In 2019, the Nordic Council of Ministers published the report “Integrating immigrants into the Nordic labour markets” (Calmfors and Sánchez Gassen 2019). A main conclusion was that Denmark, Finland, Norway, and Sweden all face serious challenges in this regard, especially with low-educated men and women from outside the EU. At the same time, a high employment rate is crucial to sustain the Nordic welfare system and to boost income and personal well-being among immigrants. The report therefore investigated which policies and measures are most successful in facilitating labour market integration, and what the Nordic countries can learn from each other.
Since the previous report was published, all Nordic countries have been affected by the Covid-19 pandemic, which initially took a substantial toll on economic growth and employment levels. First evidence shows that migrants, and especially recently arrived immigrants, may have been disproportionately affected (OECD 2020). Still, the overall picture remains fragmented.
To get a more comprehensive view, the Nordic Council of Minsters through the project Nordic co-operation on integration, commissioned Nordregio to provide an up-to-date overview of the current labour market situation of immigrant men and women in the Nordic countries. Nordregio also re-connected with some of the writers of the 2019 report, including Per Skedinger, Bert Bratsberg, Oddbjørn Raaum, and Knut Røed. The researchers reviewed their respective chapters and adjusted policy recommendations considering the impact of Covid-19. A new, final section by Professor Olof Åslund, brings the chapters together and offers cross-cutting conclusions and recommendations. The project was impeccably managed by Nora Sánchez Gassen and Oskar Penje from Nordregio.
We hope this report will contribute to Nordic learning for a more effective labour market integration of non-EU immigrants, post Covid. For more information about the Nordic co-operation on integration, visit: www.integrationnorden.org
Eva Franzén Rolf Elmér
Nordic Welfare Center Nordregio
Nora Sánchez Gassen, Nordregio, firstname.lastname@example.org
Oskar Penje, Nordregio, Nordregio, email@example.com
Bernt Bratsberg, Ragnar Frisch Centre for Economic Research, firstname.lastname@example.org
Oddbjorn Raaum, Ragnar Frisch Centre for Economic Research, email@example.com
Knut Roed, Ragnar Frisch Centre for Economic Research, firstname.lastname@example.org
Per Skedinger, Research Institute of Industrial Economics (IFN) and Centre for Discrimination and Integration Studies, Linnaeus University Växjö, email@example.com
Olof Åslund, Uppsala University, Department of Economics, firstname.lastname@example.org
In late 2019, the first cases of the infectious COVID-19 disease, caused by a novel coronavirus, were reported in the Chinese city of Wuhan. The disease has rapidly spread across the world since then. Its human cost has been dramatic, with more than 167 million registered cases of infection and more than 3.5 million registered deaths as of May 2021 (WHO 2021). To limit the spread of the disease and prevent a collapse of public health care systems, countries around the world introduced different types of containment measures, including lock-downs, travel restrictions and quarantine rules (OECD 2020a).
In the Nordic Region, Norway, Denmark, Finland and Iceland implemented more restrictive containment strategies and temporarily closed down parts of society in the spring of 2020. This included the temporary closing of schools and kindergartens, shops, bars and restaurants, sports and leisure facilities, and other public places (Hansen, Sørensen and Andersen 2021). Restrictions were gradually loosened over the summer, when infection rates were low, but reintroduced during the second and third waves of the pandemic. Restrictions in Sweden were milder than in other countries and relied less on mandatory restrictions and more on voluntary personal compliance with hygiene and social distancing recommendations. As in other countries, companies were encouraged to allow their staff to work from home. Where restrictions were imposed, these took account of local infection rates. Nonetheless, when infection rates rapidly increased during the autumn of 2020, the Swedish government introduced more restrictive measures, for instance, by limiting opening hours of restaurants and bars (Hansen, Sørensen and Andersen 2021).
The measures adopted by the Nordic countries contributed to limiting the spread of the disease, but they also had strong effects on economic activity (OECD 2020a). Lockdowns and mobility restrictions interrupted supply chains around the world. As people started to avoid public spaces and reduced their mobility, the demand for many types of products and services dropped. Hotels and restaurants, retail and the travel and culture sectors were particularly affected by declines in customers and revenue (Nordic Council of Ministers 2020). Governments in the Nordic countries intervened by adopting support measures for the hardest hit sectors and by introducing subsidised job retention schemes. Nonetheless, unemployment rates increased during the spring and summer of 2020 (OECD 2020a).
Not all population groups have been equally affected by rising unemployment. First research suggests that migrants, especially recently arrived immigrants, have been more likely to lose their jobs (OECD 2020a, Adams-Prassl et al. 2020). Several reasons have been discussed for migrants’ vulnerability: First, migrants disproportionally work in sectors such as the hospitality industry, tourism or retail, which have been particularly affected by lockdowns, social distancing recommendations and travel restrictions. Migrants are also more likely to work on temporary contracts and to have fewer years of employment than their native-born peers. Previous economic crises have shown that employees in precarious jobs and with lower seniority are often the first ones to experience redundancy (Andersson-Joona 2020a, OECD 2020b). Immigrants also tend to be overrepresented among the self-employed, often owning small businesses with few employees and low capital stock. These businesses have been at a particular risk of experiencing bankruptcy (OECD 2020a). Nonetheless, the labour demand in sectors such as health care and social care, which also employ many immigrants, has increased during the pandemic. This may have buffered rising unemployment to some extent.
Second, immigrants who have become unemployed may find it challenging to obtain new employment as the number of jobseekers is growing and competition for open position increases (OECD 2020a). Recently arrived immigrants often have less extensive professional networks and contacts than natives, which can be an important factor in finding a new employment (OECD 2020b).
Third, recently arrived immigrants who were participating in language courses, work training, or other types of education or labour market training before the pandemic may have found it difficult to continue their integration efforts when programs were put on halt or moved online. This likely resulted in delays with language learning, social integration and the transition to a first job (OECD 2020b).
Fourth, research shows that migrants have been at higher risk of becoming infected with the virus and experiencing long-term health implications and higher mortality (Lerpold and Sjöberg 2020, OECD 2020b). Reasons for these unequal health outcomes are that many immigrants work in ‘frontline’ sectors with high human contact and risk of infection, such as health care, elderly care or public transportation services (Kleine-Rueschkamp and Özgüzel 2020, Anderson Joona 2020a). Migrants are also more likely to live in crowded, multi-generational households, where it is difficult to isolate in case of infection (Sigurjónsdóttir, Sigvardsson and Oliveira e Costa 2021). In addition, language barriers have made it more difficult for public authorities to share information about the virus and personal protection strategies with immigrant groups, at least initially during the pandemic (OECD 2020b). Longer periods of sickness may have reduced migrants’ employment prospects and outcomes.
While first evidence on the unequal labour market effects of the COVID-19 pandemic is emerging, the overall picture for the Nordic countries remains somewhat fragmented. A comprehensive overview of how the COVID-19 pandemic has affected the labour market integration of immigrants, and which differences exist between the Nordic countries, is still lacking. This report aims to make a contribution and answer two research questions:
This report builds on a comprehensive study about immigrant integration into the Nordic labour markets that was published by the Nordic Council of Ministers in 2019 (Calmfors and Sánchez Gassen 2019). This study concluded that the Nordic countries face serious challenges in integrating immigrants into their labour markets, especially low-educated ones from outside the European Union (EU). The study from 2019 also analysed how immigrants' labour market integration could be improved through education, active labour market, social benefit and wage policies. A key conclusion was that no single policy will suffice. Instead, a combination of policies is needed to promote labour market integration. The study of 2019 was published around nine months before the start of the pandemic, and the context and conditions of immigrant integration into the Nordic labour markets have substantially changed since then. Therefore, the present report “Integrating immigrants into the Nordic labour markets. The impact of the COVID-19 pandemic” shall revisit some of the conclusions and policy recommendations outlined in the 2019 study. Some of the policy recommendations that were proposed then may not be feasible to implement during the conditions of the pandemic; whereas other recommendations may have become even more important.
The report is structured as follows: In the first chapter, Nora Sánchez Gassen and Oskar Penje visualize and analyse how employment and unemployment rates of foreign-born immigrants in the Nordic countries have developed during the pandemic, and how they compare to those of their native-born peers. Differences in labour market outcomes by age, sex, region of origin and sector are highlighted. The chapter also identifies similarities and differences across the Nordic countries.
In a second chapter, Bernt Bratsberg, Oddbjørn Raaum and Knut Røed discuss the implications of the COVID-19 pandemic for social insurance policies. With a focus on Norway, they argue that the inclusion of immigrant workers in the relatively generous social insurance system has been successful in preventing poverty, which attests to the importance of unified social insurance systems in the Nordic countries. Nonetheless, they also find evidence that welfare state generosity sets unintended incentives for excess churning (the simultaneous layoff and recruitment of immigrant workers) and social dumping, a pattern that was already observed prior to the COVID-19 crisis.
A third chapter by Per Skedinger reflects on the role of wage policies in facilitating the labour market integration of immigrants. The overall analysis in this chapter suggests that minimum wage cuts, combined with other active labour market policies, should be considered to reduce high unemployment rates among immigrants especially in the Swedish context.
Olof Åslund draws the discussions and analyses of this report together and offers overarching reflections in the last chapter.
In all Nordic countries, economic activity declined after the start of the pandemic. Unemployment rates increased, both due to an increase in the number of job seekers, caused by layoffs and non-renewal of temporary contracts, and a decrease in the number of new workers hired as companies put recruitments on hold (Hansen, Sørensen and Andersen 2021). In this chapter, we analyse the most recently available data from the Nordic countries to estimate how employment and unemployment rates of the foreign-born population have developed during the pandemic, and how these developments compare to the labour market outcomes of the native-born population. The focus of this chapter is the year 2020, which includes the first and the second wave of the pandemic.
This chapter is mostly based on the European Labour Force Survey (LFS). The LFS is the largest European household sample survey with 35 participating countries. These include the EU member states but also Iceland and Norway. The Åland Islands are included in the LFS as part of the Finnish survey. The other two autonomous regions, Greenland and the Faroe Islands, are not part of the LFS. The LFS is a continuous survey, which measures key indicators in each quarter of the year and spreads interviews with the sample population uniformly across all weeks of each quarter (Eurostat 2016). The participating countries are responsible for collecting the data via face-to-face interviews, telephone or web interviews, or questionnaires. The European statistical office Eurostat centrally processes the survey data. All countries use the same concepts and definitions, follow common guidelines and classifications, and record the same set of indicators and variables during the data collection (Eurostat 2021a). LFS data is therefore comparable across countries. To the largest extent possible, this chapter uses LFS data to compare employment and unemployment trends in the Nordic countries. Where LFS data is not available for indicators of interest, registry data is used, if avialable. To completely grasp and understand national employment and unemployment trends during 2020, registry-based indicators from the national statistical institutes would need to be consulted, but due to publishing delays, this has not been feasible within the time frame of this report.
In a recent report, the OECD argues that current labour market statistics may not reflect the full extent of job loss in 2020 due to difficulties of conducting surveys during the pandemic (OECD 2020d). Nonetheless, this challenge does not appear to be as severe in the Nordic Region as in other European countries. Eurostat (2021b) analysed non-response to the LFS during the pandemic. It concluded that non-response increased at EU level due to lock-down measures, but Denmark, Finland and Sweden are exceptions since no large increases in non-response rates happened there. This is likely because these countries already used remote interviewing techniques for data collection before the pandemic and continued to do so during 2020. Other countries used to collect LFS data via face-to-face interviews, and these were more strongly affected by distancing requirements and lockdowns. The response rate to the LFS in Iceland even increased during the pandemic, likely due to a change in data collection methods (Sigurðardóttir and Blöndal 2020). Data on non-response rates is not available for Norway. Nonetheless, it is likely that the impact of the pandemic has been equally limited as in the other Nordic countries, since Norway also uses remote interviewing techniques (Eurostat 2016). While response rates remained relatively stable during the pandemic in the Nordic countries, the overall achieved sample size was around 15% smaller in Denmark during the first three quarters of 2020, as compared to the same quarters of 2019.According to Eurostat, the achieved sample in 2020 was (average per month): Denmark: 6.000 persons, FI: 7.000, Iceland: 1.000, Norway: 7.000. For Sweden, information on achieved sample size is not available from the Eurostat database. Sweden reported an increase in sampling errors during the first two quarters of 2020, but the situation improved at the end of the year (Eurostat 2021b).
The Nordic labour markets are characterised by high activity rates and relatively low unemployment levels. During the period 2005-2019, unemployment rates in the Nordic countries were consistently lower than in the EU area considered as a whole (Figure 1). The only exception is Finland where higher rates were registered during parts of 2017 and 2018. In 2020, however, unemployment rates in the Nordics increased rapidly and Sweden, Finland and Iceland registered higher unemployment levels than the EU area.Between the third quarter of 2019 and the third quarter of 2020, the increase in unemployment rates, measured in percentage points, was 1.0 (EU-27), 1.2 (NO), 1.3 (DK), 1.7 (IS), 1.8 (FI) and 1.9 (SE). In Iceland, unemployment rose particularly strongly, from 3.2% in the first quarter of 2020 to 7.4% in the fourth quarter of the same year. This drastic increase has been attributed to Iceland’s strong tourism sector which has been heavily affected by international travel restrictions and changed mobility behaviours (Nordic Council of Ministers 2020). In Norway, the unemployment rate reached 4.9% in the third quarter of 2020. While this level is low in comparison to the other Nordic countries, it is the highest level measured in Norway during the 16-year period shown in Figure 1.
In Denmark, Finland, Norway, Sweden and the EU, unemployment rates peaked during the third quarter of 2020 and then declined in the fourth quarter of the same year. Nonetheless, unemployment rates so far remain at a higher level than before the crisis in all Nordic countries. In Iceland, unemployment continued to increase throughout 2020 and was at its highest level in the fourth quarter of 2020 since the financial crisis of 2008.
Figure 1. Seasonally adjusted unemployment rates in the Nordic countries and the EU (15-74 years), by quarter (2005-2020).
Data source: Labour Force Survey (Eurostat).
It is important to note that the unemployment rates presented in Figure 1 and other parts of this chapter may not capture all persons who lost their job during the pandemic (Hansen, Sørensen and Andersen 2021). In the LFS survey, persons above the age of 15 are classified as unemployed if they were not employed during a given reference week but were available for work, and if they had actively sought work during the past four weeks or had already found a job with a start date within the next three months. If interview respondents became unemployed during the pandemic but did not apply for new jobs due to fear of getting infected or the uncertain labour market situation, they were classified as ‘inactive’ instead of ‘unemployed’ (OECD 2020a). This may have inflated reported inactivity rates in the LFS survey in 2020, and depressed unemployment rates. Nonetheless, inactivity rates calculated on the basis of LFS data do not suggest that the data shown in Figure 1 strongly underestimate unemployment during the pandemic. Inactivity rates increased in all Nordic countries between 2019 and 2020, but these increases were minor.DK: 0.0 percentage points, FI: 0.2 percentage points, SE: 0.1 percentage points, NO: 0.2 percentage points. Only in Iceland did the inactivity rate increase more strongly by 2.2 percentage points (from 18.8% in 2019 to 21.0% in 2020).
