Poverty among migrant, mixed, and non-migrant households: the role of non-teleworkability and single-earnership in Germany

Hornung, Maria; Stuffolino, Emanuela; Zagel, Hannah Article — Published Version Poverty among migrant, mixed, and non-migrant households: the role of non-teleworkability and singleearnership in Germany Journal of Ethnic and Migration Studies Provided in Cooperation with: WZB Berlin Social Science Center Suggested Citation: Hornung, Maria; Stuffolino, Emanuela; Zagel, Hannah (2025) : Poverty among migrant, mixed, and non-migrant households: the role of non-teleworkability and single-earnership in Germany, Journal of Ethnic and Migration Studies, ISSN 1469-9451, Taylor & Francis, London, Vol. 51, Iss. 5, pp. 1294-1321, https://doi.org/10.1080/1369183X.2024.2404219 This Version is available at: https://hdl.handle.net/10419/310920.2 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0/ Poverty among migrant, mixed, and non-migrant households: the role of non-teleworkability and singleearnership in Germany Maria Hornung a , Emanuela Stuffolino b and Hannah Zagel c a Department of Social Sciences, Humboldt Universität zu Berlin, Berlin, Germany; b Dipartimento di Scienze Sociali e Politiche, Universita degli Studi di Milano, Milan, Italy; c Wissenschaftszentrum Berlin für Sozialforschung (WZB), Berlin, Germany ABSTRACT Migrant and mixed households have higher poverty than nonmigrant households. This is partly because single-earner twoadult households are more prevalent in migrant and mixed households and because such households have different job characteristics. One crucial job characteristic is teleworkability. Whether or not individuals can work from home has become a dividing factor in the labour market. While much research has focused on how teleworkability affects poverty in the majority population, less attention has been devoted to migrant and mixed two-adult households. Using the German Microcensus (2019), we construct work arrangements based on the number of earners in the household and their job‘s teleworkability to predict poverty for non-migrant (N = 49,507), mixed (N = 6,818), and migrant households (N = 8,922). Descriptive statistics show that, in Germany, migrant and mixed households have more singleearner and non-teleworkable work arrangements. Results from logistic regressions report higher poverty for non-teleworkable and single-earner work arrangements, putting mixed and migrant households at an increased disadvantage. Furthermore, we find that migrant (and mixed) households not only have a higher prevalence of high-poverty work arrangements but also higher poverty than non-migrant and mixed households within the same work arrangements. ARTICLE HISTORY Received 18 September 2023 Accepted 9 September 2024 KEYWORDS Teleworkability; labour market; poverty; mixed households; migration Introduction Migrants, defined here as individuals who leave their country of birth to live elsewhere, are an economically vulnerable group and face higher poverty than non-migrants in Europe (Giesecke et al. 2017; Kesler 2015). As poverty is an important obstacle to integration into society and the labour market (Barnes et al. 2002), high poverty levels are particularly problematic for migrants who lack the country-specific capital that would © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group CONTACT Maria Hornung [email protected] Supplemental data for this article can be accessed online at https://doi.org/10.1080/1369183X.2024.2404219. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/ licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. JOURNAL OF ETHNIC AND MIGRATION STUDIES 2025, VOL. 51, NO. 5, 1294–1321 https://doi.org/10.1080/1369183X.2024.2404219 facilitate their participation in wider society. While employment is important in protecting individuals against poverty, migrants’ poverty exceeds that of non-migrants, even when they are employed (Crettaz 2018; Lohmann 2009). In-work poverty has increased in recent decades, due to worsening labour-market conditions related to the increase of atypical forms of employment such as part-time, temporary, and self-employment (Filandri and Struffolino 2019). One of this study’s key motivations is to find out how this plays out for migrants’ poverty. Furthermore, digitalisation and technological advances have made teleworkability a new potential source of inequality in the labour market. This has become more visible since the Covid-19 pandemic. Jobs can be considered teleworkable if they do not require the worker to primarily work on the employer’s premises – examples include insurance agents or business analysts. In contrast, non-teleworkable jobs – like shop assistant or manufacturing worker positions – are mainly tied to a specific workplace. Differences in the degree of teleworkability have implications for skill requirements and employment demand (Bihagen et al. 2021). Previous research has shown that migrants are less likely to hold or obtain a teleworkable job than non-migrants (Fasani and Mazza 2020), suggesting that non-teleworkability is one driver of differences in poverty between migrants and non-migrants. Labour market inequalities stemming from digitalisation and technological advances often create a ‘digital divide’ (Messenger et al. 2017), and the question of how this divide affects migrants in terms of poverty outcomes is an open one. Poverty is typically measured on the household level, on the assumption that members pool their incomes. Therefore, household composition, especially regarding the number of earners, is important in assessing a household’s economic situation (Biegert and Ebbinghaus 2022; Brady, Finnigan, and Hübgen 2017). As the dual-earner model has become more common, single-earner two-adult households are increasingly exposed to higher poverty (Nieuwenhuis et al. 2020; Tamayo and Popova 2020). This study discusses having a non-teleworkable job and being in a single-earner two-adult household as poverty risks and considers their interaction in defining the different exposure to poverty for non-migrant, mixed, and migrant households. We examined the association between poverty and households’ working arrangements, as defined by the number of earners in the households alongside the teleworkability of the earners‘ jobs jointly. If one of two earners held a teleworkable job, this may avert poverty at the household level. On the other hand, poverty risks accumulate if the sole earner holds a non-teleworkable job. Migrants’ poverty risks can accumulate in households for reasons that go beyond job status and job characteristics. Previous research reports that households with one migrant and a non-migrant (henceforth referred to as mixed households) are less exposed to poverty than households with two migrants (Giesecke et al. 2017; Kesler 2015). Mixed households’ economic advantages may be linked to the non-migrant partner compensating for the migrant partner’s poverty risk or to a positive selection of migrants into mixed households. Therefore, our differentiation of household types considers that, in couple households, migrants can be either partnered with another migrant or with a non-migrant. Investigating differences between migrant, mixed, and non-migrant households, we contribute to research on the economic consequences of teleworkable or non-teleworkable jobs, which rarely considers