Elevator pitch
The adverse effects of unemployment are a cause for concern for all demographic groups but they will be most acute for those experiencing the highest unemployment rates. In particular, high levels of unemployment are observed for a range of immigrant groups across many countries. However, there is considerable variation both across and within countries. It is therefore important to determine the factors that are most likely to cause high rates of unemployment, especially from a migration perspective, and to identify appropriate policy responses (e.g. enhancing human capital and improving job search effectiveness).
Key findings
Pros
Unemployment rates are lower for some immigrant groups than for the native born population.
The impact of economic downturns is lessened for more highly educated immigrants.
Policies that enhance human capital levels can be effective in reducing the high rates of unemployment experienced by immigrant groups.
Improving language skills is crucial for increasing the employment prospects of immigrants.
Cons
Certain immigrant groups, such as those with low levels of education, experience extremely high rates of unemployment, especially during recessions.
The unemployment rates of immigrants can be higher in countries that have more rigid labor market regulations and institutions.
Comparing countries can be difficult due to differences in their migration histories.
Newly arrived immigrants are often disadvantaged in terms of their country-specific skills and their knowledge of the labor market.
Author's main message
Unemployment often produces a range of adverse consequences for the individuals experiencing it and for the communities in which it is concentrated. A variety of factors can contribute to the high rates of unemployment experienced by some immigrant groups; these include human capital, cultural and religious differences, and discrimination. Appropriate government policies should recognize such inter-group differentials and group-specific determining factors. Such policies include the scheduling of courses to enhance human capital levels, such as skills in the host country’s main language.
Motivation
In addition to being a crucially important variable from a macroeconomic perspective, unemployment often has a major impact on the unemployed themselves, and its consequences tend to increase with the amount of time spent out of work. For example, unemployment is typically accompanied by adverse effects on health and well-being, resulting from the loss/lack of income and the absence of routine that regular employment provides. It therefore follows that those groups experiencing the highest rates of unemployment will be most affected by its negative consequences. Considering the widespread issues related to the integration of immigrants into host countries and social cohesion, the high levels of unemployment experienced by immigrant groups are of particular concern.
As well as the negative consequences that unemployment can have for the individual, it can produce wider effects if it is highly concentrated within particular communities. For example, the high rates of unemployment experienced by some groups may be associated with increased levels of crime and social unrest. As a result, it is important for policymakers to be aware of how and why unemployment rates may vary between groups that originate from different countries and ethnic backgrounds, as well as the types of policies that may be effective in reducing unemployment in the most affected groups.
A large international literature now exists on unemployment differences between ethnic minority and immigrant groups, especially in comparison to the native born. However, it is difficult to summarize these findings concisely given the wide diversity in the experiences of different groups across a range of economies. In particular, comparing the overall unemployment rates for immigrants with those for the native born population typically hides a large amount of heterogeneity between the different ethnic minority and immigrant groups within any given country. It is thus worthwhile to analyze cross-country differences in unemployment between native and foreign born residents and to further examine some variations in unemployment between different immigrant groups within those countries. This analysis allows for a discussion of the main reasons why unemployment rates may vary across such groups. It should be noted that the sole focus of this article in terms of labor market outcomes is on unemployment, even though immigration is linked to many other labor market differentials including participation rates, types of employment, wages, and occupation.
Discussion of pros and cons
Unemployment differences between native and foreign born residents
There is considerable variation in unemployment rates between natives and immigrants across OECD countries, as shown in the illustration on page 1. For example, in some countries the diference is very high—due to wider labor market difficulties, especially high overall levels of unemployment (e.g. Spain and Greece)—whilst it is relatively low in others, including some with a long-established history of receiving immigrants (e.g. the UK and the US). The unemployment gap between native and foreign born residents was highest in Belgium and Sweden in 2015 (where it exceeded 10 percentage points). In contrast to most countries, immigrants actually had lower rates of unemployment than natives in both Israel and the US. In addition to cross-country variations in overall levels of unemployment between the native and foreign born, there are also gaps by gender, as shown in Figure 1 and Figure 2. In some countries, such as Spain and Greece, unemployment rates for the foreign born were relatively high in 2015 for both men and women. By contrast, there were quite large differences by gender elsewhere, including in Denmark and Italy, where rates for foreign born women were relatively high compared with those for men, and in Ireland where the reverse was the case.
