Access the full text.
Sign up today, get DeepDyve free for 14 days.
This paper explores the role of school quality in immigrants’ home countries on their earnings in Germany, using native Germans as a benchmark. We propose an empirical analysis that highlights two important insights. First, there is a substantial gap in the returns to education between natives and immigrants in Germany, especially when we consider the quality of schooling in the source country where education was obtained. In particular, lower school quality reduces the endowment advantage that immigrants possess from their education. Second, quality‑adjusted education helps us to better understand the potential driving force behind the native–immigrant wage gap. We show that this measure accounts for a substantial fraction of the unexplained part in the Oaxaca–Blinder decomposition. These findings emphasize the role of the school quality in explaining the imperfect transferability of human capital of immigrants in Germany. Keywords: School quality, Wage gap decomposition, Immigration, Earnings JEL Classification: I21, I26, J31, J61 capital transferability is the quality of schooling in immi- 1 Introduction grants’ source countries. In the aftermath of the financial crisis in 2009, Germany The majority of existing evidence on the native–immi - experienced a surge of immigration with an average grant wage gap considers education obtained abroad as a annual net flow of approximately 300,000 people, a clear perfect substitute for the education in the host country majority of whom came from new member states of the (Basilio et al. 2017). However, one year of schooling in the European Union (Dustmann et al. 2012). This consider - host country might effectively be equal to more or less ably changed the composition of source countries, indi- than a year of schooling in other countries. Therefore, it cating a shift towards European immigrants, and a skill is crucial to consider not only the years of education but structure marked by higher skill levels of immigrants in also the quality of these years of education (Rohrbach- Germany (Muysken et al. 2015). A major challenge of Schmidt and Tiemann 2016; Wößmann 2003). integrating immigrants into the host country are barriers In this paper, we deviate from the conventional to the transferability of human capital endowments and approach and provide an important measure that their educational skills (Chiswick and Miller 2008). It is accounts for the differences in the quality of human well-known from empirical evidence that these barriers capital endowments between the host and source coun- explain a substantial part of the native–immigrant wage tries of immigrants. We examine the wage assimilation gap (Friedberg 2000). In this paper, we show that one of of immigrants in Germany and the determinants of the the key factors characterizing the limitation of human native–immigrant wage gap. We adjust the educational level of immigrants by using the school quality index of Hanushek and Kimko (2000). The goal of this paper is to *Correspondence: email@example.com understand how the pattern of the wage growth changes Otto–Friedrich University of Bamberg, Bamberg, Germany when the school quality index is taken into account. In Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. 8 Page 2 of 15 H. Le‑Quang , E. Vallizadeh Using a large, representative household dataset—the Germany Socio-Economic Panel (GSOEP) 1984–2016— we show that a substantial part of the unexplained wage gap decomposition is explained by the school quality in the source countries of immigrants. Further, these data provide detailed information about the pre- and post- migration educational activities of immigrants, which serves as a basis to understand the differences in the returns to education in Germany and the returns to edu- cation abroad. Unlike Coulombe et al. (2014), who use gross domestic product (GDP) as a proxy for school qual- ity, or Basilio et al. (2017), who do not adjust the quality of foreign schooling in their wage estimations, we use a direct measure of school quality to calculate the effective years of schooling of immigrants in Germany. In particu- Fig. 1 Wage assimilation of foreign‑ educated immigrants ( Total lar, we first split the total years of education into years sample). Sample of individuals aged 18 to 64, having full‑time or regular part‑time employment. The (unadjusted) wage assimilation studying abroad and years studying in Germany. Next, we over the time since the migration of immigrants educated in foreign adjust years studying abroad by using the Hanushek and countries before arriving in Germany. The dotted lines represent Kimko (2000) school quality index. the 95 percent confidence interval. The red horizontal line on top This approach is suited for the empirical analysis for a illustrates the mean wage of natives over time. Sources: German number of reasons. First, although GDP has a strong cor- Socio‑Economic Panel (GSOEP) (wave: 1984–2016), version 33, released in 2018 relation with the amount of resources allocated to edu- cation, the question is whether this indicator measures the quantity rather than the quality of schooling. Fur- thermore, there could be a reverse causality where higher Fig. 1 we explore data from the German Socio-Economic human capital leads to higher GDP (Hanushek 2005). Panel (wave: 1984–2016) to decompose the wage assimi- Second, unlike contemporary school quality data such lation pattern of immigrants into low-quality and high- as PISA or PIACC scores, the Hanushek and Kimko quality schooling groups . As illustrated in this figure, (2000) school quality index is a comprehensive indicator, upon arrival, both groups of immigrants have, on aver- which is constructed based on six voluntary international age, lower wages than natives. However, immigrants who standardized tests in mathematics and science conducted obtained their education in countries with high school- between 1965 and 1991. Therefore, it can potentially ing quality have higher wages than the other immigrant better capture the time during which immigrants in the group. Moreover, the rate at which they catch up to sample actually went to school in their home countries . natives is faster for the high-quality schooling group of Our empirical findings provide two novel insights. immigrants. On average, their wages converge approxi- First, we show that the returns to quality-adjusted edu- mately 8 years faster to natives’ wage level than the low- cation are positive but significantly less than the returns quality schooling group of immigrants. to the unadjusted education of immigrants in Germany. From this interesting observation, this paper attempts to improve the understanding of the role of schooling quality on the native–immigrant wage gap. Particularly, two key questions at the core of our analysis are (1) How PISA (Program for International Student Assessment) from the OCED does the school quality of immigrants’ source countries measures 15-year-old students’ ability to use their reading, mathematics affect the returns to education? (2) How does school and science knowledge and skills to meet real-life challenges. The tests have quality impact the native–immigrant wage gap? been conducted in more than 90 countries since 2000 and organized every three years (OECD 2022a). The Program for the International Assessment of Adult Competencies (PIAAC) developed and conducts the Survey of Adult Skills. The survey measures adults’ proficiency in key information-processing skills—literacy, numeracy and problem solving in technology-rich environ- The low-quality schooling group consists of immigrants from countries for ments—and gathers information and data on how adults use their skills at which the Hanushek and Kimko (2000) school quality index is smaller than home, at work and in the wider community. The PIACC was initiated in 2008 the first quartile of the school quality index distribution (i.e., where the school (OECD 2022b). quality index equals 46.77). The high-quality schooling group consists of immigrants