Changing sector? Social mobility among female migrants in care and cleaning sector in Spain and Sweden

Changing sector? Social mobility among female migrants in care and cleaning sector in Spain and... Abstract This paper analyses female migrant worker’s labour mobility in Spain and Sweden by using data from the Spanish National Immigrant Survey 2007 (NIS) and the Swedish Level of Living Survey for foreign-born and their children 2010 (LNU-UFB). We examine to what extent the different institutional contexts promote or obstruct the labour mobility of immigrant women in the two countries with different migration and employment regimes. First, to identify different patterns of economic integration, we analyse the labour market entry among women who started in the care and cleaning sector, in which female migrants have acquired a special role in both countries. Secondly, we investigate what factors influences sector mobility among female migrants who started in care/cleaning jobs, and the mobility into this sector. The results show that the entry into the labour market is faster in Spain than in Sweden, and that the ethnic niche of the care/cleaning sector is more evident in Spain. The results also suggest that upward mobility (from care/cleaning job sector into professional/clerk jobs) is more feasible for migrant women in Sweden, especially if they have required country-specific human capital, and that migrant women in Spain are more likely to move into the care/cleaning job sector (regardless of education and region of origin), which reflect the higher demand for care/cleaning workers in Spain. We conclude that the two institutional contexts shape opportunities for upward and lateral mobility differently for migrant women depending on their educational level and region of origin. 1. Introduction The intensification of international migration from 2000 onwards has shaped the landscape of the Spanish and Swedish labour market in recent decades. Between 1996 and 2014, the average increase in the number of foreigners migrating to Spain was about 25 per cent per year, and reached a historical peak between 2000 and 2007 with about 570,000 new entries in the country. In Sweden, during the same period, the average increase in immigration has been around 6 per cent, with a more recent peak in 2013–14, which relates to the war in Syria, but since the year 2000, 1,314,417 immigrants have entered Sweden (Statistics Sweden 2015a). This inflow of international immigrants had a significant impact on the ethnic structure of the Spanish and Swedish labour market. In Spain, about 30 per cent of the new jobs created from 1994 to 2007 were held by immigrant workers (Reher et al. 2008). In Sweden, about 20 per cent of the work force (aged 16–65 years) in 2014 was born outside of Sweden (Statistics Sweden 2015b). In Spain, the figure was 12 per cent for the same year (INE 2016a). Female immigrants have come to play an important role in specific labour market niches. In 2013, about 45 per cent of the working female immigrant population were employed in jobs related to care and cleaning in Spain, and the figure for Sweden during the same year was about 37 per cent.1 This sectorial concentration appears to be related to the Spanish and Swedish care regimen. In the specialized literature, the concept of care regime refers to a complex web of institutional, political and cultural factors to promote or inhibit the reconciliation of work and family life (Anttonen and Sipilä 1996; Simonazzi 2009). Spain and Sweden are considered to be very different, especially in regard to different care regimes. Spain exemplifies the Mediterranean familialist model, with a relatively limited supply of public welfare services; hence, the family is considered the main provider of care for children and the elderly (León 2010; Da Roit and Weicht 2013; Hobson et al. 2015a). Sweden represents the Nordic, dual-earner model (Korpi 2000; Kvist and Peterson 2010), with extensive public provisioning of childcare and elderly care (Bettio and Plantenga 2004; Da Roit and Weicht 2013). These different institutional settings have not only had a different impact on the demand for care/cleaning services but also on opportunities for labour market integration among immigrant women. In both countries, however, the care and cleaning sector has become more and more common in the last decades. This trend appears to be associated with the aging of the Spanish and Swedish population, which poses a growing challenge in terms of long-term care for the elderly in need of attention for longer periods of their lives; and, at the same time, there is a shrinking number of family members, particularly women, who can shoulder these care responsibilities as a consequence of their increased participation in the labour market (Estévez-Abe 2015). The growing demand for workers in the care and cleaning sector, mainly as domestic workers in countries such as Spain, has been met by the international migration of women, and this sector has become the major entry point into the labour market for female migrants (Anderson 2000; Parreñas 2001; Lutz 2008; Isaksen 2010; León 2010; Vidal-Coso and Miret-Gamundi 2014). Therefore, the market expansion of care and cleaning jobs, where immigrant women have an important role, has been the result of the latent tension between the growing demand for care and insufficient state intervention to promote possibilities to reconcile work and family life, especially among Spanish and Swedish native women (Da Roit and González Ferrer 2013; Gavanas 2013). Despite its economic, social and institutional relevance, recent research has shown that a migrant worker’s situation in care and cleaning jobs is characterized by precarious employment conditions (low wages, poor working conditions, high labour instability) and informal work (ILO 2013; Hobson et al. 2015a; Strauss and McGrath 2017). In order to understand why immigrant women accept these jobs, the labour market segmentation approach states that the labour market is divided into two distinct sectors—primary and secondary—where mobility from one to another is difficult (Dickens and Lang 1993; Aysa-Lastra and Cachón-Rodríguez 2013). The secondary sector is characterized by low wages, poor working conditions, long working hours, and labour instability. In the past, native women supplied demand for the secondary sector. However, the increase of female labour force participation in the primary sector (medium- and high-skilled jobs) has led to a strong demand for foreign labour located in the secondary sector (Cachón 2006; Stanek 2011). As a consequence, this sector absorbed an increasing number of female transnational migrants, mainly during the initial stages of their settlement process. One could expect that immigrant women, over time, seek to improve their employment situation or leave the secondary sector, paralleled with social and economic integration into the host country (Chiswick 1978). Findings from previous research have shown that the Spanish and Swedish labour markets have ethnic concentrations in specific areas of economic activity and certain occupations within each area. The majority of non-European immigrant workers in Spain and Sweden are concentrated in the lower rungs of the occupational ladder, with fewer opportunities to improve their positions or salaries (Rydgren 2004; Åslund and Skans 2005; Bernardi and Martínez-Pastor, 2010; Stanek and Veira 2012; Bygren 2013; Vidal-Coso and Miret-Gamundi 2014). However, these studies rarely distinguish between more detailed occupational sections, since the ‘low-status jobs’ title does not usually distinguish between women working in care and cleaning sector and those who are not (Fernández et al. 2015; Bevelander and Irastorza 2014; Vidal-Coso and Miret-Gamundi 2014). On the other hand, there is little cross-national research that explores the process that creates this female penalty, or estimates the chances that caregivers and cleaners move to another sector. In order to fill this gap, the aim of this article is to analyse the occupational mobility of immigrant women from a wide range of different migrant origin groups who were employed in the care and cleaning sector as a first job in Spain and Sweden. Both countries represent two different institutional contexts in regard to their welfare state, migration, and employment regimes (Hobson et al. 2015a), but with a growing demand for female immigrant caregivers and cleaners. Given this contextual differences, Spain and Sweden provide an excellent international comparative testing ground to estimate to what extent the institutional context promotes or hinders the mobility of immigrant women who are employed in the care and cleaning sector relative to other sectors. Due to the contextual differences between Spain and Sweden, it is necessary to first clarify the characteristics of women whose first job was in the care and cleaning sector. This analysis seeks to identify similarities or differences that may explain the subsequent patterns of labour mobility in both countries. Secondly, we investigate what factors influences female migrant’s occupational mobility; that is, the mobility between their first job upon arrival and their current job. However, labour mobility from the care/cleaning sector cannot fully be interpreted without other points of reference; therefore, we include also women who started either in professional/clerk jobs, service jobs or other elementary jobs. Our objective is to examine if migrant women whose first job was in the care/cleaning sector are more likely than women in other sectors to abandon the ethnic niche, and if the chance of upward mobility among carers and cleaners differs between Spain and Sweden. This article contributes to the literature on immigrant integration into labour markets by assessing the labour mobility patterns among carers and cleaners from an international perspective. Most of the existing studies have explained both the labour market position and the migrants’ patterns of labour mobility in one country of destination. However, international comparative studies have been scarce; except for valuable recent exceptions (see Pereira et al. 2015). The lack of comparative studies on the labour mobility of care and cleaning workers is due to the absence of retrospective information on migrants in receiving societies (Aysa-Lastra and Cachón-Rodríguez 2013). This limitation is overcome through the Spanish National Immigrant Survey from 2007 and the Swedish Level of Living Survey of foreign-born and their children (LNU-UFB) from 2010. With this rich data we can increase our understanding of opportunities and constraints linked to the labour trajectories among workers in this sector and unveil if factors relevant for explaining social mobility are applicable across different welfare regimes. In order to provide a background, we start by presenting a brief panoramic of migration and labour market regimes in Spain and Sweden, followed by the main theoretical approaches regarding immigrant labour mobility and our hypotheses. Thirdly, the data and the methodology employed will be described. The fourth section presents the results followed by a discussion and concluding remarks. 2. Migration and employment regimes:2 the playing field for immigrant women strategies Spain and Sweden represent different welfare and care regimes. Spain exemplifies the Mediterranean familialist model, with a relatively limited supply of public welfare services; hence the family is considered the main provider of care for children and the elderly (León 2010; Hobson et al. 2015b). Sweden represents the Nordic, dual-earner model (Korpi 2000), with extensive public provisioning for childcare and the elderly (Bettio and Plantenga 2004). These different institutional settings have not only had different impact on the demand for care/cleaning services but also on opportunities for labour market integration among immigrant women. The following section will discuss the migration and employment regimes in Spain and Sweden. In comparison to Sweden, international migration is a relatively a new phenomenon in Spain. From the year 2000, the share of immigrants in the Spanish population increased from about 2 per cent to more than 13 per cent (5 million) in 20143 (INE 2016b). Before the year 2000, Spain mainly received migrants from north-western Europe and Latin America (Chile and Cuba), and today the immigrant population in the country are from a wider range of countries; non-EU countries, Latin America (Ecuador and Colombia), north Africa (Morocco) and eastern Europe (especially Romania). The rapid increase can be related to the needs of Spanish local markets regarding the expansion of the housing bubble, which acted as a pull factor for international migration in the construction sector (Reher and Requena 2009). Thus, the primary motivators behind the migration to Spain are economic reasons followed by family reasons (Reher et al. 2008; Requena and Sánchez-Domínguez 2011). By contrast, Sweden is an old immigrant country since the end of World War II. The immigration trends have changed from labour force immigration in 1940–70 to refugee/asylum immigration and family related immigration since the 1990s. Since 2000 the immigration increased especially in regard to labour market immigration from new EU countries, the European Economic Area, and refugee immigration from Iraq, Somalia and Afghanistan (Wennesjö 2013). Recently, there has been a peak in 2013–14, which relates to the war in Syria (Statistics Sweden 2015c). In 2014, family related immigration was the largest group of immigrants; 38 per cent of the residence permits were granted for family immigration, 32 per cent for refugee/asylum immigration, and 14 per cent for work immigration (Swedish Migration Agency 2015). Despite its long international migration history, Sweden has also experienced an increase of the foreign-born population since the year 2000; from about 11 per cent to 16.5 per cent (1.6 million) in 2014 (Statistics Sweden 2015d) when the largest groups were from Finland (9.9 per cent), Iraq (8.1 per cent), Poland (5.1 per cent), Iran (4.3 per cent) and former Yugoslavia (4.2 per cent) (Statistics Sweden 2015e). The different modes of entry4 significantly affect immigrants’ opportunities and changes in regard to labour market entry; for instance, the majority of those who migrated to Sweden for work reasons are employed within a year, while the transition into the labour market is considerably slower for those who migrated for family reasons and refugee/asylum reasons (Le Grand et al. 2013). Since Spain became a new receiving country of international migration, one of the most important features of the flows has been its female component, mainly featuring women from Latin America and, more recently, by women from eastern Europe (INE 2016b; Sánchez-Domínguez et al. 2011). This feminization is highly salient within the Spanish labour market, where a large proportion of migrant women work in the care/cleaning sector, which can be regarded as a response to the care deficit that has not been resolved by the state or by the family. This feminization of migrants is not prevalent in Sweden, instead the majority of the immigrants are men (50–54 per cent in 2000–14) (Statistics Sweden 2015a) As discussed, both Spain and Sweden has experienced increases in migration since the early twenty-first century. The new immigration is not only larger in volume than in earlier periods, but migratory flows from less wealthy regions have also increased, and these migrants are more likely to be in the lowest-skilled work, where care and cleaning jobs represent a significant sector (Figure 1 and 2). Figure 1. View largeDownload slide Proportion of immigrant women by different work sectors, 2001–2013. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Figure 1. View largeDownload slide Proportion of immigrant women by different work sectors, 2001–2013. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Figure 2. View largeDownload slide Immigrant women by occupation and region of origin in 2013; Spain and Sweden. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Note: Western countries include migrants from North America and Oceania. Figure 2. View largeDownload slide Immigrant women by occupation and region of origin in 2013; Spain and Sweden. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Note: Western countries include migrants from North America and Oceania. In both Spain and Sweden, we can observe that the proportion of migrant women working in the care/cleaning sector has increased over time, especially in Spain. In 2013, migrant women represented 34 per cent of the female workforce in the care/cleaning sector in Spain, and the proportion for Sweden was 21 per cent. The corresponding proportion of migrant women working in professional and clerk jobs was 7 per cent in Spain and 11 per cent in Sweden (Figure 1). The care deficit in Spain is an important factor explaining the growing demand for this sector (León 2010). However, in Sweden, the market for care/cleaning services had not expanded as a response to fill a care deficit, but as the result of a political reform, a tax deduction on such services, implemented in 2007 (Fahlén et al. 2015; Hellgren 2015). The objective was to enable men and women to combine work and family life on an equal basis, and to increase the employment rates in the formal sector and among the low skilled (Government of Sweden 2006). This indicates that the demand for care and cleaning services in Spain and Sweden are shaped rather differently. As seen, in both countries, migrant women represent a significant proportion of the workforce in the care cleaning sector. However, there are differences between Spain and Sweden in regard to the migrant women’s region of origin, and Figure 2 illustrates how the segmented labour market is related to this factor. 3. Theoretical background on migrant labour market integration In the specialized literature, multiple hypotheses to explain and predict the complex relationship between migration and labour market integration coexist (Stanek 2009). Most of these hypotheses try to identify the mechanisms underlying labour market integration from a micro, macro and meso perspective and try to determine what factors influence immigrants’ success in the host labour market. We present below the three most important sets of theoretical explanations, referred to as the hypothesis of assimilation, the hypothesis of opportunities, and the importance of the receiving context. The hypothesis of assimilation emphasizes that the pattern of immigrant integration into the labour market is U-shaped, with an initial downward mobility in the first job after migration, followed by an upward mobility paralleled with social and economic integration into the host country. According to this hypothesis, from a micro perspective, human capital, measured in terms of the skills and education of immigrant workers, has a considerable impact on immigrant’s occupational mobility in the new country (Chiswick 1978). From this perspective, occupational mobility varies in relation to immigrants’ training and experience. The lack of skills, being unable to communicate in the host country language, which facilitates adaptation to the host country and the transfer of knowledge from the country of origin, or having limited work experience generates difficulties to access the labour market as well as upward mobility. Only after a period of stay in the country, when they have acquired the native language, knowledge of the labour market, and personal and professional networks, immigrants will improve their working condition. It has also been demonstrated that training obtained in the host country improves the possibility of upward mobility, not only because it allows the immigrant to adapt more easily to the needs of the labour market, but also because it promotes a better utilization of the human capital acquired before migration (Redstone Akresh 2006). Therefore, all the disadvantages that immigrants face in their first years in the new country tend to fade after a period of stay. Numerous studies have shown the limitation of the assimilation hypothesis, in particular as regards the first stage of the settlement. As Fernández and Ortega (2006) have pointed out, immigrants have a higher risk of unemployment, higher rates of temporary employment and greater over-education than non-migrants. Even highly skilled immigrants often experience an initial downward mobility, although they recover their lost status more quickly (Riaño and Bahdadi 2007; van Riemsdijk 2013). This has been attributed to the fact that people with higher education tend to be more individualistic, have a broader social horizon and are thus less constrained by family ties and origin (Kalmijn 1998). Some studies have also linked higher levels of human capital to higher levels of integration into the host society (Portes and Rumbaut 2001). Regardless of educational level, immigrants’ risk of working in low-skilled jobs would be greater than for non-migrants, both in the first and the current job (Bevelander and Irastorza 2014; Rodríguez-Planas and Nollenberger 2014). This process appears partly to be related to discrimination by employers. Samers (2014) has suggested that ‘statistical discrimination’ plays an important role in the selection process for a job or in the workplace. This term reflects the prevalence of stereotypes, associated with the country of origin, gender