Responsibility for opinions expressed in signed articles rests solely with their authors, and publication does not constitute an endorsement by the ILO.
This article is also available in French, in Revue internationale du Travail 165 (1), and Spanish, in Revista Internacional del Trabajo 145 (1).
1. Introduction
The workforce in the platform economy, where economic activities and transactions are mediated through digital labour platforms, is still relatively small (Lenaerts et al. 2023; Piasna, Zwysen and Drahokoupil 2022; Urzì Brancati, Pesole and Fernández-Macías 2020). However, the generally poor working conditions associated with these platforms and their broader implications for the world of work have garnered significant attention from researchers and policymakers in recent years. A key question concerns the motivations that drive individuals to participate in this type of work. Platforms promise a degree of flexibility in terms of working time and location that is rarely matched by traditional, dependent employment. However, numerous studies have shown that this flexibility is often illusory and is coupled with tight surveillance and high-pressure algorithmic management practices (Pulignano et al. 2021). This has prompted a search for more structural explanations. For example, some studies find that platform work is more prevalent in local labour markets with limited access to better alternatives (Zwysen and Piasna 2024). It has also been convincingly argued that the absence of formal recruitment processes – common to many platforms – and lax legal requirements offer viable opportunities for individuals otherwise excluded from formal labour markets or at greater risk of unfair treatment and discrimination in the traditional economy (van Doorn and Vijay 2024).
Migrants constitute a notable group of such vulnerable workers, with weak bargaining power in the labour market. There is broad consensus that individuals of foreign origin are over-represented in the platform economy (Kowalik, Lewandowski and Kaczmarczyk 2022; Piasna, Zwysen and Drahokoupil 2022; van Doorn and Vijay 2024). Some types of platform work – particularly food delivery – are even equated with migrant work (see, e.g., van Doorn and Vijay 2024).
However, understanding of migrants’ involvement in the platform economy remains limited in several significant respects. First, there is a gap in knowledge regarding whether this involvement spans the entire platform economy, which includes not only relatively well-studied location-based services – such as ride-hailing, delivery and care work – but also remote work performed online. This remote work may involve simple, standardized tasks (“clickwork”) or more complex freelance work requiring professional skills. Second, both platform workers and migrants tend to differ from the general population in terms of socio-demographic characteristics – notably age, education, labour market status and family situation (Piasna, Zwysen and Drahokoupil 2022; Van Mol and de Valk 2016; Zaiceva and Zimmermann 2014). It remains unknown, however, to what extent migrants’ over-representation in platform work is attributable to these compositional factors. It is possible that migrants and platform workers share certain characteristics that help explain their significant overlap. Lastly, open questions remain as to whether migrants’ involvement in platform work is related to structural factors identified in the literature – in particular, limited access to decent and quality employment in the traditional economy – and, relatedly, whether differences exist among migrants with diverse socio-economic characteristics in their involvement in platform work.
This article aims to address these questions by bridging three strands of literature considering: (i) the heterogeneity of the platform workforce; (ii) vulnerability and dependency in the platform economy; and (iii) the embeddedness of platform work in the traditional economy. It provides a comprehensive picture of platform work, encompassing both location-based and remote services, involving simple as well as complex tasks. The strength of the article’s empirical contribution lies in the use of large scale and cross-national survey data, which enable comparisons between migrant and non-migrant populations, as well as between platform work and employment in the traditional economy. In particular, our analysis draws on the European Trade Union Institute (ETUI) 2021 Internet and Platform Work Survey (IPWS), which is representative of the resident populations (aged 15 to 65) in 14 major European Union (EU) Member States, covering 84 per cent of the EU working-age population (Piasna, Zwysen and Drahokoupil 2022). In the last part of the analysis, the IPWS is complemented by official labour market statistics from the EU Labour Force Survey (EU-LFS) for the same set of countries.
The extent to which digital labour platforms are indeed a labour market entry point for migrants is an important question, with both theoretical and practical policy implications. These platforms may offer migrants work opportunities shortly after arrival in a destination country, owing to generally lower entry barriers. However, this may come at the cost of longer-term negative effects, hindering labour market integration and progression (Kowalik, Lewandowski and Kaczmarczyk 2022; Lam and Triandafyllidou 2024). In particular, platform work can adversely affect long-term employment and career prospects by providing few opportunities for upward mobility, skill acquisition and the expansion of social networks. It may also expose workers to discrimination. Van Doorn and Vijay (2024) refer to this as the “Janus face” of platforms – welcoming newcomers but also rejecting, deceiving and disappointing them. This dynamic is linked to the concept of predatory inclusion within racial capitalism: although the system appears inclusive and open, it operates on extractive terms and results in racialized forms of oppression targeting the most vulnerable (McMillan Cottom 2020). Moreover, while embarking on work through platforms may be relatively easy, this work can create a vicious circle by hindering the accumulation of skills and social capital typically acquired through traditional (non-platform) work in the country of residence. This, in turn, further limits career progression over time (Damas de Matos 2017; Zwysen 2019). Such dynamics may lead to an accumulation of disadvantages, rendering the migrants engaged in platform work particularly vulnerable to exploitation (Rodríguez-Modroño, Agenjo-Calderón and López-Iqual 2024). With fewer alternatives available, these workers may become increasingly dependent on platform work, and the persistence of poor working conditions may thus become more pronounced than among non-migrant – and therefore, on average, less vulnerable – groups of workers. This signals serious challenges for social inclusion and integration. At the same time, the availability of vulnerable workers may perpetuate the growth of digital labour platforms, potentially exerting downward pressure on working and employment conditions in the sectors in which they operate (Graham, Hjorth and Lehdonvirta 2017).
A better understanding of migrants’ participation in the platform economy and how it is affected by the opportunities available to them in the traditional economy is crucial for developing effective policies and regulations that ensure decent work for all, including migrant populations and platform economy workers more broadly. However, the lack of representative data has made it difficult to quantify the extent of migrants’ over-representation in the platform economy, and the existence of similar patterns across different types of platform work and socio-economic groups. This study attempts to fill that gap.
The remainder of the article is organized as follows. Section 2 sets out our theoretical framework for understanding the representation of migrants in the platform economy and the factors driving their participation in this type of work organization. Section 3 presents our data, measures and methods. Section 4 examines our results: first, describing differences in platform work engagement between migrant and non-migrant groups; second, examining the influence of compositional factors in the likelihood of migrant engagement; and third, exploring the mechanisms that may contribute to migrant over-representation in the platform economy. These results are discussed in section 5, while section 6 draws some conclusions and identifies avenues for future research.
2. Theoretical framework
Platform work is typically characterized by clearly defined tasks performed by a worker at the request of a client, with the matching handled by a digital platform through a series of algorithms. Despite the considerable heterogeneity in the nature of platform work (e.g. location-based versus remote services, or high-skilled versus low-skilled tasks), the literature consistently highlights the prevalence of poor working conditions. These include unpaid work, low pay, job insecurity, unpredictability and challenges associated with algorithmic management, such as surveillance, control and information asymmetry (Berg 2016; Piasna, Zwysen and Drahokoupil 2022; Pulignano et al. 2021; Wood et al. 2019). The business model of platforms is based on the increasing informalization and hyper-externalization of labour relations, which have weakened workers’ bargaining power (Pirina, Della Puppa and Perocco 2024). The key challenge for platforms is therefore to secure a pool of workers willing to accept such low-quality jobs. This section examines the evidence on whether migrants are more inclined to engage in platform work, explores the underlying mechanisms driving this participation and considers whether certain groups are more drawn to platform work than others.
2.1. Representation of migrant workers in the platform economy
In-depth qualitative studies have repeatedly pointed to the prevalence of migrant workers in the platform economy – particularly from more vulnerable groups, such as recent arrivals and individuals without the legal right to work (Berger et al. 2019; Lam and Triandafyllidou 2024; van Doorn and Vijay 2024). This finding is partly confirmed by quantitative studies, which do indeed show an over-representation of migrants in platform work overall (e.g. Jeon, Liu and Ostrovsky 2021; Piasna, Zwysen and Drahokoupil 2022; Urzì Brancati, Pesole and Fernández-Macías 2020).