from countries for which the Hanushek and Kimko (2000) school One disadvantage of using this observed measure of cognitive skills is quality index is larger than or equals 46.77. The Hanushek and Kimko (2000) that there are confounding factors outside of formal schools such as home quality index, which is constructed based on international standardized tests background or social background factors other than the instructional char- in mathematics and science, offers the opportunity to differentiate between acteristics of schools that we cannot control for (Hill and Rowe (1996) and low- and high-quality schooling countries. Hanushek and Kimko (2000)). The returns to school‑quality‑adjusted education of immigrants in Germany Page 3 of 15 8 This finding shows that lower school quality negates the accounts for a large fraction (from 30 to 80 percent) endowment advantage that immigrants possess from of the variation in the rates of return to education not their education. Second, there is a wage gap between only between immigrants and natives but also between natives and immigrants, and education plays an impor- immigrants who acquire education in the host coun- tant role in explaining this earnings difference. Exist - tries and those who do not. In other words, the lower ing studies often use wage gap decomposition methods partial effect of schooling on earnings for immigrants breaking down mean wage differentials into explanatory from less developed countries is due to lower quality of determinants and an unexplained part. The interpreta - foreign schooling, even after allowing for differences in tion of the unexplained component is vulnerable because working experience and other factors that might influ - it captures a range of factors, such as discrimination ence earnings. effects and unobserved individual or institutional charac - Low international transferability of skills is a poten- teristics. Our analysis reveals that controlling for school tial explanation for the poor labour market outcomes of quality allows to account for important institutional immigrants in a host country due to large differences in characteristics. This factor significantly reduces the unex - human capital quality across countries (Friedberg 2000; plained wage gap. Ultimately, our results highlight the Chiswick and Miller 2009, 2010; Aleksynska and Tritah role of school quality in understanding the international 2013; Basilio et al. 2017). In particular, the differences in transferability of human capital and the returns to educa- schooling systems, unrecognized qualifications, techno - tion of immigrants in the host country. logical development, and other barriers to labour market Ever since the seminal research of Card and Krueger entry could adversely affect international skill transfer - (1992) on the causal relationship between school qual- ability (de Oliveira et al. 2000; Aleksynska and Tritah ity and earnings in the United States, economists have 2013). used different measures of school quality to investigate This paper contributes to the literature in several its transmission mechanism to earnings. The quantity of important ways. First, we use an alternative measure of schooling, which represents resources devoted to edu- years of education obtained abroad that adjusts for the cation such as the pupil–teacher ratio, expenditures per quality of schooling in the source country of migration. pupil, relative wages of teachers, and the length of school In this way, we are able to compute plausible and precise term, is easily measured, and data are widely available estimates of the returns to education of immigrants in (Hanushek 2005). However, the central concern in con- the host country. Second, we use a large representative temporary education policy revolves much more around panel dataset, covering a long period until 2016, which the quality of schooling. Compared to the quantity of allows us to include recent immigration flows to Ger - schooling, the measurement of school quality is more many, such as the refugee influx in 2015 (Eurostat 2018). challenging. This is because the question is whether we Third, using a Oaxaca–Blinder decomposition model, we can assume international standardized test scores to be show that using a quality-adjusted measure of education an appropriate measure of cognitive skills affected by the substantially reduces the unexplained part of the native– quality of schooling and whether students’ performance immigrant wage gap by approximately 20 percent. Thus, on these standardized tests has any correlation with eco- our approach reveals that school quality appears to be a nomic outcomes such as their subsequent labour market major factor in explaining the imperfect transferability performance (Hanushek 2005). of human capital and decreases the level of “ignorance” Nevertheless, Chiswick and Miller (2010) and Cou- often attributed to “discrimination” in previous literature. lombe et al. (2014) still attempt to use different meas - Nevertheless, another challenge in estimating the ures of school quality such as the PISA score and the returns to education is the omitted variable bias that may national GDP to explain the relatively lower labour arise due to unobservable variables, for example, family market performance of immigrants in host coun- background, innate ability, motivation, and other non- tries. Raaum (1998) and Friedberg (2000) estimate the cognitive skills. We assume that the direction of this bias return to foreign schooling of immigrants in Norway is the same for both immigrants and natives. Under this and Israel, respectively. They find a significantly lower assumption, we can compare the returns to education return to foreign education than to host-country educa- among natives, German-educated immigrants and for- tion. Betts and Lofstrom (2000), Bratsberg and Ragan Jr eign-educated immigrants. (2002), Bratsberg and Terrell (2002) and Chiswick and The paper proceeds as follows. Section 2 presents our Miller (2010) investigate the payoff of schooling for empirical strategies. Section 3 introduces the dataset immigrants in the United States, and Sweetman (2004) and our key variables for analysis. Section 4 analyses and Fortin et al. (2016) do so in Canada and find that the empirical findings and provides some insights of the the quality of the educational system obtained abroad results. Section 5 concludes the paper. 8 Page 4 of 15 H. Le‑Quang , E. Vallizadeh 2 Empirical strategy This component tells us that a native–immigrant wage We propose an empirical analysis that consists of two gap persists even when natives and immigrants are alike consecutive steps. In the first step, we estimate the dif - in terms of their observable attributes. In other words, ferences in returns to total years of education between this component is often referred to as “discrimination” in natives and immigrants using a Mincerian earnings equa- the literature. tion. The mean natural logarithm of hourly wages of In this analysis, we present our baseline estimates using natives and immigrants in Germany is thus represented unadjusted education and compare the results with the by the following two equations: quality-adjusted years of education. In doing so, we show (1) ln(w ) =β + β Educ + β Exp + β Exp + β X + β s + β t + ǫ N 0 1 N 2 N 3 4 N 5 6 N (2) ln(w ) =α + α Educ + α Exp + α Exp + α X + α s + α t + α c + η I 0 1 I 2 I 3 4 I 5 6 7 I Let the subscripts N and I denote natives and immi- that using the adjusted years of education provides more grants, respectively; Educ denotes total years of educa- plausible estimates and reduces the unexplained differ - tion; Exp denotes years of experience with the current ence in earnings and thus improves our understanding of firm in Germany; X denotes a vector of other control the native–immigrant wage gap. variables (gender, marital status, age, age squared, years As mentioned previously, the estimation of the returns since migration, years since migration squared, German to education using Mincerian regression brings about language skills, establishment