Chapter 7 The urban space: post registration in South Africa

Material factor: state capacity concerns in urban spaces

A key material factor influencing the hands-off approach to post-registration is simply a lack of resources and capacity. Local integration and assistance programmes for refugees are costly both in terms of resources and manpower (OECD, 2017). Furthermore, as set out below, when a country has a multitude of structural issues relating to poverty and inequality within its own voting population, these material factors combine with democratic pressures and national interests, making a potent mix.

South Africa is a dual economy, with nearly half of its population classified as chronically poor (World Bank, 2018a).¹ The World Bank reported extremely high formal unemployment rates of 26.7 per cent at the end of 2017, with the unemployment rate for youths even higher, at around 50 per cent (World Bank, 2018a). The state also has one of the highest inequality rates in the world (World Bank, 2018b). This sees the poorest 20 per cent of the South African population consuming less than 3 per cent of total expenditure, while the wealthiest 20 per cent consume 65 per cent (World Bank, 2018a). This wealth disparity is due to an ‘enduring legacy of apartheid’ (World Bank, 2018b) resulting from a history of labour exploitation and privilege built through that exploitation (Ballard et al., 2017).

At the municipality level, it is a similar picture. Johannesburg in the Gauteng province is South Africa’s largest city with more than 4.4 million residents (City of Johannesburg, 2013). The local economy has increased since the early 2000s, with the Gauteng province being the most industrialised and economically diverse region of South Africa and having the lowest poverty rate (19 per cent in 2015) in the country (Parilla and Trujillo, 2015). Nonetheless, around 20 per cent of its residents are still not in formal housing (de Wet et al., 2011). Similar structural problems persist in gaining access to quality health services for large portions of the local population (Vearey, 2017).

South Africa has attempted to respond to these disparities in socio-economic standards by running fairly generous social assistance programmes for low-income citizens. For example, in 2010 ‘one in every two households had a social assistance beneficiary, and the budget had doubled since 1994 to over 3.5% of GDP’ (Barrientos and Pellissery, 2012). One key motivation for a political settlement (such as the ANC in South Africa) to maintain and expand social protection programmes is as a means of ‘securing the acquiescence of groups that might otherwise threaten political stability and economic growth in the future or to undermine political opponents’ (Lavers and Hickey, 2015). In this way, the political elites adopt generous social assistance programmes as compensation for the ‘capital-intensive growth strategy’ that has created and sustained these high levels of inequality and unemployment seen in South Africa (Seekings and Nattrass, 2005).

By comparison, motivation for the ruling political settlement to use scarce resources on a comparatively small proportion of the urban population (namely refugees) who cannot vote, and who are often unable to contribute via taxes, remains low. Firstly, since the end of the apartheid system, all forms of international migration into South Africa have continued to increase, with urban areas such as Johannesburg being a main destination for migrants and refugees from across the continent (Landau, 2007). Consequently, the number of economic immigrants, asylum-seekers and refugees in South Africa is significant.² Yet, the total number in urban spaces is relatively small when compared to the number of internal migrants regularly moving to the city. For example, in 2011 nearly one-third of Johannesburg residents were born elsewhere in South Africa (and can vote), compared to around 13 per cent who were born outside the country (Vearey et al., 2017).³ Secondly, it has become increasingly difficult for refugees and other groups of African migrants to gain permanent residency, let alone citizenship. This means that voting rights remain a remote possibility as permanent residency status does not bring with it the right to vote in South Africa.

Thirdly, refugees and other international migrants in South Africa contribute to the local economy in numerous ways, including through employing nationals and paying some forms of tax (such as value-added taxation), which results in immigration having a positive effect on the gross domestic product (GDP) per capita (OECD, 2017). Yet, due to issues relating to the accessibility of correct documentation and the non-functional registration system, most refugees and asylum-seekers in the urban space find it virtually impossible to obtain work in the formal sector. A government official at the city level argued that this means refugees are forced to work in the informal market and/or become self-employed:

They can’t apply for jobs in private sector. So self-employed. So move around looking for better opportunities … Some stay more though, especially in the informal sector.⁴

The national government, therefore, does not see much in the way of direct revenue from urban refugees, with numerous de jure and de facto barriers preventing refugees from paying national-level taxes.

