by Poorvi Yerappureddy, Antara Madavane
South Africa’s Refugees Act guarantees the right to work, study and access basic healthcare backed by a constitutional framework and its position as a regional economic hub. This has made its major cities a key destination for people fleeing conflict and political instability from countries such as the Democratic Republic of the Congo (DRC), Somalia, South Sudan, Burundi, Ethiopia and Zimbabwe.
Within this setting, our case study follows how digital identity and data infrastructures organise everyday life for refugees and asylum seekers—from the Department of Home Affairs’ (DHA) asylum and civil‑registration systems, through banking and SIM registration, to community‑run WhatsApp groups and other forms of resilience. The central problem is not that refugees lack documents or devices, but that the systems they enter, treat their documentation as permanently conditional and their presence as inherently risky.
The downstream harm is a form of digital limbo, where people on the move must constantly re‑perform their legitimacy to keep access to banking, connectivity and basic protection. Across the report, we show how data flows centralised around DHA make refugees highly legible to enforcement and compliance systems, while avenues for correction or redress remain narrow.
South Africa’s position as one of the most industrialised economies in the continent and a middle-income democracy has made it a longstanding magnet for migrants from the Southern African region and wider sub-Saharan Africa, especially into Gauteng and the Western Cape. Armed conflict and political instability in countries such as the DRC, Somalia, Burundi and Ethiopia continue to drive displacement, alongside persecution of political or ethnic minorities and collapse of basic services. This movement accelerated after the end of apartheid, when a new constitutional order and regionally strong economy made the country a key destination for people fleeing conflict and economic collapse.
Labour migration dominates in the country, but another populace arrives seeking asylum from conflict, repression and economic collapse, creating a layered mobility regime in which work seekers and refugees occupy the same urban spaces under very different legal categories, which the policy environment does not always distinguish. This political economy is characterised by extreme inequality, persistent unemployment and a sizeable informal sector, conditions that position people on the move within low-wage labour markets and expose them to tension when competition over jobs, housing and services intensifies.
South Africa’s position as one of the most industrialised economies in the continent and a middle-income democracy has made it a longstanding magnet for migrants from the Southern African region and wider sub-Saharan Africa, especially into Gauteng and the Western Cape. Armed conflict and political instability in countries such as the DRC, Somalia, Burundi and Ethiopia continue to drive displacement, alongside persecution of political or ethnic minorities and collapse of basic services. This movement accelerated after the end of apartheid, when a new constitutional order and regionally strong economy made the country a key destination for people fleeing conflict and economic collapse.
Labour migration dominates in the country, but another populace arrives seeking asylum from conflict, repression and economic collapse, creating a layered mobility regime in which work seekers and refugees occupy the same urban spaces under very different legal categories, which the policy environment does not always distinguish. This political economy is characterised by extreme inequality, persistent unemployment and a sizeable informal sector, conditions that position people on the move within low-wage labour markets and expose them to tension when competition over jobs, housing and services intensifies.
Development of Refugee Policy
The formal asylum regime emerged in the late 1990s as part of South Africa’s post-apartheid constitutional guarantees, which extended socio-economic rights to everyone, including non-nationals. The post-1994 migration regime was built around two distinct, sometimes conflicting, legislations. The first is the Refugees Act of 1998, which outlined an urban, rights-based asylum system that rejects encampment, locates refugees in ordinary neighbourhoods and, on paper, grants them the right to work, study and move freely. In parallel, the Immigration Act of 2002 and its subsequent amendments sought to manage labour migration and irregular entry, emphasising employer sanctions, carrier liability, and border control.
In the late 2000s and early 2010s, South Africa became one of the world’s busiest asylum destinations, creating large backlogs of refugee and asylum applications and prompting a policy shift toward strategic management of international movement into the country. Over time, successive White Papers and amendment acts have reframed asylum as a vector of abuse within this broader migration system, legitimising more restrictive rules on entry, status cessation, and camp-like ‘asylum processing centres’ near borders. These shifts have not been matched by investments in administrative capacity, producing a regime that is formally rights-affirming yet practically deterrent.
Refugees Act 130 of 1998 (as amended)
Immigration Act 13 of 2002 (and amendments)
White Paper on International Migration 2017 and subsequent draft policy…
National Health Act and 2007 migrant health directive
Cybercrimes Act 19 of 2020
Narratives around foreigners
Public perception has evolved in tandem with these institutional changes. Early democratic discourse in the 1990s leaned on pan-African solidarity and South Africa’s obligation to provide refuge after decades in which neighbouring African states hosted South African anti-apartheid activists in exile. From the mid-2000s onward, however, political rhetoric increasingly cast “illegal foreigners” as drivers of crime and service collapse, a framing that crystallised in large-scale xenophobic violence in 2008, renewed attacks in 2015, and recurrent localised episodes since.
Official responses have combined condemnation with denialism and short-term policing operations, rather than sustained prevention or accountability for perpetrators. For refugees and asylum seekers, this means that the risk of attack or harassment is not an exceptional event but a background condition shaping decisions about where to live, work and send children to school. The 2023 Migration Profile report explicitly acknowledges migrants’ contributions to employment, remittances and skills transfer, and calls for data-informed policy to align migration management with development goals. Yet gaps between these commitments and implementation are persistent.
1948–1994
Apartheid-era influx control and regional migrant-labour system restricted black South Africans and foreign Africans through laws and contract labour schemes
1994
Democratic transition and adoption of a rights-based Constitution extending socio-economic rights to everyone in South Africa
1998
Refugees Act passed, domesticating international refugee law and establishing an urban asylum system
2002
Immigration Act replaces earlier Aliens Control Act, restructuring labour migration and enforcement
2008
Large-scale xenophobic attacks across townships and informal settlements, killing dozens and displacing thousands
2015
Renewed nationwide xenophobic violence targeting African migrants and refugees in major cities
2017–2020
Refugees Amendment Act and new regulations introduce more restrictive provisions on status cessation, exclusion and asylum procedures
2020
COVID-19 lockdown leads to closure of Refugee Reception Offices and introduction of an online asylum and refugee permit-renewal system
2023
Stats SA publishes its first Migration Profile report, signalling a more data-driven approach to migration
2026
Cabinet announces a new immigration policy direction emphasising digitised migration and asylum management and expanded use of biometric and integrated data systems
Failures of the System
Against this backdrop, the core systems that refugees must navigate reveal persistent implementation gaps and governance failures. In the asylum system, numerous reports document entrenched corruption at Refugee Reception Offices, with bribes reportedly demanded for entry to queues, file retrieval and permit renewal and long-running case backlogs leaving asylum seekers in prolonged precarity.
Constitutional guarantees and a 2007 directive on health care entitle all residents to primary care and HIV/TB treatment. Yet studies describe “medical xenophobia”, inconsistent application of fee schedules, and confusion among staff about treating or turning away undocumented patients.
Financial access shows a similar pattern with court decisions and FIC guidance requiring banks to accept asylum and refugee permits for KYC, but branch-level refusals and account closures remain common, pushing these communities towards cash economies or intermediated access through friends’ accounts and informal remittance channels. These gaps are central features of how the migration regime allocates risk and uncertainty.
Why it matters now
Digitalisation is now being layered onto these pre-existing dynamics. The Department of Home Affairs has embarked on a multi-year “modernisation” programme centred on biometric registration, integrated civil and migration records and, since 2020, an online permit-renewal system for asylum seekers and refugees.