or race, on the human capital of immigrants when the employer decides whether to hire or promote a person. Such stereotypes would be the source of why immigrants often work in low-skilled jobs (Reyneri and Fullen, 2011; Pereira 2013; Nergård Larsen 2016). From this theoretical framework, we derive our first two hypotheses for the empirical analysis, as follows. Hypothesis 1: We expected that education is more important in relation to the current job than in relation to the first job in Spain and Sweden. In addition, we also expected that more highly educated women, who started working in the care/cleaning sector, leave the sector after a period of stay in the country due mainly to the greater ability of finding a job that corresponds with their level of human capital. Hypothesis 2: We also expect that those who have finished their studies in the host society have higher probability to change from the low-skilled sector because they have attained a certain degree of country-specific human capital, which includes a greater chance to interact with members of the host society, with resulting resource implications in terms of become familiar with the new society. In addition, the hypothesis of opportunities, at the macro level of analysis, places the structure of demand in the labour market at the core of the explanation of labour integration. This hypothesis, linked with the labour market segmentation theory, assumes that the relatively lower labour market accomplishment among migrants is due to the lack of opportunities in the host society. According to this theoretical approach, the labour market is divided into two distinct sectors where mobility from one to another is almost chimerical, but not mobility within each of the sector (Aysa-Lastra and Cachón-Rodríguez 2013). Primary sector workers enjoy good wages, good working conditions, labour mobility, job security, and laws protecting workers. Workers in the secondary sector are deprived of these benefits. The jobs of the secondary sector, sometimes in the shadow economy, are characterized by precariousness: low wages, poor working conditions, and high labour instability (Cachón 2006). In the past, mainly women and youths worked in the secondary sector; however, during the last few decades, ethnic segregation in the secondary sector has intensified due to the strong economic growth, where native workers, especially women, are more likely to find a job in the primary sector, and thus leave positions in the secondary market open for migrant women (Cachón 2006; Vidal-Coso 2009; Stanek 2011). For non-European migrants in Europe, mostly women and irregular migrants, temporary and unskilled jobs (secondary sector) can be an important stepping-stone while they acquire receiving-country human capital (Bevelander and Irastorza 2014; Rodríguez-Planas and Nollenberger 2014) or while they are waiting to regularize their legal situation in the country. In this sense, the barriers that immigrants face in entering the labour market may also contribute to the stagnation in low-skilled jobs (Samers 2014; Pereira et al. 2015). In addition, differences in the opportunities for upward mobility also depend on other assigned or acquired social characteristics, such as gender and family situation (Raijman and Semyonov 1997; Powers and Seltzer 1998; Aysa-Lastra and Cachón-Rodríguez 2013; Bevelander and Irastorza 2014; Rodríguez-Planas and Nollenberger 2014). For instance, compared with male immigrants, female immigrants tend to hold lower occupational positions and are less likely to improve their situation, especially if they have children (Powers et al. 1998; Schrover et al. 2007; Vidal-Coso and Miret-Gamundi 2014). One explanation of this gender difference is related to prevailing gender norms, that is that women (especially mothers) settle for low-prestige and low-reward occupations as a strategy to gain easier access to the labour market and avoid conflict with traditional family roles (Raijman and Semyonov 1997). Another explanation is related to the gendered labour market segmentation, where only a limited number of jobs are accessible for female immigrants, jobs that have limited channels for upward mobility given the type of demand and the organization of the work (e.g. domestic work), or because they require substantial investment in human capital (Raijman and Semyonov 1997). This is why some authors stress that female immigrants are doubly disadvantaged in the labour market, both as immigrants and women. Alongside the structure of the labour market, migration policy of the host country is also important. Administrative rules and practices could create a discriminatory institutional framework (e.g. by restricting access to residence permits or making it difficult to validate foreign academic degrees) that limits some categories of immigrants to low-skill and low-wage employment, in response to the specific labour market demands (Cachón 2009). From this point of view, the country of origin is an important assigned characteristic that will determine the mobility opportunities of the immigrants. Non-European immigrants are more likely to occupy lower-skilled jobs, regardless of their educational level or previous work experience, than Europeans. EU citizens have a more advantageous situation due to their freedom to move between EU countries. However, it is also relevant to stress the importance of other factors explaining why non-European migrants have more limited opportunities. As recent research has pointed out, the cultural capital of immigrant women from non-European countries is often devalued in European societies (Kofman and Raghuram 2006; Riaño and Baghadi 2007; Erel 2010). We developed the following working hypotheses based on this theoretical approach on opportunities: Hypothesis 3: We expect that migrant women are more likely to remain in the secondary sector (lateral mobility or immobility) and less likely to move from the secondary sector to primary sector (upward mobility). Hypothesis 4: We expect that female immigrants arriving after the start of the intense migration (starting around 2000 in both countries) will be more likely to work in the care/cleaning sector and less likely to change the sector because the need for workers in the secondary sector. Hypothesis 5: We also expect that female immigrants who start in the care/cleaning sector with maternal responsibility, measured as women who arrived with children to Spain and Sweden, would lead them to take fewer risks in the labour market, which will reduce the likelihood of a change of sector. Hypothesis 6: Female immigrants from European countries have better jobs and greater opportunities to improve their labour situation than non-European immigrants. A third type of hypothesis, from a meso perspective, emphasizes the key role of the receiving context itself. This is the product of a number of interrelated factors such as attitudes towards immigrants, the characteristics of ethnic communities and the development of social networks (Portes and Rumbaut 1998) in the receiving societies. These elements would promote or inhibit the labour integration. With regard to attitudes towards immigrants, more positive attitudes toward immigrants can lead to fewer problems in their labour integration processes, and the social rules and control mechanisms that exist within a group can either stimulate or impede the labour mobility. Attitudes are linked to the characteristics of ethnic communities. Hence, an additional factor regarding the labour mobility in the destination country is related to the cultural, linguistic, and historical affinity existing between the society of origin and the host society. Whenever this affinity is high, it is reasonable to expect more labour opportunities and easier access to that society. When this is not the case, the perception of the ‘other’ creates social distance that may be difficult to overcome. Intermarriage is an apt indicator of social distance separating non-natives from the host society. Greater social distance between different groups will reduce the likelihood of mixed marriages (Sánchez-Domínguez, 2016). Intermarriage may allow immigrants to acquire a better knowledge of the native language as well as of the labour market and the development of personal and professional networks, which could lead to an improvement of the migrants’ working conditions. Social networks of immigrants have both positive and negative consequences for the integration of immigrants into the host societies. On the one hand, social networks reduce the economic and emotional costs and risks related with the migratory project. Membership in networks of compatriots provides job opportunities as well as contacts that go to cushion the ‘loneliness’ associated with this first stage of the migratory project. Once the person has reached the host society, its social network (family/friends) provides information to access a job, usually in an ethnic niche (Massey et al. 1994; Pereira 2013). Nevertheless, as Portes and Sensenbrenner (1993) have pointed out, these social networks also have negative consequences as migrant networks lead migrants to remain in lower positions in the labour market, where opportunities to improve their working conditions are scarce. Taking into account these additional factors that are likely to influence immigrants’ labour market trajectories, we will test the following three hypotheses: Hypothesis 7: We expect that migrants from regions with long-standing links or a smaller social distance to Spain and Sweden have greater opportunities on the host country’s labour market. Hypothesis 8: We expect that women with a native partner have a better chance to move upwards. Hypothesis 9: We also expect that those who had any contacts in Spain and Sweden upon arrival will be more likely to start working in care/cleaning jobs due to the existence of this ethnic niche and also remain in the same sector even after some time in the host country. In sum, previous studies suggest that the assumptions discussed above, and their universal applicability, very much depends on the migrant’s individual characteristics as well as the social and institutional context. These theoretical assumptions have been developed and tested within a particular country, but rarely tested across different contexts simultaneously. It is therefore more appropriate to talk about complementary hypotheses rather than mutually exclusive hypotheses. Individual characteristics, such as gender, the educational level and language skills, are important for explaining immigrants’ labour market integration and the labour market success (Chiswick and Miller 2003; Stanek 2009, 2011; Stanek and Veira 2012), as are family status and permeability of the host society towards migration (Sanchez-Dominguez 2011) and social networks in destination (Portes 2000). However, the individual characteristics and resources are interrelated with the institutional context and the labour markets. It is within these frameworks that immigrants’ opportunities and constraints regarding labour market chances are shaped. 4. Data and methods In this paper, labour trajectories of Spanish and Swedish female immigrants will be analysed with data from two sources. First, the Swedish Level of Living Survey of foreign-born and their children (LNU-UFB) from 2010–12, with a sample of 7,350 persons 18–75 years old who were born abroad, with foreign-born parents and who are not adopted and have lived in Sweden for at least five years. The sample is stratified by age and by region of origin to secure representativeness from seven regions of the origin countries. The response rate is 60.6 per cent. The second data source is the Spanish National Immigrant Survey (NIS) 2007. The total sample consists of 15,465 foreign-born respondents living in Spain at the time of the interview who were at least 16 years old and who had resided in Spain for at least a year or had the intention of doing so. The sample is stratified by age, sex and by country of origin and the sample is nationally representative. In this study, we use a sub-sample of women, aged 18–64 years at the time of immigration to Spain (n = 6,613) and Sweden (n = 1,104). Weights are used to compensate for the demographic structure (age and region of origin) of immigrants in total of Sweden and Spain. Only female labour trajectories of first generation of immigrants will be analysed.5 These data sources provide detailed information about the first and current job as well as a whole individual characteristic that are crucial to explaining immigrant’s integration and the degree of success into the labour market such as the human capital (Chiswick and Miller 2003), social capital/social networks in destination (Portes and Rumbaut 1998; Stanek and Veira 2012), family status at arrival or the current family situation (Sánchez-Domínguez et al. 2011). This allows for testing the main factors found in previous studies and to explore the links between care/cleaning sector and the migration/employment regimen at the micro level. 4.1 Variables In our analyses, two variables are of main interest; the first job sector after arrival in Sweden or Spain and the current job sector (at the time of the survey). The two variables are coded into four categories each, which are then harmonized with the International Standard Classification of Occupations, ISCO-88 codes: Professionals/clerks equal high-skilled occupation such as legislators, senior officials, managers, professionals, and technicians, associate professionals and clerks (codes 1100–4223). Care/cleaning jobs are equivalent to less-skilled, institution-based, personal-care workers (code 5132), home-based, personal-care workers (code 5133), child-care workers (code 5131), other personal care and related workers (code 5139), domestic helpers and cleaners (code 9131), helpers and cleaners in offices, hotels and other establishments (code 9132) and other cleaners and launderers (code 9133). Service jobs include all other semi- and low-skilled service occupations not included in the category care/cleaning jobs such as protective services workers, shop and market sales workers and other service workers (codes 5110–5123, 5141–5169). Elementary jobs equals other low skilled occupations not included in the category care/cleaning jobs such as agricultural and fishery workers, craft and related trades workers, plant and machine operators and assemblers, and sales and services elementary occupations (codes 6111–9120, 9141–9330). When analysing mobility, upward mobility denotes the move to professional/clerk jobs from the other job sectors. Downward mobility refers to a change from professional/clerk jobs to other sectors. Mobility between care/cleaning jobs, service jobs and other elementary jobs is considered as lateral mobility. In addition, professional/clerk jobs are equivalent to the primary sector, while the other sectors are considered to belong to the secondary sector (Aysa-Lastra and Cachón-Rodríguez 2013). The regression models also include resources, social support, year and age at arrival, region of origin and family situation. Educational attainment is divided into three categories; compulsory level (or less), upper secondary/vocational level, and tertiary level. The educational attainment refers to the current level of education. Due to data limitations, no information is available about the level of education at the time of arrival. To account for this, we include a variable whether or not the respondent attained her highest level of education in Spain and Sweden, coded as a dummy variable. This variable captures country-specific human capital and language skills and indicates the degree of integration into both societies. Region of origin includes six broad groups so as to harmonize the LNU-UFB and NIS data: Western Europe countries (EU15, the Nordic countries, Switzerland, the Anglo-speaking countries), The rest of Europe (post-socialist countries and former Yugoslavian countries), Middle East + North Africa (including Turkey), Southern Africa, Asia and Latin America. These regions of origin have strongly divergent historical, migratory and cultural links to Spain and Sweden, which is important to keep in mind when interpreting the results. Social network is operationalized by whether or not the respondent had any contacts (family, relatives, friends or other contact) in Sweden/Spain at the time of arrival. This variable measures the social capital (Portes and Rumbaut 1998). The family situation is measured with three different variables. The first measures if the respondent arrived together with her partner or not. The second measures current partnership status divided into three categories: (i) the current partner is not born in Sweden/Spain, (ii) current partner is born in Sweden/Spain, (iii) respondent has no partner. The third measures if the respondent arrived together with any children or not. Year of arrival of the respondent has been coded into three periods: prior to 1990, during the 1990s and during the 2000s (up to the moment of the surveys) so as to distinguish different migration phases. Age of arrival is divided into four categories: 18–25 years, 26–35 years, 36–45 years, and 46 years or above. The distribution of the Spanish and the Swedish sample (Table 1) shows that care/cleaning jobs is the most common job sector among migrant women in both countries, about one third in both countries started in this sector. Twenty-four per cent in the Swedish sample started as professionals/clerks, compared with 14 per cent in the Spanish sample. Considering the current job, the care/cleaning sector is still the most common job sector in Spain (25 per cent), while the most frequent current job sector in Sweden is professionals/clerks (27 per cent), followed by care/cleaning jobs (19 per cent). The educational attainment levels are similar in both samples, yet a slightly higher proportion in the Swedish sample has tertiary education. Forty per cent in the Swedish sample received their highest level of education in Sweden, compared with about 6 per cent in Spain. The different migration regimes and migration pattern is reflected in the region of origin and year of arrival. Close to 47 per cent of the immigrant women in Spain come from Latin America and 67 per cent migrated to Spain during the 2000s. The immigrant women in Sweden come mainly from Western Europe countries (31 per cent) and the rest of Europe (27 per cent), and 48 per cent migrated to Sweden prior to 1990. In both countries, the majority had some social contacts in the country of destination at the time of arrival, and the majority did not migrate with a partner or with a child, yet a higher proportion (30 per cent) in the Swedish data migrated with children, while about 51 per cent in the Spanish data left their children in the country of origin (no such information is available for the Swedish data). About 40 per cent in both samples are currently living with a partner not born in Spain or Sweden, respectively. Table 1. Descriptive statistics of the total sub-sample of women aged 18–65 at the time of arrival in Spain and Sweden Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Source: NIS and LNU-UFB. Note 1: Weighted data. The weight compensates for the demographic structure of immigrants in total Sweden and Spain respectively. Note 2: *The category ‘others not in paid work/unknown’ includes students, housewives, retired, and permanently sick. Table 1. Descriptive statistics of the total sub-sample of women aged 18–65 at the time of arrival in Spain and Sweden Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Source: NIS and LNU-UFB. Note 1: Weighted data. The weight compensates for the demographic structure of immigrants in total Sweden and Spain respectively. Note 2: *The category ‘others not in paid work/unknown’ includes students, housewives, retired, and permanently sick. 4.2 Analytical strategy In the first analysis, examining what factors are associated with the first job in care/clearing jobs, we rely on logistic regression. In the second analysis, studying the mobility between sectors, we apply multinomial regression. The multinomial models estimate the likelihood, or relative risk, of currently working in professional/clerk jobs, service jobs and elementary jobs relative to working in care/cleaning jobs. To more fully explore and illustrate the mobility between job sectors, predicted probabilities are calculated for each of the current job sectors by first job sector. Predicted probabilities are calculated as: Py=j=expLj1+expL1+…+expLn Py=refcat=11+expL1+…+expLn where (Lj) equals the estimated logit for each non-reference category (see Liao 1994). 