Care for Young Migrants in Gauteng: Challenging Normative Assumptions of Family and Institutional Care

Abstract

South Africa is host to a significant number of young migrants who are in institutional care. This study aimed to understand how migrant youth access care, who provides it and their views about the quality of care they receive, given the precarity of urban life in South African cities. The study used a mixed methods, explanatory, sequential research design. In this paper, we present some of the research findings that help understand how young migrants access care in contexts of urban self-settlement and how they create family through everyday care practices to meet their needs. The results show that they maintain ties with family members and meet their needs through a combination of NGO-provided material support and family-provided emotional support. Notably State support was largely absent, making it impossible to meet their needs when the State had a monopoly on service provision (such as documentation and asylum). Being listened to and being able to take decisions regarding their care were extremely important to young migrants and were significantly correlated with better optimism for the future, happiness with the services they receive and optimism for life in South Africa. These findings underscore the adaptive and largely effective ways that young migrants and families meet their care needs in a precarious urban setting. They provide important understanding to ensure that the design of services for young migrants recognise existing care practices that are effective and support rather than supplant them.

1. Introduction

Since democracy in 1994, South Africa has become a leading destination country on the continent for migrants, including young migrants1. Most youth migrating to South Africa are from Zimbabwe, Mozambique, the DRC, and the Horn of Africa (UNDESA, 2017). Unlike many of its neighbouring countries, South Africa has a policy of integration for all migrants, including refugees. As a result, migrant youth and families have settled primarily in urban centres, where services and opportunities are more readily available than in other areas (see Landau, 2006). Whilst this is a progressive policy approach, urban livelihoods in South African cities are extremely precarious for their residents (Bakonyi et al., 2021). Urban precarity is exacerbated by weak State service delivery, limited social protection mechanisms and difficulty in accessing social protection for migrants who may lack the required legal documents (Amit & Kriger, 2014). As a result, the child protection system in South Africa relies heavily on civil society organisations to provide care for migrant youth and fill the gaps both by providing services on behalf of the State and by providing independent services. Given this context of precarious urban settlement for migrant youth, it is important to understand better the care interventions and practices that exist for unaccompanied migrant youth in South African urban centres.

The South African Children’s Act (38 of 2005) is globally recognised as a progressive piece of legislation that relies on broad understanding of care. As such, it goes significantly beyond defining care as the provision of basic material needs and includes children’s psychosocial, cultural and educational fulfilment. To implement this Act, the Department of Social Development (DSD) relies on a network of Child and Youth Care Centres (CYCCs), often run by NGOs, across the country, where young migrants in need of care are provided with housing and other support. These NGOs are social service practitioners registered with the Department of Social Development. They are not-for-profit organisations that provide young migrants with shelter, food, therapeutic services, education, and recreational programmes (RSA, 2020, p. 7).

Drawing on mixed methods research with migrant youth living in and accessing services from these NGOs this paper seeks to understand the nature of care provided to migrant youth in the Gauteng region of South Africa. We discuss who provides which aspects of care as defined in the Children’s Act. We argue that in the South African context, a unique, ground-up model of care has emerged through several contextual factors, including weak State service delivery, a relatively well-developed NGO sector, and fluid migration of young people that exists on a continuum from completely independent to completely accompanied. These factors shape the approach taken to care and raise important questions about how the needs of young migrants in institutional care can best be met. Similarly, the research revealed a model of care for migrant youth that shows a spontaneous and adaptive manner of “doing family”, with its fluidity, diversity, and multifacetedness (Morgan, 1996, p. 186). The continuum of institutional and family care challenges notions of migrant children and youth and their needs that are dominant in the academic literature.

1.1. Conceptualising Childhood and Care

It has become taken for granted in childhood studies to critique the excessive universalisation of a normative model of childcare developed in post-World-War-2 European contexts (Prout & James, 2015; Nussbaum, 2006; Burman, 2016). The pathologising effects of this universalisation and its colonial legacies have been thoroughly critiqued (Blume Oeur & Varga, 2026; Knight, 2019). In contrast, there has been a growing acceptance of childhood as socially constructed in different ways across time and place (Holloway & Valentine, 2000, p. 764; Chin & Phillips, 2004). Thus, childhood is constructed and reconstructed both for and by children (James & Prout, 1990; Prout & James, 2015). Far from being a natural, biological state, the child is a discourse of modernity rooted in relations of power (Burman, 2021).