Figure 1 does not systematically include people who were part of job retention schemes with reduced working hours and salary compensation during the pandemic. This group of workers are generally not classified as ‘unemployed’ but as ‘employed’ in the LFS data if their work contracts remained valid during the period with reduced work time.According to the OECD (2020d), formal job attachment of temporarily laid-off workers is determined according to two criteria: first, an assurance of return to work within a period of three months (or longer than that, if the return to the job is guaranteed) and second, the receipt of half or more of the usual wage or salary from the employer. Temporarily laid-off employees who fulfil one of these criteria are categorized as employed in the LFS. Temporarily laid-off employees who do not meet these criteria are classified as unemployed if they would be available to start a new work during the next two weeks and have actively searched for a job during the last four weeks. All other laid-off persons are classified as inactive. The LFS collects separate information on people who are temporarily laid off, which allows for an estimation of the number of people who were part of job retention schemes. Such data is available for Sweden, Finland and Denmark.Data on temporary absences from work due to layoff in Norway and Iceland have low reliability and are not published on the Eurostat database. They show that the number of temporarily laid-off people increased during the first quarter and peaked during the second quarter of 2020. The numbers then decreased again, even though they remained at a higher level in the last quarter of 2020 than in pre-pandemic times (Figure 2). Data on the number of temporarily laid-off employees by country of birth is not available. Due to these gaps, temporary absences from work are not considered when discussing unemployment trends in the remainder of this chapter.
Figure 2. Absences from work due to layoffs, as a share of active population by quarter, in percent. Seasonally adjusted figures. Population 20-64 years. Iceland and Norway excluded due to low reliability in data.
Data source: Labour Force Survey (Eurostat).
Concerns have been raised that the COVID-19 pandemic will lead to an increase in long-term unemployment (OECD 2020a). Long-term unemployment is here defined as unemployment spells that last for 12 months or more. LFS data on long term unemployment by region of birth is not yet available. Nonetheless, first data from individual Nordic countries indeed suggests an increase in the labour force that is affected by long-term unemployment. For example, recent data from the Swedish Public Employment Service shows that long term unemployment has been on a sharp increase since mid-2019, i.e., before the start of the pandemic (Figure 3). At the end of March 2019, 63% of the long-term unemployed were foreign-born, which is a clear overrepresentation, given that they only account for 30% of the population in the studied age group. In the wake of the pandemic, however, the labour market situation has also become more difficult for native-born Swedes. For this group, the number of long-term unemployed has increased by 15,000 in the past year. Long-term unemployment has risen more strongly among natives than foreign-born workers, which implies that the proportion of native-born among the long-term unemployed has increased from 34% to 37% during the last year.
Figure 3. Long-term unemployment by region of birth (native-born, foreign-born and share of foreign-born), 2010-2021, Sweden.
Source: Arbetsförmedlingen (2021).
The question of how the pandemic has influenced the labour market integration of immigrants is particularly relevant for the Nordic countries, since they have all experienced high immigration numbers during the last years, and their populations have become more diverse (Figure 4). In 2021, almost 8% of the population in Finland was born abroad. In Sweden, it was close to 20%, with the other Nordic countries ranging in the middle. In Iceland, the foreign-born population has increased particularly rapidly during recent years, but data for 2021 is not yet available. In Denmark, Finland, Norway and Sweden, the share of the foreign-born population was higher in 2021 than it had been before the start of the pandemic. Hence, even though the COVID-19 pandemic strongly reduced international migration flows (OECD 2020e), this did not bring the increase in the foreign-born population in these countries to a halt.
Figure 4. Share of foreign-born people in the populations in the Nordic countries, in percent (2015-2021).
Data source: Nordic Statistics and Statistics Norway.
Table 1 shows the largest immigrant groups by country of birth. In all Nordic countries, people born in other Nordic countries or EU member states constitute important groups among the foreign-born population. In addition to this, we can identify important African and Asian countries of origin such as Syria, Iraq, Iran and Somalia. Previous research has shown that immigrant groups born in developing countries often face the strongest obstacles in obtaining employment in the Nordic countries (Calmfors and Sánchez Gassen 2019).
Table 1. Main countries of origin, foreign-born population of working ages (15-74 years) in the Nordic countries (2020).
Data source: Eurostat.
In early 2020, the pandemic was occasionally described as a potential leveller that would affect all members of society alike. This myth was quickly debunked as the differential impact of the pandemic on the health and wellbeing of different population groups became clear (Drefahl et al. 2020, Bambra et al. 2020). The labour market impact of the pandemic has also been unequal across societal groups. In all Nordic countries and the EU, immigrants have been hit harder by unemployment than their native-born peers (Figure 5).
Figure 5. Unemployment rates of native-born and foreign-born by quarter of the year, 15-74 years (2017-2020)
Data source: Labour Force Survey (Eurostat). Unemployment data for the foreign-born population is not available for 2017 in Iceland.
Already before the pandemic, immigrants in the Nordic countries and the EU had higher unemployment rates than natives. Gaps in unemployment rates somewhat decreased towards the end of 2019, for instance in Denmark, but increased during the pandemic in all Nordic countries. Increases in unemployment among the foreign born were particularly strong in Iceland (from 5% in 2020Q1 to 18% in 2020Q4) and Sweden (from 15% in 2019Q4 to 20% in 2020Q3). One reason for the strong increase in Iceland is that many immigrants work in the tourism industry which has been strongly affected by lockdowns and travel restrictions (Hansen, Sørensen and Andersen 2021).
After spikes in the second or third quarter of 2020, unemployment rates of natives started to decline again towards the end of the year in Denmark, Sweden, Finland, Norway and the EU. Unemployment rates of their foreign-born peers, by contrast, continued to increase until the last quarter of 2020 in Finland, Norway and particularly strongly in Iceland. In Sweden, Denmark and the EU, the unemployment rates of foreign-born decreased during the last quarter of 2020, but rates remained at higher levels than for the native population and higher than before the pandemic.
The full labour market effect of the pandemic, and how unemployment rates will develop in 2021, is difficult to estimate, as the third wave of infections is ongoing and the worst labour market effects were temporarily buffered through job retention schemes and other policy responses (OECD 2020a, Nordic Council of Ministers 2020). Annex 1 provides an overview of some of the most central policies and instruments that the Nordic countries implemented to support struggling companies, protect jobs and stimulate economic activity. It is beyond the scope of this study to investigate how successful these different measures were in supporting foreign-born people, their jobs and businesses. Nonetheless, this question deserves attention in future research, when data for such analyses becomes available.
Figure 6. Difference in unemployment rate between 2019Q3 and 2020Q3 in percentage points, foreign born and natives. Countries sorted from left to right by the gap between unemployment increases of natives and foreign born. Countries with low reliability data are excluded.
Data source: Labour Force Survey (Eurostat).
Figure 6 highlights the unequal effect of the pandemic on the labour market attachment of foreign-born and native-born groups. We compare the change in unemployment rates of native-born and foreign-born people between the third quarter of 2019 (2019Q3) and the same quarter in 2020 (2020Q3). Countries are ranked according to the gap between unemployment increases of the two groups.
In almost all countries shown in Figure 6, unemployment rates increased more strongly for foreign-born than for native-born people, and in Sweden and Iceland these differences were particularly pronounced. Spain is the only country where the difference in unemployment increases between these two groups is even larger. Finland and Denmark lie in the middle of the spectrum, and in Norway increases in unemployment were more similar for foreign-born and native-born people. Together with Spain, Greece and Turkey, Sweden was also one of the European countries with the highest level of unemployment among the foreign-born population during the third quarter of 2020 (Figure 7). The unemployment rate of native-born people in Sweden increased less strongly and remained more similar to other Nordic and Western European countries, such as the Netherlands or Switzerland.
Figure 7. Unemployment rate in European countries. Third quarter 2019 and third quarter 2020. Natives and foreign born.
Data source: Labour Force Survey (Eurostat).
Overall, this section suggests that the pandemic has had a disproportionate impact on immigrants’ labour market attachment, and this has especially been the case in Sweden and Iceland. Nonetheless, a certain recovery process with declining unemployment rates seems to have set in towards the end of 2020 in Denmark, Sweden and the EU area. Going forward, a strong focus on labour market integration efforts appears particularly necessary in Sweden, both to help reintegrate foreign-born people into the labour market who became unemployed during 2020, but also to reduce overall high unemployment among the foreign-born population that already existed before the pandemic. In Iceland, there may be hope that the very high unemployment levels among foreign-born workers in late 2020 will decrease during 2021, if the pandemic can be brought under sufficient control to allow for international travel to resume, and if job opportunities open up again in the tourism sector. The situation should be closely monitored, and additional support measures may be required to support the foreign-born people who lost their job during the pandemic.
Figure 8. Employment rate of natives and foreign born, ages 15-74 years. Quarterly employment rate levels indexed to baseline 2019Q1 (=100%).
Data source: Labour Force Survey (Eurostat).
The labour market situation of foreign-born immigrants looks somewhat more promising when comparing employment rates. In Figure 8, employment rates of foreign-born and native-born groups in 2019 and 2020 are indexed to the first quarter of 2019. Rates above 100% indicate increases in employment, rates below 100% show declines. In Denmark, Finland, Sweden, Norway and the EU, employment rates of natives fluctuated in 2019 and 2020 but were around the same level in the last quarter of 2020 as they had been in early 2019. Only in Iceland did employment rates of natives substantially drop in 2020. Among the foreign-born population, employment rates in the last quarter of 2020 remained below those of the native population, and somewhat below the level reached in early 2019 in Finland, Norway, the EU and particularly in Iceland. In Sweden, employment rates of the foreign-born labour force recovered during the third and fourth quarters of 2020 and were around the same level at the end of the year as in early 2019. In Denmark, the employment rate of immigrants increased even stronger than among the native population, and in late 2020 it was above the reference level from early 2019. Overall, Figure 8 also suggests that the pandemic had a stronger effect on the labour market outcomes of the foreign-born population in most Nordic countries. Nonetheless, with the exception of Iceland, declines in employment rates have not been very strong, and first signs exist that a recovery process has begun in Denmark and Sweden which has also benefitted immigrant population groups.
The foreign-born population in the Nordic countries is a diverse group and their risk to experience unemployment varies by age, gender, region of birth, educational attainment, and sector of employment (Andersson Joona 2020b). In the following section, we will analyse how employment and unemployment rates have differed across these socio-demographic factors during the pandemic.
Existing research has shown that labour market outcomes of the foreign-born labour force in the Nordic countries differ by region of birth, with immigrants born in the EU area achieving higher employment levels and lower unemployment rates than their peers born in countries outside the EU. In all Nordic countries, these differences have persisted during the pandemic (Figure 9). We compare the third quarter of 2020, when unemployment peaked in Denmark, Finland, Norway and Sweden, with the same quarter of 2019. In Iceland, separate unemployment rates for immigrants born in and outside of the EU are not available, and only a joint rate for all foreign-born people could be obtained.The category „Foreign born: outside EU27“ includes immigrants from rich, industrialized countries such asJapan, Switzerland or the USA, who usually have high employment rates in the Nordic countries. It alsoincludes immigrants from poorer, less developed countries who face stronger difficulties in integrating into theNordic labour markets. The available LFS data do not allow us to distinguish between these differentcountries of origin. For an overview of important countries of origin, see also Table 1 in this chapter.
In Finland, Norway and Sweden, immigrants born outside the EU face a higher risk of unemployment than their native-born peers and those born in the EU-27 area, both before and during the pandemic. In July-September 2019 (Q3), unemployment rates of those born outside the EU were at 10.0% in Norway and 13.2% in Finland. In Sweden, this group fell even more strongly behind other population groups, with unemployment levels of 17.5%. Immigrants born in the EU area had comparatively lower unemployment rates in 2019, between 5.1% in Norway and 8.7% in Finland. In Denmark, immigrants born in the EU area and outside the EU area had almost identical unemployment rates in the third quarter of 2019. Natives had the lowest unemployment rates among the three groups in all countries, including Iceland.
Figure 9. Unemployment rates by region of origin in 2019Q3 and 2020Q3, 15-74 years (in %).
Data source: Labour Force Survey (Eurostat).
Between the third quarters of 2019 and 2020, unemployment rates increased for all groups in the Nordic countries, but at different levels. In Sweden, unemployment increased particularly strongly among immigrants born outside the EU. Almost 24% of the labour force born in third countries were looking for a job in the third quarter of 2020, while unemployment rates for EU-born and native-born workers remained at or below 7%. In Norway, unemployment increased most strongly among immigrants from the EU area and in Finland and Denmark, both groups of foreign-born experienced similar increases in unemployment. In Iceland, unemployment increased more strongly among the foreign-born than among the native-born population. Around 10% of the foreign-born labour force was without a job during the third quarter of 2020, compared to 4.4% of the Iceland-born population.
Overall, unemployment rates of foreign-born workers from outside the EU have remained higher than those of other population groups during the pandemic. In Sweden, existing inequalities have intensified, with unemployment rates among third country immigrants reaching new highs in 2020. A specific focus on this group is needed to bring them back into employment and avoid that current labour market inequalities lead to long-term divisions in terms of career trajectories, income and quality of employment.
The difference in unemployment rates between immigrants born in the EU and outside of the EU has in part been linked to educational attainment levels (Andersson Joona 2020b, Ek and Skedinger 2019a). The share of jobs for which only elementary education is required is low in the Nordic countries, in particular in Norway, Sweden, Iceland and Finland (Calmfors and Sánchez Gassen 2019). Immigrants born in the EU are, on average, more likely than immigrants from outside the EU to have reached higher educational attainment levels and to meet the skill requirements on the Nordic labour markets. EU immigrants, therefore, often find a first job faster and benefit from more varied job opportunities than their peers from third countries. The link between educational attainment and employment is visualized in Figure 10. It shows the composition of the native and immigrant populations in the Nordic countries by educational attainment, distinguishing between immigrants born inside and outside the EU-27 area. The employment rates of the different groups are stated inside the bars of the diagram.
In all Nordic countries, immigrants born outside the EU area are more likely than natives and EU immigrants to have obtained only low education. In Sweden, the share of low educated third country immigrants is particularly high, with around 40% belonging to this group. Nonetheless, it should be noted that more than 30% of third-country immigrants in all Nordic countries have obtained high educational attainments. Among immigrants born in the EU, as many as 52% in Sweden and 51% in Denmark have obtained high levels of education.