5. Results 5.1 Labour market entry among women who started in the care/cleaning sector Considering the patterns of entry into the labour market among migrant women whose first job in the country of destination was in the care/cleaning sector (Figure 3), we find that in Spain, over 80 per cent of the migrant women found their first job within the first year since arrival, except for women from Africa. These groups have found their first job in the care/cleaning sector within two and three years, respectively. Figure 3. View largeDownload slide Time it took to find the first care/cleaning job in Spain and Sweden by the region of origin. Source: NIS and LNU-UFB. Figure 3. View largeDownload slide Time it took to find the first care/cleaning job in Spain and Sweden by the region of origin. Source: NIS and LNU-UFB. In Sweden, the pattern is quite different and the transition into the first job is much slower. Among women from Western Europe countries and the rest of Europe, about 80 per cent had found their first job within the second year since arrival. About 80 per cent of the women from Latin America found their first job within four years after arrival. For women from Asia it took about six years after arrival to reach the same employment rate, seven years for women from Middle East/North Africa, and about nine years for women from southern Africa. The diverse pattern between Spain and Sweden is highly related to the reason for migration. Significant proportions of those migrating to Sweden from all regions, except for Western Europe countries, are refugees or asylum seekers, or have migrated due to family reasons. While in Spain, work opportunities are a strong pull factor, especially for women from Latin America and Asia, and only a few percentage of the women in the Spanish sample are refugees or asylum seekers.6 5.2 Who started in the care/cleaning sector? In this section we investigate what factors influence the entry into care/cleaning jobs as the first occupation in the country of destination (Table 2) given that this is the largest first job sector among migrant women in both Spain and Sweden (Table 1). Table 2 shows that resources, region of origin, year and age of arrival, social support and family situation have somewhat different impact on the likelihood of entering care/cleaning jobs as the first job in Spain and Sweden. In Spain, those with a compulsory level of education were most likely to start working in the care/cleaning sector. Educational attainment appears to matter less in Sweden in regard to the first care/cleaning job in the country. In this model, the variable ‘highest education attained in country’ is included mainly as a control variable, hence the result should be interpreted with caution given that the first job is more likely to occur than the attainment of education in the host country. Table 2. Logistic regression on the likelihood of first job in the care/cleaning sector (b-coefficients) Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note: Categories with no information/missing values are omitted from the table. Table 2. Logistic regression on the likelihood of first job in the care/cleaning sector (b-coefficients) Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note: Categories with no information/missing values are omitted from the table. Considering ethnic niching of the care/cleaning sector, the result shows that women from Western Europe and other Anglo-speaking countries were less likely to enter care/cleaning jobs as their first job in both Spain and Sweden. In Spain, the ethnic niching of the sector is most noticeable for migrant women from the rest of Europe, Asia and Latin America. A similar pattern is apparent in Sweden, but the differences between women from Western Europe countries and the other regions are more modest. 5.3 Who moves or stays in the care/cleaning sector? The basic hypotheses stated earlier regarding mobility between the first job sector and the current job sector in the host country are tested in multinomial regression models for immigrant women and separately by country of destination. Mobility is analysed in relation to human capital, ethnic niching, and social capital. The model also controls for year of arrival and age at arrival. In the analysis (Table 3) we find a strong association between the first job and the current job in both Spain and Sweden, seen in the high relative risk ratios for the same first and current job sector. This means that migrant women tend to remain in the same sector as they began in the host country. Table 3. Multinomial regression on the likelihood of currently working in other job sectors versus care/cleaning job sector (relative risk ratios) Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note 1: The reference category is Care/cleaning jobs. Note 2: Categories with no information/missing values are omitted from the table. Table 3. Multinomial regression on the likelihood of currently working in other job sectors versus care/cleaning job sector (relative risk ratios) Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note 1: The reference category is Care/cleaning jobs. Note 2: Categories with no information/missing values are omitted from the table. Table 3 displays the likelihood of working in another job sector versus care/cleaning sector and the main effect of each independent variable, hence the understanding of the mobility is not straightforward. To clarify mobility patterns among migrant women in Spain and Sweden, predicted probabilities are calculated, based on the results from the multinomial regression (Table 3). We are interesting in the combined effects of (i) first job and human capital, (ii) first job and region of origin, and (iii) first job and social capital, on the current job sector. We also analyse the mobility in two directions; mobility from the care/cleaning sector into the other job sectors, and mobility into the care/cleaning sector from the other job sectors. Table 3 suggests that educational attainment and country-specific human capital (highest level of education attained in the host country) have different impact on the current job sector in the two institutional contexts. To fully depict the combined effects of first job and human capital on job sector mobility, predicted probabilities are presented for current job sectors by first job sector, educational attainment and country-specific human capital, holding all other variables at the reference categories (Figure 4). Figure 4. View largeDownload slide The probability of sector mobility by educational attainment and if the highest level of education was attained in Spain/Sweden. Source: NIS and LNU-UFB. Note: Comp. = compulsory level of education. Sec. = secondary level of education. Tert. = tertiary level of education. ES = Spain. SW = Sweden. Figure 4. View largeDownload slide The probability of sector mobility by educational attainment and if the highest level of education was attained in Spain/Sweden. Source: NIS and LNU-UFB. Note: Comp. = compulsory level of education. Sec. = secondary level of education. Tert. = tertiary level of education. ES = Spain. SW = Sweden. Figure 4.17 shows that the probability of remaining in the care/cleaning sector is relatively high among women with less education in both countries (53–76 per cent in Spain and 58–73 per cent in Sweden). Migrant women with tertiary education are most likely to move into other sectors, seen in high overall mobility probabilities; 70–79 per cent in Spain and 78–80 per cent in Sweden. The mobility probabilities among the low educated migrant women are slightly higher in Sweden (33–42 per cent) than in Spain (24–26 per cent). In Spain (Figure 4.1), the country-specific human capital has only a minor impact on mobility among the low-educated migrant women, who are more likely to move to service jobs. It matters more for the middle- and high-educated migrant women. Having attained the highest level of education in Spain increases the probability to move to professional/clerk jobs with about 10 percentage points for the middle-educated and by about 17 percentage points for the high-educated women, when compared with those who did not received their education in Spain. However, the middle-educated women are most likely to move to service jobs, regardless of country-specific human capital. In Sweden (Figure 4.1), low-educated migrant women who did not receive their education in Sweden are more likely to move to elementary jobs, compared with women who attained their education in Sweden. A similar pattern is found also for the middle-educated migrant women. For the high-educated women, having attained the highest level of education in Sweden increases the probability of upward mobility by about 6 percentage points, when compared with those who did not attain their education in Sweden. The mobility from other job sectors into the care/cleaning job sector is more obvious in Spain than in Sweden across all educational groups and regardless of country-specific human capital (Figure 4.2). The probabilities for downward mobility, from professional/clerk jobs into the care/cleaning job sector, is also more apparent in Spain than in Sweden, especially among less-educated migrant women who did not receive their highest level of education in the host country. In Spain, we also see an extensive the lateral mobility, from service jobs and elementary jobs into care/cleaning jobs among low- and middle-educated migrant women. For instance, a low-educated woman, with country-specific human capital, who started in elementary jobs has a 58 per cent probability to move into a care/cleaning job, and if the education is not attained in Spain, this mobility probability is about 63 per cent. The difference in country-specific human capital holds also for women with a secondary level of education. The corresponding mobility probability for the same example of the low-skilled in Sweden is 18 and 9 per cent. Although the lateral mobility probability is much lower in Sweden, the result suggests that country-specific human capital can increase the opportunities for lateral mobility among migrant women in Sweden. This is also true for women with a secondary level of education. Considering the ethnic niching, Table 3 shows that in Spain women from Western Europe countries are less likely to work in care/cleaning jobs than in professional/clerk occupations or service jobs, compared with women from other regions of origin. In Sweden, we do not find the same ethnic niching between women from Western Europe countries and the other regions of origin. To further analyse the ethnic niching, and the occupational mobility among women from different region of origin, predicted probabilities are calculated for each of the current job sectors by first job sector and region of origin, holding all other variables at the reference category (Figure 5). Figure 5. View largeDownload slide The probability of sector mobility by region of origin, in Spain and Sweden. Source: NIS and LNU-UFB. Figure 5. View largeDownload slide The probability of sector mobility by region of origin, in Spain and Sweden. Source: NIS and LNU-UFB. Figure 5:18 shows that the probability of remaining in the care/cleaning sector is very high; 74–95 per cent in Spain and slightly lower in Sweden; 68–86 per cent. Women from Western Europe countries have the highest probability of mobility from the care/cleaning sector into other sectors in Spain, but the differences across regions are more modest in Sweden. In Spain (Figure 5.1), across all regions of origin, women are most likely to move into service jobs, especially women from Western Europe countries and Asia, and the upward mobility (move to professional/clerk jobs) is very low across all regions of origin (0.2–5 per cent). In Sweden, we can observe higher upward mobility probabilities (4–16 per cent), yet less so for women from southern Africa and Asia who generally show very low mobility out of the care/cleaning job sector relative to other regions of origin. The mobility from other job sectors into the care/cleaning job sector is more evident in Spain than in Sweden across all region of origin (Figure 5.2). The probabilities for downward mobility from professional/clerk jobs into care/cleaning jobs is higher in Spain (32–83 per cent) than in Sweden (4–19 per cent). In Spain, women from Western Europe countries display the lowest probabilities for downward mobility (32 per cent) and highest among women from the rest of Europe (83 per cent) and Latin America (78 per cent). Even if the probabilities for downward mobility are low in Sweden, migrant women from Southern Africa and Asia are slightly more likely to experience downward mobility (10 per cent and 19 per cent respectively). In Spain, we also see an extensive lateral mobility, from service jobs and elementary jobs into care/cleaning jobs across all regions, yet less so for women from Western Europe countries and Asia. For instance, women from the rest of Europe and Latin America, who started in elementary jobs have an 85 per cent probability of moving into care/cleaning jobs, and women from southern Africa, who started in the service job sector have an 82 per cent probability of moving into the care/cleaning sector. In Sweden, women from southern Africa and Asia, who started in elementary jobs, have the highest probability of moving into care/cleaning jobs (51 per cent and 42 per cent respectively). The probability to move from the service sector into care/cleaning jobs is slightly higher among women from Asia (40 per cent), compared with women from the other regions. Table 3 suggests that social capital is a relevant factor determining the current job sector. To further analyse the impact of social capital on the occupational mobility, predicted probabilities are calculated for each of the current job sectors by first job sector, social network at arrival and the origin of the current partner. Figure 6.19 shows that the probability of remaining in the care/cleaning sector is more noticeable in Spain (51–74 per cent) than in Sweden (33–68 per cent), regardless of the combined effect of social capital. Figure 6. View largeDownload slide The probability of sector mobility by social contacts and origin of the partner, in Spain and Sweden. Source: NIS and LNU-UFB. Note: ES = Spain; SW = Sweden. Figure 6. View largeDownload slide The probability of sector mobility by social contacts and origin of the partner, in Spain and Sweden. Source: NIS and LNU-UFB. Note: ES = Spain; SW = Sweden. Considering the mobility from care/cleaning jobs into other job sectors, in Spain, it appears to be the lack of social contacts at arrival rather than their partner’s origin that results in higher mobility probabilities from care/cleaning jobs to other sectors (Figure 6.1). Nevertheless, highest probability is found for the mobility to service jobs; especially for women who had no contact when arriving in Spain and had a partner not born in Spain. In contrast, lowest mobility probability is found for women who had social contacts in Spain at arrival and a partner not born in Spain. In Sweden, it is the partner’s origin rather than social contacts at arrival that results in higher mobility probabilities from care/cleaning jobs to other sectors. Having a partner born in Sweden increases the probability of mobility, especially to professional/clerk jobs (Figure 6.1). As previous analysis has shown, the mobility from other job sectors into the care/cleaning job sector, by social capital, is more evident in Spain than in Sweden (Figure 6.2). The probabilities for downward mobility, from professional/clerk jobs into care/cleaning jobs, is higher in Spain (11–32 per cent) than in Sweden (1–4 per cent), regardless of social capital. In Spain, the downward mobility, from professional/clerk jobs into the care/cleaning sector is more linked to their partner’s origin than to social contact at arrival, and it is foremost for women with a partner not born in Spain, who display a higher probability of downward mobility into care/cleaning jobs (Figure 6.2). As we have seen in the other analysis, the downward mobility is very low in Sweden. However, women with a partner not born in Sweden have a slightly higher risk of downward mobility. Considering the lateral mobility, from the service job sector into care/cleaning jobs, we find that the combination of no social contact at arrival and having a partner not born in the host country generate the highest mobility probabilities in both countries. This holds also for the mobility from the elementary jobs into the care/cleaning sector in Sweden. In Spain, social contact at arrival and partner’s origin have similar influence on the mobility from the elementary jobs into the care/cleaning sector (Figure 6.2). 6. Conclusions and discussion The aim of this paper was to examine to what extent the different institutional contexts promote or obstruct the mobility of immigrant women in two countries with different migration and employment regimes. Our main focus has been on migrant women in the care/cleaning sector, with the objective of examining patterns of economic integration, factors influencing the likelihood of working in this sector and the mobility out of and into this sector. The analyses on the transition to the first job (among women whose first job was in the care/cleaning job sector) shows great variation between the countries regarding the time it took the migrant women from different regions to find their first job. In Spain, the majority of the migrant women, regardless of region of origin, found their first job within the first year after the arrival. However, women from Africa appear to face greater difficulties for economic integration, reflected in a longer transition to the first job. In Sweden, the transition to the first job is substantially slower, especially among women from the Middle East/North Africa and Southern Africa, for whom it took 7-9 years until the majority had found their first job. This diverse labour market integration pattern between Spain and Sweden highly related to the different migration regimes. A significant proportion of the immigrants in Sweden are refugees or asylum seekers, or have migrated due to family reasons, and previous studies on Sweden have shown that these factors impact immigrants’ opportunities in regard to labour market entry. In contrast, a very small proportion of the immigrants in Spain are refugees. Instead, various work opportunities have been a strong motivation for migrating to Spain. In the analysis on factors influencing the entry into care/cleaning jobs, we found that resources, region of origin, year and age of arrival, social support and family situation have different impact in Spain and Sweden. For instance, low educated migrant women in Spain is most likely to start working in the care/cleaning sector, while educational attainment have no significant impact on the first job in Sweden. The result also shows a noticeable ethnic niching of the care/cleaning sector in Spain, where women from the rest of Europe, Asia and Latin America are more likely to enter care/cleaning jobs as their first job, compared with migrant women from Western Europe. A similar pattern was observed also for Sweden, but the differences between regions of origin were more modest. This suggests that the care/cleaning sector in Spain evidently is an ethnic divider, between women from Western Europe and women from other regions, while this is less evident in Sweden. We also found that having social contacts in Spain, arriving without a partner, but with children, increase the prospects to find a care/cleaning job in Spain. These factors had no significant impact on the first job in Sweden. The results regarding the mobility out of and into the care/cleaning job sector reveals vast differences between Spain and Sweden. First, the results suggest that country-specific human capital increase the likelihood for upward mobility, from care/cleaning jobs into professional/clerk jobs in both countries. However, the upward mobility is more achievable for the low and high educated in Sweden, especially among those who attained their education in the host country. This can be explained by the fact that by attaining some education in Sweden also includes acquiring knowledge in the Swedish, a language rarely spoken by people not from Sweden. The immobility and lateral mobility (mobility into service jobs or elementary jobs) is more apparent in Spain than in Sweden, except for the low educated migrant women. The mobility into the care/cleaning job sector is more obvious in Spain than in Sweden across all educational groups and regardless of country-specific human capital, which mirrors the high demand for care/cleaning workers in Spain, and migrant women’s labour market opportunities within that context. Second, the results also indicate that there exists a tendency towards ethnic niching, especially in Spain. The mobility out of the care/cleaning sector is relatively low in both countries, except among women from Western Europe/countries, who appears to have greater opportunities to reach a higher position. In Spain, women across all regions tend to move into service jobs. Upward mobility is more feasible in Sweden than in Spain, but migrant women from Southern Africa and Asia appear to face more difficulties to move upwards. In addition, the mobility into the care/cleaning sector is more evident in Spain than in Sweden, regardless of women’s region of origin, again reflecting the high demand of care/cleaning workers in Spain. For the developing countries, this lateral mobility is informing us that the care/cleaning sector is a strategical sector itself because it could be an easy way of earning money to send remittances, but offers these women few mobility opportunities in the longer term. Women who are in this particular situation are from the rest of Europe and Latin America in Spain, and women from Southern Africa and Asia in Sweden. Considering the impact of social capital on the mobility from the care/cleaning sector, the result shows a diverse pattern between Spain and Sweden. In Spain, it is the lack of social contacts at arrival that impacts the mobility from care/cleaning jobs into other sectors, while in Sweden, having a partner born in Sweden increase the likelihood of this mobility. Considering the mobility into the