Aycan Çelikaksoy
Eskil Wadensjö
Catching Up? Country Studies on Intergenerational Mobility and Children of Immigrants
Report
Table of contents
Share
Facebook
Twitter
LinkedIn
Chapter 5. Sweden: Intergenerational mobility patterns in immigrant and native families
Abstract
This chapter investigates the labour market situation of Swedish native-born sons and daughters in immigrant and native families with regard to their parents’ education, as well as intergenerational educational mobility patterns for these families. The latter are compared in order to ascertain whether the roles played by institutions and family background vary across these two groups. Rather than focussing solely on father-son pairs, the chapter looks at all family combinations, including mothers and daughters and mixed couples. Since there can be differences across countries of origin for those families with an immigrant background, all country groups of interest are analysed separately. Transmission patterns are also investigated separately for different household types with regard to parental composition.
Main findings
There is a significant and positive relationship between the educational attainment of parents and their children in Sweden. This relationship holds for both immigrants’ and natives’ families. Yet, the strength of the relationship is weaker for immigrant families – especially for those with two immigrant parents – suggesting a higher degree of educational mobility for those with an immigrant background compared to those with a native background – a common finding for several countries.
Comparing immigrant and native-born families, Swedish studies consistently find a higher degree of upward mobility for immigrant families with less well educated parents and a higher degree of persistence at the higher end of the parental educational distribution.
The composition of the parents in terms of origin countries plays a role in the transmission of education. There is a higher degree of educational mobility for families in which both parents are foreign-born compared to those with a mixed or a native background.
There is a clear pattern of gender segregation in terms of educational attainment, regardless of maternal education and migration background of the parents. Results indicate a higher upward educational mobility for daughters than for sons. In both immigrants’ and natives’ families, daughters attain a higher level of education and indeed the highest level of education in the family. In fact, daughters of low‑educated mothers are over ten percentage points more likely than sons to finish a long tertiary degree, for both children of immigrants and children of native-born.
Daughters from Iran and Morocco with two foreign-born parents stand out in their upward mobility and indeed, there is no significant relationship between their education and that of their mothers. In contrast, low levels of the mothers’ education especially seem to negatively impact on the sons’ educational attainment in immigrant families.
Daughters’ employment rates are lower than those of sons in both immigrant and native families, but increase with the level of education of the mother, with the biggest jump observed for daughters of immigrants whose mother has obtained high school (compared with at most compulsory education). Among all groups, daughters with an immigrant background, display the lowest level of employment, especially when their mothers have only a compulsory level of education.
Descriptive figures show that maternal education is not associated with differences in the employment rates of sons with native-born parents, in contrast to immigrant parents. However, the occupations of both sons and daughters of mothers – both immigrant and native-born – are correlated with the mother’s level of education.
Introduction
The intergenerational transmission of a wide range of characteristics from parents to offspring has long been discussed in biology, medicine, psychology, sociology and economics. In social sciences the interest in these transmissions is generated by the objective of guiding social and economic policy to counteract inequality of opportunity. Parents with greater resources are able to invest more in their children. These investments take different forms and can be direct or indirect as well as monetary and non-monetary. Children in families of high socio-economic status (SES) have access to high-quality child care, early childhood education, certain beneficial social network and neighbourhood characteristics, high-quality public schools or private schools, and universities. In addition, these children have greater access to books and technology in the form of laptops, tablets, etc., as well as certain social skills. Also, in interacting with their children, highly educated parents transmit to them a different set of cognitive and non‑cognitive attributes, such as ambition. These are only a few of the pathways that throw into sharp relief the contrasting situation of children from low SES families. Theirs is an environment of restricted opportunities, with the parents transmitting the inequality they themselves experienced.
Positive assortative matching of single individuals via a set of characteristics, and transmission of these characteristics from parent to child, can be seen as a major source of inequality in society (Smits, Ultee and Lammers, 1998; Fernandez, Guner and Knowles, 2005). The reproduction of social and economic disadvantage from one generation to the next has been studied in relation to several outcomes, such as education, social networks, social class, socio-economic status, occupational status and income. In terms of economic inequality, the intergenerational persistence of education and income levels is an important indicator of the extent of transmission of economic disadvantage across families. Several pathways transmitting disadvantage, suggested through contrast above, can be counteracted and compensated by public policies such as universal provision of high-quality child care, early childhood education and high-quality public schools. Such policies can provide equal opportunities through government investments in children’s human capital (Solon, 2002). So it is that several cross-country studies have focused on ongoing intergenerational economic disadvantage and the degree of economic mobility across generations in order to investigate differences across societies.
Intergenerational transmission in the case of ethnic minorities has been discussed in the context of maintaining group-related ethnic, religious or cultural traits over generations, which leads to diverging and distinct group characteristics within a country. Some of these studies include analyses of intergenerational cultural transmission; the majority find evidence that parental inputs significantly affect individual outcomes (Bisin and Verdier, 2000 and 2000). Several characteristics are acquired by children through an imitation-and-adoption process depending on the socialisation actions of the parents and the environment in which the children live. How these characteristics are valued and transformed into material payoffs in the working depends also on the structure of the relevant labour market. However, regardless of the pathways, origins and mechanisms of transmission of disadvantage or advantage, public policies that aim at equality of opportunity will lead to social cohesion and economic efficiency. At the same time, actually identifying the specific pathways of these transmissions would lead to effective and efficient implementation of the policies. Education is an important determinant of outcomes throughout the life course; reproduction of educational inequality from the parents’ to the child’s generation will have major social as well as economic consequences. This chapter investigates the labour market situation of native-born sons and daughters in immigrant and native families with regard to their parents’ education, as well as intergenerational educational mobility patterns for these families. The focus is not only on parent-child pairs, but also on family composition.
Literature overview
Intergenerational mobility
One way of studying intergenerational mobility in terms of social class is to create a socio-economic index based on occupations that ranks occupational levels and then correlates this index between parents and their offspring. This method has been used for cross-country analyses (Ganzeboom and Treiman, 1996). Other methods have also been employed, such as aggregating occupations according to the employment conditions experienced. In a comparison of three countries, Sweden is found to have greater class mobility than France or England (Erikson, Goldthorpe and Portocarero, 1979). Modelling social class mobility by using two-way tables to measure the moves between classes has been proved to be difficult, since structural shifts across generations will increase the appearance of absolute mobility by forcing families away from the diagonal. Thus measures of relative mobility have been utilised – such as log-linear models, which do not vary with compositional changes across generations (Erikson and Goldthorpe, 1992).