Questions remain, however, as to the nature of this over-representation and whether similar patterns are observed across different socio-economic groups and types of platform work. Surveys among the platform workforce suggest that it differs in notable ways from the general working-age population. For instance, platform workers tend to be younger, better educated and live in larger cities (e.g. Piasna, Zwysen and Drahokoupil 2022; Urzì Brancati, Pesole and Fernández-Macías 2020). Similarly, immigrants are found to differ from the native-born populations in terms of their socio-demographic characteristics, although these differences vary considerably across migration waves and regions (Van Mol and de Valk 2016; Zaiceva and Zimmermann 2014). It is still unclear to what extent these characteristics make migrants more similar to an average platform worker, thus explaining their higher prevalence in this type of work.
Existing literature on migrants’ involvement in platform work relies largely on qualitative methods based on small groups and tends to focus on a limited selection of location-based platforms – particularly in ride-hailing and delivery, but also in cleaning and personal services (Altenried 2024; Holtum et al. 2022; Pais and Marcolin 2024; van Doorn 2017; van Doorn, Ferrari and Graham 2023). This fails to provide a comprehensive picture of the platform economy, which – even within the same type of services – is highly fragmented, with a multitude of platform companies following divergent business and employment models (Pirina, Della Puppa and Perocco 2024). Moreover, there is little research on migration and remote platform work (Zwysen and Piasna 2024), making it difficult to generalize earlier findings about the over-representation of migrants to the platform economy as a whole.
The emphasis on location-based platform workers in previous studies can be explained by the relative ease of recruiting these workers as research participants. They are more visible and accessible, and are also more likely to form communities or engage in collective bargaining, making it easier to establish contacts through personal networks (Vandaele 2021). However, location-based platform work may be particularly attractive to migrants owing to its generally lower skill and language requirements. In contrast, freelance work performed through remote labour platforms offers work at a range of skill levels, many tasks requiring specific qualifications. Nevertheless, remote platform work may also be generally appealing to migrants, as it offers access to a much larger cross-border labour market. This suggests that language and cultural barriers in the host country may be less significant than in comparable jobs in the traditional economy (Kässi and Lehdonvirta 2018; Munoz, Sawyer and Dunn 2022).
These considerations point to several areas where more research and new data are needed to complement existing literature on the migrant platform workforce. In particular, it remains unclear to what extent the over-representation of migrants in platform work is related to compositional factors and thus can be explained by their particular socio-demographic profile. Moreover, we expand on prior research by considering different types of platform work, both location-based and remote. These issues are addressed in the first part of the analysis.
2.2. Mechanisms guiding migrants to platform work
The second set of questions examined in this article concerns the reasons why individuals engage in platform work and whether any such dynamics are specific to, or more prevalent among, migrant populations. In particular, the literature points to the lack of better alternatives in the traditional economy – often due to structural barriers or discrimination – as a key factor in the greater uptake of platform work. We explore this mechanism in detail, considering the role of migrants’ individual and group characteristics, as well as contextual factors.
Several characteristics of platform work have been linked to its particular appeal to groups facing labour market disadvantage or vulnerability. For instance, it is relatively easy to start working on platforms owing to a straightforward registration process that typically replaces more rigorous recruitment, places less emphasis on language skills or prior experience, and involves a lighter administrative burden (Holtum et al. 2022; van Doorn and Vijay 2024). Although platforms often check work permits, it is not uncommon for workers to share or lend accounts, enabling those without permits to work through the app – sometimes in exchange for a fee paid to the account holder (Altenried 2024; van Doorn, Ferrari and Graham 2023). Lax oversight by platforms may therefore make it easier to circumvent policies and regulations.
Migrant labour is generally characterized by a high degree of job insecurity, frequent horizontal mobility with limited upward mobility (Fellini and Guetto 2019), and low recognition of educational qualifications and prior work experience acquired abroad (Zwysen and Demireva 2025). Migrants also tend to earn lower average wages than native-born workers in comparable jobs (Pirina, Della Puppa and Perocco 2024). Therefore, for migrants who have fewer opportunities and encounter distinct – and usually much more pronounced – barriers in the traditional labour market, platform work may offer a viable alternative owing to its typically lower entry requirements (van Doorn, Ferrari and Graham 2023). This ease of entry may also stem from the transnational nature of many platforms and the similarities in their user interfaces and operations. As a result, migrants may already have experience of working on platforms in their home countries and feel comfortable using them in the host country (van Doorn, Ferrari and Graham 2023; van Doorn and Vijay 2024).
Platforms try to attract potential workers with promises of flexible working hours and locations, which may be particularly important for individuals who need to fit work around study or other commitments that limit their ability to work regular full-time hours (Adams and Berg 2021; Lehdonvirta 2018; Peticca-Harris, deGama and Ravishankar 2020). Migrant workers are found to be more exposed to such constraints, increasing their preference for flexible employment (Lam and Triandafyllidou 2024). However, this argument should be treated with caution in the light of research that reveals the often misleading nature of flexibility in platform work, especially in situations of higher economic dependence, heightening individuals’ vulnerability to platforms’ disciplining and exploitative practices (Piasna and Drahokoupil 2021). A study of Uber drivers in Australia by Holtum et al. (2022) finds that migrants experience flexibility very differently from native-born workers. Flexibility is necessary for migrants to fit work in with other commitments, such as studying or job-searching, but they are also more economically dependent on platform work. This higher dependence on platforms is found to compel workers to commit to longer hours and effectively limits their ability to exercise the flexibility that platforms claim to offer (Schor et al. 2020; Wood et al. 2019).
These features make platform work more or less attractive depending on the socio-demographic characteristics of the workers and their embeddedness in the traditional labour market (Piasna and Drahokoupil 2021). While there are several relevant aspects to heterogeneity among migrants, we focus on two factors, linked to differences in opportunities in the traditional labour market.
The first factor is educational attainment. Migrants may face greater challenges in having their qualifications recognized and are often employed in jobs for which they are overqualified (Damas de Matos and Liebig 2014). This is particularly the case for higher-skilled migrants seeking jobs that require advanced levels of education, specialized knowledge and high skill levels. When suitable job opportunities are not locally available or accessible – especially where discriminatory barriers hinder the recognition of their foreign qualifications in the traditional labour market – professional work via remote labour platforms may appear relatively more attractive to higher-skilled migrants. In contrast, low-skilled migrants may have more alternatives in low-paid and low-skilled jobs in the traditional economy, especially where recognition of formal qualifications plays a smaller role. Based on the literature, we therefore expect the mechanism of having fewer alternatives to be more relevant for high-skilled migrant workers, particularly where remote platforms provide access to high-skilled work.
The second factor is the degree of embeddedness and integration in the traditional labour market, as manifested by having a non-platform job. Workers who do not already have a non-platform job tend to have fewer employment alternatives and may be more dependent on the platform economy to access work. They may lack country-specific resources, such as language skills or recognized qualifications, which are often needed to obtain a job in the traditional economy (Dustmann and Fabbri 2003; Zwysen 2019). We therefore expect migrants who are not employed in the traditional economy to display a greater propensity to engage in platform work.