care/cleaning job sector, not having a partner born in the host country increase the risk of downward mobility, from professional/clerk jobs into care/cleaning sector. This is more evident in Spain than in Sweden, where the downward mobility is very low. The combination of no social contact at arrival and having a partner not born in the host country increase the risk of lateral mobility, from the service job sector into care/cleaning jobs. This is also found for the mobility from the elementary jobs into the care/cleaning sector in Sweden, while in Spain, social contacts and partner’s origin have equal influence on the mobility from the elementary jobs into the care/cleaning sector. Based on these results, we conclude that the institutional context is important to understand the underlying dynamics of immigrant women's labour integration process. The Swedish context promotes mobility of immigrant women, while the Spanish contexts obstruct mobility. The Swedish welfare state provides a range of tools to facilitate social integration and reduce difficulties immigrants may encounter in the first stage of their settlement. For instance, through programs such as ‘the Swedish for immigrants’ (SFI), which main aim is to enable migrants to move faster in learning Swedish to study or get a job in Sweden, the Swedish welfare state aim at supporting or accelerate the integration process. SFI is a bridge between the society of origin and Sweden, but also one explanation why immigrants tend to take longer to get their first job. Immigrants not only learn the language but also how the society and the labour market work. This type of national program does not exist in Spain. For many immigrant women, the bridge between their country of origin and Spain is the care and cleaning sector. This sector is a stepping-stone while acquiring knowledge about the labour market and the Spanish society, and, as demonstrated in this paper, this temporary situation becomes the only employment alternative for these women. Funding This work was supported by Riksbankens Jubileumsfond (Grants P11-1111:1) for the research project Do Welfare Regimes Matter? Migration and Care/domestic work in two institutional contexts, Sweden and Spain: A Multi-Tier Design, led Professor Barbara Hobson, Department of Sociology, Stockholm University. This paper has also received funding from the Ministerio de Economía y Competitividad - FEDER: CSO2014-53903-C3-3-R: “Family and Ageing in Spain”. Conflict of interest statement. None declared. Acknowledgements We are especially grateful for comments and suggestions from Barbara Hobson and Rickard Sandell. Special thanks to the anonymous reviewers for their valuable comments that have greatly improved the article. Footnotes 1. The Spanish Data are from the Labour Force Survey, and the Swedish data come from Statistics Sweden’s occupational register (authors’ own calculation). accessed date 2. Due to the space limitation, authors cannot deepen all that they would like in the Spanish and Swedish institutional contexts in regard to the care, migration and employment regimes. For a more complete description, we strongly suggest a reading of Hobson, B., Hellgren, Z., and Bede, L. (2015a) ‘How institutional contexts matter: Migration and domestic care services and the capabilities of migrants in Spain and Sweden.’ Working paper 46. Families and Societies <http://www.familiesandsocieties.eu/wp-content/uploads/2015/11/WP46HobsonEtAl2015.pdf> accessed 31 Jul 2017. 3. Data from the Spanish Municipal Register of Population (Padrón Municipal de Habitantes). 4. See Hobson, B., Hellgren, Z., and Bede, L. (2015a) How institutional contexts matter: Migration and domestic care services and the capabilities of migrants in Spain and Sweden. Working paper 46. Families and Societies, 11–15 <www.familiesandsocieties.eu/wp-content/uploads/2017/03/WorkingPaper46.pdf> accessed 31 Jul 2017. 5. In order to make comparable samples, we have excluded second generation immigrants. While Sweden has a large second generation, the second generation in Spain are still of school age. It is therefore necessary, when comparing immigrant career paths between these two countries to exclude this population. 6. Reason for migration was excluded from the analyses due to high collinearity with region of origin. 7. Figure 4:1 shows mobility from the care/cleaning sector. The probabilities within each bar add up to 100 per cent, and the remaining fraction equals the probability for immobility from the care/cleaning sector. 8. See note 7. 9. See note 7. References Anderson B. ( 2000 ) Doing the Dirty Work? The Global Politics of Domestic Labour . London : Zed Books . Anttonen A. , Sipilä J. ( 1996 ) ‘ European Social Care Services: Is it Possible to Identify Models? Journal of European Social Policy, 6 / 2 : 87 – 100 . Google Scholar Crossref Search ADS Åslund O. , Skans O. N. ( 2005 ) Will I see you at work? Ethnic workplace segregation in Sweden 1985-2002. Working paper 2005:24. Uppsala: The Institute for Labour Market Policy Evaluation (IFAU). <http://www.ifau.se/globalassets/pdf/se/2005/wp05-24.pdf> accessed 31 Jul 2017. Aysa-Lastra M. , Cachón-Rodríguez L. ( 2013 ) ‘ Determinantes de la movilidad ocupacional segmentada de los inmigrantes no comunitarios en España ’, Revista Internacional de Sociología , 71 / 2 : 383 – 413 . Bernardi F. , Martínez-Pastor J. I. ( 2010 ) ‘ Falling at the Bottom: Unskilled Jobs at Entry in the Labour Market in Spain Over Time and in a Comparative Perspective’, International Journal of Comparative Sociology , 51 / 4 : 289 – 307 . Google Scholar Crossref Search ADS Bettio F. , Plantenga J. ( 2004 ) ‘ Comparing Care Regimes in Europe’, Feminist Economics , 10 / 1 : 85 – 113 . Google Scholar Crossref Search ADS Bevelander P. , Irastorza N. ( 2014 ) Catching Up: The Labor Market Outcomes of New Immigrants in Sweden . Washington, DC and Geneva : Migration Policy Institute and International Labour Office Washington, DC and Geneva: Migration Policy Institute and International Labour Office <http://www.migrationpolicy.org/research/catching- labor-market-outcomes-new-immigrants-sweden> accessed 31 Jul 2017. Bygren M. ( 2013 ) ‘ Unpacking the causes of segregation across workplaces’ Acta Sociológica , 56 / 1 : 3 – 19 . Google Scholar Crossref Search ADS Cachón . L. ( 2006 ) ‘Los inmigrantes en el mercado de trabajo’, in Aja, E. and Arango, J (eds) Veinte años de inmigración en España. Perspectiva jurídica y sociológica (1985–2004, pp. 175–201. Spain: CIDOB. Cachón L. ( 2009 ) La“ España inmigrante”: marco discriminatorio, mercado de trabajo y políticas de integración . Anthropos . Chiswick B. R. ( 1978 ) ‘ The Effect of Americanization on the Earnings of Foreign- born Men’, The Journal of Political Economy , 86 / 5 : 897 – 921 . Google Scholar Crossref Search ADS Chiswick B. R. , Miller P. W. ( 2003 ) ‘ The Complementarity of Language and Other Human Capital: Immigrant Earnings in Canada’, Economics of Education Review , 22 / 5 : 469 – 80 . Google Scholar Crossref Search ADS Da Roit B. , Weicht B. ( 2013 ) ‘ Migrant Care Work and Care, Migration and Employment Regimes. A Fuzzy Set Analysis’, Journal of European Social policy , 23 / 5 : 469 – 86 . Google Scholar Crossref Search ADS Da Roit B. , González Ferrer A. , Moreno-Fuentes F. J. ( 2013 ) ‘ The Southern European Migrant-Based Care Model: Long-Term Care and Employment Trajectories in Italy and Spain’, European Societies , 15 / 4 : 577 – 96 . Google Scholar Crossref Search ADS Dickens W. T. , Lang K. ( 1993 ) Labor Market Segmentation Theory: Reconsidering the Evidence . Netherlands : Springer, 141–80 . Erel U. ( 2010 ) ‘ Migrating Cultural Capital: Bourdieu in Migration Studies ’, Sociology , 44 / 4 : 642 – 60 . Google Scholar Crossref Search ADS Estévez-Abe M. ( 2015 ) ‘ The Outsourcing of House Cleaning and Low Skill Immigrant Workers ’, Social Politics , 22 / 2 : 147 – 69 . Google Scholar Crossref Search ADS Fahlén S. , Hobson B. , Sánchez-Domínguez M. ( 2015 ) What Do Private Markets in Domestic Services Do? Widening gap in inequalities in Spain and Sweden. Paper presented at the Annual RC19 conference in Bath, UK, 26–28 August 2015. Fernández-Macías E. , Grande R. , del Rey Poveda A. , Antón J. I. ( 2015 ) ‘ Employment and occupational mobility among recently arrived immigrants: The Spanish case 1997–2007’, Population Research and Policy Review , 34 / 2 : 243 – 77 . Google Scholar Crossref Search ADS Fernández C. , Ortega C. ( 2006 ) ‘Asimilación laboral de los inmigrantes en España: ¿Trabajo a costa de malos empleos?’, in Dolado J. J. , Vázquez P. (eds) Ensayos sobre los efectos económicos de la inmigración en España , pp . 52 – 68 . Madrid : Fundación de Estudios de Economía Aplicada . Gavanas A. ( 2013 ) ‘ Migrant Domestic Workers, Social Network Strategies and Informal Markets for Domestic Services in Sweden’, Women’s Studies International Forum , 36 : 54 – 64 . Google Scholar Crossref Search ADS Government of Sweden ( 2006 ) Proposition 2006/07: 94. Skattelättnader för hushållstjänster, m.m. [Tax Relief for Household Services, etc.] <https://www.skatteverket.se/download/18.6efe6285127ab4f1d2580006907/1359707449188/15209.pdf> accessed 31 Jul 2017. Hellgren Z. ( 2015 ) ‘ Markets, Regimes, and the Role of Stakeholders: Explaining Precariousness of Migrant Domestic/Care Workers in Different Institutional Frameworks’, Social Politics , 22 / 2 : 220 – 41 . Google Scholar Crossref Search ADS Hobson B. , Hellgren Z. , Bede L. ( 2015a ) How Institutional Contexts Matter: Migration and Domestic Care Services and the Capabilities of Migrants in Spain and Sweden. Working paper 46 (2015). Families and Societies. <http://www.familiesandsocieties.eu/wp-content/uploads/2015/11/WP46HobsonEtAl2015.pdf> accessed 31 Jul 2017. Hobson B. , Hellgren Z. , Serrano I. ( 2015b ) How Private Markets Work in Two Divergent Welfare Regimes: Shaping the Capabilities and Agency of Migrant Care/Domestic Workers in Sweden and Spain. Paper presented at the 22d CES Conference at Sciences Po in Paris, 8–10 July 2015. ILO ( 2013 ) Domestic Workers Across the World: Global and Regional Statistics and the Extent of Legal Protection. International Labour Organization. <www.ilo.org/wcmsp5/groups/public/–-dgreports/–-dcomm/–-publ/documents/publication/wcms_173363.pdf> accessed 9 Dec 2014. INE ( 2016a ) Activos por nacionalidad, sexo y ocupación [Active population of foreign nationality by sex and occupation. <http://www.ine.es/jaxiT3/Tabla.htm?t=4080andL=0> accessed 17 Mar 2017. INE ( 2016b ) Padrón Municipal de Habitantes [Spanish Municipal Register of Population]. <http://www.ine.es/dyngs/INEbase/es/operacion.htm?c=Estadistica_C&cid= 1254736177012&menu=ultiDatos&idp=1254734710990> accessed 31 Jul 2017. Isaksen L. W. (ed.) ( 2010 ) Global Care Work—Gender and Migration in Nordic Societies . Lund : Nordic Academic Press . Kalmijn M. ( 1998 ) ‘ Intermarriage and Homogamy: Causes, Hatterns, Trends’, Annual Review of Sociology , 24 / 1 : 395 – 421 . Google Scholar Crossref Search ADS Kofman E. , Raghuram P. ( 2006 ) ‘ Gender and Global Labour Migrations: Incorporating Skilled Workers’, Antipode , 38 / 2 : 282 – 303 . Google Scholar Crossref Search ADS Korpi W. ( 2000 ) ‘ Faces of Inequality: Gender, Class, and Patterns of Inequalities in Different Types of Welfare States . Social Politics , 7 / 2 : 127 – 91 . Google Scholar Crossref Search ADS Kvist E. , Peterson E. ( 2010 ) ‘ What Has Gender Equality Got to Do with It? An Analysis of Policy Debated Surrounding Domestic Services in the Welfare State of Spain and Sweden’, Nordic Journal of Feminist and Gender Research , 18 / 3 : 185 – 203 . Google Scholar Crossref Search ADS León M. ( 2010 ) ‘ Migration and Care Work in Spain: The Domestic Sector Revisited’, Social Policy and Society , 9 / 3 : 409 – 18 . Google Scholar Crossref Search ADS Liao T. F. ( 1994 ) Interpreting Probability Models: Logit, Probit, and Other Generalized Linear Models. Sage University Paper series on Quantitative Application in the Social Sciences, 07-101. Thousand Oaks, CA: Sage Publications, Inc. Le Grand C. et al. ( 2013 ) Vid arbetslivets gränser: Sysselsättning, matchning, barriärer 1974–2010. Underlagsrapport nr 12. Stockholm: Parlamentariska socialförsäkringsutredningen (S 2010:04) <www.psfu.se/uploads/files/12_vid_arbetslivets_granser.pdf> accessed 23 May 2014. Lutz H. ( 2008 ) Migration and Domestic Work – A European perspective on Global Theme. Aldershot: Ashgate Publishing Limited. Massey D. S. et al. ( 1994 ) ‘ An Evaluation of International Migration Theory: The North American Case’, Population and Development Review , 20 : 699 – 751 . Google Scholar Crossref Search ADS Nergård Larsen E. ( 2016 ) Ethnic Discrimination, the Structure of the Labour Market and Economic Conditions <http://gemm2020.eu/wp-content/uploads/2016/12/Report-Ethnic-Discrimination-The-Structure-of-the-Labour-Market-and-Economic-Condistions-1.pdf> accessed 31 Jul 2017. Parreñas R. S. ( 2001 ) Servants of globalization: Women, migration, and domestic work . Palo Alto, CA : Stanford University Press . Pereira S. ( 2013 ) Replacement Migration and Changing Preferences: Immigrant workers in Cleaning and Domestic Service in Portugal . Journal of Ethnic and Migration Studies , 39 / 7 : 1141 – 58 . Google Scholar Crossref Search ADS Pereira S. , Snel E. , ’t Hart M. ( 2015 ) ’Economic Progress, Stagnation, or Decline? Occupational Mobility of Non-Eu Immigrants in Europe ’, Immigration and Work , 27 : 129 – 65 . Google Scholar Crossref Search ADS Portes A. ( 2000 ) ‘ Globalization from below: the rise of transnational communities’, The ends of globalization: bringing society back in , 253 – 70 . Portes A. , Rumbaut R. ( 1998 ) Immigrant America: A Portrait . Berkeley, Los Angeles : University of California Press . Portes A. , Rumbaut R. G. ( 2001 ) Legacies: The story of the immigrant second generation . Berkeley : University of California Press and Russell Sage Foundation . Portes A. , Sensenbrenner J. ( 1993 ) ‘ Embeddedness and Immigration: Notes on the Social Determinants of Economic Action’, American Journal of Sociology , 98 / 6 : 1320 – 50 . Google Scholar Crossref Search ADS Pereira S. ( 2000 ) ‘Social Capital: Its Origins and Applications in Modern Sociology’, in Lesser E. L. (ed.) Knowledge and Social Capital , pp. 43 – 67 . Boston : Butterworth-Heinemann . Pereira S. , Rumbaut R. G. ( 1998 ) Immigrant America: A Portrait . Berkeley : University of California Press . Powers M. G. , Seltzer W. ( 1998 ) ‘ Occupational Status and Mobility among Undocumented Immigrants by Gender’, International Migration Review , 32 ( 1 ): 21 – 55 . Google Scholar Crossref Search ADS PubMed Powers M. G. , Seltzer W. , Shi J. ( 1998 ) ‘ Gender Differences in the Occupational Status of Undocumented Immigrants in the United States: Experience Before and after Legalization ’, International Migration Review , 32 ( 4 ): 1015 – 1046 . Google Scholar Crossref Search ADS Raijman R. , Semyonov M. ( 1997 ) ‘ Gender, Ethnicity, and Immigration: Double Disadvantage and Triple Disadvantage among Recent Immigrant Women in the Israeli Labor Market’, Gender and Society , 11 / 1 : 108 – 25 . Redstone Akresh I. ( 2006 ) Occupational mobility among legal immigrants to the United States. International Migration Review , 40 / 4 : 854 – 84 . Google Scholar Crossref Search ADS Reher D. S. et al. ( 2008 ) Informe Encuesta Nacional de Inmigrantes (ENI-2007). Documentos de Trabajo. 2(08). Reher D. , Requena M. ( 2009 ) ‘ The National Immigrant Survey of Spain: A new data source for migration studies in Europe’, Demographic Research , 20 : 253 . Google Scholar Crossref Search ADS Requena M. , Sánchez-Domínguez M. ( 2011 ) ’ Las familias inmigrantes en España’, Revista Internacional de Sociología (RIS) , 69 / M1 : 79 – 104 . Google Scholar Crossref Search ADS Reyneri E. , Fullin G. ( 2011 ) ‘ Labour Market Penalties of New Immigrants in New and Old Receiving West European Countries. International Migration , 49 / 1 : 31 – 57 . Google Scholar Crossref Search ADS Riaño Y. , Baghdadi N. ( 2007 ) ‘ Understanding the Labour Market Participation of Skilled Immigrant Women in Switzerland: The Interplay of Class, Ethnicity, and Gender’, Journal of International Migration and Integration/Revue de l'integration et de la migration internationale , 8 / 2 : 163 . Google Scholar Crossref Search ADS Rodríguez-Planas N. , Nollenberger N. ( 2014 ) A Precarious Position: The Labor Market Integration of New Immigrants in Spain . Washington, DC and Geneva : Migration Policy Institute and International Labour Office . <www.migrationpolicy.org/research/precarious-position-labor-market-integration-new-immigrants-spain> accessed 30 September 2014. Rydgren J. ( 2004 ) ‘ Mechanisms of Exclusion: Ethnic Discrimination in the Swedish Labour Market . Journal of Ethnic and Migration Studies , 30 / 4 : 697 – 716 . Google Scholar Crossref Search ADS Sanchez Dominguez M. ( 2011 ) Estrategias matrimoniales y procesos de integración social de los inmigrantes en España (Doctoral dissertation, Universidad Complutense de Madrid) . Sánchez-Domínguez M. ( 2016 ) ‘ Los senderos hacia la integración: matrimonios interétnicos en España. En Miyar-Busto, M “El nuevo escenario migratorio en España” ’, Panorama Social , 24 : 73 – 88 . Sánchez-Domínguez M. , De Valk H. , Reher D. ( 2011 ) ‘ Marriage Strategies among Immigrants in Spain’, Revista Internacional de Sociología , 69 / M1 : 139 – 66 . Google Scholar Crossref Search ADS Samers M. ( 2014 ) ‘How to Understand the Incorporation of Immigrants in European Labour Markets’, in Martiniello M. , Rath J. (eds), An Introduction to Immigration Studies , pp. 171 – 94) . Amsterdam : Amsterdam University Press . Schrover M. , Van der Leun J. , Quispel C. ( 2007 ) ‘ Niches, Labour Market Segregation, Ethnicity and Gender’, Journal of Ethnic and Migration Studies , 33 / 4 : 529 – 40 . Google Scholar Crossref Search ADS Simonazzi A. ( 2009 ) ‘ Care Regimes and National Employment Models ’, Cambridge Journal of Economics , 33 / 2 : 211 – 32 . Google Scholar Crossref Search ADS Stanek M. ( 2009 ) Los inmigrantes polacos en la Comunidad de Madrid y su inserción laboral. Doctoral thesis. Madrid: Universidad Complutense de Madrid. Stanek M. ( 2011 ) ‘Nichos étnicos y movilidad socio-ocupacional. El caso del colectivo polaco en Madrid [Ethic Niches and Socio-Occupational Mobility. The Case of the Poles in Madrid], REIS, 135 : 69 – 88 . Stanek M. , Veira A. ( 2012 ) ‘ Ethnic Niching in a Segmented Labour Market: Evidence from Spain ’, Migration Letters , 9 / 3 : 249 – 62 . Statistics Sweden ( 2015a ) Invandring efter år [Immigration by Year]. <www.statistikdatabasen.scb.se> accessed 18 Dec 2015. Statistics Sweden ( 2015b ) Befolkningen 15–74 år (AKU), 1000-tal efter arbetskraftstillhörighet, inrikes/utrikes född, kön, ålder och år [Population aged 15–74 (LFS), 1000s after Working Status, Domestic/Foreign Born, Sex, Age and Year]. <www.statistikdatabasen.scb.se> accessed 18 Dec 2015. Statistics Sweden ( 2015c ) Kriget i Syrien medförde rekordstor invandring [The War in Syria Led to Record-High Immigration]. Number 2015: 106, published 7 April 2015. <https://www.scb.se/sv_/Hitta-statistik/Artiklar/Kriget-i-Syrien-medforde-rekordstor-invandring/> accessed 31 Jul 2017. Statistics Sweden ( 2015d ) Antal personer efter region, utländsk/svensk bakgrund och år. [Number of Persons by Region, Foreign/Swedish Background and Year]. <www.statistikdatabasen.scb.se> accessed 18 Dec 2015. Statistics Sweden ( 2015e ) Utrikes födda efter födelseland, år och ålder [Foreign Born by Country of Birth, Year and Age. <www.statistikdatabasen.scb.se> accessed 18 Jun 2015. Strauss K. , McGrath S. ( 2017 ) ‘ Temporary Migration, Precarious Employment and Unfree Labour Relations: Exploring the Continuum of Exploitation in Canada’s Temporary Foreign Worker Program ’, Geoforum , 78 : 199 – 208 . <http://doi.org/10.1016/j.geoforum.2016.01.008> accessed 31 Jul 2017. Google Scholar Crossref Search ADS Swedish Migration Agency ( 2015 ) Beviljade uppehållstillstånd 1980–2014 [Granted Permits of Residence 1980–2014]. <www.migrationsverket.se/Om-Migrationsverket/Statistik/Oversikter-och-statistik-fran-tidigare-ar.html> accessed 18 Jun 2015. van Riemsdijk M. ( 2013 ) ‘ Everyday Geopolitics, the Valuation of Labour and the Socio-Political Hierarchies of Skill: Polish Nurses in Norway’, Journal of Ethnic and Migration Studies , 39 / 3 : 373 – 90 . Google Scholar Crossref Search ADS Vidal-Coso E. ( 2009 ) Activitat i complementarietat sociodemográfica entre les dones immigrades i les no immigrades a Espanya. Doctoral dissertation. [Tesis Doctoral, Universitat Autónoma de Barcelona, Facultat de Filosofia i Lletres, Departament de Geografia, Bellaterra]. Vidal-Coso E. , Miret-Gamundi P. ( 2014 ) ‘ The Labour Trajectories of Immigrant Women in Spain: Are There Signs of Upward Social Mobility?’, Demographic Research , 31 / 13 : 337 – 80 . Google Scholar Crossref Search ADS Wennesjö A. ( 2013 ) Flera skäl till invandring idag [Several Reasons for Immigration Today]. Välfärd 1/2013. Stockholm: Statistics Sweden. <www.scb.se/valfard> accessed 31 Jul 2017. © The Authors 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Migration Studies Oxford University Press

Changing sector? Social mobility among female migrants in care and cleaning sector in Spain and Sweden

Loading next page...
 
/lp/ou_press/changing-sector-social-mobility-among-female-migrants-in-care-and-AApLyT1fbD
Publisher
Oxford University Press
Copyright
© The Authors 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
2049-5838
eISSN
2049-5846
D.O.I.