The recently increasing interest in intergenerational mobility and cross-country comparisons in economics literature is mainly due to data availability. These studies focus on earnings as the main variable, although doing so has its drawbacks where non-labour income is not acknowledged. In addition, those without paid employment are neglected, and majority of the literature centres on the relationship between fathers’ and sons’ earnings, which excludes females from the analyses. This literature has shown the importance of parental background and the persistence of disadvantage from parent to child, especially for the United States. The studies have mainly utilised intergenerational earnings (IGE), a common measure of the average persistence of earnings from the parents’ generation to the offspring’s generation. The importance of using long-run earnings for both parents and children in the income-generating ages is emphasised in these studies to avoid underestimating the IGE (Solon, 1992, 2002; Zimmerman, 1992). Early studies show that Nordic countries are characterised by significantly higher intergenerational income mobility than the United States or the United Kingdom (Björklund and Jäntti, 1997; Österbacka, 2001; Solon, 2002). Some of these studies have employed different data structures as well as different methodologies, making it difficult to draw solid conclusions. However, later studies using the same methods (intergenerational earnings elasticity, correlation coefficients and quintile mobility matrices but with standardised data structures) have also found that the Nordic countries have higher intergenerational income mobility than the United Kingdom and the United States, and that the latter has the lowest (Jäntti et al., 2006). Even though income remains substantially the same over the generations in all countries, there is a significant difference between the Nordic countries and the United States and the United Kingdom. Similar to the trends in income, in Nordic countries generational education levels are found to persist significantly less than in non-Nordic nations (Hertz et al., 2007). Universal access to human capital formation – including early child care and early childhood education – and redistributive policies are two of the factors that reduce ongoing disadvantage from one generation to the next. Equalised standards of living and expanded opportunities for further education have diminished the influence of social origin on education over time. This trend toward equalisation in Sweden has been documented (Erikson and Jonsson, 1996); the question is whether it influences all groups in a similar way – including those with an immigrant background – in the Swedish labour market, in the educational system, and in Scandinavian societies in general.
Intergenerational mobility of individuals with an immigrant background
Compared to reports on intergenerational mobility in earnings and education the literature on those with an immigrant background is relatively more recent. Both sets of studies focus on father-son transmission, especially in the case of earnings but also in education. Borjas (1992) and Card, DiNardo and Estes (2000) find that the transmission of earnings from immigrant fathers to sons in the United States is slightly weaker than for the country’s overall population. Aydemir, Chen and Corak (2009) finds similar patterns in Canada for immigrant and native fathers and their sons but a weaker association for fathers and daughters in terms of earnings. On the other hand, Dustmann (2005) finds a stronger link among immigrant fathers and their sons compared to the general population. The results are mixed in the case of Sweden. Österberg (2000) finds similar results for immigrants and natives, while Hammarstedt (2008) and Hammarstedt and Palme (2012) report a stronger relationship between immigrant fathers and their sons compared to native fathers and their sons. This would suggest lower mobility among immigrant families compared to native families. Children born into immigrant families would thus seem to have fewer opportunities in the labour market and be shackled to the characteristics of their family background to a greater extent than those born to native families, which violates norms of equal opportunities. However, to answer whether this is due to greater persistence of educational attainment among immigrant families or whether it is due to, for example, discrimination in the labour market requires further research.
There are additional challenges in fully addressing earning mobility patterns in the case of immigrant families. Some of these are related to measuring lifetime earnings, where pre‑immigration earnings history is not available. Low female labour force participation for immigrant groups is another concern in terms of measuring mother-daughter transmissions. In addition, due to imperfect transferability of the skills of immigrants, parents’ earnings do not necessarily reflect their full earnings potential in the destination country. There is evidence from several countries that immigrants are overeducated in their jobs. Discrimination too is a very important factor that could explain stronger ongoing disadvantage in the labour market among immigrant families. Thus, analysing educational mobility patterns has advantages over those of earnings as it is a reliable measure relatively early in life. Education has also been found to be a good proxy for overall well-being, although the gaps across employees in relation to the payoff for educational attainment are a major source of dissatisfaction among employees (Oreopoulos and Salvanes, 2011).
Unlike the literature on earnings mobility for immigrant families, that on the transmission of education is less contradictory, since the vast majority of studies find a weaker relationship among immigrant fathers and their sons compared to native families for Canada and Germany as well as Sweden (Aydemir, Chen and Corak, 2009; Dustmann, 2005; Eriksson, 2006; Borjas, 1992; Card, DiNardo and Estes, 2000). This suggests weaker ties between fathers’ and sons’ educational attainments for immigrant families compared to native families, which in turn may imply socio-economic integration. Few studies have analysed mother-daughter educational persistence, or included daughters in analyses of immigrant families (Gang and Zimmerman, 2000; Aydemir, Chen and Corak, 2009; Niknami, 2012). There can of course be differences across family members as well as by country of origin, and so all possible links will be included in the following analysis.
Immigration to Sweden
Looking at just two generations, approximately 30% of Sweden’s population has a foreign background. Within the population of those with a foreign background, the proportion of those foreign-born is 57%; 17% are born in Sweden to two foreign-born parents; and 25% are born in Sweden to one foreign-born and one native-born parent (Çelikaksoy, 2016). There have been three main sources of immigration to Sweden. The first concerns migration from the Nordic countries during and after the Second World War. With the agreement signed in 1954, a common Nordic labour market was formalised, although it had already existed in practice prior to the signing of the agreement (Wadensjö, 2012). The main purposes of the agreement were maintaining full employment and achieving balanced regional development. From this period onward labour migration, primarily from Finland, continued. The second source concerned labour migrants from southern and eastern European countries in the 1950s and 1960s, recruited to work in the manufacturing sector, booming at the time. Thirdly, after the early 1970s labour migration became more restrictive, and refugee migration and family (re)unification became the largest sources of migration to Sweden. Refugees came from Estonia in 1944; Hungary in the late 1950s; Czechoslovakia and Poland in the late 1960s; Latin America, the Middle East and Africa in the 1970s; Iran in the 1980s; Yugoslavia (mainly Bosnia and Herzegovina) in the 1990s; and Iraq in the early 2000s. In 2005, the five largest immigrant groups in Sweden originated from Finland, Iraq, Yugoslavia, Iran and Bosnia and Herzegovina.
Sweden did not apply any transitional rules for immigrants coming from the ten new Member States with the EU enlargement in 2004. Immigration from these countries also increased, especially from Poland, Hungary and Baltic states. In 2014, the five largest groups were from Finland, Iraq, Poland, Iran and Yugoslavia (Çelikaksoy, 2016). However, in the 2016 statistics, Syria was among the five largest immigrant groups due to the current Syrian refugee crisis, the others being Finland, Iraq, Poland and Iran. Gender, age and civil status compositions vary across types of migration and countries of origin. Sweden has also been receiving the largest number of asylum applications by unaccompanied minors compared to all EU countries (Çelikaksoy and Wadensjö, 2015, 2017). Previous studies on refugee children arriving without their parents, where the majority were from Afghanistan, show that a higher proportion of this group remains in education at each age and continues in the educational system for a relatively longer period compared to refugee children who arrived in the Netherlands and Germany with their families (Crul et al., 2016; Çelikaksoy and Wadensjö, 2015). Later streaming and a relatively more flexible system in Sweden as, well as the wide availability of adult education compared to other countries, are arguably some of the reasons for the differences observed (Crul et al., 2016). Refugee students in Sweden have lower levels of compulsory school achievement compared to native students, although this difference disappears when the parental socio-economic background is taken into account in addition to neighbourhood of residence (Grönqvist and Niknami, 2017). Thus, investigating mobility patterns and the mechanisms behind these patterns is crucial.