Employment rates are linked to educational attainment (Ek and Skedinger 2019a). In all Nordic countries and across all population groups shown in Figure 10, those with high educational attainment achieved higher employment rates than their peers with low or medium educational attainment in 2020. The only exception is Iceland, where low-educated immigrants born outside the EU have higher employment rates (74%) than those with medium educational attainment (71%). Among highly educated immigrants born outside the EU, employment levels reached between 74% in Finland and 78% in Norway. Among immigrants born in the EU area, higher education also entails high levels of employment, from 82% in Iceland and 88% in Sweden. Among lower educated groups, employment rates of native-born people and foreign-born immigrants from outside the EU were similar in Denmark (52% and 50%), Finland (36% and 38%) and Sweden (42% and 40%). Overall, Figure 10 suggests that higher education facilitates labour market integration of natives and immigrant groups in the Nordic countries. In order to accelerate the labour market re-integration of immigrants who lost their jobs during the pandemic, a focus should, therefore, be placed on increasing educational attainments.
Figure 10. Employment rates by educational attainment levels and region of birth in 2020 (in %). Population aged 15-64 years.
Low, medium and high education refers to lower secondary education or lower, upper secondary and post-secondary non-tertiary education, and tertiary education, respectively. The numbers printed inside the bars show the employment level of the different population groups.
Data source: Labour Force Survey (Eurostat).
The emerging literature has also started to analyse gender differences in labour market outcomes during the pandemic, and different hypotheses have been raised as to whether women have been more likely to become unemployed than men. On the one hand, some female-dominated sectors have seen stable or increased labour demand during the pandemic, such as health care or elderly care. Women in these sectors may have been less at risk of losing their jobs or may even have profited from new employment opportunities (OECD 2020a, Andersson Joona 2020b). Conversely, other sectors that employ many women, such as the hospitality and retail sectors, have been hard hit by the pandemic and many jobs have been lost (OECD 2020a). Women, and foreign-born women in particular, are also more likely than men to work on temporary contracts, and these groups tend to be among the first to lose their job during economic crises (Andersson-Joona 2020b, OECD 2020e).
No clear picture on gender differences emerges based on the LFS data. In the four Nordic countries shown in Figure 11, unemployment rates increased for both male and female native-born and foreign-born groups between 2019 and 2020, with the exception of foreign-born men in Denmark.For Iceland, gaps exist in the data on unemployment rates by gender. Iceland is therefore not included in this section. Among the native born, unemployment increased more strongly for women than for men in Finland, but the reverse is true in Sweden. In Denmark and Norway, unemployment rates increased almost equally strongly for men and women. Among the foreign-born population, women’s unemployment rates increased more strongly than those of their male counterparts in Denmark, Norway and Sweden, but not in Finland. Overall, a clear gendered pattern in unemployment increases does not appear.
Figure 11. Changes in unemployment rates by gender and region of birth between 2019 and 2020 (percentage points).
Data source: Labour Force Survey (Eurostat).
People who lost their jobs during the pandemic may not always have registered as unemployed. Instead, they may have temporarily dropped out of the labour market, for instance out of fear of getting infected at a new workplace, or to wait for better employment prospects after the pandemic. Early evidence from several European countries suggest that women may have been more likely to become economically inactive during the pandemic to care for children, sick family members or dependent elderly people (OECD 2020a). Nonetheless, LFS data does not suggest that inactivity rates among the foreign-born, and among women in particular, increased during the pandemic (Figure 12). On the contrary, inactivity rates of foreign-born women declined between 2019 and 2020 in all countries. Among foreign-born men, inactivity rates increased mildly in Denmark and more strongly in Norway. In Finland and Sweden, inactivity rates of foreign-born men declined. While we see stronger changes in inactivity rates among the foreign-born populations, the pandemic had a rather limited effect on inactivity rates of native-born men. Among native-born women, inactivity increased moderately.
Figure 12. Changes in inactivity rates by gender and region of birth between 2019 and 2020 (percentage points)
Data source: Labour Force Survey (Eurostat).
Inactivity rates are expressed as the number of inactive persons in relation to the total reference population. A decline in inactivity rates can, therefore, be caused by a decrease in the number of inactive persons or an increase in population size. LFS data suggests that both factors play a role in explaining the declines in inactivity rates among foreign-born women in Finland and Norway. In Denmark, the decline in inactivity rates is completely driven by a decline in inactive women, since the number of foreign-born women declined between 2019 and 2020. In Sweden, the number of inactive foreign-born women increased between 2019 and 2020, and the decline in the inactivity rate was driven by population growth alone.
A possible explanation for the decline in the number of inactive foreign-born women in several of the Nordic countries may be that some of them started to look for work as the labour market prospects of their male partners became more insecure during the pandemic. In Denmark and Finland, employment numbers and employment rates of foreign-born women were indeed higher in 2020 than they had been during 2019 (Table 2). Increases in female employment to compensate for loss of family income have been described in the context of previous crises, for instance in Greece (Papapetrou and Bakas 2013), Spain and Italy (Addabbo, Rodríguez-Modroño and Gálvez Muñoz 2015). The literature also suggests that many of these women may leave the labour market when the income of the primary earner in the family becomes more stable again. The current situation could, therefore, be considered a chance for female labour market integration in the Nordic countries. To promote employment of foreign-born women in the longer term, a focus should be placed on encouraging those women who have taken up employment to remain part of the labour force even when the labour market prospects of their partners recover.
|Employment rate 2019||Employment rate 2020||Difference|
Table 2. Employment rates of foreign-born men and women, 2019 and 2020.
Data source: Labour Force Survey (Eurostat).
First research evidence also indicates that the labour market impacts of the COVID-19 pandemic differ across age groups. Various studies have concluded that unemployment rates have increased and employment rates have decreased particularly strongly among young people between the ages of 15 to 24 years (OECD 2020a, Hansen, Sørensen and Andersen 2021). Several reasons explain the strong impact of the crisis on younger people. Firstly, sectors that provide entry jobs for young people have been heavily affected during the COVID-19 pandemic, including the hospitality sector, tourism and retail. These sectors could not offer the usual entry jobs for young people during the crisis (Arbetsförmedlingen 2020). Secondly, young people are overrepresented among employees with temporary contracts (Andersson-Joona 2020). Evidence from previous crises shows that these groups are usually the first to become unemployed as their contracts are not renewed (Arbetsförmedlingen 2020).
The described age-effects are indeed visible among the native-born population in all Nordic countries (Figure 13, Panel a). Employment rates decreased for all age groups between 2019 and 2020, except for those 50 years and older in Sweden. In all Nordic countries, decreases in employment were strongest among young adults aged 15 to 24 years.
Among the foreign born, the picture is not as clear (Panel b). Employment rates of most groups decreased during the pandemic. Nonetheless, in Finland and Norway, the effects were strongest among the oldest age group (50+ years), while in Sweden and Iceland, employment rates decreased most strongly among the middle age group (25-49 years). In Denmark, employment rates of the middle-aged and older groups were higher in 2020 than in 2019, and only young people (15-24 years) were less likely to be employed in 2020 than in the previous year. With the exception of Denmark, employment rates declined less strongly during the pandemic among the young foreign born than among some of their older peers. This pattern has, to our knowledge, not yet been described in the literature.
Figure 13. Difference in employment rates of natives and foreign born between 2019 and 2020 (in percentage points), by age group.
Data source: Labour Force Survey (Eurostat).
The employment rate is calculated as the number of employed people in relation to the total reference population. A decline in employment rates could be caused by a decline in the number of employed persons or an increase in the reference population. LFS data suggests that population growth partly or wholly explains the decline in employment rates among the oldest foreign-born age group in Norway and Finland, as well as among the younger and middle age groups in Iceland.Numbers are not shown here, but are availble from the authors and from the Eurostat database. In order to explain employment trends during 2020, it is, therefore, important to consider demographic developments in addition to labour market trends. Some of the changes in the employment rates shown in Figure 13 are the result of demographic changes, and not only due to changes in employment.
One of the characteristics of the COVID-19 pandemic is that it had a very unequal impact on labour demand in different industrial sectors. Throughout the Nordics, people decreased their mobility and reduced social contacts after the start of the pandemic which heavily affected business activity in sectors that depend on human contact or mobility. Mobility restrictions and social distancing rules also led to a sharp shift towards telework and accelerated the ongoing digitalization process of the Nordic societies. This shift occurred particularly rapidly in sectors that do not require physical proximity, such as research institutions or consultancies. Naturally, these sectors have avoided many of the negative impacts of the crisis.
To assess increases and declines in jobs in different sectors, and the impact these trends had on the native and foreign-born labour force, we have compared employment numbers during the second and third quarter in 2020 with the pre-pandemic situation, which is represented by an average from the same quarters in 2017, 2018 and 2019. This choice of baseline value is intended to smooth the normal annual variation and detect the actual pandemic effect. Registry based employment data for 2020 is available from Denmark, Iceland, and Norway, but not yet from Sweden and Finland.Statistics Sweden (SCB) will release register-based employment data for November 2020 in November 2021. The statistical office of Finland will publish employment data by sector for the year 2020 in November 2022. Since the employment data comes from different national sources, they are not directly comparable. In the following section, we present the data separately for each country. The grouping of employment follows the international standard classification of industries NACE rev 2. Categories like ‘unspecified’, ‘unknown’ or ‘other’ have been removed from the figures.
Figure 14 shows employment by industry and employee background for Denmark. The largest negative proportional change was within ‘Accommodation and food service activities’, which saw a 6% decrease of immigrants’ jobs and a 10% decrease of jobs for natives between the baseline period and 2020. In absolute numbers, the decrease in this sector amounted to 4745 jobs for the native group of employees and 1228 jobs for immigrants, which makes it the industry with the largest drop in job opportunities for immigrants in absolute terms.
Nonetheless, in many sectors job opportunities also increased in 2020 in comparison to the baseline period. Sectors with the largest proportional increases of jobs for foreign-born workers include ‘Electricity, gas, steam and air conditioning supply’, ‘Information and communication’ as well as ‘Construction’. In absolute numbers, the largest increases in jobs for immigrants occurred in ‘Human health and social work’ (3835 additional jobs) and in Construction (2546 additional jobs). When comparing job growth and decline across all industries between the baseline and 2020, the number of jobs for natives declined by 0.5%. The number of immigrants in employment increased by 6% during the same period.
Figure 14. Denmark. Percent change in jobs per industry, 2020 compared with 2017-2019 average (2nd and 3rd quarter average all reference years). Population aged 16-74 years by country background. Industries are sorted from top to bottom by the change in the number of jobs in absolute terms.
Source: Statistics Denmark, table LBESK62.
In contrast to Denmark, the number of jobs in Iceland decreased in most sectors between the baseline period and 2020, for native and for foreign-born workers (Figure 15). The largest negative proportional changes were registered in tourism-related sectors such as ‘Travel agency, tour operator and other reservation service and related activities’, ‘Accommodation’ and ‘Tourism industries’. The number of jobs held by immigrants within the first-mentioned sector decreased by 46%, while the number jobs held by natives decreased by 43%. ‘Tourism industries’ was also the sector with the single largest loss of jobs compared to the baseline average. The number of immigrants working in this sector decreased by 3688 persons. This industry also saw the largest drop for natives, with 5270 fewer jobs in 2020. Job growth occurred only in a few sectors. For the foreign-born labour force, the largest proportional gains were in craft sectors such as ‘Electricity, gas, steam and air conditioning supply’ as well as ‘electricity and water supply’. Across all industries, the number of jobs held by natives declined by 3% between the baseline period and 2020, and by 10% for immigrants.
Figure 15. Iceland. Percent change in jobs per industry, 2020 compared with 2017-2019 average (2nd and 3rd quarter average all reference years). Population aged 16-74 years by country background. Industries are sorted from top to bottom by the change in the number of jobs in absolute terms.
Source: Statistics Iceland, table VIN10021.
Similar to Denmark and Iceland, sectors that rely on tourism and travel have been strongly affected by job decline in Norway (Figure 16). Between the baseline period and 2020, the largest negative proportional changes occurred within ‘Accommodation’ and ‘Food and beverage service activities’. The number of jobs within Accommodation decreased by 26% for immigrants and 23% for natives. In absolute numbers, however, the industry with the largest loss of jobs for immigrants was ‘Domestic trade’ (8498 jobs lost). In ‘Transportation and storage’, 7957 jobs were lost in 2020 compared to the 2017-2019 average. While the number of available jobs declined in around half of the sectors shown in Figure 16, increases in the number of jobs were registered in others. Most notably, the number of immigrants in employment increased by more than 15% in ‘Financial and insurance activities’ and ‘Power and water supply, sewerage/remediation activities’. When aggregating all job categories, the difference between the time periods is a 0.6% decrease for natives, but a 3.3% increase for immigrants.
Figure 16. Norway. Percent change in jobs per industry, 4th quarter 2020 compared with 4th quarter average 2017-2019. Population aged 15-74 years by country background. Industries are sorted from top to bottom by the change in the number of jobs in absolute terms.
Source: SSB table 13215.
Some of the absolute employment numbers behind the graphs shown in this section are very small. This is the case especially for Iceland. For these industries, a change that represents 100 jobs will show as a large relative change and could risk skewing the overall picture for the reader. It should also be noted that changes between the baseline period and 2020 could partly be due to demographic factors, especially in-migration or out-migration, occurring between 2017 and 2020. Another factor that should be considered is that a positive change in the number of jobs could result from active and successful integration activities, both at state level and at industry level. The COVID-19 pandemic may hence explain part of the changes in jobs we are seeing, but other contextual factors will likely also play a role.
The COVID-19 pandemic has had an immediate and strong impact on the labour markets of the Nordic countries but, as this chapter has illustrated, some population groups have been more strongly affected than others. In all Nordic countries, foreign-born people have experienced stronger increases in unemployment than their native-born peers. Differences in the impact of the pandemic have been particularly pronounced in Iceland and Sweden. Immigrants born outside the EU, many of whom have only obtained low levels of educational attainment, have faced the largest challenges in finding and keeping employment in 2020. These findings have largely confirmed expectations and early analyses that have been published since the start of the pandemic.