10.1093/migration/mnx052
Publisher site
See Article on Publisher Site

Abstract

Abstract This paper analyses female migrant worker’s labour mobility in Spain and Sweden by using data from the Spanish National Immigrant Survey 2007 (NIS) and the Swedish Level of Living Survey for foreign-born and their children 2010 (LNU-UFB). We examine to what extent the different institutional contexts promote or obstruct the labour mobility of immigrant women in the two countries with different migration and employment regimes. First, to identify different patterns of economic integration, we analyse the labour market entry among women who started in the care and cleaning sector, in which female migrants have acquired a special role in both countries. Secondly, we investigate what factors influences sector mobility among female migrants who started in care/cleaning jobs, and the mobility into this sector. The results show that the entry into the labour market is faster in Spain than in Sweden, and that the ethnic niche of the care/cleaning sector is more evident in Spain. The results also suggest that upward mobility (from care/cleaning job sector into professional/clerk jobs) is more feasible for migrant women in Sweden, especially if they have required country-specific human capital, and that migrant women in Spain are more likely to move into the care/cleaning job sector (regardless of education and region of origin), which reflect the higher demand for care/cleaning workers in Spain. We conclude that the two institutional contexts shape opportunities for upward and lateral mobility differently for migrant women depending on their educational level and region of origin. 1. Introduction The intensification of international migration from 2000 onwards has shaped the landscape of the Spanish and Swedish labour market in recent decades. Between 1996 and 2014, the average increase in the number of foreigners migrating to Spain was about 25 per cent per year, and reached a historical peak between 2000 and 2007 with about 570,000 new entries in the country. In Sweden, during the same period, the average increase in immigration has been around 6 per cent, with a more recent peak in 2013–14, which relates to the war in Syria, but since the year 2000, 1,314,417 immigrants have entered Sweden (Statistics Sweden 2015a). This inflow of international immigrants had a significant impact on the ethnic structure of the Spanish and Swedish labour market. In Spain, about 30 per cent of the new jobs created from 1994 to 2007 were held by immigrant workers (Reher et al. 2008). In Sweden, about 20 per cent of the work force (aged 16–65 years) in 2014 was born outside of Sweden (Statistics Sweden 2015b). In Spain, the figure was 12 per cent for the same year (INE 2016a). Female immigrants have come to play an important role in specific labour market niches. In 2013, about 45 per cent of the working female immigrant population were employed in jobs related to care and cleaning in Spain, and the figure for Sweden during the same year was about 37 per cent.1 This sectorial concentration appears to be related to the Spanish and Swedish care regimen. In the specialized literature, the concept of care regime refers to a complex web of institutional, political and cultural factors to promote or inhibit the reconciliation of work and family life (Anttonen and Sipilä 1996; Simonazzi 2009). Spain and Sweden are considered to be very different, especially in regard to different care regimes. Spain exemplifies the Mediterranean familialist model, with a relatively limited supply of public welfare services; hence, the family is considered the main provider of care for children and the elderly (León 2010; Da Roit and Weicht 2013; Hobson et al. 2015a). Sweden represents the Nordic, dual-earner model (Korpi 2000; Kvist and Peterson 2010), with extensive public provisioning of childcare and elderly care (Bettio and Plantenga 2004; Da Roit and Weicht 2013). These different institutional settings have not only had a different impact on the demand for care/cleaning services but also on opportunities for labour market integration among immigrant women. In both countries, however, the care and cleaning sector has become more and more common in the last decades. This trend appears to be associated with the aging of the Spanish and Swedish population, which poses a growing challenge in terms of long-term care for the elderly in need of attention for longer periods of their lives; and, at the same time, there is a shrinking number of family members, particularly women, who can shoulder these care responsibilities as a consequence of their increased participation in the labour market (Estévez-Abe 2015). The growing demand for workers in the care and cleaning sector, mainly as domestic workers in countries such as Spain, has been met by the international migration of women, and this sector has become the major entry point into the labour market for female migrants (Anderson 2000; Parreñas 2001; Lutz 2008; Isaksen 2010; León 2010; Vidal-Coso and Miret-Gamundi 2014). Therefore, the market expansion of care and cleaning jobs, where immigrant women have an important role, has been the result of the latent tension between the growing demand for care and insufficient state intervention to promote possibilities to reconcile work and family life, especially among Spanish and Swedish native women (Da Roit and González Ferrer 2013; Gavanas 2013). Despite its economic, social and institutional relevance, recent research has shown that a migrant worker’s situation in care and cleaning jobs is characterized by precarious employment conditions (low wages, poor working conditions, high labour instability) and informal work (ILO 2013; Hobson et al. 2015a; Strauss and McGrath 2017). In order to understand why immigrant women accept these jobs, the labour market segmentation approach states that the labour market is divided into two distinct sectors—primary and secondary—where mobility from one to another is difficult (Dickens and Lang 1993; Aysa-Lastra and Cachón-Rodríguez 2013). The secondary sector is characterized by low wages, poor working conditions, long working hours, and labour instability. In the past, native women supplied demand for the secondary sector. However, the increase of female labour force participation in the primary sector (medium- and high-skilled jobs) has led to a strong demand for foreign labour located in the secondary sector (Cachón 2006; Stanek 2011). As a consequence, this sector absorbed an increasing number of female transnational migrants, mainly during the initial stages of their settlement process. One could expect that immigrant women, over time, seek to improve their employment situation or leave the secondary sector, paralleled with social and economic integration into the host country (Chiswick 1978). Findings from previous research have shown that the Spanish and Swedish labour markets have ethnic concentrations in specific areas of economic activity and certain occupations within each area. The majority of non-European immigrant workers in Spain and Sweden are concentrated in the lower rungs of the occupational ladder, with fewer opportunities to improve their positions or salaries (Rydgren 2004; Åslund and Skans 2005; Bernardi and Martínez-Pastor, 2010; Stanek and Veira 2012; Bygren 2013; Vidal-Coso and Miret-Gamundi 2014). However, these studies rarely distinguish between more detailed occupational sections, since the ‘low-status jobs’ title does not usually distinguish between women working in care and cleaning sector and those who are not (Fernández et al. 2015; Bevelander and Irastorza 2014; Vidal-Coso and Miret-Gamundi 2014). On the other hand, there is little cross-national research that explores the process that creates this female penalty, or estimates the chances that caregivers and cleaners move to another sector. In order to fill this gap, the aim of this article is to analyse the occupational mobility of immigrant women from a wide range of different migrant origin groups who were employed in the care and cleaning sector as a first job in Spain and Sweden. Both countries represent two different institutional contexts in regard to their welfare state, migration, and employment regimes (Hobson et al. 2015a), but with a growing demand for female immigrant caregivers and cleaners. Given this contextual differences, Spain and Sweden provide an excellent international comparative testing ground to estimate to what extent the institutional context promotes or hinders the mobility of immigrant women who are employed in the care and cleaning sector relative to other sectors. Due to the contextual differences between Spain and Sweden, it is necessary to first clarify the characteristics of women whose first job was in the care and cleaning sector. This analysis seeks to identify similarities or differences that may explain the subsequent patterns of labour mobility in both countries. Secondly, we investigate what factors influences female migrant’s occupational mobility; that is, the mobility between their first job upon arrival and their current job. However, labour mobility from the care/cleaning sector cannot fully be interpreted without other points of reference; therefore, we include also women who started either in professional/clerk jobs, service jobs or other elementary jobs. Our objective is to examine if migrant women whose first job was in the care/cleaning sector are more likely than women in other sectors to abandon the ethnic niche, and if the chance of upward mobility among carers and cleaners differs between Spain and Sweden. This article contributes to the literature on immigrant integration into labour markets by assessing the labour mobility patterns among carers and cleaners from an international perspective. Most of the existing studies have explained both the labour market position and the migrants’ patterns of labour mobility in one country of destination. However, international comparative studies have been scarce; except for valuable recent exceptions (see Pereira et al. 2015). The lack of comparative studies on the labour mobility of care and cleaning workers is due to the absence of retrospective information on migrants in receiving societies (Aysa-Lastra and Cachón-Rodríguez 2013). This limitation is overcome through the Spanish National Immigrant Survey from 2007 and the Swedish Level of Living Survey of foreign-born and their children (LNU-UFB) from 2010. With this rich data we can increase our understanding of opportunities and constraints linked to the labour trajectories among workers in this sector and unveil if factors relevant for explaining social mobility are applicable across different welfare regimes. In order to provide a background, we start by presenting a brief panoramic of migration and labour market regimes in Spain and Sweden, followed by the main theoretical approaches regarding immigrant labour mobility and our hypotheses. Thirdly, the data and the methodology employed will be described. The fourth section presents the results followed by a discussion and concluding remarks. 2. Migration and employment regimes:2 the playing field for immigrant women strategies Spain and Sweden represent different welfare and care regimes. Spain exemplifies the Mediterranean familialist model, with a relatively limited supply of public welfare services; hence the family is considered the main provider of care for children and the elderly (León 2010; Hobson et al. 2015b). Sweden represents the Nordic, dual-earner model (Korpi 2000), with extensive public provisioning for childcare and the elderly (Bettio and Plantenga 2004). These different institutional settings have not only had different impact on the demand for care/cleaning services but also on opportunities for labour market integration among immigrant women. The following section will discuss the migration and employment regimes in Spain and Sweden. In comparison to Sweden, international migration is a relatively a new phenomenon in Spain. From the year 2000, the share of immigrants in the Spanish population increased from about 2 per cent to more than 13 per cent (5 million) in 20143 (INE 2016b). Before the year 2000, Spain mainly received migrants from north-western Europe and Latin America (Chile and Cuba), and today the immigrant population in the country are from a wider range of countries; non-EU countries, Latin America (Ecuador and Colombia), north Africa (Morocco) and eastern Europe (especially Romania). The rapid increase can be related to the needs of Spanish local markets regarding the expansion of the housing bubble, which acted as a pull factor for international migration in the construction sector (Reher and Requena 2009). Thus, the primary motivators behind the migration to Spain are economic reasons followed by family reasons (Reher et al. 2008; Requena and Sánchez-Domínguez 2011). By contrast, Sweden is an old immigrant country since the end of World War II. The immigration trends have changed from labour force immigration in 1940–70 to refugee/asylum immigration and family related immigration since the 1990s. Since 2000 the immigration increased especially in regard to labour market immigration from new EU countries, the European Economic Area, and refugee immigration from Iraq, Somalia and Afghanistan (Wennesjö 2013). Recently, there has been a peak in 2013–14, which relates to the war in Syria (Statistics Sweden 2015c). In 2014, family related immigration was the largest group of immigrants; 38 per cent of the residence permits were granted for family immigration, 32 per cent for refugee/asylum immigration, and 14 per cent for work immigration (Swedish Migration Agency 2015). Despite its long international migration history, Sweden has also experienced an increase of the foreign-born population since the year 2000; from about 11 per cent to 16.5 per cent (1.6 million) in 2014 (Statistics Sweden 2015d) when the largest groups were from Finland (9.9 per cent), Iraq (8.1 per cent), Poland (5.1 per cent), Iran (4.3 per cent) and former Yugoslavia (4.2 per cent) (Statistics Sweden 2015e). The different modes of entry4 significantly affect immigrants’ opportunities and changes in regard to labour market entry; for instance, the majority of those who migrated to Sweden for work reasons are employed within a year, while the transition into the labour market is considerably slower for those who migrated for family reasons and refugee/asylum reasons (Le Grand et al. 2013). Since Spain became a new receiving country of international migration, one of the most important features of the flows has been its female component, mainly featuring women from Latin America and, more recently, by women from eastern Europe (INE 2016b; Sánchez-Domínguez et al. 2011). This feminization is highly salient within the Spanish labour market, where a large proportion of migrant women work in the care/cleaning sector, which can be regarded as a response to the care deficit that has not been resolved by the state or by the family. This feminization of migrants is not prevalent in Sweden, instead the majority of the immigrants are men (50–54 per cent in 2000–14) (Statistics Sweden 2015a) As discussed, both Spain and Sweden has experienced increases in migration since the early twenty-first century. The new immigration is not only larger in volume than in earlier periods, but migratory flows from less wealthy regions have also increased, and these migrants are more likely to be in the lowest-skilled work, where care and cleaning jobs represent a significant sector (Figure 1 and 2). Figure 1. View largeDownload slide Proportion of immigrant women by different work sectors, 2001–2013. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Figure 1. View largeDownload slide Proportion of immigrant women by different work sectors, 2001–2013. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Figure 2. View largeDownload slide Immigrant women by occupation and region of origin in 2013; Spain and Sweden. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Note: Western countries include migrants from North America and Oceania. Figure 2. View largeDownload slide Immigrant women by occupation and region of origin in 2013; Spain and Sweden. Source: Prepared by the authors on the basis of the Spanish Labour Force Survey and Swedish Occupational register at Statistics Sweden. Note: Western countries include migrants from North America and Oceania. In both Spain and Sweden, we can observe that the proportion of migrant women working in the care/cleaning sector has increased over time, especially in Spain. In 2013, migrant women represented 34 per cent of the female workforce in the care/cleaning sector in Spain, and the proportion for Sweden was 21 per cent. The corresponding proportion of migrant women working in professional and clerk jobs was 7 per cent in Spain and 11 per cent in Sweden (Figure 1). The care deficit in Spain is an important factor explaining the growing demand for this sector (León 2010). However, in Sweden, the market for care/cleaning services had not expanded as a response to fill a care deficit, but as the result of a political reform, a tax deduction on such services, implemented in 2007 (Fahlén et al. 2015; Hellgren 2015). The objective was to enable men and women to combine work and family life on an equal basis, and to increase the employment rates in the formal sector and among the low skilled (Government of Sweden 2006). This indicates that the demand for care and cleaning services in Spain and Sweden are shaped rather differently. As seen, in both countries, migrant women represent a significant proportion of the workforce in the care cleaning sector. However, there are differences between Spain and Sweden in regard to the migrant women’s region of origin, and Figure 2 illustrates how the segmented labour market is related to this factor. 3. Theoretical background on migrant labour market integration In the specialized literature, multiple hypotheses to explain and predict the complex relationship between migration and labour market integration coexist (Stanek 2009). Most of these hypotheses try to identify the mechanisms underlying labour market integration from a micro, macro and meso perspective and try to determine what factors influence immigrants’ success in the host labour market. We present below the three most important sets of theoretical explanations, referred to as the hypothesis of assimilation, the hypothesis of opportunities, and the importance of the receiving context. The hypothesis of assimilation emphasizes that the pattern of immigrant integration into the labour market is U-shaped, with an initial downward mobility in the first job after migration, followed by an upward mobility paralleled with social and economic integration into the host country. According to this hypothesis, from a micro perspective, human capital, measured in terms of the skills and education of immigrant workers, has a considerable impact on immigrant’s occupational mobility in the new country (Chiswick 1978). From this perspective, occupational mobility varies in relation to immigrants’ training and experience. The lack of skills, being unable to communicate in the host country language, which facilitates adaptation to the host country and the transfer of knowledge from the country of origin, or having limited work experience generates difficulties to access the labour market as well as upward mobility. Only after a period of stay in the country, when they have acquired the native language, knowledge of the labour market, and personal and professional networks, immigrants will improve their working condition. It has also been demonstrated that training obtained in the host country improves the possibility of upward mobility, not only because it allows the immigrant to adapt more easily to the needs of the labour market, but also because it promotes a better utilization of the human capital acquired before migration (Redstone Akresh 2006). Therefore, all the disadvantages that immigrants face in their first years in the new country tend to fade after a period of stay. Numerous studies have shown the limitation of the assimilation hypothesis, in particular as regards the first stage of the settlement. As Fernández and Ortega (2006) have pointed out, immigrants have a higher risk of unemployment, higher rates of temporary employment and greater over-education than non-migrants. Even highly skilled immigrants often experience an initial downward mobility, although they recover their lost status more quickly (Riaño and Bahdadi 2007; van Riemsdijk 2013). This has been attributed to the fact that people with higher education tend to be more individualistic, have a broader social horizon and are thus less constrained by family ties and origin (Kalmijn 1998). Some studies have also linked higher levels of human capital to higher levels of integration into the host society (Portes and Rumbaut 2001). Regardless of educational level, immigrants’ risk of working in low-skilled jobs would be greater than for non-migrants, both in the first and the current job (Bevelander and Irastorza 2014; Rodríguez-Planas and Nollenberger 2014). This process appears partly to be related to discrimination by employers. Samers (2014) has suggested that ‘statistical discrimination’ plays an important role in the selection process for a job or in the workplace. This term reflects the prevalence of stereotypes, associated with the country of origin, gender or race, on the human capital of immigrants when the employer decides whether to hire or promote a person. Such stereotypes would be the source of why immigrants often work in low-skilled jobs (Reyneri and Fullen, 2011; Pereira 2013; Nergård Larsen 2016). From this theoretical framework, we derive our first two hypotheses for the empirical analysis, as follows. Hypothesis 1: We expected that education is more important in relation to the current job than in relation to the first job in Spain and Sweden. In addition, we also expected that more highly educated women, who started working in the care/cleaning sector, leave the sector after a period of stay in the country due mainly to the greater ability of finding a job that corresponds with their level of human capital. Hypothesis 2: We also expect that those who have finished their studies in the host society have higher probability to change from the low-skilled sector because they have attained a certain degree of country-specific human capital, which includes a greater chance to interact with members of the host society, with resulting resource implications in terms of become familiar with the new society. In addition, the hypothesis