Social policy and institutions in Sweden
Social policy and institutions underwent major changes in the country in the 1960s and 1970s; this is the period during which the sons and daughters of the immigrants who studied here grew up in Sweden. The educational system expanded quickly. Earlier, many young people left for the labour market after completing just primary school. Now, an increasing share continues to secondary school and many more than before also continue to higher education. Earlier, boys continued to secondary school and higher education in greater numbers than girls. During these two decades the situation changed – the girls began to continue their education for more years than the boys. Schools are publicly funded in Sweden. Tertiary education is also free of charge. In addition, student aid – which is income-tested – is available for a maximum period of six academic years; the aid is partly a grant and partly a loan. In addition, several studies point to the importance of early child care and early childhood education as a means of counteracting ongoing educational disadvantage within families. These institutions are crucial in terms of early childhood socialisation, cognitive development and language skills. Child care in Sweden has been accorded high priority. Universal high-quality child care is provided, financed out of public funds and organised mainly by municipalities. The pedagogical dimension of early child care has been extremely important. What guarantees the quality are well-educated personnel with a high degree of pedagogical competence, and a pedagogical culture of preschool education that has developed over a lengthy period and that continues to develop in relation to the current needs of society.
In the first decades after the Second World War, most married women were housewives. Female participation in the labour force was low, certainly much lower than for men. That changed to a very large extent in the decades of interest for this study, the 1960s and the 1970s. Factors driving that change were a large expansion of inexpensive child care organised, as mentioned above, by the municipalities and a new income tax system with separate taxation of the spouses. Expansion of the public sector (education, health care) at the same time led to a demand for labour in occupations that mainly had a female workforce.
Data
The data come from the database Stativ, which is administered by Statistics Sweden (SCB). The high-quality register data in Sweden, where every person has a record that is a by-product of registers held for administrative purposes. The population registry is administered by the Swedish Tax Agency. The main sources of the data used are education registers and multigenerational registers, which make it possible to construct parent-child pairs. Also included is information originally collected by the Migration Board. Personal identification numbers are anonymised and cannot be observed by the researcher. The dataset at hand covers the entire population of native-born individuals between ages 16 and 65 with at least one foreign-born parent, as well as the entire population of foreign-born individuals who lived in Sweden at some point during 2003‑12. The birth cohort chosen is 1958-78 for the native-born with an immigrant background; these individuals were between the ages of 25 and 54 during the observation period. Their parents immigrated to Sweden during the 1960s and 70s, which coincides with the main labour migration period. Thus the countries of origin chosen correspond to this type of immigration during this period. Only for some countries is the exact country of origin noted; for the rest there is information on groups of countries, due to an ethical measure SCB uses to avoid dissemination of too detailed information. Thus, parents’ countries of origin included in the analyses are the former Yugoslavia, Turkey, Iran, Morocco and the 13 countries that joined the EU after the first 15: Cyprus, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia, Slovenia, Rumania, Bulgaria and Croatia. Since educational attainment is the key measure for the analyses in this paper, the sample is restricted to individuals whose parents have available information on educational attainment. Excluding missing information on maternal education (6% of the sample) and paternal education (16%), there are 585 550 observations, as well as a 10% random sample for the year 2008 for the native‑born population with a native background, i.e. both parents were born in Sweden. The same set of restrictions is used for the native background population. These pertain to cohort as well as excluded missing information on maternal education (8% of the sample) and paternal education (14%). These restrictions leave a sample of 70 561. Both datasets show the educational level of the individuals and their parents in five categories: compulsory education, high school, short tertiary education, long tertiary education, and doctoral-level education.
Analytic strategy
Comparing mobility patterns for native-born individuals in immigrant versus native families helps to understand whether the role of institutions and family background vary across these two groups, and whether their intergenerational transmission patterns differ. The choice of educational over earnings transmission is explained by the former’s reliability in measuring parent-child pairs for the entire population of interest. Also, it is not possible to observe the parents at ages when they were active in the labour market with the data available. However, as it is of interest to investigate both outcomes for the groups, the chapter looks at employment rates and occupational categories by maternal education, in order to have an idea of the groups’ situation in the labour market in relation to the mother’s education level. Furthermore, since neither detailed educational categories nor the length of studies for the parents of the random sample with the native background are available, the chapter employs educational categories without transforming them into number of years. It thus provides a crude absolute measure for these categorical outcomes through cross-tabulation tables for the parent-child pairs and subsequent comparison of offspring with native and immigrant backgrounds. The measure’s sensitivity to differences in the marginal distributions of education for the two generations is addressed by testing the correlation between the educations of the parent-child pairs. This is done through use of Kendall’s tau, which is the appropriate correlation measure between two ranked discrete variables.
Results
Table 5.1 shows the highest educational attainment of sons and daughters according to five educational categories, broken down by foreign and native backgrounds. The group of origin countries selected for this chapter is referred to as +13 since they are the last 13 countries that joined the EU, even though they were not part of the EU during the time frame of the parents’ immigration to Sweden. Since these countries appear as a group, it is not possible to identify each country of origin. The +13 group represents 68% of the mothers and 66% of the fathers in the SCB study, whose other countries included the former Yugoslavia (24% of mothers and 25% of fathers); Turkey (6% and 7%, respectively); Iran (1% and 1%); and Morocco (1% and 2%).
Table 5.1. Highest educational attainment of native-born offspring with immigrant and native parents
2003-2012, percentages
|
Immigrant parents |
Native parents |
|||
|---|---|---|---|---|
|
Education |
Sons |
Daughters |
Sons |
Daughters |
|
Compulsory education |
11 |
8 |
8 |
5 |
|
High school |
49 |
45 |
53 |
44 |
|
Short tertiary |
10 |
7 |
7 |
6 |
|
Long tertiary |
28 |
39 |
30 |
44 |
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
Those who immigrated from Iran and from the +13 countries in the period of interest for this study were mainly refugees, and those immigrating from the former Yugoslavia, Turkey and Morocco were mainly labour migrants (or relatives of persons who had earlier arrived as labour migrants).
Table 5.1 shows that those with an immigrant background are more likely to complete only compulsory school education compared to those with a native background. The difference is 3 percentage points on average for both males and females. In the case of both immigrant and native backgrounds, females are more educated than males. Among those with an immigrant background, 47% of daughters as opposed to 40% of sons have reached a higher level of education than high school; the figures are 51% for daughters and 38% of sons for those with a native background. This trend of girls outperforming boys both in terms of grades and in length of education is seen in most OECD countries. Some of the explanation regarding grades is related to reading: girls tend to read longer hours than boys outside the classroom. There is also some evidence in Sweden that there is a larger gap between teachers’ evaluations on the one hand and test scores on the other for girls and non-native students compared to boys and native students (Lindahl, 2007). This could be due to the fact that female and non-native students have a harder time reflecting their real knowledge and potential in the tests and the teachers compensate for this, or it could be a result of positive discrimination. However, girls outperform boys in test scores other than those that involve teacher evaluations.