Figure 1 and Figure 2 show unemployment rates for 2007 (immediately prior to the Great Recession) and 2015 (the most recent year for which such information is available). Immigrants experienced a far larger increase in unemployment rates in some countries over this period, most notably in Greece and Spain (as well as in Ireland and Italy to a lesser extent)—all countries that have been severely affected by the Eurozone crisis. By contrast, in other countries unemployment rates fell for some immigrant groups, such as for foreign born men and women in Germany and the UK.
Cross-country variations in the size of the overall differences between immigrants and natives are likely due to a combination of factors, and these can account for why some immigrants may experience lower and others higher unemployment rates in certain countries. Such factors include issues relating to immigrants arriving from different countries of origin, which can be associated with migration cohort effects and labor market discrimination. For example, there may be variations in the characteristics of immigrants—including their skill levels—according to the time of arrival in the host country, whilst employers may display different amounts of prejudice. However, there are likely to be other important cross-country differences that will influence the relative levels of unemployment experienced by immigrants. In particular, some economies have more flexible labor markets. In these countries, natives and immigrants may experience similar unemployment rates, but other labor market differentials may be in play, such as with regard to wages. This is consistent with the situation observed in the US, as well as in the UK, where high proportions of immigrants are employed but often work in low-paying sectors.
Moreover, the evidence suggests that the labor market assimilation of immigrants can be influenced by institutions, given that they can affect how flexible the labor market is. For example, differences are observed between the US and Australia, where wages in Australia are less flexible and unemployment benefits are relatively more generous [1]. Cross-country differences in the generosity of welfare programs might also account for some of the variation in unemployment gaps, given that there is some evidence to suggest that immigrants are disproportionately attracted to countries where welfare payments are higher [2]. However, immigrants with relatively high levels of education who do find work can experience significant occupational downgrading after arriving in a new country [3]. Therefore, the presence of more rigid labor market regulations and institutions may help to explain the relatively high levels of unemployment amongst immigrants in some continental European countries, such as France, Italy, and Spain.
The criteria used to admit immigrants can also have an impact on their unemployment rates. This is connected to both the admissions policies for different categories of immigrants and to policies used to select economic migrants. For example, some countries may allow a higher proportion of immigrants for family (reunification) reasons than others, which can have an impact on their immigrant population’s labor market outcomes, often in an adverse way. By contrast, human capital levels will typically be higher for immigrants in countries that operate points-based systems of immigration, since, in these systems, immigrants are typically selected according to the amount of human capital (e.g. formal education and language skills) that they have. For example, the proportion of foreign born residents with tertiary levels of education is highest in Canada and Australia, both of which operate selective migration policies [4]. In the EU, freedom of movement regulations imply that there is the potential for unlimited migration between countries, although transitional arrangements have been imposed by some member states following recent enlargements. Such arrangements can restrict the number and type of employment opportunities available to immigrants from new member states.
The size of the immigrant population already residing in the host country could also have an effect on unemployment rates, with the potential for rates to be comparatively low in countries where there is only a relatively small number of foreign born residents. This likely explains why low rates could be recorded for the foreign born population in some countries in certain years.
Variations between immigrant groups
Combining the foreign born into a single category is not ideal given the diversity of immigrants, especially when considering their countries of origin. The composition of the immigrant population can vary considerably between host countries according to their socio-demographic characteristics, which are often heavily influenced by the length of residence in the country. Such factors impact an individual’s probability of unemployment. However, making comparisons between immigrants across countries is often challenging because of differences in migration histories, especially with respect to when and from where immigrants have typically arrived. Immigration is also a very dynamic phenomenon, and the situation in both sending and receiving countries can change quickly, as has occurred following EU enlargements [2].