household context and migrant status beyond control variables. JOURNAL OF ETHNIC AND MIGRATION STUDIES 1295 The context of our study is Germany, which has one of the largest immigrant populations in Europe (Destatis 2023) and is the largest economy in the European Union (Eurostat 2024). This means we can provide insights into a significant share of the European workforce. In light of the historical low-skilled immigration patterns to Germany from Southern Europe, Turkey, and Northern Africa post-WWII and given the substantial presence of migrant workers in manufacturing, the German labour market is an excellent context to monitor persistent economic differences between migrants and non-migrants. This is because digitalisation and advances in information and communication technologies (ICT) have reduced the need for routine manual workers and negatively impacted low-skilled production and manufacturing workers’ labour market opportunities (Hötte, Somers, and Theodorakopoulos 2023; Wiedner and Giesecke 2022). In 2019, Germany’s degree of teleworkability, measured as the share of employees who usually or sometimes perform telework, was close to the EU average (Sostero et al. 2020). In this ranking, Sweden has the highest prevalence of teleworkability and Bulgaria the lowest. Using large-scale representative data from the German Microcensus (2019), we construct household-level work arrangements by including the number of earners and the teleworkability of the earners’ jobs. We measure teleworkability using a novel index of the teleworkability of occupations (Gädecke et al. 2021) based on a task-focused employee survey in Germany. The questions guiding this article are: a.) How are work arrangements distributed among migrant, mixed, and non-migrant households? and b.) How does poverty vary by work arrangement for migrant, mixed, and non-migrant households? The analysis consists of three parts. First, we map differences in work arrangements between migrant, mixed, and non-migrant households. Second, we estimate poverty for migrant, mixed, and non-migrant households. Finally, we investigate how poverty varies for migrant, mixed, and non-migrant households by work arrangements that account for teleworkability and the number of earners in the household. Migrants’ poverty risk In European countries, migrants are exposed to higher poverty than non-migrants (Kesler 2015). Although poverty differentials between migrants and non-migrants are partially driven by higher unemployment rates, employed migrants also have higher poverty levels than non-migrants (Crettaz 2018; Lohmann 2009). Migrants’ in-work poverty is driven by their weak position in the labour market, which is often ascribed to individual-level factors, such as human capital characteristics and discrimination, or to aspects of the labour market structure. Individual-level factors One often-mentioned factor is that, on average, migrants have lower educational attainment than non-migrants. Previous studies have found that migrants’ allocation to lowwage employment can often be explained when controlling for educational attainment (Granato and Kalter 2001). More recent data, however, have shown an increase in the number of migrants with medium and high education levels throughout the last decades, which has altered the historical prevalence of low-skilled migrants in 1296 M. HORNUNG ET AL. Germany. Although this trend in educational attainment points to potential improvements in migrants’ economic situation, human capital is not easily transferable across countries, and migrants face difficulties in getting their qualifications recognised (Sommer 2021). Migrants often lack country-specific human capital, such as language skills or cultural knowledge, which means that their educational and vocational attainment is less valued (Konietzka and Kreyenfeld 2002). This contributes to their placement in the lower strata of the labour market. In addition, migrants may employ different human capital investment strategies than non-migrants. As migrants may plan to return to their countries of origin and prefer prompt financial returns over long-term financial gains, they could select low-status or precarious employment and not invest in receiving country-specific human capital ). Second, employer discrimination is a barrier for migrants in accessing (strong) labour market positions. Field experiments that involve sending applications for real jobs have identified ethnic, religious, and racial discrimination dynamics. Individuals from countries with substantial Muslim populations (Di Stasio et al. 2021), veiled women (Weichselbaumer 2016), foreign-born minorities, and minorities from culturally very distant countries (Veit and Thijsen 2021) get lower call-back rates from employers than the majority population in Germany. However, discrimination varies between countries and groups. For example, people from Turkish migration backgrounds face less hiring discrimination in Germany than in the Netherlands (Thijssen et al. 2021). Labour market structure Labour market structures are a further factor explaining migrants’ economically poor position. The German labour market has a dual structure, meaning it differentiates between labour market insiders and outsiders. Insider employment is stable, well-paid, and offers opportunities for professional development, while outsider employment is casual, temporary, poorly paid, and associated with higher poverty levels. These two labour markets are not permeable, and segregation can be observed across and within industries and firms. Previous studies report high occupational segregation for migrants in Germany. Migrants are: i) overrepresented among labour market outsiders (Constant and Massey 2005); ii) more likely to work in blue-collar occupations and in the hospitality and restaurant industries (Drever and Hoffmeister 2008); and iii) more likely to be employed in more volatile sectors with less secure seasonal and temporary employment (Bogoeski 2022). As the labour market structure has changed due to deindustrialisation, tertiarisation, and decreasing labour market regulation, so-called atypical jobs have become more prevalent, and the low-wage sector has expanded. This development has further resulted in an increase in temporary contracts, part-time work, irregular working hours, and poorly protected employment. Since low work intensity and low pay are the main drivers of poverty, employment has partly lost its protective effect. In addition, the emergence of ICT has provoked new discussions on labour market segregation along a digital divide (Bihagen et al. 2021; Messenger et al. 2017). More emphasis is being put on skilled labour, meaning that low– and medium-skilled workers are being assigned less value (Acemoglu and Autor 2011). A study on Germany confirms that migrants from Turkey have been disadvantaged by Germany’s educational expansion and structural JOURNAL OF ETHNIC AND MIGRATION STUDIES 1297 labour market changes and experienced worsening labour market positions due to their low skill sets and historically high employment in manufacturing positions (Wiedner and Giesecke 2022). And in the Covid-19 pandemic and the associated lockdowns, workers in a non-teleworkable job were at risk of poverty as non-teleworkable jobs had higher furlough rates (Adams-Prassl et al. 2020; Fasani and Mazza 2020). Previous studies have