Now turning briefly to the labour market situation in relation to the parents’ education, Table 5.2 shows employment rates of sons and daughters by level of maternal education for the year 2008. Overall, the employment rate is lower for those whose mothers have only compulsory-level education, except for sons whose mothers are native-born. Female employment rates are lower than those of sons in the case of both immigrant and native backgrounds, but increase with the level of education of the mother. However, these differences are small. The lowest level of employment is observed for daughters with an immigrant background, especially where the mother has the lowest level of education. Low levels of maternal education do not seem to play a role in the employment rate of sons with a native background, thus indicating that it does not seem to be a disadvantage – at least in terms of finding a job. Looking at occupational categories however, the disadvantages become clearer.
Table 5.2. Employment rates of sons and daughters
By maternal education and immigration background, 2008, percentages
|
Immigrant parents |
Native parents |
|||
|---|---|---|---|---|
|
Mother's education |
Sons |
Daughters |
Sons |
Daughters |
|
Compulsory education |
86 |
82 |
92 |
87 |
|
High school |
88 |
86 |
92 |
88 |
|
Short tertiary |
88 |
88 |
93 |
90 |
|
Long tertiary |
90 |
86 |
92 |
90 |
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
Table 5.3 shows detailed categories of occupations for sons and daughters who were employed in 2008 by maternal level of education in relation to parental background. There is a clear pattern of occupational gender segregation regardless of maternal education or parental background. It can be seen that a lower proportion of females are in management positions compared to males, and a higher proportion are in service, care and sales occupations when compared to males. The situation is similar for the construction and manufacturing sector, with an even more striking gender gap. It is also clear that the offspring’s occupations are highly correlated with the mother’s level of education. Around 50% of the offspring whose mothers have long tertiary education are in management positions or have work that requires specialist knowledge. In the case of mothers with a doctoral degree this figure is 62%: 55% for sons and 69% for daughters with an immigrant background, and 59% and 66% respectively for sons and daughters with a native background. On the other hand, in the case of mothers with only compulsory-level education, a higher proportion of the offspring are in unskilled jobs: 19% in the case of sons with an immigrant background and 26% of sons with a native background. Table 5.3 also shows that among sons whose mothers had only compulsory education, 41% of those with an immigrant background and 35% of those with a native background have managerial jobs, work that requires specialist knowledge, or work that requires higher education.
Table 5.3. Occupations of native-born sons and daughters of immigrants
By maternal education and immigration background, 2008, percentages
|
(a) Immigrant parents |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
Sons |
Daughters |
||||||||
|
|
Mother's education |
Mother's education |
||||||||
|
Occupations |
Compulsory education |
High school |
Short tertiary |
Long tertiary |
PhD |
Compulsory education |
High school |
Short tertiary |
Long tertiary |
PhD |
|
Military work |
0 |
1 |
1 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
|
Management |
8 |
9 |
13 |
12 |
13 |
4 |
5 |
8 |
6 |
6 |
|
Work that requires specialist knowledge |
14 |
19 |
31 |
40 |
42 |
15 |
22 |
34 |
43 |
63 |
|
Work that requires higher education |
19 |
23 |
24 |
23 |
21 |
22 |
24 |
26 |
26 |
21 |
|
Office and customer service work |
6 |
5 |
4 |
3 |
4 |
15 |
14 |
11 |
8 |
5 |
|
Service, care, and sales work |
11 |
9 |
7 |
6 |
7 |
29 |
24 |
14 |
11 |
6 |
|
Work in farming, gardening, forestry, and fishing |
1 |
1 |
1 |
1 |
1 |
0 |
1 |
0 |
0 |
0 |
|
Unskilled labour in construction or manufacturing |
13 |
13 |
8 |
4 |
1 |
1 |
1 |
1 |
1 |
0 |
|
Machine operator work, transport work, etc. |
19 |
13 |
7 |
5 |
3 |
5 |
3 |
1 |
1 |
0 |
|
Work without special training requirements |
6 |
4 |
2 |
2 |
3 |
6 |
4 |
2 |
2 |
0 |
|
Unknown |
3 |
3 |
3 |
3 |
5 |
3 |
3 |
2 |
2 |
0 |
|
(b) Native-born parents |
||||||||||
|
|
Sons |
Daughters |
||||||||
|
|
Mother's education |
Mother's education |
||||||||
|
Occupations |
Compulsory education |
High school |
Short tertiary |
Long tertiary |
PhD |
Compulsory education |
High school |
Short tertiary |
Long tertiary |
PhD |
|
Military work |
0 |
1 |
1 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
|
Management |
6 |
6 |
8 |
9 |
9 |
3 |
4 |
4 |
5 |
8 |
|
Work that requires specialist knowledge |
12 |
17 |
30 |
41 |
50 |
15 |
21 |
34 |
47 |
58 |
|
Work that requires higher education |
17 |
21 |
23 |
23 |
22 |
22 |
25 |
29 |
26 |
15 |
|
Office and customer service work |
5 |
5 |
4 |
4 |
5 |
13 |
12 |
10 |
7 |
6 |
|
Service, care, and sales work |
9 |
9 |
8 |
7 |
6 |
33 |
29 |
17 |
10 |
8 |
|
Work in farming, gardening, forestry, and fishing |
2 |
2 |
1 |
1 |
0 |
1 |
1 |
1 |
0 |
1 |
|
Unskilled labour in construction or manufacturing |
21 |
18 |
11 |
6 |
2 |
2 |
1 |
0 |
1 |
0 |
|
Machine operator work, transport work, etc. |
21 |
17 |
9 |
6 |
3 |
5 |
3 |
2 |
1 |
1 |
|
Work without special training requirements |
5 |
4 |
2 |
2 |
1 |
6 |
4 |
2 |
1 |
1 |
|
Unknown |
2 |
2 |
1 |
2 |
2 |
1 |
1 |
1 |
1 |
2 |
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
These differences could indicate a higher degree of mobility for those with an immigrant background. However, a lower proportion with an immigrant background in this group, that is, those with a mother who has compulsory education, is actually employed, so this could be a more selected group, those who were able to find jobs. Overall, around 40% of those whose mothers have only compulsory education are in management positions, work that requires specialist knowledge or work that requires higher education. This figure is 41% for those with an immigrant background for both sons and daughters, and 40% for daughters with a native background. Also, it must be remembered that the highest proportion of employment is observed for sons with a native background in the group whose mothers have only compulsory education.
