Chapter 5. Foreign-born workers and their labour market outcomes

This chapter focuses on how foreign-born workers fare in the host-country labour market. It focuses on earnings, occupational status and the extent to which foreign-born workers’ skills are used in the workplace. It also discusses the factors that could affect these outcomes, including immigrants’ proficiency in literacy, numeracy and the host-country language, their country of origin, and where they acquired their education.

    

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

The capacity to gauge and exploit written material (literacy), and the ability to process and treat mathematical information (numeracy), is essential in order to take part in today’s knowledge-based societies. While people’s competencies must be considered when examining how workers perform in the labour market, the lack of reliable data on workers’ skills has made it impossible to do so directly. Until recently, educational attainment was almost the sole measure of human capital.

But measuring human capital by educational attainment only can be misleading. This is especially true for foreign-born workers who, because they have probably been exposed to disparate education systems in their home countries, offer a different set of skills than their native-born peers for a given level of education. In fact, not all education systems are equal: there are significant differences across countries in the quality of education provided. Obtaining a primary education in one region does not necessarily imply holding the same set of literacy and numeracy skills as someone who obtained a primary education in another region. Indeed, spending the same numbers of years in education does not imply the same human capital or skills across countries.

In contrast to much of the previous economic literature on the subject, this chapter includes detailed individual-level information on language, literacy and numeracy proficiency in its analyses. Moreover, the chapter measures the labour market performance of workers not only in terms of earnings, but also regarding workers’ occupation. The socio-economic status of workers’ occupation is an important aspect of analysing the labour market performance of foreign-born workers. Given the global concern about the correct allocation of human capital, it is crucial to know the kind of jobs immigrant workers obtain in their destination country. Indeed, the jobs workers do have considerable consequences for workers’ well-being, work-life balance and job satisfaction (Rose, 2003[1]).

The Survey of Adult Skills (PIAAC) allows for determining not only whether foreign-born workers are paid salaries commensurate with their human capital, but also in which occupation they work in their host country and to what extent their skills are used in their jobs. This analysis is particularly important as it aims to identify the obstacles that impede foreign-born adults from obtaining better jobs.

This chapter shows that the different returns to education on earnings between foreign-born and native-born adults are not directly due to the poor quality of foreign-born workers’ education or the non-transferability of the skills they learned back home, but rather to differences in their occupations in the destination country. In turn, the kinds of jobs foreign-born workers obtain partially reflect their skills: accounting for workers’ literacy and numeracy proficiency reduces the observed differences in status in the occupations of foreign-born and native-born adults. Yet, even after accounting for such skills, foreign-born workers with foreign qualifications still show somewhat lower occupational status than native-born workers. In some countries, foreign-born workers reported that they feel that their skills are underutilised.

Description of the data

Reliable cross-country surveys capturing skills heterogeneity among adult workers are scarce, particularly those covering foreign-born workers.1 Given its clear advantages compared with other datasets, the Survey of Adult Skills (PIAAC) is attracting increasing attention from labour economists (Levels, Van der Velden and Allen, 2014[2]; Nieto and Ramos Lobo, 2014[3]; Hanushek et al., 2015[4]; Montt, 2017[5]).

But the survey does have its limitations, especially with respect to research on immigrant workers (see Bonfanti and Xenogiani (2014[6]), for more details). For example, once the immigrant population is decomposed by its own characteristics – for example, region of origin or educational attainment – the samples could be small. Hence, in order to reduce measurement error, in this report all econometric results based on less than 30 observations are reported as missing (see Perry et al. (2014[7]) for a similar approach). Also, following Bonfanti and Xenogiani (2014[6]), countries where the immigrant population is less than 2.5% of the total population – namely Japan, Korea, Poland the Russian Federation and the Slovak Republic – are excluded from the analysis. Chile and the Czech Republic must also be excluded as the disaggregation of their immigrant population by country of education or by region of origin leads to very small samples. Information on country of birth and spoken languages is missing for Australia and Germany, which are therefore dropped from consideration, as such variables are central for the analysis. Thus the results discussed in this chapter cover the following 22 countries: Austria, Belgium, Canada, Cyprus2,3, Denmark, Estonia, Finland, France, Greece, Ireland, Israel, Italy, Lithuania, the Netherlands, New Zealand, Norway, Singapore, Slovenia, Spain, Sweden, the United Kingdom and the United States.4

The returns to education on earnings of foreign-born and native-born adults

A long history of debate

The relative position of foreign-born workers in the earnings distribution of a country defines the extent to which those workers contribute to the host-country economy. If foreign-born adults earn higher wages, they contribute more to tax and benefit systems, thereby raising overall aggregate income (Dustmann and Glitz, 2011[8]). In addition, the earnings of foreign-born workers are an important indicator of their integration in the host economies, and they could have considerable impact on their compatriots’ decisions on whether or not to emigrate themselves.

The seminal works by Chiswick (1978[9]), Borjas (1985[10]), and LaLonde and Topel (1991[11]) fuelled the debate on the earning patterns of foreign-born workers and their “quality” – i.e. schooling. In addition to educational attainment, researchers have shown that the earnings differential between native-born and foreign-born adults is linked to a variety of factors. For instance, due to credit constraints, foreign-born workers are likely to accept any available job at the beginning of their stay in the host country, thereby earning lower wages than what their educational attainment would predict (Eckstein and Weiss, 2004[12]). And not only are foreign-born workers less likely to use all of their skills upon arrival, but the skills acquired by native-born adults are likely more easily adaptable to technological changes in their own country (Lam and Liu, 2002[13]).

Overall, the existing evidence suggests that education and experience obtained in countries of origin are not fully valued in destination countries, thereby resulting in apparently well-qualified foreign-born adults holding low-paying and low-quality jobs. There is also the possibility that firms in destination countries discriminate against foreign-born workers, so that they pay these workers less than they pay native-born workers with similar skills.5 In order to disentangle the two factors, direct measures of skills are needed.6

Few studies have looked at the skills of foreign-born adults and these workers’ labour market performance across countries. A notable example is Clarke and Skuterud (2016[14]), who exploit the Adult Literacy and Lifeskills (ALL) dataset to compare literacy test scores and their impact on wages and employment of foreign-born workers in Australia, Canada and the United States. Overall, they find no convincing evidence that the labour market returns to literacy skills of migrant workers with a foreign mother tongue in Australia and Canada are statistically different from those of their native-born counterparts. In contrast, they find greater returns to literacy for foreign-born workers in the United States whose mother tongue is different from English and Spanish. This result reflects complementarities between language and skills. Jerrim (2015[15]) uses PIAAC information on British nationals working abroad to study their labour market outcomes, including prolonged periods out of work, earnings and overqualification. He finds that British emigrants earn more than workers who stay in the United Kingdom. But this is largely due to longer working hours abroad compared to that in Britain, rather than to a different skills set and educational attainment.

This chapter uses PIAAC data to examine the returns to education among foreign-born and native-born adults in 22 countries, and how those returns are related to language, literacy and numeracy skills. This analysis considers the country in which foreign-born adults earned their highest qualification, as those who completed their studies in the host country might realise returns that are more similar to native-born adults than foreign-born workers who earned their highest degree in their home countries (Schaafsma and Sweetman, 2001[16]; Bratsberg and Ragan Jr, 2002[17]). A qualification earned in the host country is more likely to be valued by local employers. In addition, foreign-born adults educated in the host country might have stronger social networks that can help them find better jobs, and might have greater proficiency in the local language and better knowledge of the social norms pervasive in the socio-economic landscape of the country. Thus the analysis uses the age at arrival in the host country along with the age at which the highest educational qualification was earned to distinguish between those foreign-born workers who were educated in the host country and those who were educated elsewhere.