Nonetheless, this chapter also pointed to some findings that have – to our knowledge – not yet been discussed in the available literature. One such finding relates to age patterns. While young adults have experienced the strongest increases in unemployment among the native-born population, this pattern does not seem to apply among immigrants. With the exception of Denmark, middle (25-49 years) or older (50+ years) age groups have experienced the highest decreases in employment between 2019 and 2020 according to the LFS data. The effect of gender on labour market attachment during the pandemic has also led to somewhat unexpected conclusions. LFS data suggests that inactivity rates among foreign-born women decreased in several Nordic countries, while employment rates increased. We have speculated that this trend may reflect women’s strategies to support family income as the labour market position of the main household earners in the family became more uncertain during the pandemic. If this hypothesis is correct, then the current situation should be considered an opportunity to improve the labour market integration of foreign-born women. Efforts should then be placed on encouraging these women to remain part of the labour force even if the labour market situation of the main family earner(s) becomes more secure again.
Further research is needed to corroborate the different trends and results shown in this chapter with the help of national register data, when these become available for all Nordic countries. This chapter mostly relied on the European Labour Force Survey. The LFS has a rather large sample size and generally provides reliable statistics on labour market indicators. Nonetheless, there is some uncertainty. Just as other surveys, the LFS results rely on the answers of a selected sample of the population. In principle, many different population samples could be interviewed, who might give different answers to the survey questions and thereby influence overall survey results. The results of the LFS therefore have to be viewed as providing estimations, rather than exact results, of the labour market attachment and outcomes of different population groups.
Finally, this chapter highlighted some trends which suggest that the worst labour market effects of the pandemic may have happened and that a recovery period has started. Total unemployment rates reached their peak in the autumn of 2020 and started to decline again in the fourth quarter of the year in all Nordic countries except Iceland. Employment rates of immigrants have started to recover and reached or surpassed pre-pandemic levels in Denmark and Sweden. In Denmark and Norway, a growth in the number of jobs held by immigrants occurred in craft and manufacturing sectors, as well as public service and education sectors. These jobs could to some extent buffer the job loss that occurred in sectors related to tourism and hospitality. In Iceland, however, the situation remains serious with unemployment rates rapidly increasing all through the end of 2020, and employment rates of natives and foreign-born workers dropping far below pre-pandemic levels.
The overall impact of the pandemic on the labour market integration of immigrants is difficult to predict since the pandemic is ongoing and the worst shocks on businesses and employment have been buffered by public policies and support measures (see also Annex 1). Future labour market developments in the Nordic countries will be contingent on how infection rates and new virus variants will develop, whether jobs that were lost during the pandemic will be re-created or disappear, and the effect of different types of economic support measures that are still in place in the Nordic countries (Arbetsförmedlingen 2020). Previous economic crises have shown that early spells of unemployment can have long-lasting negative consequences on labour market trajectories of young people and recently arrived immigrants (OECD 2020b). Public policies and investments are required to ensure that people who lost their jobs or were not able to find a first job during the crisis do not end in situations of long-term unemployment. This would risk setting back the achievements in the integration of immigrants that were gained since the European migrant crisis of the years 2015-16 (OECD 2020b).
As shown in this chapter, people with low educational attainment face larger challenges in obtaining employment in the Nordic countries. Educational policies and opportunities for up-skilling and re-skilling should be prioritised for immigrants who lack upper secondary educational attainments (Arbetsförmedlingen 2020). In addition, integration measures and language courses should be quickly re-instated and accelerated for recently arrived immigrants who may not yet have mastered the language of their host country or may lack educational achievements. If immigrants can obtain skills and competencies that are required on the labour market, their chances to find employment should increase. Due to demographic changes and investments made in sectors linked to the green transition, demand for highly educated labour is expected to increase in the coming years, for instance in the mining and mineral industry, the steel and petroleum industry (Moodie et al. 2021), but also in health care and social care.
All Nordic countries adopted measures and instruments during the pandemic to support the most affected sectors and firms and prevent bankruptcies and job losses (OECD 2020b). Hansen and colleagues (2021) distinguish between three types of measures. Table 1 provides an overview of the number of measures and instruments that were adopted in each Nordic country in 2020 (adapted from Hansen, Sørensen and Andersen 2021).
Type 1: Measures aimed at supporting liquidity and preventing bankruptcies among viable businesses. These include subsidies, access to capital and loans, and deferrals of payments and obligations. Such measures were implemented in all Nordic countries except for the last-mentioned measure, which was not used in Finland. Norway and Iceland implemented many instruments in this area (Hansen, Sørensen and Andersen 2021, OECD 2020a).
Type 2: Measures to maintain employment and provide income security for employees. This includes job retention schemes which allowed companies to temporarily lay off employees or reduce their work hours instead of permanently dismissing them. Part or all of the costs of hours not worked were covered by the state to relieve financial pressure on companies and protect workers’ income. Iceland introduced a new job retention scheme, while the other Nordic countries expanded and simplified access to existing instruments. Finland, Norway, and Sweden also increased benefits and access to schemes for workers in non-standard jobs (OECD 2020c). Other instruments in this category aimed at securing income for workers, including entitlements for sick leave which were extended in Finland and Sweden (OECD 2020a). Support measures were also introduced for parents, care givers, unemployed workers and people not covered by existing schemes. Overall, Sweden and Denmark implemented most measures in this category (Hansen, Sørensen and Andersen 2021).
Type 3: Measures to stimulate economic activity and job creation for instance via active labour market policies and financial aid packages to stimulate the economy. The Finnish government was particularly active in this area, implementing several measures which aimed at making the labour market more flexible; for example, temporarily providing companies the ability to terminate contracts during probationary periods due to declines in production. The notice period for mass redundancies was also shortened (Hansen, Sørensen and Andersen 2021).
Available evidence suggests that these measures were successful in preventing mass unemployment during the first phases of the pandemic (Hansen, Sørensen and Andersen 2021, Nordic Council of Ministers 2020, OECD 2020c). The different measures were mostly time limited and intended to act as a buffer for the worst labour market effects of the pandemic and mitigate rising unemployment without distorting long-term market conditions.
|Type 1||Measures to support businesses||6||6||12||14||5|
|Access to capital||3||3||3||5||1|
|Deferral of payments or liabilities||1||5||2||1|
|Type 2||Measures to protect jobs and support employees||9||2||4||5||8|
|Flexibilization of working hours||1|
|Income security of employees||2||1||1|
|Income security of unemployed workers||1||1||1|
|Income security for individuals without scheme coverage||3||2||3||1|
|Support for parents and caregivers||2||1||1|
|Type 3||Measures to stimulate economic activity||2||9||2||2||1|
|Active labour market policies, subsidies and job creation||1||2||2||1|
|Flexibility and security||6|
|Stimulation of economy||1||2|
|Protection of vulnerable groups (in addition to income security)||1|
Annex Table 1. Nordic support measures to avoid increases in unemployment.
Source: Adapted from Hansen, Sørensen and Andersen (2021).
In our contribution to the report prepared for the Nordic Council of Ministers in 2019 (Bratsberg, Raaum and Røed 2019), we argued that social insurance design may have profound effects both on migration patterns and labour supply behaviour of migrants once established in the host country. In particular, we emphasized the risks that large differences in welfare state generosity within a common labour market may trigger migration flows that are socially inefficient and lead to excess creation of precarious jobs with low pay. We also provided empirical evidence indicating that the degree of social insurance generosity has a large influence on the labour supply behaviour of immigrants because their employment and career prospects are on average much poorer than those of natives. Still, we concluded that such evidence does not defend reduced social insurance generosity nor tighter regulations of migration across Europe. We advised against a two-tier social insurance system with different rules for immigrants and natives. Instead, we called for greater labour market regulation to prevent the proliferation of short-term jobs with low productivity and pay, and more activity-oriented social insurance policies that involve education, training, or supported employment.
During the last year, the COVID-19 pandemic has put the common European labour market under severe stress due to historically high levels of unemployment, underemployment and soaring fiscal deficits. As the fight against the pandemic involves reducing human mobility, the immediate short-term consequences for labour migration have been restrictive, however, it remains unclear how the pandemic will influence long-term migration and residential patterns in Europe.
Do our experiences from the COVID-19 crisis call for a rethink of social insurance and migration policies in welfare states like Norway? The current crisis certainly calls for a reassessment of social insurance policies. In this chapter, we first provide a brief description of how the crisis hit immigrant and native workers in Norway. This description shows that immigrants from new EU member countries in Central and Eastern Europe have been vastly overrepresented among job losers. Hence, the inclusion of these workers in a relatively generous social insurance system has been critical for the prevention of poverty and for minimizing demand-driven knock-on effects from income decline in industries directly affected by the crisis. At the same time, we suspect from past experiences that many immigrants that have lost their jobs during the pandemic will have a hard time finding new employment (Bratsberg, Raaum and Røed 2018a). The COVID-19 crisis has also revealed how Norwegian firms have become reliant on migrant labour, particular within construction, farming, and health care, thereby exposing the country’s vulnerability with respect to disruptions of labour movements. The cost of hiring migrant labour has increased during the crisis due to requirements of testing, quarantines and adequate housing conditions for temporary workers. Hiring costs of foreign labour are likely to remain higher than before the crisis due to concerns about future pandemics. From the migrant worker perspective, the utility cost of migration is likely to be higher, partly due to increased uncertainty. The gains for domestic employers (and consumers) from labour migration are, therefore, likely to be lower on a permanent basis.
All in all, we conclude that the crisis has underscored the need for a generous and uniform social insurance system, but also highlighted the importance of maintaining a certain degree of “self-sufficiency” of labour in critical sectors of the economy. The arguments for labour market regulations preventing the development of an underpaid immigrant “underclass” have not become less relevant, although increased costs associated with the use of migrant labour may reduce socially inefficient migration.
Starting in March 2020, the coronavirus crisis led to a massive reduction in economic activity and historically high levels of unemployment. In Norway, the rate of unemployed, or underemployed, job seekers increased from 2.8% to 15.4% of the labour force in just about four weeks (Dahl et al. 2021).These numbers deviate considerably from the unemployment rates calculated from labour force surveys, as they also include furloughed workers who expect to return to their employer once the crisis is over and hence do not search actively for alternative employment. By the end of the year, one half million workers had filed for unemployment insurance (UI) benefits. In a familiar pattern from past economic fluctuations, immigrants were much harder hit than natives (Alstadsæter et al. 2020).
In response to the unprecedented rise in unemployment, the parliament quickly legislated a number of adjustments to the unemployment insurance system. The replacement ratio was (temporarily) raised from 62.4% to 80% for annualized earnings up to approximately NOK 300,000 (but remained at 62.4% for earnings between 300,000 and 600,000). Eligibility requirements were modified to ensure that UI entitlement was extended to workers with low past earnings, to the self-employed, and to freelancers. Employers’ obligation to pay the full wage during the first 20 days of furlough (temporary layoff) periods was cut to only 2 days, and the maximum duration of UI claims (including furloughs) was extended.
Based on individual report cards submitted bi-weekly by UI claimants, Figure 1 shows the total number of UI claimants from early March 2020 to mid-February 2021 among those aged 18-66 who worked before the onset of the crisis (around March 12th). For each week, all UI spells that started before or during the week are categorized as either completed or ongoing (active), and ongoing spells are again categorized on the basis of the degree of unemployment relative to pre-crisis hours worked.A completed spell in most cases (approximately 90%) involves return to employment, but there are also some transitions out of the labour force (including to education) and out-migration from Norway. UI benefits are reduced proportionally to reported hours worked, and Figure 1 illustrates that partial unemployment has been an important feature of the labour market during the pandemic. The unemployment patterns for the three immigrant groupsThe groups are defined in the figure note. depicted are similar to those of natives, but the scales differ dramatically. For all immigrant groups, the pandemic has hit harder and the job losses have lasted longer than for natives. Immigrants from new EU countries, such as Poland, have been hit particularly hard. As many as one in three of the female employees from these countries have been (temporarily or permanently) laid off at some stage during the crisis, compared to 13% among native women. It is also evident from Figure 1 that a smaller fraction of immigrant UI spells were completed by the end of the observation period. For example, by mid-February 2021, 63% of the UI claimant spells among male immigrants from the new EU countries were completed, compared to 75% for native men. A study of drivers behind the immigrant-native differentials in Figure 1 would be highly informative but falls outside the limits of this chapter. We know, however, that the sector of employment is key as the pandemic affected service sectors with many migrant workers more than other sectors (Alstadsæter et al. 2020).
Figure 1. Unemployment status of new UI applicants since March 1, 2020, by gender, immigrant group, and week of UI report card. Weeks ending March 8, 2020 – February 14, 2021.
Note: UI status is classified from weekly report cards filed by the UI claimant. Classification of claimants with positive hours worked is relative to hours in pre-crisis employment file. Population is restricted to individuals with a valid record (with non-zero hours and pay) in the January-March 2020 files of the employer-employee register, and who were residents and age 18-66 as of January 1, 2020. Old EU refers to the 15 pre-2004 EU member states; New EU to the 12 countries that joined in 2004 and 2007. Observation counts are 1,061,534 (native men), 51,894 (old EU/OECD men), 76,492 (New EU men), 108,729 (Other immigrant men), 1,015,240 (native women), 40,606 (old EU/OECD women), 46,983 (New EU women), and 107,953 (Other immigrant women). Major source countries are Sweden (26%), Germany (17%), and Denmark (11%) in the Old EU/OECD group; Poland (52%), Lithuania (22%), and Romania (8%) in the New EU group; and the Philippines (7%), Thailand (6%), Eritrea (5%), Russia (5%), Iraq (5%), and Somalia (5%) in the other immigrant group.
Source: Own tabulations from register data provided by the project “The coronavirus crisis and its economic consequences.”
While the pandemic has thrown many new workers into unemployment, it has also made the situation much more difficult for those who were already unemployed at the onset of the crisis. Figure 2 focuses on the employment status of workers who were registered as fulltime unemployed in February 2020 (i.e., prior to the coronavirus-initiated lockdowns). For comparison, we also include the same statistics for the fulltime unemployed observed exactly one year earlier. Again, we see that all immigrant groups have lower rates of return to employment than natives, and that these differentials increased during the pandemic. Employment rates are lower in 2020 than in 2019 for all groups, however, by the end of 2020, 61% of native women and 63% of native men in the extract had not regained employment. The comparable non-employment figures for the New EU group are 69% for women and 68% for men. For immigrants from other (low-income) countries, as much as 76% of the initially unemployed women and 70% of the men had not returned to employment by the end of the year.
Figure 2. Employment status over time among people registered as full-time unemployed in February 2019 vs 2020. By immigrant group and gender.
Note: Populations consist of individuals age 18-66 who were (fulltime) unemployed in February 2019 and 2020. Job is defined as having a valid record with non-zero pay and hours worked in the monthly employer-employee register file. Observation counts in the eight panels are 20,259/1,486/4,762/6,571 (top panels) and 12,969/852/3,060/7,140 (bottom panels) in 2019, and 18,993/1,318/4,445/6,548 (top panels) and 11,833/817/2,737/6,393 (bottom panels) in 2020.