size and industry dum- several challenges. First, years of education and the mies); s denotes federal state fixed effects; t denotes the school quality index may be subject to measurement survey year fixed effects; c denotes the country of ori - errors. Particularly, the years of education in the GSOEP gin fixed effects; and ǫ and η denote idiosyncratic error are self-reported and converted from the highest quali- terms. We control for these fixed effects to capture the fication attained, which may induce problems of hetero - common development in a specific state or country and geneous reporting behaviours. The school quality index at a specific point in time. is assumed to be constant over the sampling period but The only difference between these two equations is that may change over time in practice. for foreign-educated immigrants, we adjust the quality of Second, our estimation may suffer from omitted vari - schooling for the years of education abroad, while we do able bias because we are unable to observe and meas- not do so for natives and German-educated immigrants. ure innate ability, motivation, and non-cognitive skills. In the second step of our analysis, we decompose the Years of experience with the current firm in Germany, native–immigrant wage gap using Oaxaca–Blinder or tenure, is also potentially an endogenous variable. decomposition method (Blinder 1973; Oaxaca 1973). Burdett (1978) and Jovanovic (1979) show in theoretical This method separates the wage gap into the explained search models that a high-productivity job–skill match is component due to differences in observed characteris - unlikely to end and that firm tenure is positively corre - tics between natives and immigrants and the unexplained lated with employees’ productivity and wages. Likewise, component often attributed to discrimination or differ - establishment size and industry may also be endogenous ences in unobserved characteristics between natives and because workers can self-select into certain industries immigrants. and establishments of a given size. ln(w ) − ln(w ) = (X − X )β + (β − β )X I N I N N I N I mean wage gap explained by means of regressors unexplained component The first term on the right-hand side of this equation Systematic measurement errors may be of concern measures the explained component of the wage gap. This if there are systematic differences in reporting hetero - component includes the portion of the wage gap due to geneity or unobservable characteristics across popula- differences in the observed characteristics of immigrants tion groups. Since our empirical strategy is to compare and natives, evaluated by the coefficients of the natives. returns to education obtained abroad with returns to The second term on the right-hand side of the equation education obtained in Germany, the problem of system- measures the unexplained component of the wage gap. atic reporting bias or omitted variable bias should be less The returns to school‑quality‑adjusted education of immigrants in Germany Page 5 of 15 8 of a concern. This is particularly true if we assume that of outliers that could bias our estimation results, we win- there are not many differences in unobserved characteris - sorize wages at the 5th and 95th percentiles. That is, we tics explaining the educational attainment levels between set all data below the 5th percentile to the 5th percentile individuals who obtained their education abroad and and data above the 95th percentile to the 95th percen- those who obtained their education in Germany. To over- tile—a method documented by Amann and Klein (2012). come some of the heterogeneity problems, we add coun- Years of quality-adjusted education is the single most try of origin fixed effects and federal state fixed effects important variable in this research. One drawback of the to control for migration motives and regional economic survey data is that they do not contain information about conditions, respectively. the source country where immigrants obtained their edu- cation; hence, we make the following assumptions. First, 3 Data and variables all native Germans and immigrants who migrated to Ger- 3.1 The German socio‑economic panel (GSOEP) many before the age of six acquire education exclusively We use the German Socio-Economic Panel (GSOEP) for in Germany. These immigrants are called German-edu - the years 1984–2016. The German Socio-Economic Panel cated immigrants. Second, for immigrants who migrated (SOEP) is a wide-ranging representative longitudinal to Germany after the age of six, we observe whether study of private households, conducted by the German they either acquire education exclusively in their home Institute for Economic Research, DIW Berlin. Starting countries or partly in Germany and partly in their home from 1984, every year in Germany, approximately 25,000 countries. These immigrants are called foreign-edu - respondents in nearly 15,000 households are interviewed cated immigrants. For the foreign-educated immigrants, by Kantar Public Germany (Wagner et al. 2007). The data we split the total years of education into years studying provide information on all household members, con- abroad and years studying in Germany based on their age sisting of Germans living in the Old and New German at migration and adjust the quality of schooling accord- States, foreigners, and recent immigrants to Germany. ingly. We normalize the Hanushek and Kimko (2000) One of the advantages of this dataset is that it has rich school quality index to the German level by dividing the information at the individual level about education and school quality of country c by that of Germany as follows: labour market performance. This allows us to account for SQ different individual characteristics when we compare the q = SQ DE labour market performance of immigrants and natives. Immigrants are defined as people who were born out - This normalization of the index implies that the school side Germany or have at least one parent with a migra- quality in Germany is 1 and the school quality in other tion background (Bundesministerium für Bildung und countries could be smaller or larger than 1 depending Forschung 2019). We consider individuals who are on whether the quality of education in those countries is between 18 and 64 years old and are employed on either better or worse than that in Germany. a full- or regular part-time basis. We exclude people in education, retirement, civil or military service and self- 3.3 The Hanushek and Kimko (2000) school quality index employment because of their irregular employment The Hanushek and Kimko (2000) quality index is con - activities and unreliable wage information. After applying structed based on six voluntary international standard- these sample selection criteria, we are left with 224,867 ized tests in mathematics and science conducted between person–year observations (188,065 natives and 36,802 1965 and 1991. Altogether, they use information about 26 immigrants). 3.2 Variables We do not differentiate primary/secondary schooling quality from ter - The natural logarithm of hourly wages measures the tiary education quality (quality of vocational training and universities) in our financial rewards of a job, which may reflect the indi - adjustment process. This is in line with empirical findings showing a highly vidual productivity or the returns to skills and education positive correlation between primary/secondary education and higher edu- cation. Michaelowa (2007) argues that the primary and secondary education of an individual. Using the GSOEP data, we compute the systems of a country influence the knowledge and attitudes of individuals who logarithm of hourly wages from the self-reported gross enter higher education. In other words, without a pool of qualified second - monthly wages and working hours per week. We deflate ary graduates, one is unlikely to have a qualified pool of students available for higher education. Using PISA scores to compute primary/secondary educa- the wage variable to 2010 prices. To prevent the effects tion quality and two alternative university rankings for tertiary education quality, she finds a positive relationship between the primary/secondary edu - cation and tertiary education quality. This positive relationship remains strong after adding other control variables such as GDP, population, and enrolment This paper uses data from the Socio-Economic Panel (SOEP) for the years rates. 