doi.org/10.1093/isq/sqz033.

[30] USCIS (2020), H-1B Authorized-to-Work Population Estimate, https://www.uscis.gov/tools/reports-studies/understanding-our-data. (accessed on 29 June 2020).

[29] USCIS (2019), Approved L-1 Petitions by Employer-FY2018, USCIS.

[24] Wolf, M. (2016), Computer Science Workforce Trends and Data, BLS.

Annex 3.A. Technical annex

Industry classification

The industry classification is based on the sector for which the worker’s company generates most of their revenue and profits. Two types of structures of the International Standard Industrial Classification of all economic activities (ISIC Rev4) have been used. The ISIC broad structure (classification into 21 sectors) is the basis for the analysis. In some section showing more detailed analysis, the following classification has been used:

A Agriculture, Forestry and Fishing.

B Mining and Quarrying.

CA Manufacture of food products, beverages and tobacco products.

CB Manufacture of textiles, wearing apparel, leather and related products.

CG Manufacture of rubber, plastics and other non-metallic mineral products.

CH Manuf. of basic metals and fabricated metal products, except mach. & equip.

CI Manufacture of computer, electronic and optical products.

CJ Manufacture of electrical equipment.

CK Manufacture of machinery and equipment not else classified.

CL Manufacture of transport equipment.

CX Other manufacturing.

D Electricity, Gas, Steam and Air Conditioning Supply.

E Water Supply; Sewerage, Waste Management and Remediation Activities.

F Construction.

GA Wholesale and retail trade and repair of motor vehicles and motorcycles.

GB Wholesale trade, except of motor vehicles and motorcycles.

GC Retail trade, except of motor vehicles and motorcycles.

HA Land transport and transport via pipelines.

HD Warehousing and support activities for transportation.

HX Other transport and storage.

IA Accommodation.

IB Food and Beverage Service Activities.

JA Publishing, audio-visual, broadcasting activities and telecommunications.

JC IT and other information services.

K Financial and insurance activities.

L Real estate activities.

MA Legal, accounting, management, architecture, engineering activities.

MC Scientific and technical activities.

NA Other administrative and support service activities.

NE Security, services to buildings and other business support activities.

O Public administration and defence, compulsory social security.

P Education.

QA Human health activities.

QB Residential care and social work activities.

RA Arts, cultural activities, gambling and betting activities.

RB Sports activities and amusement and recreation activities.

SA Activities of membership organisations.

SC Other service activities.

T Activities of households as employers, undifferentiated goods and services, producing activities of households for own use.

U Activities of Extraterritorial Organisations and Bodies.

The category RB is not available in Canada, Japan and the United States and category T is not available in Japan. Given the international nature of the category U, the latter has not be taken into account in the analysis.

For Australia and New Zealand, the Australia and New Zealand Standard Industrial Classification (ANZSIC, in 19 divisions) has been used. Only sectors matching with ISIC are shown in the figures for these countries.

ISIC has been matched for both Canada and the United States with the North American Industry Classification System (NAICS) as follows:

A Agriculture, Forestry, Fishing, and Hunting (NAICS 111-115).

B Mining, quarrying, and oil and gas extraction (211-123).

CA Food, beverages and tobacco products manufacturing (311-312).

CB Textile and textile product mills; Clothing, leather and allied product manufacturing (313-316).

CG Plastics, rubber and Non-metallic mineral product manufacturing (326-327).

CH Primary metal and fabricated metal product manufacturing (331-332).

CI Computer and electronic product manufacturing (334).

CJ Electrical equipment, appliance and component manufacturing (335).

CK Machinery manufacturing (333).

CL Transportation equipment manufacturing (336).

CX Other manufacturing (321, 322-325, 337-339, 8 113).

D Electric power generation, transmission and distribution; natural gas distribution (2211-2212).

E Water, sewage and other systems (2 213).

F Construction (236-238).

GA Motor vehicle and motor vehicle parts and accessories merchant wholesalers and dealers, automotive repair and maintenance (4231, 441, 8 111).

GB Durable and non-durable goods wholesale, except motor vehicle and motor vehicle parts (423-425, except 4 231).

GC Retail trade, except of motor vehicles and parts (442-454).

HA Land transport and transport via pipelines.

HD Warehousing and support activities for transportation.

HX Other transport and storage.

IA Accommodation.

IB Food and Beverage Service Activities.

JA Publishing, audio-visual, broadcasting activities and telecommunications.

JC IT and other information services.

K Financial and insurance activities.

L Real estate activities.

MA Legal, accounting, management, architecture, engineering activities.

MC Scientific and technical activities.

NA Other administrative and support service activities.

NE Security, services to buildings and other business support activities.

O Public administration and defence, compulsory social security.

P Education.

QA Human health activities.

QB Residential care and social work activities.

RA Arts, entertainment, recreation, libraries and archives.

SA Activities of membership organisations.

SC Other service activities.

T Activities of households as employers, undifferentiated goods and services, producing activities of households for own use.

U Activities of Extraterritorial Organisations and Bodies.