Similarly, critiques of normative conceptions of families have challenged how we understand the family. Notably, there have been calls to acknowledge the diversity of family composition and the fluidity of family relationships, in which people are seen as constituting family in changing ways throughout their lifetimes (Williams, 2004). This contrasts with conceptualising family as static and singular in its form and function. To address these critiques, researchers have increasingly defined family through its practices rather than its composition (see Morgan, 1996; Finch, 2007; Ribbens et al., 2010). For example, Morgan (1996) focused on the interactions and meanings people give to family relationships, shifting analysis away from family as a structure that individuals belong to towards a set of activities that take on a particular meaning at a given point in time. Family would therefore be a facet of social life, not a social institution, that represents “a quality rather than a thing” (Morgan, 1996, p. 186). Morgan’s concept of family practices incorporates several key concepts that other scholars have used to analyse contemporary families, including fluidity, diversity, and multifacetedness. This then reorients our understanding towards doing family in everyday activities and routines, where individuals constitute certain actions and activities as family practices. Families are thus understood in relation to the socio-economic and cultural context, countering the singular concept of the child that is too often integrated into care arrangements and protected by international legal and development discourses (Abebe, 2019). Meanings attributed to and generated through interactions between family members are therefore constructed within a given socio-cultural and economic context rather than in relation to naturalistic reproductive and socialisation functions (Smart, 2004; Smart & Neale, 1999). Thus, analysing how people do family is at the core of understanding how people constitute and understand family (Ribbens et al., 2010). These reconceptualisations of family have enabled the appreciation and analysis of diverse family forms, seeing them as time- and context-appropriate rather than lacking in relation to an elusive yet normalised ideal. Examples of such bounded family functionality include extended families (Verhoef, 2005; Kassa, 2017), fostering and collective care arrangements (Abebe, 2019, p. 3), families constituted only of children (Phillips, 2007), non-cohabitating families (Ansell & Young, 2003; Verhoef, 2005; Dyer, 2007, Serpell & Adamson-Holley, 2017) and other family practices beyond the normalised nuclear family.

Regardless of its form, family is inherently conceptualised as a space where there is an expectation of care, even when this care is not provided in practice. Standing (2001, p. 17) defines care as the act of meeting a person’s “physical, psychological, emotional and developmental needs”. It thus embraces a range of human experiences and includes relationships of obligation, trust, loyalty and commitment concerned with the wellbeing of others (Graham, 1983). Yet, physical, psychological and emotional care are not separate but integrated and mutually constituting. For McGovern and Devine (2016) “affective practices do not operate autonomously but are strongly interconnected with access to economic, social and cultural resources and are deeply intertwined with the structuring of inequalities in the wider society” (p. 39). These can give rise to substantive differences in the “care worlds” of children (Lynch et al., 2009, p. 41). Thus, care is shaped strongly by structural conditions including social norms, migration status, State infrastructure, xenophobia, etc. Whilst the breadth of these debates is beyond the scope of this paper, five key critical questions stemming from them are relevant to understanding the care practices of migrant youth in South Africa. These are: What constitutes a good family for whom? What forms of familial care are excluded or pathologised through these notions of family? What constitutes care, and how can care be implemented in contexts of mobility? Who is ‘in need of care’? Finally, who should provide this care? By attending to these questions in the South African context, this paper provides much-needed detail about the forms that this socially constructed and reconstructed childhood takes outside the global North.

1.2. Care for Children and Youth in Contexts of Migration

Following the debates above, research on migrant children and families has challenged normative family models by questioning the assumption that a positive family structure has cohabiting, sedentary members (Baldassar et al., 2014). Despite this, much migration research has still treated transnational families as lacking, with particular concern for the (lack of) care for children and youth remaining behind when parents migrate (Yeoh & Lam, 2007). In contrast, however, in her studies of unaccompanied children in the US, Heidbrink (2014) argues that children who remain in the destination country when their parents migrate are nevertheless embedded in communities of care that often transcend national borders and are fluid. Undertaking perilous journeys for the sake of caring responsibilities, circulation within extended kin networks, and the provision of remittances to non-migrant family members questions the notion of children being “untethered to family” (Heidbrink, 2014, p. 137). Similarly, in their studies of Vietnamese families in Russia, Hoang et al. (2015) make a case for a notion of care that looks beyond the family to the macro-level socio-economic structures and processes that shape the practice and ethics of care. Experiences of Venezuelan children and youth documented by Vega (2016) show how children assume a family-provider role in active defiance of roles traditionally assigned to them. The research in this area reveals the benefits of analysing how care may shift to and from the child and the family, and across borders, within socio-economic structures and processes that shape the practice of care.