1984–2016, version 33, released in 2018, doi:https:// doi. org/ 10. 5684/ soep. v33. We assume that immigrants do not study in a third country. 8 Page 6 of 15 H. Le‑Quang , E. Vallizadeh Fig. 2 Hanushek and Kimko (2000) School quality index. Sources: Hanushek and Kimko (2000) separate test score series from different age groups, sub -3.4 Summary statistics fields and years to generate human quality indices for 39 Table 1 presents summary statistics for the full sample of countries participating at least once in these international natives and the two groups of immigrants under study. In tests and extrapolate to other 111 countries based on the foreign-educated immigrant group, we further split information provided in the National Assessment of Edu- by countries with a high quality of schooling and coun- cational Progress (NAEP) (Hanushek and Kimko 2000; tries with a low quality of schooling. Wößmann 2003). The resulting quality measure reflects Figure 3 shows that natives have, on average, slightly the weighted average of all test scores available for each higher wages than both groups of immigrants. This fig - country where the weights are the normalized inverse of ure also shows that the wage distribution of immigrants the country-specific standard error of each test, assum - has longer tails than that of natives, indicating that immi- ing that the higher the standard error is, the less accurate grants’ wages are more extreme than those of natives. the information it conveys (Wößmann 2003). This list Having a longer tail on the left-hand side (below 2.14) lacks some countries in Eastern Europe; consequently, by also means that there are many immigrants earning less matching the country-level data with the individual-level than the minimum wage in 2016. Table 1 shows that data, the number of countries decreases to 87. while the mean difference in wages between natives and Figure 2 displays the summary of the constructed German-educated immigrants is only 0.1 log points, the school quality index for different regions. The overall mean difference in wages between natives and foreign- mean is 51.28, and the median is 46.77 with a stand- educated immigrants from countries with a high quality ard deviation of 13.48. Countries in the Asian–Ocean- of schooling is 7.4 log points and that between natives ian region have the largest range of school quality: from and immigrants from countries with a low quality of 20.80 in India to 67.06 in Singapore. Generally speaking, schooling is 10.9 log points (equivalent to 1 and 1.12 countries in East Asia such as Japan, Hong Kong, Sin- euros per hour, respectively). gapore, China and Taiwan consistently have the highest While the years of unadjusted education for natives ranking throughout the period from 1965 to 1991 (the is higher than those for immigrants (12.48 and 11.36 maximum score is 67.06). In the bottom are countries in respectively), this result is reversed when we take the Latin America (Chile, El Salvador, Paraguay) and Africa years of quality-adjusted education into considera- (Algeria, Bahrain, Ghana). The country with lowest score tion (12.48 and 13.63, respectively). This result seems is Iran at 18.26 points. Germany has a slightly higher surprising at first; however, when we separate the for - score than the mean score of all countries (55.74 com- eign-educated immigrants into two sub-groups, those pared to 51.28). The minimum wage in Germany in 2016 was 8.50 euros. The returns to school‑quality‑adjusted education of immigrants in Germany Page 7 of 15 8 Table 1 Summary statistics. Source: SOEP (wave: 1984–2016), version 33, released in 2018 Variables Natives German‑ educated Foreign‑ educated immigrants immigrants Low quality High quality Mean Std. Dev Mean Std. Dev Mean Std. Dev Mean Std. Dev Log(hourly wages) 2.678 0.525 2.677 0.504 2.569 0.430 2.604 0.444 Years of education (unadjusted) 12.488 2.561 12.226 2.711 10.097 2.250 11.034 2.421 Years of education (adjusted) 12.488 2.561 12.226 2.711 9.967 2.442 15.762 3.690 Males 0.543 0.498 0.564 0.495 0.728 0.444 0.550 0.497 Married 0.668 0.470 0.555 0.496 0.887 0.316 0.807 0.394 Age 42.266 10.498 36.192 10.512 39.544 9.574 42.641 9.611 Years since migration – – 35.541 10.770 18.814 8.569 18.710 9.587 Year of arrival – – 1974 8.273 1984 11.921 1987 14.122 Experience with current firms 10.894 9.762 7.875 8.115 8.809 7.940 8.972 8.240 Good German language skills – – 0.988 0.106 0.624 0.484 0.777 0.416 Qualifications No vocational degrees 0.095 0.177 0.572 0.356 Vocational degrees 0.739 0.671 0.335 0.475 University degrees 0.166 0.152 0.093 0.169 Establishment size Up to 10 workers 0.168 0.167 0.114 0.149 Up to 20 0.132 0.112 0.099 0.107 Up to 200 0.273 0.244 0.275 0.301 Above 200 0.427 0.477 0.512 0.443 Industry Agriculture 0.055 0.039 0.094 0.071 Mining 0.148 0.185 0.303 0.238 Manufacturing 0.084 0.118 0.141 0.124 Construction 0.093 0.055 0.072 0.084 Trade 0.149 0.172 0.152 0.142 Transport and financial services 0.103 0.106 0.077 0.064 Public administration 0.145 0.123 0.058 0.097 Health and Education 0.178 0.163 0.081 0.145 Private household 0.045 0.039 0.022 0.035 Occupation Untrained/Semi‑trained workers 0.138 0.207 0.649 0.476 Trained workers 0.186 0.164 0.180 0.172 Trained employees with simple tasks 0.089 0.102 0.027 0.063 Qualified professionals 0.299 0.281 0.053 0.147 Highly qualified professionals 0.189 0.171 0.040 0.087 Civil servants 0.099 0.071 0.051 0.055 Person–year observations 188,065 14,002 4,992 17,808 Sample of individuals aged 18 to 64, having full‑time or regular part ‑time employment coming from countries with a low quality of schooling such as Japan, Hong Kong, Singapore, China and Tai- have only 9.967 years of quality-adjusted education, wan consistently have the highest ranking on the school while the others from countries with a high quality of quality index throughout the period from 1965 to 1991. schooling have almost 16 years of quality-adjusted edu- This result is also reflected in the highest attained qual - cation. The reason is that many countries in Western ifications that they possess, i.e., vocational training Europe such as Denmark, Finland, Austria, France, and and university graduation. For German natives, almost the United Kingdom as well as countries in East Asia three-quarters have vocational degrees, and only 9.5 8 Page 8 of 15 H. Le‑Quang , E. Vallizadeh To shed light on the sub-sample of immigrants in our dataset, Table 2 presents some key statistics such as the school quality index, the mean birthyear, the mean years of (unadjusted) education and the mean year of arrival of immigrants from the top 20 sending countries. The total number of observations from these top 20 coun- tries accounts for 91.46 percent of the total observations of the immigrants in our dataset (33,656 and 36,802, respectively). Most of these immigrants come from Europe, except for a quite substantial portion coming from Turkey (14.11 percent). In this sub-sample, Poland has the highest school quality index (64.37), and Turkey has the lowest school quality index (39.72). Recall that the Hanushek and Kimko (2000) school quality index Fig. 3 Density functions of natives and immigrants (log) hourly is constructed based on the international standardized wages. Kernel density wage estimation of natives and immigrants using Epanechnikov kernel function. Sources: SOEP (wave: 1984– tests conducted between 1965 and 1991, and the range of 2016), version 33, released in 2018 mean birthyear of immigrants in the sample is from 1955 (Croatia) to 1972 (Kosovo-Albania and Ukraine); this strengthens our argument that the Hanushek and Kimko (2000) index is better than PISA or PIACC scores in cap- percent do not have school-leaving