suggested that teleworkability is not evenly distributed in the labour market; migrants less frequently work in teleworkable jobs (Alipour et al. 2021; Fasani and Mazza 2020). For Germany, research has shown that being a migrant is negatively associated with always or frequently working from home (Alipour et al. 2021). In the pandemic, a discussion emerged over whether working from home primarily reflects existing labour market inequalities or creates a new divide among workers (Sostero et al. 2020). Teleworkability reflects previous labour market advantages, in that individuals who can work from home tend to be highly educated, to have more work experience, to work in higher-paid and higher-level occupations, to have permanent work contracts, to work full-time, and to be more autonomous (Alipour et al. 2021; Brussevich, Dabla-Norris, and Khalid 2020; Sostero et al. 2020). Further factors associated with teleworkability include factors related to the work organisation, firm size, the level of employer’s trust, and the time spent commuting to work (Sostero et al. 2020). In the pandemic, heterogeneity in terms of various jobs teleworkability potential became especially visible (Fasang, Struffolino, and Zagel 2023; Fasani and Mazza 2020). In households with two adults, poverty declines considerably if both are employed (Tamayo and Popova 2020). Therefore, the rise in female employment in OECD countries in recent decades has attenuated increasing poverty levels by providing households with a second earner (Nieuwenhuis et al. 2020). Nevertheless, the second earner is typically employed in less secure, less typical, and lower-paid jobs, minimising the possible effects of poverty reduction. Furthermore, the growth in female employment is selective and mainly occurs in households which already engage in employment. Consequently, partners’ homogamy in employment prevents substantial poverty reductions between households through women’s labour force participation, which increases polarisation between employment-intense and jobless households (Gregg and Wadsworth 2008). A theoretical explanation for employment intensity in migrant households is the family investment hypothesis, which assumes that newly arrived migrants must invest in receiving-country-specific human capital, financed by the family. Accordingly, one partner (in different-sex couples, this is mostly the man) invests in education or job training while the other partner works in dead-end jobs. For Germany, Basilio, Bauer, and Sinning (2009) have not found evidence supporting the family investment hypothesis, as partners’ wages increased at similar rates with time spent in the destination country. However, other studies in Germany have reported gender differences in migrant employment as migrant women are, on average, less often employed than migrant men and non-migrant women (Fleischmann and Höhne 2013; Salikutluk, Giesecke, and Kroh 2020). Furthermore, migrant women are often underemployed or do not find adequate employment, especially when they have small children (Rubin et al. 2008). Taken together, these factors indicate that migrant households exhibit lower labour market attachment, partly driven by women’s lower employment rates. 1298 M. HORNUNG ET AL. To summarise theoretical considerations and previous empirical findings, we could expect differences in poverty between migrant and non-migrant households to be driven by migrants’ lower likelihood of having a teleworkable job and their higher prevalence of single-earner work arrangements. Mixed households’ poverty risk Compared to migrant couples, mixed households where one person is a migrant and the other is a non-migrant have lower poverty rates (Giesecke et al. 2017; Kesler 2015); sometimes, they even have lower poverty rates than non-migrant households (Bostic and Hyde 2023). The influence of the non-migrant partner explains some of the economic advantages of migrants in mixed households. A non-migrant partner facilitates access to non-migrant social networks and enables faster economic integration into the receiving country. For Sweden, Dribe and Lundh (2008) have found that, for migrants, being married to a non-migrant is positively associated with higher employment rates and higher individual and household income. Using Danish longitudinal data and distributed fixed effects, Elwert and Tegunimataka (2016) showed that cohabiting with a native Dane positively affects migrants’ incomes. Meng and Meurs (2009) and Meng and Gregory (2005) have found intermarriage premiums for migrants who intermarry in France and Australia. To some extent, the better labour market outcomes of migrants in mixed unions have been attributed to selection (Kantarevic 2004), which means migrants with specific characteristics are particularly likely to partner with non-migrants. Furthermore, migrants in mixed unions may differ from migrants in migrant unions in terms of both their partnership status when migrating and the length of the relationship at the moment of the interview. In Germany, gender, country of birth, and religious affiliation explain selection patterns into mixed partnerships (Haug 2010; Schroedter 2013). Migrants in mixed unions have, on average, higher educational levels, better countryspecific language skills, and longer residence in the receiving country (Haug 2010; Schroedter 2013). Some studies have found that the positive effect of intermarriage disappears when accounting for selection (Nottmeyer 2011), while others report a positive effect of intermarriage on labour market outcomes, even when considering selection (Elwert and Tegunimataka 2016). From a household perspective, studies in Germany have shown a lower prevalence of dual-earners among mixed households compared to non-migrant ones (Braack, Milewski, and Trappe 2022; Nottmeyer 2011), but dual earners are still more common in mixed than in migrant households (Nottmeyer 2011). One explanation for this is human capital differences. For instance, in Germany, the migrant and the nonmigrant partner in mixed households have more similar levels of education than migrant households (Nottmeyer 2011). As most studies take an individual approach when examining migrants in mixed unions, they disregard the role of the non-migrant partner in compensating for poverty. Nevertheless, non-migrants who intermarry seem to be selective. Evidence from Spain has shown that non-migrant men are more likely to intermarry if they are unemployed and low-skilled, suggesting a negative selection of non-migrant men into mixed households; this cannot be found for non-migrant women (González-Ferrer et al. 2018). However, non-migrants often are sole earners in mixed households (Braack, Milewski, and Trappe 2022). Therefore, we might expect JOURNAL OF ETHNIC AND MIGRATION STUDIES 1299 that, although non-migrants in mixed households (especially men) will be negatively selected, they will not face the same labour market barriers as their migrant partners. Hypotheses In the first step of the present study, we will identify the distribution of work arrangements, accounting for teleworkability and the number of earners for migrant, mixed, and non-migrant households. We expect the prevalence of dual-earner households to be the highest among non-migrant households, in the middle among mixed households, and lowest among migrant households (Hypothesis 1a). Further, we expect