At the same time, South Africa has experienced major data breaches in government and private organisations. Enforcement of the Protection of Personal Information Act and the Cybercrimes Act remains inconsistent. This results in an environment where new systems often expand data collection quicker than they strengthen governance. For refugees and asylum seekers, becoming visible to the state increasingly entails inclusion in databases whose security, accountability and responsiveness were not designed with their vulnerabilities at the centre.
Together, these elements form the terrain for the report’s digital and data analysis. Scarcity and inequality shape how migrants and refugees are discussed and governed. The formal legal regime that sets broad rights is implemented through under-resourced institutions with documented corruption risks. Public narratives range from solidarity to hostility, and episodes of xenophobic violence influence which policy options come onto the agenda. Digital infrastructures introduce new tools for managing migration and service access and operate within systems where governance deficits are already well described. The subsequent sections trace how data architectures, flows and governance arrangements within this context shape the everyday experiences of refugees and asylum seekers from across the region.
Population Profile
Fieldwork for this case study involves 50 participants between the ages of 14 and 65, predominantly women, originating from the DRC, Kenya, Ethiopia, Somalia and Zimbabwe. Participants include asylum seekers, recognised refugees and people awaiting decisions on their status, reflecting a range of documentation situations but not including South African citizens, longstanding permanent residents or internal migrants.
Across these groups, participants describe living in multi-lingual, multi-status households where members hold different permits or none at all. The need to navigate South African cities through documentation regimes and digital systems that do not recognise them as primary users, is an experience shared by participants from various origin countries in strikingly similar terms.
Journey map: navigating life into and through South Africa
Journeys after arrival are rarely a one-time passage to safety, they are a series of negotiations with specific institutions. Many first filed their asylum claims at border-proximate Refugee Reception Offices and then moved to Gauteng or the Western Cape in search of work, schooling and community, while their files remain anchored to the original office.
Journeys after arrival are rarely a one-time passage to safety, they are a series of negotiations with specific institutions. Many first filed their asylum claims at border-proximate Refugee Reception Offices and then moved to Gauteng or the Western Cape in search of work, schooling and community, while their files remain anchored to the original office.
Renewals then required expensive travel back to that office, sometimes across provinces, with people sleeping outside reception centres because they could not afford accommodation while waiting in queues. Participants report files being misplaced when transferred between offices. Often, rejections arrive with little explanation when officers mistype names or file numbers. So legal status feels contingent on clerical accuracy and available cash.
Language barriers turned these logistical hurdles into substantive injustices. Application forms and written notices at reception offices are typically available only in English. Participants described interviews conducted without professional interpreters, relying instead on ad-hoc translation by other asylum seekers or community volunteers. For people with limited English or literacy skills, this means signing documents they cannot fully read and struggling to narrate complex persecution histories in a second or third language. When an officer mistranscribes a name, date or account of events, those inconsistencies later appear as grounds for rejection, even though applicants never receive a linguistically fair chance to present their claims.
Queer refugees faced additional hurdles at the same point in the process. South African law recognises persecution based on sexual orientation or gender identity as grounds for asylum, yet queer participants who fled homophobic violence described interviews in which officials asked them to ‘prove’ their sexuality through intrusive and discriminatory questioning. Some officers accepted community letters or general accounts of criminalisation in countries of origin while others demanded specific incidents or documentation that was impossible to obtain. People with similar histories report very different outcomes depending on which reception officer they encounter first. Thus, for queer applicants, the office that hears the claim often matters as much as the substance of their claim.
Patterns also diverged by nationality. Participants note that applicants from countries with widely recognised conflicts, such as the DRC or Somalia, more often have their need for protection understood, while those from countries where persecution is localised, politicised or under-reported internationally, face higher scepticism and heavier evidentiary demands. Some eventually receive recognition while others do not. The same story may be treated as self-evident in one case and implausible in another depending on how closely it matches officials’ dominant image of a ‘real’ refugee. These dynamics recurred across origin-country groups in our conversations.
Interactions with digital systems
Digitalisation layers email inboxes, portals and device requirements onto this already fragile system. The email-based renewal system introduced during the COVID-19 period is open to anyone with a permit, yet assumes stable access to smartphones, data, scanners and English language interfaces. Each renewal depends on an informal chain of intermediaries whose incentives and capabilities vary. Older members of the community and women with limited digital literacy described relying on internet cafés or younger relatives to handle the process, from composing emails to attaching scans and monitoring responses.
When a message goes unanswered or an attachment is rejected, applicants struggle to know whether the problem lies in the intermediary’s handling, their own data or technical errors, because feedback from the system is sparse and inconsistent. The same infrastructure that promises remote access to the state also disperses responsibility across actors who never meet. Participants from different origin countries describe this aspect as a source of chronic uncertainty.
Movement between cities is shaped by these administrative and digital constraints. Some participants relocate closer to reception offices or to areas where community networks provide accompaniment and translation for renewals and hearings, while others remain in places with better work prospects but weaker institutional access and accept longer periods of irregularity between permit expiries and renewals.
People describe the decision to engage with or withdraw from the asylum system as based on the perceived benefits of renewed status outweighing the financial, emotional and practical costs of each interaction with the state.
Accessing services through precarious documentation
Health facilities, schools, banks and welfare offices formally recognise certain permits, yet their staff apply these rules unevenly and often work with partial or outdated information about the validity of documents. Participants describe presenting the same permit at different clinics, schools or branches and receiving conflicting answers, illustrating how information asymmetries between central policies and front-line staff shape access. Clinic staff sometimes deny treatment to refugees whose permits have expired or contain spelling errors, despite regulations that guarantee primary care in emergencies regardless of status.
Children are turned away from schools until an advocate or community worker intervenes with circulars clarifying that documentation should not be a barrier. In financial services, particular banks and branches develop reputations within communities for being more willing to work with refugee documents, while others routinely refuse to open accounts or block access to them once a permit approaches expiry.
Mobile-money and wallet ecosystems add options but also new complexities. Somali, Congolese and other communities favour different apps and operators for remittances and local payments, usually chosen because someone in the network has successfully used them rather than because terms are transparent. Where formal accounts are unavailable, people use employer or friend accounts or keep most income in cash. This makes participation in digital transactions dependent on others’ goodwill and increases vulnerability to theft respectively.
In effect, access to almost every service from health care, education or banking depends on the current status of one’s papers. Also, whether these papers are recognised as sufficient by an intermediary at the counter or in the community. This pattern was confirmed by participants across origin-country groups in our fieldwork.
Community networks as infrastructure
Within this landscape, community networks operate more as essential infrastructure than as background support. Origin-country associations, religious congregations and neighbourhood committees are often the first point of contact for new arrivals, helping with immediate needs such as finding a bed, explaining transport routes and sharing basic information about legal status. Participants from different countries shared similar experiences when describing how they first orient themselves in South African cities.
Religious institutions play a particular bridging role. Congregations collect small contributions to cover the cost of transport for renewals, accompany people to offices as informal interpreters and mediate disputes with landlords or employers. These networks frequently determine whether people are able to approach state institutions at all, even though statutory systems formally remain in place.
Digitally-mediated networks extend and densify this support. WhatsApp groups managed by community leaders circulate updates about reception offices, changes in renewal procedures, new NGO programmes and alerts about xenophobic protests in specific areas. Facebook pages act as noticeboards: where landlords known to accept refugee documentation advertise rooms, where community organisations announce workshops on online renewals, and where missing persons or threats can be flagged. The boundary between social networks and critical infrastructure often blurs as communities fill the vacuum created by institutions that design and run these systems.