certificates. Ger - turing the time during which immigrants in the sample man-educated immigrants follow the same pattern in actually went to school. terms of education qualifications with 67.1 percent hav - In addition, immigrants from France have the highest ing vocational degrees, but the portion of people hav- average years of (unadjusted) education (13.22 years), ing no school-leaving certificates is remarkably higher compared to Italy with only 9.78 years. Among these at approximately 18 percent. immigrants, the largest proportion is those from Ger- Despite having higher years of quality-adjusted edu- many with almost 31 percent of the total observations. In cation, immigrants from both groups in our sample are other words, they are either first-generation immigrants more likely to work in untrained or semi-trained occu- who have already been naturalized or second-generation pations (39.79 percent), especially foreign-educated immigrants (born in Germany having at least one par- immigrants (64.9 percent), while natives are more likely ent with a migration background). If immigrants indicate to work in qualified professional occupations (29.9 per - that Germany is their country of origin, they would not cent), which, as we argue, explains part of the wage gap be asked about their year of arrival but instead be catego- between natives and immigrants. rized in the group of “Born in Germany or immigrated As mentioned previously, the data in the survey do before 1950”. These immigrants also have the highest not allow us to specifically determine the country where mean years since migration (35.54 years); on the other immigrants obtained their education. We therefore make hand, immigrants from Eastern European countries such the assumption that immigrants who migrated to Ger- as Bulgaria, Romania and Ukraine migrated to Germany many before the age of six acquire education exclusively approximately 22 years ago, on average. in Germany (German-educated immigrants) and immi- grants who migrated to Germany after the age of six either acquire education exclusively in their home coun- 4 Empirical results tries or partly in Germany and partly in their home coun- In this section, we first present our findings from the OLS tries (foreign-educated immigrants) based on their age regressions for the groups of natives and immigrants in at migration and total years of education. Among these the sample. Our interest is to see how years of education foreign-educated immigrants, approximately 31 percent (after adjustment) affect immigrants’ wages in Germany. already migrated to Germany when they were at least 18 Second, we present the results of the Oaxaca–Blinder years old, 63.69 percent did so when they were at least 25, decomposition model, in which we show that using qual- and 80.80 percent did so when they were at least 30. If ity-adjusted education yields a better understanding of we assume that immigrants do not obtain any further ter- the native–immigrant wage gap in Germany. tiary education in Germany when they are older than 30 years old, there will be approximately 20 percent of for- 4.1 Returns to education: OLS estimation eign-educated immigrants who acquire their education Table 3 presents the OLS regression results for natives exclusively outside Germany. using unadjusted years of education. Column (1) presents The returns to school‑quality‑adjusted education of immigrants in Germany Page 9 of 15 8 Table 2 Summary Statistics—Top 20 sending countries. Source: SOEP (wave: 1984–2016), version 33, released in 2018 and Hanushek and Kimko (2000) Country of origin Person–year Percentage School quality Mean birthyear Mean years of Mean observations index education (unadjusted) year of arrival Germany 11,397 30.970 55.740 1970 12.440 <1950 Turkey 5193 14.110 39.720 1963 9.880 1980 Italy 2687 7.300 49.410 1958 9.780 1976 Poland 2360 6.410 64.370 1969 11.900 1992 Kazakhstan 1954 5.310 54.650 1971 11.260 1995 Russia 1831 4.980 54.650 1969 11.680 1996 Greece 1524 4.140 50.880 1956 9.910 1975 Croatia* 1065 2.890 53.970 1955 10.230 1975 Romania* 1063 2.890 62.800 1971 12.020 1998 Bosnia–Herzegovina* 841 2.290 53.970 1958 10.040 1980 Spain 830 2.260 51.920 1958 10.120 1975 Kosovo‑Albania* 506 1.370 51.280 1972 10.230 1992 Ex‑ Yugoslavia 465 1.260 53.970 1952 9.860 1975 Serbia* 441 1.200 53.970 1959 10.100 1979 Ukraine 312 0.850 54.650 1972 12.040 1998 Hungary 251 0.680 61.230 1966 13.180 1993 Bulgaria* 247 0.670 62.800 1968 12.030 1998 France 245 0.670 56.000 1964 13.220 1988 USA 235 0.640 46.770 1961 13.120 1983 United Kingdom 209 0.570 62.520 1961 12.710 1985 Sample of individuals aged 18 to 64, having full‑time or regular part ‑time employment. *indicates six countries for which we do not have any information about their school quality index, namely, Croatia, Romania, Bosnia–Herzegovina, Kosovo–Albania, Serbia and Bulgaria with total observations of 4163, which accounts for 11 percent of full immigrant sample. We impute missing values based on geographical proximity. Ex‑ Yugoslavia consists of Bosnia–Herzegovina, Croatia, Macedonia, Montenegro, Serbia and Slovenia; hence these countries take the school‑ quality index of the former Yugoslavia (i.e., 53.97). Romania and Bulgaria take the average value of Hungary and Poland (i.e., 62.80) the results of the restricted model where we only con- 5.3 percentage points (Column 2), which is lower than trol for years of education. Columns (2) and (3) gradually the 6.6 percentage points of natives. Second, when dis- add other control variables, where we take into account tinguishing between the German-educated and foreign- all observable characteristics that determine wages. The educated immigrant groups, we see that the return to results reveal that education has a positive and significant each year of education for German-educated immi- impact on earnings. In particular, one year of education grants is approximately 6.2 percentage points, while the is associated with an approximately 6.6 percentage point returns to each year of education for foreign-educated increase in hourly wages, ceteris paribus. Other variables immigrants are significantly lower. In particular, each such as experience, age, gender and marital status also year of education in Germany is associated with a 2.6 show expected impacts on wages. percentage point increase in earnings, and each year of Table 4 presents the OLS regression results, first for all education outside Germany is associated with only a 1.4 immigrants and then separately for German-educated percentage point increase in earnings, ceteris paribus. and foreign-educated immigrants. For foreign-educated These results highlight the low transferability of human immigrants, we divide their total years of education into capital and the significantly low returns to foreign educa - years of education abroad and years of education in Ger- tion. Furthermore, among immigrants, especially those many and adjust the years of foreign education using who partly conducted their education outside Germany, the Hanushek and Kimko (2000) school quality index. are additionally punished in terms of wages, even when Table 4 only shows the results of our preferred models we take their quality of schooling in consideration. with all control variables. First, the conditional returns Tables 3 and 4 strongly support the notion that there is a to education are positive and statistically significant. As substantial gap in the returns to education between natives expected, they are also lower than those of natives. For and immigrants. This gap is enlarged when we consider all immigrants, the return to one year of education is the school quality in immigrants’ home countries. 