to find the lowest share of work arrangements with teleworkable jobs for migrant households, followed by mixed households (Hypothesis 1b). These hypotheses are based on two insights from the existing literature discussed above. First, migrants are more likely than nonmigrants to have jobs in the lower strata of the labour market and, consequently, are less likely to hold a teleworkable job. Second, migrants in mixed households are positively selected and display higher educational homogamy than migrant households, suggesting higher labour force participation and higher levels of teleworkability. In the second step, to examine differences in poverty, we will map predicted poverty probabilities for migrant, mixed, and non-migrant households. In line with previous studies, we expect migrant households to experience the highest poverty, followed by mixed households and non-migrant households (Hypothesis 2). We will investigate how differentials in poverty for migrant, mixed, and non-migrant households are associated with different work arrangements, net of observed human capital and household and job characteristics. As non-teleworkability and single-earnership have commonly been identified as poverty risks (Fasang, Struffolino, and Zagel 2023; Fasani and Mazza 2020), we expect household arrangements with single earners or workers in non-teleworkable jobs to have a higher poverty risk. Consequently, we expect the higher prevalence of migrant and mixed households in work arrangements with higher poverty risks to explain poverty differentials by household type (Hypothesis 3). Finally, we expect to find higher poverty for migrant and mixed households than for non-migrant households within the same work arrangement (Hypothesis 4). This may be explained by unobserved heterogeneity in variables not contained in our data, such as salaries, differences within teleworkable and non-teleworkable jobs, or further disadvantages faced by migrants, such as employer discrimination (see discussion above). Data and methods Data and sample We used data from the German Microcensus 2019 (RDC of the Federal Statistical Office and Statistical Offices of the Federal States of Germany 2022, DOI: 10.21242/ 12211.2019.00.00.3.1.0), an annually conducted representative household survey administered by the Federal Statistical Office. The Microcensus covers approximately 1% of the German population, and participation is mandatory for the selected households. The anonymised scientific use file for researchers comprises 70% of the data. 1 Following common German poverty measurements, households are selected at their main 1300 M. HORNUNG ET AL. residence, with communal accommodation being excluded (Boehle 2015). We restricted the sample to different-sex, two-adult couple households consisting of individuals of working age (19–65) and (if present) children below 18. Households with adult children are therefore excluded in our analyses. For the households to be further included in the sample, at least one person in the household had to be employed. Employment was defined as pursuing a professional activity for at least one hour per week. We excluded 1,747 non-migrant, 269 mixed, and 1,062 migrant households in which none of the two adult household members was employed. 2 Variables and models The dependent variable was a household’s probability of being poor. A household was considered poor if the equalised net disposable household income was below 60% of the median household income in a given context. The Microcensus reports net monthly household income in the month preceding the survey before tax and social insurance payments. This income can come from various sources, such as unemployment benefits (Arbeitslosengeld I, Arbeitslosengeld II), child and accommodation allowance, investment income, and retirement benefits, although employment, on average, contributed the most (Hochgürtel 2019). We calculated the median monthly income at the country level of the total Microcensus scientific use file and not only our analytical sample. As the Microcensus uses income classes to measure household income, we used a procedure developed by Stauder and Hüning (2004) and edited by Boehle (2015) to impute the monthly household income within each income class. This procedure assumes that income is evenly distributed within each income class and ascribes each individual in the indicated income class a different possible income value that depends on the total number of individuals in the class and the width of the income class. Household type was our core independent variable. We differentiated between three types of households: migrant households with two migrants, mixed households with one migrant, and non-migrant households with two non-migrants. A migrant was defined as a person who migrated themselves, i.e. a first-generation migrant. The second independent variable was work arrangement: it combines information on whether the households consist of one or two earner(s) and whether the earner(s)’ job was teleworkable. This differentiation yielded eight different work arrangements for different-sex couple households (Table 1), ranging from work arrangements with low expected poverty risk (two earners with a teleworkable job) to work arrangements with high expected poverty risk (one earner with a non-teleworkable job). Table 1. Typology of work arrangements in different-sex households. Highest risk ←− Lowest risk N of earners Man Woman 2 teleworkable teleworkable 2 teleworkable non-teleworkable 2 non-teleworkable teleworkable 2 non-teleworkable non-teleworkable 1 teleworkable not employed 1 not employed teleworkable 1 non-teleworkable not employed 1 not employed non-teleworkable JOURNAL OF ETHNIC AND MIGRATION STUDIES 1301 partners with migration experience (assortative mating) may increase economic inequalities between migrants and non-migrants. Our study shows that if, however, a migrant is partnered with a non-migrant, these mixed unions have a lower frequency of non-teleworkable work arrangements and somewhat lower poverty than migrant households. Therefore, it is vital to consider partnership formation behaviour and the number of migrants in a household when discussing the economic situation of households. Specifically, the role of non-migrant partners in averting poverty in mixed unions is an underexplored and important avenue for future research. Furthermore, research should explore to what extent mixed households impact processes of economic integration of migrants and polarisation between migrants concerning poverty in the long term. Observing the economic situation of non-migrant, mixed, and migrant households over time is also relevant, as children of mixed partnerships are more likely to partner with individuals with a migrant background than children with non-migrant parents (Irastorza and Elwert 2021). Our findings further emphasise strong heterogeneity within non-teleworkable jobs, with migrants in non-teleworkable jobs being especially prone to high poverty. The high heterogeneity in non-teleworkable jobs alludes to labour market segregation beyond teleworkability. In our models, we controlled for type of contract, employment type, leadership responsibility, and shift work, but other job characteristics, like the industry or the level of unionisation, may be important mechanisms in explaining differences between migrant, mixed, and