ResTech in this corridor rarely appears as a platform built specifically for refugees. Instead, communities use appropriate general-purpose tools such as WhatsApp, Facebook, Google Drive and mobile wallets to meet needs overlooked by formal systems. Participants’ descriptions of these adaptations highlight both the ingenuity and the burden carried by communities who must compensate for systems that do not recognise them as primary users.
Community-based Organisations (CBOs) sometimes formalise these practices into digital clinics or hotlines where volunteers assist with drafting emails to the DHA, completing online forms or checking bank and wallet requirements. These initiatives redistribute technical skills within communities and create small counter-weights to institutional opacity. They also reveal the unpaid labour involved in staying digitally connected to the state as someone from the community translates jargon, manages devices and data, and absorbs the emotional cost when systems fail.
Stratified solidarity and contested online spaces
Community support is uneven and stratified. Access to key WhatsApp groups and Facebook pages is mediated by gatekeepers who also decide what is shared. Many networks are structured along national or linguistic lines, so information flows readily within origin-country circles but seldom across them. In some groups with internal denominational or political differences, participants describe parallel networks that do not always cooperate or share information.
Those who lack smartphones, data, literacy in dominant languages or personal ties to gatekeepers may be physically surrounded by compatriots yet digitally excluded from the channels through which opportunities and warnings circulate.
Digital spaces are also contested terrain in the broader politics of xenophobia and vigilantism. Community leaders who use Facebook to document police raids, landlord abuses or service refusals, sometimes find themselves targeted by online harassment and offline intimidation. Participants recount instances where xenophobic actors infiltrate WhatsApp groups to spread rumours about impending “clean-up” campaigns or to blame particular nationalities for crime. This deepens mistrust between community groups and makes coordinated responses more difficult.
These dynamics mirror national patterns in which anti-foreigner narratives circulate online before materialising in physical attacks and boycotts. For people on the move, digital networks function simultaneously as tools for survival and as channels through which they are watched and misrepresented, a tension that emerges across origin-country groups in our conversations.
Laying the ground for a digital and data analysis
Community networks and digital groups mediate almost every interaction with the state and market, while also encoding internal hierarchies and exclusions. These realities set the stage for the next section, which turns from experience to technical architecture. It traces how data about refugees and asylum seekers moves through state, commercial and community systems. It also asks whose priorities are embedded in the infrastructures that now govern movement, access and risk in this corridor.
Patterns of harm and protection along this corridor depend on who systems recognise, whose claims they delay and whose records are circulated across institutions. The same digital and legal infrastructures that stabilise status for some people leave some others relying on intermediaries or avoiding state spaces altogether. IOM’s Determinants of Migrant Vulnerability (DoMV) frames vulnerability as the product of individual, household and family, community, and structural factors, rather than as a fixed trait of particular migrants.
Applying the determinants helps translate earlier findings regarding data invisibility, mixed-status households, community gatekeeping and xenophobic violence into a concise map of risk. It highlights how design choices in digital and legal systems interact with gender, age, work, care responsibilities and local politics to aggravate or mitigate harm. Situating the research within this framework also allows comparison with other contexts that use DoMV, while keeping the focus on locally-specific determinants and lived experience.
Determinants of Migrant Vulnerability
01
Individual factors
Who the person is, and what they can do alone.
GENDER, AGE AND SEXUALITY
Young men in street-based work face frequent stops and checks; older people and women with limited schooling struggle with English-only forms and portals; queer refugees often avoid state spaces after past experiences with hostility.
DIGITAL AND LANGUAGE SKILLS
Access to smartphones, email and English literacy shapes who can use online renewals without intermediaries; limited skills lead to dependence on café operators or relatives for legal processes.
02
Household and family factors
Decisions taken for several people at once.
MIXED LEGAL STATUS
Households combine recognized refugees, asylum seekers, people with lapsed permits and undocumented members; decisions about who engages which institution carry consequences for everyone.
CARE RESPONSIBILITIES AND INCOME
Primary caregivers fit queues, hearings and clinic visits around childcare and informal work; missed renewals due to these responsibilities result in frozen accounts thereafter affecting rent and food.
03
Community factors
Where information actually travels.
NETWORKS AND GATEKEEPING
Origin-country associations, churches and neighbourhood committees provide information, translation and referrals; access to WhatsApp groups and Facebook pages with alerts about raids or documentation changes often depends on language, personal ties and moderators’ decisions.
04
Structural factors
The systems that set the terms for all of the above.
DOCUMENTATION REGIMES AND DATA PRACTICES
Legal hierarchies are implemented through backlogged, partially digitised systems; people may appear in enforcement databases far earlier than in protection-oriented systems.
XENOPHOBIA, POLICING, LABOUR AND HOUSING
Anti-migrant campaigns, document checks, precarious work and overcrowded, informally rented housing cluster risk in specific neighbourhoods; digital system design (online-only, English-only channels, opaque data sharing) amplifies the effects of each breakdown.
*the figure has been derived from
Interplay of the various factors
Legal hierarchies between citizens, permanent residents, recognised refugees, asylum seekers and undocumented people structure who can claim which rights. Yet these categories are administered through overburdened, partially digitised systems that routinely fail to renew permits or update status in time.
Each mechanism is formally neutral, but in combination with xenophobia, gendered care burdens and informal housing, selectively locks out those who cannot safely or credibly present the right documents in the right way. For these groups, invisibility is not the absence of data but a skewed presence in it; they appear in enforcement records before they appear in protection records. Yet they cannot use DHA databases, SAPS systems or financial records to prove identity, correct errors or claim entitlements. They are on the system enough to be stopped, questioned or profiled, but not enough to trigger protection or redress. It is datafication without rights.
Spatial segregation, overcrowded rental markets and dependence on informal intermediaries for housing, jobs and access to basic services magnify the effects of each bureaucratic or digital breakdown. In this corridor, vulnerability is a verdict produced at the intersection of law, labour markets, digital architectures and community politics.
Information about where raids take place, which officials demand bribes, and which remittance operators accept particular documents rarely enters formal databases. Community members instead log and circulate these details through WhatsApp groups, congregation networks and refugee-led organisations, building an internal evidence base about institutional behaviour over time.
These flows form a shadow analytics layer in which communities generate knowledge that guides risk-mitigation when trustworthy official information is absent. Participants from different origin countries describe using these insights to decide which reception offices to approach, which branches to avoid and when to stay indoors. At the same time, formal systems render refugees and asylum seekers highly legible to enforcement and compliance actors, while refugee perspectives reach decision-makers only indirectly, filtered through advocacy channels and community-mediated insights.
The digital ecosystem shaping life for refugees and asylum seekers in South Africa is dense and unevenly built. Bordering operates through KYC checks, sanction-screening APIs and SIM-registration rules as much as through fences and uniforms, with management and control as the dominant organising logic. Service delivery or protection is added only where donor funding or political will allows for it. Refugees and asylum seekers navigate an environment in which they are intensely knowable to multiple institutions yet rarely protected by any of them.
Understanding the crowded ecosystem of actors is a prerequisite for tracing how data moves. Many institutions occupy more than one role over time, but actors are grouped here by their primary function in relation to people on the move and to digital or data infrastructures. The following categories provide the backdrop for the data-flows map and clarify who sets rules, who implements them, who contests them, and who is impacted by them.
Ecosystem Map
Regulators
State institutions are vested with a constitutional and/or legal mandate, led by an intention to manage migration and borders in South Africa. They establish documentation requirements, manage asylum processes, conduct verification and identity checks, regulate financial services, and oversee data protection.