A snapshot of foreign-born workers’ earnings

In spite of their overall high educational attainment (see Chapter 2), foreign-born adults earn significantly less than their native-born peers. Figure 5.1 looks at median hourly earnings, including bonuses (in PPP-corrected USD). Across almost all countries, native-born workers earn more than foreign-born workers. In the sample studied, the hourly wage earned by native-born adults is 5% higher, on average, than that earned by foreign-born workers who were educated in the host country, and 16% higher than the hourly wage earned by foreign-born workers who were educated elsewhere. For instance, in the United States, foreign-born adults who were educated elsewhere earn only 72% of the hourly wage of native-born adults (a finding consistent with the previous literature; see (OECD, 2008[18])). In Spain, workers in the former group earn 71%, and in Lithuania they earn 65% of the hourly wages of native-born adults in the respective countries.

Figure 5.1. Median hourly earnings
picture

1. Note by Turkey: The information in this document with reference to “Cyprus” relates to the southern part of the Island. There is no single authority representing both Turkish and Greek Cypriot people on the Island. Turkey recognises the Turkish Republic of Northern Cyprus (TRNC). Until a lasting and equitable solution is found within the context of the United Nations, Turkey shall preserve its position concerning the “Cyprus issue”.

2. Note by all the European Union Member States of the OECD and the European Union: The Republic of Cyprus is recognised by all members of the United Nations with the exception of Turkey. The information in this document relates to the area under the effective control of the Government of the Republic of Cyprus.

Source: (OECD, 2015[19]) Survey of Adult Skills (PIAAC) (2012, 2015), Table A6.1, www.oecd.org/skills/piaac/publicdataandanalysis

 StatLink http://dx.doi.org/10.1787/888933846080

Why do foreign-born adults earn less than their native-born peers? As mentioned above, an important factor that has been overlooked in earlier studies is the heterogeneity of the skills set held by workers, especially foreign-born workers. This chapter analyses measures of workers’ numeracy and literacy proficiency.7 Respondents’ skills levels are measured on a scale of 0 to 500. The left (right) panel of Figure 5.2 presents different percentiles of the distribution of literacy (numeracy) test scores by country of origin. In contrast with foreign-born adults’ educational attainment, their proficiency in literacy and numeracy is lower than that of natives across the entire distribution.

Figure 5.2. Distribution of literacy and numeracy test scores, by percentile
picture

Source: (OECD, 2015[19]) Survey of Adult Skills (PIAAC) (2012, 2015), Table A6.1, www.oecd.org/skills/piaac/publicdataandanalysis

 StatLink http://dx.doi.org/10.1787/888933846099

Differences in returns to education

Looking only at the correlation between hourly earnings, educational attainment and assessed skills is insufficient. Differences in age patterns of the foreign-born and native-born populations, as well as language fluency and other individual characteristics might also explain disparities in returns to education. This chapter considers the interactions between the migration variable (which can take three values: native-born, foreign-born educated in the host country, and foreign-born educated elsewhere) and the educational level (which can also take three values: primary, secondary and post-secondary schooling).

Table 5.A.1 (see Annex A) presents the estimation of these interactions for each PIAAC-participating country with available information.8 Language and skills are not included in this analysis. Controls include both immigration background (native-born, foreign-born educated in the host country, and foreign-born educated elsewhere) and education (primary, secondary and post-secondary schooling) variables, age, gender, marital status, number of children, and dummies for having two or more jobs, for public employment, for having an indefinite job contract, and for industries.9

In most countries, the returns to education are not statistically different between native-born and foreign-born adults. However, the returns to upper secondary education among foreign-born workers educated in the host country are lower in Cyprus10, Greece and New Zealand, while those of foreign-born workers who were educated elsewhere are lower in Singapore and Spain. The results are similar when considering tertiary education. In Belgium, Cyprus11, the Netherlands and New Zealand, these results are associated with significantly lower wages for foreign-born workers educated in the host country and, in France and Spain, for foreign-born workers educated elsewhere.12

So which factors are associated with foreign-born workers earning smaller returns to education than natives? If schooling in the country of origin provided foreign-born workers with fewer or poorer-quality skills, then the returns to their educational attainment in the host country might be lower. This is not only the case among foreign-born adults who were educated entirely elsewhere, but also for those foreign-born workers who earned their highest qualification in the host country. In order to disentangle these factors, the analysis in Table 5.A.2 (see Annex A) includes controls for workers’ proficiency in numeracy and literacy.13 These variables at least partially represent the quality of the education provided. If relevant, they should reduce the difference in the returns to education between native-born and foreign-born workers, particularly those who were educated outside the host country. In addition, as language skills have been proved to be important determinants of the labour market performance of foreign-born workers (McManus, Gould and Welch, 1983[20]; Dustmann, 1994[21]; Bleakley and Chin, 2004[22]), Table 5.A.2 also includes a dummy variable for whether the language of the survey is the same as the respondent’s first, second or most often spoken language.14

Results in Table 5.A.2 suggest that skills and proficiency in the host-country language do not fully explain the lower returns to education among foreign-born workers compared to native-born workers.15 In fact, the returns to education for foreign-born adults remain approximately the same – both in magnitude and statistical significance – as those shown in Table 5.A.1. Therefore, the hypothesis that the different returns to schooling between native and foreign-born adults are wholly due to the quality of the education provided or to differences in language proficiency can be at least partially rejected.

An alternative explanation of the disparities in the impact of education on earnings lies in workers’ occupations. In some countries, foreign-born workers might be segregated into occupations where the returns to education are particularly low, while native-born adults may be working in more rewarding jobs at the opposite end of the salary spectrum.16 Results shown in Table 5.A.3 (see Annex A) confirm that, in most countries, controlling for occupation eliminates the statistical difference between the returns to education among natives and the returns to education among foreign-born workers, implying that foreign-born adults are indeed more likely to work in occupations with lower returns. This is particularly the case in Austria, France, Greece, Singapore and the United States. In contrast, two countries still show lower returns to both upper secondary- and tertiary-educated foreign-born workers: New Zealand, among foreign-born workers educated in New Zealand, and in Spain, among foreign-born workers educated elsewhere. These results might reflect discrimination in the labour market (as suggested by Solé and Parella (2003[23]), for Spain, and Harris et al. (2006[24]) for New Zealand), or other factors, such as lack of established social networks, migrants’ lack of self-confidence during the hiring process, or non-recognition of foreign education qualifications.

Workers’ skills and occupation

Occupations held by foreign-born workers

A look at the distribution of foreign-born workers across occupations shows that they are over-represented in low-skilled jobs (Figure 5.3). In spite of their educational attainment, foreign-born adults who were educated outside the host country are more than twice as likely as native-born adults to be employed in low-skilled occupations. In several countries, namely Belgium, Estonia and Finland, this pattern is also seen among foreign-born workers who were educated in the host country.

Figure 5.3. Share of workers in low-skilled occupations
picture

Note: Low-skilled occupations are defined as those occupations under the ISCO code category 9, i.e. “elementary occupations”.

1. See notes 1, 2 in Figure 5.1

Source: (OECD, 2015[19]) Survey of Adult Skills (PIAAC) (2012, 2015), Table A6.1, www.oecd.org/skills/piaac/publicdataandanalysis

 StatLink http://dx.doi.org/10.1787/888933846118

This chapter uses the International Socio-Economic Index (ISEI) of occupational status developed by Ganzeboom et al. (1992[25]), recently updated by Ganzeboom (2010[26]), when considering the socio-economic status of occupations. The ISEI index is based on the idea that occupation is the activity that links education and income. As such, the index is constructed to maximise the indirect influence of education and skills on income. The index has been widely used not only in the sociological literature, but also in migration economics research (Euwals et al., 2010[27]; Dustmann and Frattini, 2013[28]; Zorlu, 2013[29]; Postepska and Vella, 2017[30]).17

Figure 5.4 shows the median ISEI score by worker’s background for each of the PIAAC-participating countries. Overall, native-born adults are employed in jobs ranked higher on the occupational status scale than those in which foreign-born adults are employed. This difference is particularly large for foreign-born workers with foreign education qualifications. On average, their ISEI score is 32 – a score representing such occupations as domestic housekeeper or stock clerk – compared to the average score of native workers (44), and the score of foreign-born workers educated in the host country (41) – scores that generally represent such occupations as office clerk. A notable exception is observed in North America, where foreign-born adults educated in the host country are employed in slightly better-ranked jobs than natives, and in New Zealand and Singapore, where the median score among natives is lower than that among foreign-born adults.