In turn, by limiting refugees and other forced migrants to the informal economy, patterns of short-term residency and onward and circular movement between different urban settlements appear. As confirmed by the same government official, individuals have to continually ‘move on’ to find better opportunities:

In and out movement – temporary stay for migrants. Data on migration – hard to track. One day is different to the other. Not even sure how long they will stay.⁵

Comparable observations were made by civil society actors in both Johannesburg and Cape Town, with one NGO employee noting that many refugees and other forced migrants do not see the city as a long-term home; instead, ‘a lot of people come to Cape Town with the idea of just being a midway point to somewhere else’.⁶ From a national governance point of view, the temporary and cyclical nature of the movement in and out of urban areas makes it hard to monitor and plan for this highly mobile population (Landau, 2006). It also heightens varied barriers to access to public services and tax systems.

Ultimately, capacity issues are helping frame refugees and other forced migrants as low priorities for the democratically elected government. Certainly, the ruling political settlement is more interested in assisting the large sections of their voting public who are themselves unable to access essential rights or fundamental freedoms. Nevertheless, the lack of any discernible investment in the mid- to long-term reception of refugees contributes to the broader issues faced by the government in terms of cross-border migration into urban spaces. The absence of post-registration assistance directly impacts the number of asylum-seekers, refugees and forced migrants who work in the informal sector as well as those who are obliged to regularly move between different urban spaces in South Africa in search of work and opportunities. This movement has the potential to create additional tension and instability within state structures at the local and national level, while at the same time resulting in losses in revenue from taxes.

Material factor: the capacity of UNHCR and the global refugee regime in urban spaces

Capacity issues, including a scarcity of resources, are not exclusively a state issue. This section examines the limited financial capacity of UNHCR to engage with the reception of refugees post registration in urban spaces. In 2018, UNHCR estimated its funding gap at USD 4.5 billion (UNHCR, 2018a). The budget for UNHCR Southern Africa followed this broad trend with large reductions occurring between 2012 and 2016.⁷ The shortfall globally means that most funding received is used to respond to emergencies. The ‘refugee situation’ in South Africa is not categorised as an emergency and as such has witnessed large budget cuts. For example, a 52 per cent gap in the funding was reported in 2015 (UNHCR, 2016a). With refugees permitted, at least in policy if not always in practice, to access economic opportunities in the territory, South Africa is seen as less of a priority to the global refugee regime than other states. Providing humanitarian assistance to large, diverse and mobile populations is extremely challenging in an urban environment (HPN/ODI, 2018). Thus, these financial restraints play a key role in dictating the type of work that UNHCR, and its implementing partners, can conduct in urban spaces in South Africa.

Partly because of these funding deficits, UNHCR in-country work is now mainly focused on educational initiatives, capacity building and hosting forums. For example, the agency runs a Protection Working Group (PWG), which is a semi-regular forum including government bodies and civil society, that meets to discuss protection issues.⁸ Yet, with little funding available and the options for formal resettlement limited, the scope and effectiveness of the Working Group was heavily criticised by civil society.

In addition, a small number of implementing partners have a mandate from UNHCR to run its social assistance policy, which includes emergency social assistance relating to food, access to healthcare and education in large urban areas. As a manager of a key implementing partner explained, funds are minimal:

to give you an example, I have funding to assist with rent and food for seven families and seven individuals, that’s not even a drop in the ocean.⁹

Furthermore, due to cuts, implementing partners have had to reserve their remaining funds and resources for new arrivals (that is, a focus on initial registration) and the most ‘vulnerable’ in the urban space. The same manager commented:

with UN funding, we can assist people who have been in the country for less than two years. So it’s basically, I think that [it’s] newcomers and vulnerable groups – as funding gets more restricted going forward. The newcomers will drop out and we’ll only have funding to assist people with disabilities … and unaccompanied minors.¹⁰

As a result, most refugees and other forced migrants in South Africa are unable to access any assistance from UNHCR, or the global refugee regime, post the initial welcome and registration phase.¹¹

Ideational factor: a ‘generous reception’ in urban spaces

This section now switches to examining an ideational factor operating at the national level that influences the policy of non-interference with refugees and forced migrants after the initial registration period of reception, specifically, the notion within relevant government departments that the form of reception already offered to refugees in South Africa is in actuality very generous.

A particular form of conditional mid- to long-term reception exists in urban spaces in South Africa. During public events and closed meetings with civil society, high-ranking officials within Home Affairs regularly commented on the gracious hospitality afforded to refugees by the state. This often came up in the context of comparing South Africa’s ‘generous’ free-settlement approach to reception with the more restrictive approaches seen in neighbouring SADC states. Civil society actors confirmed this framing by high-up officials within government departments who see the maintenance of the ‘open door’ policy, in combination with no refugee camps on the territory, as being highly praise-worthy and generous.