The labour market disadvantage of not being employed in the traditional economy may stem from individual-level characteristics and circumstances, but it may also be attributed to discrimination or prejudice related to a worker’s group of origin. The ability to find and access decent work in the traditional economy is further influenced by the institutions and regulations that govern migration and employment policies. Depending on their country of origin and how they enter a country, some migrants face restrictions on their right to work and other regulatory barriers that hinder labour market integration after arrival. In the case of the EU, this applies in particular to migrants from non-EU countries. Such barriers may limit access to employment in the traditional labour market and increase the need for working-time flexibility or other non-standard arrangements, which may lead migrants to engage in platform work. Importantly, despite their stronger legal position, even intra-EU migrants face disadvantages in the labour market (Zwysen and Akgüç 2023). Migrants also differ in their relative labour market opportunities based on the regulatory framework, their resources in the host country – such as networks and language skills – and the effects of potential discrimination (Algan et al. 2010; Zwysen 2019; Zwysen, Di Stasio and Heath 2021). We therefore examine whether migrants who belong to a group that is generally more disadvantaged in the traditional labour market on the basis of region of origin – reflected in larger unexplained differences in outcomes compared to non-migrants – are more likely to engage in platform work.
3. Data, measures and methods
This study uses data from the spring and autumn waves of the 2021 ETUI Internet and Platform Work Survey (IPWS) (Piasna, Zwysen and Drahokoupil 2022). This is a cross-national survey representative of the adult population (aged 15 to 65), measuring the prevalence of digitally mediated work in 14 EU Member States,1 with at least 1,750 respondents surveyed per country. It uses random digit dialling as a sampling technique. After excluding a small number of cases owing to missing information for at least one of the key individual characteristics, the sample for analysis includes 35,440 working-age respondents. Of these, 4,148 (11.7 per cent) were born outside their country of residence and are considered to be migrants in the analysis.
While the IPWS is, to our knowledge, the best available resource for a cross-national analysis of the prevalence of platform work, it poses some limitations for this particular study. The sample may not include recent migrants or undocumented migrants, as they are less likely to have been contacted or to have responded when contacted. As the survey instrument was carried out in the official language of the country, those migrant workers who did not speak the language were not included. The results are therefore likely to underestimate the migrant population. On the other hand, nationals of a country who were born abroad may be included in the sample as migrants. These limitations are addressed in greater detail in the discussion section, which presents the robustness checks performed on the main results. A further limitation is that the survey does not include information about when individuals arrived in the host country or where they obtained their qualifications. This means that some nuance within the migrant group may be missed.
The survey includes a series of questions on different types of income-generating activities carried out via online platforms, websites or mobile apps. Based on the type of tasks performed, we classify platform work into four broad groups: (1) remote clickwork, performing generally short, standardized and low-skilled tasks, such as data entry or transcription; (2) remote professional work, involving higher-skilled projects carried out online, such as IT tasks, sales and marketing support, copy-editing or other creative work; (3) location-based transport and delivery work, performed in public spaces, such as food or goods delivery or ride-hailing; and (4) other location-based work involving services provided in the private sphere, such as general maintenance, cleaning, tutoring or babysitting. Respondents who report service work not otherwise classified are included in the “other location-based work” category. Platform work is defined as the performance of at least one of these tasks for pay in the 12 months prior to the survey. In the analysed sample, 4,050 individuals (11.4 per cent) are classified as platform workers.
In the analysis, we control for demographic and socio-economic characteristics, including: sex; age (categorized as 18–24, 25–34, 35–44, 45–54, 55–65); place of residence (large city, town, rural area); highest educational level attained (lower secondary, upper secondary or post-secondary non-tertiary, tertiary); the presence of a child aged 12 or younger in the household; and main employment status (employed, unemployed, student, other inactive).
We also examine the role of structural and institutional factors. The aim is to capture labour market disadvantage in economic terms and in employment opportunities for migrants with similar backgrounds to non-migrant respondents. This “labour market penalty” (Heath and Cheung 2006) is estimated as the difference between each broad migrant origin group and non-migrants in the probability of being employed and in occupational status, measured using the International Socio-Economic Index of Occupational Status (ISEI) (Ganzeboom and Treiman 1996). These differences are calculated at the country level, with comparisons made between migrants and non-migrants of the same sex, highest qualification, age group and place of residence. The greater the gaps in favour of non-migrants, the more disadvantaged the migrants from a given group are in that country. Such disparities may indicate discrimination, lack of networks or human capital, or other forms of disadvantage or stereotyping that affect labour market integration (see, e.g., Zwysen, Di Stasio and Heath 2021). These estimates are based on data from the 2019–21 EU-LFS, a large cross-national representative dataset collected and distributed by Eurostat. Lastly, we compare EU-born migrants with those born outside the EU, as the latter face higher formal barriers to accessing work in the EU.
The aim of the analysis is to explore the propensity of migrants to engage in platform work. First, we describe differences in the incidence of platform work between migrant and non-migrant groups and highlight heterogeneity in these differences by type of platform work and by the socio-demographic characteristics of migrants. Second, we analyse the influence of compositional factors on the propensity of migrants to engage in platform work. We do this by modelling the probability of having engaged in platform work over the last 12 months using a binary logistic regression, with migrant status as the dependent variable and controlling for individual characteristics (age, sex, the interaction of age and sex, highest qualification attained, place of residence, and the presence of young children in the household) and country fixed effects.
Lastly, we explore what mechanisms may contribute to the over-representation of migrants in platform work. We do this in two ways. First, we analyse individual characteristics by including interaction terms between migrant status and employment status, and between migrant status and education level. This tests whether migrants are more likely to engage in platform work if they are not employed in the traditional economy, and if they have higher levels of educational attainment. Second, we consider group-related characteristics by testing whether migrants, who are expected to be more disadvantaged in the labour market based on their origin, are also more likely to participate in platform work. This disadvantage is approximated in two ways: (i) by identifying migrants who belong to groups facing relatively higher labour market penalties, as estimated from the EU-LFS; and (ii) by distinguishing between migrants born in and outside the EU. A disadvantaged group is here defined as one which, based on age, sex, qualification and place of residence, experiences the greatest penalties (the most disadvantaged quartile) in employment or occupational status relative to non-migrants.
All analyses are weighted to be representative of the population and are carried out separately for the four different types of platform work identified above, as well as for all platform workers jointly. Importantly, as some respondents participate in multiple types of platform work (multihoming), they may be present in several of the platform-specific models.
4. Results
4.1. Patterns of engagement in the platform economy among migrants and non-migrants
In our sample, migrants are slightly younger (36 per cent aged 18 to 34 compared with 31 per cent of non-migrants) and more often live in large cities (49 per cent compared with 42 per cent of non-migrants). Their education levels are more polarized: 24 per cent of migrants have low educational attainment (compared with 16 per cent of non-migrants), while 34 per cent have a university degree (compared with 29 per cent of non-migrants).
As shown in table 1, 13.4 per cent of migrants (n = 554) engage in platform work compared with 11.2 per cent of non-migrants (n = 3,496). This difference of 2.2 percentage points means that migrants are nearly 20 per cent more likely to undertake platform work than non-migrants.
Table 1. Migrant and non-migrant platform and non-platform workers by socio-demographic characteristics
| Non-migrant | Migrant | |||
| Non-platform | Platform | Non-platform | Platform | |
| Total | 27 796 | 3 496 | 3 594 | 554 |
| Sex (%) | ||||
| Men | 50 | 53 | 52 | 51 |
| Women | 50 | 47 | 48 | 49 |
| Young children in household (%) | ||||
| No | 68 | 69 | 69 | 64 |
| Yes | 32 | 31 | 31 | 36 |
| Age (%) | ||||
| 18–24 | 11 | 21 | 12 | 18 |
| 25–34 | 19 | 25 | 23 | 27 |
| 35–44 | 23 | 21 | 25 | 31 |
| 45–54 | 24 | 19 | 22 | 16 |
| 55–65 | 25 | 14 | 18 | 8 |
| Residence (%) | ||||
| City | 41 | 50 | 48 | 54 |
| Town | 32 | 28 | 27 | 23 |
| Rural area | 28 | 22 | 24 | 23 |
| Level of education (%) | ||||
| Low | 17 | 10 | 26 | 14 |
| Middle | 55 | 53 | 42 | 36 |
| High | 29 | 37 | 32 | 51 |
-
Note: Percentages may not total 100 per cent on account of rounding.
Source: Our own calculations based on 2021 IPWS data.