non-migrant households. Throughout the pandemic, employment in essential infrastructure jobs played a protective role. Research on such occupations in Germany indicates high variation within these jobs, with migrants being overrepresented in jobs with bad working conditions (Nivorozhkin and Poeschel 2022). Given that the impact of technological change on households differs by institutional context (Minardi et al. 2023), our findings for Germany have to be considered in light of the country’s particular setting. However, we argue that Germany is an exemplary case for the study of poverty gaps between migrant, mixed, and non-migrant households depending on the access to teleworkable jobs, because the country has the largest EU economy and has a long history of immigration, enabling us to define labour market integration strategies over time. Therefore, our results may differ substantially in countries with less powerful economies, less occupational segregation, more recent immigration histories and differences in the selection of migrants. By using data from 2019, the present study sets the baseline for future analyses on how the new economic scenario during the Covid-19 pandemic might have altered the patterns we identified. Recent studies suggest an expansion in teleworkability above what was available during the pandemic (Sostero et al. 2020). Therefore, our measure of teleworkability is conservative. It may be particularly worthwhile to examine whether the expansion of teleworkable tasks increased inequalities between migrants and non-migrants, since migrants face higher risks of unemployment in economic recessions (Kogan 2004). With mobile information and communication technology (ICT) becoming more important in the working environment, teleworkability is presumably not only a safer occupational feature in a pandemic like the Covid-19 one; indeed, literacy in ICT will become more important in the future labour market. Monitoring workers’ skill sets and offering 1308 M. HORNUNG ET AL. further ICT literacy training to migrants may be important to decrease labour market inequalities and poverty. This study does not come without limitations. Our data do not enable us to disentangle the internal-household mechanisms used to decide the number of earners. Whether having more single-earner households is due to cultural norms or structural difficulties in accessing employment is an avenue for future research. Although research in Germany has refuted the theory of family investment among migrants (Basilio, Bauer, and Sinning 2009), households’ decisions to opt for a single-earner model could be linked to labour market disadvantages. The fact that single-earner work arrangements with a man in a teleworkable job are very common among third-country nationals may allude to specific migration patterns and work arrangements of migrant main earners from these regions. Further, the observation that in mixed households, the non-migrant partner is more likely to be the single earner (Braack, Milewski, and Trappe 2022) may hint at migrants’ labour market disadvantages and a decision to prevent economic risks with the non-migrant partner taking over (sole) employment. We cannot rule out selection into different household types. For example, studies show that single-earner couples are more likely to migrate than dual-earner couples (Vidal et al. 2017). Furthermore, certain migrant groups seem to be overrepresented in certain work arrangements, raising questions about economic differences between migrants. These selection effects may be visible within mixed households as well. For instance, the number of earners in mixed households may differ by region of birth and other migration-specific characteristics (Schroedter 2013). This study defines non-migrants as individuals without migration experience to Germany. The definition of a nonmigrant includes individuals who have at least one parent who migrated, that is, migrants’ descendants. Although research shows that second-generation migrants differ in their labour market outcomes (Granato and Kalter 2001) and have different mating behaviour than first-generation migrants and the majority population (Schroedter 2013), we have not distinguished between migrants and their descendants because this would have added too much complexity to the analysis. Finally, we have focused on two-adult households and households with individuals of working age (19–65). Consequently, our conclusions are confined to this specific group. Economic disadvantages that older migrants (Steinbach 2018) or single households may face due to low pension entitlements and less risk-sharing by a partner are not considered in this study. Given the ageing migrant population in Germany and the increasing prevalence of single households, studies that look at these households will provide a more comprehensive understanding of the specific economic challenges. We further excluded households with adult children. In Germany, multigenerational households are more common among migrants (Flake 2012), and, for instance, children of migrants from Turkey leave the parental home later than children with non-migrant parents (Windzio and Aybek 2015). Therefore, investigating households with adult children is an important avenue for future research, particularly because forming multigenerational households can help mitigate economic disadvantages. Despite these limitations, this study underlines the importance of taking a household perspective to gauge poverty for migrant, mixed, and non-migrant households. By focusing on two-adult households where one adult is employed, we have conceptualised and empirically substantiated non-teleworkability and single-earner work arrangements as JOURNAL OF ETHNIC AND MIGRATION STUDIES 1309 poverty risks, especially for migrant households. As the higher prevalence of single-earner households among migrant and mixed households is one driver of their poverty, incentivising dual-earnership among these households may decrease poverty in the long term. Regarding teleworkability, improving migrants’ skill levels, especially in ICT, is an important policy recommendation. If the aim is to reduce differences between migrant and nonmigrant households, merely targeting households by teleworkability/non-teleworkability might not suffice. Particularly because migrant households generally seem to benefit less from social policies (Bostic and Hyde 2023), such policies should be more attuned to migration experience as a labour-market disadvantage. The results showed that migrant households in non-teleworkable (single-earner) work arrangements displayed higher poverty in 2019. The pandemic might have exacerbated these economic vulnerabilities. By looking at households where at least one individual was employed, we have likely underreported the severity of the poverty differential between migrant and non-migrant households. Our results can therefore be considered conservative regarding the actual economic disadvantage of mixed and migrant households in Germany. Promoting employment, especially among these households, is the first important step towards improving migrants’ economic well-being. However, especially since the Covid-19 pandemic accelerated the use of ICT, it is important to monitor to what extent different skill sets and less adaptation to a (more) digitalised labour market might further fuel differences between migrants and non-migrants in poverty over time. Notes 1. The Microcensus scientific use file (SUF) includes a 70% sample of the complete Microcensus. Due to anonymisation requirements, the German Statistical Offices provide only a 70% sample for use outside their premises. The selection process of the 70% subsample follows the sampling design of the Microcensus closely. The subsample is solely altered for reasons of enlarging feature groups to maintain factual anonymisation (Statistisches Bundesamt 2022). The Federal Statistical Office further provides weights to preserve representativeness. 