This conceptualisation of reception in South Africa as being expansive and magnanimous is also reinforced at the international level. Following numerous public statements of praise, a high-ranking UNHCR official in the in-country office stressed the generosity of the state (and frequently expressed gratitude to the government) in their interview:¹²

The regime in South Africa has been very good … Soon as you come and claim asylum in South Africa, you are permitted to work, permitted to go to school.¹³

This generosity lens has fed down to the local and sub-local level and into the narratives of refugee and migrant groups in South Africa. Groups representing these categories of migrants regularly adopt varying approaches based on a theme of the ‘good migrant’. Repeatedly in interviews and workshops, refugee and migrant leaders expressed their gratitude for the welcome of the host state. In addition, they demanded (at least in public and in interviews) that their fellow refugees and migrants learn to abide by the laws and find ways to integrate and become useful members of the community.

In contrast, demands for the state to offer services beyond registration or better access to rights contained within the global refugee regime were kept to a minimum – even in private and anonymised interviews. As an example, during an annual meeting of refugee and migrant groups in Johannesburg in 2017, the greater part of the meeting was spent discussing how to stop refugees working in informal markets, rather than engaging in pressing issues around a lack of rights or protection.

The notion that refugees should be grateful for the reception they receive (regardless of how conditional and temporary it is) is not a new phenomenon. Nevertheless, specific to South Africa, this perception of magnanimous reception sees the state as only obliged to allow refugees onto the territory and grant freedom of movement. These acts alone should make the refugees grateful and accommodating guests. In this way a delicate relationship forms, whereby obligations within the urban space fall mainly on the guest. The state has completed its side of the ‘bargain’ at the point of registration, by allowing the refugee onto the territory and granting access to registration procedures. In return, as reception moves to post-registration, the refugee agrees to essentially become an ‘invisible’ guest removed (at least to a certain extent) from the political life of the state.

In the context of South Africa, historical ideational factors add a further layer of conditionality to the mid- to long-term reception of refugees and migrants. As Landau and Freemantle (2017:291) argue, with the continued suffering of black South Africans post apartheid, foreigners are often framed solely as either helping or hindering the goal of ‘economic freedom and transformation for South Africans’. Again, here refugees are constructed entirely as guests whose stay is seen as conditional rather than as equal to nationals (regardless of the time spent in urban spaces). In fact, the reception offered remains dependent upon refugees performing a useful purpose in the urban space. This seemingly one-sided relationship has only been heightened since the recent policy shifts post 2011, with new draft asylum-seeker forms making explicit reference to the potential economic contribution of the asylum-seeker:¹⁴

you can see lots of places where ‘refugees are useful to us’ … The claim is not dependent on persecution, rather who are valuable. In the draft of the form it asks questions on bank account, money, wage slips, all sorts of things that are irrelevant or should be irrelevant. [It’s about] issues of national interest.¹⁵

In conclusion, this framing of refugees as useful and grateful guests removes many obligations and responsibilities from the state, particularly pertaining to protection issues. In its place, a delicate if contested relationship emerges in the urban space between host and guest, whereby obligations fall mainly on the visitor. Refugees are being essentially commodified, whereby their status is more focused on their duties to the state rather than a set of rights or obligations owed to them by the state.¹⁶ As examined further below, this delicate relationship between refugees and state bodies in urban spaces is contested further still, if the number of refugees in those spaces increases to levels perceived to cause instability.

Ideational factor: the global refugee regime and urban refugees in South Africa

The section now investigates how the construction of the urban refugee by UNHCR guides the agency’s response to – and informs the national government’s conceptualisation of – refugees in these reception sites beyond initial registration procedures. As introduced previously, the global refugee regime in Southern Africa regularly equates the ‘refugee’ in urban spaces with independence and self-reliance. In doing so, once a refugee arrives in an urban area there is an assumption that there is little need for protection or assistance.

This framing is based on the premise that if a refugee manages to make it to an urban area in Southern Africa, then implicitly they will have the necessary skills and agency to survive on their own. There is a logic to certain aspects of this, albeit state-centric, interpretation of cross-border/continental movement. If a refugee reaches an urban centre in Southern Africa from ongoing conflicts in East Africa or the Horn of Africa, then at this point in their journey the likelihood of needing immediate humanitarian protection is greatly reduced. Undoubtedly, many of the most vulnerable refugees will remain immobile or seek immediate protection and humanitarian assistance in one of the refugee camps which they are likely to pass on their journey south. Thus, in essence, this line of argument proposes that only the most resilient refugees will ever reach a city like Johannesburg, in the southern-most state of Africa.¹⁷

The result of this viewpoint is that once refugees arrive in urban spaces in South Africa, UNHCR urban refugee policy, for all intents and purposes, ceases to apply. As a case in point, top officials in UNHCR South Africa did not see the 2009 and 2014 UNHCR urban policies applying to refugees in Johannesburg. It is ‘not really an issue in South Africa because now people can live where they want – most people are actually living in urban centres’,¹⁸ the argument being that as refugees have freedom of movement, assistance from the global regime is not required.¹⁹