of opportunities, at the macro level of analysis, places the structure of demand in the labour market at the core of the explanation of labour integration. This hypothesis, linked with the labour market segmentation theory, assumes that the relatively lower labour market accomplishment among migrants is due to the lack of opportunities in the host society. According to this theoretical approach, the labour market is divided into two distinct sectors where mobility from one to another is almost chimerical, but not mobility within each of the sector (Aysa-Lastra and Cachón-Rodríguez 2013). Primary sector workers enjoy good wages, good working conditions, labour mobility, job security, and laws protecting workers. Workers in the secondary sector are deprived of these benefits. The jobs of the secondary sector, sometimes in the shadow economy, are characterized by precariousness: low wages, poor working conditions, and high labour instability (Cachón 2006). In the past, mainly women and youths worked in the secondary sector; however, during the last few decades, ethnic segregation in the secondary sector has intensified due to the strong economic growth, where native workers, especially women, are more likely to find a job in the primary sector, and thus leave positions in the secondary market open for migrant women (Cachón 2006; Vidal-Coso 2009; Stanek 2011). For non-European migrants in Europe, mostly women and irregular migrants, temporary and unskilled jobs (secondary sector) can be an important stepping-stone while they acquire receiving-country human capital (Bevelander and Irastorza 2014; Rodríguez-Planas and Nollenberger 2014) or while they are waiting to regularize their legal situation in the country. In this sense, the barriers that immigrants face in entering the labour market may also contribute to the stagnation in low-skilled jobs (Samers 2014; Pereira et al. 2015). In addition, differences in the opportunities for upward mobility also depend on other assigned or acquired social characteristics, such as gender and family situation (Raijman and Semyonov 1997; Powers and Seltzer 1998; Aysa-Lastra and Cachón-Rodríguez 2013; Bevelander and Irastorza 2014; Rodríguez-Planas and Nollenberger 2014). For instance, compared with male immigrants, female immigrants tend to hold lower occupational positions and are less likely to improve their situation, especially if they have children (Powers et al. 1998; Schrover et al. 2007; Vidal-Coso and Miret-Gamundi 2014). One explanation of this gender difference is related to prevailing gender norms, that is that women (especially mothers) settle for low-prestige and low-reward occupations as a strategy to gain easier access to the labour market and avoid conflict with traditional family roles (Raijman and Semyonov 1997). Another explanation is related to the gendered labour market segmentation, where only a limited number of jobs are accessible for female immigrants, jobs that have limited channels for upward mobility given the type of demand and the organization of the work (e.g. domestic work), or because they require substantial investment in human capital (Raijman and Semyonov 1997). This is why some authors stress that female immigrants are doubly disadvantaged in the labour market, both as immigrants and women. Alongside the structure of the labour market, migration policy of the host country is also important. Administrative rules and practices could create a discriminatory institutional framework (e.g. by restricting access to residence permits or making it difficult to validate foreign academic degrees) that limits some categories of immigrants to low-skill and low-wage employment, in response to the specific labour market demands (Cachón 2009). From this point of view, the country of origin is an important assigned characteristic that will determine the mobility opportunities of the immigrants. Non-European immigrants are more likely to occupy lower-skilled jobs, regardless of their educational level or previous work experience, than Europeans. EU citizens have a more advantageous situation due to their freedom to move between EU countries. However, it is also relevant to stress the importance of other factors explaining why non-European migrants have more limited opportunities. As recent research has pointed out, the cultural capital of immigrant women from non-European countries is often devalued in European societies (Kofman and Raghuram 2006; Riaño and Baghadi 2007; Erel 2010). We developed the following working hypotheses based on this theoretical approach on opportunities: Hypothesis 3: We expect that migrant women are more likely to remain in the secondary sector (lateral mobility or immobility) and less likely to move from the secondary sector to primary sector (upward mobility). Hypothesis 4: We expect that female immigrants arriving after the start of the intense migration (starting around 2000 in both countries) will be more likely to work in the care/cleaning sector and less likely to change the sector because the need for workers in the secondary sector. Hypothesis 5: We also expect that female immigrants who start in the care/cleaning sector with maternal responsibility, measured as women who arrived with children to Spain and Sweden, would lead them to take fewer risks in the labour market, which will reduce the likelihood of a change of sector. Hypothesis 6: Female immigrants from European countries have better jobs and greater opportunities to improve their labour situation than non-European immigrants. A third type of hypothesis, from a meso perspective, emphasizes the key role of the receiving context itself. This is the product of a number of interrelated factors such as attitudes towards immigrants, the characteristics of ethnic communities and the development of social networks (Portes and Rumbaut 1998) in the receiving societies. These elements would promote or inhibit the labour integration. With regard to attitudes towards immigrants, more positive attitudes toward immigrants can lead to fewer problems in their labour integration processes, and the social rules and control mechanisms that exist within a group can either stimulate or impede the labour mobility. Attitudes are linked to the characteristics of ethnic communities. Hence, an additional factor regarding the labour mobility in the destination country is related to the cultural, linguistic, and historical affinity existing between the society of origin and the host society. Whenever this affinity is high, it is reasonable to expect more labour opportunities and easier access to that society. When this is not the case, the perception of the ‘other’ creates social distance that may be difficult to overcome. Intermarriage is an apt indicator of social distance separating non-natives from the host society. Greater social distance between different groups will reduce the likelihood of mixed marriages (Sánchez-Domínguez, 2016). Intermarriage may allow immigrants to acquire a better knowledge of the native language as well as of the labour market and the development of personal and professional networks, which could lead to an improvement of the migrants’ working conditions. Social networks of immigrants have both positive and negative consequences for the integration of immigrants into the host societies. On the one hand, social networks reduce the economic and emotional costs and risks related with the migratory project. Membership in networks of compatriots provides job opportunities as well as contacts that go to cushion the ‘loneliness’ associated with this first stage of the migratory project. Once the person has reached the host society, its social network (family/friends) provides information to access a job, usually in an ethnic niche (Massey et al. 1994; Pereira 2013). Nevertheless, as Portes and Sensenbrenner (1993) have pointed out, these social networks also have negative consequences as migrant networks lead migrants to remain in lower positions in the labour market, where opportunities to improve their working conditions are scarce. Taking into account these additional factors that are likely to influence immigrants’ labour market trajectories, we will test the following three hypotheses: Hypothesis 7: We expect that migrants from regions with long-standing links or a smaller social distance to Spain and Sweden have greater opportunities on the host country’s labour market. Hypothesis 8: We expect that women with a native partner have a better chance to move upwards. Hypothesis 9: We also expect that those who had any contacts in Spain and Sweden upon arrival will be more likely to start working in care/cleaning jobs due to the existence of this ethnic niche and also remain in the same sector even after some time in the host country. In sum, previous studies suggest that the assumptions discussed above, and their universal applicability, very much depends on the migrant’s individual characteristics as well as the social and institutional context. These theoretical assumptions have been developed and tested within a particular country, but rarely tested across different contexts simultaneously. It is therefore more appropriate to talk about complementary hypotheses rather than mutually exclusive hypotheses. Individual characteristics, such as gender, the educational level and language skills, are important for explaining immigrants’ labour market integration and the labour market success (Chiswick and Miller 2003; Stanek 2009, 2011; Stanek and Veira 2012), as are family status and permeability of the host society towards migration (Sanchez-Dominguez 2011) and social networks in destination (Portes 2000). However, the individual characteristics and resources are interrelated with the institutional context and the labour markets. It is within these frameworks that immigrants’ opportunities and constraints regarding labour market chances are shaped. 4. Data and methods In this paper, labour trajectories of Spanish and Swedish female immigrants will be analysed with data from two sources. First, the Swedish Level of Living Survey of foreign-born and their children (LNU-UFB) from 2010–12, with a sample of 7,350 persons 18–75 years old who were born abroad, with foreign-born parents and who are not adopted and have lived in Sweden for at least five years. The sample is stratified by age and by region of origin to secure representativeness from seven regions of the origin countries. The response rate is 60.6 per cent. The second data source is the Spanish National Immigrant Survey (NIS) 2007. The total sample consists of 15,465 foreign-born respondents living in Spain at the time of the interview who were at least 16 years old and who had resided in Spain for at least a year or had the intention of doing so. The sample is stratified by age, sex and by country of origin and the sample is nationally representative. In this study, we use a sub-sample of women, aged 18–64 years at the time of immigration to Spain (n = 6,613) and Sweden (n = 1,104). Weights are used to compensate for the demographic structure (age and region of origin) of immigrants in total of Sweden and Spain. Only female labour trajectories of first generation of immigrants will be analysed.5 These data sources provide detailed information about the first and current job as well as a whole individual characteristic that are crucial to explaining immigrant’s integration and the degree of success into the labour market such as the human capital (Chiswick and Miller 2003), social capital/social networks in destination (Portes and Rumbaut 1998; Stanek and Veira 2012), family status at arrival or the current family situation (Sánchez-Domínguez et al. 2011). This allows for testing the main factors found in previous studies and to explore the links between care/cleaning sector and the migration/employment regimen at the micro level. 4.1 Variables In our analyses, two variables are of main interest; the first job sector after arrival in Sweden or Spain and the current job sector (at the time of the survey). The two variables are coded into four categories each, which are then harmonized with the International Standard Classification of Occupations, ISCO-88 codes: Professionals/clerks equal high-skilled occupation such as legislators, senior officials, managers, professionals, and technicians, associate professionals and clerks (codes 1100–4223). Care/cleaning jobs are equivalent to less-skilled, institution-based, personal-care workers (code 5132), home-based, personal-care workers (code 5133), child-care workers (code 5131), other personal care and related workers (code 5139), domestic helpers and cleaners (code 9131), helpers and cleaners in offices, hotels and other establishments (code 9132) and other cleaners and launderers (code 9133). Service jobs include all other semi- and low-skilled service occupations not included in the category care/cleaning jobs such as protective services workers, shop and market sales workers and other service workers (codes 5110–5123, 5141–5169). Elementary jobs equals other low skilled occupations not included in the category care/cleaning jobs such as agricultural and fishery workers, craft and related trades workers, plant and machine operators and assemblers, and sales and services elementary occupations (codes 6111–9120, 9141–9330). When analysing mobility, upward mobility denotes the move to professional/clerk jobs from the other job sectors. Downward mobility refers to a change from professional/clerk jobs to other sectors. Mobility between care/cleaning jobs, service jobs and other elementary jobs is considered as lateral mobility. In addition, professional/clerk jobs are equivalent to the primary sector, while the other sectors are considered to belong to the secondary sector (Aysa-Lastra and Cachón-Rodríguez 2013). The regression models also include resources, social support, year and age at arrival, region of origin and family situation. Educational attainment is divided into three categories; compulsory level (or less), upper secondary/vocational level, and tertiary level. The educational attainment refers to the current level of education. Due to data limitations, no information is available about the level of education at the time of arrival. To account for this, we include a variable whether or not the respondent attained her highest level of education in Spain and Sweden, coded as a dummy variable. This variable captures country-specific human capital and language skills and indicates the degree of integration into both societies. Region of origin includes six broad groups so as to harmonize the LNU-UFB and NIS data: Western Europe countries (EU15, the Nordic countries, Switzerland, the Anglo-speaking countries), The rest of Europe (post-socialist countries and former Yugoslavian countries), Middle East + North Africa (including Turkey), Southern Africa, Asia and Latin America. These regions of origin have strongly divergent historical, migratory and cultural links to Spain and Sweden, which is important to keep in mind when interpreting the results. Social network is operationalized by whether or not the respondent had any contacts (family, relatives, friends or other contact) in Sweden/Spain at the time of arrival. This variable measures the social capital (Portes and Rumbaut 1998). The family situation is measured with three different variables. The first measures if the respondent arrived together with her partner or not. The second measures current partnership status divided into three categories: (i) the current partner is not born in Sweden/Spain, (ii) current partner is born in Sweden/Spain, (iii) respondent has no partner. The third measures if the respondent arrived together with any children or not. Year of arrival of the respondent has been coded into three periods: prior to 1990, during the 1990s and during the 2000s (up to the moment of the surveys) so as to distinguish different migration phases. Age of arrival is divided into four categories: 18–25 years, 26–35 years, 36–45 years, and 46 years or above. The distribution of the Spanish and the Swedish sample (Table 1) shows that care/cleaning jobs is the most common job sector among migrant women in both countries, about one third in both countries started in this sector. Twenty-four per cent in the Swedish sample started as professionals/clerks, compared with 14 per cent in the Spanish sample. Considering the current job, the care/cleaning sector is still the most common job sector in Spain (25 per cent), while the most frequent current job sector in Sweden is professionals/clerks (27 per cent), followed by care/cleaning jobs (19 per cent). The educational attainment levels are similar in both samples, yet a slightly higher proportion in the Swedish sample has tertiary education. Forty per cent in the Swedish sample received their highest level of education in Sweden, compared with about 6 per cent in Spain. The different migration regimes and migration pattern is reflected in the region of origin and year of arrival. Close to 47 per cent of the immigrant women in Spain come from Latin America and 67 per cent migrated to Spain during the 2000s. The immigrant women in Sweden come mainly from Western Europe countries (31 per cent) and the rest of Europe (27 per cent), and 48 per cent migrated to Sweden prior to 1990. In both countries, the majority had some social contacts in the country of destination at the time of arrival, and the majority did not migrate with a partner or with a child, yet a higher proportion (30 per cent) in the Swedish data migrated with children, while about 51 per cent in the Spanish data left their children in the country of origin (no such information is available for the Swedish data). About 40 per cent in both samples are currently living with a partner not born in Spain or Sweden, respectively. Table 1. Descriptive statistics of the total sub-sample of women aged 18–65 at the time of arrival in Spain and Sweden Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Source: NIS and LNU-UFB. Note 1: Weighted data. The weight compensates for the demographic structure of immigrants in total Sweden and Spain respectively. Note 2: *The category ‘others not in paid work/unknown’ includes students, housewives, retired, and permanently sick. Table 1. Descriptive statistics of the total sub-sample of women aged 18–65 at the time of arrival in Spain and Sweden Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Spain Sweden Freq. % Freq. % First job in country Professionals/clerks 610 9.2 269 24.3 Care/cleaning jobs 1769 26.7 325 29.4 Service jobs 849 12.8 108 9.8 Elementary jobs 462 7.0 222 20.0 First job unknown 2923 44.2 180 16.6 Current job in country Professionals/clerks 915 13.8 303 27.4 Care/cleaning jobs 1673 25.3 210 19.0 Service jobs 850 12.9 58 5.3 Elementary jobs 400 6.1 70 6.3 Unemployed 513 7.8 70 6.3 Others not in paid work/unknown* 2261 34.2 392 35.7 Education Compulsory level (or less) 1585 24.0 283 25.6 Upper secondary/vocational level 3411 51.6 504 45.8 Tertiary level 1561 23.6 317 28.6 Education unknown 57 0.9 – – Highest education attained in country No 5636 85.2 638 57.9 Yes 409 6.2 442 39.9 No information 568 8.6 24 2.2 Region of origin Western Europe/countries 1221 18.5 344 31.4 The rest of Europe 1312 19.8 302 27.3 Middle East and North Africa 621 9.4 224 20.2 The rest of Africa 140 2.1 53 4.8 Asia 232 3.5 115 10.4 Latin America 3087 46.7 66 5.9 Social network at arrival Had contacts in county at arrival 4804 72.6 835 75.4 Had no contacts in country at arrival 1009 15.3 265 24.2 No information 800 12.1 4 0.4 Family situation Arrived with partner 747 11.3 263 28.8 Did not arrived with partner 5845 88.4 316 71.2 No information 21 0.3 525 – Arrived with children 168 2.5 333 30.1 Did not arrive with children 6445 97.5 771 69.9 Current partner not born in country 2500 37.8 452 40.8 Current partner born in country 798 12.1 282 25.5 No partner 3032 45.9 370 33.7 No information 283 4.3 – – Year of arrival Prior to 1990 788 11.9 535 48.3 During the 1990s 1393 21.1 359 32.4 During the 2000s 4432 67.0 210 19.3 Age at arrival 18–25 2317 35.0 488 44.1 26–35 2294 34.7 401 36.2 36–45 1117 16.9 141 12.7 46+ 886 13.4 74 7.0 N 6613 1104 Source: NIS and LNU-UFB. Note 1: Weighted data. The weight compensates for the demographic structure of immigrants in total Sweden and Spain respectively. Note 2: *The category ‘others not in paid work/unknown’ includes students, housewives, retired, and permanently sick. 4.2 Analytical strategy In the first analysis, examining what factors are associated with the first job in care/clearing jobs, we rely on logistic regression. In the second analysis, studying the mobility between sectors, we apply multinomial regression. The multinomial models estimate the likelihood, or relative risk, of currently working in professional/clerk jobs, service jobs and elementary jobs relative to working in care/cleaning jobs. To more fully explore and illustrate the mobility between job sectors, predicted probabilities are calculated for each of the current job sectors by first job sector. Predicted probabilities are calculated as: Py=j=expLj1+expL1+…+expLn Py=refcat=11+expL1+…+expLn where (Lj) equals the estimated logit for each non-reference category (see Liao 1994). 