Table 5.4 shows the educational attainment for sons and daughters whose mothers have at most compulsory school education. We can see that sons with an immigrant background are the largest group among those reaching no more than compulsory education. This could indicate a lasting disadvantage for immigrants’ sons. In contrast, daughters with a native background more frequently reach higher levels of education. There is, more generally, an indication of greater mobility for daughters compared to sons regardless of their background. Around 29% of daughters with a mother who only has compulsory education have completed long tertiary education compared to sons, at around 18%. Overall for those with a mother with only compulsory level education, 28% of sons and 35% of daughters with an immigrant background, and 24% of sons and 34% of daughters with a native background, have tertiary education or higher.
Table 5.4. Educational attainment of sons and daughters whose mothers had only compulsory level education
2003-2012, percentages
|
Immigrant parents |
Native parents |
|||
|---|---|---|---|---|
|
Education |
Sons |
Daughters |
Sons |
Daughters |
|
Compulsory education |
15 |
10 |
13 |
8 |
|
High school |
58 |
56 |
64 |
57 |
|
Short tertiary |
8 |
6 |
6 |
5 |
|
Long tertiary |
19 |
28 |
17 |
29 |
|
PhD. |
1 |
1 |
1 |
0 |
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
These results are consistent with the general finding that girls outperform boys in terms of grades as well as length of studies. However, the results also show that this represents a higher degree of educational mobility for girls given maternal education. Furthermore, in the case of those with an immigrant background, there is a possible indication of both a higher degree of persistent disadvantage as well as a higher degree of mobility when compared to natives whose mothers have the same level of education. While one cannot draw conclusions from a descriptive table, it is important to investigate whether these differences are due to performance or decisions regarding further education given performance, in addition to investigating the mechanisms behind each pathway. The discussion now turns to the issue of intergenerational transmission of education from parents to their offspring in a more formal way, via the next tables.
Looking at Kendall’s tau for the correlation of educational attainment between the parent and the child in Table 5.5, it is clear that intergenerational educational transmission between all family member pairs is significant and positive for those with native and immigrant backgrounds. This shows that there is a significant and moderate positive relationship between the educational attainment of parents and their children. Comparing those with an immigrant and a native background, the strength of the relationship is slightly lower for those with an immigrant background. The results thus suggest a slightly lower level of ongoing disadvantage for those with an immigrant background, a finding Sweden in fact shares with other countries such as Canada and Germany (Eriksson, 2006; Niknami, 2012; Aydemir, Chen and Corak, 2009; Dustmann, 2005; Borjas, 1992; Card, DiNardo and Estes, 2000; Gang and Zimmerman, 2000). These results accordingly indicate a higher degree of mobility in the case of immigrant families compared to native families.
Table 5.5. Correlation between the educational attainment of parents and their children
2013-2012, Kendall’s rank correlation coefficient
|
|
Immigrant parents |
Native parents |
|---|---|---|
|
Mother-daughter |
0.23*** |
0.27*** |
|
Mother-son |
0.22*** |
0.28*** |
|
Father-son |
0.23*** |
0.30*** |
|
Father-daughter |
0.21*** |
0.25*** |
Note: Immigrant parents include one or two parents born abroad. *** indicates significance at the 1% level, and ** and * at the 5% and 10% levels, respectively.
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
However, the patterns could vary by country of origin for those with an immigrant background. Furthermore, as discussed in Trejo and Duncan (2011), intergenerational transmission patterns depend to a great extent on family composition – that is, whether both of the parents were born abroad or one of the parents is native-born – and the real patterns could be misrepresented if the family types are not analysed in detail. Table 5.6 therefore shows the country of origin of the parents, which proves a variable that introduces wide variation in composition. It can be seen that a large proportion of those with a mother or father from +13 countries and Iran have a native-born parent. In contrast, of those with a mother from Turkey only 3% have a native-born father; 91% have a father also from Turkey. That figure is 79% and 86% respectively for the former Yugoslavia and Morocco. What follows therefore is a detailed analysis of parent-child correlations for the different parental compositions.
Table 5.6. Family composition
By country of origin, 2003-2012, percentages
|
Mother |
Sweden (native-born) |
Same as spouse |
Other |
|---|---|---|---|
|
+13 |
62 |
27 |
11 |
|
Former Yugoslavia |
10 |
79 |
11 |
|
Turkey |
3 |
91 |
6 |
|
Iran |
51 |
42 |
7 |
|
Morocco |
7 |
86 |
7 |
|
Father |
|||
|
+13 |
71 |
21 |
8 |
|
Former Yugoslavia |
30 |
60 |
10 |
|
Turkey |
24 |
67 |
9 |
|
Iran |
73 |
16 |
11 |
|
Morocco |
60 |
26 |
14 |
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
The detailed correlations in Table 5.7 show that the ethnic composition of the parents plays a role in the transmission of education. The table indicates that there is a higher degree of mobility for those families in which both parents were born abroad. There is little evidence in the literature with regard to how children from mixed backgrounds fare in the Scandinavian countries, and the international evidence is mixed. Ramakrishan (2004) and Chiswick and DebBurman (2004) find that the native-born with mixed parentage have better educational outcomes than those who have two foreign-born parents in the United States, while Furtado (2009) finds the opposite. Hoge and Petrillo (1978) find that children from mixed backgrounds have a weaker educational inheritance compared to children from two foreign-born parents. On the other hand, in terms of educational inheritance our results show that children from foreign-born unions have weaker persistence in terms of educational attainment, suggesting a higher degree of educational mobility in foreign-born households. This could be due to positive selection into intermarriage noted in previous studies, where immigrants from the higher end of the educational distribution tend to intermarry (Kantaravic, 2004, Nekby, 2010). Since previous findings show that there is a higher degree of educational mobility for those families who are at the lower end of the educational distribution, these results could partly be reflecting a higher degree of educational mobility for those families at the lower end of the educational distribution.
Table 5.7. Correlation between the educational attainment of parents and their children
By family pairs and household composition, 2003-2012, Kendall’s rank correlation coefficient
|
|
Immigrant parents |
Native parents |
||
|---|---|---|---|---|
|
All |
Both parents born abroad |
One parent born abroad |
||
|
Mother-daughter |
0.23*** |
0.19*** |
0.25*** |
0.27*** |
|
Mother-son |
0.22*** |
0.17*** |
0.26*** |
0.28*** |
|
Father-son |
0.23*** |
0.18*** |
0.25*** |
0.30*** |
|
Father-daughter |
0.21*** |
0.17*** |
0.22*** |
0.25*** |
Note: *** indicates significance at the 1% level, and ** and * at the 5% and 10% levels, respectively.
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
The discussion therefore turns to the detailed educational attainment of parents and their children for all sub-categories. This is done to have an idea of the educational outcomes of the different groups. It is not possible to use the length of education for analytic purposes, since this information is not available for the native-born parents in the data set that we have. However, the length of education has been computed from detailed educational categories and from the information on the length of each study for the native-born population as well as for the foreign-born parents, which can be seen in Table 5.8.