This construction of refugee movement into the urban space by the agency has led civil society in South Africa to conclude that UNHCR’s approach to urban displacement is essentially that ‘onward movement means being left on your own’.²⁰ A former Department of Home Affairs manager certainly felt this:

If you take initiative and travel further than you should in line with what is expected … then you have too much agency to be a refugee – if you demonstrate [this] agency then you don’t need our help.²¹

This quote aptly illustrates the risk inherent with this type of approach. As highlighted by civil society, when the topic of better access to services/protection for refugees is brought up with government officials, the typical response from the Department of Home Affairs is to reiterate the importance of self-integration.

Furthermore, this framing of refugees in urban areas as essentially having too much agency to be ‘regime refugees’ is inadvertently supporting recent shifts at the national level that are shrinking the asylum space and removing refugees from both the global regime and the national refugee framework. The risk is that as the influence of the refugee regime reduces, a void is created which is subsequently filled by the national immigration framework. Thus, a form of regime shifting similar to that examined previously in the context of Zambia, and re-introduced in the preceding chapter, is occurring in South Africa. As Betts (2009a) describes it, regime shifting is a form of forum shopping whereby a state attempts to address ‘problems’ which normally fall within the purview of one regime (in this case the refugee regime) by addressing them through another (in this case, the broader immigration regime). Whether by design or inference, the national government in South Africa is gradually detaching refugees from their refugee status, and from the refugee regime, by responding to all African migrants through using the national immigration framework. By essentially abstaining from engagement in the urban space and reinforcing the construction of urban refugees as entirely self-reliant, the UN agency is in danger of tacitly affirming this new approach to state-based reception.

Finally, this construction of urban refugees in South Africa as a solution to displacement fosters concerns that echo those that were brought up in relation to UNHCR’s past pushes for self-reliance.²² Similarly to previous approaches, UNHCR in Southern Africa is framing the ‘refugee camp’ as a site of reception and protection in a way that sets it up as antithetical to the ‘urban space’. By extension, this also positions concepts of ‘vulnerability’ and ‘self-reliance’ in opposition to one another. This means that the mere presence of refugees in urban areas is enough to ascribe self-reliance, without the need for intervention from the regime. As such, rather than ‘creating appropriate conditions for refugee self-sufficiency’ (Meyer, 2006:14) in urban areas, UNHCR focuses its attention and resources on providing assistance in refugee camps. By contrast, in these confined spaces, individuals are framed as regime refugees, entirely dependent on aid and the regime. Ultimately, these opposing conceptualisations of the urban space/refugee camp by the UN agency have the potential to profoundly affect how the regime is understood at the sub-regional and regional level.

In conclusion, the relevance of UNHCR in the everyday practice of reception in South Africa post registration appears minimal. By moving to urban areas in South Africa, refugees are understood by UNHCR as essentially no longer needing assistance. This in turn is resulting in refugees being detached from the protection of the international governance regime, or even the national refugee legal framework. In this way, urban refugees are increasingly seen as part of the broader population of economic and illegal migrants. Thus, this approach is adding to the overall shrinking of the asylum space, with refugees running greater risks of arrest and harassment in urban centres. From a state perspective, this urban lens also removes obligations in relation to offering protection to refugees on its territory. Indeed, as set out in the previous section, the obligations at this stage of reception appear to fall mostly on the ‘guest’.

The effect of national-run post-registration reception in urban spaces

This section considers the consequences of the institutional, material and ideational factors set out above, in terms of the role the national government and UNHCR play in post-registration reception in South Africa. In particular, how contemporary approaches to refugee reception in urban areas by key actors shape a refugee’s ability to interact with local communities and economies in an attempt to pursue their own personal and economic aims. The section concludes by reflecting on how this overarching approach to refugees at the level of the urban space diverges from the broader approach seen at the national level to the perceived destabilising effect of large movements of refugees and migrants into cities post 2011.

Firstly, the assumption that, through agency and movement, minimal protection issues exist in cities in South Africa is not supported by the empirical research. This is particularly evident over the last decade, with policy changes at the national level shrinking the asylum space and increasing securitisation of all forms of African migrants in cities such as Johannesburg and Cape Town. Numerous interviews with leaders of local NGOs and migrant groups raised concerns over the general reception and protection being experienced by their clients/members in urban areas. A number of these issues appeared unique to refugees and forced migrants, for example, a lack of documentation or general confusion surrounding the validity of documentation (for refugees and law enforcement officers); issues surrounding opening bank accounts; gaining access to schools; and accessing healthcare.