• Department of Home Affairs (DHA)
• Border Management Authority (BMA)
• South African Police Service (SAPS)
• National Director of Public Prosecutions (NDPP)
• South African Reserve Bank (SARB)
• Department of Social Development/South African Social Security Agency (DSD/SASSA)
• Department of Health and provincial health departments
• Department of Basic Education and provincial education departments
• Statistics South Africa (Stats SA)
• Information Regulator (POPIA)
Facilitators
Internationally-funded and relatively well-resourced organisations operating in South Africa with humanitarian protection or oversight mandates. They provide shelter, healthcare, case management and legal referrals to people on the move. They often fill gaps emanating from state capacity constraints and operate in urban settlements where people on the move concentrate. Their work is shaped by international humanitarian standards and donor priorities.
• UN High Commissioner for Refugees (UNHCR)
• International Organization for Migration (IOM)
• South African Human Rights Commission (SAHRC)
Service Providers
Formal commercial institutions and regulated businesses in South Africa offering services where access requires identity verification or proof of legal status. They operate within the formal economy and include financial institutions, telecom operators, private healthcare providers, education providers and intermediaries and verifiers in the employment sectors.
• Banking institutions (commercial banks, mutual banks and their digital channels)
• Mobile-money and remittance operators (e.g., Mukuru, Mama Money, Hello Paisa, Shop2Shop), regulated as ADLAs or bank partners
• Mobile-network operators and ISPs (SIM registration, customer databases)
• Hospitals, clinics and private healthcare providers
• Schools and universities
• Large employers in sectors with heavy migrant labour (construction, hospitality, retail, logistics, private security)
Advocacy Organisations
Organisations operating in South Africa, focused on legal representation, rights protection and policy reform for people on the move. They operate at both grassroots and systemic levels.
• Scalabrini Centre of Cape Town
• Lawyers for Human Rights (LHR)
• Consortium for Refugees and Migrants in South Africa (CoRMSA)
• African Diaspora Forum (ADF)
• Africa Unite/African Solidarity Network and similar civic coalitions
• Kopanang Africa Against Xenophobia (KAAX)
Restrictors
Actors in South Africa that deliberately harm, exclude, exploit or perpetuate violence against people on the move.
• Operation Dudula
• March and March
• ‘Put South Africa First’ movement and similar social media campaigns
• Local business forums, street-level vigilante groups and informal ‘community policing’ formations
• Exploitative landlords, employers and intermediaries
Community Actors
Networks and organisations created by and/or accountable to people on the move in South Africa, ranging from informal peer networks to established grassroots organisations. They are grounded in cultural identity, shared nationality, trust-based relationships and collective survival strategies. They may or may not be formally registered but are characteristically community-accountable and self-directed.
• Refugee-led community-based organisations (CBOs) organised by nationality or language (e.g., Congolese, Somali, Ethiopian, Zimbabwean associations)
• WhatsApp safety networks sharing alerts on raids, xenophobic incidents and documentation changes
• Faith-based organisations and congregations
• Informal peer groups and mutual-aid networks
• African Diaspora Forum’s member associations and other refugee community forums
Research Organisations
Academic institutions, think tanks and research institutes based in or working on South Africa that produce evidence about people on the move through research. Their work examines migration patterns, technology use, data governance, policy impacts and community experiences within the South African context. Their research informs policy advocacy, community understanding and international discourse about SA’s migration systems.
• African Centre for Migration & Society (ACMS), Wits University
• Scalabrini Institute for Human Mobility in Africa (SIHMA)
• Southern African Migration Programme (SAMP)
• FinMark Trust/Cenfri and partners
• University-based centres and scholars
Funders
Institutions providing financial resources to support migration-related work in South Africa. Their funding priorities, conditions, and timelines shape what work gets supported and how organisations operate. Some funders are embedded within South African institutions, while others are external. Their resources are critical to sustaining migration work.
• Multilateral and UN funders: UNHCR, IOM, UNICEF, OHCHR and the World Bank, which fund protection programmes, social-protection pilots and research on refugees’ access to services.
• Bilateral donors: Governments that support migration, human-rights and digital-governance work through development-cooperation programmes (e.g., EU institutions and European states that fund UNHCR and civil-society partners).
• Philanthropic foundations and international NGOs: Foundations and international NGOs supporting anti-xenophobia campaigns, legal aid, community organising and digital-rights work (for example, those partnering with ACMS, Scalabrini, CoRMSA and FinMark Trust).
Mapping the stack: what the data-flows map shows
The map reveals an ecosystem in which formal data about refugees and asylum seekers concentrates around DHA, while interpretive power over that data disperses across a dense network of institutions that shape policy, enforcement and access.
• DHA’s asylum and civil-registration systems capture biographical data, biometrics and case histories, then issue documents such as asylum permits, refugee status documents and identity cards that unlock almost all other services. Service providers including banks, hospitals, schools and mobile-money operators depend on DHA-recognised identifiers to perform KYC, enrol students and treat patients.
• The heaviest red flows run inward from refugees and asylum seekers into DHA and outward into enforcement and compliance systems, while the thickest blue flows run from communities and advocates back into regulators as attempts to correct, contextualise or contest what those data mean. The architecture encodes a structural asymmetry where data primarily moves inward to support control and compliance functions. Only in narrower channels does data return as protection or redress, for example through litigation, targeted social-protection programmes or remedial policy shifts.
For people on the move, this pattern diverges from the spirit of South Africa’s constitutional and legislative commitments, which frame refugees and asylum seekers as rights-holders entitled to equality, dignity and access to basic services. Laws such as the Refugees Act and POPIA articulate duties of fair processing, non-discrimination and due process, yet the operational data architecture prioritises institutional risk management and regulatory comfort over giving individuals clear pathways to challenge decisions, correct records or have harms repaired.
How data moves: formal and informal pathways
Service providers layer rules on top of identifiers. Banks and ADLAs build KYC engines around DHA-recognised documents; some remittance firms explicitly advertise that they accept passports, foreign IDs and asylum or refugee documents, while others apply narrower thresholds. Remittance services maintain their own customer databases and compliance processes in line with SARB requirements. Thus, each transaction generates a record that can evidence economic contribution and simultaneously feed risk-scoring.
Informal pathways overlay this formal architecture. Community organisations and NGOs maintain their own records of cases, referrals and complaints, while WhatsApp groups and social-media pages act as real-time alert systems for xenophobic protests and shifts in administrative practice, and as conduits through which rumours and anxieties circulate.
In this configuration, digital and data infrastructures for refugee governance do not simply mediate access; they combine into a dispersed bordering apparatus where power sits across state, private and intermediary actors. Participants describe being constructed as potential security risks and exposed to distinctive data harms through these systems, even as they rely on the same infrastructures to survive everyday xenophobia and exclusion.
Payments and financial infrastructures as bordering tools
The payments ecosystem, particularly cross-border remittances, makes the dispersed bordering dynamic concrete. Formal remittance services such as Mukuru, Mama Money and Hello Paisa are widely used by refugees from Zimbabwe, the DRC, Nigeria, Uganda and elsewhere in South Africa and operate as authorised dealers in foreign exchange or licensed ADLAs under SARB and FIC oversight.
Where providers explicitly accept asylum and refugee documents as valid IDs, barriers fall for those who hold current papers, while people with irregular or no documentation remain outside all formal channels. Every formal remittance creates a data point that documents cross-border economic ties and can also train surveillance systems on particular corridors and communities.
Domestic payments follow similar patterns. Because only banks can issue e-money in South Africa, mobile-money and wallet products must partner with banks and integrate into the national payments system, which reinforces bank-grade KYC even when front-end products appear more flexible.