These studies complicate the notion of the unaccompanied child migrant, or the left-behind child, by making visible the fluid range of care practices that form a transnational family. The advancement of communication and social media has shaped how people care and what and how aspects of care are provided and by whom. For example, recent research has shown that families can provide care through technology that creates a sense of presence (Kim, 2017). However, this is not to romanticise migrant caregiving practices. Indeed, in many resource-poor contexts, this ability to create continuity in care, given the necessity of labour migration and the poorly developed ICS technologies in the region, is limited and causes significant distress to families who continue to value a model of care vastly different from their own (Takaindisa & Palmary, 2020). Southern African research shows how children often pay substantial emotional costs to meet their economic needs (Marchetti-Mercer, 2017) whilst their own caregivers provide for the emotional needs of others through what Hochschild (2003, p. 17) terms a “care drain”. For many children, their lived experiences mean that a child’s socially and culturally valued role cannot be taken for granted as they move between family positionings as both carers and care recipients (Hirschfeld, 2002). The focus in Southern Africa on care work provided by those who migrate for work (see Daly, 2001; Yeates, 2005; Sarti, 2008; Williams, 2012; Razavi & Staab, 2010) has led to an awareness that little is known about care for children and youth who move independently and what this means for care arrangements in the destination and origin country (for instance, Peng & Wong, 2016, Gheaus, 2013: Hoang et al., 2015; Poeze et al., 2017) as well as questions of separation and reunification of children in migrant families (see Suárez-Orozco et al., 2012; Zentgraf & Chinchilla, 2012).

This complex combination of separation and connection stands in stark contrast to some of the existing literature focusing on unaccompanied migrant children. Primarily triggered by the Syrian conflict in 2015, there has been a revival of research in the global North on unaccompanied and separated children and youth and the care systems they live in (see, for instance, Jensen, 2018; Derluyn, 2018; Migliorini et al., 2022). The literature on care and wellbeing from European and US contexts tends to assume migrant children and youth are far away from their family caregivers, unable to reconnect with them and live in contexts where the State provides (relatively) extensive care for such children and youth. In South Africa, porous borders and the fluidity of cross-border livelihoods mean that the presence of different caregivers will be closer, even if corruption and restrictive policies make the documents needed for migration more difficult to access (Klaaren & Ramji, 2001). In this context, civil society organisations step in to provide a complex set of care services that at times blur State and non-State functions.

2. Materials and Methods

This article is based on a mixed methods study, integrating qualitative and quantitative research methods. The choice of mixed methods provided a more generalised picture of how and where young migrants access care than has been available in the existing Southern African literature whilst the qualitative component ensured we did not lose sight of the complexity and meanings of care. In this paper, the focus was primarily on describing the patterns of care for young migrants in Gauteng, South Africa, and elaborating the implications for the better implementation of their care (see Perez-Sanchez, 2005 for more details on the overall study).

The research took an explanatory, sequential approach, following a nested sampling design in which the results from the quantitative phase were used to select the individuals or cases for the subsequent qualitative phase (Collins et al., 2007; Evertsson, 2017). Quantitative data was collected through a survey of young migrants. The NGOs that provide care for young migrants facilitated access to them. These NGOs included a bridging school with more than two hundred young migrants regularly attending classes, a Centre which hosts around 200 vulnerable youth and a network of shelters providing for migrant youth. Throughout the research, NGO staff were asked about other service providers to ensure a comprehensive list of providers had been generated. This paper focuses on describing the broad pattern of care that migrant youth access, how they access it and young migrants’ views on the quality of care provided and who provided care.

The sampling began by identifying all facilities providing care to young people in Gauteng province in South Africa. There are 92 facilities with the capacity to accommodate 5339 individuals, both migrants and South Africans. Based on estimates from IOM and Save the Children, we expected that migrant youth represented approximately 3.7% of children and youth in care in Gauteng (Scalabrini, 2019). As the estimated migrant youth population in Gauteng is 642,000 (UNDESA, 2017), an appropriate estimate of the population of migrant youth in institutional care out of the total migrant youth population in South Africa was calculated at 23,754 (3.7% of 642,000). For a population of 23,754 migrant youth in residential care, the corresponding sample size is 378 children and youth with a 95% confidence level and 5% margin of error, as a standard level in social science (Cramer & Howitt, 2004, pp. 32–36). Following the policies regarding provision care among the NGOs, we did not exclude those above the legal age of 18 but considered them part of the sample if they were still in care. Thus, the participants ranged from 6 to 21 years old. Reaching the desired sample size proved challenging as the research took place at the height of the COVID-19 lockdown and due to the high levels of mobility among the young migrants. Ultimately, 100 youth were surveyed from the NGOs in Gauteng.

The objective of the qualitative data collection was to build on the quantitative results to confirm and better explain some of the emerging findings. The first author administered four focus groups with 21 migrant children and youth using purposive sampling. The sample for the qualitative focus group was designed to ensure diversity in nationality, age and gender. The focus groups began with a game. After that the questions focused on how young migrants feel about the care they receive, their experiences of integration and belonging and their views on how laws, policies and NGO practices should change. These questions were chosen because of the results from the preliminary survey data analysis. Following on from the focus groups, semi-structured interviews were conducted with 10 stakeholders, purposively selected from NGOs that supported the children and youth. The questions focused on the nature of the services they provided to young migrants, the care needs that these migrants had, and the challenges and constraints they faced in their work. The focus groups and interviews were transcribed verbatim and, where necessary, translated as many participants switched across languages during the discussions. They were coded and analysed using thematic analysis (Braun & Clarke, 2006; Naeem et al., 2023).