8 Page 10 of 15 H. Le‑Quang , E. Vallizadeh Table 3 OLS—Estimated returns to education—Natives. Source: Immigrants from the OECD and Eastern European/ SOEP (wave: 1984–2016), version 33, released in 2018 FSU countries have statistically significant returns to education. In particular, one additional year of educa- Dep. Var: Log(hourly wages) (1) (2) (3) tion in an OECD country or in a Eastern European/FSU Years of education 0.067*** 0.068*** 0.066*** country is associated with a 2.2 percentage point and a (0.001) (0.001) (0.001) 1.3 percentage point increase in the hourly wages in Ger- Experience 0.019*** many, respectively. These results indicate that proxim - (0.001) ity of educational systems between immigrants’ source (Experience‑squared)/100 − country and Germany could partly explain the difference 0.033*** in the returns to education among different groups of (0.002) immigrants. Our findings are in line with the empirical Male 0.210*** 0.170*** evidence on the impact of institutional and geographical (0.005) (0.005) proximity on economic growth improvement (Ahmad Age 0.040*** 0.026*** and Stephen 2012). In contrast, investment in education (0.001) (0.001) in Germany is associated with positive and significant (Age–squared)/100 − 0.038*** − returns across all country groups. In particular, immi- 0.027*** grants from East Europe/FSU benefit the most (a 3.1 per - (0.001) (0.001) centage point increase in hourly wages for each year of Married 0.031*** education in Germany), while the returns to education in (0.005) Germany is the lowest for immigrants from Turkey (only Constant 1.842*** 0.811*** 0.696*** 1.5 percentage point increase in hourly wages for each (0.015) (0.030) (0.031) year of education in Germany). Person‑ year observations 188,065 188,065 188,065 Similar to Basilio et al. (2017), we also find that education R‑squared 0.107 0.282 0.372 in Germany yields higher returns than education abroad for Dependent variable: Natural logarithm of hourly wages. Sample of natives aged all immigrant groups. This result holds true even for immi - 18 to 64, having full‑time or regular part ‑time employment. Clustered robust standard errors at the individual level are reported in parentheses. ***, ** and grants from high-income OECD countries where education * indicate significance at the 1%, 5% and 10% level, respectively. In addition to systems, industrial structures and technology are compara- the covariates shown in the table, models (2) and (3) control for federal state ble to those in Germany. Therefore, the quality of education fixed effects and survey year fixed effects. Model (3) additionally controls for occupation, establishment size and industry fixed effects in immigrants’ home countries is important, but everyone could benefit from extra education in Germany. On the one hand, German employers are more familiar with the aca- Recall that on average, immigrants have higher adjusted demic qualifications in Germany, and hence they reward years of education than natives, which means that the qual- German education more than foreign education. On the ity of schooling in their home countries must be higher other hand, the knowledge, experience and language skills than that in Germany. Hence, this result may first appear that immigrants receive during their education and train- surprising because we should expect that given a higher ing in Germany could also be more relevant for the German quality of schooling, immigrants should receive higher labour market than what they learned in their home coun- rewards in terms of wages. To address this puzzling result, tries. Not limited to using a similar approach to Basilio et al. we examine the heterogeneity of effects across immigrants’ (2017) in the heterogeneity analysis, our paper adjusts the country of origin. We argue that not only school quality but quality of schooling in the home countries of immigrants also the relative closeness to the German educational sys- to gain more plausible estimates of the returns to educa- tem matter for skill transferability. In doing so, we divide tion. Additionally, our paper uses a more updated version of the country of origin of foreign-educated immigrants into GSOEP (version 33) where the data span until 2016 (the for- five groups: OECD, Turkey, East Europe/former Soviet mer paper uses GSOEP data until 2013 only). With a wider Union (FSU), Ex-Yugoslavia, and Others. The results of the span of data, we are able to include and analyse the inflow OLS regression for each group are presented in Table 5. of new immigrants into Germany, especially after the surge of more than one million migrants and asylum seekers in OECD countries consist of Australia, Austria, Belgium, Canada, Denmark, 2015, which fundamentally changed the pool of immigrants Finland, France, Greece, the United Kingdom, Ireland, Israel, Italy, Japan, Benelux, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, the in Germany (Eurostat 2018). Netherlands and the USA. East Europe/former Soviet Union consists of Rus- sia, Ukraine, Belarus, Uzbekistan, Kazakhstan, Georgia, Lithuania, Azerbaijan, Moldavia, Latvia, Kyrgyzstan, Tajikistan, Armenia, Turkmenistan, Estonia, Bulgaria, Czech Republic, Hungary, Poland, Romania and Slovakia. Ex-Yugo- slavia consists of Bosnia and Herzegovina, Croatia, Macedonia, Montenegro, Serbia and Slovenia. The returns to school‑quality‑adjusted education of immigrants in Germany Page 11 of 15 8 Table 4 OLS—Estimated returns to education—Immigrants. Source: SOEP (wave: 1984–2016), version 33, released in 2018 Dep. Var: Log(hourly wages) All immigrants German‑ educated immigrants Foreign‑ educated immigrants (1) (2) (3) (4) (5) (6) Years of education 0.057*** 0.053*** 0.060*** 0.062*** (0.002) (0.002) (0.003) (0.003) Years of education in Germany 0.023*** 0.026*** (0.004) (0.003) Years of education abroad (adjusted) 0.002 0.014*** (0.002) (0.003) Experience 0.018*** 0.020*** 0.016*** (0.002) (0.003) (0.002) − − − (Experience‑squared)/100 0.030*** 0.030*** 0.029*** (0.001) (0.001) (0.001) Male 0.234*** 0.199*** 0.235*** 0.177*** 0.234*** 0.217*** (0.009) (0.010) (0.016) (0.018) (0.012) (0.012) Age 0.042*** 0.026*** 0.050*** 0.039*** 0.037*** 0.015*** (0.003) (0.003) (0.005) (0.014) (0.004) (0.004) − − − − − − (Age‑squared)/100 0.043*** 0.032*** 0.051*** 0.063*** 0.038*** 0.022*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Married 0.033*** 0.023 0.030** (0.010) (0.014) (0.013) Year since migration 0.002* 0.004 0.006*** (0.001) (0.014) (0.002) (Years since migration‑squared)/100 0.002 0.002 0.005 (0.003) (0.003) (0.006) Good German language skills 0.052*** 0.075 0.081*** (0.009) (0.047) (0.010) Constant 0.948*** 0.907*** 0.675*** 0.577*** 1.515*** 1.447*** (0.056) (0.059) (0.098) (0.111) (0.078) (0.076) Person–year observations 36,802 36,802 14,002 14,002 22,800 22,800 R‑squared 0.236 0.336 0.278 0.366 0.171 0.291 Dependent variable: Natural logarithm of hourly wages. Sample of immigrants aged 18 to 64, having full‑time or regular part ‑time employment. Clustered robust standard errors at the individual level are reported in parentheses. ***, ** and * indicate significance at the 1%, 5% and 10% level, respectively. In addition to the covariates shown in the table, all models control for federal state fixed effects, survey year fixed effects, and country of origin fixed effects. Models (2), (4) and (6) additionally control for occupation, establishment size and industry fixed effects 4.2 Na tive–immigrant wage gap: Oaxaca–blinder feature of our approach is that we are able to exam- decomposition ine the determining factor behind the lower returns to In this section, we examine the contribution of school foreign education and the imperfect transferability of quality to the native–immigrant wage gap using the Oax- human capital of immigrants in the host country. More aca–Blinder decomposition method. Table 6 reports important, we show that lower school quality is one of the results of this decomposition. Overall, the observed the major reasons why immigrants earn significantly less wage gap between natives and immigrants is − 5.1 log than natives in Germany. In particular, when we adjust points. Existing studies have provided several reasons the years of education by the school quality measure, for this wage gap, such as lower returns to human capi- the share of the unexplained component of the native– tal obtained abroad, imperfect transferability of skills, immigrant wage gap declines (from – 0.183 to – 0.166 and