2. Our main focus is on differences in job characteristics between household types and not the selection into employment. However, we acknowledge that by excluding households with no earners from our analysis, we underestimate the overall poverty of migrant households in Germany. We present the substantive poverty gap of migrant households with no earners in Table S1 of the Supplementary Materials. 3. For detailed information on the distribution of the teleworkability index between household types see Figure S1 in the Supplementary Materials. 4. FLC include: Italy, Spain, Greece, Turkey, Morocco, Portugal, Montenegro, Serbia, Macedonia, Bosnia and Herzegovina and Croatia, EU-15 include: Iceland, Liechtenstein, Norway, Switzerland, Denmark, Finland, Sweden, France, Switzerland, Ireland, United Kingdom, Belgium, Luxemburg, Netherlands, Austria. EU-enlargement include: Bulgaria, Romania, Lithuania, Latvia, Czech Republic, Hungary, Poland. FSU include: Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan, Moldova (San Marino, Andorra, Vatican), Russian Federation, Kyrgyzstan, Tajikistan, Turkmenistan, Uzbekistan, Ukraine. Some allocations are due to the categorisation of countries in the SUF of the Microcensus 2019. For more information on the region of birth and the year of migration, see Figure S2 of the Supplementary Material. 5. Following deHaan et al. (2023), we estimate a model where we interacted all controls (except region of birth) and household types. The results presented in Figure 3 are robust to this specification. This finding also holds if we consider interacted controls (see Model 6, Table A4 in the Appendix). Results are available upon request. 1310 M. HORNUNG ET AL. Acknowledgements This research was funded by the Support Network for Interdisciplinary Social Policy Research (FIS) by the German Federal Ministry of Social and Labour Affairs (BMAS) as part of the research project “Household structures and economic risks during the COVID-19 pandemic in East and West Germany: Compensation or accumulation? (KOMPAKK)”, PIs: Anette Fasang, Emanuela Struffolino, and Hannah Zagel. We thank Martin Gädecke and Jonas Braun for excellent research assistance, and the editor as well as two anonymous reviewers for their helpful comments. Disclosure statement No potential conflict of interest was reported by the author(s). Data availability statement The replication code for all analysis is available here: https://osf.io/n47fq/. The data supporting this study’s findings derive from the Scientific Use Files (SUF) of the German Microcensus 2019 (DOI: 10.21242/12211.2019.00.00.3.1.0), available from the Research Data Centre of the Federal Statistical Office. More information on access and restrictions to the availability of these data, see: https:// www.forschungsdatenzentrum.de/en/access. ORCID Maria Hornung http://orcid.org/0000-0002-2905-2707 Emanuela Stuffolino http://orcid.org/0000-0002-6635-8748 Hannah Zagel http://orcid.org/0000-0002-5307-3380 References Acemoglu, D., and D. Autor. 2011. “Chapter 12 – Skills, Tasks and Technologies: Implications for Employment and Earnings.” In Bd. 4 of Handbook of Labor Economics, 1043–1171, edited by D. Card, and O. Ashenfelter. Amsterdam: Elsevier. Adams-Prassl, A., T. Boneva, M. Golin, and C. Rauh. 2020. “Inequality in the Impact of the Coronavirus Shock: Evidence from Real Time Surveys.” Journal of Public Economics 189:104245. https://doi.org/10.1016/j.jpubeco.2020.104245. Alipour, J., O. Falck, S. Schüller, H. Fadinger, J. Schymik, O. Falck, A. Peichl, and S. Sauer. 2021. “My Home is My Castle – The Benefits of Working from Home During a Pandemic Crisis.” Journal of Public Economics 196 (13152): 104373. https://doi.org/10.1016/j.jpubeco.2021. 104373. Arntz, M., S. Ben Yahmed, and F. Berlingieri. 2020. “Working from Home and COVID-19: The Chances and Risks for Gender Gaps.” Intereconomics 55 (6): 381–386. https://doi.org/10. 1007/s10272-020-0938-5. Barnes, M., C. Heady, S. Middleton, J. Millar, F. Papadopoulos, G. Room, and P. Tsakloglou. 2002. Poverty and Social Exclusion in Europe. Cheltenham, UK/Northampton, MA: Edward Elgar Publishing. https://www.elgaronline.com/display/1840643757.xml. Basilio, L., T. K. Bauer, and M. Sinning. 2009. “Analyzing the Labor Market Activity of Immigrant Families in Germany.” Labour Economics 16 (5): 510–520. https://doi.org/10.1016/j.labeco. 2009.03.002. Biegert, T., and B. Ebbinghaus. 2022. “Accumulation or Absorption? Changing Disparities of Household non-Employment in Europe During the Great Recession.” Socio-Economic Review 20 (1): 141–168. https://doi.org/10.1093/ser/mwaa003. JOURNAL OF ETHNIC AND MIGRATION STUDIES 1311 Bihagen, E., A. Dicks, M. Ehlert, J. Härkönen, T. Korpi, A. Künn-Nelen, L. Menze, R. Montizaan, and H. Pöyilö. 2021. “Skills, Automation, and Earnings: Employment on Technology Drivenlabor Markets (Version 3.0).” Maastricht, Maastricht University, School of Business and Economics, ROA. Boehle, M. 2015. Armutsmessung mit dem Mikrozensus: Methodische Aspekte und Umsetzung für Querschnittsund Trendanalysen. GESIS Papers, 16. Bogoeski, V. 2022. “Continuities of Exploitation: Seasonal Migrant Workers in German Agriculture During the COVID-19 Pandemic.” Journal of Law and Society 49 (4): 681–702. https://doi.org/10.1111/jols.12389. Bostic, A., and A. Hyde. 2023. “Social Spending, Poverty, and Immigration: A Systematic Analysis of Welfare State Effectiveness and Nativity in 24 Upperand Middle-Income Democracies.” Social Currents 10, https://doi.org/10.1177/23294965231169253. Braack, M. K., N. Milewski, and H. Trappe. 2022. “Crossing Social Boundaries in an Immigration Context: Exogamy and Gendered Employment Patterns in Unions in Germany.” Gender Issues 39 (2): 142–176. https://doi.org/10.1007/s12147-021-09281-8. Brady, D. J., R. Finnigan, and S. Hübgen. 2017. “Rethinking the Risks of Poverty: A Framework for Analyzing Prevalences and Penalties.” American Journal of Sociology 123 (3): 740–786. https:// doi.org/10.1086/693678. Brussevich, M., E. Dabla-Norris, and S. Khalid. 2020. “Who Will Bear the Brunt of Lockdown Policies? Evidence from Tele-Workability Measures Across Countries.” IMF Working Papers 20 (88), https://doi.org/10.5089/9781513546285.001. Constant, A., and D. S. Massey. 2005. “Labor Market Segmentation and the Earnings of German Guestworkers.” Population Research and Policy Review 24 (5): 489–512. https://doi.org/10.1007/ s11113-005-4675-z. Crettaz, E. 2018. “In-work Poverty among Migrants.” In Handbook on In-Work Poverty, edited by H. Lohmann, and I. Marx, 89–106. Cheltenham, UK/Northampton, MA: Edward Elgar Publishing. deHaan, E., J. Moon, J. E. Shipman, Q. T. Swanquist, and R. L. Whited. 2023. Control Variables in Interactive Models (SSRN Scholarly Paper No. 3959861). Destatis. 2023. 