The lower-than-desired quantitative sample size may increase the risk of type II error and places some limits on the extent of the statistical analysis that is possible. However, given the dearth of knowledge on access to care in African self-settled contexts, the survey provides an important exploratory description of the needs and experiences of migrant youth that can inform care practices. Triangulation with the qualitative findings helps with interpreting and confirming some of the survey findings. In this way it accomplishes the intended task of understanding how young migrants do family (Morgan, 1996) in hard-to-access and under-researched contexts.

Ethical approval was obtained from the University of Johannesburg, Faculty of Humanities, Research Ethics Committee (protocol number: REC-254-2020). All participants were given detailed information about the study and assents/consents were obtained from all participants (Alderson & Morrow, 2004). In addition to the assents, if participants had guardians residing with them in South Africa, we sought the parents’/guardians’ consents if the child was under 18 years old. Those over 18 years old gave informed consent to participate in the study. The identifying information of participants and their families was changed to ensure confidentiality. Furthermore, the names of the NGOs were pseudonymized for publication. All NGOs were given feedback on the findings of the study.

3. Results

3.1. Patterns of Care for Migrant Youth

In this section, we describe the care arrangements for young migrants. The table below shows the age, gender, length of time in South Africa, countries of origin and care arrangements for the youth surveyed.

All participants accessed care provided by NGOs as described in Table 1 below. A total of 36 participants lived full-time in institutional care. The remaining participants received support from the NGOs but also lived with their parents (n = 43), an adult caregiver different from their parents (n = 11), and in other arrangements, including family-based foster care or another NGO (n = 2). Among those living with their families, fathers tend to be largely absent with only 2% of the participants reporting living with their fathers at any time.

Table 1. Profile of the sample.

Age	Gender	Length of Time in South Africa (Years)	Country of Origin	Residence
Mean = 14.47 Min = 6 Max = 21 Range = 15	M = 56 F = 44	Mean = 6 Min = 1 Max = 21 Range = 20	Mozambique = 30 Zimbabwe = 16 African Great Lakes region = 47 Other = 7	NGO-sponsored shelter = 36 With Parents = 43 With other relatives = 11 Family-based foster care = 2 No fixed living place = 1 Others = 7

These findings show that although the young migrants live in fluid arrangements, 90% of them had a relative or next-of-kin in South Africa. A total of 57% were in contact with their mothers, while the majority were also in touch with their brothers (46%) or sisters (36%). Most youth were still in contact with those relatives that travelled with them to South Africa, and most managed to remain united with their siblings, with 85% of the youth reporting living with their siblings either in institutional care or, at times, with their parents.

3.2. Who Provides Which Needs for Young Migrants?

Obtaining information about assistance can be challenging for migrant youth (Migliorini et al., 2022). The participants were asked how they typically found out about services available to them. Respondents tended to obtain information about services from family members (50%), NGO staff (18%) and other migrant youth (9%) followed by social media (6%). This indicates the importance of social networks in accessing care, even for youth who are unaccompanied. State agencies such as the Department of Social Development are notably absent as a source of information for unaccompanied youth.

In keeping with the broad definition of care described above, youth were asked whether their emotional and physical needs were met, who provided this care, what they thought about the quality of care that they were provided with and how much input they had in decisions regarding their care.

Access to material needs was measured using a series of questions shown in Figure 1 below. The participants were asked to respond on a scale from strongly agree to strongly disagree to the statement: “I have someone I could ask for…”. Overall, migrant youth reported having someone who can assist with most of their basic material needs.

Figure 1. Responses to the question “I have someone I can ask if I ever need” for various forms of support.

What is most noteworthy however is that the availability of material care decreased regarding services that the State has a monopoly on such as provision of migration documents, access to asylum, and social services (which would include State-provided social grants such as the childcare grant). Indeed, 53% of young migrants reported never having access to social services, 56% never accessing assistance with legal documentation, and 69% never accessing the asylum system.

Young migrants’ feelings of being unable to access State services due to their documentation is summarised by the following quote by Emmanuel:

We receive basic support at the centre, the school, and the clinic, but things should improve for us; we often feel that the support depends on our documents or our status, and this is very tiring.

(Emmanuel, DRC, 13 years old)

Similarly, Melany, from the DRC, expressed her feelings about the asylum system in South Africa in the following statement:

It is just not normal: now I have been six years waiting for my asylum interview; who can wait that long? I do not understand why it is taking this long, and in the meantime, I do not have my documents; I feel hopeless.