Figure 5.4. Median occupational status score
picture

Note: The ISEI is a continuous index expressed in a 10-90 metric.

1. See notes 1, 2 in Figure 5.1

Source: (OECD, 2015[19]) Survey of Adult Skills (PIAAC) (2012, 2015), Table A6.1, www.oecd.org/skills/piaac/publicdataandanalysis

 StatLink http://dx.doi.org/10.1787/888933846137

The relationship among migrant background, occupation and skills

In order to determine whether the occupations of foreign-born adults in the host labour market are of lower status than those of native-born adults, even after taking into consideration differences in individual characteristics, the analysis estimates, for each PIAAC-participating country, the relationship between the ISEI index and the regions of origin and education of those workers. Seven different categories of origin are created: (1) natives; (2) immigrants from EU countries educated in the host country; (3) immigrants from EU countries educated elsewhere; (4) immigrants from European non-EU countries educated in the host country; (5) immigrants from European non-EU countries educated elsewhere; (6) immigrants from outside Europe educated in the host country; and (7) immigrants from outside Europe educated elsewhere.

While the results in Table 5.A.4 (see Annex A) indicate that each country has its own pattern, common trends can be observed. For instance, immigrants from European countries (whether those countries are part of the European Union or not) educated in the host country mostly hold jobs of the same status as natives. This is also observed among immigrants coming from other continents and educated in the host countries, although Denmark, France, Israel, Sweden and the United Kingdom are noteworthy exceptions.

The situation is reversed when looking at immigrants with foreign education qualifications. Regardless of their region of origin, members of this group hold jobs with significantly lower occupational status than natives. However, in Spain, neither region of origin nor country of education has an impact on the status of foreign-born workers’ occupation. In both Austria and Ireland, immigrants (in Austria, those from non-EU countries; in Ireland, those from EU countries) with foreign education qualifications are employed in jobs with lower status than their native-born peers.

Table 5.A.5 (see Annex A) shows results after accounting for language and skills. These results not only confirm that immigrants educated in the host country tend to work in occupations whose status is similar to those in which native-born adults work, they also show that in certain countries, namely Canada, Spain and the United States, immigrants hold jobs with higher status. Once immigrants’ proficiency in literacy and numeracy are taken into account, differences between immigrants and natives in the status of their occupations shrink. For instance, in the Netherlands, New Zealand, Slovenia and the United Kingdom, these differences are fully explained by workers’ skills.

Nonetheless, there are still factors that can disadvantage foreign-born workers, especially those who were not educated in the host country. One commonly cited is that immigrants’ education and work experience in their country of origin are not recognised in the host country (Dustmann and Frattini, 2013[28]). However, in Denmark, Ireland, Italy and Sweden, immigrants from other EU countries who have foreign qualifications are still found in lower-status occupations – even though EU regulations mandate the recognition of diplomas across member states. Other factors, such as attitudes towards immigrants or the lack of social networks among immigrants, lie at the heart of such findings.

To gain better insights into how immigrants are welcomed – or not – into the labour market, this analysis also seeks to determine whether the status of the occupations in which immigrants are employed is related to the immigrants’ country of origin. Occupations are divided into four groups according to their ISEI score. The first quartile includes jobs of the lowest occupational quality; the fourth quartile contains the most prestigious jobs. The comparisons involve the same seven categories of immigrant background listed above. By looking at workers with average skills and average individual characteristics (age, gender, education, etc.), it is possible to compute the probability that each of the seven hypothetical individuals gets a job in an ISEI quartile.

For the sake of simplicity, results are presented for the two extreme cases: the probability of low-educated workers landing in occupations within the first (low quality) ISEI quartile and the probability of highly educated workers having a job in the fourth (high quality) ISEI quartile. Figure 5.5 presents results for the pooled sample of PIAAC countries. As expected, workers with lower secondary as their highest level of educational attainment are mostly employed in jobs in the first ISEI quartile. And in line with the previous findings, immigrants educated in the host country have a probability similar to that of native-born adults of working in a certain occupation. But compared to a 54% chance that native-born adults are employed in an occupation in the lowest ISEI quartile, the likelihood for immigrants with a foreign education to be so employed is 67% for those from EU countries, 86% for those from non-EU European countries, and 63% for those from elsewhere. The only statistically significant differences are those between native-born adults and immigrants from European (whether EU or non-EU) countries with foreign education qualifications.

The case of tertiary graduates is particularly interesting, since both the media and previous academic literature have often focused on the so-called “brain waste” of highly educated immigrants. Looking at the probabilities of working in an occupation within the fourth (highest) ISEI quartile, there is no statistically significant difference between native-born and foreign-born adults. This implies that skills matter: all else being equal, if tertiary-educated workers all have the same levels of language, literacy and numeracy proficiency, then the country of origin of immigrant workers would not significantly affect the likelihood of their working in the best jobs.

Figure 5.5. Probability of working in a particular occupation
picture

Note: Specifications control for language, numeracy and proficiency, literacy skills proficiency, country of interview dummies, age, age squared, age cubed, gender, marital status, number of children, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies.

Source: (OECD, 2015[19]) Survey of Adult Skills (PIAAC) (2012, 2015), Table A6.1, www.oecd.org/skills/piaac/publicdataandanalysis

 StatLink http://dx.doi.org/10.1787/888933846156

Subjective skills mismatch and foreign-born workers

Do immigrants, themselves, perceive that their full potential is not exploited in host labour markets? Examining this question can shed additional light on the obstacles that foreign-born workers face in their host countries. There are several reasons why an analysis of skills mismatch is relevant. At the individual level, skills mismatch affects workers’ earnings and job satisfaction; at the firm level, it hinders productivity and increases turnover; at the macroeconomic level, it reduces income growth due to the drop in productivity and the loss of human capital (Quintini, 2011[31]). Skills mismatch should not be confused with qualifications mismatch. In fact, it is perfectly possible that the two do not coincide. For example, a tertiary-educated worker might hold a job requiring only secondary education, but might lack some of the skills required to be hired in a graduate position: the person is overqualified but not overskilled for his job.18

The Survey of Adult Skills (PIAAC) contains two questions that can be exploited to measure self-assessed skills mismatch: “Do you feel that you have the skills to cope with more demanding duties than those you are required to perform in your current job?” and “Do you feel that you need further training in order to cope well with your present duties?”. The first question assesses skills underutilisation; the second measures skills deficits.19 If the results discussed above hold, it is expected that foreign-born workers will have higher rates of skills underuse and lower rates of skills deficits.

The share of respondents who perceive their skills to be underutilised is presented in Figure 5.6. The picture is mixed. In certain countries, immigrants are more likely than native-born adults to believe that they have the skills to cope with more challenging tasks. For instance, in Sweden, immigrants educated in that country are 4 percentage points more likely, and immigrants with foreign qualifications are 8 percentage points more likely than native-born adults to feel overskilled. Such patterns might be due to the composition of the Swedish immigrant population, which includes numerous refugees. There is a similar situation in Denmark; and in Singapore, immigrants are 4 percentage points more likely to feel overskilled, regardless of their country of education.

Nonetheless, there are still cases where immigrants are much less likely to report that their skills are underutilised, namely in Estonia, Israel and the Netherlands (although Israel is an outlier in most of the analysis, due to its singular immigrant composition). There are also cases where immigrants educated in the host country show a similar pattern as natives, while immigrants with a foreign education are less likely to perceive that their skills are underutilised – e.g. Austria, Slovenia and the United States. Finally, in Canada, New Zealand and the United Kingdom, there appears to be no significant difference in the proportion of self-perceived skill underutilisation between the native- and the foreign-born.