Police and other law enforcement officers also regularly single out refugees and other forced migrants due to their status, intending to elicit bribes.²³ Experiencing extortion when trying to gain access to an RRO is all too frequent, with an implementing partner of UNHCR noting that their clients regularly have to ‘pay a bribe to [a] security guard to get onto the premises’.²⁴ Finally, in the relation to localised forms of reception seen at the local or sub-local level, a human rights lawyer in Cape Town observed how these forms of urban citizenship do not remove the need for legal protection:

I mean if you … if you’re up at 4 a.m., 3 a.m. in the morning you must, you must walk … and see … in fact people have been camping out there because for them being documented, having an extension of their payment, is literally a lifeline.²⁵

Thus, functioning state structures during the stages of registration and beyond remain indispensable to many refugees in urban spaces in South Africa, even if others prefer to find and rely entirely on alternative localised solutions.

Secondly, due to the attitude of non-interference in urban centres (shared by the national government and UNHCR), responsibility for the protection, support and supply of essential services falls on civil society. In reality, civil society is implementing key elements of the global refugee regime for substantial populations of forced migrants. By way of illustration, a local Catholic organisation in Johannesburg runs a busy shelter for refugee women and children, while also assisting their clients with livelihood projects. Moreover, the shelter regularly receives referrals of refugees in need of shelter and assistance from national-level state entities.

This example underscores two significant points concerning reception in the urban space. Firstly, not all refugees or refugee communities can independently meet their essential needs (including protection) in the urban space in South Africa. Indeed, with obligations imposed on them within the space by the state (including that of being a gracious and ‘useful’ guest), the granting of certain key norms, such as freedom of movement and the right to work, are not sufficient on their own to enable all refugees to achieve self-reliance. Secondly, civil society is replacing the functions and obligations of the state and the global refugee regime by implementing key elements of the global regime at the local level. In this way, implementation is in effect skipping the international and national levels and re-emerging at the sub-local level. This concept of implementation skipping levels is examined further in the second half of the chapter.

Finally, due to the inactivity by the state and UN agencies, combined with the increasingly hostile environment awaiting refugees and asylum-seekers upon arrival in urban spaces in South Africa, a UNHCR implementing partner has started ‘resettling’ refugees to refugee camps in neighbouring states. Specifically, small numbers of refugees are being relocated from Johannesburg to refugee camps in Botswana and Mozambique. The manager of the implementing partner organisation explained:

Now recently I’ve had people coming to me saying, oh they can’t look after their families and they need protection and they don’t get it here. They would like to go to the refugee camp in Botswana and I’m working with [redacted] at UNHCR to see if we can move them there … listen – it is happening lots.²⁶

This type of assistance is aimed at refugees who are struggling to adapt to the reality of a frenetic urban environment in South Africa. Indeed, in townships and informal settlements in Johannesburg and Cape Town where many refugees and other forced migrants move, they live with local communities who also find themselves cut off from any form of assistance or protection. Thus, for some refugees, the prospect (at least in the short-term) of living in a camp where services are provided is seen as the best available option.

Interviews confirmed that this form of sub-regional resettlement is being conducted in an informal ad hoc manner by one UNHCR implementing partner. Put in the context of the total number of forced migrants living in South Africa, this form of assistance is happening on a very small scale (contrary to the implication expressed in the quote above). Nevertheless, this phenomenon, which connects the urban space with the refugee camp in arguably new and somewhat surprising ways, merits further research. Also, and significantly for these purposes, it aptly illustrates the range of long-term issues that exist for refugees in urban spaces such as Johannesburg.

The first half of this chapter has examined the key factors behind why the government and UNHCR have maintained a non-interference approach to refugees in urban spaces in South Africa (beyond the initial registration phase). These causal mechanisms need to also be considered via a democratic lens. Scarce resources mean that the host state has little incentive to divert funds to a (relatively) small section of the population that lacks the right to vote. Thus, refugees have less of a voice at the national level than the voting public, and this public is currently harbouring increasingly strong anti-immigration sentiments. This results in refugees and forced migrants in cities in South Africa having very limited political space.²⁷

The ongoing capacity issues in the urban space, the framing of an already ‘generous’ reception, and the conceptualisation of ‘urban refugees’ versus ‘regime refugees’ all also feed into the changes seen in refugee policy in South Africa since 2011. Indeed, the lack of a structured approach to mid- to long-term refugee reception, and the resulting temporary and cyclical nature of the movement in and out of urban spaces, have reinforced the current perception of instability in cities in South Africa. The delicate relationships that emerge between refugees, state bodies and local communities in this contested space are then challenged further when the number of newcomers increases. Thus, conflicting conceptions at the heart of refugee reception in urban spaces in South Africa interact with each other. The construction of the individual urban refugee (adopted to justify a policy of non-interference by the state and UNHCR) is ultimately being contested by an overarching national approach to cross-border migration which views all African migrants in the urban space through a bifocal security/stability lens.