(Melany, 12 years old)

The impact of the prolonged periods of waiting, which creates a limbo status amidst asylum-seeker youth, was evident in the focus group discussions. Young migrants seeking asylum felt powerless and hopeless about their inability to regularise their status and obtain documentation and they were able to connect it to an inability to obtain other services.

In contrast, youth had better access to material resources which could be provided by non-State actors such as medical services, food, school uniforms and shelter. In the South African context there is a network of private and NGO-run healthcare and education facilities that young people can access (albeit at a cost) which likely explains the relatively good access participants had to these services. This suggests that despite youth being ‘in need of care’ as defined in the legislation, they rely heavily on non-State actors to assess their material needs.

The results of Mann–Whitney and Kruskall–Wallis tests showed no statistically significant differences based on gender or country of origin in their response to any of these questions. However, Spearman’s rho correlation, as represented in Table 2, showed a statistically significant relationship between age and young people having someone they could ask for food (p = 0.05; correlation coefficient = 0.281) and someone they could ask for help with legal documentation (p = 0.048; correlation coefficient = 0.201). As described above, family and other social networks were the main ways that young migrants accessed information about how to meet their needs. This may suggest increasing social capital as they get older, which helps them to navigate often opaque and challenging State bureaucracy.

Table 2. Correlation between age and support for material needs. ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

			Age	I Ever Need Food	I Need to Borrow R100	I Ever Need School Uniforms	I Ever Need Medical Assistance	I Need a Place to Stay	I Need Assistance with Legal Documentation	I Need Assistance with Asking for Asylum	I Need Access to Social Services
Spearman’s rho	Age	Correlation Coefficient	1.000	0.281 **	0.029	0.089	−0.019	0.074	0.201 *	0.180	0.020

Beyond material needs, most young migrants also reported having supportive people in their lives who met their emotional needs (see Figure 2 below). Around 80% of them said they always have someone who would listen to them if they had a problem with, and 86% reported that they have someone who shows them affection. Similarly, 79% had someone they could trust to confide a secret to, while 82% had someone they could seek advice from, and 90% had someone they could play with. Although there is clearly a small group of young migrants who are extremely isolated with few emotional support structures, the majority felt their emotional needs were met.

Figure 2. Responses to the question “I have someone…” for various forms of emotional support.

When tested across age (Spearman’s rho correlation), gender (Mann–Whitney test) and country of origin (Kruskal–Wallis test) there were no statistically significant differences found in young migrants’ access to this emotional support. Overall, the lack of differences by age, gender, and country of origin paints a positive picture of equitable access to care regardless of age, gender and nationality.

3.3. Young Migrants’ Perspectives on Who Provides Care/Services

A section of the questionnaire was related to young migrants’ perspectives on who provides the assistance that they received, including staff from NGOs, State social workers, various adults, and other young migrants. Regarding their material needs, the young migrants identified staff from the NGOs and family members as the ones who typically provided for these needs. They relied on family members to provide emergency cash (79%), food (67%) and accommodation (67%) with NGOs assisting with school uniforms (70%), legal support (62%) and social services (60%). The graph in Figure 3 below shows the distribution of responses.

Figure 3. The main providers of material needs.

Figure 3 shows that a combination of NGOs and relatives provide for young migrants’ material needs. Notably, even for those services that are entirely State functions, such as providing access to asylum and social services, young migrants still relied on NGOs. Reading alongside their inability to obtain these services, as described above, this places NGO staff in a difficult situation whereby young migrants expect them to provide services that they have little or no legal mandate to.

Similarly, the young migrants were asked who provides for their emotional needs (Figure 4). Notably, those people who cover their emotional needs are first and foremost their mothers. In total, 40% of young migrants identified their mother as the person who would talk about their problems, while 17% identified their mother as the person to seek advice from, and 36% identified their mother as the person who shows them affection. Furthermore, 48% of young migrants see mothers as the most important person that shows them affection. Other people who provided emotional needs included other relatives and fellow foreign youth, who were particularly relevant as playmates (47%), as people to trust to confide secrets to (36%), and as people to seek advice from (20%).

Figure 4. The main providers of emotional needs.

In contrast, staff from the NGOs perform a much more negligible role in providing for young migrants’ emotional needs. For instance, only 10.1% of the participants reported replying to NGO staff when they want someone to listen to a problem they faced, 11.5% when they need someone to talk to, 8.2% as someone who would show them affection, 2.3% as someone they can play with, 5.8% as someone to confide a secret and 9.2% as someone to seek advice from. Thus, there was a stark division in care provision with relatives and friends playing an important emotional role whilst NGO staff predominantly provided for their material needs.