Figure 5.6. Share of workers who feel that their skills are underutilised
picture

Note: See notes 1, 2 in Figure 5.1

Source: (OECD, 2015[19]) Survey of Adult Skills (PIAAC) (2012, 2015), Table A6.1, www.oecd.org/skills/piaac/publicdataandanalysis

 StatLink http://dx.doi.org/10.1787/888933846175

Figure 5.7 shows the share of workers who feel the need for further training in order to cope well with the duties of their current job. In Greece, Lithuania and Spain, foreign-born workers are less likely than their native-born peers to report skills deficits, regardless of the country in which they completed their education. In Austria, Belgium, Israel and Slovenia, immigrants educated in the host country and native-born workers reported similarly, while foreign-born workers who were educated elsewhere were less likely to report that they need additional training. For the remaining countries, the relationship is more mixed, and may well hide large differences across countries of origin. A rigorous regression analysis is thus needed in order to understand whether immigrants’ perception of skills mismatch is really greater than that of their native peers once individual characteristics have been accounted for.

Figure 5.7. Share of workers who feel they need further training
picture

Note: See notes 1, 2 in Figure 5.1

Source: (OECD, 2015[19]) Survey of Adult Skills (PIAAC) (2012, 2015), Table A6.1, www.oecd.org/skills/piaac/publicdataandanalysis

 StatLink http://dx.doi.org/10.1787/888933846194

Table 5.A.6 (see Annex A) illustrates the correlation between immigrant background by region of origin and country of education, and self-assessed skills underuse. After accounting for language, numeracy and literacy skills (as well as for the usual set of individual characteristics, such as age, education and gender), immigrants and natives show no statistical difference in their responses. Only in a few countries, and particularly among immigrants who arrived from outside the European Union, the likelihood of perceiving that they have the skills to cope with more demanding duties is greater for immigrants. This is especially the case in Canada, Denmark, Italy, New Zealand and Sweden. The results are similar if the analysis does not control for language, literacy and numeracy proficiency.

Self-perceived skills deficits are examined in Table 5.A.7 (see Annex A). Results are presented after accounting for language, numeracy and literacy proficiency.20 Again, one out of two countries in the sample, namely Austria, Estonia, Finland, France, Greece, Italy, Lithuania, the Netherlands, Norway, Singapore and Slovenia, shows no statistically significant difference between native-born and foreign-born workers. By contrast, in Belgium, Denmark, Israel and Spain, immigrants are less likely than natives to believe that they need additional training, confirming that foreign-born workers have still-untapped potential. In the remaining countries, namely Canada, New Zealand, Sweden, the United Kingdom and the United States, immigrants, especially those from outside Europe, were more likely than native-born workers to report that they require further training.

Policy implications

The results of this chapter stress the key role played by language proficiency and literacy and numeracy skills as drivers of the labour market performance of foreign-born adults. The findings also suggest that being educated in the host country can go a long way towards helping foreign-born adults gain access to jobs of similar status as those held by their native-born peers. Hence, policies that attract students should be adopted or strengthened, as should programmes that allow international students to remain in the host country for longer after graduation. For those immigrants who arrive as adults (either through labour of family migration programmes), participation in language training and some form of short-cycle qualification programme can give prospective employers a signal about the skills that these adults can offer to the labour market.

Moreover, the findings of the chapter challenge the notion of immigrants’ brain waste. The lower earnings and lower-status occupations of foreign-born workers are largely explained by these workers’ actual set of skills, although other factors, such as discriminatory practices in the labour market and a lack of information, still work against certain categories of immigrants in some host countries. Further research is needed in order to better understand the forces behind immigrants’ poor labour market performance.

References

[22] Bleakley, H. and A. Chin (2004), “Language skills and earnings: Evidence from childhood immigrants”, Review of Economics and Statistics, Vol. 86/2, pp. 481-496.

[6] Bonfanti, S. and T. Xenogiani (2014), “Migrants' skills: Use, mismatch and labour market outcomes – A first exploration of the International Survey of Adult Skills (PIAAC)”, in Matching Economic Migration with Labour Market Needs, OECD Publishing, Paris.

[10] Borjas, G. (1985), “Assimilation, changes in cohort quality, and the earnings of immigrants”, Journal of Labour Economics, Vol. 3/4, pp. 463-489.

[17] Bratsberg, B. and J. Ragan Jr (2002), “The impact of host-country schooling on earnings: A study of male immigrants in the United States”, Journal of Human Resources, Vol. 37/1, pp. 63-105.

[9] Chiswick, B. (1978), “The effect of Americanization on the earnings of foreign-born men”, Journal of Political Economy, Vol. 86/5, pp. 897-921.

[14] Clarke, A. and M. Skuterud (2016), “A comparative analysis of immigrant skills and their utilization in Australia, Canada, and the USA”, Journal of Population Economics, Vol. 29/3, pp. 849-882.

[21] Dustmann, C. (1994), “Speaking fluency, writing fluency and earnings of migrants”, Journal of Population Economics, Vol. 7/2, pp. 133-156.

[28] Dustmann, C. and T. Frattini (2013), “Immigration: The European experience”, in Card, D. and S. Raphael (eds.), Immigration, Poverty, and Socioeconomic Inequality, Russell Sage Foundation, New York.

[8] Dustmann, C. and A. Glitz (2011), “Migration and education”, in Hanushek, A. (ed.), Handbook of the Economics of Education, North-Holland, Amsterdam.

[12] Eckstein, Z. and Y. Weiss (2004), “On the wage growth of immigrants: Israel, 1990–2000”, Journal of the European Economic Association, Vol. 2/4, pp. 665-695.

[27] Euwals, R. et al. (2010), “Citizenship and labour market position: Turkish immigrants in Germany and the Netherlands”, International Migration Review, Vol. 44/3, pp. 513-538.

[26] Ganzeboom, H. (2010), “A new international socio-economic index [ISEI] of occupational status for the international standard”, Paper presented at the Annual Conference of International Social Survey Programme, Lisbon.

[25] Ganzeboom, H., P. De Graaf and D. Treiman (1992), “A standard international socio-economic index of occupational status”, Social Science Research, Vol. 21/1, pp. 1-56.

[4] Hanushek, E. et al. (2015), “Returns to skills around the world: Evidence from PIAAC”, European Economic Review, Vol. 73, pp. 103-130.

[24] Harris, R. et al. (2006), “Racism and health: The relationship between experience of racial discrimination and health in New Zealand”, Social Science & Medicine, Vol. 63/6, pp. 1426-1441.

[15] Jerrim, J. (2015), “Emigrants from Great Britain: What do we know about their lives?”, Department of Quantitative Social Science Working Paper , No. 15-02, University College London, London.

[11] LaLonde, R. and R. Topel (1991), “Immigrants in the American labour market: Quality, assimilation, and distributional effects”, American Economic Review Papers & Proceedings, Vol. 81/2, pp. 297-302.

[13] Lam, K. and P. Liu (2002), “Earnings divergence of immigrants”, Journal of Labour Economics, Vol. 20/1, pp. 86-104.

[2] Levels, M., R. Van der Velden and J. Allen (2014), “Educational mismatches and skills: New empirical tests of old hypotheses”, Oxford Economic Papers, Vol. 66/4, pp. 959-982.

[20] McManus, W., W. Gould and F. Welch (1983), “Earnings of Hispanic men: The role of English language proficiency”, Journal of Labour Economics, Vol. 1/2, pp. 101-130.

[5] Montt, G. (2017), “Field-of-study mismatch and overqualification: labour market correlates and their wage penalty”, IZA Journal of Labour Economics, Vol. 6/1, pp. 1-20.

[3] Nieto, S. and R. Ramos Lobo (2014), “Overeducation, skills and wage penalty: Evidence for Spain using PIAAC data”, Social Indicators Research, Vol. 134/1, pp. 219-236.

[19] OECD (2015), Survey of Adult Skills (PIAAC) (database), http://www.oecd.org/skills/piaac/publicdataandanalysis.