One study examines this variation by presenting information on unemployment rates for native and foreign born residents in Germany and the UK, splitting the foreign born into those originating from inside and outside the OECD [5]. Germany and the UK have long and diverse histories of immigration, with substantial inflows of immigrants, especially from the second half of the 21st century onwards. In 2001, immigrants from non-OECD countries accounted for 33% of all immigrants in Germany and 64% in the UK. The composition of immigrant groups varies considerably, as indicated by the largest country of origin in each category. In Germany, 42% of OECD immigrants originated from Turkey, whereas in the UK, 20% were born in Ireland. For non-OECD immigrants, those from the former Yugoslavia accounted for 44% in Germany, while immigrants from India and Pakistan accounted for 24% of this category in the UK. In line with the evidence shown in Figure 1 and Figure 2, the difference between unemployment rates for natives and immigrants was lower in the UK than in Germany, especially for immigrants born in the OECD (as this group had very similar rates to natives). These differences can at least be partly explained by the characteristics of the immigrant populations in the two countries, including their levels of education [5]. Higher unemployment rates were experienced by immigrants in the UK from non-OECD countries, especially men, and some groups of immigrant women, especially Muslim women, have very low levels of labor market participation, as will be discussed further below [6]. In contrast, unemployment for immigrants from both inside and outside the OECD was higher in Germany than in the UK, with a higher rate observed for men from non-OECD countries. Educational and cultural differences are again likely to be important here.
Further explanation of the differences
As with other labor market outcomes, the probability of experiencing unemployment can be affected by a wide range of personal characteristics and socio-economic factors. These influences often, but not always, impact the native and foreign born in similar ways. For example, the likelihood of unemployment will typically fall with the amount of education that an individual has acquired. However, the extent to which unemployment is affected by specific factors varies across immigrant groups, such as when comparing Turkish immigrants in Germany with Chinese immigrants in the US. This implies that there is heterogeneity in the quantitative impact of personal and group characteristics. Moreover, there are factors that are specific to certain immigrant groups, including language, culture, and religion. These factors can impact the behavior of group members, e.g. including in relation to the types of occupations that may be viewed as appropriate, as well as influencing levels of discrimination and prejudice. There can also be interactions between migrant-specific influences (e.g. ethnic background) and/or personal characteristics (e.g. education) which may exacerbate the extent of unemployment, as will be highlighted in the following discussion.
Human capital
Human capital has a major influence on the chance that an individual is employed, which is true for the foreign and native born alike. Thus, unemployment rates vary considerably for immigrants depending on their skills and qualifications. This occurs not only because of the conventional link between education and the probability of employment, but also because human capital may not be transferable across national boundaries [7]. In particular, new migrants tend to be disadvantaged compared with native born residents with respect to country-specific skills (including poorer knowledge of the host country’s customs and being less likely to hold relevant occupational licenses) and information about labor market opportunities. These disadvantages should narrow the longer the migrant stays in the host country [7]. In addition to other country-specific skills, given that immigrants often move to countries that have a different main language to their own, fluency in the principal language of the host country is an extremely important determinant of labor market success. Interactions also exist between skill levels, language fluency, and residential location within a country. For instance, immigrants often cluster in large cities where there can be greater competition for jobs between low-skilled immigrants (as well as natives). Moreover, immigrants are prone to settling in areas with relatively high concentrations of compatriots (who generally speak the same language), which can influence the types of employment opportunities that are available to them [6]. For example, the deprived nature of many ethnic/immigrant enclaves may provide only limited job opportunities and access to relatively low-paying jobs and may also restrict the potential for job search outside of the enclave.