5. Results 5.1 Labour market entry among women who started in the care/cleaning sector Considering the patterns of entry into the labour market among migrant women whose first job in the country of destination was in the care/cleaning sector (Figure 3), we find that in Spain, over 80 per cent of the migrant women found their first job within the first year since arrival, except for women from Africa. These groups have found their first job in the care/cleaning sector within two and three years, respectively. Figure 3. View largeDownload slide Time it took to find the first care/cleaning job in Spain and Sweden by the region of origin. Source: NIS and LNU-UFB. Figure 3. View largeDownload slide Time it took to find the first care/cleaning job in Spain and Sweden by the region of origin. Source: NIS and LNU-UFB. In Sweden, the pattern is quite different and the transition into the first job is much slower. Among women from Western Europe countries and the rest of Europe, about 80 per cent had found their first job within the second year since arrival. About 80 per cent of the women from Latin America found their first job within four years after arrival. For women from Asia it took about six years after arrival to reach the same employment rate, seven years for women from Middle East/North Africa, and about nine years for women from southern Africa. The diverse pattern between Spain and Sweden is highly related to the reason for migration. Significant proportions of those migrating to Sweden from all regions, except for Western Europe countries, are refugees or asylum seekers, or have migrated due to family reasons. While in Spain, work opportunities are a strong pull factor, especially for women from Latin America and Asia, and only a few percentage of the women in the Spanish sample are refugees or asylum seekers.6 5.2 Who started in the care/cleaning sector? In this section we investigate what factors influence the entry into care/cleaning jobs as the first occupation in the country of destination (Table 2) given that this is the largest first job sector among migrant women in both Spain and Sweden (Table 1). Table 2 shows that resources, region of origin, year and age of arrival, social support and family situation have somewhat different impact on the likelihood of entering care/cleaning jobs as the first job in Spain and Sweden. In Spain, those with a compulsory level of education were most likely to start working in the care/cleaning sector. Educational attainment appears to matter less in Sweden in regard to the first care/cleaning job in the country. In this model, the variable ‘highest education attained in country’ is included mainly as a control variable, hence the result should be interpreted with caution given that the first job is more likely to occur than the attainment of education in the host country. Table 2. Logistic regression on the likelihood of first job in the care/cleaning sector (b-coefficients) Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note: Categories with no information/missing values are omitted from the table. Table 2. Logistic regression on the likelihood of first job in the care/cleaning sector (b-coefficients) Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 Spain Sweden Resources     Compulsory level (or less) ref. ref.     Upper secondary level –0.18 * 0.20     Tertiary level –0.74 *** –0.02     Highest education attained in country ref. ref.     Highest education not attained in country 0.49 *** –0.54 *** Region of origin     Western Europe countries ref. ref.     The rest of Europe 2.07 *** 0.64 ***     Middle East/North Africa 1.30 *** 0.34     The rest of Africa 1.54 *** 0.96 **     Asia 2.08 *** 0.75 **     Latin America 2.50 *** 1.03 *** Social support     Had social contacts at arrival ref. ref.     No contacts in country at arrival –0.35 *** 0.07 Family situation     Did not arrive with partner ref. * ref.     Arrived with partner –0.27 ** –0.02     Arrived with children ref. ref.     Did not arrive with children 0.42 (*) –0.14 Year at arrival     Prior to 1990 ref. ref.     During the 1990s –0.53 –0.28     During the 2000s –1.08 0.08 Age at arrival     Age at migration 18–25 ref. ref.     Age at migration 26–35 0.04 0.12     Age at migration 36–45 0.22 ** 0.28     Age at migration 46+ –0.20 (*) –0.26 Constant –2.58 –0.99 Nagelkerke R square 0.18 0.07 –2 LLR 6828.86 *** 1282.65 *** N 6613 1104 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note: Categories with no information/missing values are omitted from the table. Considering ethnic niching of the care/cleaning sector, the result shows that women from Western Europe and other Anglo-speaking countries were less likely to enter care/cleaning jobs as their first job in both Spain and Sweden. In Spain, the ethnic niching of the sector is most noticeable for migrant women from the rest of Europe, Asia and Latin America. A similar pattern is apparent in Sweden, but the differences between women from Western Europe countries and the other regions are more modest. 5.3 Who moves or stays in the care/cleaning sector? The basic hypotheses stated earlier regarding mobility between the first job sector and the current job sector in the host country are tested in multinomial regression models for immigrant women and separately by country of destination. Mobility is analysed in relation to human capital, ethnic niching, and social capital. The model also controls for year of arrival and age at arrival. In the analysis (Table 3) we find a strong association between the first job and the current job in both Spain and Sweden, seen in the high relative risk ratios for the same first and current job sector. This means that migrant women tend to remain in the same sector as they began in the host country. Table 3. Multinomial regression on the likelihood of currently working in other job sectors versus care/cleaning job sector (relative risk ratios) Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note 1: The reference category is Care/cleaning jobs. Note 2: Categories with no information/missing values are omitted from the table. Table 3. Multinomial regression on the likelihood of currently working in other job sectors versus care/cleaning job sector (relative risk ratios) Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 Spain Sweden Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning Prof./ clerks vs. care/ cleaning Service jobs vs. care/ cleaning Elementary jobs vs. care/ cleaning First job sector in Spain/Sweden     First—Care/cleaning jobs 1 1 1 1 1 1     First—Professionals/clerks 21.31 *** 2.68 *** 4.11 *** 92.76 *** 3.97 * 1.80     First—Service jobs 4.35 *** 3.21 *** 1.81 ** 3.13 * 18.52 *** 2.40     First—Elementary jobs 2.43 *** 1.47 * 6.63 *** 3.40 ** 3.38 ** 21.93 *** Resources     Compulsory level (or less) 1 1 1 1 1 1     Upper secondary/vocational level 6.53 *** 1.75 *** 0.82 0.90 0.93 0.52     Tertiary level 44.80 *** 3.20 *** 0.77 15.55 *** 3.21 * 0.45     Highest education attained in country 1 1 1 1 1 1     Highest education not attained in country 0.50 *** 1.04 0.63 (*) 0.85 1.21 2.69 * Region of origin     Western Europe countries 1 1 1 1 1 1     The rest of Europe 0.03 *** 0.14 *** 0.59 1.01 1.09 0.66     Middle East/North Africa 0.10 *** 0.29 *** 1.11 0.69 0.76 0.70     Southern Africa 0.13 *** 0.16 *** 0.98 0.38 1.14 0.09 *     Asia 0.16 *** 0.94 0.44 0.19 *** 0.73 0.29 (*)     Latin America 0.07 *** 0.20 *** 0.42 ** 1.00 1.06 0.51 Social support     Had contact in country at arrival 1 1 1 1 1 1     No contacts in country at arrival 1.48 * 1.29 * 1.12 1.05 1.06 0.73 Family situation     Arrived with children 1 1 1 1 1 1     Did not arrive with children 1.77 2.17 4.38 (*) 0.68 0.85 0.36 *     Partner not born in Spain/Sweden 1 1 1 1 1 1     Partner born in Spain/Sweden 3.16 *** 1.72 ** 1.27 3.61 *** 1.38 3.58 **     No partner 0.84 1.01 0.78 (*) 1.25 1.19 1.35 Year at arrival     Prior to 1990 1 1 1 1 1 1     During the 1990s 8.18 (*) 2.82 7.20 (*) 0.96 0.94 0.89     During the 2000s 5.16 2.89 5.52 0.42 * 0.92 0.99 Age at arrival     18–25 1 1 1 1 1 1     26–35 0.81 0.78 * 0.94 0.80 0.78 1.57     36–45 0.38 *** 0.47 *** 0.90 0.17 ** 0.93 0.33     46+ 0.37 *** 0.44 *** 0.58 * 0.63 0.07 2.31 *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; (*) p ≤ 0.1. Source: NIS and LNU-UFB. Note 1: The reference category is Care/cleaning jobs. Note 2: Categories with no information/missing values are omitted from the table. Table 3 displays the likelihood of working in another job sector versus care/cleaning sector and the main effect of each independent variable, hence the understanding of the mobility is not straightforward. To clarify mobility patterns among migrant women in Spain and Sweden, predicted probabilities are calculated, based on the results from the multinomial regression (Table 3). We are interesting in the combined effects of (i) first job and human capital, (ii) first job and region of origin, and (iii) first job and social capital, on the current job sector. We also analyse the mobility in two directions; mobility from the care/cleaning sector into the other job sectors, and mobility into the care/cleaning sector from the other job sectors. Table 3 suggests that educational attainment and country-specific human capital (highest level of education attained in the host country) have different impact on the current job sector in the two institutional contexts. To fully depict the combined effects of first job and human capital on job sector mobility, predicted probabilities are presented for current job sectors by first job sector, educational attainment and country-specific human capital, holding all other variables at the reference categories (Figure 4). Figure 4. View largeDownload slide The probability of sector mobility by educational attainment and if the highest level of education was attained in Spain/Sweden. Source: NIS and LNU-UFB. Note: Comp. = compulsory level of education. Sec. = secondary level of education. Tert. = tertiary level of education. ES = Spain. SW = Sweden. Figure 4. View largeDownload slide The probability of sector mobility by educational attainment and if the highest level of education was attained in Spain/Sweden. Source: NIS and LNU-UFB. Note: Comp. = compulsory level of education. Sec. = secondary level of education. Tert. = tertiary level of education. ES = Spain. SW = Sweden. Figure 4.17 shows that the probability of remaining in the care/cleaning sector is relatively high among women with less education in both countries (53–76 per cent in Spain and 58–73 per cent in Sweden). Migrant women with tertiary education are most likely to move into other sectors, seen in high overall mobility probabilities; 70–79 per cent in Spain and 78–80 per cent in Sweden. The mobility probabilities among the low educated migrant women are slightly higher in Sweden (33–42 per cent) than in Spain (24–26 per cent). In Spain (Figure 4.1), the country-specific human capital has only a minor impact on mobility among the low-educated migrant women, who are more likely to move to service jobs. It matters more for the middle- and high-educated migrant women. Having attained the highest level of education in Spain increases the probability to move to professional/clerk jobs with about 10 percentage points for the middle-educated and by about 17 percentage points for the high-educated women, when compared with those who did not received their education in Spain. However, the middle-educated women are most likely to move to service jobs, regardless of country-specific human capital. In Sweden (Figure 4.1), low-educated migrant women who did not receive their education in Sweden are more likely to move to elementary jobs, compared with women who attained their education in Sweden. A similar pattern is found also for the middle-educated migrant women. For the high-educated women, having attained the highest level of education in Sweden increases the probability of upward mobility by about 6 percentage points, when compared with those who did not attain their education in Sweden. The mobility from other job sectors into the care/cleaning job sector is more obvious in Spain than in Sweden across all educational groups and regardless of country-specific human capital (Figure 4.2). The probabilities for downward mobility, from professional/clerk jobs into the care/cleaning job sector, is also more apparent in Spain than in Sweden, especially among less-educated migrant women who did not receive their highest level of education in the host country. In Spain, we also see an extensive the lateral mobility, from service jobs and elementary jobs into care/cleaning jobs among low- and middle-educated migrant women. For instance, a low-educated woman, with country-specific human capital, who started in elementary jobs has a 58 per cent probability to move into a care/cleaning job, and if the education is not attained in Spain, this mobility probability is about 63 per cent. The difference in country-specific human capital holds also for women with a secondary level of education. The corresponding mobility probability for the same example of the low-skilled in Sweden is 18 and 9 per cent. Although the lateral mobility probability is much lower in Sweden, the result suggests that country-specific human capital can increase the opportunities for lateral mobility among migrant women in Sweden. This is also true for women with a secondary level of education. Considering the ethnic niching, Table 3 shows that in Spain women from Western Europe countries are less likely to work in care/cleaning jobs than in professional/clerk occupations or service jobs, compared with women from other regions of origin. In Sweden, we do not find the same ethnic niching between women from Western Europe countries and the other regions of origin. To further analyse the ethnic niching, and the occupational mobility among women from different region of origin, predicted probabilities are calculated for each of the current job sectors by first job sector and region of origin, holding all other variables at the reference category (Figure 5). Figure 5. View largeDownload slide The probability of sector mobility by region of origin, in Spain and Sweden. Source: NIS and LNU-UFB. Figure 5. View largeDownload slide The probability of sector mobility by region of origin, in Spain and Sweden. Source: NIS and LNU-UFB. Figure 5:18 shows that the probability of remaining in the care/cleaning sector is very high; 74–95 per cent in Spain and slightly lower in Sweden; 68–86 per cent. Women from Western Europe countries have the highest probability of mobility from the care/cleaning sector into other sectors in Spain, but the differences across regions are more modest in Sweden. In Spain (Figure 5.1), across all regions of origin, women are most likely to move into service jobs, especially women from Western Europe countries and Asia, and the upward mobility (move to professional/clerk jobs) is very low across all regions of origin (0.2–5 per cent). In Sweden, we can observe higher upward mobility probabilities (4–16 per cent), yet less so for women from southern Africa and Asia who generally show very low mobility out of the care/cleaning job sector relative to other regions of origin. The mobility from other job sectors into the care/cleaning job sector is more evident in Spain than in Sweden across all region of origin (Figure 5.2). The probabilities for downward mobility from professional/clerk jobs into care/cleaning jobs is higher in Spain (32–83 per cent) than in Sweden (4–19 per cent). In Spain, women from Western Europe countries display the lowest probabilities for downward mobility (32 per cent) and highest among women from the rest of Europe (83 per cent) and Latin America (78 per cent). Even if the probabilities for downward mobility are low in Sweden, migrant women from Southern Africa and Asia are slightly more likely to experience downward mobility (10 per cent and 19 per cent respectively). In Spain, we also see an extensive lateral mobility, from service jobs and elementary jobs into care/cleaning jobs across all regions, yet less so for women from Western Europe countries and Asia. For instance, women from the rest of Europe and Latin America, who started in elementary jobs have an 85 per cent probability of moving into care/cleaning jobs, and women from southern Africa, who started in the service job sector have an 82 per cent probability of moving into the care/cleaning sector. In Sweden, women from southern Africa and Asia, who started in elementary jobs, have the highest probability of moving into care/cleaning jobs (51 per cent and 42 per cent respectively). The probability to move from the service sector into care/cleaning jobs is slightly higher among women from Asia (40 per cent), compared with women from the other regions. Table 3 suggests that social capital is a relevant factor determining the current job sector. To further analyse the impact of social capital on the occupational mobility, predicted probabilities are calculated for each of the current job sectors by first job sector, social network at arrival and the origin of the current partner. Figure 6.19 shows that the probability of remaining in the care/cleaning sector is more noticeable in Spain (51–74 per cent) than in Sweden (33–68 per cent), regardless of the combined effect of social capital. Figure 6. View largeDownload slide The probability of sector mobility by social contacts and origin of the partner, in Spain and Sweden. Source: NIS and LNU-UFB. Note: ES = Spain; SW = Sweden. Figure 6. View largeDownload slide The probability of sector mobility by social contacts and origin of the partner, in Spain and Sweden. Source: NIS and LNU-UFB. Note: ES = Spain; SW = Sweden. Considering the mobility from care/cleaning jobs into other job sectors, in Spain, it appears to be the lack of social contacts at arrival rather than their partner’s origin that results in higher mobility probabilities from care/cleaning jobs to other sectors (Figure 6.1). Nevertheless, highest probability is found for the mobility to service jobs; especially for women who had no contact when arriving in Spain and had a partner not born in Spain. In contrast, lowest mobility probability is found for women who had social contacts in Spain at arrival and a partner not born in Spain. In Sweden, it is the partner’s origin rather than social contacts at arrival that results in higher mobility probabilities from care/cleaning jobs to other sectors. Having a partner born in Sweden increases the probability of mobility, especially to professional/clerk jobs (Figure 6.1). As previous analysis has shown, the mobility from other job sectors into the care/cleaning job sector, by social capital, is more evident in Spain than in Sweden (Figure 6.2). The probabilities for downward mobility, from professional/clerk jobs into care/cleaning jobs, is higher in Spain (11–32 per cent) than in Sweden (1–4 per cent), regardless of social capital. In Spain, the downward mobility, from professional/clerk jobs into the care/cleaning sector is more linked to their partner’s origin than to social contact at arrival, and it is foremost for women with a partner not born in Spain, who display a higher probability of downward mobility into care/cleaning jobs (Figure 6.2). As we have seen in the other analysis, the downward mobility is very low in Sweden. However, women with a partner not born in Sweden have a slightly higher risk of downward mobility. Considering the lateral mobility, from the service job sector into care/cleaning jobs, we find that the combination of no social contact at arrival and having a partner not born in the host country generate the highest mobility probabilities in both countries. This holds also for the mobility from the elementary jobs into the care/cleaning sector in Sweden. In Spain, social contact at arrival and partner’s origin have similar influence on the mobility from the elementary jobs into the care/cleaning sector (Figure 6.2). 6. Conclusions and discussion The aim of this paper was to examine to what extent the different institutional contexts promote or obstruct the mobility of immigrant women in two countries with different migration and employment regimes. Our main focus has been on migrant women in the care/cleaning sector, with the objective of examining patterns of economic integration, factors influencing the likelihood of working in this sector and the mobility out of and into this sector. The analyses on the transition to the first job (among women whose first job was in the care/cleaning job sector) shows great variation between the countries regarding the time it took the migrant women from different regions to find their first job. In Spain, the majority of the migrant women, regardless of region of origin, found their first job within the first year after the arrival. However, women from Africa appear to face greater difficulties for economic integration, reflected in a longer transition to the first job. In Sweden, the transition to the first job is substantially slower, especially among women from the Middle East/North Africa and Southern Africa, for whom it took 7-9 years until the majority had found their first job. This diverse labour market integration pattern between Spain and Sweden highly related to the different migration regimes. A significant proportion of the immigrants in Sweden are refugees or asylum seekers, or have migrated due to family reasons, and previous studies on Sweden have shown that these factors impact immigrants’ opportunities in regard to labour market entry. In contrast, a very small proportion of the immigrants in Spain are refugees. Instead, various work opportunities have been a strong motivation for migrating to Spain. In the analysis on factors influencing the entry into care/cleaning jobs, we found that resources, region of origin, year and age of arrival, social support and family situation have different impact in Spain and Sweden. For instance, low educated migrant women in Spain is most likely to start working in the care/cleaning sector, while educational attainment have no significant impact on the first job in Sweden. The result also shows a noticeable ethnic niching of the care/cleaning sector in Spain, where women from the rest of Europe, Asia and Latin America are more likely to enter care/cleaning jobs as their first job, compared with migrant women from Western Europe. A similar pattern was observed also for Sweden, but the differences between regions of origin were more modest. This suggests that the care/cleaning sector in Spain evidently is an ethnic divider, between women from Western Europe and women from other regions, while this is less evident in Sweden. We also found that having social contacts in Spain, arriving without a partner, but with children, increase the prospects to find a care/cleaning job in Spain. These factors had no significant impact on the first job in Sweden. The results regarding the mobility out of and into the care/cleaning job sector reveals vast differences between Spain and Sweden. First, the results suggest that country-specific human capital increase the likelihood for upward mobility, from care/cleaning jobs into professional/clerk jobs in both countries. However, the upward mobility is more achievable for the low and high educated in Sweden, especially among those who attained their education in the host country. This can be explained by the fact that by attaining some education in Sweden also includes acquiring knowledge in the Swedish, a language rarely spoken by people not from Sweden. The immobility and lateral mobility (mobility into service jobs or elementary jobs) is more apparent in Spain than in Sweden, except for the low educated migrant women. The mobility into the care/cleaning job sector is more obvious in Spain than in Sweden across all educational groups and regardless of country-specific human capital, which mirrors the high demand for care/cleaning workers in Spain, and migrant women’s labour market opportunities within that context. Second, the results also indicate that there exists a tendency towards ethnic niching, especially