Table 5.8. Years of education
By country of origin and parental composition, average years of education, 2003-12
|
(a) Native-born with two foreign-born parents |
||||
|---|---|---|---|---|
|
Parental country of origin |
Mother |
Daughter |
Father |
Son |
|
EU +13 |
10.07 (2.58) |
12.72 (2.28) |
10.67 (2.78) |
12.30 (2.34) |
|
Former Yugoslavia |
9.27 (2.34) |
12.57 (2.12) |
9.99 (2.32) |
12.12 (2.05) |
|
Turkey |
8.25 (1.84) |
12.14 (2.10) |
9.02 (2.25) |
11.52 (2.24) |
|
Iran |
12.82 (2.49) |
14.49 (2.41) |
13.65 (2.77) |
13.63 (2.68) |
|
Morocco |
8.75 (1.90) |
13.01 (2.30) |
9.41 (2.44) |
12.09 (2.31) |
|
(b) Native-born with one foreign-born parent |
||||
|
Parental country of origin |
Mother |
Daughter |
Father |
Son |
|
EU +13 |
11.05 (2.76) |
13.01 (2.31) |
11.71 (2.86) |
12.59 (2.39) |
|
Former Yugoslavia |
10.54 (2.71) |
12.93 (2.16) |
10.66 (2.47) |
12.34 (2.26) |
|
Turkey |
11.52 (3.16) |
13.34 (2.28) |
10.85 (3.00) |
13.58 (2.20) |
|
Iran |
13.77 (3.31) |
14.77 (2.59) |
13.75 (2.70) |
13.31 (2.48) |
|
Morocco |
12.43 (2.73) |
14.32 (2.49) |
10.89 (2.61) |
11.39 (2.05) |
|
(c) Native-born with two native-born parents |
||||
|
Parental country of origin |
Daughter |
Son |
||
|
Sweden |
13.13 (2.14) |
12.57 (2.18) |
||
Note: Standard deviations are reported in italics in parentheses.
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
These results show that both sons and daughters of mixed parents from all countries of origin analysed in this study have more years of education than those whose parents were both born abroad, except for sons of mothers from Iran and Morocco. Further, the middle table shows that mothers from all foreign countries of origin who have a native spouse have more years of education when compared to the upper table, that is, those who have a foreign-born spouse. Again this suggests a higher degree of educational persistence among the more highly educated families, and a higher degree of mobility for families with lower levels of education – a highly desirable situation. This is consistent with previous findings for Sweden pointing to the nonlinear relationship between the education of parents and their children (Niknami, 2012). Mothers from Turkey who have a native spouse have 3.27 years’ longer education on average than mothers from Turkey who have a foreign-born spouse. This figure is 3.68 and 1.27 for mothers from Morocco and the former Yugoslavia, respectively. In the case of Iran and +13 countries, these figures are 0.95 and 0.98. It can be seen that for countries where intermarriage is common, the gap across household types in years of education is smaller. On the other hand, the education gaps between household types are larger for countries where intermarriage is very rare, thus reflecting a higher degree of selection. Only 3% of the mothers from Turkey and 7% of mothers from Morocco have a native-born spouse. The selection dynamic in these unions can partly be explained by women having to attain high education levels to cross stronger cultural barriers, but also by their having to achieve these levels to counteract discrimination in the marriage (Çelikaksoy, Nielsen and Verner, 2006; Çelikaksoy, 2016).
Foreign-born fathers who have a native-born spouse also have more years of education than those with foreign-born spouses, but fewer years than in the case of mothers. Another important finding here is that for all countries of origin and for both family types, daughters have more years of education than sons, with one exception – the daughters of mothers from Turkey who have native-born spouses. It can also be seen that both daughters and sons of mixed households have more years of education than those who have a fully native background for all countries of origin, except for sons of mothers from Morocco; which might indicate positive selection in the case of native parents as well.
Table 5.9 shows intergenerational educational mobility patterns for all the groups analysed, revealing positive and significant educational transmission for all groups and for both household types. However, daughters from Iran and Morocco whose parents are both foreign-born stand out in their mobility patterns. Correlations of the education levels of mothers and daughters from these two countries are not significant, indicating that there is no significant relationship. Indeed, Table 5.8 showed that for those who have foreign-born parents, daughters with a background from Iran and Morocco have the longest years of education. Furthermore, father-daughter correlations for those from Morocco are significant only at the 10% level. In addition, the mother daughter correlation for those with a mother from Iran and a native-born father is significant at the 5% level. The indication is that daughters of immigrants from Morocco and Iran are the groups with the highest mobility patterns.
Table 5.9. Correlation between the educational attainment of immigrant parents and their native-born children
By country of origin, family pairs and household composition, 2013-2012, Kendall’s rank correlation coefficient
|
Parental country of origin |
Two foreign-born parents |
One foreign-born parent |
|---|---|---|
|
EU+13 |
||
|
Mother-daughter |
0.16*** |
0.26*** |
|
Mother-son |
0.15*** |
0.27*** |
|
Father-son |
0.20*** |
0.25*** |
|
Father-daughter |
0.17*** |
0.22*** |
|
Former Yugoslavia |
||
|
Mother-daughter |
0.14*** |
0.20*** |
|
Mother-son |
0.10*** |
0.19*** |
|
Father-son |
0.10*** |
0.15*** |
|
Father-daughter |
0.11*** |
0.15*** |
|
Turkey |
||
|
Mother-daughter |
0.18*** |
0.50*** |
|
Mother-son |
0.14*** |
0.26*** |
|
Father-son |
0.09*** |
0.29*** |
|
Father-daughter |
0.12*** |
0.27*** |
|
Iran |
||
|
Mother-daughter |
0.02 |
0.24** |
|
Mother-son |
0.18*** |
0.51*** |
|
Father-son |
0.14*** |
0.17*** |
|
Father-daughter |
0.17*** |
0.21*** |
|
Morocco |
||
|
Mother-daughter |
-0.03 |
0.69*** |
|
Mother-son |
0.15*** |
0.76*** |
|
Father-son |
0.11*** |
0.15*** |
|
Father-daughter |
0.05* |
0.15*** |
Note: Only families in which both parents are foreign-born are included.
Source: All tables are the author’s own calculations based on Stativ data as described above in section “Data”.
As discussed before, an overall explanation for the lower degree of educational transmission for families with a foreign background compared to families with native or mixed backgrounds could be that the parents in these households tend to have lower levels of education, and so there is a higher degree of mobility in the lower end of the educational distribution. This is, as noted before, an ideal case, and previous studies have pointed to such cases in Sweden. However, the results from Iran show that this is not the only explanation, since mothers from that country have the highest level of education regardless of household type.
Thus, the findings for Moroccan and Iranian families with an immigrant background provide examples of two household types with higher and lower educational levels. Moroccan mothers in these households have the shortest years of education, whereas Iranian mothers in these households have the longest. However, in both cases the daughters in these households have the longest years of education. That might reflect certain value systems and attitudes towards educational attainment in these households or ethnic groups that are gender-specific. Further qualitative research could shed light on the specific mechanisms that lead to higher degrees of upward educational mobility in certain groups.