In addition, there is variation in the effect of human capital over the business cycle. In particular, there is evidence that the unemployment rates of low-skilled immigrant workers are affected to a higher degree by the state of the economy than is true for comparable native workers. This is because immigrants are more likely to lose their jobs during a recession. For example, the unemployment response of immigrants from non-OECD countries in Germany was 5.7 and 6.7 times higher for workers with intermediate and low levels of education, whereas the equivalent figures for the native born were 2.4 and 4.5 [5]. Similar magnitudes were found for the UK and a comparable situation is found for Mexican immigrants in the US through analyzing the impact of the national business cycle on unemployment for three skill groups [8]. The effects of the business cycle are largest for Mexican immigrants with low levels of education—i.e. having only a high school diploma and not having one at all. The unemployment levels experienced by Hispanic natives are also more responsive to the state of the economy than they are for non-Hispanic white natives across each of the educational categories. According to this study the differences within educational categories can at least be partly the result of direct or indirect discrimination by employers, as will be discussed further below. Moreover, immigrants who do not have legal status may be more affected by economic downturns, as employers may be more reluctant to hire this group in recessions [8].
Finally, from a general perspective, age plays an important role in human capital formation due to the accumulation of skills over time through formal training programs as well as by gaining on-the-job experience. As a result, the high unemployment rates experienced by some immigrant groups may in part be explained by their relatively young age profiles. Age upon arrival in the host country can also affect human capital accumulation since the earlier an individual arrives in a country the more time they will spend in the host country’s educational system and to become proficient in the host country’s main language [9]. In addition, the health of the economy on arrival can influence future unemployment since immigrants arriving during a recession have been found to experience higher levels of unemployment, although this may only be a relatively short-term effect [10].
Other factors
Individuals from ethnic minority and immigrant groups are known to have suffered from the discriminatory attitudes displayed by some employers. This is more of an issue for visible minorities, whose skin color or other traits are different from that of the majority population. For example, there is evidence from field studies indicating that employers are more prejudiced against certain ethnic minority and religious groups such as Muslims. This can be accompanied by discrimination of a more indirect nature at the hiring stage, such as employers failing to recognize the value of qualifications or experience that have been obtained in the country of origin [7].
It is sometimes difficult to separate out potentially discriminatory practices from cultural and religious differences. For example, norms and expectations that are prevalent amongst certain immigrant groups can influence the occupations that are chosen by some members of their communities and can result in relatively high rates of unemployment, such as those observed for Muslim women in many OECD countries. However, cultural and religious differences can change over time, and this may partly explain the lower rates of unemployment that some of the more established groups of immigrants are starting to experience because of expanding occupational choice. There is also evidence to indicate that culture can have an impact on unemployment. For example, immigrant women reporting a stronger German identity are less likely to be unemployed in Germany, although the magnitude of the effect is quite small and it is not observed for men [11].
It is also important to note some of the more dynamic aspects of differences in unemployment rates between immigrants and natives. In particular, immigrants could experience higher levels of unemployment stemming either from longer spells of unemployment or from shorter periods in work. Evidence from Germany suggests that unemployed immigrants tend to need more time than natives to find employment but, once they have found a job, male immigrants do not have less stable jobs. However, there is some heterogeneity between groups, especially with respect to first and second generation Turkish immigrants in Germany, who are found to be significantly less likely to move from unemployment into paid work [12]. The immigrants most likely to find employment and work in more stable jobs are typically those who originate from countries that are similar to the host country.
The evidence presented so far highlights the crucial role that human capital plays in determining the unemployment rates of immigrants. However, immigrants may also encounter discrimination in the labor market, which means that their human capital may not be rewarded as favorably as it is for natives, thereby reducing their employment prospects.