Source: Own tabulations from register data provided by the project “The coronavirus crisis and its economic consequences.”
The pandemic dramatically reduced labour demand and raised recruitment costs of migrant labour for domestic employers as well as the migration costs for foreign workers. Combined with the high unemployment rates among foreign workers already in the country, one might expect that the inflow of new migrant workers would come to a halt. To take a closer look at labour migration patterns during the crisis, Table 1 reports the number of employed workers in Norway at the end of the year by immigration status. The first rows of the table illustrate the reduction in overall labour demand, as the number of employed natives and resident immigrants dropped by 0.6% and 0.5%, respectively, from 2019 to 2020. The reduction in the number of employed temporary migrants (i.e., foreign employees in Norwegian firms without residency, i.e., with a primary residence abroad) who also held a job in Norway the prior year was much higher, 15.7%. And the inflow of new migrant workers, with or without residency, dropped by as much as one quarter. For comparison, we also include the pre-crisis growth rates from 2018 to 2019.
All in all, Table 1 shows that although the net inflow of migrant labour did fall during the pandemic, it did not come to a complete stop. It is also worth noting that a closer inspection of the register data used to create Table 1 reveals that many of the new labour migrants took jobs in sectors with high rates of temporary and permanent layoffs. Looking at the industry-specific rates of layoffs and recruitment of new migrants (at the two-digit NACE-level), we actually find a positive (and statistically significant) correlation between the two measures.The correlation coefficient between the fraction of the pre-crisis workforce laid off as of December 2020 and the fraction of the workforce that were new migrants recruited during the year is 0.33; and the regression coefficient from a simple regression of the new migrant rate on the layoff rate is 0.104 with standard error 0.032; n=86. Although we have not been able to investigate this pattern at the firm level, the observed combination of layoffs and new recruitment at the sectoral level suggests that the problem of excess churning (the simultaneous layoff and recruitment of similar workers, partly motivated by welfare state subsidies embedded in the social insurance system), may have been relevant also during the pandemic (as discussed in Bratsberg, Raaum, and Røed 2021).
|Headcount||Change (%) from prior year|
|Temporary migrants with employment record in prior year||47,999||-15.7||11.8|
|Migrant workers who arrived during year||27,625||-25.7||-6.7|
Table 1. Employment at end of year and employment change in relation to prior year, age 18-66.
Note: 2019 change statistics are relative to employment in December 2018.
Source: December files of the employer-employee register and demographic data from Statistics Norway.
The current economic crisis has hit immigrant workers much harder than natives, confirming that immigrant labour is particularly sensitive with respect to cyclical fluctuations. This implies that a dual social insurance system with poorer coverage for immigrants would also be a system with less effective insurance and weaker automatic stabilizers. Given that immigrants typically also have lower income and wealth than natives, a dual system would also be a source of increased economic inequality and more poverty. If anything, the crisis has underscored the importance of an efficient and unified social insurance system.
The crisis has also highlighted the importance of - and challenges that come with – extensive labour migration. Over the past fifteen years, the migrant worker share in many Norwegian industries has grown. A likely consequence of our recent experiences is that the cost of foreign labour will increase because of the threat presented by new pandemics that will require more adequate housing conditions for temporary migrants, and because the utility cost of migration may increase due to a higher perceived risk of disruption to the free movement across borders. The need for ensuring the availability of labour with critical functions – particularly within the health care sector – has also become more evident.
The crisis has not eliminated incentives to exploit cross-country differences in welfare state generosity as an implicit wage subsidy, potentially generating socially inefficient migration flows. There may still be a case for strengthening policies aimed at preventing social dumping, such as higher minimum wages and other standard requirements. There may also still be a case for raising questions about European regulations that make social insurance entitlements completely (and immediately) transferable across countries upon labour migration. A social insurance system that to a large extent links entitlements based on past contributions to the country (or countries) where the contributions have been made, is not at odds with the principle of unified social insurance, nor will it obstruct free movement of labour. However, it may to some extent weaken the social insurance system’s role as an automatic stabilizer in times of very high unemployment. A unified social insurance system should in any case ensure that basic income support entitlements (such as social assistance) are independent of past contributions and are granted on equal terms to all residents, regardless of immigrant background.
While we expect the cost of using physically present immigrant labour to rise, the crisis may have paved the way for increasing the use of foreign labour in jobs that do not require a physical presence within the workplace. The crisis has probably precipitated a game-changing break-through for work from a distance (including from another country), due to improved distance communication technology and skills. This may trigger offshoring of jobs that have previously been protected from that type of competition and raise new intricate questions for the design of social insurance systems.
In Ek and Skedinger (2019a), my co-author Simon Ek and I showed that there are large groups of low-skilled immigrants in the Nordic countries and that labour market integration is problematic for these groups. We also documented relatively high wage floors in industries that employ many low-skilled workers, concluding that the productivity of numerous low-skilled immigrants does not seem to match the wage levels associated with the Nordic labour market models.Another explanation for poor labour market integration could be discrimination. While much research indicates that such discrimination exists in Nordic countries, there is also evidence that foreign born with high literacy skills have employment rates on a par with those of natives with similar skills, as discussed in, e.g., Ek et al. (2020). Lack of human capital, rather than discrimination, thus seems to be the main driver behind low employment rates for foreign born.
Within these circumstances, the findings in the research literature on the employment effects of collectively agreed minimum wages in the Nordic countries, as well as legitimate concerns regarding wage inequality with across-the-board minimum wage reductions, we argued that there is a case for targeted cuts of minimum rates. These cuts should apply only to newly created permanent jobs for the low skilled, like janitors, receptionists, handymen, caretakers, manual help/assistants in construction and “pick and pack” jobs in warehouses. We also advocated larger differentiation of minimum wages according to experience in existing jobs, as this would facilitate entry without necessarily lowering wages in a career perspective. In our opinion, lower minimum wages should preferably be negotiated between unions and employers and not be introduced by legislation, the latter being the common form of wage regulation in non-Nordic countries. Furthermore, we noted that Sweden stood out among the Nordics with the largest number of immigrants born outside the (pre-Brexit) European Union (EU28) relative to the working-age population, the largest employment gap between natives and those born outside the EU28 and the highest minimum wages in the low-wage sectors that we examined. The case for minimum wage cuts therefore appeared to be strongest in this country.
Finally, we argued that minimum wage reductions should not be seen as a substitute to other measures aimed at improving the labour market situation for immigrants, such as education, training and employment subsidies, but rather as a complement, in addition to these measures. The employment effect of each individual measure is uncertain and likely not very large, which, combined with the gravity of the situation, speaks to the necessity of using all available instruments in the policy toolbox.
Since we wrote our chapter in the report, the breakout of the COVID-19 pandemic has had profound effects on the labour markets in the Nordic countries and all over the world. Given that the pandemic is far from over at the time of writing, as well as lags in the availability of data, it is too early to draw definitive conclusions regarding its impact on the labour market for immigrants in the Nordic countries in the short run, let alone in the longer term. With these shortcomings in mind, I will try to assess what recent developments imply for the conclusions we reached regarding the labour market situation for immigrants and for policy analysis.
A first thing to note is that the share of immigrants born outside the (post-Brexit) EU27 in the working-age population has continued to increase in all of the Nordics, also during the pandemic, amounting to 19.6 percent in Sweden, 13.3 percent in Norway, 9.5 percent in Denmark, 5.6 percent in Finland and 5.1 percent in Iceland in 2020, according to Eurostat. While immigration flows have diminished since the start of the pandemic, differences in age structure between natives and foreign-born incumbents, the latter being younger on average, still contribute to increasing the shares of persons born outside the EU27 in working ages. Compared to 2016, the year referred to in Ek and Skedinger (2019a), the most recent figures from 2020 represent increases of the share of immigrants born outside the EU of approximately 3.2 percentage points for Sweden, 3.7 for Norway, 1.4 for Denmark, 1.3 for Finland and 2.1 for Iceland.
Figure 1 displays employment rates for immigrants born outside the EU27 and natives in the Nordic countries as well as other European countries in 2020 (panel a) and 2019 (panel b), updating the figure for 2017 in Ek and Skedinger (2019a). Employment rates for natives remain high in the Nordic countries in international comparison and it is still the case that corresponding rates for immigrants are similar to those in other countries (with the exception of Iceland), implying relatively large employment gaps between natives and immigrants. Compared to 2019, employment rates for both natives and those born outside the EU27 have decreased in most countries, and generally more so for the latter group. This has meant increased employment gaps during the first year of the pandemic in three of the Nordics: Sweden (2.2 percentage points), Iceland (2.1 percentage points) and Norway (1.0 percentage point). Denmark and Finland, however, saw increasing employment rates for immigrants from outside the EU27 (as did the UK), implying that employment gaps narrowed by 1.7 and 1.1 percentage points, respectively. Table 2 in Chapter 1, showing employment rates by gender in 2019 and 2020, suggests that the differential development in employment gaps during the pandemic between Sweden and Norway, on the one hand, and Denmark and Finland, on the other hand, is driven by increases in employment rates for foreign-born females in the latter group of countries.
Figure 1. Employment rates for natives and immigrants born outside the EU27, 20-64 years, percent of population group.
Note: The data for the UK in 2020 refer to first to third quarters only. The countries within each group are ordered in both panels by the 2020 difference in employment rates between natives and immigrants born outside the EU27.
Compared to 2017, in Ek and Skedinger (2019a), Figure 1 in this chapter shows that the large gaps in the Nordics have increased even further in Sweden (from 19.3 in 2017 to 21.9 percent in 2020) and Norway (from 15.0 to 16.3 percent). As discussed previously, these countries have also seen the largest increases in the share of immigrants during the period. The earlier conclusion that Iceland does not have a serious integration problem remains, although the country experienced the largest increase in the employment gap among the Nordics since 2017, from 2.9 to 8.1 percent.
We showed in Ek and Skedinger (2019a) that the collectively agreed minimum wages in 2016 for low-skilled workers in the hotels and restaurants industry and retail – two major low-wage sectors – were relatively high in the Nordics compared to other countries, and especially so in Sweden.Minimum wages in cleaning, another important low-wage sector, are regulated in a large number of different collective agreements. In absolute terms and expressed in common currency, however, minimum wages in Norway and Denmark exceeded those in Sweden.
How have minimum wages in the Nordics developed since then? Table 1 reports hourly minimum wages in national currencies from 2017 up to the latest available year. Lags in data availability preclude relating the rates to recent information on median wages in the economy as a whole, but data for 2018 show that relative minimum wages in Sweden, ranging between 67 and 69 percent of median wages, are higher than in the other Nordic countries. Increases in the absolute level of minimum wages in Sweden during 2017–2021 – 7.2 percent in hotels and restaurants and 8.5 percent in retail – were either similar to or higher than those in the other Nordics. The hotels and restaurants industry is also one of the sectors that have been hardest hit in terms of job losses during the COVID-19 crisis.
In response to the COVID-19 pandemic, minimum wages in the industries examined in Table 1 were frozen during 2020 to some extent in all of the Nordics, except Denmark. This was a consequence of postponement of some collective agreements – from April to November in Sweden and from May to October for hotels and restaurants in Finland (although the latter rate was unchanged also for several years prior to 2020). At least as far as Sweden is concerned, for which I have the longest time series of minimum wage data, such freezes are very unusual and did not occur for adult workers even during the deep recessions in the early 1990s and 2008–2009.Annual inflation rates in Sweden were considerably higher during 1990–1993 than in 2020 and somewhat higher during 2008–2009, so a minimum wage freeze would have resulted in larger declines in the real minimum wage in the two earlier crises. The postponement of collective agreements and the economic downturn have also affected average wage growth in the economy. Sweden experienced a palpable reduction of wage growth in 2020, since a majority of wage earners were subject to postponed agreements.See National Mediation Office (2020) for more details.
a) Hotels and restaurants
|Percent of median wage, 2018||60.8||57.5||64.5||66.6|
|Percent increase of minimum wage, 2017-2021||7.5||0.9||7.9a||7.2|
|Percent of median wage, 2018||57.7||62.0||58.9||68.8|
|Percent increase of minimum wage, 2017-2021||7.9||6.7||3.8a||8.5|
Table 1. Hourly minimum wages in the Nordic countries, national currencies.
Note: The minimum rates are in nominal terms, refer to May each year and apply to low-skilled non-trainee manual workers (except cleaners, if covered by collective agreement) without experience and at least 20 years of age. If the hourly minimum wage is not stated in the collective agreement, it has been converted from the monthly minimum using the explicit formulas in the agreements. Minimum wages for Norway are not specified beyond the first contract year but may be adjusted based on the economic situation as well price and wage developments. Median wages are included here for the latest available year and refer to industry, construction and services (except public administration, defence and compulsory social security).
Source: Own compilation from various collective agreements for minimum wages. For details on the collective agreements included, see Ek and Skedinger (2019b). Eurostat for median wages and exchange rates.
To sum up, the conclusion in Ek and Skedinger (2019a) that Sweden stands out among the Nordics in a way that makes the integration problem more severe is supported also by later developments, up to the first year of the pandemic. If anything, the situation has become even more pressing, due to a larger increase in the share of immigrants born outside the EU27 than in the other Nordics (except Norway) as well as a larger increase in the employment gap between natives and immigrants (except Iceland). This is not to say that Sweden is performing decidedly worse than the other Nordic countries when it comes to integration, but the larger size of the immigrant population also makes the integration problem larger. Because of this, I will focus on Sweden in the policy discussion below, but many arguments are also applicable to the other Nordic countries.
The policy discussion in Ek and Skedinger (2019a) emphasised the importance of education and training in lowering the risk that workers get stuck in low-wage jobs. We also argued for the introduction of financial study support given directly to individuals and targeted earned income tax credits to reduce net-of-tax wage inequality and stimulate labour supply to such jobs. Basic education, like language training, is essential for enhancing the possibilities for immigrants to enter the labour market in the first place. The fact that linguistic distance, in terms of vocabulary, grammar, pronunciation and in some cases also alphabet, is high for immigrants from Asia and Africa in the Nordic countries underscores the potential importance of language training. According to studies from many countries, including the Nordics, there is a positive association between immigrants’ language proficiency and employment and wages.See Ek et al. (2020) for a recent survey. Causal relationships between language training for immigrants and labour market integration have been established in two recent studies, by Arendt et al. (2020) for Denmark and Lochmann et al. (2019) for France.