Decentralisation in South Africa

Decentralisation has spread rapidly across the majority world in the last few decades (Crook, 2003). This commitment to devolving decision-making and financial resources involves local governments managing mandates and budgets, running local government elections, raising taxes and then spending them locally (Smit and Pieterse, 2014). In South Africa, the National Constitution laid out the framework for local government and was followed by the 2000 municipal elections (Wittenberg, 2003). As a result, local government in South Africa retains an element of independence, with cities such as Johannesburg having power and authority in areas such as water and sanitation, and municipality planning. In terms of the reception of refugees and other migrants, most of the services available to them (for example, access to most forms of healthcare and education) falls on the national or provincial government to provide. Nevertheless, the Constitution does give some responsibility to the municipality for the social and economic development of the community (Landau, 2011).²⁹

In terms of the actual implementation of policy based on these devolved powers, the results in South Africa are however quite poor (Koelble and Siddle, 2013). As seen more broadly in the region, this localised power is frequently undermined by financial constraints (Kasim and Agbola, 2017). Equally, at least until very recently, local government in South Africa was dominated by the ruling party (the ANC), which as an organisation remains highly centralised (Landau, 2011). Power has started to shift, however, at the city level, with the Democratic Alliance (DA) winning control of Cape Town in the Western Cape Province in 2006 and then Johannesburg in 2016.

Finally, while not the focus of this section, the provincial level also retains a degree of independence from the national government (as protected by the National Constitution) in South Africa. The nine provincial governments manage key social services such as education, health and social grants. Yet, when discussed, interviewees regularly portrayed the provincial level as merely an extension of the national level and thus it was not seen as playing an independent role in the reception of refugees in South Africa. Nevertheless, when the political party in charge of a provincial government is different from the ruling party at the national level, there appears to be some scope for alternative policies and approaches to refugee reception to emerge. For example, after winning control of Cape Town in 2006, the DA then won the broader provincial elections in the Western Cape in 2009. In interviews, UNHCR officials observed how new working relationships with government bodies at the provincial level in the Western Cape had resulted in improved localised responses to xenophobic violence.

Ideational and institutional factors at the city level

Decentralisation creates the scope for achieving progressive humanitarian and development goals ‘at sub-national levels, even during a period when leadership from central levels is limited’ (Levy et al., 2015:6). Thus, while a national-level approach to a topic or issue (in this case refugee reception) may remain weak, the transfer of power to the local level opens up alternative possibilities. If this devolution and shift of power is matched with contrasting political ideals at the municipal level, then it may be possible to see real practical changes in how refugee reception is implemented in these localised sites.

Notably, from the late 2000s, the City of Johannesburg started to accept international movement (including refugees and forced migrations) as part of the fabric of the city. When referring to this shift in policy, a director of an NGO in Johannesburg suggested that there came a point when simply ignoring migrants or relying on self-integration tactics was no longer seen as a viable option at the city level:

The city has a very important role to play and ignoring a section of the population means that you are not catering for them – and it could be a public health issue – there are lots of issues which [if] you then don’t consider … could blow up in your face.³⁰

This ideational shift in how migration into the city was conceptualised, coupled with the strong personalities of two ANC mayors – Amos Masondo (2000–2011) and Parks Tau (2011–16) – resulted in marked adjustments being made to how all migrants were received in Johannesburg.

Numerous initiatives were created in the late 2000s and early 2010s, which had a direct impact on the post-registration of refugees and their ability to find ways to engage with local communities and markets in neighbourhoods in Johannesburg. These initiatives included setting up a Migrant Help Desk in April 2007. Then, in 2010, the Johannesburg Migration Advisory Panel (JMAP) and the Johannesburg Migrants’ Advisory Committee (JMAC) were both founded. At the time of writing, all three schemes remain active, albeit to varying degrees. JMAP is a forum of non-governmental organisations and city departments that meets monthly ‘to look at challenges that confront migrants’ (City of Johannesburg, 2017). The forum is used to determine where funding from the city’s social funding initiative should be used. These discussions can then get elevated to JMAC, (a higher-level committee) which has the goal of ‘facilitating the integration of plans of the various key departments that have an impact on the migrants in the various spheres of government’ (City of Johannesburg, 2017:2). JMAC then also feeds up to the Mayor’s Office.