labour market discrimination that immigrants usu- in the restricted model and from − 0.109 to − 0.087 in ally experience in the host country (Coulombe et al. the unrestricted model). This decline in the unexplained 2014; Bartolucci 2014; Aldashev et al. 2012). The key component is equivalent to 9–20 percent. In other words, 8 Page 12 of 15 H. Le‑Quang , E. Vallizadeh Table 5 Heterogeneity analysis by regions of origin—Foreign‑ educated immigrants. Source: SOEP (wave: 1984–2016), version 33, released in 2018 Dep. Var: Log(hourly wages) OECD Turkey East Europe/FSU Ex‑ Yugoslavia Others Years of education in Germany 0.029*** 0.015** 0.031*** 0.017* 0.024** (0.008) (0.007) (0.005) (0.009) (0.012) Years of education abroad (adjusted) 0.022** 0.007 0.013*** 0.010 0.004 (0.007) (0.008) (0.004) (0.006) (0.010) Experience 0.012*** 0.019*** 0.021*** 0.0087** 0.014** (0.004) (0.003) (0.003) (0.004) (0.004) − − − − − (Experience‑squared)/100 0.017 0.039*** 0.032*** 0.015*** 0.014*** (0.001) (0.001) (0.001) (0.001) (0.001) Male 0.260*** 0.251*** 0.181*** 0.222*** 0.218*** (0.026) (0.025) (0.019) (0.030) (0.051) Age 0.016* 0.011 0.014** 0.021** 0.028* (0.009) (0.008) (0.006) (0.009) (0.015) − − − − − (Age‑squared)/100 0.019 0.017** 0.023*** 0.033*** 0.027 (0.001) (0.001) (0.001) (0.001) (0.001) Married 0.028 0.048* 0.020 0.018 0.092** (0.027) (0.027) (0.019) (0.031) (0.042) Year since migration 0.003 0.004 0.004 0.012* 0.007 (0.005) (0.005) (0.003) (0.006) (0.008) − − − (Years since migration‑squared)/100 0.009 0.005 0.014 0.003 0.004 (0.001) (0.001) (0.001) (0.001) (0.001) Good German language skills 0.067*** 0.056*** 0.123*** 0.007 0.036 (0.022) (0.020) (0.016) (0.027) (0.037) Constant 1.415*** 1.693*** 1.171*** 1.478*** 1.491*** (0.202) (0.140) (0.147) (0.202) (0.304) Other controls Yes Yes Yes Yes Yes Person–year observations 5649 4158 8529 2707 1757 R‑squared 0.300 0.351 0.292 0.238 0.404 Dependent variable: Natural logarithm of hourly wages. Sample of foreign‑ educated immigrants aged 18 to 64, having full‑time or regular part ‑time employment. Clustered robust standard errors at the individual level are reported in parentheses. ***, ** and * indicate significance at the 1%, 5% and 10% level, respectively. In addition to the covariates shown in the table, all models control for occupation, establishment size, industry fixed effects, federal state fixed effects, survey year fixed effects, and country of origin fixed effects using quality-adjusted education we are able to account finding confirms the point made by Bonikowska et al. for differences in the unexplained part. This highlights (2008) that what is frequently attributed to labour market the importance of differences in educational systems in discrimination may simply reflect the lower human capi - explaining the native–immigrant wage gap and provides tal quality of immigrants in comparison to natives. a rationale behind the unexplained component of decom- position methods, which is often referred to as “labour 5 Conclusions market discrimination” in the literature. In recent decades, the pattern of immigration has sub- In general, immigrants receive lower returns to edu- stantially shifted towards better skilled, mostly Euro- cation than natives in the German labour market. These pean immigrants in Germany (Dustmann et al. 2012). returns are further reduced if their education was However, the wage assimilation of immigrants has pro- obtained outside Germany. Similar to Coulombe et al. gressed at a slow pace, leading to a substantial difference (2014), we find that a lower quality of education reduces in earnings between immigrants and natives. The goal of the endowment advantage that immigrants earn during this paper is to understand the determinants of the wage these years of education. We also observe that by control- assimilation of immigrant workers and to what extent the ling for quality-adjusted education, the wage gap between international transferability of human capital explains natives and immigrants widens, but the share of the their low initial earnings compared to natives. unexplained component in this wage gap declines. This The returns to school‑quality‑adjusted education of immigrants in Germany Page 13 of 15 8 Table 6 Oaxaca–Blinder Decomposition of immigrant wage gaps. Source: SOEP (wave: 1984–2016), version 33, released in 2018 Unadjusted years of education Adjusted years of education Restricted model Unrestricted model Restricted model Unrestricted model − − − − Observed gap 0.051*** 0.051*** 0.051*** 0.051*** (0.007) (0.005) (0.007) (0.005) Explained gap − − − − Education 0.070*** 0.072*** 0.073*** 0.074*** (0.003) (0.003) (0.004) (0.003) − − Experience 0.049*** 0.049*** (0.003) (0.003) Experience‑squared 0.025*** 0.025*** (0.002) (0.002) Male 0.007*** 0.007*** (0.001) (0.001) Married 0.002*** 0.002*** (0.000) (0.000) − − Age 0.064*** 0.064*** (0.005) (0.005) (Age‑squared)/100 0.061*** 0.061*** (0.005) (0.005) − − Year since migration 0.043*** 0.043*** (0.015) (0.015) (Years since migration‑squared)/100 0.006 0.006 (0.015) (0.015) − − Good German language skills 0.007*** 0.007*** (0.001) (0.001) Unexplained gap − − − − Education 0.183*** 0.109*** 0.166*** 0.087*** (0.033) (0.027) (0.033) (0.029) − − Experience 0.013 0.015 (0.015) (0.015) Experience‑squared 0.007 0.007 (0.009) (0.009) Male 0.004 0.007 (0.006) (0.006) Married 0.005 0.006 (0.007) (0.008) Age 0.006 0.023 (0.128) (0.128) − − (Age‑squared)/100 0.061 0.058 (0.069) (0.070) Year since migration 0.136*** 0.120*** (0.042) (0.042) − − (Years since migration‑squared)/100 0.026 0.022 (0.031) (0.031) Good German language skills 0.062*** 0.056*** (0.010) (0.010) Sample of individuals aged 18 to 64, having full‑time or regular part ‑time employment. Clustered robust standard errors at the individual level are reported in parentheses. ***, ** and * indicate significance at the 1%, 5% and 10% level, respectively. In addition to the covariates shown in the table, columns (2) and (4) also control for occupation, establishment size, industry fixed effects, federal state fixed effects, survey year fixed effects, and country of origin fixed effects 8 Page 14 of 15 H. Le‑Quang , E. Vallizadeh Acknowledgements Understanding the determinants of the native–immi- We would like to thank Silke Anger, Elizabeth Cascio, Bob Hart, Stephen grant wage gap has important policy implications for at Machin and participants in the International Workshop on Applied Economics least two reasons. First, slow wage assimilation rates may of Education 2017 in Catanzaro and the VII Workshop on Economics of Educa‑ tion in 2017 in Barcelona and especially two anonymous reviewers for helpful have adverse impacts on the long-term integration pro- comments and suggestions. Huy Le Quang also gratefully acknowledges cess of immigrants by inducing adverse incentives regard- the financial support from the Graduate School (GradAB) at the Institute for ing human capital investment by immigrants in the host Employment Research. All remaining errors are our own. The views, opinions, findings and conclusions expressed in this article are strictly those of the country. Second, a large wage gap between immigrants authors. and natives may also reflect significant labour market barriers for immigrants in the host country. Author contributions We equally contributed to completing this research. In particular, HLQ was We propose an empirical strategy that accounts for the responsible for idea generation, data cleaning, data analysis and paper draft‑ quality of schooling in immigrants’ source countries. Our ing, and EV was responsible for defining the scope of study, proofreading and empirical findings highlight several new insights. First, editing the last version of the paper. Both authors read and approved the final manuscript. the quality of the schooling system is an important pre- dictor of the speed of wage assimilation for immigrants. Funding We show that the earnings of immigrants who obtained We did not receive any funding specifically related to this research. their education in countries with high-quality schooling