Among the native-born population, platform workers are somewhat more likely to be men (53 per cent) and younger (46 per cent are aged 18 to 34). They also tend to have higher levels of education, with 37 per cent holding university qualifications compared with 29 per cent of non-platform workers. Among migrants these patterns differ somewhat: platform workers are more gender balanced and more likely to have young children. They also tend to be younger than non-platform workers, with a considerable share (31 per cent) aged 35 to 44. Platform workers are more likely to live in large cities – this holds true for both native-born (50 per cent) and migrant workers (54 per cent). Lastly, platform workers are more likely to hold university degrees than non-platform workers, but the difference is much more pronounced among migrants: 51 per cent of migrant platform workers hold a university degree, compared with 32 per cent of non-platform migrant workers.
These results indicate some commonalities in the way that migrant and non-migrant populations differ from each other, and in the way that platform workers differ from the general workforce. Overall, migrant platform workers stand out in terms of belonging to younger cohorts and having particularly high education levels.
To explore the mechanisms that may lead migrants into platform work, in the next step we analyse their embeddedness in the traditional labour market and the characteristics that may account for their greater vulnerability and exposure to discrimination.
As shown in table 2, students are somewhat over-represented among platform workers – both among migrants (by 3 percentage points) and non-migrants (by 6 percentage points). Nevertheless, students constitute only a minority of platform workers, accounting for 11 per cent of non-migrants and 8 per cent of migrants. Overall, the majority of platform workers are also employed in the traditional economy. Among non-migrants, platform workers are more likely than non-platform workers to be employed in the traditional economy, but less likely to be permanent (37 per cent compared with 48 per cent) and more likely to be self-employed (21 per cent compared with 10 per cent). Among migrants, platform workers are somewhat less likely to be employed in the traditional economy than their non-platform counterparts. When employed, they are also less likely to be permanent (31 per cent compared with 42 per cent) and more likely to be self-employed (20 per cent compared with 11 per cent). In addition, migrant platform workers are more likely than non-platform migrants to be otherwise unemployed (21 per cent compared with 15 per cent), while there is no such difference among non-migrants. Lastly, platform workers – both migrant and non-migrant – work shorter hours than non-platform workers. However, among migrants, platform work makes little difference in this respect, as they generally work fewer hours than non-migrants, regardless of platform involvement. These findings reveal differences in labour market embeddedness, with platform work serving as an alternative to traditional employment – particularly for migrants.
Table 2. Migrant and non-migrant platform and non-platform workers by socio-economic vulnerability
| Non-migrant | Migrant | |||
| Non-platform | Platform | Non-platform | Platform | |
| Total | 27 796 | 3 496 | 3 594 | 554 |
| Employment status (%) | ||||
| Employed – permanent | 48 | 37 | 42 | 31 |
| Employed – temporary | 13 | 14 | 16 | 15 |
| Self-employed | 10 | 21 | 11 | 20 |
| Unemployed | 9 | 9 | 15 | 21 |
| Student | 5 | 11 | 5 | 8 |
| Inactive | 15 | 8 | 11 | 5 |
| Weekly working hours (%) | ||||
| <20 hours | 6 | 8 | 9 | 10 |
| 20–34 hours | 16 | 20 | 22 | 22 |
| 35+ hours | 75 | 68 | 66 | 64 |
| Varying | 3 | 3 | 3 | 4 |
| Country of birth (%) | ||||
| Country of residence | 100 | 100 | 0 | 0 |
| EU | 0 | 0 | 34 | 36 |
| Europe – Other | 0 | 0 | 25 | 24 |
| Africa and Middle East | 0 | 0 | 21 | 14 |
| Asia | 0 | 0 | 5 | 8 |
| Americas, Australia | 0 | 0 | 14 | 18 |
| Labour market penalty (%) | ||||
| High employment penalty | n/a | n/a | 22 | 29 |
| Low employment penalty | n/a | n/a | 78 | 71 |
| High occupational (ISEI) penalty | n/a | n/a | 21 | 29 |
| Low occupational (ISEI) penalty | n/a | n/a | 79 | 71 |
-
Notes: Table shows column percentages; n/a = not applicable. Percentages may not total 100 per cent on account of rounding.
Source: Our own calculations based on 2021 IPWS data.
Among migrants, there is some variation in the propensity to engage in platform work depending on their region of origin, with those doing platform work somewhat more likely to be born in another EU Member State, Asia, or the Americas and Australia.
Lastly, we explore whether migrants who are expected to face relatively high penalties in terms of employment or occupational status, compared to otherwise similar native-born workers, resort to platform work more often. The results show that more disadvantaged migrant groups are over-represented among platform workers, with 29 per cent in the quartile of largest penalties, in both employment and occupational status2 – compared with approximately 20 per cent among non-platform workers.
While thus far all types of platform work have been considered jointly, it is crucial to recognize that platform tasks vary greatly in terms of skill level, complexity, pay and place of work. Accordingly, in the next step we explore the differences between migrants and non-migrants in their patterns of engagement in various types of platform work.
As shown in table 3, migrants are generally less likely than non-migrants to engage in remote platform work at all skill levels, but they are much more likely to perform location-based transport and delivery work. A quarter of migrant platform workers are engaged in transport and delivery services, compared with 13.3 per cent of non-migrants. Migrants are also more likely to undertake multiple types of platform work (20.5 per cent compared with 15.6 per cent). For native-born platform workers, higher levels of education are associated with greater participation in higher-skilled remote professional work (22.8 per cent) and lower involvement in transport and delivery services (9.8 per cent). Among migrants, there appears to be less of a skill match in this respect: only 14.9 per cent of the highly educated perform remote professional work, while 24.6 per cent work in transport and delivery.
Table 3. Heterogeneity of platform workers, by migration status and type of platform work
| Non-migrants | Count | Remote clickwork (%) | Remote professional (%) | Location-based transport and delivery (%) | Other location-based work (%) | Multiple (%) |
| Total | 3 496 | 36.4 | 17.4 | 13.3 | 17.4 | 15.6 |
| Level of education | ||||||
| Low | 177 | 42.7 | 7.8 | 15.6 | 21.8 | 12.1 |
| Middle | 1 462 | 33.6 | 15.4 | 15.3 | 20.2 | 15.4 |
| High | 1 857 | 38.6 | 22.8 | 9.8 | 12.1 | 16.7 |
| Employment status | ||||||
| Employed – permanent | 1 437 | 40.0 | 16.6 | 11.9 | 15.9 | 15.6 |
| Employed – temporary | 470 | 32.4 | 17.4 | 13.3 | 22.3 | 14.6 |
| Self-employed | 722 | 25.2 | 22.7 | 18.1 | 11.5 | 22.4 |
| Unemployed | 272 | 32.7 | 16.7 | 16.9 | 22.9 | 10.8 |
| Student | 345 | 44.1 | 16.1 | 8.3 | 18.8 | 12.8 |
| Inactive | 250 | 49.6 | 9.4 | 10.1 | 22.4 | 8.5 |
| Weekly working hours | ||||||
| <20 hours | 232 | 24.6 | 19.8 | 5.8 | 17.7 | 32.0 |
| 20–34 hours | 535 | 34.9 | 23.9 | 10.9 | 14.9 | 15.3 |
| 35+ hours | 1 765 | 35.5 | 16.9 | 15.8 | 16.1 | 15.8 |
| Varying | 75 | 32.4 | 18.7 | 13.5 | 12.2 | 23.3 |
| Total | 554 | 27.1 | 11.5 | 24.7 | 16.2 | 20.5 |
| Level of education | ||||||
| Low | 43 | 32.1 | 0.0 | 30.3 | 19.3 | 18.3 |
| Middle | 161 | 27.9 | 11.1 | 22.6 | 20.6 | 17.8 |
| High | 350 | 25.2 | 14.9 | 24.6 | 12.3 | 23.0 |
| Employment status | ||||||
| Employed – permanent | 166 | 39.5 | 7.4 | 22.4 | 14.2 | 16.5 |
| Employed – temporary | 84 | 24.8 | 19.1 | 22.8 | 15.4 | 18.0 |
| Self-employed | 123 | 19.9 | 10.7 | 26.7 | 18.6 | 24.0 |
| Unemployed | 106 | 8.2 | 13.7 | 30.5 | 18.5 | 29.1 |
| Student | 46 | 42.4 | 17.4 | 16.9 | 9.0 | 14.3 |
| Inactive | 29 | 37.7 | 0.0 | 24.6 | 23.0 | 14.7 |
| Weekly working hours | ||||||
| <20 hours | 39 | 14.2 | 7.7 | 34.7 | 10.1 | 33.3 |
| 20–34 hours | 76 | 34.9 | 5.1 | 27.9 | 17.1 | 15.0 |
| 35+ hours | 239 | 32.1 | 13.0 | 21.6 | 16.7 | 16.5 |
| Varying | 19 | 16.9 | 19.6 | 11.0 | 8.4 | 44.1 |
| Country of birth | ||||||
| EU | 179 | 32.1 | 10.5 | 22.4 | 19.4 | 15.6 |
| Europe – Other | 130 | 18.7 | 8.4 | 30.5 | 15.4 | 27.0 |
| Africa and Middle East | 88 | 24.1 | 12.1 | 23.2 | 7.5 | 33.1 |
| Asia | 40 | 29.7 | 13.7 | 31.0 | 19.8 | 5.9 |
| Americas, Australia | 117 | 29.7 | 16.4 | 19.6 | 15.8 | 18.5 |
| Labour market penalties | ||||||
| High employment penalty | 165 | 29.4 | 14.3 | 30.8 | 6.8 | 18.8 |
| Low employment penalty | 363 | 25.2 | 10.3 | 23.1 | 20.1 | 21.4 |
| High occupational (ISEI) penalty | 150 | 32.7 | 8.4 | 22.4 | 14.4 | 22.1 |
| Low occupational (ISEI) penalty | 365 | 21.9 | 13.1 | 27.2 | 17.4 | 20.4 |
-
Note: Percentages may not total 100 per cent on account of rounding.