17.3% of Germany’s Population has Immigrated Since 1950. Federal Statistical Office. https://www.destatis.de/EN/Press/2023/03/PE23_080_12.html. Di Stasio, V., B. Lancee, S. Veit, and R. Yemane. 2021. “Muslim by Default or Religious Discrimination? Results from a Cross-National Field Experiment on Hiring Discrimination.” Journal of Ethnic and Migration Studies 47 (6): 1305–1326. https://doi.org/10.1080/1369183X. 2019.1622826. Drever, A. I., and O. Hoffmeister. 2008. “Immigrants and Social Networks in a Job-Scarce Environment: The Case of Germany.” International Migration Review 42 (2): 425–448. https://doi.org/10.1111/j.1747-7379.2008.00130.x. Dribe, M., and C. Lundh. 2008. “Intermarriage and Immigrant Integration in Sweden: An Exploratory Analysis.” Acta Sociologica 51 (4): 329–354. https://doi.org/10.1177/ 0001699308097377. Elwert, A., and A. Tegunimataka. 2016. “Cohabitation Premiums in Denmark: Income Effects in Immigrant–Native Partnerships.” European Sociological Review 32 (3): 383–402. https://doi.org/ 10.1093/esr/jcw018. Eurostat. 2024. Gross Domestic Product at Market Prices [Dataset]. Eurostat. Fasang, A. E., E. Struffolino, and H. Zagel. 2023. “Household-level Prevalence and Poverty Penalties of Working in Non-Teleworkable and Non-Essential Occupations: Evidence from East and West Germany in 2019.” Zeitschrift Für Sozialreform 69 (2): 85–117. https://doi.org/ 10.1515/zsr-2022-0107. Fasani, F., and J. Mazza. 2020. “Being on the Frontline? Immigrant Workers in Europe and the COVID-19 Pandemic.” SSRN Electronic Journal 13963, https://doi.org/10.2139/ssrn.3755380. Filandri, M., and E. Struffolino. 2019. “Individual and Household in-Work Poverty in Europe: Understanding the Role of Labor Market Characteristics.” European Societies 21 (1): 130– 157. https://doi.org/10.1080/14616696.2018.1536800. 1312 M. HORNUNG ET AL. Flake, R. 2012. “Multigenerational Living Arrangements Among Migrants.” SSRN Electronic Journal, https://doi.org/10.2139/ssrn.2159706. Fleischmann, F., and J. Höhne. 2013. “Gender and Migration on the Labour Market: Additive or Interacting Disadvantages in Germany?” Social Science Research 42 (5): 1325–1345. https://doi. org/10.1016/j.ssresearch.2013.05.006. Gädecke, M., E. Struffolino, H. Zagel, and A. Fasang. 2021. KOMPAKK Index of Occupations’ Teleworkability in Germany. SowiDataNet|datorium -GESIS Data Archive, Cologne. Version 1.0.0. Dataset. https://doi.org/10.7802/2263. Giesecke, J., M. Kroh, I. Tucci, A.-L. Baumann, and N. El-Kayed. 2017. “Armutsgefährdung bei Personen mit Migrationshintergrund – Vertiefende Analysen auf Basis von SOEP und Mikrozensus.” SOEPpapers 907:67. González-Ferrer, A., O. Obućina, C. Cortina, and T. Castro-Martín. 2018. “Mixed Marriages Between Immigrants and Natives in Spain: The Gendered Effect of Marriage Market Constraints.” Demographic Research 39:1–32. https://doi.org/10.4054/DemRes.2018.39.1. Granato, N., and F. Kalter. 2001. “Die Persistenz ethnischer Ungleichheit auf dem deutschen Arbeitsmarkt.” KZfSS Kölner Zeitschrift für Soziologie und Sozialpsychologie 53 (3): 497–520. https://doi.org/10.1007/s11577-001-0076-4. Gregg, P., and J. Wadsworth. 2008. “Two Sides to Every Story: Measuring Polarization and Inequality in the Distribution of Work.” Journal of the Royal Statistical Society Series A: Statistics in Society 171 (4): 857–875. https://doi.org/10.1111/j.1467-985X.2008.00542.x. Haug, S. 2010. Interethnische Kontakte, Freundschaften, Partnerschaften und Ehen von Migranten in Deutschland. Nürnberg: Bundesamt für Migration und Flüchtlinge (BAMF) Forschungszentrum Migration, Integration und Asyl (FZ). Hochgürtel, T. 2019. Einkommensanalysen mit dem Mikrozensus. Wiesbaden: Statistisches Bundesamt. Hötte, K., M. Somers, and A. Theodorakopoulos. 2023. “Technology and Jobs: A Systematic Literature Review.” Technological Forecasting and Social Change 194:122750. https://doi.org/ 10.1016/j.techfore.2023.122750. Irastorza, N., and A. Elwert. 2021. “Like Parents, Like Children? The Impact of Parental Endogamy and Exogamy on Their Children’s Partner Choices in Sweden.” Journal of Ethnic and Migration Studies 47 (4): 895–915. https://doi.org/10.1080/1369183X.2019.1654160. Kantarevic, J. 2004. “Interethnic Marriages and Economic Assimilation of Immigrants.” IZA Discussion Paper, 1142. Kesler, C. 2015. “Welfare States and Immigrant Poverty: Germany, Sweden, and the United Kingdom in Comparative Perspective.” Acta Sociologica 58 (1): 39–61. https://doi.org/10. 1177/0001699314560238. Kogan, I. 2004. “Last Hired, First Fired? The Unemployment Dynamics of Male Immigrants in Germany.” European Sociological Review 20 (5): 445–461. https://doi.org/10.1093/esr/jch037. Konietzka, D., and M. Kreyenfeld. 2002. “The Performance of Migrants in Occupational Labour Markets.” European Societies 4 (1): 53–78. https://doi.org/10.1080/14616690220130991 Lohmann, H. 2009. “Welfare States, Labour Market Institutions and the Working Poor: A Comparative Analysis of 20 European Countries.” European Sociological Review 25 (4): 489–504. https://doi.org/10.1093/esr/jcn064. Meng, X., and R. G. Gregory. 2005. “Intermarriage and the Economic Assimilation of Immigrants.” Journal of Labor Economics 23 (1): 135–174. https://doi.org/10.1086/425436. Meng, X., and D. Meurs. 2009. “Intermarriage, Language, and Economic Assimilation Process: A Case Study of France.” International Journal of Manpower 30 (1/2): 127–144. https://doi.org/10. 1108/01437720910948447. Messenger, J., O. Llave Vargas, L. Gschwind, S. Böhmer, G. Vermeylen, M. Wilkens, and (with Internationales Arbeitsamt, & European Foundation for the Improvement of Living and Working Conditions). 2017. Working Anytime, Anywhere: The Effects on the World of Work. Luxembourg: Publications Office of the European Union. Minardi, S., C. Hornberg, P. Barbieri, and H. Solga. 2023. “The Link Between Computer use and job Satisfaction: The Mediating Role of job Tasks and Task Discretion.” British Journal of Industrial Relations 61 (4): 796–831. https://doi.org/10.1111/bjir.12738. JOURNAL OF ETHNIC AND MIGRATION STUDIES 1313 Nieuwenhuis, R., W. Van Lancker, D. Collado, and B. Cantillon. 2020. “Trends in Women’s Employment and Poverty Rates in OECD Countries: A Kitagawa–Blinder–Oaxaca Decomposition.” Italian Economic Journal 6 (1): 37–61. https://doi.org/10.1007/s40797-01900115-x. Nivorozhkin, A., and F. Poeschel. 2022. “Working Conditions in Essential Occupations and the Role of Migrants.” Economic Analysis and Policy 74:250–261. https://doi.org/10.1016/j.eap. 2022.02.002. Nottmeyer, O. 2011. “Couple’s Relative Labor Supply in Intermarriage.” SSRN Electronic Journal, https://doi.org/10.2139/ssrn.1790678. RDC of the Federal Statistical Office and Statistical Offices of the Federal States of Germany. 2022. Mikrozensus 2019, SUF, Version 0 (Version 0) [Dataset]. RDC of the Federal Statistical Office and the statistical offices of the Länder (RDC). Rohrbach-Schmidt, D., and A. Hall. 2020. BIBB/BAuA-Erwerbstätigenbefragung 2018. BIBB-FDZ Datenund Methodenbericht 1/2020. urn:nbn:de:0035-0818-6. Rubin, J., M. S. Rendall, L. Rabinovich, F. Tsang, C. van Oranje-Nassau, and B. Janta. 