Based on ISIC Rev.4, the statistical classification of economic activities in the European Community (NACE), which has been used to establish the classification explained above for European countries and Israel, has been revised in 2008, to better reflect the technological developments and the structural changes of the economy (Eurostat, 2008[33]). Breaks in series occurred in some industries and data are not available for these industries before 2008. Industries with a significant break were the following: Electricity, Gas, Steam and Air Conditioning Supply; Water Supply; Sewerage, Waste Management and Remediation Activities; Transportation and Storage; Information and Communication; Financial and Insurance Activities; Real Estate Activities; Professional, Scientific and Technical Activities; Administrative and Support Service Activities; Arts, Entertainment and Recreation; Other Service Activities.

It is worth noting that the traditional classification of companies into sectors or industries is increasingly challenged by new business models, which combine IT platforms with other business activities. In general, companies whose activities extend across several sectors are assigned to the sector that best reflects their principal business. Platform-based companies do not lend themselves easily into that scheme, as it is precisely the combination between different activities that is their business model (IT, trade and logistics in the case of Amazon, for example).

The Gross Value Added (GVA)

The gross value added is the value of output less the value of intermediate consumption. It is a measure of the contribution to the GDP (Gross Domestic Product) made by an individual sector. Value added is reported at basic prices, i.e. excluding taxes less subsidies on products, so it can be split among sectors, whereas the GDP cannot. Indeed, GDP is valued at market prices (as paid on the market by the purchaser) on the whole economy, therefore it includes taxes less subsidies on products (mainly VAT paid by the purchaser). Data quality on taxes are good with respect to the total economy but their detail across industries and institutional sectors is not.

GVA is therefore the best indicator to identify economically declining and growing industries. However, analysing the evolution of the GVA share by sector is limited by the fact that there is no internationally comparable data distinguishing the different components of productivity growth by industry (labour force composition, capital intensity and multifactorial productivity). Besides, labour force productivity cannot be broken down by country of birth.

The industry classification of GVA statistics is the same as the one used elsewhere in the chapter (see above), with the exception of accommodation and food service activities, which cannot be distinguished.

Measurement issues in the analysis of the industry structure of the migrant labour force

The standard source for comparable data on labour force are Labour Force Surveys, which include for all countries the same definition of the “population in employment” and provide details on job characteristics, including the occupation and the industry/sector.

However, a key problem when using household surveys to describe the foreign-born population is the coverage of migrants. In addition to the basic issues affecting the coverage of migrants (e.g. lower response rate because of a lack of host-country language, under-declaration of the number of residents), further issues relate to the sample size of interviewed migrants and the actual representativeness of migrant workers in surveys.

Sample size issues grow with the detail of the analysis. Therefore, looking at foreign-born working in a specific small sector or subsector is often not possible in countries with a small migrant population. In order to remain over the reliability threshold in as many sectors as possible, two types of structures of the International Standard Industrial Classification of all economic activities (ISIC Rev4) have been used in the chapter (see above). To obtain sufficient sample size for foreign-born, three years of survey were merged in the detailed structural breakdown.

Some categories of migrant workers are not well covered by standard population surveys. By definition, temporary migrant workers with a very short work contract may not be around at the time of the survey data collection. Even those available might not be included if they are not considered to be usual residents. These issues affect in particular temporary migration categories, such as seasonal foreign workers, intra-company transferees, posted workers and different categories of short-term contract foreign workers. Some of those categories contribute more or less significantly to certain sectors (e.g. seasonal workers in agriculture and tourism, posted workers in construction).

In addition, labour force surveys only cover people living in ordinary households. Therefore, seasonal workers living e.g. in housing structures directly within farms or in tourism resorts are not covered at all. This leads to a large underestimation of migrant workers in seasonal agriculture, where most seasonal workers live in non-ordinary households.

Data on seasonal workers is also difficult to obtain from administrative sources. While most countries with a seasonal worker programmes collect data on foreign seasonal workers, these data are hardly broken down by sector.

However, the biggest issue, as far as the estimate of seasonal workers is concerned, is the counting of migrants who benefit from free mobility, who can take any job available without the requirement of a residence permit or a work authorisation. This is a particular issue in Europe, where seasonal work is often done by migrants from other EU countries, who do not necessarily register in the host country when the stay is short.

What is more, individual seasonal workers may be counted multiple times over the same year, since most data collection counts contracts, rather than the number of persons. Yet, the same individual may have several contracts if he/she works for several employers. This may even be the case if he/she works only for one employer but has several separate working periods throughout the year.

In some sectors, especially agriculture, construction, domestic services, and hospitality, illegal employment of foreign workers is also an issue, which raises a separate set of coverage and measurement issues (see OECD (2018[13])for an in-depth discussion of this issue).