[18] OECD (2008), International Migration Outlook 2008, OECD Publishing, Paris, http://dx.doi.org/10.1787/migr_outlook-2008-en.

[7] Perry, A., S. Wiederhold and D. Ackermann-Piek (2014), “How can skill mismatch be measured? New approaches with PIAAC”, Methods, Data, Analyses, Vol. 8/2, pp. 137-174.

[30] Postepska, A. and F. Vella (2017), “Persistent occupational hierarchies among immigrant worker groups in the United States labour market”, IZA Discussion Paper , No. 10514, Institute for the Study of Labour (IZA), Bonn.

[31] Quintini, G. (2011), “Over-qualified or under-skilled: A review of existing literature”, OECD Social, Employment, and Migration Working Papers, No. 121, OECD, Paris.

[1] Rose, M. (2003), “Good deal, bad deal? Job satisfaction in occupations”, Work, Employment and Society, Vol. 17/3, pp. 503-530.

[16] Schaafsma, J. and A. Sweetman (2001), “Immigrant earnings: Age at immigration matters”, Canadian Journal of Economics, Vol. 34/4, pp. 1066-1099.

[23] Solé, C. and S. Parella (2003), “The labour market and racial discrimination in Spain”, Ethnic and Migration Studies, Vol. 29/1, p. 121.

[29] Zorlu, A. (2013), “Occupational adjustment of immigrants in the Netherlands”, Journal of International Migration and Integration, Vol. 14/4, pp. 711-731.

Annex 5.A. Tables
Annex Table 5.A.1. The impact of educational attainment and migration on hourly wages (without language and skills controls)

Medium education

High education

Migrants educated at destination

Migrants educated elsewhere

Migrants educated at destination

Migrants educated elsewhere

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Austria

-0.124

0.088

 

0.007

0.062

 

-0.281

0.122

*

0.028

0.095

 

Belgium

0.043

0.081

 

0.040

0.088

 

 

 

 

0.037

0.106

 

Canada

-0.062

0.084

 

-0.022

0.060

 

-0.030

0.078

 

-0.076

0.055

 

Cyprus (1,2)

-0.615

0.218

**

-0.050

0.154

 

-0.559

0.210

**

-0.259

0.163

 

Denmark

0.069

0.153

 

0.011

0.063

 

-0.085

0.107

 

0.009

0.061

 

Estonia

0.122

0.131

 

0.082

0.112

 

-0.142

0.120

 

-0.077

0.117

 

Finland

0.059

0.077

 

 

 

 

 

 

 

 

 

 

France

0.105

0.063

 

-0.027

0.094

 

0.079

0.063

 

-0.175

0.079

*

Greece

-0.300

0.111

**

 

 

 

 

 

 

 

 

 

Ireland

0.020

0.126

 

0.045

0.140

 

0.157

0.127

 

0.133

0.151

 

Israel

0.281

0.101

**

0.468

0.211

*

0.334

0.096

***

0.122

0.178

 

Italy

 

 

 

-0.148

0.083

 

 

 

 

 

 

 

Lithuania

-0.014

0.219

 

 

 

 

 

 

 

 

 

 

Netherlands

-0.132

0.114

 

-0.046

0.162

 

-0.215

0.099

*

 

 

 

New Zealand

-0.222

0.087

*

-0.092

0.107

 

-0.258

0.065

***

-0.155

0.097

 

Norway

0.181

0.095

 

-0.034

0.062

 

0.164

0.089

 

-0.065

0.114

 

Singapore

-0.064

0.138

 

-0.215

0.103

*

-0.106

0.109

 

-0.120

0.083

 

Slovenia

0.142

0.170

 

-0.098

0.065

 

 

 

 

 

 

 

Spain

 

 

 

-0.227

0.092

*

 

 

 

-0.251

0.118

*

Sweden

-0.060

0.090

 

-0.013

0.058

 

-0.046

0.082

 

0.093

0.071

 

United Kingdom

0.025

0.135

 

0.036

0.107

 

-0.037

0.132

 

-0.083

0.114

 

United States

0.247

0.136

 

-0.213

0.343

 

0.292

0.140

*

-0.181

0.323

 

Notes: (1) ***, **, and * represent 1%, 5% and 10% significance levels, respectively. (2) The dependent variable is the logarithm of hourly wages, including bonuses, in PPP-corrected USD. Specifications control for age, age squared, age cubic, gender, marital status, number of children, education, migrant status, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies. (3) All specifications are weighted by the sampling weights provided in the dataset.

Note: See notes 1, 2 in Figure 5.1

Source: Survey of Adult Skills (PIAAC) (2012, 2015).

 StatLink http://dx.doi.org/10.1787/888933846612

Annex Table 5.A.2. The impact of educational attainment and migration on hourly wages (with language and skills controls)

Medium Education

High Education

Migrants educated at destination

Migrants educated elsewhere

Migrants educated at destination

Migrants educated elsewhere

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Austria

-0.145

0.089

 

-0.023

0.065

 

-0.291

0.120

*

-0.054

0.093

 

Belgium

0.048

0.094

 

-0.045

0.089

 

 

 

 

-0.044

0.108

 

Canada

-0.084

0.084

 

-0.023

0.065

 

-0.057

0.082

 

-0.101

0.063

 

Cyprus (1,2)

-0.614

0.225

**

-0.060

0.152

 

-0.569

0.213

**

-0.248

0.167

 

Denmark

0.055

0.154

 

-0.002

0.066

 

-0.105

0.109

 

-0.003

0.064

 

Estonia

0.152

0.142

 

0.092

0.119

 

-0.096

0.132

 

-0.088

0.119

 

Finland

-0.071

0.081

 

 

 

 

 

 

 

 

 

 

France

0.073

0.057

 

-0.028

0.107

 

0.050

0.058

 

-0.177

0.080

*

Greece

-0.270

0.137

*

 

 

 

 

 

 

 

 

 

Ireland

0.049

0.122

 

0.066

0.147

 

0.181

0.121

 

0.130

0.157

 

Israel

0.266

0.100

**

0.429

0.209

*

0.313

0.095

**

0.095

0.178

 

Italy

 

 

 

-0.147

0.086

 

 

 

 

 

 

 

Lithuania

0.045

0.205

 

 

 

 

 

 

 

 

 

 

Netherlands

-0.155

0.117

 

-0.112

0.158

 

-0.254

0.099

*

 

 

 

New Zealand

-0.202

0.090

*

-0.132

0.107

 

-0.218

0.063

***

-0.201

0.096

*

Norway

0.143

0.094

 

-0.063

0.062

 

0.111

0.089

 

-0.090

0.110

 

Singapore

-0.156

0.130

 

-0.200

0.097

*

-0.197

0.107

 

-0.144

0.083

 

Slovenia

0.140

0.174

 

-0.067

0.070

 

 

 

 

 

 

 

Spain

 

 

 

-0.218

0.092

*

 

 

 

-0.238

0.115

*

Sweden

-0.078

0.092

 

-0.022

0.060

 

-0.066

0.085

 

0.084

0.069

 

United Kingdom

-0.087

0.147

 

-0.089

0.087

 

-0.131

0.140

 

-0.196

0.097

*

United States

0.225

0.135

 

-0.216

0.345

 

0.252

0.133

 

-0.218

0.310

 

Notes: (1) ***, **, and * represent 1%, 5% and 10% significance levels, respectively. (2) The dependent variable is the logarithm of hourly wages, including bonuses, in PPP-corrected USD. Specifications control for language, numeracy and literacy skills, age, age squared, age cubic, gender, marital status, number of children, education, migrant status, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies. (3) All specifications are weighted by the sampling weights provided in the dataset.

Note: See notes 1, 2 in Figure 5.1

Source: Survey of Adult Skills (PIAAC) (2012, 2015).