2008. Migrant Women in the European Labour Force: Current Situation and Future Prospects. Santa Monica, CA: RAND Corporation. https://www.rand.org/pubs/technical_reports/TR591.html. Salikutluk, Z., J. Giesecke, and M. Kroh. 2020. The Situation of Female Immigrants on the German Labour Market: A Multi-Perspective Approach. SOEPpapers on Multidisciplinary Panel Data Research, 1072. Schroedter, J. H. 2013. Ehemuster von Migranten in Westdeutschland. Wiesbaden: Springer Fachmedien Wiesbaden. Sommer, I. 2021. “Recognition of Foreign Qualifications in Germany: Selectivity and Power in reMaking Professionals.” International Migration 59 (4): 26–41. https://doi.org/10.1111/imig. 12831. Sostero, M., S. Milasi, J. Hurley, E. Fernandez-Macías, and Martina Bisello. 2020. Teleworkability and the COVID-19 Crisis: A New Digital Divide?. Seville, Spain: EU Science Hub. Statistisches Bundesamt. 2022. Datenhandbuch zum Mikrozensus Scientific Use File 2019. https:// www.gesis.org/missy/files/documents/MZ/DHB_2020.pdf. Stauder, J., and W. Hüning. 2004. “Die Messung von Äquivalenzeinkommen und Armutsquoten auf der Basis des Mikrozensus.” Statistische Analysen und Studien {NRW} 13:9–31. Steinbach, A. 2018. “Older Migrants in Germany.” Journal of Population Ageing 11 (3): 285–306. https://doi.org/10.1007/s12062-017-9183-5. Tamayo, H. X. J., and D. Popova. 2020. “Second Earners and In-Work Poverty in Europe.” Journal of Social Policy 50 (3): 470–492. https://doi.org/10.1017/S0047279420000227. Thijssen, L., B. Lancee, S. Veit, and R. Yemane. 2021. “Discrimination Against Turkish Minorities in Germany and the Netherlands: Field Experimental Evidence on the Effect of Diagnostic Information on Labour Market Outcomes.” Journal of Ethnic and Migration Studies 47 (6): 1222–1239. https://doi.org/10.1080/1369183X.2019.1622793. Veit, S., and L. Thijsen. 2021. “Almost Identical but Still Treated Differently: Hiring Discrimination Against Foreign-Born and Domestic-Born Minorities.” Journal of Ethnic and Migration Studies 47 (6): 1285–1304. https://doi.org/10.1080/1369183X.2019.1622825. Vidal, S., F. Perales, P. M. Lersch, and M. Brandén. 2017. “Family Migration in a Cross-National Perspective: The Importance of Institutional and Cultural Context.” Demographic Research 36:307–338. https://doi.org/10.4054/DemRes.2017.36.10. Weichselbaumer, D. 2016. Discrimination Against Female Migrants Wearing Headscarves (SSRN Scholarly Paper No. 2842960). Wiedner, J., and J. Giesecke. 2022. “Immigrant Men’s Economic Adaptation in Changing Labor Markets: Why Gaps Between Turkish and German Men Expanded, 1976–2015.” International Migration Review 56 (1): 176–205. https://doi.org/10.1177/01979183211029903. Windzio, M., and C. M. Aybek. 2015. “Marriage, Norm Orientation and Leaving the Parental Home: Turkish Immigrant and Native Families in Germany.” Comparative Population Studies 40 (2), https://doi.org/10.12765/CPoS-2015-07. 1314 M. HORNUNG ET AL. Appendix Table A1. Descriptive statistics (total n = 65,247). Non-migrant Mixed Migrant (N = 49,507) (N = 6,818) (N = 8,922) Household Composition Dual Earner 89 83 73 Single Earner 11 17 27 Total 100 100 100 Work Arrangement Dual earner both teleworkable 22 20 7 man teleworkable 12 12 5 woman teleworkable 22 20 11 both non-teleworkable 32 31 49 Single earner man teleworkable 3 5 4 woman teleworkable 1 2 1 man non-teleworkable 5 8 19 woman non-teleworkable 2 3 4 Source: German Microcensus Scientific Use File 2019, weighted column percentages, total numbers are not weighted. If the partner is not mentioned, he/she is either in a non-teleworkable job (dual-earner households) or not employed (single-earner households). Due to rounding, the percentages presented in the tables and figures may not add up to exactly 100%. JOURNAL OF ETHNIC AND MIGRATION STUDIES 1315 Table A2. Characteristics of the main earner by work arrangements: dual-earner households. Both teleworkable Man teleworkable Woman teleworkable Both non-teleworkable Non-migrant Mixed Migrants Non-migrant Mixed Migrants Non-migrant Mixed Migrants Non-migrant Mixed Migrants Woman 19% 22% 21% 14% 16% 14% 26% 31% 24% 17% 19% 14% Average Age 44,56 41,92 40,19 44,67 41,75 42,45 44,70 42,02 41,16 44,47 42,23 43,96 Education Low 1 2 4 2 5 11 3 7 10 6 15 28 Medium 35 26 24 47 43 35 56 50 53 71 63 56 High 64 72 72 51 52 54 41 43 37 24 23 16 Type of contract Not applicable 14 19 19 12 12 17 10 13 14 7 9 7 Fixed-term 3 5 8 4 5 7 3 5 4 4 8 9 Permanent 82 76 73 85 82 76 87 82 82 88 84 84 Employment type Self-employment 14 19 18 12 12 17 10 13 14 7 9 7 PT-Atypical employment 1 2 2 2 2 4 2 3 1 3 3 4 FT-Atypical employment 3 4 8 4 5 7 3 6 6 5 9 12 PT-Employment 5 4 4 5 5 2 5 5 3 4 5 3 Full-time employment 76 71 68 78 75 70 79 74 75 80 75 74 Leadership responsibility No 50 53 62 54 59 62 60 62 72 67 70 86 Supervisor 20 20 16 21 20 21 21 19 14 21 19 9 Manager 30 27 22 24 21 17 19 19 13 12 11 5 Shift work Every day 1 1 1 4 5 4 10 9 15 19 21 22 >= half of the days 0 0 0 1 1 1 2 2 4 4 4 5 < half of the days 0 1 0 1 1 2 1 2 1 2 2 2 No 98 98 99 95 93 92 87 87 79 76 73 70 Regions of countries of birth Germany 100 64 0 100 62 0 100 55 0 100 57 0 FLC 0 7 16 0 7 19 0 14 16 0 15 22 EU-15 0 9 11 0 8 5 0 6 4 0 5 1 EU-enlargement 0 9 23 0 9 29 0 11 29 0 10 34 Former Soviet Union Countries 0 4 25 0 5 27 0 7 39 0 6 29 Other 0 7 25 0 8 19 0 8 13 0 7 14 Row percentages 25 24 9 14 15 8 24 22 14 36 39 68 N10,454 1,199 518 5,687 756 431 9,930 1,115 790 14,976 1,976 3,761 Source: German Microcensus Scientific Use File 2019, own calculations, percentages weighted, total numbers and row percentages are not weighted. FLC: former labour recruitment countries. If the partner is not mentioned, he/she is in a non-teleworkable job. 1316 M. HORNUNG ET AL. Table A3. Characteristics of the main earner by work arrangements: single-earner households. Man teleworkable Woman teleworkable Man non-teleworkable Woman non-teleworkable Non-migrant Mixed Migrants Non-migrant Mixed Migrants Non-migrant Mixed Migrants Non-migrant Mixed Migrants Woman 3% 1% 1% 100% 100% 100% 4% 5% 2% 100% 100% 100% Average Age 46,15 42,61 39,25 48,75 41,62 40,28 45,81 41,76 40,54 48,01 42,59 45,60 Education Low 2 3 9 3 4 7 9 22 37 11 23 29 Medium 39 34 19 56 42 32 68 54 45 70 51 50 High 59 64 71 41 54 61 23 24 18 19 26 21 Type of contract Not applicable 11 14 15 4 7 11 5 7 6 5 3 2 Fixed-term 4 5 11 6 11 16 6 8 14 8 9 11 Permanent 84 81 75 90 81 74 89 85 81 87 88 86 Employment type Self-employment 11 14 14 4 7 10 5 7 6 5 3 2 PT-Atypical employment 2 1 3 6 8 8 3 4 9 14 16 21 FT-Atypical employment 4 5 10 4 9 11 7 11 15 5 5 8 PT-Employment 2 4 2 18 12 10 3 3 3 23 23 14 Full-time employment 81 76 71 68 64 61 82 76 67 53 53 55 Leadership responsibility No 51 55 65 74 71 76 70 73 88 79 86 89 Supervisor 21 21 18 17 21 16 19 17 8 14 8 10 Manager 28 23 17 9 8 7 11 10 4 7 7 1 Shift work Every day 2 2 2 1 3 3 18 21 20 20 18 21 >= half of the days 0 0 1 0 1 0 4 6 5 5 4 8 < half of the days 0 0 0 0 1 1 1 2 2 3 0 1 No 97 97 97 98 96 96 77 70 74 72 78 71 Regions of countries of birth Germany 100 72 0 100 71 0 100 52 0 100 61 0 FLC 0 5 19 0 5 14 0 25 27 0 11 21 EU-15 0 6 7 0 3 6 0 4 1 0 4 2 EU-enlargement 0 6 14 0 12 27 0 4 23 0 8 25 Former Soviet Union Countries 0 3 18 0 6 13 0 5 15 0 8 34 Other 0 7 43 0 3 39 0 10 34 0 8 19 Row percentages 28 31 16 10 8 2 51 54 74 11 7 8 N 2,388 546 535 839 141 76 4,318 960 2,549 915 125 262 Source: German Microcensus Scientific Use File 2019, own calculations, percentages weighted, total numbers and row percentages are not weighted. FLC: former labour recruitment countries. If the partner is not mentioned, he/she is not employed. JOURNAL OF ETHNIC AND MIGRATION STUDIES 1317