Refugees who cannot meet proof-of-address requirements or whose permits have lapsed have fewer viable options on formal payment rails. They rely more heavily on cash, including in situations where they would prefer digital tools for security or convenience. Many also actively choose cash for reasons such as budgeting, privacy or mistrust of institutions. The difficulty lies in the absence of choice over when and how to shift between cash and digital instruments, which heightens exposure to theft and keeps much of their economic life invisible to institutions that only read formal transaction data.
Parallel infrastructures: community data and ResTech
Alongside formal systems, community actors sustain a parallel digital layer that is mostly absent from official diagrams and central to lived experience. Refugee-led CBOs, faith-based organisations and informal peer networks maintain WhatsApp safety groups that circulate information about raids and xenophobic incidents, effectively running a community-level incident-reporting system in the absence of a comparable state-run mechanism.
These groups share templates for letters to schools and clinics, photographs of circulars on rights to access services. They offer practical advice on which offices, branches or remittance outlets have recently accepted specific documents. The data here is largely qualitative and relational. Administrators track who reliably verifies rumours, and organisations keep records of assistance from emergency shelter to transport for renewals, using these records in advocacy with funders and state actors.
Community infrastructures therefore function as grassroots digital infrastructure. They manage identity by recognising who belongs to the network, route information and resources, and generate insights that sometimes travel back through advocacy organisations to regulators.
These systems carry their own power dynamics and vulnerabilities. Group administrators decide who gains entry and what circulates, internal politics shapes whose cases receive sustained attention. Reliance on WhatsApp and Facebook means that data generated in these networks remains technically accessible to corporate actors and, in some circumstances, law-enforcement agencies.
Participants recount episodes where xenophobic actors infiltrate groups and spread misinformation, showing that community infrastructures sit inside the wider politics of surveillance and hate rather than outside it. Community ResTech mitigates some harms produced by formal systems and still leaves people exposed to the broader data economy that those systems inhabit.
Data governance and opacity
Governance across this ecosystem is uneven and, for those it governs, largely invisible. DHA operates under the Refugees Act, immigration legislation and general administrative-law principles, while POPIA offers broad data-protection standards without a dedicated, publicly accessible framework tailored to asylum and immigration systems.
The Information Regulator holds formal oversight, yet refugees in our fieldwork say they have never heard of POPIA or seen signage explaining their rights in DHA offices. Public documents do not set out audit mechanisms for how biometric and case data are accessed within DHA or shared with SAPS and other agencies. Whatever internal controls exist are opaque to the people whose data they concern.
Banks and telecoms fall under clearer regimes such as FICA and sectoral laws that mandate due-diligence, record-keeping and reporting, and corporate-governance frameworks that require risk-management and compliance functions. Algorithmic risk models, data-sharing agreements with analytics vendors and decisions about which documents to treat as valid nonetheless remain proprietary. So, those affected cannot readily inspect or contest them.
Digital public-infrastructure principles such as open standards, transparent governance and user-centred interoperability appear selectively, mainly in strategy documents. Design processes, vendor choices and technical architectures stay closed. There is no public register of APIs or data-sharing agreements across asylum, civil-registration, banking and telecom domains. Refugees lack a simple, low-cost complaint pathway to challenge how their data is processed outside of courts.
State of play: fragmentation, interdependencies and system architecture
The digital ecosystem governing refugees and asylum seekers in South Africa operates as a patchwork of powerful but misaligned systems rather than a coherent, rights-centred infrastructure. Each major actor runs robust platforms for its own mandate, and cross-cutting coordination appears mainly through specific projects or bilateral integrations. Where interdependencies exist, they often compound harm: a permit that fails to renew online cascades into banking and SIM systems as account closures or service denial. A mis-entered date of birth in DHA records reappears as an obstacle in every subsequent verification.
Certain systems function as infrastructural keystones. DHA’s asylum database acts as the single source of truth for legal status. When it fails or becomes inaccessible, the legal persona of refugees collapses in the eyes of banks, employers and police. Community infrastructures, in turn, depend on commercial platforms whose governance refugees cannot influence. A unilateral change in WhatsApp’s encryption policy or content-moderation rules could alter community security overnight.
In this configuration, refugees and asylum seekers are legible enough to be managed, monitored and sometimes targeted through data, while remaining insufficiently visible as rights-holders with agency over how their digital lives are organised.
Limitations of the mapping exercise
The map has important limitations that should be stated plainly. First, it focuses on organisations with which the research team had direct engagement or for which reliable public documentation of roles and data practices exists. Many actors that shape refugees’ digital lives are not individually named. Second, arrows primarily represent known or reported flows; they do not capture the full complexity of real-time data exchanges, especially within proprietary systems such as telecoms metadata, fraud-analytics platforms or global social-media infrastructures. Third, informal and illicit actors generate and use information about refugees but are not easily representable in a formal map, even though their decisions affect people’s safety.
These constraints mean the map should be read as a partial but analytically useful snapshot rather than as an exhaustive blueprint. It identifies where the formal power over refugee data is concentrated, how community-level infrastructures intersect with that power, and where further investigation is needed to understand how bordering operates through code and databases.
Within this case study, the Migration Infrastructure Assessment (MIA) framework is used to take a structured view of how digital systems intersect with movement along the corridor into South Africa. The framework assesses five pillars—digital infrastructure, governance of digital infrastructure, community agency, policy environment and information landscape—and situates each on a three-stage trajectory from nascent to emerging to mature.
Each pillar is evaluated in isolation, using predefined indicators. In this section, we apply those indicators to South Africa to understand how refugee-facing digital systems are currently designed and implemented, and how far they speak to refugee realities across law, infrastructure and lived experience.
Snapshot of the Assessment
Digital infrastructure and the policy environment move fastest. South Africa is pushing toward digitised asylum procedures, linked banking and SIM registration, and integrated civil‑registration systems, while migration and refugee law formally affirms substantive rights. As detailed in the following sections, our fieldwork revealed that governance of this infrastructure and the public information landscape lag behind. Participants described opaque data flows, limited oversight and inconsistent guidance from institutions that set the rules. Community networks are dense but improvised with refugee‑run WhatsApp groups, overstretched legal clinics and local advocacy coalitions filling gaps the institutional system creates rather than components it intentionally delegates. Taken together, these dynamics produce an ecosystem in which the policy rails for digital migration management are robust, but the terms on which refugees travel them remain fragile, contingent and difficult to contest.
Entity
Nascent
Emerging
Mature
State of Digital Infrastructure
South Africa has digitised systems nationwide for identity, asylum processing, banking and SIM registration, with relatively stable connectivity in main urban centres. For refugees and asylum seekers, these systems accept their documentation in some channels and reject or question in others. Coverage and usability for refugee documents remain partial on MIA indicators.
Governance of Digital Infrastructure
POPIA and an Information Regulator establish a general data‑protection regime, and sectoral rules exist for banking, education and health data. Yet oversight of migration databases is limited, enforcement capacity is uneven, refugee‑specific guidance is fragmented, and practical avenues for access, correction and redress are weak.
Community Agency
Refugee‑led CBOs, faith communities, and peer networks sustain dense support structures, and informal representation, which function continuously. These actors increasingly shape advocacy coalitions and policy debates, yet internal stratification and resource scarcity mean access to community infrastructures and decision‑making remains uneven, keeping community agency strong but not universally enabling across all refugee sub‑groups. The state does not recognise or support these groups.