 StatLink http://dx.doi.org/10.1787/888933846631

Annex Table 5.A.3. The impact of educational attainment and migration on hourly wages (with language and skills controls and occupation fixed effects)

Medium Education

High Education

Migrants educated at destination

Migrants educated elsewhere

Migrants educated at destination

Migrants educated elsewhere

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Austria

-0.066

0.089

 

-0.020

0.061

 

-0.226

0.116

 

-0.042

0.082

 

Belgium

0.034

0.089

 

-0.037

0.092

 

 

 

 

-0.076

0.090

 

Canada

-0.075

0.079

 

-0.019

0.065

 

-0.048

0.077

 

-0.063

0.061

 

Cyprus (1,2)

-0.428

0.216

*

-0.013

0.150

 

-0.333

0.201

 

-0.181

0.159

 

Denmark

0.056

0.153

 

0.029

0.065

 

-0.103

0.107

 

0.034

0.061

 

Estonia

0.122

0.140

 

0.062

0.115

 

-0.135

0.124

 

-0.082

0.113

 

Finland

-0.034

0.071

 

 

 

 

 

 

 

 

 

 

France

0.086

0.055

 

-0.058

0.106

 

0.044

0.055

 

-0.121

0.076

 

Greece

-0.188

0.164

 

 

 

 

 

 

 

 

 

 

Ireland

0.128

0.129

 

0.123

0.137

 

0.222

0.123

 

0.185

0.152

 

Israel

0.130

0.086

 

0.313

0.206

 

0.253

0.090

**

0.084

0.173

 

Italy

 

 

 

-0.081

0.085

 

 

 

 

 

 

 

Lithuania

0.063

0.190

 

 

 

 

 

 

 

 

 

 

Netherlands

-0.176

0.114

 

-0.072

0.162

 

-0.283

0.096

**

 

 

 

New Zealand

-0.213

0.088

*

-0.168

0.101

 

-0.188

0.068

**

-0.179

0.094

 

Norway

0.056

0.078

 

-0.079

0.067

 

0.042

0.078

 

-0.083

0.119

 

Singapore

-0.115

0.135

 

-0.097

0.089

 

-0.165

0.105

 

-0.087

0.078

 

Slovenia

0.137

0.168

 

-0.050

0.068

 

 

 

 

 

 

 

Spain

 

 

 

-0.219

0.091

*

 

 

 

-0.224

0.110

*

Sweden

-0.022

0.088

 

0.012

0.053

 

-0.021

0.082

 

0.171

0.066

*

United Kingdom

-0.060

0.137

 

-0.122

0.087

 

-0.111

0.133

 

-0.200

0.107

 

United States

0.229

0.131

 

-0.200

0.334

 

0.235

0.127

 

-0.204

0.287

 

Notes: (1) ***, **, and * represent 1%, 5% and 10% significance levels, respectively. (2) The dependent variable is the logarithm of hourly wages, including bonuses, in PPP-corrected USD. Specifications control for occupation dummies, language, numeracy and literacy skills, age, age squared, age cubic, gender, marital status, number of children, education, migrant status, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies. (3) All specifications are weighted by the sampling weights provided in the dataset.

Note: See notes 1, 2 in Figure 5.1

Source: Survey of Adult Skills (PIAAC) (2012, 2015).

 StatLink http://dx.doi.org/10.1787/888933846650

Annex Table 5.A.4. The impact of migration by origin on occupational placement (without language and skills controls)

Migrants - Europe (EU)

Migrants - Europe (non-EU)

Migrants - Outside Europe

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Austria

-0.131

2.117

 

-1.141

1.512

 

-3.743

1.972

 

-11.010

1.266

***

 

 

 

 

 

 

Belgium

-3.930

3.243

 

0.118

1.503

 

 

 

 

 

 

 

 

 

 

 

 

 

Canada

2.522

1.259

*

-3.592

1.713

*

-0.011

2.584

 

-8.300

2.997

**

1.092

0.763

 

-6.800

0.693

***

Cyprus (1,2)

-2.709

1.883

 

-3.384

1.939

 

 

 

 

-6.447

3.012

*

2.388

1.777

 

 

 

 

Denmark

1.240

1.781

 

-7.477

1.639

***

0.861

1.848

 

-8.606

1.746

***

-3.474

1.552

*

-13.680

1.598

***

Estonia

 

 

 

 

 

 

-2.947

1.058

**

-5.993

1.463

***

 

 

 

 

 

 

Finland

-1.950

2.829

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

France

-2.446

1.334

 

-3.130

1.695

 

 

 

 

 

 

 

-3.255

1.044

**

-9.072

1.331

***

Greece

-2.587

1.459

 

 

 

 

-4.409

2.124

*

-13.090

2.772

***

 

 

 

 

 

 

Ireland

-1.230

1.624

 

-5.627

1.102

***

 

 

 

 

 

 

0.976

2.392

 

0.024

3.148

 

Israel

1.548

2.112

 

 

 

 

0.358

1.339

 

-14.890

1.532

***

-3.454

1.139

**

-7.527

2.067

***

Italy

 

 

 

-6.990

1.417

***

 

 

 

-7.027

1.529

***

 

 

 

-5.154

1.341

***

Lithuania

 

 

 

 

 

 

1.314

2.136

 

 

 

 

 

 

 

 

 

 

Netherlands

 

 

 

 

 

 

 

 

 

 

 

 

0.273

2.351

 

-10.110

3.600

**

New Zealand

1.704

1.478

 

-0.959

1.439

 

 

 

 

 

 

 

-0.193

1.153

 

-4.395

1.030

***

Norway

-2.706

1.915

 

-6.723

1.533

***

 

 

 

 

 

 

-2.686

1.739

 

-16.160

2.557

***

Singapore

 

 

 

 

 

 

 

 

 

 

 

 

1.151

0.625

 

-0.878

0.572

 

Slovenia

0.226

2.546

 

 

 

 

-1.574

1.990

 

-5.090

1.281

***

 

 

 

 

 

 

Spain

 

 

 

-4.241

2.315

 

 

 

 

 

 

 

3.231

1.732

 

-2.072

1.384

 

Sweden

1.644

1.725

 

-7.037

2.194

**

 

 

 

-15.670

2.954

***

-4.727

1.566

**

-12.400

1.780

***

United Kingdom

0.599

2.287

 

-8.028

1.771

***

 

 

 

 

 

 

-3.204

1.608

*

-5.959

1.869

**

United States

 

 

 

 

 

 

 

 

 

 

 

 

1.187

1.266

 

-5.642

1.186

***

Notes: (1) ***, **, and * represent 1%, 5% and 10% significance levels, respectively. (2) The dependent variable is the ISEI index. Specifications control for age, age squared, age cubic, gender, marital status, number of children, education, migrant status, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies. (3) All specifications are weighted by the sampling weights provided in the dataset.

Note: See notes 1, 2 in Figure 5.1

Source: Survey of Adult Skills (PIAAC) (2012, 2015).

 StatLink http://dx.doi.org/10.1787/888933846669

Annex Table 5.A.5. The impact of migration by origin on occupational placement (with language and skills controls)

Migrants – Europe (EU)

Migrants – Europe (non-EU)

Migrants – Outside Europe

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Austria

0.572

2.086

 

0.629

1.544

 

-0.465

2.141

 

-4.193

1.800

*

 

 

 

 

 

 

Belgium

-3.449

3.043

 

1.539

1.509

 

 

 

 

 

 

 

 

 

 

 

 

 

Canada

3.157

1.269

*

-0.601

1.540

 

1.082

2.536

 

-2.600

2.761

 

3.345

0.841

***

-1.963

0.840

*

Cyprus (1,2)

-3.020

1.774

 

-3.635

1.902

 

 

 

 

-7.586

2.875

**

1.814

1.851

 

 

 

 

Denmark

0.942

1.788

 

-6.825

1.869

***

1.953

1.931

 

-7.063

1.890

***

-2.090

1.763

 

-10.039

1.965

***

Estonia

 

 

 

 

 

 

-2.502

1.098

*

-4.839

1.497

**

 

 

 

 

 

 

Finland

-1.955

2.577

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

France

-1.504

1.384

 