Source: Our own calculations based on 2021 IPWS data.
Differences between migrants and non-migrants in the types of platform work they perform also vary by employment status in the traditional economy. In particular, unemployed migrants are more likely to engage in transport and delivery (30.5 per cent) and multihoming (29.1 per cent), but they are less likely to perform remote clickwork (8.2 per cent). In contrast, clickwork is the most common activity among unemployed non-migrants (32.7 per cent). Whereas 22.7 per cent of self-employed native-born workers engage in remote professional work, only 10.7 per cent of self-employed migrants do this type of work; instead, they are most likely to engage in transport and delivery (26.7 per cent).
By country of birth, remote clickwork is most prevalent among EU-born migrants and those from Asia, the Americas and Australia, while transport and delivery is most common among migrants born in Asia and in other European (non-EU) countries. Migrants facing the greatest occupational penalties are relatively more likely to engage in remote clickwork, while those with the highest employment penalties are more likely to engage in both types of remote platform work as well as transport and delivery.
4.2. Mapping the over-representation of migrants in platform work accounting for compositional differences
The overview presented in the previous section highlights substantial differences between the migrant and native-born populations in our sample, both in terms of their socio-demographic characteristics and patterns of engagement in the platform economy. Migrant workers are more likely to participate in platform work – especially location-based transport and delivery – but less likely to engage in remote platform work. Migrants perceived to be in a more vulnerable and disadvantaged situation in the host country – in terms of relative employment opportunities and being born outside the EU – are, on average, more likely to engage in platform work, particularly in location-based delivery and transport. However, as migrants differ from the native-born population in terms of educational attainment, age, family situation and employment status, they may have different work-related needs and opportunities. These differences may influence their participation in both the traditional and platform economies. This highlights the need to account for compositional differences between the migrant and native-born populations to allow for comparisons between otherwise similar individuals.
Figure 1 illustrates the relative difference in the probability of platform work engagement between migrants and non-migrants, accounting for their socio-demographic differences.3 The results indicate that migrants are 13 per cent more likely to engage in platform work, corresponding to a difference of 1.3 percentage points in the incidence of such work. Accordingly, accounting for compositional differences reduces migrants’ over-representation in platform work by almost half, from 2.2 to 1.3 percentage points. By type of platform work, migrants are twice as likely to engage in transport and delivery (a 99 per cent higher likelihood compared with non-migrants), and 28 per cent more likely to carry out other location-based work.
Figure 1. Relative difference in the probability of platform work engagement between migrants and non-migrants (percentages)
Notes: Weighted migrant gap in platform work engagement with 90 per cent confidence interval, controlling for sex, age, sex by age, place of residence, level of education, having a young child, wave of survey and country fixed effects. Expressed as a percentage difference compared with the non-migrant population, with positive values indicating over-representation of migrants.
Source: Our own calculations based on 2021 IPWS data.
4.3. Heterogeneity among migrants by labour market disadvantage
We now explore the over-representation of migrants in the platform economy, as identified in the previous section, in light of the drivers proposed in the literature relating to labour market embeddedness and opportunities in the traditional economy. We thus test which characteristics play a greater role among migrants when compared with non-migrants, and within the migrant group, accounting for compositional differences between these populations.
Figure 2 shows how migrant gaps vary by employment status.4 When they are not employed in the traditional labour market, migrants are generally more likely (by 22 per cent) to engage in platform work compared with similar non-migrant respondents, while there is no statistically significant difference among those who are employed elsewhere. This finding is consistent with previous research showing that platform work is more likely to be taken up by migrants in place of, rather than in addition to, offline opportunities (Altenried 2024), but the tasks that they are able to access are in the lowest paid activities.
Figure 2. Relative difference in the probability of platform work engagement between migrants and non-migrants, by employment status (percentages)
Notes: Weighted migrant gap in platform work engagement with 90 per cent confidence interval, controlling for sex, age, sex by age, place of residence, level of education, having a young child, wave of survey and country fixed effects. Expressed as a percentage difference compared with the non-migrant population, with positive values indicating over-representation of migrants.
Source: Our own calculations based on 2021 IPWS data.
Figure 3 reveals that, while migrants are, on average, more likely to engage in platform work than non-migrants, this is closely related to their educational attainment. Migrants with a university degree are much more likely (by 26 per cent) to engage in platform work than their non-migrant counterparts, while there is no significant difference among those without a university education. This pattern holds across all types of platform work. It should be noted, however, that the data do not distinguish between higher qualifications obtained in the country of origin and those acquired in the host country. Qualifications earned abroad are more likely to be discounted or not recognized (Zwysen and Demireva 2025).
Figure 3. Relative difference in the probability of platform work engagement between migrants and non-migrants, by education (percentages)
Notes: Weighted migrant gap in platform work engagement with 90 per cent confidence interval, controlling for sex, age, sex by age, place of residence, level of education, having a young child, wave of survey and country fixed effects. Expressed as a percentage difference compared with the non-migrant population, with positive values indicating over-representation of migrants.
Source: Our own calculations based on 2021 IPWS data.
Next, we explore whether migrants’ over-representation in platform work is related to group-level differences capturing the penalties and disadvantages faced by migrants with similar origins and socio-demographic characteristics. We first analyse the difference between migrants born in another EU Member State and those born in non-EU countries, who may face more barriers to employment in the host country and hence be more likely to resort to more accessible platform economy jobs. Contrary to our expectations, figure 4 shows that, after accounting for compositional differences, EU-born migrants are generally more likely to engage in platform work. This seems to be mainly driven by higher involvement of EU-born migrants in remote clickwork, but small sample sizes do not allow us to draw firm conclusions in this regard.5
Figure 4. Relative difference in the probability of platform work engagement between migrants and non-migrants, by country of birth (percentages)
Notes: Weighted migrant gap in platform work engagement with 90 per cent confidence interval, controlling for sex, age, sex by age, place of residence, level of education, having a young child, wave of survey and country fixed effects. Expressed as a percentage difference compared with the non-migrant population, with positive values indicating over-representation of migrants.