Language requirements are, on average, lower in low-skilled jobs than in other jobs. Figure 2 shows the distribution of language requirements, as reported by employees in the United States, in all occupations (panel a) and in jobs roughly corresponding to those requiring primary education at the most (panel b), according to the Swedish Standard Classification of Occupations (SSYK).It is not obvious that tasks in the elementary jobs in SSYK are identical to those in their US counterparts, which may imply differences also in language requirements. However, the fact that there is a strong positive correlation in the US data between occupation-level language requirements and individual wages (see Handel, 2016), together with the observation that the elementary occupations in SSYK are relatively low paid (see Ek et al., 2020), is consistent with lower language requirements on average in these jobs also in the Nordic labour markets. The creation of more low-skilled jobs could lower the language-related thresholds to the labour market for immigrants. If such jobs are created as part of our previously proposed scheme with targeted minimum wage cuts, negative spillover effects on the wages of other workers in these sectors would be minimised.
Figure 2. Distribution of language requirements (scale 0–100) across occupations, percent.
Note: The vertical axes indicate percent and the horizontal axes score values. The data refer to the average of 968 occupations in panel a) and 39 occupations in panel b) where workers in the United States answered the question "What level of English language is required to perform your current job?". Elementary occupations roughly correspond to those that require at most primary education according to the Swedish Standard Classification of Occupations (SSYK). The data have been standardised to a score scale of 1–100. Mean values: panel a) 53.7, panel b) 38.7.
Source: Ek et al. (2020), based on data from O*NET.
In Ek and Skedinger (2019a), we argued that high wage floors constitute impediments for integration of low-skilled immigrants. High minimum wages do not only price many low-skilled jobs out of the labour market but could also lead to excessive requirements regarding language proficiency and other qualifications in the jobs that remain. With an increasing supply of workers to low-skilled jobs following a minimum wage hike, employers may raise standards to reduce the number of applicants.Analysing job postings in the United States, Clemens et al. (2020) find that employers are more likely to require a high school diploma when minimum wages increase. Language requirements were not investigated. Such adjustments from employers are likely to be more harmful to immigrants than to natives.
The COVID-19 crisis has shown that it is unlikely that collectively agreed minimum wages in Sweden’s low-wage industries will be adjusted in a way that facilitates integration even in the face of an extremely deep recession, without external pressure on unions and employer organisations to do so. The so-called establishment jobs, a government initiative before the COVID-19 crisis to subsidise low-skilled jobs, targeted on newly arrived immigrants and the long-term unemployed, have so far failed to materialise, adding to the seriousness of the situation.See Calmfors et al. (2018) and Swedish Trade Union Confederation (2017) for a discussion of the establishment jobs. The reasons for the tardiness are not entirely clear, but a contributing factor could be that the focus during the COVID-19 crisis has shifted to protecting existing jobs, and away from creating new employment opportunities.
There are good reasons to be skeptical towards government intervention in wage formation, or other matters subject to collective agreement. This is primarily because of the superior information of unions and employers and the fact that collective agreements are more conducive to stable long-term solutions than legislation, which may be changed for opportunistic reasons in connection with, for example, election campaigns. Even so, threats of legislation seem to have worked in other contexts in Sweden, when nothing else has convinced the negotiating parties about the necessity to reach an agreement. The Saltsjöbaden Agreement in 1938, regulating negotiation procedures, and the Industrial Agreement in 1997, establishing that the negotiated wage increase in the manufacturing sector should not be exceeded in other sectors, are two important examples from the previous century. A recent example is the agreement on reformation of employment protection, following an official report from the Swedish government, after a decades-long deadlock in negotiations.See SOU (2020) and Confederation of Swedish Enterprise (2020). Setting up a government commission with directives to propose the creation of permanent, new jobs with lower, legally mandated minimum wages, unless unions and employers agree on other solutions, could prove effective also for improving integration.
What will the future look like for the integration of immigrants in the Nordic countries, after the COVID-19 crisis? On the positive side, and as argued in Forslund (2020), many industries that suffered large job losses and employ many immigrants, such as hotels and restaurants and other providers of personal services, are not in structural decline and should be able to recuperate in due course (even though some firms in these sectors will not survive). On the negative side, there is evidence that exposure to high unemployment rates have detrimental, long-run effects on the labour market prospects for new entrants.See, for example, Åslund and Rooth (2007) for Sweden and Raaum and Røed (2006) for Norway. As the economy recovers, a large number of low-skilled refugee immigrants will attempt to gain a foothold on the labour market, at the same times as many young natives, who have the advantage of fluency in Swedish, will be looking for entry-level jobs. Consequently, competition for such jobs may turn out to be fierce. Even before the pandemic, the Swedish Public Employment Service forecasted an increase in excess supply of labour for several elementary occupations for the period up to 2024.See Swedish Public Employment Service (2019), which is the most recent detailed forecast.
Should the positive side prevail, with a return to a “normal” situation in the Nordic labour markets, the conclusions in Ek and Skedinger (2019a) still remain valid. Under the more negative scenario, the case for a multi-pronged approach, including minimum wage cuts and other measures to improve the integration of immigrants, is strengthened even further.
The labor market integration of immigrants has for a long period of time been a major policy concern in the Nordic countries. The COVID-19 pandemic has made social and economic inequalities even more pronounced and imposed new challenges for policy in relation to the functioning of labor markets. This chapter provides some reflections on the contributions in this volume and discusses recent related and topical issues in the policy and public debates based on recent research.
The chapter by Sánchez Gassen and Penje presents a variety of statistics illustrating recent labor market developments in the Nordic countries, with a particular focus on immigrant-native unemployment and employment differentials. The overall picture is in line with other international evidence suggesting that the pandemic hit the labor market prospects of the foreign-born particularly hard (Fasani and Mazza 2020; Borjas and Cassidy 2020). This trend follows a common experience from previous economic downturns where marginal groups are typically more affected than those with a stronger foothold. Furthermore, the pandemic has had a particularly adverse impact on industries and occupations employing many migrant workers and providing ports of entry to the labor market, such as the hospitality and retail sectors.
But the picture is not uniform across countries and groups, and the impression varies somewhat depending on the outcome in focus. For example, the development of unemployment rates is more dramatic and suggestive of the foreign-born being hit harder, whereas employment rates in many cases are back on pre-pandemic levels. For young people, the pandemic has in many Nordic countries seen a greater decline in the employment rates of natives than among the foreign-born (although from different initial levels). There are signs that labor market conditions are improving, which raises hopes that the situation for the foreign-born can bounce back. However, it is important to note that we are yet unlikely to have the full picture of the labor market consequences of the pandemic. As discussed in the chapter, there may be issues with unreliability in survey data stemming from low sample rates in certain sub-populations. Also, one should be cautious in interpreting short-term differential trends since changes in the composition of the foreign-born population (e.g. duration of residence in the host country), as well as people entering and leaving public programs, could have an impact. Even though figures are correct and show what they show, they do not necessarily reflect e.g. changes in the functioning of the labor market.
More important for future developments, we do not yet know what the long-term consequences of the pandemic will be for labor markets. First, substantial public measures supporting workers and firms in economic distress may mask challenges that will become more apparent with time. For example, immigrants often work in low-productive firms (Åslund et al. 2020), which may further influence immigrant-native disparities to the extent that it affects firm hiring and performance prospects in the recovery. Another issue is the potential long-term consequences of the greater health impact on the foreign-born and the more adverse educational effects on children and young people living in poorer socioeconomic conditions. Such effects are hard to gauge based on the current data, but they are at least likely not to improve the labor market chances among immigrants and their descendants.
The overall message is that slow labor market entry of many recently arrived immigrants and economic gaps lasting over generations have for a long time been a significant social and economic concern, and will likely continue to be so. Recent developments related to the pandemic appear to have exacerbated the situation. But there may also be possibilities for a more positive perspective; we will return to this towards the end of the chapter.
While the Nordic countries share many of the problems (and opportunities), some issues are more urgent and influential in specific countries than others. The two remaining contributions in the volume can be seen as an illustration of this. Bratsberg, Raaum and Røed focus on Norway and discuss whether experiences from the pandemic call for a reform of the welfare system and/or migration policies. While the evidence suggests that immigrant workers, especially those from “new” EU member states in Central and Eastern Europe, have been much more likely to lose their jobs, the authors argue for a unified social insurance system providing support for individuals and automatic stabilizers for the economy as a whole. But they also conclude that it is still relevant to consider how the design of welfare systems may create incentives for socially inefficient migration flows exploiting cross-country variations in income levels and welfare generosity. The authors also note that the crisis has illuminated the potential risks of reliance on foreign labor and argue that an increased degree of self-sufficiency may be needed if the future holds greater uncertainty and costs for international mobility. In other words, the chapter considers issues most relevant for Norway, a country that has been the destination for labor migrants covered by free movement and transferability of social rights.
The chapter by Skedinger considers the role of skill differences and wage floors for immigrant-native employment gaps. Here, the focus is on Sweden where the unemployment of the foreign-born increased more than in the other Nordic countries during 2020, already starting from a high level. The country also has a larger population of recently arrived immigrants from countries where previous immigrant cohorts have been less successful in the Swedish labor market. The author concludes that skill requirements in terms of language proficiency impede immigrant opportunities in an economy characterized by high-productive and high-wage jobs. He advocates targeted minimum wage cuts applicable to new permanent jobs in low-skilled occupations and argues for the need for greater differentiation based on experience in existing jobs. Furthermore, despite the slowdown of wage growth in 2020, and the temporary freeze of wages in some collective agreements, the development of entry wages in the years before the pandemic signals that the relatively high minimum wages in Sweden remain. Skedinger also discusses the pros and cons of political interventions in wage determination. He concludes that collective agreements are unlikely to be adjusted to facilitate integration through differentiated wages without pressure on the social partners, and points to the potential importance of threats of legislation.
Even though concerns of both types are applicable to both Norway and Sweden (and the other Nordic countries), it is notable that the Norwegian contribution to some extent focuses on what is needed to maintain an economy with high minimum wages, whereas the Swedish case emphasizes the potential negative consequences for immigrant labor market integration. This can be seen as an illustration of the fact that economic policy has different sides, and that current conditions may influence how the focus is set. When economic migration is more influential, concerns about the impact on wages receive more weight compared to when the more dominating issue is the labor market entry of humanitarian migrants.
One could argue that migration and integration policy should be separated. Migration policy determines who can enter under which conditions, and tries to regulate the composition and size of migration flows. Integration policy attempts to influence the outcomes of migrants and their interactions with, and impacts on, the receiving society and population. But this separation is in theory. In practice, the two policy areas are connected in the implementation of legislation and support systems, as well as in the political debate.
In a broad perspective, one aspect of the current debate within the Nordic countries considers what a well-functioning welfare state looks like in a society characterized by diversity in region of origin and where many workers continue to have strong ties to more than one country. What support systems are both sustainable and offer reasonable benefits and protection? How should one consider inequality across groups and contexts, and what mechanisms affect behavior and outcomes?
These issues are central to the discussion of Bratsberg et al. in this volume. In previous work, the same authors have found a greater responsiveness to benefit generosity among immigrant workers (Bratsberg, Raaum, and Røed 2018b) and documented long-term excess entry into support systems among the foreign-born (Bratsberg, Raaum, and Røed 2010; 2014). As much of the discussion today focuses on humanitarian and family reunification migrants, it is important to note that their findings also pertain to early cohorts of labor migrants. In fact, recent results for Sweden show that the long-term overrepresentation of immigrants relative to natives in early retirement has been decreasing across cohorts at the same time as the composition of immigrants has shifted from labor migrants from nearby countries towards asylum seekers from more distant locations (Laun, Liljeberg, and Åslund 2020). This suggests there may be factors in the characteristics of labor market and social experiences among migrants that go beyond the vast differences in economic outcomes seen shortly after immigration across migrant groups. Also, the evidence indicates that welfare reforms may have a differential impact on the immigrant population.
A related issue is whether migration flows are affected by welfare generosity. A recent study from Denmark that has received considerable attention finds that the number of asylum seekers to Denmark has responded significantly in response to changes in the benefits available to those granted asylum (Agersnap, Jensen, and Kleven 2020). Using variation in different dimensions, the authors find that benefit cuts decrease the number of asylum seekers and that higher benefits have the opposite effect. They also make another point with particular relevance for a Nordic context, arguing that their results should be seen as affecting destination choice conditional on migrating; i.e., benefits do not trigger the decision to flee/migrate, but influence where migrants go. Since other Nordic countries may be likely alternative destination choices, some form of coordination may be warranted. Indeed, recent research suggests spillover effects from migration policy reforms in the Nordic region (Bratu et al. 2020). Furthermore, the authors of the Danish study point to the risk of a “race-to-the-bottom” where countries attempt to deter unwanted migrants. One could imagine that such processes could lower benefits to a level which may be suboptimal both considering prevailing preferences regarding inequality and for promoting labor market and social integration.
A closely connected discussion considers how welfare systems affect economic and social integration. In addition to arguments on fairness and correspondence between contribution and eligibility, the current political discussion tends to emphasize the incentives brought by less generous welfare state arrangements. Here, recent research based on Danish experiences provides some relevant perspectives (Andersen, Dustmann, and Landersø 2019)A revised manuscript is being circulated under the title “Refugee Benefit Cuts” (dated March 2021).. In line with previous findings based on the same 2002 reform, lowered benefits for refugees appear to have had positive employment effects, although relatively short-lived. The impact was greater in local labor markets where there was a demand for the skills brought by refugees. This illustrates the point that reforms affecting incentives on the margin are most likely to affect those closer to the labor market in the first place. For those further away, it may not matter how much one gains from moving from welfare to work.
Other results in the study by Andersen et al. (2019) are even more intriguing. First, the authors claim that the reform embedded disincentives which caused female migrants to leave the labor force. In other words, the reform may in effect have constituted a gender inequality trap.The authors argue that one mechanism was means-testing transfers at the household level in combination with requirements on participation in integration programs, which decreased incentives for females to remain in the labor force when their spouses enter employment more rapidly. Another mechanism affected households where one spouse arrived before the reform and one after. In these cases, all transfers were allocated to the first-arriving spouse, which was typically the male. This also decreased the incentives for female labor force participation. This mechanism was of course temporary (limited to the time around the reform), but still illustrative of the complex issues at hand. Second, the study concludes that the sharp drop in disposable income brought by the reform for a substantial share of the households in fact increased crime rates. For adults, the effect mostly pertained to property crime relatively shortly after immigration. For adolescents, also violent and gang related crime increased significantly over time. These results do not necessarily mean that welfare reforms reducing benefits are counter-productive. Rather, they illustrate that policy reforms, in addition to intended and desired consequences, may have unintended and damaging impacts on individuals and society.