-0.063

2.013

 

 

 

 

 

 

 

-0.586

1.018

 

-5.078

1.288

***

Greece

-2.634

1.693

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ireland

-1.363

1.626

 

-4.034

1.306

**

 

 

 

 

 

 

1.806

2.399

 

1.868

3.160

 

Israel

1.115

2.104

 

 

 

 

-0.641

1.433

 

-14.188

1.535

***

-2.527

1.089

*

-6.610

2.074

**

Italy

 

 

 

-5.167

1.690

**

 

 

 

-5.570

1.864

**

 

 

 

-2.444

1.974

 

Lithuania

 

 

 

 

 

 

1.145

2.189

 

 

 

 

 

 

 

 

 

 

Netherlands

 

 

 

 

 

 

 

 

 

 

 

 

2.253

2.599

 

-5.136

3.868

 

New Zealand

1.532

1.512

 

-0.785

1.357

 

 

 

 

 

 

 

1.238

1.180

 

-1.634

1.178

 

Norway

-2.011

2.063

 

-2.274

1.744

 

 

 

 

 

 

 

1.060

2.018

 

-9.131

3.076

**

Singapore

 

 

 

 

 

 

 

 

 

 

 

 

1.189

0.621

 

0.444

0.563

 

Slovenia

0.304

2.539

 

 

 

 

-1.070

1.904

 

-3.062

1.589

 

 

 

 

 

 

 

Spain

 

 

 

-2.471

2.552

 

 

 

 

 

 

 

3.877

1.742

*

-0.863

1.387

 

Sweden

1.760

1.809

 

-4.962

2.229

*

 

 

 

-13.065

3.079

***

-2.097

1.670

 

-7.990

2.151

***

United Kingdom

2.399

2.390

 

-3.172

2.284

 

 

 

 

 

 

 

0.059

1.460

 

-1.423

1.724

 

United States

 

 

 

 

 

 

 

 

 

 

 

 

2.656

1.317

*

-1.666

1.245

 

Notes: (1) ***, **, and * represent 1%, 5% and 10% significance levels, respectively. (2) The dependent variable is the ISEI index. Specifications control for language, numeracy and literacy skills, age, age squared, age cubic, gender, marital status, number of children, education, migrant status, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies. (3) All specifications are weighted by the sampling weights provided in the dataset.

Note: See notes 1, 2 in Figure 5.1

Source: Survey of Adult Skills (PIAAC) (2012, 2015).

 StatLink http://dx.doi.org/10.1787/888933846688

Annex Table 5.A.6. The impact of migration by origin on self-assessed skill underutilisation

Migrants - Europe (EU)

Migrants - Europe (non-EU)

Migrants - Outside Europe

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Austria

-0.025

0.041

 

0.033

0.020

 

-0.003

0.040

 

-0.085

0.056

 

 

 

 

 

 

 

Belgium

0.021

0.063

 

0.063

0.043

 

 

 

 

 

 

 

 

 

 

 

 

 

Canada

0.038

0.021

 

0.019

0.033

 

0.091

0.019

***

0.033

0.052

 

0.021

0.017

 

0.063

0.017

***

Cyprus (1,2)

-0.067

0.048

 

0.049

0.031

 

 

 

 

-0.119

0.069

 

0.068

0.031

*

 

 

 

Denmark

0.024

0.038

 

-0.022

0.041

 

0.143

0.029

***

0.072

0.046

 

0.022

0.042

 

0.019

0.042

 

Estonia

 

 

 

 

 

 

-0.011

0.028

 

-0.023

0.036

 

 

 

 

 

 

 

Finland

0.050

0.071

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

France

0.038

0.048

 

0.012

0.056

 

 

 

 

 

 

 

0.008

0.032

 

0.000

0.041

 

Greece

-0.021

0.040

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ireland

-0.039

0.034

 

0.001

0.023

 

 

 

 

 

 

 

-0.050

0.062

 

0.017

0.041

 

Israel

-0.037

0.045

 

 

 

 

-0.014

0.024

 

-0.051

0.034

 

0.006

0.021

 

-0.063

0.044

 

Italy

 

 

 

0.042

0.050

 

 

 

 

-0.032

0.077

 

 

 

 

0.142

0.044

**

Lithuania

 

 

 

 

 

 

0.033

0.038

 

 

 

 

 

 

 

 

 

 

Netherlands

 

 

 

 

 

 

 

 

 

 

 

 

-0.002

0.066

 

-0.060

0.092

 

New Zealand

-0.009

0.030

 

-0.008

0.028

 

 

 

 

 

 

 

0.021

0.019

 

0.052

0.025

*

Norway

-0.047

0.054

 

-0.017

0.051

 

 

 

 

 

 

 

-0.023

0.046

 

-0.076

0.073

 

Singapore

 

 

 

 

 

 

 

 

 

 

 

 

0.018

0.019

 

0.030

0.019

 

Slovenia

0.040

0.029

 

 

 

 

0.007

0.037

 

0.013

0.033

 

 

 

 

 

 

 

Spain

 

 

 

0.049

0.038

 

 

 

 

 

 

 

-0.026

0.045

 

0.028

0.019

 

Sweden

0.065

0.037

 

0.071

0.042

 

 

 

 

0.162

0.045

***

0.065

0.035

 

0.092

0.048

 

United Kingdom

-0.014

0.045

 

0.019

0.046

 

 

 

 

 

 

 

0.030

0.025

 

-0.012

0.046

 

United States

 

 

 

 

 

 

 

 

 

 

 

 

0.014

0.023

 

-0.060

0.043

 

Notes: (1) ***, **, and * represent 1%, 5% and 10% significance levels, respectively. (2) The dependent variable is a dummy being one if the respondent feels that he/she has the skills to cope with more demanding duties than those he/she is required to perform in his/her current job. Specifications control for language, numeracy and literacy skills, age, age squared, age cubic, gender, marital status, number of children, education, migrant status, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies. (3) All specifications are weighted by the sampling weights provided in the dataset.

Note: See notes 1, 2 in Figure 5.1

Source: Survey of Adult Skills (PIAAC) (2012, 2015).

 StatLink http://dx.doi.org/10.1787/888933846707

Annex Table 5.A.7. The impact of migration by origin on self-assessed skill deficit

Migrants - Europe (EU)

Migrants - Europe (non-EU)

Migrants - Outside Europe

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

Educated at destination

Educated elsewhere

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Coef.

S.E.

 

Austria

0.018

0.078

 

0.039

0.050

 

0.047

0.073

 

-0.031

0.060

 

 

 

 

 

 

 

Belgium

-0.004

0.068

 

-0.107

0.042

*

 

 

 

 

 

 

 

 

 

 

 

 

Canada

-0.009

0.033

 

0.019

0.042

 

0.109

0.093

 

0.083

0.066

 

0.108

0.032

***

0.111

0.026

***

Cyprus (1,2)

-0.169

0.065

**

0.010

0.066

 

 

 

 

-0.098

0.074

 

-0.067

0.100

 

 

 

 

Denmark

-0.083

0.032

**

-0.028

0.034

 

0.019

0.059

 

-0.084

0.054

 

0.014

0.044

 

0.004

0.043

 

Estonia

 

 

 

 

 

 

0.027

0.035

 

-0.040

0.042

 

 

 

 

 

 

 

Finland

-0.068

0.070

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

France

-0.028

0.045

 

-0.029

0.053

 

 

 

 

 

 

 

0.030

0.034

 

0.033

0.041

 

Greece

-0.054

0.075

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ireland

0.083

0.038

*

0.036

0.032

 

 

 

 

 

 

 

0.048

0.076

 

0.085

0.062

 

Israel

0.074

0.063

 

 

 

 

-0.049

0.038

 

-0.115

0.039

**

0.005

0.039

 

0.043

0.059

 

Italy

 

 

 

-0.081

0.073

 

 

 

 

-0.037

0.102

 

 

 

 

-0.101

0.108

 

Lithuania

 

 