Source: Our own calculations based on 2021 IPWS data.
For a more comprehensive view on the role of the labour market penalties in platform work engagement, in the next step we differentiate migrants based on group-related disadvantage in employment opportunities and in occupational status (Heath and Cheung 2006; Zwysen, Di Stasio and Heath 2021). The results indicate that migrant groups who face relatively high penalties in employment (figure 5) are substantially (and statistically significantly) more likely to engage in the platform economy: the gap is 41 per cent compared to non-migrants, while it amounts to only 6 per cent (and is not statistically significant) for migrants with low penalties in employment. These effects are statistically significant for platform work in general and are particularly pronounced in transport and delivery, whereas they are much weaker for remote clickwork and remote professional work.
Figure 5. Relative difference in the probability of platform work engagement between migrants and non-migrants, by employment penalty (percentages)
Notes: Weighted migrant gap in platform work engagement with 90 per cent confidence interval, controlling for sex, age, sex by age, place of residence, level of education, having a young child, wave of survey and country fixed effects. Expressed as a percentage difference compared with the non-migrant population, with positive values indicating over-representation of migrants.
Source: Our own calculations based on 2021 IPWS and 2019–21 EU-LFS data.
Lastly, figure 6 illustrates a similar analysis for penalties in terms of occupational status (ISEI). Overall, migrants who are expected to face greater disadvantage in the traditional labour market are more likely to work on platforms – on average, they are 40 per cent more likely to do so compared with non-migrants, while the gap amounts to only 4 per cent for migrants facing lower levels of disadvantage. This holds true for platform work in general as well as for its different types, especially remote clickwork.
Figure 6. Relative difference in the probability of platform work engagement between migrants and non-migrants, by occupational status penalty (percentages)
Notes: Weighted migrant gap in platform work engagement with 90 per cent confidence interval, controlling for sex, age, sex by age, place of residence, level of education, having a young child, wave of survey and country fixed effects. Expressed as a percentage difference compared with the non-migrant population, with positive values indicating over-representation of migrants.
Source: Our own calculations based on 2021 IPWS and 2019–21 EU-LFS data.
On the whole, the results reveal the role of structural, rather than just individual-level, mechanisms in increasing the probability of platform work among migrants. Those who have fewer alternatives, facing either a lack of recognition of their experience and qualifications or more discrimination, are more likely to work outside the traditional labour market (Pager and Pedulla 2015; Zwysen, Di Stasio and Heath 2021).
5. Discussion
The aim of this study was to analyse the prevalence of platform work among migrants across Europe by testing the generalizability of previous research findings across socio-demographic groups and different types of platform work. In general, we have found that migrants are approximately 20 per cent more likely to engage in platform work than the non-migrant population – corresponding to a 2.2 percentage point difference in the prevalence of platform work – predominantly working in food delivery and ride-hailing. The analysis has confirmed that the over-representation of migrants in platform work is at least partly related to their particular socio-demographic characteristics: they are younger, have higher levels of education and are less likely to be employed in the traditional economy, making them more similar to the overall platform workforce. After accounting for such compositional differences, the over-representation of migrants in platform work drops to 13 per cent, or 1.3 percentage points, but does not disappear completely.
The findings are in line with expectations and previous research indicating that migrants are drawn to platform work owing to a lack of suitable alternatives in the traditional labour market. This may stem from barriers to employment in general or difficulty in finding jobs matching their educational qualifications. The pattern is particularly evident among migrants with higher educational attainment and higher skills, who are more likely to engage in digitally mediated labour markets. In contrast, lower-skilled migrants often have more alternatives in the form of low-paid jobs in the traditional economy. Furthermore, migrants belonging to socio-demographic groups that experience poorer labour market outcomes in host countries are particularly likely to engage in platform work. A greater need for flexibility to fit in with other commitments – as in the case of students – is another explanation for the higher involvement in the platform economy of migrants without traditional, non-platform employment (Lam and Triandafyllidou 2024; van Doorn and Vijay 2024). However, the higher proportion of non-employed students may also be explained by the use of student visas by economic migrants, who apply for them because they are easier to obtain, thus (in)formally combining work with education.
By extending the analysis to include different types of platform work, the results show that the over-representation of migrants in the platform economy, which has mainly been observed in location-based work (Kowalik, Lewandowski and Kaczmarczyk 2022) cannot easily be generalized to other forms of platform work. Migrants are particularly more likely to work in location-based delivery or transport work, where they are almost twice as prevalent as non-migrants, and they are around 30 per cent more likely to work in other location-based work. However, there is no migrant over-representation in remote work. This is contrary to our expectations and can be surprising given migrants’ relatively high education levels and access to the larger cross-national labour market that this type of work offers.
Therefore, our findings provide robust empirical support for earlier, mostly qualitative studies finding that migrants are more likely to engage in platform work owing to its relatively low entry barriers (Altenried 2024; van Doorn and Vijay 2024). Furthermore, migrants who are more vulnerable in terms of not having a job in the traditional economy – which increases their economic dependence on platforms – and those who face a higher relative penalty in traditional employment in terms of their occupational status, are more likely to engage in platform work.
There are some limitations to this analysis related to data availability. It is based on a cross-national survey designed to capture a representative sample of the general population aged 18 to 65 and is not specifically targeted at migrants. Although anyone with a phone number could have been sampled, some of the most vulnerable migrants are likely to have been excluded – either because they did not yet have a local phone number, did not know the language well enough to take part in the survey, or were more reluctant to respond. As a result, our estimates are likely to represent the lower bound of the differences between migrants and non-migrants, as the more vulnerable migrants could be expected to have even higher rates of participation in the platform economy (see, e.g., van Doorn, Ferrari and Graham 2023). Moreover, we are only able to identify first-generation migrants. Other migrants are considered jointly with the native-born population as a comparison group, even though discrimination and differences in labour market integration are likely to persist into subsequent generations.
To address these limitations, we conducted a number of robustness tests. First, we assessed whether migrants were under-sampled in the IPWS by comparing it with the 2020–21 EU-LFS, which is a large representative survey feeding into official European statistics. Table SA5 in the supplementary online appendix compares the shares of migrants in the IPWS and the EU-LFS and shows that the IPWS generally slightly overestimates the foreign-born population, especially in the case of third-country migrants, but that there is no systematic bias in it. We then reweighted the IPWS data so that the proportions of migrants by demographic characteristics (age and sex) and region of origin matched those in the EU-LFS and repeated the main analyses using the reweighted data. The analysis presented in figure SA1 in the supplementary online appendix shows consistent results, with some of the migrant gaps being somewhat higher after reweighting.
Lastly, this study focuses on a selection of measures of employment opportunities and economic and social conditions. It therefore does not account for other factors that may influence migrants’ labour market outcomes and participation in platform work. In particular, the general quality of employment opportunities is expected to influence the uptake of platform work (Zwysen and Piasna 2024). Future iterations of this analysis could incorporate additional characteristics of local labour markets to better capture these dynamics. The heterogeneity of the platform economy also warrants further exploration, especially in terms of cross-national variation in business and employment models. Even within the EU, significant differences exist in regulation and worker protection, with the same platform companies using different employment arrangements in different countries. These differences may contribute to variation in migrants’ engagement in platform work. In this respect, the effects of supranational regulation and standard-setting – such as the recently adopted EU Directive on Platform Work,6 harmonizing employment status classification to some extent – deserves further investigation. Taken together, these considerations point to potential avenues for further research, for which this study provides an empirical basis.