Availability of data and materials systems catch up about eight years faster to natives than The datasets used during the current study are available from the correspond‑ immigrants who received their education in countries ing author on reasonable request. with low-quality schooling systems. Second, our findings indicate that one potential reason Declarations for the low international transferability of human capital Consent for publication may be driving by the poor quality of school systems in We confirm that this work is original and has not been published elsewhere, immigrants’ home countries. On average, the return to nor is it currently under consideration for publication elsewhere. We declare that all authors have approved the manuscript for submission. education for foreign-educated immigrants is approxi- mately 3.6 percentage points lower compared than the Competing interests returns to education of German-educated immigrants We have no conflicts of interest. (2.6 and 6.2 percentage points, respectively). However, Author details when accounting for school quality, the return to foreign 1 2 Otto–Friedrich University of Bamberg, Bamberg, Germany. I nstitute education decreases further to 1.4 percentage points. for Employment Research (IAB), Nuremberg, Germany. These results indicate that further educational invest - Received: 24 June 2021 Accepted: 21 June 2022 ment in the host country yields significant returns for immigrants. Third, using the Oaxaca–Blinder decomposition method, our empirical findings reveal that the school References quality index explains a remarkable part of the unex- Ahmad, M., Stephen, G.H.: Economic growth and convergence: Do institu‑ tional proximity and spillovers matter? J. Policy Model 39(6), 1065–1085 plained component of the native–immigrant wage gap (2017) decomposition: school-quality-adjusted foreign educa- Aldashev, A., Gernandt, J., Thomsen, S.L.: The immigrant‑native wage gap in tion reduces the unexplained component by approxi- Germany. J. Econ. Stat. 232(5), 490–517 (2012) Aleksynska, M., Tritah, A.: Occupation‑ education mismatch of immigrant work‑ mately 20 percent. This finding highlights that school ers in Europe: Context and policies. Econ. Educ. Rev. 36, 229–244 (2013) quality in the home country of immigrants can substan- Amann, R.A., Klein, T.J.: Returns to type or tenure? Journal of the Royal Statisti‑ tially explain differences in unobservables. cal Society. Seri. A Stat. Soc. 175(1), 153–166 (2012) Bartolucci, C.: Understanding the native‑immigrant wage gap using matched To summarize, our findings highlight that the integra - employer employee data: Evidence from Germany. ILR Rev. 67(4), tion of immigrants into the host society depends largely 1166–1202 (2014) on institutional features and educational policy in their Basilio, L., Bauer, T.K., Kramer, A.: Transferability of human capital and immigrant assimilation: An analysis for Germany. Labour 31(3), 245–264 (2017) home countries. Particularly, the quality of schooling in Betts, J.R., Lofstrom, M.: The educational attainment of immigrants: Trends and their home countries is a strong predictor of their perfor- implications. NBER Chapters, in: Issues in the Economics of Immigration, mance in the host country’s labour market. Furthermore, 51–116 (2000) Blinder, A.S.: Wage discrimination: Reduced form and structural estimates. J. our analysis shows that there is substantial heterogene- Hum. Resour. 8(4), 436–455 (1973) ity in the quality of schooling across immigrants’ home Bonikowska, A., Green, D.A., Riddell, W.C.: Literacy and the Labour Market: Cog‑ countries. Taking these heterogeneities into considera- nitive Skills and Immigrant Earnings. Statistics Canada Ottawa, Canada (2008) tion improves our knowledge about the determinants of Bratsberg, B., Ragan, J.F., Jr.: The impact of host‑ country schooling on earnings: the wage assimilation of immigrants and why they ini- A study of male immigrants in the United States. J. Hum. Resour. 37(1), tially earn less than natives. 63–105 (2002) The returns to school‑quality‑adjusted education of immigrants in Germany Page 15 of 15 8 Bratsberg, B., Terrell, D.: School quality and returns to education of US immi‑ grants. Econ. Inq. 40(2), 177–198 (2002) Bundesministerium für Bildung und Forschung: Weiterbildungsverhalten in Deutschland 2018. Ergebnisse des Adult Education Survey, AES‑ Trend‑ bericht (2019) Burdett, K.: A theory of employee job search and quit rates. Am. Econ. Rev 68, 212–220 (1978) Card, D., Krueger, A.B.: Does school quality matter? Returns to education and the characteristics of public schools in the United States. J. Polit. Econ. 100(1), 1–40 (1992) Chiswick, B.R., Miller, P.W.: Why is the payoff to schooling smaller for immi‑ grants? Labour Econ. 15(6), 1317–1340 (2008) Chiswick, B.R., Miller, P.W.: The international transferability of immigrants’ human capital. Econ. Educ. Rev. 28(2), 162–169 (2009) Chiswick, B.R., Miller, P.W.: The effects of school quality in the origin on the payoff to schooling for immigrants. Migration and Culture, 67–103 (2010) Coulombe, S., Grenier, G., Nadeau, S.: Human capital quality and the immigrant wage gap. IZA J. Migr. 3(1), 14 (2014) de Oliveira, M.M., Santos, M.C., Kiker, B.F.: The role of human capital and techno‑ logical change in overeducation. Econ. Educ. Rev. 19(2), 199–206 (2000) Dustmann, C., Frattini, T., Lanzara, G.: Educational achievement of second‑ generation immigrants: An international comparison. Econ. Policy 27(69), 143–185 (2012) Eurostat: Einwanderung Nach Alter und Geschlecht. http:// ec. europa. eu/ euros tat/ de/ data/ datab ase Accessed 2022‑05‑12 Fortin, N., Lemieux, T., Torres, J.: Foreign human capital and the earnings gap between immigrants and Canadian‑born workers. Labour Econ. 41, 104–119 (2016) Friedberg, R.M.: You can’t take it with you? Immigrant assimilation and the portability of human capital. J. Labour Econ. 18(2), 221–251 (2000) Hanushek, E.A.: The economics of school quality. German Econ. Rev. 6(3), 269–286 (2005) Hanushek, E.A., Kimko, D.D.: Schooling, labour‑force quality, and the growth of nations. Am. Econ. Rev. 90(5), 1184–1208 (2000) Hill, P.W., Rowe, K.J.: Multilevel modelling in school effectiveness research. Sch. E. S ff ch. Improv. 7(1), 1–34 (1996) Jovanovic, B.: Job matching and the theory of turnover. J. Polit. Econ. 87(5), 972–990 (1979) Michaelowa, K.: The impact of primary and secondary education on higher education quality. Qual. Assur. Educ. 15(2), 215–236 (2007) Muysken, J., Vallizadeh, E., Ziesemer, T.: Migration, unemployment, and skill downgrading. Scand. J. Econ. 117(2), 403–451 (2015) Oaxaca, R.: Male‑female wage differentials in urban labour markets. Int. Econ. Rev. 14(3), 693–709 (1973) OECD: Program for International Student Assessment. https:// www. oecd. org/ pisa/. Accessed 12 May 2022 OECD: Program for the International Assessment of Adult Competencies. https:// www. oecd. org/ skills/ piaac/ Accessed 12 May 2022 Raaum, O.: Returns to education and Norwegian language skills among immi‑ grants in Norway. SNF‑ Oslo, Oslo (1998) Rohrbach‑Schmidt, D., Tiemann, M.: Limited transferability of human capital across countries–the case of workers with foreign qualifications in Ger ‑ many. Federal Institute for Vocational Education and Training Discussion Paper (2016) Sweetman, A.: Immigrant source country educational quality and Canadian labour market outcomes. Analytical Studies, Statistics Canada (2004) Wagner, G.G., Frick, J.R., Schupp, J.: The German Socio‑Economic Panel Study (SOEP): Scope, Evolution and Enhancements. Schmollers Jahrbuch : Journal of Applied Social Science Studies / Zeitschrift für Wirtschafts‑ und Sozialwissenschaften 127(1), 139–169 (2007) Wößmann, L.: Specifying human capital. J. Econ. Surv. 17(3), 239–270 (2003) Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations.
Journal for Labour Market Research – Springer Journals
Published: Dec 1, 2022
Keywords: School quality; Wage gap decomposition; Immigration; Earnings; I21; I26; J31; J61
Access the full text.
Sign up today, get DeepDyve free for 14 days.