Limitations and gaps
This article has examined unemployment amongst immigrants in developed countries. Some of the discussion may also be relevant to developing countries, although, for all host countries, much depends on the countries from which immigrants originate. Furthermore, the types of employment opportunities that are available in developing countries will be quite different since there will typically be far fewer vacancies for highly skilled foreign born workers. Moreover, there are significant issues associated with measuring unemployment in developing countries, which is likely to make comparisons rather difficult.
In addition, the focus of this article has been on a single labor market outcome: unemployment. Although this is a very important indicator, immigrant differences in unemployment should also be viewed in conjunction with other variables, especially wages, where significant differences have also been found. This is because, in addition to employment status, the pay and quality of jobs is important (e.g. in determining poverty and inequality levels). Some groups, such as Muslim women in many OECD countries, also have low levels of labor market participation, which will further reduce employment rates.
Finally, migration is an evolving process that will affect not only the number of immigrants residing in a particular country, but also the socio-economic characteristics of these groups. Therefore, unemployment rates for certain groups can change fairly rapidly, particularly at different points in the economic cycle, and this will have an impact on recommendations for the appropriate policy responses.
Summary and policy advice
Although unemployment tends to be higher for immigrants than for the native born in most OECD countries, the extent to which this is the case varies considerably. There are also substantial differences in unemployment between immigrant groups within countries owing to a range of factors, including varying levels of education and labor market discrimination. Some clear policy messages emerge from this analysis, especially with respect to the importance of human capital. Increasing human capital formation, especially in relation to education and language ability, is vital for reducing unemployment. This particularly relates to raising the low levels of education possessed by some immigrant communities, as well as improving fluency in the host country’s main language. The importance of formal education is likely to continue to rise as employment in advanced economies becomes ever more concentrated in service-based industries. Moreover, beyond simply encouraging children from immigrant backgrounds to achieve high levels of education , attempts should also be made to further increase educational levels amongst adults. In this regard, it is particularly important to target individuals who would benefit most from adult educational and training programs, such as groups (e.g. Muslim women) that traditionally have high unemployment rates. Similarly, given the importance of proficiency in the host country’s main language for reducing unemployment, the provision of adult language courses for recent immigrants can be effective in enhancing employability [13]. However, the efficiency of such courses could be improved if these were tailored to meet the requirements of specific groups. For example, certain courses could be designed in ways that recognize the traditional family roles of some immigrant women by scheduling classes at convenient times, coordinating with childcare provision, and offering women-only classes [13].
A range of other factors also contribute to the higher levels of unemployment experienced by some immigrant groups and these are affected by ethnic/immigrant and host country-specific issues. For example, despite achieving higher levels of education, children from some immigrant groups, such as Asians in some OECD countries, are still likely to encounter obstacles to their progression in the labor market, including as a result of direct discrimination by employers. Therefore, organizations should be encouraged to monitor their workforce’s demographics and to consider proactive measures if certain groups are found to be under-represented.
Other policy measures could be aimed at improving the employability of groups experiencing the highest unemployment rates, especially by making job search more effective. Examples include counseling individuals about the importance of, and access to, social networks for obtaining jobs. The effectiveness of such policies is likely to be enhanced by the availability of appropriate childcare provision [13]. Furthermore, given that there is often a lack of recognition in the host country of the skills and qualifications that immigrants have obtained overseas [7], policies could be designed to assist with increasing the rewards for immigrants’ existing human capital. The introduction of programs to help improve the utilization of these skills in a different country should enhance employability and matching within the labor market.
Acknowledgments
The author thanks three anonymous referees and the IZA World of Labor editors for many helpful suggestions on earlier drafts. The author would like to thank the colleagues that he has collaborated with on examining ethnic and immigrant differences in labor market outcomes, especially David Blackaby, Ken Clark, Derek Leslie, Phil Murphy, and Nigel O’Leary.
Competing interests
The IZA World of Labor project is committed to the IZA Guiding Principles of Research Integrity. The author declares to have observed these principles.
© Stephen Drinkwater