3.4. The Quality of Care That Young Migrants Receive

The young migrants were asked to rate how helpful the support that they received was. Participants were most likely to receive support for education (n = 71), shelter assistance (n = 66) and health (n = 56), and these forms of support received fairly good ratings from young migrants. Significantly lower numbers of young migrants received support for protection against xenophobia (n = 10), guidance to seek asylum (n = 16) and psychological assistance (n = 20), and these sources of support received poor ratings for the level of helpfulness. Figure 5 below shows the young migrants’ perceptions of the quality of the assistance received.

Figure 5. Perceptions of the helpfulness of different forms of support received.

Overall, young migrants rated the assistance received for documentation and legal counselling negatively, with 73% and 66.7%, respectively, rating them as completely unhelpful. They felt more positive about the educational assistance received with 77.5% rating it as moderately or very helpful and shelter assistance with 77% rating it as moderately or very helpful. The assistance received for recreational activities was also rated as moderately or very helpful by 71% of the young migrants. They had polarised opinions on the protection they received against xenophobia2 with the same proportion of respondents rating it as helpful and unhelpful, respectively.

3.5. Agency and Young Migrants’ Role in Decision-Making

When questioned about their capacity to make decisions regarding their care, young migrants expressed minimal involvement in decision-making. Only 8% and 19% respectively reported being able to always or often take decisions regarding their care (see Figure 6).

Figure 6. I am able to take my own decisions regarding my care.

Understandably, being able to take their own decisions was significantly correlated with age as seen in Table 3 below:

Table 3. Correlation between age and being involved in decisions about their care. ** Correlation is significant at the 0.01 level (2-tailed).

			Age	I Am Able to Take Decisions Regarding My Care
Spearman’s rho	Age	Correlation Coefficient	1.000	0.294 **
		Sig. (2-tailed)	.	0.003

Similarly, as indicated in Table 4 below, young migrants from Mozambique were significantly more likely to feel they could take decisions about their care than those from the African Great Lakes region. There was no statistically significant gender difference (based on Mann–Whitney test) in their belief that they could take decisions about their care.

Table 4. Relationship between nationality and ability to take their own decisions about care. ^a Significance values have been adjusted by the Bonferroni correction for multiple tests.

Sample 1–Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig. a
Other–Great Lakes region	7.906	11.175	0.707	0.479	1.000
Other–Zimbabwe	10.027	12.500	0.802	0.422	1.000
Other–Mozambique	26.314	11.578	2.273	0.023	0.138
Great Lakes region–Zimbabwe	2.121	7.984	0.266	0.790	1.000
Great Lakes region–Mozambique	18.409	6.446	2.856	0.004	0.026
Zimbabwe–Mozambique	16.288	8.539	1.907	0.056	0.339

Each row tests the null hypothesis that the Sample 1 and Sample 2 distributions are the same. Asymptotic significances (2-sided tests) are displayed. The significance level is 0.050.

Linked to this finding, when asked whether they feel that NGO staff listen to them regarding different aspects of their care, most respondents said they did not, as reflected in Figure 7 below. Generally, 50% or more of the respondents reported not being listened to at all, or only to a small extent, regarding their care arrangements. Most notable is that they felt their views were particularly excluded from decisions regarding recreational activities with 62% saying they were listened to “to no extent or a limited extent”. Similarly, regarding housing arrangements, 58% of the young migrants indicated that they were listened to “to no extent or a limited extent”.

Figure 7. Responses to the question “I feel listened to by NGO service providers” for various support services.

This finding resonates with impressions during the focus groups. John, from the DRC, shared the following reflection:

I feel like being ordered to do many things here at the centre; I would like people working with us to also ask me questions about how I feel things could be improved at the centre.

(John, 13 years old, DRC)

This finding is particularly surprising as it runs contrary to the ideals of youth as active participants in their care that many of the NGOs embraced. As one of the key informants noted:

With these children, we need to work with a focus on a child-centred approach, and it is difficult for adults to fully follow this, but you need to give young people the power to make decisions that affect them and enhance agency, using a child-centred approach.

(KII10)

The value placed on young migrants’ participation and the need to support their agency was frequently referenced by the NGO staff. And yet this did not translate into a sense by young migrants that they had the power to take decisions or that they were listened to.

To understand this finding further, a factor analysis was used to combine the questions in Figure 7 into an item measuring young migrants’ perceptions of their role in decision-making about their care. All items in the factor analysis had correlations above 0.8 indicating sufficient correlation between items to proceed with factor analysis. The Cronbach’s alpha coefficient was 0.945 indicating that these items formed a highly reliable scale.

Kruskal–Wallis tests between the factor “being listened to” and country of origin (Table 5) show that young migrants from Mozambique were statistically more likely to feel listened to than those from the African Great Lakes region. However, gender and age were not significantly associated with feeling listened to.

Table 5. Relationship between country of origin and feeling listened to. ^a Significance values have been adjusted by the Bonferroni correction for multiple tests.