6. Conclusions
The emergence and growth of the platform economy around the world have been heralded as the next step in the changing nature of work in capitalist economies, unfolding alongside technological transformation. Thanks to remarkable research efforts, we now have a thorough understanding of working and employment conditions in platform work, which has allowed a mapping of the new challenges associated with technological innovation against continuing forms of worker exploitation and asymmetrical power relations. However, some gaps in knowledge persist. One of them concerns the mechanisms behind the growth of the platform economy and the factors that drive people to platform work. This study contributes to filling that gap by exploring the prevalence of platform work among a particularly vulnerable segment of the workforce – first-generation migrants – and by identifying the social and economic conditions associated with their increased propensity to join the platform economy.
In doing so, we shed new light on the dynamics behind these emerging forms of work. While earlier processes of employment fragmentation, outsourcing and subcontracting were aimed at relocating work to places with lower labour costs, the strategy of platform companies seems to be based on finding pools of cheap labour in places that also offer access to a sufficiently affluent customer base. In attempting to identify the mechanisms that push migrants towards the platform economy, we have found a clear pattern: vulnerable workers who, on average, have fewer resources and belong to groups that are more disadvantaged in employment are more likely to engage in platform work. The analysis also provides some support for the role of lower entry barriers as a key mechanism driving migrant participation. However, it is not low barriers in platform work per se that account for this trend, but rather the relative ease of joining a platform compared with traditional employment in a given country or region.
Previous research has highlighted the generally poor quality of platform work. It provides limited, if any, access to worker rights and social protection, is typically characterized by precarious working conditions, and plays only a restricted role as a stepping stone to employment in the traditional economy (Altenried 2024; Lenaerts et al. 2023). Therefore, while the platform economy may offer short-term labour market integration for the foreign-born population and other vulnerable groups, in the long term it risks severely harming their prospects and contributing to the accumulation of disadvantage and economic and social exclusion.
This article is, to the best of our knowledge, the first to examine the over-representation of migrants in platform work across Europe using a representative population sample. Its contribution also lies in its inclusion of different types of platform work, beyond food delivery or ride-hailing. This provides some much-needed context on the importance of this type of work for migrants.
First, we have challenged the view that platform work is predominantly migrant work. While foreign-born workers are more likely to engage in ride-hailing and delivery services than native-born workers, the majority of platform workers in our dataset are actually native-born. This distinction is important when discussing the need to regulate and set standards for platform work, which is sometimes dismissed as entry-level employment for migrants or as a transitional phase towards greater labour market integration. Importantly, platform work is not confined to a single type of worker but is widespread among groups of workers in precarious situations. Second, we have found evidence that migrants are more likely to engage in platform work when confronted with a paucity of alternative options matching their education or qualifications. This finding is in line with previous research indicating that internet and platform work are more prevalent in regions where better employment options in the traditional labour market are limited (Zwysen and Piasna 2024).
Platform work thus appears to be a symptom of the challenges of labour market integration faced by migrants who are unable to secure suitable jobs in the traditional economy, as well as more general deficits in access to decent work. A fundamental policy consideration should therefore be to pursue a more integrated approach targeting labour standards in the traditional economy and compliance in the platform economy, coupled with the prevention of the exploitation of more vulnerable workers.
Notes
- Countries included in the survey: Austria, Bulgaria, Czechia, Estonia, France, Germany, Greece, Hungary, Ireland, Italy, Poland, Romania, Slovakia and Spain. ⮭
- There is some overlap between the penalties in terms of employment and occupational status, but these do not concern all the same people. Of those with high penalties in terms of employment, 42 per cent also have high penalties in terms of occupational status, while 58 per cent do not. ⮭
- See supplementary online appendix, table SA1 for the underlying calculations, and table SA2 for the detailed regression results by type of platform work. ⮭
- Detailed regression results for engagement in any type of platform work are reported in table SA3 in the supplementary online appendix. Further results are available upon request. ⮭
- Detailed regression results for engagement in any type of platform work are shown in table SA4 in the supplementary online appendix. ⮭
- European Union, Directive (EU) 2024/2831 of the European Parliament and of the Council of 23 October 2024 on improving working conditions in platform work. ⮭
Competing interests
The authors declare that they have no competing interests.
References
Adams, Abigail, and Janine Berg. 2021. “When Home Affects Pay: An Analysis of the Gender Pay Gap among Crowdworkers”. In The Future of Diversity, edited by Christiane Kuptsch and Éric Charest, 95–119. Geneva: ILO.
Algan, Yann, Christian Dustmann, Albrecht Glitz, and Alan Manning. 2010. “The Economic Situation of First and Second-Generation Immigrants in France, Germany and the United Kingdom”. Economic Journal 120 (542): F4–F30. http://doi.org/10.1111/j.1468-0297.2009.02338.x.
Altenried, Moritz. 2024. “Mobile Workers, Contingent Labour: Migration, the Gig Economy and the Multiplication of Labour”. Environment and Planning A: Economy and Space 56 (4): 1113–1128. http://doi.org/10.1177/0308518X211054846.
Berg, Janine. 2016. “Income Security in the On-Demand Economy: Findings and Policy Lessons from a Survey of Crowdworkers”. Comparative Labor Law & Policy Journal 37 (3): 543–573.
Berger, Thor, Carl Benedikt Frey, Guy Levin, and Santosh Rao Danda. 2019. “Uber Happy? Work and Well-Being in the ‘Gig Economy’”. Economic Policy 34 (99): 429–477. http://doi.org/10.1093/epolic/eiz007.
Damas de Matos, Ana. 2017. “Firm Heterogeneity and Immigrant Wage Assimilation”. Applied Economics Letters 24 (9): 653–657. http://doi.org/10.1080/13504851.2016.1218421.
Damas de Matos, Ana, and Thomas Liebig. 2014. “The Qualifications of Immigrants and Their Value in the Labour Market: A Comparison of Europe and the United States”. In Matching Economic Migration with Labour Market Needs, 187–228. Paris: OECD.
Dustmann, Christian, and Francesca Fabbri. 2003. “Language Proficiency and Labour Market Performance of Immigrants in the UK”. Economic Journal 113 (489): 695–717. http://doi.org/10.1111/1468-0297.t01-1-00151.
Fellini, Ivana, and Raffaele Guetto. 2019. “A ‘U-Shaped’ Pattern of Immigrants’ Occupational Careers? A Comparative Analysis of Italy, Spain, and France”. International Migration Review 53 (1): 26–58. http://doi.org/10.1177/0197918318767931.
Ganzeboom, Harry B.G., and Donald J. Treiman. 1996. “Internationally Comparable Measures of Occupational Status for the 1988 International Standard Classification of Occupations”. Social Science Research 25 (3): 201–239. http://doi.org/10.1006/ssre.1996.0010.
Graham, Mark, Isis Hjorth, and Vili Lehdonvirta. 2017. “Digital Labour and Development: Impacts of Global Digital Labour Platforms and the Gig Economy on Worker Livelihoods”. Transfer: European Review of Labour and Research 23 (2): 135–162. http://doi.org/10.1177/1024258916687250.
Heath, Anthony, and Sin Yi Cheung. 2006. “Ethnic Penalties in the Labour Market: Employers and Discrimination”. Research Report No. 341. Leeds: Department for Work and Pensions.
Holtum, Peter James, Elnaz Irannezhad, Greg Marston, and Renuka Mahadevan. 2022. “Business or Pleasure? A Comparison of Migrant and Non-Migrant Uber Drivers in Australia”. Work, Employment and Society 36 (2): 290–309. http://doi.org/10.1177/09500170211034741.
Jeon, Sung-Hee, Huju Liu, and Yuri Ostrovsky. 2021. “Measuring the Gig Economy in Canada Using Administrative Data”. Canadian Journal of Economics/Revue canadienne d’économique 54 (4): 1638–1666. http://doi.org/10.1111/caje.12558.
Kässi, Otto, and Vili Lehdonvirta. 2018. “Online Labour Index: Measuring the Online Gig Economy for Policy and Research”. Technological Forecasting and Social Change 137 (December): 241–248. http://doi.org/10.1016/j.techfore.2018.07.056.