Conclusion
Intergenerational mobility means that the disadvantages of parents will not find their way into the lives of their children, while intergenerational persistence signifies that parental disadvantage will be passed on to the children and be reflected in their educational attainment as well as labour market position and earnings, with inequalities persisting over generations. This chapter has discussed mobility patterns with regard to the labour market and education focusing on Sweden. The country provides an especially interesting case study in terms of mobility patterns due to its publicly funded policies supporting individuals and families in a variety of ways. Public insurance and health care, free education and a comprehensive system of support for families with young children are parts of these social welfare policies, the aim of which is to equalise living conditions among households. Thus, Sweden has been found to have a higher degree of mobility in the labour market with regard to earnings as well as occupations when compared to non-Scandinavian countries.
However, an important question is whether these social policies have a similar influence across different groups within society. Thus there have been an increasing number of international studies comparing immigrant and native families with regard to their mobility patterns in the labour market.
Overall, the Swedish income distribution is considerably more compressed than the US income distribution, and Sweden is considered a country with high levels of intergenerational income mobility. However, there are substantial differences across Swedish local labour markets. In fact, Heidrich (2015) finds that upward mobility varies greatly across these markets and that location matters: there can be especially large differences in outcomes for children from the lower end of income distribution depending on the region. Furthermore, the results here draw attention to the importance of detailed analyses within groups as well as across groups in relation to several categories. For instance, although a larger proportion of sons with an immigrant background whose mothers have only compulsory-level education are not in employment, a higher proportion of all those who are employed are in occupations requiring higher qualifications when compared to sons with a native background. Although this can partly be explained by selection into employment, detailed knowledge of the mobility pathways for different groups would increase our understanding of the mechanisms that facilitate or hinder mobility for different types of households. The results further show that a large proportion (15%) of sons with an immigrant background whose mothers have only compulsory school education also attain only a compulsory level of education. On the other hand, this group also has a higher proportion continuing with tertiary education when compared to native sons for those whose mothers have the same low level of education. Thus, there is a need for further analyses investigating factors that influence mobility in much greater detail.
Prior Swedish studies that analysed educational mobility patterns beyond simply comparing native versus immigrant families are scarce. Niknami (2012) finds that educational mobility is nonlinear, consistently finding a higher degree of upward mobility for those with less well educated parents and a higher degree of persistence at the higher end of the parental educational distribution, in principal an ideal case. That is, while the well-educated manage to pass on their advantage to their children, the less educated have children that tend to outperform their parents. Factors found to play a major role are streaming in the educational system, early school entry, and greater access to preschool and kindergarten programmes (Bauer and Ripahn, 2013). Thus, the provision of publicly funded universal child care, a publicly funded educational system that includes tertiary education, the availability of student aid and a flexible system where one can continue to tertiary education from all lines of high school are social policies in Sweden that work towards increasing the mobility for disadvantaged groups.
The relatively open meritocratic school system in Sweden, with its late and less selective streaming system, offers a flexible structure to continue to post-secondary and higher education. In addition, additional support for children who need it that is part of the regular educational system is an important factor that facilitates mobility; such additional support takes forms such as a support teacher and extra language courses. However, even though the differences appear to be smaller in Sweden compared to other countries, they are still substantial across municipalities and local communities within the municipalities in terms of school quality and neighbourhood characteristics, including with respect to residential segregation.
The findings here demonstrate that there is a significant and positive correlation between parent-child pairs, pointing to moderately persisting educational attainment for those with a native or immigrant background. The results also demonstrate that intergenerational transmission of education is weaker within immigrant families than within native families. This is a common finding for Sweden as well as for other countries, suggesting a higher degree of mobility for immigrant families. The majority of studies analysing mobility patterns study father-son pairs; this chapter looks at all family combinations, including mothers and daughters. There can also be differences across countries of origin for those families with an immigrant background, and so all groups of interest are analysed separately. Furthermore, previous studies suggest that family structure is likely to play a role in transmission mechanisms; here, transmission patterns are also investigated separately for different household types with regard to parental composition.
The results show that native-born sons and daughters in immigrant families have slightly lower education levels compared to those with a native background for the selected groups of immigrants: on average, 0.14 fewer years for daughters and 0.24 fewer years for sons. It is the opposite case for families with a mixed background, where one parent is foreign-born and the other native-born. Daughters of mixed parentage have 0.54 more years’ education on average than daughters with a fully native background, whereas this figure is 0.07 years in the case of sons. Another important result is that daughters in all family types and from both immigrant and native-born backgrounds have higher educational attainment than sons. They also have more years of education than their mothers and fathers, and thus have the highest educational attainment in the family. In terms of intergenerational transmission, in decomposing the group of families with an immigrant background into mixed versus a fully immigrant background, the results for mixed families mimic those of the native families to a large extent. In the case of families with a fully immigrant background, transmission of educational attainment is weaker not only for fathers and sons but also for all family pairs, thus implying a higher degree of mobility for this group.
Furthermore, when looking at the educational transmission patterns by country of origin, it can be seen that varying patterns emerge with regard to educational persistence across countries of origin, parent-child pairs and family type. In particular, in families where both of the parents are born abroad, where the mothers are from Iran or Morocco there is no systematic relationship between the educational attainments of mothers and daughters. Those results imply upward mobility especially for daughters in these families. Overall, the results in relation to the labour market as well as educational mobility show that there are varying patterns both within and across the different groups that cannot be explained by broad analyses. Although national policies to facilitate equality and mobility patterns are crucial, further in-depth knowledge is required with regard to the specific pathways that lead to mobility or persistence, especially for the disadvantaged groups, to create more effective and efficient support systems.
References
Aydemir, A., W.H. Chen and M. Corak (2009), “Intergenerational earnings mobility among the children of Canadian Immigrants”, Review of Economics and Statistics, Vol. 91, No. 2, pp. 377-97.
Bauer, P.C. and R.T. Riphahn (2013), “Institutional determinants of intergenerational education transmission – Comparing alternative mechanisms for natives and immigrants”, Labour Economics, Vol. 25(C), pp. 110-22.
Bisin, A. and T. Verdier (2000), “Beyond the melting pot: Cultural transmission, marriage, and the evolution of ethnic and religious traits”, Quarterly Journal of Economics, Vol. 115, No. 3, pp. 955-88.
Bisin, A. and T. Verdier (2001), “The economics of cultural transmission and the dynamics of preferences”, Journal of Economic Theory, Vol. 97, No. 2, pp. 298-319.
Björklund, A. and M. Jäntti (1997), “Intergenerational income mobility in Sweden compared to the United States”, American Economic Review, Vol. 87/ 5, pp. 1009-18.
Borjas, G.J. (1992), “Ethnic capital and intergenerational mobility”, Quarterly Journal of Economics, Vol. 107/1, pp. 123-50.
Card, D., J.E. DiNardo and E. Estes (2000), “The more things change: Immigrants and children of immigrants in the 1940s, the 1970s, and the 1990s”, in G. Borjas (ed.), Issues in the Economics of Immigration, University of Chicago Press, Chicago.