The Help Desk was opened to provide numerous services to assist with the integration of cross-border migrants and to reduce the spread of xenophobia in the city.³¹ One of the main roles it undertakes is the provision of information to migrants and refugees on where they can gain access to key services such as counselling, legal advice, housing and health. Significantly, the Help Desk has a policy of not asking the legal status of any cross-border migrant before assisting them. This differs sharply from approaches within the national government, where refugees and forced migrants consistently need to show documentation and evidence of their status. Without providing this they are often unable to access services available to citizens and run the risk of detention or deportation.

By utilising city-level funding, the three initiatives (Help Desk, JMAC and JMAP) have assisted in the creation and coordination of numerous events, dialogues, workshops and other initiatives aimed at improving the reception and integration of all forms of cross-border migrants. In addition, numerous attempts at improving social cohesion between local communities and migrant communities have been undertaken to help fight the increasing number of xenophobic outbreaks in neighbourhoods and townships around Johannesburg.

This willingness to engage and collaborate with migrants on the ground (that is, at the local and sub-local level) is also replicated at the international level. When speaking to UNHCR officials in South Africa it was evident that the UN agency has a healthier working relationship at this level of the state than with specific national-level government departments.

At the local level – Johannesburg – we have had co-operations with municipalities in terms of fighting xenophobia … we do not have a problem there.³²

Finally, several key stakeholders noted the importance of former Mayor Parks Tau in particular, as the driving force behind these initiatives. The former mayor managed to gain sufficient support for these ideational shifts by stressing that Johannesburg was ‘a city built on migration’ and these migrants ‘were here and they needed to be integrated into the city’.³³

The idea … is quite a good thing, the idea of acknowledging Johannesburg as a city built on migration … tagline, ‘how migrants can help the city and how the city can help migrants’.³⁴

It is evident that by the late 2000s – at least in policy and approach – the City of Johannesburg had started to move away from the national-level approach to refugee reception in urban spaces. Indeed, this contrast between the form of reception at the local level and that offered at the national level highlights the lack of a ‘clear coordinated coherent policy, to achieve some kind of social integration’ for migrants or refugees at the national level.³⁵

Continuing contestation

The previous section set out some key ideational and institutional differences in how the city level in South Africa (specifically Johannesburg) has understood and framed refugee and migrant reception, compared with national bodies and structures. In turn, these factors have been the main driving force for new reception policies to emerge in Johannesburg. Yet, the actual reception refugees receive at the city level via local state structures inevitably depends on the successful implementation of these policies. From key informant interviews and attendance at the helpdesk and numerous JMAP meetings, it is evident that there have been several points of blockage that have affected the execution of these city-run initiatives in the urban space. Significantly, the reasons for much of the contestation at the municipality level are a result of similar material, institutional and ideational factors found at the national level.

A mixture of local, national and provincial government departments provide services at the local level. Thus, refugees’ access to services can differ vastly based on who the provider is and on the approach taken more broadly to the inclusion of refugees. This is particularly pertinent to several key public services that should, by law, be available to all refugees after initial registration.

The problem is some of the services are not at the municipal level. Depending on what it is, for example education is provincial.³⁶

As educational services are run and funded at the national and provincial level, a former manager of an RRO noted how various structural factors emanating from those levels create access issues (including exclusion mechanisms) for refugees and forced migrants.³⁷ Based on this division of labour, the city has had only limited success in unblocking certain barriers to inclusion (which have increased since 2011) that regularly prevent refugees from accessing key services such as education.³⁸ Another example is key health services, where the local government in Johannesburg is ‘quite keen to provide services to asylum-seekers and refugees, [but] they have no control over the admission policy of a public hospital’.³⁹

Finally, even when a large city like Johannesburg runs specific services, such as small-scale public health clinics, blockages regularly still occur at the point of delivery. This is often due to material and ideational factors such as capacity constraints in terms of funding and training, or anti-migrant sentiments at the point of service. This was the view of the manager of a UNHCR implementing partner organisation in Johannesburg:

I think at the municipality level, they discuss these things and they come up with good resolutions but you know if you look for instance at health … it’s up to them – to the staff – at a particular clinic, whether they are going to implement it or not.⁴⁰

Due to these institutional, ideational and material constraints, a great deal of the work conducted by the city in terms of post-registration reception (such as the Migrant Help Desk) is focused on orientation, facilitating and sharing of information. Beyond this exchange of information, approval and coordination with national-level institutions (such as the Department of Home Affairs or Health) are usually required for refugees to see a real practical benefit. Collaboration between the different levels of the state is the part of the process where the system generally breaks down. As a case in point, national institutions frequently do not attend the JMAC. As a former employer of the Department of Home Affairs observed, there is an overall ‘lack of coordination at the horizontal and vertical levels of government’.⁴¹