The literature on active labor market policies targeting the foreign-born is encouraging in the sense that various interventions appear to be able to contribute positively (Åslund, Forslund, and Liljeberg 2017; Hernes et al. 2020; Sarvimäki and Hämäläinen 2016). Recent findings confirm that more extensive language training (Arendt et al. 2021) as well as programs directly involving employers (Dahlberg et al. 2020) can be effective tools. But even if traditional interventions have statistically and economically significant effects, substantial gaps and disadvantages remain. As a result, debate and discussion need to also consider other factors and policy alternatives.
One idea that seems to be gaining momentum is that restricting immigration improves integration. Denmark has recently received international attention (and critique) for proposals explicitly aimed at reducing the number of asylum seekers. Sweden is currently developing new legislation in the area, and a central part of the discussion concerns its impact on future immigration. Leaving ethical concerns and fulfillment of international commitments aside, it is interesting to note that in both countries, restrictive policy reforms are motivated by their expected positive impact on integration. Is there evidence supporting such ideas? In the aggregate, part of the answer is obvious: the larger a poor-performing group is, the greater an issue it is at a societal level. But policy makers would probably be wrong to hope that fewer migrants in itself would improve outcomes among those who do come. Theoretically, effects could go both ways. Resource constraints or limitations in job creation could make it harder for those who arrive to a country experiencing higher immigration. But there may also be economies of scale in interventions, or that a larger number of migrants make them more important and relevant for labor supply, affecting employer behavior and hiring strategies.
The question of how the size of migration flows affects integration is very hard to address empirically, but at least three observations urge caution. First, the employment gaps seen between refugees and natives (or economic migrants) are rather similar in countries experiencing very different levels of humanitarian migration (Brell, Dustmann, and Preston 2020). This type of observation holds also within the Nordic countries in the sense that outcomes converge and diverge in ways that are not directly related to the size of immigrant inflows (Hernes et al. 2020). Second, the stability of long-term outcomes over time seen e.g. in Sweden, despite variation in the size and composition of migrant inflows, signals that other factors are more important for economic integration (Åslund, Forslund, and Liljeberg 2017). Third, a detailed comparison of Finland and Sweden shows remarkable similarities in labor market entry and continued integration processes, despite the countries having very different levels of immigration, as well as immigration histories (Ansala, Åslund, and Sarvimäki 2021).
Partly related recent research provides some input to the discussion on the impact of temporary vs. permanent residence permits. In 2013, Sweden moved from temporary to permanent permits for most Syrians receiving asylum. Analysis suggests that short-term employment rates were higher under temporary protection, whereas those granted permanent protection were more likely to enroll in regular education (Jutvik and Robinson 2020). This finding is in line with migration regulation affecting short-term vs. long-term perspectives among refugees. A study of the effects of the general shift to temporary permits in 2015/2016 finds employment effects in the same direction, but (contrary to Jutvik and Robinson) also concludes that the probability of participating in education and recognition of foreign education increased (Blomqvist, Skogman Thoursie, and Tyrefors 2020). The authors interpret their results in terms of the increased rewards to labor market integration and the possibility that educational investments are considered a path to labor market entry. Finally, a study of a Danish change in the expected time before one could receive permanent protection found no clear labor market impact in any direction (Kilström, Larsen, and Olme 2021). Given that theoretical expectations are ambiguous, it is not very surprising that the empirical evidence is mixed. Also, the literature is limited, and long-term effects including health and family-related outcomes are yet to be investigated.
There is also a broader issue of carrots vs sticks in integration policy. Advocates of restrictive policies emphasize the role of incentives and the importance of immigrants “earning” rights in the new country (e.g. permanent employment as a criterion for family reunification, language requirements for citizenship). The opposite view argues that opening up institutions and making general conditions more favorable for migrants enhance the prospects of successful integration. While these issues connect to philosophical perspectives on rights and duties of individuals and groups, and there is unlikely to be a single correct scientific answer, recent research contains some support for inclusionary policies in a broad sense.
Several studies suggest that making citizenship more accessible improves the economic and social integration of immigrants. Evaluating reforms in Switzerland (Hainmueller, Hangartner, and Ward 2019; Hainmueller, Hangartner, and Pietrantuono 2017) and Germany (Gathmann and Keller 2018), researchers find positive impacts on earnings and employment, as well as on measures of social belonging and interactions.Naturalization does, however, not seem to be a way of avoiding discrimination (Vernby and Dancygier 2019). Furthermore, effects tend to be greater for marginalized groups of migrants.
There is also evidence that opening up labor markets to asylum seekers have positive effects both directly after being granted asylum and in a long-term perspective. A study of European countries between 1985 and 2012 suggests that exposure to an employment ban (i.e. not having the right to work) reduces employment in post-ban years by 15 percent, with effects lasting up to ten years (Fasani, Frattini, and Minale 2021). An evaluation of a German court ruling reducing the length of employment bans suggests 20 percentage points reduction in employment over five years for those who had to wait an additional seven months (Marbach, Hainmueller, and Hangartner 2018). Similarly, there are studies pointing to the labor market advantages of reducing asylum waiting times (Hainmueller, Hangartner, and Lawrence 2016; Hvidtfeldt et al. 2018), something which may also affect health positively (Hvidtfeldt, Petersen, and Norredam 2019).
Taken together, these studies indicate the importance of quicker access to labor market and other opportunities in the host country. Even though one can claim that the starting point may matter and that Nordic countries already have inclusionary policies in these areas, it is relevant to consider the direction of reforms and their impact.
The evidence presented and discussed in this volume paints a rather bleak picture. The labor market attachment of immigrants has been a concern in the Nordic countries for decades, and the COVID-19 pandemic does at least not seem to have made things better. In addition, the restrictions on schools and limitations on teaching opportunities can be expected to harm children and youth in marginalized groups even more. Also, we have yet to see the consequences of the differential impact on health, as well as what will happen to support systems and active policies when public budget discipline is reinstated.
Can one nevertheless hope for a better future? The pandemic has emphasized the need for labor in sectors and organizations already seeing strong hiring needs. Even though public (or publicly funded) employers in health and elderly care, child care, and education may hope that digitalization and increased hours among present workers might solve some needs, the realities of high retirement rates and an overall aging population remain. These are also activities where many (in particular female) migrants have found employment in the past decades, signaling the potential for reducing gender inequality in immigrant-native comparisons. The need for “self-sufficiency” in labor supply discussed in this volume can also be seen as an argument for making better use of all resources available.
Improving immigrant labor market integration and providing high quality in labor intense public services are two fundamental challenges facing Nordic countries. To make them become each other’s solution to some extent will require creativity and new ways of thinking. Employers will have to consider whether skill requirements can be adjusted to include broader groups of workers and reorganizing to make better use of critical competences.This is related to the ”unbundling” of tasks discussed in the academic literature (Fredriksson, Hensvik, and Nordström Skans 2018) and in policy contexts (Nordström Skans 2020). To the extent that necessity is the mother of invention, such progress can be more likely when organizations face greater restrictions on certain types of labor supply at the same time as demand for services increases.
Policy discussions on matters of great significance, especially in times of crisis, tend to look for an Alexandrian solution (i.e., a single and unexpected measure that solves the problem). The labor economist answer is that such solutions are rarely available when it comes to e.g. lowering unemployment or improving immigrant labor market integration. But on the positive side, there are several margins and mechanisms of importance, which can also be affected by policy. This includes supply side factors, such as the skill composition among marginal workers and incentives brought by welfare state arrangements. Demand side components include wages/labor costs, overall employer conditions, and mechanisms of discrimination/exclusion. There are also interventions and reforms to improve the matching between supply and demand in the labor market. All of these are thoroughly discussed in numerous previous reports (Calmfors and Sánchez Gassen 2019; Åslund, Forslund, and Liljeberg 2017).
Yet, 15 months after the onset of the COVID-19 pandemic, there are reasons to reflect upon what issues and measures are most important to consider right now. A first priority should be to limit the extent of long-term unemployment in marginal groups and the risk of people leaving the labor force. This includes monitoring whether the economic recovery leads to a demand for the labor supplied by immigrant groups with marginal labor market positions, or whether targeted (additional) support through lowered labor costs (subsidies or differentiated wages) are needed. Increased educational and training interventions may also be warranted. Such efforts should have a strong focus on providing skills that are in demand and make labor market (re-)entry possible. A second priority should be on minimizing the adverse consequences from school closures, online classes, and other restrictions on the educational system. Such effects are likely to be more severe among disadvantaged groups (Engzell, Frey, and Verhagen 2021), where immigrant children and youth are overrepresented. Since school performance is a crucial determinant of labor market prospects, increased short-term efforts can yield substantial long-term gains.
Finally, as illustrated by the contributions in this issue, designing well-functioning policies and support systems is difficult and involves trade-offs between different goals and groups. Policy reforms need to consider possible interactions and spillover effects between government agencies, support structures, and across sectors. Identifying and targeting margins and measures that really matter may also be harder when there are strong political tensions. But policy makers hoping to improve immigrant labor market integration would be wise to consider the existing evidence without regard of current popularity.
Adams-Prassl, A., Boneva, T., Golin, M. and Rauh, C. (2020): The large and unequal impact of COVID-19 on workers. Published on VOX, CEPR Policy Portal, 8 April 2020. Available at: The large and unequal impact of COVID-19 on workers | VOX, CEPR Policy Portal (voxeu.org).
Addabbo, T., Rodríguez-Modroño, P. and Gálvez Muñoz, L. (2015): Gender differences in labor force participation rates in Spain and Italy under the Great Recession. Revista de Economia Mundial, number 41, pp. 21-42. Available at: https://www.redalyc.orgnord2021-050.pdf866/86643415002.pdf.
Agersnap, O., Jensen, A. and Kleven. H. (2020): The Welfare Magnet Hypothesis: Evidence from an Immigrant Welfare Scheme in Denmark. American Economic Review: Insights, vol. 2, issue 4, pp. 527–42. Available at: https://doi.org/10.1257/aeri.20190510.
Alstadsæter, A., Bratsberg, B., Eielsen, G., Kopczuk, W., Markussen, S., Raaum, O. and Røed, K. (2020): Who got hit, when and why? Covid Economics, issue 15, CEPR Press and NBER Working Paper No. 27131.
Andersen, L., Dustmann, C. and Landersø, R. (2019): Lowering Welfare Benefits: Intended and Unintended Consequences for Migrants and Their Families. Rockwool Fondens Forskningsenhed. Arbejdspapir.
Andersson-Joona, P. (2020a): Pathways to work for refugees and family migrants. SNS Research report. Available at: https://www.sns.se/en/articles/pathways-to-work-for-refugees-and-family-migrants/.
Andersson-Joona, P. (2020b): Coronakrisen: Hur påverkas tudelningen på arbetsmarknaden? Artikelserie DEL 5 Externa Perspektiv. Segregation och COVID-19, Huddinge: Delmos. Available at: Coronakrisen: Hur påverkas tudelningen på arbetsmarknaden? (delmos.se)
Ansala, L., Åslund, O. and Sarvimäki, M. (2021): Immigration History, Entry Jobs and the Labor Market Integration of Immigrants. Journal of Economic Geography, no. lbaa038 (April). Available at: https://doi.org/10.1093/jeg/lbaa038.
Arbetsförmedlingen (Swedish Public Employment Service, 2021): Långtidsarbetslöshetens utveckling i spåren av pandemin. Analysavdelningen 2021-04-13, Diarienummer: Af-2021/0006 4010, Arbetsförmedlingen analys 2021:6.
Arbetsförmedlingen (Swedish Public Employment Service, 2020): Arbetsmarknadsutsikterna våren 2020. Available at: https://arbetsformedlingen.se/download/18.2bef8e33170a57d956513766/1592291999584/arbetsmarknadsutsikterna-varen-2020.pdf.
Arendt, J- N., Bolvig, I., Foged, M., Hasager, L. and Peri, G. (2021): Language Training and Refugees’ Integration. Working Paper 26834. Working Paper Series. National Bureau of Economic Research. Available at: https://doi.org/10.3386/w26834.
Arendt, J. N., Bolvig, I., Foged, M., Hasager, L. and Peri, G. (2020): Integrating refugees: Language training or work-first incentives? Working Paper 26834, Cambridge, MA: National Bureau of Economic Research.
Åslund, O., Bratu, C., Lombardi, S. and Thoresson, A. (2020): Firms, Productivity, and the Immigrant-Native Earnings Gap. In: Wages and Their Impact on Individuals, Households and Firms. Department of Economics, Uppsala University. Available at: http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-425941.
Åslund, O., Forslund, A. and Liljeberg, L. (2017): Labour Market Entry of Non-Labour Migrants–Swedish Evidence. Nordic Economic Policy Review, pp. 95–130.
Åslund, O., and Rooth, D.-O. (2007): Do where and when matter? Initial labour market conditions and immigrant earnings. Economic Journal, vol. 117, pp. 422–448.
Bambra, C., Riordan, R., Ford, J. and Matthews, F. (2020): The COVID-19 pandemic and health inequalities. Journal of Epidemiology & Community Health, vol. 74, issue 11, pp. 964-968.
Blomqvist, N., Skogman Thoursie, P. and Tyrefors, B. (2020): Restricting Residence Permits. Short-Run Evidence from a Swedish Reform. In Essays on Labor Economics: The Role of Government in Labor Supply Choices. Stockholm: Department of Economics, Stockholm University. Available at: http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-180933.
Borjas, G. J. and Cassidy, H. (2020): The Adverse Effect of the COVID-19 Labor Market Shock on Immigrant Employment. NBER Working Paper 27243.
Bratsberg, B., Raaum, O. and Røed, K. (2021): Excess churn in integrated labour markets. Journal of Population Economics, vol. 34, issue 3, pp. 865-892. Available at: https://doi.org/10.1007/s00148-020-00795-1.
Bratsberg, B., Raaum, O. and Røed, K. (2019): Social Insurance Design and the Economic Integration of Immigrants. In: L. Calmfors and N. Sánchez-Gassen (Eds.): Integrating immigrants into the Nordic labour markets, Nord 2019:024, Nordic Council of Ministers.
Bratsberg, B., Raaum, O. and Røed, K. (2018a): Job Loss and Immigrant Labor Market Performance. Economica, vol. 85, issue 337, pp. 124–151. Available at: https://doi.org/10.1111/ecca.12244.
Bratsberg, B., Raaum, O. and Røed. K. (2018b): Immigrant Responses to Social Insurance Generosity. SSRN Scholarly Paper ID 3170271. Rochester, NY: Social Science Research Network. Available at: https://papers.ssrn.com/abstract=3170271.
Bratsberg, B., Raaum, O. and Røed. K. (2014): Immigrants, Labour Market Performance and Social Insurance. The Economic Journal, vol. 124, issue 580, pp. F644–83. Available at: https://doi.org/10.1111/ecoj.12182.