Sample 1–Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig. a
Great Lakes region–Other	−11.869	11.626	−1.021	0.307	1.000
Great Lakes region–Zimbabwe	15.735	8.306	1.894	0.058	0.349
Great Lakes region–Mozambique	34.831	6.706	5.194	<0.001	0.000
Other–Zimbabwe	3.866	13.005	0.297	0.766	1.000
Other–Mozambique	22.962	12.046	1.906	0.057	0.340
Zimbabwe–Mozambique	19.096	8.884	2.150	0.032	0.190

Each row tests the null hypothesis that the Sample 1 and Sample 2 distributions are the same. Asymptotic significances (2-sided tests) are displayed. The significance level is 0.050.

Given the importance associated with young people’s agency in the literature described above, we investigated the impact of not feeling listened to or being able to take their own decisions further. Correlations were performed between this scale and three other questions in the questionnaire. These questions were: How is life in South Africa compared with how you imagined it would be on a scale from ‘much worse’ to ‘much better’? Overall, how happy are you with the level of support available to migrant children in South Africa on a scale from ‘very happy’ to ‘very unhappy’? The final question was if they have high hopes for their future in South Africa on a scale of ‘strongly agree’ to ‘strongly disagree’.

Table 6 below shows the results of Spearman’s rho test to assess the correlations.

Table 6. Correlation between feeling listened to and being able to take their own decisions with optimism for their future, happiness with care and life in South Africa relative to their expectations. *** indicates significance at the 0.001 level. ** indicates significance at the 0.01 level.

	Optimism for Their Future in South Africa	Happiness with Care in South Africa	Life in South Africa Compared to Their Expectations
Degree to which I feel listened to	0.312 **	0.646 **	0.393 **
I am able to take decisions about my care	0.012	0.338 ***	0.293 **

Table 6 above shows that feeling listened to was significantly and positively associated with young migrants’ optimism for their future, their happiness with levels of care in South Africa and their assessment of life in South Africa in relation to their expectations. Similarly, being able to take their own decisions was significantly and positively associated with their happiness with levels of care in South Africa and their assessment of life in South Africa in relation to their expectations. The feelings of agency young migrants have therefore had a strong impact on their overall happiness and wellbeing in South Africa. This has important implications for young migrants’ adjustment and integration in South Africa given their difficult migration conditions and the well documented challenges of integration after migration.

4. Discussion

The above results show how young migrants access care in a context of self-settlement in precarious urban centres. One of the most striking lessons from this data is that the young migrants could not be neatly defined as accompanied or unaccompanied. This is contrary to the assumptions in policy and many academic literature studies. Given the circular nature of migration in South Africa, young migrants’ accompanied status was fluid, challenging existing research that assumes a clear division between accompanied and unaccompanied. This has implications for the form that young migrants’ care takes as they move in and out of institutional care. Even those young migrants who lived entirely at the shelters had social networks and family members who continued to play an active role in their care. Young migrants thus met their needs through a combination of carers, both familial and institutional. Documenting these forms of care allows us to respond to the call to focus on family as a set of practices rather than a social institution and documents how family care is enacted through the everyday work of doing family (Morgan, 1996).

What emerged from the questions regarding provision of material and emotions needs of young migrants is a complex picture where material needs are provided by a combination of relatives and NGOs, but emotional needs are provided almost entirely by family, often mothers. Although they live across NGO-run shelters and family care, young migrants mostly managed to meet their emotional and material needs through this ‘ground-up’ system of care. This suggests a contextually responsive environment where NGOs and families work together to provide care. In keeping with the literature above, a focus on doing family rather than family composition brings into focus the complex everyday practices of care that can mean young people are provided for even if they appear to be separated from family at first glance. This extends the findings of authors such as Vega (2016), Kim (2017) and Takaindisa and Palmary (2020) by showing how the multifaceted nature of care is not just a phenomenon of transnational families but equally evident for young migrants.

Importantly, the young migrants in our study also enact care. They played a central role in providing companionship for each other in a context where there are widespread exclusion and violence against foreigners (Misago & Landau, 2023). Similarly, young migrants shared secrets and addressed problems together. This is an often underacknowledged form of agency that shows how young migrants create supportive environments despite instability in their living arrangements. Whilst the risks of parentification have been well documented in the literature, this study showed that care was enacted in ways that complemented family and NGO care (Bergset et al., 2025).

What is notable from the results is that the State is almost entirely absent from the lives of young migrants. The State was not seen as a source of information about services for young migrants. The services that the State has a monopoly on, such as provision of access to asylum and legal documents, are the ones that the young migrants were least able to access and rated most poorly. Despite having a progressive Children’s Act and despite the self-settlement policy in South Africa, the reality was one where the State did not meet its legal mandate to provide for young migrants in need of care. The observation of young migrants lacking basic documentation shows a State that is struggling to enact the rights-based approach imagined in law. This is in contrast to many literature studies from the global North which assumes that unaccompanied children are cared for almost exclusively by the State (Wade et al., 2012; Daly, 2001). This paper thus shows how young migrants survive in contexts where State provision is weak and family systems are fluid.