Kowalik, Zuzanna, Piotr Lewandowski, and Paweł Kaczmarczyk. 2022. “Job Quality Gaps between Migrant and Native Gig Workers: Evidence from Poland”. IBS Working Paper 09/2022. Warsaw: Instytut Badań Strukturalnych.
Lam, Laura, and Anna Triandafyllidou. 2024. “Road to Nowhere or to Somewhere? Migrant Pathways in Platform Work in Canada”. Environment and Planning A: Economy and Space 56 (4): 1150–1169. http://doi.org/10.1177/0308518X221090248.
Lehdonvirta, Vili. 2018. “Flexibility in the Gig Economy: Managing Time on Three Online Piecework Platforms”. New Technology, Work and Employment 33 (1): 13–29. http://doi.org/10.1111/ntwe.12102.
Lenaerts, Karolien, Karen Van Aerden, Dirk Gillis, Jessie Gevaert, Ignace Pollet, Katrien Steenssens, Ine Smits, and Christophe Vanroelen. 2023. Personen met een (grote) afstand tot de arbeidsmarkt en de platformeconomie in Vlaanderen. Leuven: KU Leuven.
McMillan Cottom, Tressie. 2020. “Where Platform Capitalism and Racial Capitalism Meet: The Sociology of Race and Racism in the Digital Society”. Sociology of Race and Ethnicity 6 (4): 441–449. http://doi.org/10.1177/2332649220949473.
Munoz, Isabel, Steve Sawyer, and Michael Dunn. 2022. “New Futures of Work or Continued Marginalization? The Rise of Online Freelance Work and Digital Platforms”. CHIWORK ’22: Proceedings of the 1st Annual Meeting of the Symposium on Human–Computer Interaction for Work: Article No. 18. http://doi.org/10.1145/3533406.3533412.
Pager, Devah, and David S. Pedulla. 2015. “Race, Self-Selection, and the Job Search Process”. American Journal of Sociology 120 (4): 1005–1054. http://doi.org/10.1086/681072.
Pais, Ivana, and Arianna Marcolin. 2024. “Digital Platforms in the Italian Domestic Care Sector: The Emergence of an Unprecedented Corporate Logic and Its Implications for Workers’ Social Protection”. International Labour Review 163 (3): 397–415. http://doi.org/10.1111/ilr.12433.
Peticca-Harris, Amanda, Nadia deGama, and M.N. Ravishankar. 2020. “Postcapitalist Precarious Work and Those in the ‘Drivers’ Seat: Exploring the Motivations and Lived Experiences of Uber Drivers in Canada”. Organization 27 (1): 36–59. http://doi.org/10.1177/1350508418757332.
Piasna, Agnieszka, and Jan Drahokoupil. 2021. “Flexibility Unbound: Understanding the Heterogeneity of Preferences among Food Delivery Platform Workers”. Socio-Economic Review 19 (4): 1397–1419. http://doi.org/10.1093/ser/mwab029.
Piasna, Agnieszka, Wouter Zwysen, and Jan Drahokoupil. 2022. “The Platform Economy in Europe: Results from the Second ETUI Internet and Platform Work Survey”. ETUI Working Paper No. 2022.05. Brussels: European Trade Union Institute.
Pirina, Giorgio, Francesco Della Puppa, and Fabio Perocco. 2024. “The Time–Space–Energy Nexus in the Gig Economy: The Work and Everyday Lives of Migrant Food Delivery ‘Walkers’ in Venice”. Journal of Ethnic and Migration Studies 50 (15): 3822–3838. http://doi.org/10.1080/1369183X.2024.2379648.
Pulignano, Valeria, Agnieszka Piasna, Markieta Domecka, Karol Muszyński, and Lander Vermeerbergen. 2021. “Does It Pay to Work? Unpaid Labour in the Platform Economy”. ETUI Policy Brief 2021.15. Brussels: European Trade Union Institute.
Rodríguez-Modroño, Paula, Astrid Agenjo-Calderón, and Purificación López-Igual. 2024. “A Social Reproduction Analysis of Digital Care Platform Work”. New Political Economy 29 (4): 616–627. http://doi.org/10.1080/13563467.2024.2317703.
Schor, Juliet B., William Attwood-Charles, Mehmet Cansoy, Isak Ladegaard, and Robert Wengronowitz. 2020. “Dependence and Precarity in the Platform Economy”. Theory and Society 49 (5–6): 833–861. http://doi.org/10.1007/s11186-020-09408-y.
Urzì Brancati, Maria Cesira, Annarosa Pesole, and Enrique Fernández-Macías. 2020. New Evidence on Platform Workers in Europe: Results from the Second COLLEEM Survey. Luxembourg: Publications Office of the European Union.
van Doorn, Niels. 2017. “Platform Labor: On the Gendered and Racialized Exploitation of Low-Income Service Work in the ‘On-Demand’ Economy”. Information, Communication & Society 20 (6): 898–914. http://doi.org/10.1080/1369118X.2017.1294194.
van Doorn, Niels, Fabian Ferrari, and Mark Graham. 2023. “Migration and Migrant Labour in the Gig Economy: An Intervention”. Work, Employment and Society 37 (4): 1099–1111. http://doi.org/10.1177/09500170221096581.
van Doorn, Niels, and Darsana Vijay. 2024. “Gig Work as Migrant Work: The Platformization of Migration Infrastructure”. Environment and Planning A: Economy and Space 56 (4): 1129–1149. http://doi.org/10.1177/0308518X211065049.
Van Mol, Christof, and Helga de Valk. 2016. “Migration and Immigrants in Europe: A Historical and Demographic Perspective”. In Integration Processes and Policies in Europe: Contexts, Levels and Actors, edited by Blanca Garcés-Mascareñas and Rinus Penninx, 31–56. Cham: Springer International Publishing.
Vandaele, Kurt. 2021. “Collective Resistance and Organizational Creativity amongst Europe’s Platform Workers: A New Power in the Labour Movement?” In Work and Labour Relations in Global Platform Capitalism, edited by Julieta Haidar and Maarten Keune, 206–235. Cheltenham: Edward Elgar.
Wood, Alex J., Mark Graham, Vili Lehdonvirta, and Isis Hjorth. 2019. “Good Gig, Bad Gig: Autonomy and Algorithmic Control in the Global Gig Economy”. Work, Employment and Society 33 (1): 56–75. http://doi.org/10.1177/0950017018785616.
Zaiceva, Anzelika, and Klaus F. Zimmermann. 2014. “Migration and the Demographic Shift”. IZA Discussion Paper No. 8743. Bonn: Institute for the Study of Labor.
Zwysen, Wouter. 2019. “Different Patterns of Labor Market Integration by Migration Motivation in Europe: The Role of Host Country Human Capital”. International Migration Review 53 (1): 59–89. http://doi.org/10.1177/0197918318767929.
Zwysen, Wouter, and Mehtap Akgüç. 2023. “Analysing the Flows and Labour Market Outcomes of Short-Term Mobile Workers in the EU”. ETUI Working Paper No. 2023.07. Brussels: European Trade Union Institute.
Zwysen, Wouter, and Neli Demireva. 2025. “The Importance of Host Country Human Capital for the Labour Market Integration of Different Migrant Groups in Europe”. ISER Working Paper Series, No. 2025-01. Colchester: University of Essex.
Zwysen, Wouter, Valentina Di Stasio, and Anthony Heath. 2021. “Ethnic Penalties and Hiring Discrimination: Comparing Results from Observational Studies with Field Experiments in the UK”. Sociology 55 (2): 263–282. http://doi.org/10.1177/0038038520966947.
Zwysen, Wouter, and Agnieszka Piasna. 2024. “Pushed Online: What Characteristics of Regional Offline Labour Markets Influence the Expansion of Internet and Platform Work?” New Technology, Work and Employment 39 (1): 63–88. http://doi.org/10.1111/ntwe.12277.