Bratsberg, B., Raaum, O. and Røed. K. (2010): When Minority Labor Migrants Meet the Welfare State. Journal of Labor Economics, vol. 28, issue 3, pp. 633–76. Available at: https://doi.org/10.1086/650546.
Bratu, C., Dahlberg, M., Engdahl, M. and Nikolka, T. (2020): Spillover Effects of Stricter Immigration Policies. Journal of Public Economics, vol. 190 (October), p. 104239. Available at: https://doi.org/10.1016/j.jpubeco.2020.104239.
Brell, C., Dustmann, C. and Preston, I. (2020): The Labor Market Integration of Refugee Migrants in High-Income Countries. Journal of Economic Perspectives, vol. 34, issue 1, pp. 94–121. Available at: https://doi.org/10.1257/jep.34.1.94.
Calmfors, L., and Sánchez Gassen, N. (2019): Integrating Immigrants into the Nordic Labour Markets. Copenhagen: Nordic Council of Ministers.
Calmfors, L., Danielsson, P., Ek, S., Kolm, A.-S., Pekkarinen, T. and Skedinger, P. (2018): Hur ska fler komma in på arbetsmarknaden? Stockholm: Dialogos.
Clemens, J., Kahn, L. B. and Meer, J. (2020): Dropouts need not apply? The minimum wage and skill upgrading. Working Paper 27090, Cambridge, MA: National Bureau of Economic Research.
Confederation of Swedish Enterprise (2020): Huvudavtal om trygghet, omställning och anställningsskydd. Svenskt Näringsliv, Förhandlings- och samverkansrådet PTK, LO, Stockholm.
Dahl, E.S., Furuberg, J., Helde, I., Kalstø, Å.M., Kann, I.C., Myhre, A., Nicolaisen, H., Nossen, J.P., and Sohlman, M. (2021): Ett år med korona. Utvikling og utsikter for NAVs ytelser og brukere. Arbeid og velferd, No. 1/2021.
Dahlberg, M., Egebark, J., Vikman, U. and Özcan, G. (2020): Labor Market Integration of Low-Educated Refugees. 2020:21. IFAU - Institute for Evaluation of Labour Market and Education Policy. Available at: https://www.ifau.se/Forskning/Publikationer/Working-papers/2020/labor-market-integration-of-low-educated-refugees/.
Drefahl, S., Wallace, M., Mussino, E., Aradhya, S., Kolk, M., Brandén, M., Malmberg, B. and Andersson, G. (2020): A population-based cohort study of socio-demographic risk factors for COVID-19 deaths in Sweden. Nature Communications, vol. 11, article number 5097.
Ek, S., Hammarstedt, M. and Skedinger, P. (2020): Enkla jobb och kunskaper i svenska – nycklar till integration? Stockholm: Centre for Business and Policy Studies (SNS).
Ek, S. and Skedinger, P. (2019a): Wage policies and the integration of immigrants. In: L. Calmfors and N. Sánchez-Gassen (Eds.): Integrating immigrants into the Nordic labour markets, Nord 2019:024, Nordic Council of Ministers.
Ek, S and Skedinger, P. (2019b): Wage policies and the integration of immigrants. Policy Paper 86, Stockholm: Research Institute of Industrial Economics (IFN).
Engzell, P., Frey, A. and Verhagen, M. D. (2021): Learning Loss Due to School Closures during the COVID-19 Pandemic. Proceedings of the National Academy of Sciences, vol. 118, issue 17. Available at: https://doi.org/10.1073/pnas.2022376118.
Eurostat (2021a): Employment and unemployment (LFS) – Overview. Available at: https://ec.europa.eu/eurostatlfs.
Eurostat (2021b): Sample size and non-response – quarterly statistics. Available at: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Sample_size_and_non-response_-_quarterly_statistics.
Eurostat (2016): Labour force survey in the EU, candidate and EFTA countries. Main characteristics of national surveys, 2015. Luxembourg: Publications Office of the European Union.
Fasani, F., Frattini, T. and Minale, L. (2021): Lift the Ban? Initial Employment Restrictions and Refugee Labour Market Outcomes. Journal of the European Economic Association, issue jvab021 (May). Available at: https://doi.org/10.1093/jeea/jvab021.
Fasani, F., and Mazza, J. (2020): Being on the Frontline? Immigrant Workers in Europe and the COVID-19 Pandemic. 13963. IZA Discussion Papers. Institute of Labor Economics (IZA). Available at: https://ideas.repec.org/p/iza/izadps/dp13963.html.
Forslund, A. (2020): Coronakrisen och den svenska arbetsmarknaden. Ekonomisk Debatt, vol. 48, pp. 55–62.
Fredriksson, P., Hensvik, L. and Nordström Skans, O. (2018): Mismatch of Talent: Evidence on Match Quality, Entry Wages, and Job Mobility. American Economic Review, vol. 108, issue 11, pp. 3303–38. Available at: https://doi.org/10.1257/aer.20160848.
Gathmann, C. and Keller, N. (2018): Access to Citizenship and the Economic Assimilation of Immigrants. The Economic Journal, vol. 128, issue 616, pp. 3141–81.
Hainmueller, J., Hangartner, D. and Ward, D. (2019): The Effect of Citizenship on the Long-Term Earnings of Marginalized Immigrants: Quasi-Experimental Evidence from Switzerland. Science Advances , vol. 5, issue 12, eaay1610. Available at: https://doi.org/10.1126/sciadv.aay1610.
Hainmueller, J., Hangartner, D. and Pietrantuono, G. (2017): Catalyst or Crown: Does Naturalization Promote the Long-Term Social Integration of Immigrants? American Political Science Review, vol. 111, issue 2, pp. 256–76. Available at: https://doi.org/10.1017/S0003055416000745.
Hainmueller, J., Hangartner, D. and Lawrence, D. (2016): When Lives Are Put on Hold: Lengthy Asylum Processes Decrease Employment among Refugees. Science Advances, vol. 2, issue 8, e1600432. Available at: https://doi.org/10.1126/sciadv.1600432.
Handel, M. J. (2016): The O*NET content model: Strengths and limitations. Journal for Labour Market Research, vol. 49, pp. 157–176.
Hansen, M. E., Sørensen, S. Y. and Andersen, T. (2021): Analyse af udvikling og respons i nordiske arbejdsmarkeder som følge af COVID-19-pandemien. Aarhus: Teknologisk Institut. Available at: https://www.teknologisk.dk/ydelser/analyse-af-udvikling-og-respons-i-nordiske-arbejdsmarkeder-som-foelge-af-COVID-19-pandemien/42769.
Hernes, V., Arendt, J., Andersson Joona, P. and Tronstad, K. (2020): Rapid or Long-Term Employment? A Scandinavian Comparative Study of Refugee Integration Policies and Employment Outcomes. Journal of European Public Policy, latest articles, pp. 1–21. Available at: https://doi.org/10.1080/13501763.2020.1824011.
Hvidtfeldt, C., Petersen, J. H. and Norredam, M. (2019): Prolonged Periods of Waiting for an Asylum Decision and the Risk of Psychiatric Diagnoses: A 22-Year Longitudinal Cohort Study from Denmark. International Journal of Epidemiology, vol. 49, issue 2, pp. 400-409. Available at: https://doi.org/10.1093/ije/dyz091.
Hvidtfeldt, C., Schultz-Nielsen, M. L., Tekin, E. and Fosgerau, M. (2018): An Estimate of the Effect of Waiting Time in the Danish Asylum System on Post-Resettlement Employment among Refugees: Separating the Pure Delay Effect from the Effects of the Conditions under Which Refugees Are Waiting. PLoS ONE, vol. 13, issue 11. Available at: https://doi.org/10.1371/journal.pone.0206737.
Jutvik, K., and Robinson, D. (2020): Permanent or Temporary Settlement? A Study on the Short-Term Effects of Residence Status on Refugees’ Labour Market Participation. Comparative Migration Studies, vol. 8, article number 44. Available at: https://doi.org/10.1186/s40878-020-00203-3.
Kilström, M., Larsen, B. and Olme, E. (2021): Temporary Refugee Protection and Labor-Market Outcomes. Available at: http://www.matildakilstrom.com/wp-content/uploads/2021/03/main.pdf.
Kleine-Rueschkamp, L. and Özgüzel, C. (2020): COVID-19 and key workers: The role of migrants across regions and cities. Published on VOX, CEPR Policy Portal, 9 December 2020. Available at: https://voxeu.org/article/COVID-19-and-key-workers-role-migrants-across-regions-and-cities.
Laun, L., Liljeberg, L., and Åslund, O. (2020): Utrikes Födda Och Välfärdssystemen. IFAU Rapport 2020:15.
Lerpold, L. and Sjöberg, Ö. (2020): Those most negatively impacted by COVID-19. Preprint from the book „Sweden thorugh the crisis“. Available at: a49.pdf (hhs.se).
Lochmann, A., Rapoport, H. and Speciale, B. (2019): The effect of language training on immigrants’ economic integration: Empirical evidence from France. European Economic Review, vol. 113, pp. 265–296.
Marbach, M., Hainmueller, J. and Hangartner, D. (2018): The Long-Term Impact of Employment Bans on the Economic Integration of Refugees. Science Advances, vol. 4, issue 9, eaap9519. Available at: https://doi.org/10.1126/sciadv.aap9519.
Moodie, J., Tapia, C., Lofving, L., Sánchez Gassen, N. and Cedergren, E. (2021): Towards a territorially just climate transition – assessing the Swedish EU Just Transition Action Plan process. Paper under review for publication at Sustainability.
National Mediation Office (2020): Pandemin och löneutvecklingen. Stockholm.
Nordic Council of Ministers (2020): Nordic Economic Outlook 2020. Copenhagen. Available at: https://www.norden.org/en/publication/nordic-economic-outlook-2020.
Nordström Skans, O. (2020): En Arbetsmarknad För Ett Starkare Sverige. In Idéer För Ett Starkare Sverige. Omstartskommissionen. Ekerlids Förlag.
OECD (2020a): OECD Employment Outlook 2020. Worker security and the COVID-19 Crisis. Paris: OECD Publishing. Available at: https://doi.org/10.1787/1686c758-en.
OECD (2020b): OECD Policy Responses to Coronavirus (COVID-19). What is the impact of the COVID-19 pandemic on immigrants and their children? Paris: OECD Publishing. Available at: https://www.oecd.org/coronavirus/policy-responses/what-is-the-impact-of-the-COVID-19-pandemic-on-immigrants-and-their-children-e7cbb7de/.
OECD (2020c): Job retention schemes during the COVID-19 lockdown and beyond. Paris: OECD Publishing. Available at: https://read.oecd-ilibrary.org/view/?ref=135_135415-6bardplc5q&title=Job-retention-schemes-during-the-COVID-19-lockdown-and-beyond.
OECD (2020d): OECD unemployment rate up sharply in March 2020 with data for April pointing to an unprecedented rise. News Release: March 2020, Paris, 13 May 2020. Available at: https://www.oecd.org/sdd/labour-stats/unemployment-rates-oecd-05-2020.pdf.
OECD (2020e): International migration outlook 2020. Paris: OECD Publishing. Available at: https://www.oecd-ilibrary.org/social-issues-migration-health/international-migration-outlook-2020_ec98f531-en.
Papapetrou, E. and Bakas, D. (2013): The Greek labour market during the crisis: Unemployment, employment and labour force participation. Economic Bulletin 38, November 2013, pp. 65-83. Available at: https://www.researchgate.net/profile/Evangelia-Papapetrou/publication/326479951_The_Greek_labour_market_during_the_crisis_Unemployment_employment_and_labour_force_participation/links/5b5078c0a6fdcc8dae2cf908/The-Greek-labour-market-during-the-crisis-Unemployment-employment-and-labour-force-participation.pdf.
Raaum, O. and Røed, K. (2006): Do business cycle conditions at the time of labor market entry affect future employment prospects? Review of Economics and Statistics, vol. 88, pp. 193-210.
Sarvimäki, M. and Hämäläinen, K. (2016): Integrating Immigrants: The Impact of Restructuring Active Labor Market Programs. Journal of Labor Economics, vol. 34, issue 2, pp. 479–508. Available at: https://doi.org/10.1086/683667.
Sigurðardóttir, J. M. and Blöndal, L. (2020): Response rate during a pandemic. Response rate of the Labour Force Survey in Iceland during the COVID-19 pandemic in 2020. Conference of European Statisticians. Data collection and the impact, challenges and opportunities of the COVID-19 pandemic virtual workshop, 5-8 October, Virtual. Available at: https://unece.org/fileadmin/DAM/stats/documents/ece/ces/ge.58/2020/mtg4/DC2020_D2-4_Iceland_-_Sigur%C3%B0ard%C3%B3ttir_AD.pdf.
Sigurjónsdóttir, H. R., Sigvardsson, D. and Oliveira e Costa, S. (2021): Who is left behind? The impact of place on the possibility to follow COVID-19 restrictions, Copenhagen: Nordic Council of Ministers.
SOU (2020): En moderniserad arbetsrätt. 2020:30. Stockholm: Ministry of Labour.
Swedish Public Employment Service (2019): Var finns jobben? Bedömning för 2019 och på fem års sikt. URA 2019:1, Stockholm.
Swedish Trade Union Confederation (2017): LO:s förslag om utbildningsjobb för stärkt etablering. Stockholm.
Vernby, K. and Dancygier, R. (2019): Can Immigrants Counteract Employer Discrimination? A Factorial Field Experiment Reveals the Immutability of Ethnic Hierarchies. PLoS ONE, vol. 14, issue 7. Available at: https://doi.org/10.1371/journal.pone.0218044.
WHO (2021): World Health Organization Coronavirus Dashboard. Available at: https://covid19.who.int/.
Nora Sánchez Gassen, Oskar Penje, Bernt Bratsberg, Oddbjorn Raaum, Knut Roed, Per Skedinger and Olof Åslund
ISBN 978-92-893-7093-6 (PDF)
ISBN 978-92-893-7094-3 (ONLINE)
© Nordic Council of Ministers 2021
Layout: Louise Jeppesen
Cover photo: Ricky John Molloy/norden.org
Photos: Ricky John Molloy/norden.org, expect chapter 3: Mostphotos
Nordic co-operation is one of the world’s most extensive forms of regional collaboration, involving Denmark, Finland, Iceland, Norway, Sweden, and the Faroe Islands, Greenland and Åland.
Nordic co-operation has firm traditions in politics, economics and culture and plays an important role in European and international forums. The Nordic community strives for a strong Nordic Region in a strong Europe.
Nordic co-operation promotes regional interests and values in a global world. The values shared by the Nordic countries help make the region one of the most innovative and competitive in the world.
The Nordic Council of Ministers
Ved Stranden 18
Read more Nordic publications on www.norden.org/publications