in Spain. The mobility out of the care/cleaning sector is relatively low in both countries, except among women from Western Europe/countries, who appears to have greater opportunities to reach a higher position. In Spain, women across all regions tend to move into service jobs. Upward mobility is more feasible in Sweden than in Spain, but migrant women from Southern Africa and Asia appear to face more difficulties to move upwards. In addition, the mobility into the care/cleaning sector is more evident in Spain than in Sweden, regardless of women’s region of origin, again reflecting the high demand of care/cleaning workers in Spain. For the developing countries, this lateral mobility is informing us that the care/cleaning sector is a strategical sector itself because it could be an easy way of earning money to send remittances, but offers these women few mobility opportunities in the longer term. Women who are in this particular situation are from the rest of Europe and Latin America in Spain, and women from Southern Africa and Asia in Sweden. Considering the impact of social capital on the mobility from the care/cleaning sector, the result shows a diverse pattern between Spain and Sweden. In Spain, it is the lack of social contacts at arrival that impacts the mobility from care/cleaning jobs into other sectors, while in Sweden, having a partner born in Sweden increase the likelihood of this mobility. Considering the mobility into the care/cleaning job sector, not having a partner born in the host country increase the risk of downward mobility, from professional/clerk jobs into care/cleaning sector. This is more evident in Spain than in Sweden, where the downward mobility is very low. The combination of no social contact at arrival and having a partner not born in the host country increase the risk of lateral mobility, from the service job sector into care/cleaning jobs. This is also found for the mobility from the elementary jobs into the care/cleaning sector in Sweden, while in Spain, social contacts and partner’s origin have equal influence on the mobility from the elementary jobs into the care/cleaning sector. Based on these results, we conclude that the institutional context is important to understand the underlying dynamics of immigrant women's labour integration process. The Swedish context promotes mobility of immigrant women, while the Spanish contexts obstruct mobility. The Swedish welfare state provides a range of tools to facilitate social integration and reduce difficulties immigrants may encounter in the first stage of their settlement. For instance, through programs such as ‘the Swedish for immigrants’ (SFI), which main aim is to enable migrants to move faster in learning Swedish to study or get a job in Sweden, the Swedish welfare state aim at supporting or accelerate the integration process. SFI is a bridge between the society of origin and Sweden, but also one explanation why immigrants tend to take longer to get their first job. Immigrants not only learn the language but also how the society and the labour market work. This type of national program does not exist in Spain. For many immigrant women, the bridge between their country of origin and Spain is the care and cleaning sector. This sector is a stepping-stone while acquiring knowledge about the labour market and the Spanish society, and, as demonstrated in this paper, this temporary situation becomes the only employment alternative for these women. Funding This work was supported by Riksbankens Jubileumsfond (Grants P11-1111:1) for the research project Do Welfare Regimes Matter? Migration and Care/domestic work in two institutional contexts, Sweden and Spain: A Multi-Tier Design, led Professor Barbara Hobson, Department of Sociology, Stockholm University. This paper has also received funding from the Ministerio de Economía y Competitividad - FEDER: CSO2014-53903-C3-3-R: “Family and Ageing in Spain”. Conflict of interest statement. None declared. Acknowledgements We are especially grateful for comments and suggestions from Barbara Hobson and Rickard Sandell. Special thanks to the anonymous reviewers for their valuable comments that have greatly improved the article. Footnotes 1. The Spanish Data are from the Labour Force Survey, and the Swedish data come from Statistics Sweden’s occupational register (authors’ own calculation). accessed date 2. Due to the space limitation, authors cannot deepen all that they would like in the Spanish and Swedish institutional contexts in regard to the care, migration and employment regimes. For a more complete description, we strongly suggest a reading of Hobson, B., Hellgren, Z., and Bede, L. (2015a) ‘How institutional contexts matter: Migration and domestic care services and the capabilities of migrants in Spain and Sweden.’ Working paper 46. Families and Societies <http://www.familiesandsocieties.eu/wp-content/uploads/2015/11/WP46HobsonEtAl2015.pdf> accessed 31 Jul 2017. 3. Data from the Spanish Municipal Register of Population (Padrón Municipal de Habitantes). 4. See Hobson, B., Hellgren, Z., and Bede, L. (2015a) How institutional contexts matter: Migration and domestic care services and the capabilities of migrants in Spain and Sweden. Working paper 46. Families and Societies, 11–15 <www.familiesandsocieties.eu/wp-content/uploads/2017/03/WorkingPaper46.pdf> accessed 31 Jul 2017. 5. In order to make comparable samples, we have excluded second generation immigrants. While Sweden has a large second generation, the second generation in Spain are still of school age. It is therefore necessary, when comparing immigrant career paths between these two countries to exclude this population. 6. Reason for migration was excluded from the analyses due to high collinearity with region of origin. 7. Figure 4:1 shows mobility from the care/cleaning sector. The probabilities within each bar add up to 100 per cent, and the remaining fraction equals the probability for immobility from the care/cleaning sector. 8. See note 7. 9. See note 7. References Anderson B. ( 2000 ) Doing the Dirty Work? The Global Politics of Domestic Labour . London : Zed Books . Anttonen A. , Sipilä J. ( 1996 ) ‘ European Social Care Services: Is it Possible to Identify Models? Journal of European Social Policy, 6 / 2 : 87 – 100 . Google Scholar Crossref Search ADS Åslund O. , Skans O. N. ( 2005 ) Will I see you at work? Ethnic workplace segregation in Sweden 1985-2002. Working paper 2005:24. Uppsala: The Institute for Labour Market Policy Evaluation (IFAU). <http://www.ifau.se/globalassets/pdf/se/2005/wp05-24.pdf> accessed 31 Jul 2017. Aysa-Lastra M. , Cachón-Rodríguez L. ( 2013 ) ‘ Determinantes de la movilidad ocupacional segmentada de los inmigrantes no comunitarios en España ’, Revista Internacional de Sociología , 71 / 2 : 383 – 413 . Bernardi F. , Martínez-Pastor J. I. ( 2010 ) ‘ Falling at the Bottom: Unskilled Jobs at Entry in the Labour Market in Spain Over Time and in a Comparative Perspective’, International Journal of Comparative Sociology , 51 / 4 : 289 – 307 . Google Scholar Crossref Search ADS Bettio F. , Plantenga J. ( 2004 ) ‘ Comparing Care Regimes in Europe’, Feminist Economics , 10 / 1 : 85 – 113 . Google Scholar Crossref Search ADS Bevelander P. , Irastorza N. ( 2014 ) Catching Up: The Labor Market Outcomes of New Immigrants in Sweden . Washington, DC and Geneva : Migration Policy Institute and International Labour Office Washington, DC and Geneva: Migration Policy Institute and International Labour Office <http://www.migrationpolicy.org/research/catching- labor-market-outcomes-new-immigrants-sweden> accessed 31 Jul 2017. Bygren M. ( 2013 ) ‘ Unpacking the causes of segregation across workplaces’ Acta Sociológica , 56 / 1 : 3 – 19 . Google Scholar Crossref Search ADS Cachón . L. ( 2006 ) ‘Los inmigrantes en el mercado de trabajo’, in Aja, E. and Arango, J (eds) Veinte años de inmigración en España. Perspectiva jurídica y sociológica (1985–2004, pp. 175–201. Spain: CIDOB. Cachón L. ( 2009 ) La“ España inmigrante”: marco discriminatorio, mercado de trabajo y políticas de integración . Anthropos . Chiswick B. R. ( 1978 ) ‘ The Effect of Americanization on the Earnings of Foreign- born Men’, The Journal of Political Economy , 86 / 5 : 897 – 921 . Google Scholar Crossref Search ADS Chiswick B. R. , Miller P. W. ( 2003 ) ‘ The Complementarity of Language and Other Human Capital: Immigrant Earnings in Canada’, Economics of Education Review , 22 / 5 : 469 – 80 . Google Scholar Crossref Search ADS Da Roit B. , Weicht B. ( 2013 ) ‘ Migrant Care Work and Care, Migration and Employment Regimes. A Fuzzy Set Analysis’, Journal of European Social policy , 23 / 5 : 469 – 86 . Google Scholar Crossref Search ADS Da Roit B. , González Ferrer A. , Moreno-Fuentes F. J. ( 2013 ) ‘ The Southern European Migrant-Based Care Model: Long-Term Care and Employment Trajectories in Italy and Spain’, European Societies , 15 / 4 : 577 – 96 . Google Scholar Crossref Search ADS Dickens W. T. , Lang K. ( 1993 ) Labor Market Segmentation Theory: Reconsidering the Evidence . Netherlands : Springer, 141–80 . Erel U. ( 2010 ) ‘ Migrating Cultural Capital: Bourdieu in Migration Studies ’, Sociology , 44 / 4 : 642 – 60 . Google Scholar Crossref Search ADS Estévez-Abe M. ( 2015 ) ‘ The Outsourcing of House Cleaning and Low Skill Immigrant Workers ’, Social Politics , 22 / 2 : 147 – 69 . Google Scholar Crossref Search ADS Fahlén S. , Hobson B. , Sánchez-Domínguez M. ( 2015 ) What Do Private Markets in Domestic Services Do? Widening gap in inequalities in Spain and Sweden. Paper presented at the Annual RC19 conference in Bath, UK, 26–28 August 2015. Fernández-Macías E. , Grande R. , del Rey Poveda A. , Antón J. I. ( 2015 ) ‘ Employment and occupational mobility among recently arrived immigrants: The Spanish case 1997–2007’, Population Research and Policy Review , 34 / 2 : 243 – 77 . Google Scholar Crossref Search ADS Fernández C. , Ortega C. ( 2006 ) ‘Asimilación laboral de los inmigrantes en España: ¿Trabajo a costa de malos empleos?’, in Dolado J. J. , Vázquez P. (eds) Ensayos sobre los efectos económicos de la inmigración en España , pp . 52 – 68 . Madrid : Fundación de Estudios de Economía Aplicada . Gavanas A. ( 2013 ) ‘ Migrant Domestic Workers, Social Network Strategies and Informal Markets for Domestic Services in Sweden’, Women’s Studies International Forum , 36 : 54 – 64 . Google Scholar Crossref Search ADS Government of Sweden ( 2006 ) Proposition 2006/07: 94. Skattelättnader för hushållstjänster, m.m. [Tax Relief for Household Services, etc.] <https://www.skatteverket.se/download/18.6efe6285127ab4f1d2580006907/1359707449188/15209.pdf> accessed 31 Jul 2017. Hellgren Z. ( 2015 ) ‘ Markets, Regimes, and the Role of Stakeholders: Explaining Precariousness of Migrant Domestic/Care Workers in Different Institutional Frameworks’, Social Politics , 22 / 2 : 220 – 41 . Google Scholar Crossref Search ADS Hobson B. , Hellgren Z. , Bede L. ( 2015a ) How Institutional Contexts Matter: Migration and Domestic Care Services and the Capabilities of Migrants in Spain and Sweden. Working paper 46 (2015). Families and Societies. <http://www.familiesandsocieties.eu/wp-content/uploads/2015/11/WP46HobsonEtAl2015.pdf> accessed 31 Jul 2017. Hobson B. , Hellgren Z. , Serrano I. ( 2015b ) How Private Markets Work in Two Divergent Welfare Regimes: Shaping the Capabilities and Agency of Migrant Care/Domestic Workers in Sweden and Spain. Paper presented at the 22d CES Conference at Sciences Po in Paris, 8–10 July 2015. ILO ( 2013 ) Domestic Workers Across the World: Global and Regional Statistics and the Extent of Legal Protection. International Labour Organization. <www.ilo.org/wcmsp5/groups/public/–-dgreports/–-dcomm/–-publ/documents/publication/wcms_173363.pdf> accessed 9 Dec 2014. INE ( 2016a ) Activos por nacionalidad, sexo y ocupación [Active population of foreign nationality by sex and occupation. <http://www.ine.es/jaxiT3/Tabla.htm?t=4080andL=0> accessed 17 Mar 2017. INE ( 2016b ) Padrón Municipal de Habitantes [Spanish Municipal Register of Population]. <http://www.ine.es/dyngs/INEbase/es/operacion.htm?c=Estadistica_C&cid= 1254736177012&menu=ultiDatos&idp=1254734710990> accessed 31 Jul 2017. Isaksen L. W. (ed.) ( 2010 ) Global Care Work—Gender and Migration in Nordic Societies . Lund : Nordic Academic Press . Kalmijn M. ( 1998 ) ‘ Intermarriage and Homogamy: Causes, Hatterns, Trends’, Annual Review of Sociology , 24 / 1 : 395 – 421 . Google Scholar Crossref Search ADS Kofman E. , Raghuram P. ( 2006 ) ‘ Gender and Global Labour Migrations: Incorporating Skilled Workers’, Antipode , 38 / 2 : 282 – 303 . Google Scholar Crossref Search ADS Korpi W. ( 2000 ) ‘ Faces of Inequality: Gender, Class, and Patterns of Inequalities in Different Types of Welfare States . Social Politics , 7 / 2 : 127 – 91 . Google Scholar Crossref Search ADS Kvist E. , Peterson E. ( 2010 ) ‘ What Has Gender Equality Got to Do with It? An Analysis of Policy Debated Surrounding Domestic Services in the Welfare State of Spain and Sweden’, Nordic Journal of Feminist and Gender Research , 18 / 3 : 185 – 203 . Google Scholar Crossref Search ADS León M. ( 2010 ) ‘ Migration and Care Work in Spain: The Domestic Sector Revisited’, Social Policy and Society , 9 / 3 : 409 – 18 . Google Scholar Crossref Search ADS Liao T. F. ( 1994 ) Interpreting Probability Models: Logit, Probit, and Other Generalized Linear Models. Sage University Paper series on Quantitative Application in the Social Sciences, 07-101. Thousand Oaks, CA: Sage Publications, Inc. Le Grand C. et al. ( 2013 ) Vid arbetslivets gränser: Sysselsättning, matchning, barriärer 1974–2010. Underlagsrapport nr 12. Stockholm: Parlamentariska socialförsäkringsutredningen (S 2010:04) <www.psfu.se/uploads/files/12_vid_arbetslivets_granser.pdf> accessed 23 May 2014. Lutz H. ( 2008 ) Migration and Domestic Work – A European perspective on Global Theme. Aldershot: Ashgate Publishing Limited. Massey D. S. et al. ( 1994 ) ‘ An Evaluation of International Migration Theory: The North American Case’, Population and Development Review , 20 : 699 – 751 . Google Scholar Crossref Search ADS Nergård Larsen E. ( 2016 ) Ethnic Discrimination, the Structure of the Labour Market and Economic Conditions <http://gemm2020.eu/wp-content/uploads/2016/12/Report-Ethnic-Discrimination-The-Structure-of-the-Labour-Market-and-Economic-Condistions-1.pdf> accessed 31 Jul 2017. Parreñas R. S. ( 2001 ) Servants of globalization: Women, migration, and domestic work . Palo Alto, CA : Stanford University Press . Pereira S. ( 2013 ) Replacement Migration and Changing Preferences: Immigrant workers in Cleaning and Domestic Service in Portugal . Journal of Ethnic and Migration Studies , 39 / 7 : 1141 – 58 . Google Scholar Crossref Search ADS Pereira S. , Snel E. , ’t Hart M. ( 2015 ) ’Economic Progress, Stagnation, or Decline? Occupational Mobility of Non-Eu Immigrants in Europe ’, Immigration and Work , 27 : 129 – 65 . Google Scholar Crossref Search ADS Portes A. ( 2000 ) ‘ Globalization from below: the rise of transnational communities’, The ends of globalization: bringing society back in , 253 – 70 . Portes A. , Rumbaut R. ( 1998 ) Immigrant America: A Portrait . Berkeley, Los Angeles : University of California Press . Portes A. , Rumbaut R. G. ( 2001 ) Legacies: The story of the immigrant second generation . Berkeley : University of California Press and Russell Sage Foundation . Portes A. , Sensenbrenner J. ( 1993 ) ‘ Embeddedness and Immigration: Notes on the Social Determinants of Economic Action’, American Journal of Sociology , 98 / 6 : 1320 – 50 . Google Scholar Crossref Search ADS Pereira S. ( 2000 ) ‘Social Capital: Its Origins and Applications in Modern Sociology’, in Lesser E. L. (ed.) Knowledge and Social Capital , pp. 43 – 67 . Boston : Butterworth-Heinemann . Pereira S. , Rumbaut R. G. ( 1998 ) Immigrant America: A Portrait . Berkeley : University of California Press . Powers M. G. , Seltzer W. ( 1998 ) ‘ Occupational Status and Mobility among Undocumented Immigrants by Gender’, International Migration Review , 32 ( 1 ): 21 – 55 . Google Scholar Crossref Search ADS PubMed Powers M. G. , Seltzer W. , Shi J. ( 1998 ) ‘ Gender Differences in the Occupational Status of Undocumented Immigrants in the United States: Experience Before and after Legalization ’, International Migration Review , 32 ( 4 ): 1015 – 1046 . Google Scholar Crossref Search ADS Raijman R. , Semyonov M. ( 1997 ) ‘ Gender, Ethnicity, and Immigration: Double Disadvantage and Triple Disadvantage among Recent Immigrant Women in the Israeli Labor Market’, Gender and Society , 11 / 1 : 108 – 25 . Redstone Akresh I. ( 2006 ) Occupational mobility among legal immigrants to the United States. International Migration Review , 40 / 4 : 854 – 84 . Google Scholar Crossref Search ADS Reher D. S. et al. ( 2008 ) Informe Encuesta Nacional de Inmigrantes (ENI-2007). Documentos de Trabajo. 2(08). Reher D. , Requena M. ( 2009 ) ‘ The National Immigrant Survey of Spain: A new data source for migration studies in Europe’, Demographic Research , 20 : 253 . Google Scholar Crossref Search ADS Requena M. , Sánchez-Domínguez M. ( 2011 ) ’ Las familias inmigrantes en España’, Revista Internacional de Sociología (RIS) , 69 / M1 : 79 – 104 . Google Scholar Crossref Search ADS Reyneri E. , Fullin G. ( 2011 ) ‘ Labour Market Penalties of New Immigrants in New and Old Receiving West European Countries. International Migration , 49 / 1 : 31 – 57 . Google Scholar Crossref Search ADS Riaño Y. , Baghdadi N. ( 2007 ) ‘ Understanding the Labour Market Participation of Skilled Immigrant Women in Switzerland: The Interplay of Class, Ethnicity, and Gender’, Journal of International Migration and Integration/Revue de l'integration et de la migration internationale , 8 / 2 : 163 . Google Scholar Crossref Search ADS Rodríguez-Planas N. , Nollenberger N. ( 2014 ) A Precarious Position: The Labor Market Integration of New Immigrants in Spain . Washington, DC and Geneva : Migration Policy Institute and International Labour Office . <www.migrationpolicy.org/research/precarious-position-labor-market-integration-new-immigrants-spain> accessed 30 September 2014. Rydgren J. ( 2004 ) ‘ Mechanisms of Exclusion: Ethnic Discrimination in the Swedish Labour Market . Journal of Ethnic and Migration Studies , 30 / 4 : 697 – 716 . Google Scholar Crossref Search ADS Sanchez Dominguez M. ( 2011 ) Estrategias matrimoniales y procesos de integración social de los inmigrantes en España (Doctoral dissertation, Universidad Complutense de Madrid) . Sánchez-Domínguez M. ( 2016 ) ‘ Los senderos hacia la integración: matrimonios interétnicos en España. En Miyar-Busto, M “El nuevo escenario migratorio en España” ’, Panorama Social , 24 : 73 – 88 . Sánchez-Domínguez M. , De Valk H. , Reher D. ( 2011 ) ‘ Marriage Strategies among Immigrants in Spain’, Revista Internacional de Sociología , 69 / M1 : 139 – 66 . Google Scholar Crossref Search ADS Samers M. ( 2014 ) ‘How to Understand the Incorporation of Immigrants in European Labour Markets’, in Martiniello M. , Rath J. (eds), An Introduction to Immigration Studies , pp. 171 – 94) . Amsterdam : Amsterdam University Press . Schrover M. , Van der Leun J. , Quispel C. ( 2007 ) ‘ Niches, Labour Market Segregation, Ethnicity and Gender’, Journal of Ethnic and Migration Studies , 33 / 4 : 529 – 40 . Google Scholar Crossref Search ADS Simonazzi A. ( 2009 ) ‘ Care Regimes and National Employment Models ’, Cambridge Journal of Economics , 33 / 2 : 211 – 32 . Google Scholar Crossref Search ADS Stanek M. ( 2009 ) Los inmigrantes polacos en la Comunidad de Madrid y su inserción laboral. Doctoral thesis. Madrid: Universidad Complutense de Madrid. Stanek M. ( 2011 ) ‘Nichos étnicos y movilidad socio-ocupacional. El caso del colectivo polaco en Madrid [Ethic Niches and Socio-Occupational Mobility. The Case of the Poles in Madrid], REIS, 135 : 69 – 88 . Stanek M. , Veira A. ( 2012 ) ‘ Ethnic Niching in a Segmented Labour Market: Evidence from Spain ’, Migration Letters , 9 / 3 : 249 – 62 . Statistics Sweden ( 2015a ) Invandring efter år [Immigration by Year]. <www.statistikdatabasen.scb.se> accessed 18 Dec 2015. Statistics Sweden ( 2015b ) Befolkningen 15–74 år (AKU), 1000-tal efter arbetskraftstillhörighet, inrikes/utrikes född, kön, ålder och år [Population aged 15–74 (LFS), 1000s after Working Status, Domestic/Foreign Born, Sex, Age and Year]. <www.statistikdatabasen.scb.se> accessed 18 Dec 2015. Statistics Sweden ( 2015c ) Kriget i Syrien medförde rekordstor invandring [The War in Syria Led to Record-High Immigration]. Number 2015: 106, published 7 April 2015. <https://www.scb.se/sv_/Hitta-statistik/Artiklar/Kriget-i-Syrien-medforde-rekordstor-invandring/> accessed 31 Jul 2017. Statistics Sweden ( 2015d ) Antal personer efter region, utländsk/svensk bakgrund och år. [Number of Persons by Region, Foreign/Swedish Background and Year]. <www.statistikdatabasen.scb.se> accessed 18 Dec 2015. Statistics Sweden ( 2015e ) Utrikes födda efter födelseland, år och ålder [Foreign Born by Country of Birth, Year and Age. <www.statistikdatabasen.scb.se> accessed 18 Jun 2015. Strauss K. , McGrath S. ( 2017 ) ‘ Temporary Migration, Precarious Employment and Unfree Labour Relations: Exploring the Continuum of Exploitation in Canada’s Temporary Foreign Worker Program ’, Geoforum , 78 : 199 – 208 . <http://doi.org/10.1016/j.geoforum.2016.01.008> accessed 31 Jul 2017. Google Scholar Crossref Search ADS Swedish Migration Agency ( 2015 ) Beviljade uppehållstillstånd 1980–2014 [Granted Permits of Residence 1980–2014]. <www.migrationsverket.se/Om-Migrationsverket/Statistik/Oversikter-och-statistik-fran-tidigare-ar.html> accessed 18 Jun 2015. van Riemsdijk M. ( 2013 ) ‘ Everyday Geopolitics, the Valuation of Labour and the Socio-Political Hierarchies of Skill: Polish Nurses in Norway’, Journal of Ethnic and Migration Studies , 39 / 3 : 373 – 90 . Google Scholar Crossref Search ADS Vidal-Coso E. ( 2009 ) Activitat i complementarietat sociodemográfica entre les dones immigrades i les no immigrades a Espanya. Doctoral dissertation. [Tesis Doctoral, Universitat Autónoma de Barcelona, Facultat de Filosofia i Lletres, Departament de Geografia, Bellaterra]. Vidal-Coso E. , Miret-Gamundi P. ( 2014 ) ‘ The Labour Trajectories of Immigrant Women in Spain: Are There Signs of Upward Social Mobility?’, Demographic Research , 31 / 13 : 337 – 80 . Google Scholar Crossref Search ADS Wennesjö A. ( 2013 ) Flera skäl till invandring idag [Several Reasons for Immigration Today]. Välfärd 1/2013. Stockholm: Statistics Sweden. <www.scb.se/valfard> accessed 31 Jul 2017. © The Authors 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Migration StudiesOxford University Press

Published: Nov 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off