Policy Environment
A rights‑affirming legal framework domesticates international refugee instruments and recognises persecution, including on sexual‑orientation grounds, as valid for asylum. Strategic litigation and court rulings meaningfully influence implementation, while recent policies foreground digital migration management. Implementing regulations, directives and operational circulars remain fragmented, slow to update and unevenly communicated, particularly on procedures, timelines and data use, which leaves significant space for discretion and contestation in day‑to‑day practice.
Information Landscape
Hostile political and media narratives that frame migrants as abusers of the asylum system or as security risks shape the wider information environment and influence how institutions and the public interpret refugee presence. Undocumented and never‑registered migrants remain largely absent from official figures. Institutional communication on rights, procedures and digital changes is often English‑only, sparse and poorly disseminated through channels refugees actually use. Civil‑society research and rich community reporting circulate in fragmented, sometimes misinformation‑prone channels, leaving the overall information landscape only partially reliable or accessible for refugees.
The central problem is not that refugees lack documents or devices, but that the systems they enter treat their documentation as permanently conditional and their presence as inherently risky. The downstream harm is a form of digital limbo, where people on the move must constantly re-perform their legitimacy to keep access to banking, connectivity and basic protection.
Compared to other corridors in our work, South Africa sits at an unusual point in the MIA landscape. It combines a relatively advanced digital infrastructure and policy environment with a politicised, enforcement-heavy migration regime as well as strong but under-resourced community networks. This case therefore forces a sharper question onto the collective research agenda, not whether digital systems can include refugees in principle, but on whose terms inclusion occurs, and what forms of visibility people on the move can safely afford.
The following recommendations focus on shifting how data and digital systems treat refugees and asylum seekers in South Africa by intervening at the points where DHA-anchored data flows feed into various infrastructures. Permanent change requires regulatory mandates, using service providers, funders, research organisations and community actors as critical implementation partners. Most suggestions address immediate harms produced by current digital and data practices rather than longer-term integration or status reform. They are not exhaustive. They identify urgent, digitally-mediated pressure points where regulatory action can rebalance a system that presently channels information inward for control and compliance while offering refugees and asylum seekers few reliable ways to exercise rights, seek correction or obtain redress.
While grounded in the South African reality, these recommendations speak to any context where payments infrastructure fragments along documentation, or where communities have built parallel digital layers to survive the gaps in formal systems. Some of these conditions are present across East Africa, mixed-status migrant populations in Southeast Asia, and post-crisis digital transitions in the West Asia region. The recommendations have been outlined in the sheet here.
IMPLEMENTING STAKEHOLDER
CATEGORY
RECOMMENDATION
EXAMPLE & LINKS/REFERENCES
Regulator, Advocacy Organisations
Policy
Amending asylum and immigration regulations to explicitly delink minor clerical errors and missing origin-country documents from status cessation; would reduce wrongful rejections without weakening substantive credibility assessments.
After court challenges, South Africa's regulations and practice on bank access for asylum seekers were adjusted to prevent blanket exclusion, illustrating that targeted legal changes can correct over-broad interpretations without compromising institutional control. A parallel amendment for clerical and documentary failures in RSD would likely attract criticism framed around fraud risk; clear definitions of the narrow category of errors covered, combined with audit mechanisms, would make the intended scope concrete and harder to misrepresent.
Regulator, Advocacy Organisations
Policy
Mandating that all substantive asylum interviews use interpreters drawn from an accredited pool co-managed with civil-society organisations would improve accuracy and safety in narrative-based claims, including SOGI cases.
In the UK, organisations such as Rainbow Migration and Micro Rainbow have developed guidance and training for interpreters working with LGBTQI+ asylum seekers, illustrating both the harms produced by untrained interpreters and the feasibility of queer-sensitive interpretation standards in asylum systems. Applying accredited-pool requirements in South Africa would increase interview preparation time and cost, and would require explicit safeguards, including refusal and complaint mechanisms, for cases where an interpreter's conduct imports homophobic or transphobic bias into proceedings.
Regulator, Research Organisations
Policy
Embedding questions that distinguish between documented, lapsed and never-registered migrants and refugees in national surveys would improve visibility of populations outside administrative systems and support evidence-based planning.
Stats SA's first Migration Profile report began integrating multiple data sources to estimate migrant numbers, indicating institutional appetite for more granular categorisation even amid political sensitivities. Introducing status-disaggregated questions into established surveys such as the General Household Survey or Community Survey would require careful questionnaire design to ensure reliability, and a communication plan for how findings will be released that addresses the political sensitivity around officially acknowledging larger or more varied migrant populations.
Regulator
Policy
Designating xenophobic vigilantism as a specific aggravating factor in public-order offences and mandating reporting on such cases would signal zero tolerance for organised attacks on migrants and create a basis for tracking trends.
South African courts have already treated hate-crime elements as aggravating in sentencing, and SAHRC investigations of xenophobic violence have established analytical and legal tools that could be extended explicitly to vigilantism against migrants. Formalising this as a reporting obligation would strain already stretched prosecutorial resources and would face resistance in local communities aligned with vigilante groups, making implementation most viable when paired with dedicated prosecution capacity and community-safety programmes that address the underlying grievances that vigilante mobilisation exploits.
Community Actors, Regulator
Community
Convening regular migrant municipality consultation forums that feed into city safety, housing and service-delivery plans would surface early warnings and practical fixes to local xenophobic pressure points.
The African Diaspora Forum's engagements with municipal authorities during previous xenophobic flare-ups showed that organised migrant forums can shape local responses even when representation remains uneven. Sustaining these forums as formal consultation channels rather than one-off engagements would require deliberate outreach mechanisms to include smaller, less visible communities and queer or undocumented migrants who are often absent from the better-organised nationality associations that tend to dominate such spaces.
Regulator, Facilitator
Structural Vulnerabilities
Aligning eligibility and enrolment procedures for social grants, schooling and primary health care with the realities of mixed-status households would reduce the cascading impact of a single lapsed document across multiple service domains.
UNHCR's work with South African authorities to clarify grant and schooling access for recognised refugees shows that targeted guidance can shift practice where departments acknowledge gaps. Extending that clarity to cover the full range of household documentation situations, including children without papers and households with mixed status categories, would require tailored guidance for frontline staff and a communications strategy that addresses concerns that flexible verification constitutes special treatment rather than a correction of over-exclusion.
Regulator, Advocacy Organisations
Structural Vulnerabilities
After court challenges, South Africa’s regulations and practice on bank access for asylum seekers were adjusted to prevent blanket exclusion, illustrating that targeted legal changes can correct over-broad interpretations without collapsing control.
Municipal-level safety audits conducted with civil-society partners in Johannesburg and Durban have been used to design more inclusive responses to xenophobic violence, even where implementation has been uneven. Extending this approach into formal urban-management and disaster-risk frameworks, with explicit attention to migrant-dense areas, would involve patient engagement with residents who fear that such planning privileges foreigners and would need communication strategies that frame these measures as investments in overall neighbourhood safety.
Regulator
Institutional Accountability
Embedding independent third-party monitors at Refugee Reception Offices (RROs) to observe queues, interviews and decision-making would deter petty corruption and abusive practices.
SAHRC and civil-society monitoring of polling stations and detention facilities in South Africa has improved compliance with rights standards, indicating that observed environments can shift frontline behaviour. Applying this to RROs through accredited entities or consortia, with rotating teams, unannounced visits and mixed methods (on-site observation plus anonymous feedback channels), would help counter the tendency for audits to become ritualised and gamed over time. Such a scheme would likely require legislative or MoU-based mandates and clear protocols to manage tensions with officials who see external oversight as interference.
Regulator, Advocacy Organisations
Institutional Accountability
Publishing disaggregated asylum-processing statistics, including rejection reasons, office-level turnaround times and appeal outcomes, would expose systemic bias and guide reform.