Annex 3.B. Statistical annex

Annex Table 3.B.1. Ranking of sectors by share of migrants, OECD countries, 2016-18
Share among the total population in employment

Note: 1: highest share among all sectors; 39: lowest share among all sectors. The table is sorted by the ranking for the EU-28. Data for Japan refer to foreign nationals, not foreign-born. Survey results based on a non-significant sample are marked “ns”. Data not available are marked “na”. Data for Japan have been calculated from micro-data of the Japanese population census 2015 in line with the statistics act, Article 33 (1), funded by JSPS Grant-in-aid for scientific research, 17H04785.
Source: OECD Secretariat calculations (see Annex 3.A) on the basis of European Union Labour Force Survey (EU-LFS 2016-18), Current Population Survey (CPS 2016-18), Canadian Census 2016, Israel LFS 2019, Japanese Census 2015.
StatLink https://stat.link/c5eio6

Annex Table 3.B.2. Ranking of sectors by share of migrants born in EU countries, OECD countries, 2016-18
Share among the total population in employment

Note: 1: highest share among all sectors; 39: lowest share among all sectors. The table is sorted by the ranking for the EU-28. Survey results based on a non-significant sample are marked “ns”. Data not available are marked “na”.Source: EU-LFS 2016-18. See Annex 3.A for a detailed explanation of the classification used.
StatLink https://stat.link/wt9o71

Annex Table 3.B.3. Ranking of sectors by share of migrants born in non-EU countries, OECD countries, 2016-18
Share among the total population in employment

Note: 1: highest share among all sectors; 39: lowest share among all sectors. The table is sorted by the ranking for the EU-28. Survey results based on a non-significant sample are marked “ns”. Data not available are marked “na”.
Source: EU-LFS 2016-18. See Annex 3.A for a detailed explanation of the classification used.
StatLink https://stat.link/pal1si

Annex Table 3.B.4. Ranking of sectors by share of recent migrants, OECD countries, 2016-18
Share among the total population in employment

Note: 1: highest share among all sectors; 39: lowest share among all sectors. The table is sorted by the ranking for the EU-28. Recent migrants are those with at most five years of residence. Survey results based on a non-significant sample are marked “ns”. Data not available are marked “na”.
Source: OECD Secretariat calculations (see Annex 3.A) on the basis of European Union Labour Force Survey (EU-LFS 2016-18), Current Population Survey (CPS 2016-18) 2018, Canadian Census 2016, Israel LFS 2019.
StatLink https://stat.link/cdq80y

Annex Table 3.B.5. Ranking of sectors by share of settled migrants, OECD countries, 2016-18
Share among the total population in employment

Note: 1: highest share among all sectors; 39: lowest share among all sectors. The table is sorted by the ranking for the EU-28. A settled migrant is a foreign-born living in the host country for ten years or more. Survey results based on a non-significant sample are marked “ns”. Data not available are marked “na”.
Source: OECD Secretariat calculations (see Annex 3.A) on the basis of European Union Labour Force Survey (EU-LFS 2016-18), Current Population Survey (CPS 2016-18) 2018, Canadian Census 2016, Israel LFS 2019.
StatLink https://stat.link/3w70xm

Notes

← 1. In this chapter, “sector” generally refers to a larger, general part of the economy, while the term “industry” is more specific and refers to smaller sub-sectors. That notwithstanding, the terms “industry structure” and “sectoral structure” are used synonymously in this chapter.

← 2. The terms “migrant” and “foreign-born” are used synonymously in this chapter.

← 3. See, for instance, the Norwegian example (Shin et al., 2019[34]).

← 4. The terms “under-represented” and “overrepresented” refer to the share of migrants in the respective sector compared with their share in overall employment.

← 5. Data for Japan refer to foreign nationals and are tabulated from micro-data of the Japanese population census 2015 in line with the statistics act, Article 33 (1), funded by JSPS Grant-in-aid for scientific research, 17H04785.

← 6. Public service sectors include the sectors O, P and Q, according to the ISIC, i.e. Public administration and defence, compulsory social security; Education; Human health, residential care and social work activities.

← 7. The share of highly skilled workers exceeds 50% in most OECD countries in the following sectors: information and communication technology; finance; professional, scientific and technical activities; education; human health and social work.

← 8. There is a break in series in 2008 for some industries. See Annex 3.A for more details.

← 9. Sectors where the share of low- and medium skilled workers exceed two thirds in most OECD countries include water supply; sewerage, waste management and remediation activities; construction; wholesale and retail trade; accommodation and food services; transport and storage; administrative and support service; activities of households as employers of domestic personnel.

← 10. In Ireland, there is a special programme for meat-deboners – an occupation exclusive to the meatpacking industry.

← 11. The primary units to be classified in the International Standard Classification of Occupations (ISCO) are jobs, and industries in the International Standard Industry Classification (ISIC). Jobs are classified in ISCO on the basis of the type of work performed, that is, the task and duties to be carried out. ISIC provides categories for the classification of units based on the activities carried out by these units. ISIC and ISCO have different functions and conceptions. However, when similarities and differences between certain groups in ISCO are based on the type of distinctions that are reflected in ISIC (i.e. between the type of products, namely, goods and services, that are being produced or sold), the ISCO groups are defined in a manner that is generally consistent with the definition of these goods and services in ISIC.

← 12. According to the Silicon Valley Institute for Regional Studies, the Silicon Valley is made up by the Santa Clara and San Mateo Counties.

╳

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/ec98f531-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.