Crul, M., Keskiner, E., Schneider, J., Lelie, F. and Ghaeminia, S. (2016) “No lost generation: Education for refugee children: A comparison between Swede, Germany, The Netherlands, and Turkey”, EUI Forum on Migration, Citizenship and Demography in Florence.
Çelikaksoy, A. (2016), “Household formation behavior: An analysis of relative education and exogamy for descendants of immigrants in Sweden”, Ethnicities, Vol. 16/4, pp. 547-67.
Çelikaksoy, A. (2012), “Intergenerational transmission of interethnic union formation patterns in Sweden”, Migration Letters, Vol. 9/2, pp. 101-14.
Çelikaksoy, A. and E. Wadensjö (2017), “Refugee youth in Sweden who arrived as unaccompanied minors and separated children”, Journal of Refugee Studies, forthcoming.
Çelikaksoy, A. and E. Wadensjö (2015), “The unaccompanied refugee minors and the Swedish labor market”, IZA Discussion Paper No. 9306, Institute for the Study of Labor.
Çelikaksoy, A., H.S. Nielsen and M. Verner (2006), ”Marriage migration: Just another case of positive assortative matching?”, Review of Economics of the Household, Vol. 4/3, pp. 253-75.
Chiswick, B. and N. DebBurman (2004), “Educational attainment: An analysis by immigrant generation”, Economics of Education Review, Vol. 23/4, pp. 361-79.
Dustmann, C. (2005), “Intergenerational mobility and return migration: Comparing sons of foreign and native born fathers”, CReAM Discussion Paper No. 05/05, Centre for Research and Analysis of Migration.
Erikson, R. and J.H. Goldthorpe (1992), The Constant Flux: A Study of Class Mobility in Industrial Societies. Clarendon, Oxford.
Erikson, R. and J.O. Jonsson (1996), Can Education Be Equalized? The Swedish Case in Comparative Perspective. Westview Press, London.
Erikson, R., J.H. Goldthorpe and L. Portocarero (1979), “Intergenerational class mobility in three western countries: England, France and Sweden”, British Journal of Sociology, Vol. 30/4, pp. 415-41.
Eriksson, T. (2006), “Following your father’s footsteps? Intergenerational mobility and ethnic capital among second generation immigrants in Sweden”, MSc Thesis, Department of Economics, Lund University.
Fernandez, R., N. Guner and J. Knowles (2005), “Love and money: A theoretical and empirical analysis of household sorting and inequality”, Quarterly Journal of Economics, Vol. 120/1, pp. 273-344.
Furtado, D. (2009), “Cross nativity marriages and human capital levels of children”, Research in Labor Economics, Vol. 29/2, pp. 273-96.
Gang, I.N. and K.F. Zimmerman (2000), “Is child like parent? Educational attainment and ethnic origin”, Journal of Human Resources, Vol. 35/3, pp. 550-69.
Ganzeboom, H. and D. Treiman (1996), “Internationally comparable measures of occupational status for the 1988 International Standard Classification of Occupations”, Social Science Research, Vol. 25/3, pp. 201-39.
Grönqvist, H. and Niknami, S. (2017) “The School Achievements of Refugee Children: Lessons from Sweden”, Nordic Economic Policy Review, pp. 159-83.
Hammarstedt, M. (2009) “Intergenerational mobility and the earnings position of first-, second- and third-generation immigrants in Sweden”, Kyklos, Volume 62/2, pp. 275-92.
Hammarstedt, M. and M. Palme (2012), “Intergenerational mobility, human capital transmission and the earnings of second-generation immigrants in Sweden”, IZA Discussion Paper No. 1943.
Heidrich, S. (2015), “Intergenerational mobility in Sweden: A regional perspective”, Department of Economics, Umeå University.
Hertz, T., T. Jayasundera, P. Piraino, S. Selcuk, N. Smith and A. Verashchagina (2007), “The inheritance of educational inequality: International comparisons and fifty-year trends”, B.E. Journal of Economic Analysis & Policy, Vol. 7/2, pp. 1-45.
Hoge, D.R. and G.H. Petrillo (1978), “Determinants of church participation and attitudes among high school youth”, Journal for the Scientific Study of Religion, Vol. 17/4, pp. 359-79.
Jäntti, M., B. Bratsberg, K. Roed, O. Raaum, R. Naylor, E. Österbacka, A. Björklund and T. Eriksson (2006), “American exceptionalism in a new light: A comparison of intergenerational earnings mobility in the Nordic countries, the United Kingdom and the United States”, IZA Discussion Paper No. 1938.
Kantaravic, J. (2004), “Interethnic marriages and economic assimilation of immigrants”, IZA Discussion Paper No. 1142.
Lindahl, E. (2007), “Comparing teachers’ assessments and national test results: Evidence from Sweden”, IFAU Working Paper 2007:24, Institute for Evaluation of Labour Market and Education Policy.
Nekby, L. (2010), “Inter- and intra-marriage premiums revisited: It’s probably who you are not who you marry!”, IZA Discussion Paper No. 5317.
Niknami, S. (2012), “Essays on inequality and social policy”, PhD Thesis, Swedish Institute for Social Research, Stockholm University.
Oberdabernig, D. and A. Schneebaum (2017), “Catching up? The educational mobility of migrants’ and natives’ children in Europe”, Applied Economics, Vol. 49/37, pp. 3701-3728.
Oreopoulos, P. and K.G. Salvanes (2011), “Priceless: The nonpecuniary benefits of schooling”, Journal of Economic Perspectives, Vol. 25/1, pp. 159-84.
Österbacka, E. (2001), “Family background and economic status in Finland”, Scandinavian Journal of Economics, Vol. 103/3, pp. 467-84.
Österberg, T. (2000), “Intergenerational income mobility among individuals with different ethnic origins”, PhD Thesis, Department of Economics, Gothenburg University.
Ramakrishan, S.K. (2004), “Second generation immigrants? The 2.5 generation in the United States”, Social Science Quarterly, Vol. 85/2, pp. 380-99.
Smits, J., W. Ultee and J. Lammers (1998), “Educational homogamy in 65 countries: An explanation of differences in openness using country-level explanatory variables”, American Sociological Review, Vol. 63/2, pp. 264-85.
Solon, G. (2002), “Cross-country differences in intergenerational earnings mobility”, Journal of Economic Perspectives, Vol. 16/3, pp. 59-66.
Solon, G. (1992), “Intergenerational income mobility in the United States”, American Economic Review, Vol. 82, No. 3, pp. 393-408.
Trejo, S.J. and B. Duncan (2011), “Tracking intergenerational progress for immigrant groups: The problem of ethnic attrition, American Economic Review, Vol. 101/3, pp. 603-08.
Wadensjö, E. (2012), “Framtidens migration”, Report 5 to The Future Commission, The Future Commission, Stockholm.
Zimmerman, D.J. (1992), “Regression toward mediocrity in economic stature”, American Economic Review, Vol. 82/3, pp. 409-29.