Corruption Watch and Lawyers for Human Rights have used access-to-information requests and court records to reveal systemic problems at RROs. Where granular data were available, those records underpinned credible advocacy for change. A regularised publication regime that includes narrative guidance on how to interpret the figures would create tools for policy design and civil-society monitoring, while also requiring communication strategies that anticipate politicised readings of high rejection rates or backlogs at particular offices.
Regulator, Community Actors
Institutional Accountability
Providing recurring training for police units in migrant- and refugee-dense areas on lawful document checks, hate-crime indicators and non-collaboration with vigilante groups, alongside linked training for local social workers, would reduce everyday harassment and arbitrary detention and strengthen referral pathways.
Targeted SAPS training programmes developed with civil society after earlier xenophobic pogroms have shown modest improvements in local practice when followed up over time. A refreshed programme that pairs police training with social-work training and creates joint protocols for responding to xenophobic incidents would need performance incentives and sanctions for repeat offenders, as well as monitoring that tracks whether learning translates into fewer unlawful stops and more appropriate referrals.
Regulator
Institutional Accountability
Ensuring that court orders and laws which prohibit movements such as Operation Dudula from blocking clinics, schools or businesses are consistently enforced would reaffirm rights to access services for refugees and asylum seekers.
South African courts have issued interdicts against Operation Dudula’s attempts to block foreigners from accessing services, and in some instances police have enforced these orders to disperse unlawful gatherings. Turning this into a stable practice would require dedicated security resources, clear operational plans for crowd management and explicit measures to protect pro-migrant officials, health workers, educators and community leaders who face threats when they uphold inclusive service provision.
Regulator, Facilitator
Policy
Embedding SOGI-specific credibility guidelines, developed with queer-led organisations, into RSD policy and quality-assurance reviews would reduce reliance on stereotypes and office-by-office heuristics in asylum decisions.
UNHCR’s SOGI guidance and UK SOGI guidelines, combined with targeted quality-assurance projects in several asylum systems, have moved practice away from invasive questioning and towards more structured assessments that still leave room for adjudicator judgement. Applying a similar model in South Africa would involve co-designing guidelines with local queer-led groups, integrating them into file reviews and mentoring, and updating them as case law evolves so that quality-assurance mechanisms reinforce rather than dilute substantive protections.
Regulator, Research Organisations
Structural Vulnerabilities
Creating a migration- and asylum-focused fact-check and pre-bunking unit jointly run by government communicators and independent fact-checkers would blunt viral falsehoods that drive xenophobic mobilisation.
India’s PIB Fact Check Unit and South Africa-based Africa Check demonstrate that dedicated structures can systematically verify and debunk misinformation, including on migration, when their methods, datasets and corrections are transparently published and open to external audit. A joint unit focused on refugee and migrant narratives would need a governance model that protects editorial independence, clear escalation channels into security and social-cohesion actors, and communications strategies that explain how corrections are produced so that they are not dismissed as state propaganda.
Regulator, Service Provider
Data Governance
Requiring large social-media platforms to provide anonymised, aggregate reports on xenophobic incitement and hate-speech trends would help advocacy and oversight bodies anticipate and respond to risks of violence.
Collaborations between Africa Check, social-media platforms and newsrooms in South Africa have already relied on platform-provided data and fact-checking partnerships to identify harmful narratives and correct them. A regulatory requirement that focuses on major social-media platforms, rather than telecom operators, and specifies privacy-preserving aggregation, clear definitions of xenophobic content and proportionate thresholds would create early-warning indicators while limiting the risk that reporting morphs into broad surveillance of online speech.
Regulator, Advocacy Organisations
Data Governance
Establishing a standardised data-access and correction-request form and tracking system, co-designed with legal NGOs, would make it practically possible for refugees and asylum seekers to exercise POPIA rights over their records.
Community-based advice offices have long used standardised forms to lodge social-grant and housing appeals, and those templates have helped government departments process and respond more consistently to citizen claims. Applying this practice to asylum and civil-registration records, with simple forms translated into key languages, digital and paper submission channels and status-tracking mechanisms, would likely generate an initial spike in complex cases for DHA back-office teams and would therefore need phased implementation and capacity-building plans.
Regulator
Data Governance
Flagging immigration-related stops, searches and arrests in police information systems with a dedicated code and publishing anonymised statistics would make patterns of discriminatory enforcement visible and create a basis for oversight.
Xenowatch and SAHRC analyses of xenophobic violence have shown that coded categories in police and complaint datasets can reveal enforcement trends, even when those categories were originally designed for other purposes. Applying a dedicated code for immigration-related police contacts would require buy-in from SAPS leadership and robust anonymisation protocols, since the same data that surfaces systemic profiling could, if poorly governed, expose individuals or be used to justify intensified enforcement of specific communities.
Regulator, Community Actors
Data Governance
Creating a formal mechanism for refugee-led organisations to submit periodic dossiers on documented cases of mis-spelled names, lost files and erroneous decisions to DHA’s internal quality-improvement processes would translate community evidence into system fixes.
Corruption Watch and Lawyers for Human Rights have compiled structured case evidence on abuses in Refugee Reception Offices (RROs), and where authorities engaged with those submissions, they contributed to targeted remedial actions. A formalised dossier mechanism would work best when accompanied by transparent feedback loops that show submitting organisations how their evidence influenced change, reducing the risk that community-generated evidence is received as adversarial rather than as a resource for institutional improvement.
Regulator, Facilitator
Policy
Translating asylum application forms and core information materials into the main languages spoken by applicants and formally accrediting community interpreters to assist with completion would make RSD processes more accurate and accessible.
The EUAA’s Let’s Speak Asylum toolbox and UNHCR country operations in India and Ukraine offer multi-lingual portals and guidance, demonstrating that multi-lingual information provision in asylum systems is feasible when paired with standardised templates and interpreter protocols. Implementing this in South Africa would require central coordination over version control and updates so that translated materials remain consistent across offices, and accreditation standards for interpreters would need safeguards against bias, including specific standards for SOGI and other sensitive claim types.
Regulator, Service Provider
Digital Infrastructure
Standardising the acceptance of refugee and asylum‑seeker documents across KYC and account‑opening procedures would reduce arbitrary refusals and account closures for people on the move.
After litigation and FIC guidance, South African banks were required to accept asylum‑seeker and refugee permits for account opening, which showed that regulatory clarification can shift KYC practice without breaching AML rules. Implementing a standardised acceptance regime at scale would involve short‑term operational adjustments, including staff retraining and system updates, and would work best alongside simple redress pathways that allow refugees and asylum seekers to contest refusals and have compliant branches corrected when they misapply the rules.
Regulator, Facilitator
Digital Infrastructure
Piloting multi-lingual, human‑assisted digital renewal booths at or near Refugee Reception Offices would reduce dependence on internet cafés and lower error rates in email‑based extensions for asylum seekers and refugees.
UNHCR’s joint project with DHA to clear the asylum backlog created dedicated processing teams and tailored communication channels, indicating that collaborative operational innovations are feasible within the South African asylum system. A renewal‑booth model that relies on trained staff or accredited assistants, rather than AI tools, to help applicants complete online processes in their own languages. It would need careful scheduling and eligibility criteria to prevent overcrowding and queuing from simply shifting into a new space.
Regulator, Funders
Digital Infrastructure
Expanding the purpose of DHA’s existing Mobile Home Affairs Units to cover asylum‑seeker and refugee renewals in major urban neighbourhoods, on fixed and publicly communicated schedules, would materially reduce travel time and support regular renewals for people on the move.