This relatively positive picture of complementary care practices among families and NGOs alongside weak State provision comes with considerable risks. Firstly, the absence of the State in young migrants’ lives resulted in many of them believing that NGOs were responsible for all provision, even those services that the State had the sole mandate to provide. Thus, the fact that the State was so absent in the lives of young migrants meant they tended to expect NGOs to provide services that they simply could not provide. This risks increasing tension between young migrants and NGOs, because they believe the NGOs are failing in these areas. In addition, NGOs often operate with limited funding and are themselves sustaining this very precarious model, creating risks for the sustainability of this model of care.

Throughout the research, young migrants and NGO service providers emphasised the importance of young migrants having a say in their care. Many NGOs saw this as fundamental to their philosophy and believed that they were exceptional in this regard, relative to how other adults treat young migrants. Despite this underlying value, the young migrants in this study did not feel listened to by service providers and did not feel they could take decisions about their care. This sense of not having agency had significant connections to their sense of wellbeing and adaptation to the context of migration. The more young migrants felt listened to, the more likely they were to feel optimistic about their future, happy about the care they receive and that life in South Africa was better than their expectations. Similarly, young migrants who could take their own decisions about their care were more likely to feel happier about the care they received and feel that life in South Africa was better than they had expected. This suggests that inclusion in decisions about their care makes a significant difference to young migrants’ sense of wellbeing and happiness in contexts of migration. The existing literature has focused on children’s and youth’s participation and agency as a moral obligation (Abebe, 2019) and yet this study shows the practical costs of failing to take participation seriously. It also shows how challenging meaningful agency is in practice as NGO workers were committed to the ideal but were not, in the participants’ view, implementing this value. The difference among national groups in their sense of whether they could take their own decisions and whether they felt listened to may well reflect social and cultural differences in how much young people expect to have a say in their care. This further shows the complexity of implementing the principle of agency in practice as adequate agency for one national group may not be seen as adequate by another.

5. Conclusions

This article sought to understand how young migrants’ care takes place in the context of precarious urban self-settlement. It describes who provides what kind of care and their perceptions of this care. In doing so this paper contributes towards a model of care that recognises and values the multifaceted forms of care that take place in migration settings that are precarious due to weak State implementation, social and economic exclusion and self-settlement. Recognising that material and emotional needs of young migrants were met by a combination of NGOs and their families suggests the importance of these two systems working together. This is further reinforced by the fact that although technically these are young people in need of care, family networks remain important in their lives and they continue to participate in the emotional life of the family. Supporting NGOs and families as partners in ‘doing family’ is central to quality care. Similarly, the emphasis young migrants placed on being listened to and the impact it had on their wellbeing is a reminder that this needs to be better integrated into care. The patterns of care observed are reciprocal and interdependent, with young migrants displaying agency and playing an active role in enacting care to their parents and peers. The absence of the State in the provision of young migrants’ care, despite the enabling policy environment, is a significant risk to young migrants because it means many of them cannot access documentation. This is an intractable risk that has been documented in many other studies in South Africa.

What emerged from this study is a model of care for young migrants that shows a spontaneous and adaptative manner of ‘doing family’ that can be enhanced and further supported by the State rather than surpassed. Families are responding to the surrounding conditions left by an absent State by relying largely on NGO services for the provision of material needs, playing a critical role in the provision of emotional needs, and building upon the agency and resilience of young migrants themselves, who play a meaningful role in this model of care. By supporting what is a largely positive spontaneous form of care, and addressing some of its weaknesses, young migrants can have their care needs met even in contexts where there is limited formal State provision and socio-economic precarity for young migrants. This means that migrant families need to be recognised as important decision-makers in young migrants’ care even when they meet the legal definition of being unaccompanied. This would require a shift from the State away from punitive approaches to migration towards the implementation of the child protection system that currently is provided for by the law but poorly enacted. Most significantly, the State does not need to create entirely new systems of service delivery to meet their legal obligations but instead can draw from the strengths of a ground-up model of care that has emerged from the precarity of life in Gauteng and the relatively strong NGO sector.

The delivery of care by NGOs needs support to better implement the existing philosophies of youth participation. The extent to which agency differed by nationality suggests that youth’s sense of agency is shaped by social norms and culturally contextual factors that influence their expected involvement in their care. This means that implementing participation is complex and a constantly negotiated terrain that requires ongoing child engagement. Recognising this as a process that is constantly underway can help NGO staff to implement child-friendly platforms such as children’s and youth’s desks or forums to support their participation.