 

 

 

 

0.007

0.082

 

 

 

 

 

 

 

 

 

 

Netherlands

 

 

 

 

 

 

 

 

 

 

 

 

0.014

0.062

 

-0.009

0.083

 

New Zealand

0.037

0.042

 

-0.004

0.045

 

 

 

 

 

 

 

0.068

0.034

*

0.081

0.038

*

Norway

-0.047

0.06

 

-0.062

0.050

 

 

 

 

 

 

 

-0.029

0.062

 

0.056

0.070

 

Singapore

 

 

 

 

 

 

 

 

 

 

 

 

0.014

0.028

 

0.018

0.026

 

Slovenia

0.095

0.073

 

 

 

 

0.033

0.068

 

-0.076

0.063

 

 

 

 

 

 

 

Spain

 

 

 

-0.145

0.063

*

 

 

 

 

 

 

0.006

0.072

 

0.020

0.042

 

Sweden

0.081

0.055

 

-0.023

0.059

 

 

 

 

-0.036

0.084

 

0.115

0.047

*

0.112

0.072

 

United Kingdom

-0.040

0.057

 

0.189

0.075

*

 

 

 

 

 

 

0.081

0.044

 

0.036

0.066

 

United States

 

 

 

 

 

 

 

 

 

 

 

 

0.085

0.037

*

0.117

0.053

*

Notes: (1) ***, **, and * represent 1%, 5% and 10% significance levels, respectively. (2) The dependent variable is a dummy being one if the respondent feels that he/she needs further training in order to cope well with his/her present duties. Specifications control for language, numeracy and literacy skills, age, age squared, age cubic, gender, marital status, number of children, education, migrant status, number of jobs, a dummy for public-sector workers, a dummy for indefinite contract, industry dummies. (3) All specifications are weighted by the sampling weights provided in the dataset.

Note: See notes 1, 2 in Figure 5.1

Source: Survey of Adult Skills (PIAAC) (2012, 2015).

 StatLink http://dx.doi.org/10.1787/888933846726

Notes

← 1. Notable exceptions are the International Adult Literacy Survey (IALS) and the Adult Literacy and Life Skills Survey (ALL), produced during the 1990s and 2000s, respectively (see OECD/Statistics Canada (2000[35]); McIntosh and Vignoles (2001[36]); and Clarke and Skuterud, (2016[14]), among others). Compared to such skills datasets, however, the Survey of Adult Skills (PIAAC) is better suited to capturing the different skills that occupations (particularly complex occupations) require of workers. For example, PIAAC contains much more information on numeracy skills, and a more granular definition of literacy, which also includes an assessment of the ability to read digital texts.

← 2. Note by Turkey:

The information in this document with reference to “Cyprus” relates to the southern part of the Island. There is no single authority representing both Turkish and Greek Cypriot people on the Island. Turkey recognises the Turkish Republic of Northern Cyprus (TRNC). Until a lasting and equitable solution is found within the context of the United Nations, Turkey shall preserve its position concerning the “Cyprus issue”.

← 3. Note by all the European Union Member States of the OECD and the European Union:

The Republic of Cyprus is recognised by all members of the United Nations with the exception of Turkey.

← 4. Data for United Kingdom were collected only in England and Northern Ireland, while data for Belgium refers to Flanders only.

← 5. For example, as shown by Toomet (2011[34]), even a command of the official destination language is not enough to eliminate the wage differential. As for African-Americans in the United States or Latin American immigrants in Spain, the members of a minority still suffer from the glass-ceiling effect.

← 6. Despite its important policy implications, only few works have exploited direct skills measures to study immigrants’ labour market performance. Ferrer et al. (2006[33]) use the Ontario Immigrant Literacy Survey to reject the hypothesis that immigrants receive different returns to literacy skills than natives, suggesting that natives’ earnings premium cannot be explained by discrimination against foreigners, but rather by different skills endowments between the two groups. Canada is again at the centre of the analysis by Coulombe and Tremblay (2009[32]), which adopts information from the 2003 International Adult Literacy and Skills Survey (IALSS) to show that, while using education data suggests a brain drain that benefits the Canadian economy, results are reversed if using skills data: international migrants have overall lower skills levels than similarly educated natives.

← 7. Note that respondents of the Survey of Adult Skills (PIAAC) were also assessed in their ability to solve problems in technology-rich environments. However, not all countries included this exercise in their questionnaires (France, Italy and Spain, for example, are excluded). Moreover, the share of respondents to this section is as low as 24% in certain countries, such as Estonia. As analysing problem solving in technology-rich environments would imply focusing only on a selected sub-sample of workers, it is not considered in the analysis.

← 8. Note that empty cells represent missing data or estimations with fewer than 30 observations.

← 9. Some of the controls included in the analysis may be endogenous to the outcome of interest, that is, they may be correlated with unobservable characteristics also affecting wages, thereby biasing the parameters of interest. Thus, as a check of robustness, regressions have also been estimated without such controls – namely marital status, number of children, dummies for having two or more jobs, for public employment, for indefinite job contract, and for industries. Results are robust to this test. In addition, findings remain both quantitatively and qualitatively similar also after including a dummy for being a recent immigrant (less than five years since arrival at the destination country).

← 10. See notes 2 and 3.

← 11. See notes 2 and 3.

← 12. A notable exception is Israel, where returns to education are overall greater for immigrants than for natives. This is likely to be due to the singular composition of its foreign-born workforce in the PIAAC survey. For instance, the country has experienced a large influx of skilled immigrants, and almost one in two immigrants comes from the former Soviet Union. Likewise, in the United States, immigrants educated there seem to have slightly higher returns to tertiary education than natives. A similar result has also been found by Bonfanti and Xenogiani (2014[6]), who explain it as possibly linked to the fact that the model presupposes constant, rather than decreasing, returns to schooling.

← 13. Throughout the chapter, controls for language and numeracy proficiency are included simultaneously in the specification. Importantly, results remain qualitatively similar if language and numeracy skills are included separately.

← 14. There are several reasons why host-language proficiency is an important factor behind immigrants’ assimilation in destination labour markets. From an economic perspective, weak linguistic abilities decrease immigrants' productivity and hence earnings, and reduce the range and quality of jobs available to foreigners. From a social perspective, a lack of proficiency in the host-country language may foster discrimination and isolation.

← 15. Literacy and numeracy skills in PIAAC are measured in the host-country language. Although the analysis of this chapter controls for whether the individual speaks the language of the survey, a fluency effect may remain.

← 16. There are several explanations for a possible over-representation of immigrants in less-prestigious occupations. Employers might consider foreign qualifications of a lesser quality and hence offer the foreign-born adults who hold them low-prestige jobs. Credit constraints might force immigrants to accept the first available job offer and thus they end up with potentially poorer matches.

← 17. The literature on the approaches to occupational stratification is vast. This analysis relies on the ISEI measure as it is practical, and it has been widely exploited in the socio-economic literature (see for instance Marks (2005[37]), and Raitano and Vona (2015[38]), among others). This does not imply that a similar analysis would not have been possible using different indices, both categorical and continuous.

← 18. Extensive reviews of the literature on overqualification and skills mismatch can be found in Quintini (2011[31]), Prokic-Breuer and McManus (2016[39]), and Pellizzari and Fichen (2017[40]).

← 19. Clearly, as noticed by Pellizzari and Fichen (2017[40]), subjective measures of mismatch do not come without caveats, the most obvious being that individuals might misreport their true skills due to overconfidence. In response to this potential drawback, the OECD has developed a more complex measure of skills mismatch, which classifies workers as well-matched in literacy or numeracy if their proficiency score in that domain is between the minimum and maximum score observed among workers who answered “no” to both questions in the same occupation and country (OECD, 2013[41]). A detailed analysis for this measure of skills mismatch – as well as other measures – for foreign-born workers is presented in Bonfanti and Xenogiani (2014[6]).

← 20. Findings remain similar without controls for language and skills.

End of the section – Back to iLibrary publication page