DHA Mobile Home Affairs Units already deliver Smart IDs and birth registrations across rural South Africa, demonstrating that outreach‑based civil‑registration is administratively viable within the department’s existing operating model. Adapting that model for urban refugee neighbourhoods would extend a proven mechanism to a population that currently bears disproportionate access costs . It would require attention to staff rostering, local political dynamics and, where relevant, stakeholder communication in areas where migrant presence is contested.
Regulator, Community Actors
Digital Infrastructure
Accrediting selected community‑based paralegals and organisers as documentation assistants attached to mobile DHA teams would extend renewal, correction and basic advice services deeper into migrant neighbourhoods.
Ward‑based primary health‑care outreach teams that rely on trained community health workers supervised by nurses, and community advice offices that use accredited paralegals for frontline justice support, show that community‑embedded, state‑linked paraprofessionals can deliver sensitive services at scale when oversight mechanisms exist. A documentation‑assistant model would need clear ethical standards, supervision, zero‑fee rules and accessible complaint hotlines so that communities can report misconduct and so that local gatekeeping or capture is identified and addressed early.
Regulator, Service Provider
Digital Infrastructure
Introducing tiered KYC products, explicitly calibrated for refugees and asylum seekers, would expand safe use of remittance and wallet services while maintaining AML safeguards.
FinMark Trust’s work on low‑value, risk‑based KYC in the South African remittance market has already informed the design of products by Mukuru and Mama Money, suggesting that regulators and providers can collaborate on calibrated inclusion for cross‑border transfers. Extending this approach to a broader set of payment products for refugees and asylum seekers would entail revising risk‑scoring models, communicating clearly about thresholds and conditions, and working with supervisors to discourage blanket de‑risking by institutions that are uneasy with differentiated compliance categories.
Regulator
Data Governance
Publishing data‑sharing agreements and forward‑looking risk assessments for asylum, border and civil‑registration systems – including how refugee and asylum data may be used in AI and analytics models – would clarify which agencies access refugee data and on what legal basis.
The Information Regulator’s investigations into large data breaches have already led some departments and firms to disclose processing activities and remedial plans, showing that transparency mandates can be enforced in South Africa’s context. Extending this to structured publication of agreements and pre‑emptive risk assessments on onward sharing and AI training would likely surface politically sensitive practices and expose departments to litigation or public criticism for historical non‑compliance, yet it would also create a basis for expert consultation on mitigation measures before harms crystallise.
Regulator, Advocacy Organisations
Data Governance
Establishing a dedicated refugee and migrant data‑rights complaints window within the Information Regulator would create a concrete route for people on the move to contest misuse of their personal data.
Civil‑society organisations have already brought POPIA‑related complaints on consumer data to the Information Regulator, demonstrating that complaints‑driven enforcement can shift practice when channels are visible and accessible. A specialised window focused on refugees and migrants would need clear case‑intake criteria, guidance for NGOs that prepare complaints and resourcing plans so that complex cases do not overwhelm a small regulator.
Regulator, Research Organisations
Data Governance
Instrumenting DHA’s asylum and ID systems with auditable logs, including threshold‑based alerts and periodic supervisory reviews, would make it possible to detect systemic biases and error patterns in decision‑making and data access.
The African Centre for Migration & Society’s Xenowatch project has used crowd‑sourced and official data to map xenophobic incidents, illustrating the value of independent analytics layered over state information in South Africa. Applying a similar analytic lens to well‑designed DHA logs, with quarterly or semi‑annual review cycles and automatic alerts when rejection rates, file accesses or specific outcomes cross agreed caps, would require upfront investment in logging infrastructure and would need careful governance to reassure officials that scrutiny aims at systemic improvement rather than individual scapegoating.
Regulator, Facilitator
Policy
Codifying detailed refugee status‑determination guidelines for sensitive claims, and integrating them into officer training, would reduce arbitrary credibility tests faced by queer refugees and other applicants whose persecution experiences are often misunderstood.
UNHCR’s global SOGI RSD guidelines and their uptake in jurisdictions such as Kenya and Canada show that formalised standards can move adjudication away from stereotype‑driven questioning towards more consistent credibility assessments. Adapting these approaches for South Africa, with locally grounded examples and monitored training, would require engagement with adjudicators who are used to broad discretion and would likely generate political pushback in some settings, yet it would anchor decisions more firmly in rights‑based criteria.
Behind the Research
This case study uses qualitative methods to understand how people on the move in South Africa experience digital infrastructure, data practices and access services. The research was designed to centre lived experience while situating it within the country’s institutional and DPI landscape. The aim was to balance analytical depth with care, given the precarious positions many participants occupy.
A participatory and trauma-informed methodology was used, which sought to minimise extractive dynamics and give participants control over how they shared their stories. The primary methods were semi-structured focus group discussions organised around personas that reflected common migration trajectories and documentation scenarios. This design allowed participants to describe how they encountered digital systems, various demands for their data and bureaucratic processes in ways that felt grounded in their own lives, while enabling comparison across different legal statuses and service pathways.
Fieldwork involved 50 participants between the ages of 14 and 65, predominantly women, originating from the Democratic Republic of the Congo, Kenya, Ethiopia, Somalia and Zimbabwe. Participants included asylum seekers, recognised refugees and people awaiting decisions on their status, reflecting a range of documentation situations but not including South African citizens, permanent residents of long standing, or internal migrants. Interviews were not conducted with undocumented migrants who had never engaged the asylum system, people in immigration detention, or officials working inside state agencies; their perspectives appear indirectly, through secondary sources and community narratives, rather than as direct testimony. Discussions took place in community settings with the support of local organisations, interpreters where required and facilitators trained to recognise and respond to distress. In parallel, desk research was undertaken and we had conversations with various stakeholders including researchers, advocacy organisations, civil society actors and community-based organisations working on migration or digital rights to better understand how systems are designed and implemented in practice.
Material was analysed using an iterative and thematic approach. There were recurring patterns around documentation, access to services, data collection and sharing, community support and trust in institutions. Insights from desk research and expert conversations were used to contextualise and challenge our interpretations of field narratives. The goal was to produce a grounded and analytical account of digital and data infrastructures as experienced by people on the move.
Consent and risk were handled with particular attention to participants’ insecurity. All participants were provided with an honorarium for their time. Participation was voluntary, with clear oral explanations of the study aims, the topics to be discussed and the use of anonymised findings. Participants could skip questions or withdraw at any point without consequence. There were no audio or video recordings as this could increase fear of surveillance, relying instead on detailed note-taking and immediate de-briefing among the research team. Identifying details were not collected to reduce the risk to individuals. These choices limit what can be claimed about prevalence and precise trajectories, but they are essential to keeping potential harms proportionate to the value of the research.
Limitations and scoping
This study has several important limitations that shape how its findings should be read. The fieldwork sample is relatively small and concentrated in specific communities and locations and does not capture the full diversity of South Africa’s migrant and refugee populations. Our access to institutional perspectives was uneven. While desk research and expert conversations provided insight into formal policies and system design, we were not able to conduct a full set of interviews with all relevant government departments, financial institutions or technology providers. As a result, there may be gaps between formal rules and actual practice that we could not fully verify.
We focused primarily on digital identity, payments and selected service access points. Other parts of South Africa’s digital public infrastructure, such as education or health information systems, were not examined in equivalent depth. The research reflects a specific moment in an evolving policy and technological landscape, so the findings indicate broader patterns and tensions rather than an exhaustive map.
© 2026 Aapti Institute. All rights reserved.