Shift in ideational approach at the city level

In 2016, power at the city level in Johannesburg shifted from the ANC, who had run the municipal council for twenty-two years since the end of apartheid, to the DA party. This change in power brought in a new city mayor, Herman Mashaba. At the time of his election, the city was suffering from several systemic issues concerning poverty, crime and unemployment. Part of his plan to tackle this overall sense of instability in the city was the adoption of a dramatically different approach to the movement of cross-border migrants into the urban space. Indeed, following his appointment there was a striking ideological shift in how all African migrants in the urban space were framed at the city level:

I have seen a big shift recently … there was a recognition that people were here and they needed to be integrated into the city – but that has all changed with the new DA mayor.⁴²

At the core of these policy shifts was a marked increase in the securitisation of all foreign nationals (excluding the highly skilled) and the insertion of xenophobic sentiment into public speeches and policy. Unquestionably, a key message from the Mayor’s Office was the blaming of structural and systemic issues within the city (such as increases in crime, lack of jobs and the lack of housing) on refugees and other African migrants.

In terms of reception policies aimed at migrants and refugees, the Migrant Help Desk was effectively still running in the early 2020s, as well as JMAP and JMAC. Yet core changes in the overall policy at the city level were evident. These included: repeated raids on ‘illegal’ migrants; attempts to change legislation to expedite the removal of migrants from their accommodation, particularly from so-called ‘hijacked buildings’; and stoking up xenophobic violence and attacks (Wilhelm-Solomon, 2017). This change in approach restricts access to the urban space for refugees.

An illustration of this shift in approach comes from an infamous speech covering cross-border migrants given by Herman Mashaba in 2017. In it, the mayor noted, ‘they’re holding our country to ransom and I’m going to be the last South African to allow it. I’ve got constraints as local government, because the national government has opened our borders to criminality’ (as quoted in Mail and Guardian, 2017). This quote aptly demonstrates the engagement with the securitisation of cross-border migrants at the city level. It also shows how the municipality started reframing the movement of migrants into the city as the responsibility of the national government.⁴³ In doing so, this has the effect of removing responsibility from the Mayor’s Office in relation to the reception of refugees in Johannesburg.

This change in approach, unsurprisingly, saw city-based reception on the ground regress further from global refugee regime norms. Indeed, the shifts seen in the national government’s approach to the spatial restrictions of refugee reception were (and still are) being embraced and replicated at the city level (including through numerous community-level exclusion mechanisms). These policies ultimately feed into the impression that a refugee’s stay in Johannesburg is inherently temporary, or worse, illegal. Furthermore, the approach of the Mayor’s Office caused further ruptures between refugees and local communities, which inevitably impacts on a refugee’s ability to successfully find forms of reception at the local and sub-local level and ultimately achieve their own personal and economic aims.

Reception at the city level: a mixed bag

This examination of local-level state-run refugee reception policies in urban spaces reveals some key issues that are worthy of further investigation. Firstly, the section tentatively suggests the possibility of regime-norm implementation skipping levels of the state where political structures are causing blockages or contestation and then reappearing and being implemented at lower levels (where different structures and pressures may exist). Forms of local-level state reception in Johannesburg have been able to reach refugees and other forced migrants who lack the correct legal paperwork – or who simply never attempted to engage with national-level processes. This preliminary finding also raises questions around how ‘the state’ is understood in relation to the global refugee regime and refugee reception and whether ‘the state’ and its interaction with the regime need rethinking/reconceptualising.

Secondly, the section shows how material, ideational and institutional factors that influence state-run reception are not exclusive to the national level. Indeed, through using the ‘democracy-asylum’ nexus, the recent regressive policy shifts noted at the city level were at least partially foreseeable. The appointment of the first-ever DA mayor came at a time of extremely low employment figures, high crime rates and increasing anxiety within the urban population about the perceived instability being caused by the numbers of African migrants in the city. Thus, the increased refugee movement, as permitted by national and local reception policies, created a disruptive presence in the urban space and adversely affected the opinion of the voting public. Concerned with the short-term gains of remaining in power, it is therefore unsurprising that the then mayor adopted approaches similar to the national level by blaming ongoing structural and systemic issues on ‘non-rooted’ persons such as migrants and refugees.⁴⁴

This section contributes to the growing body of literature examining the role of the city as a possible space of improved reception and protection for refugees and other migrants. These new avenues of research are particularly pertinent now, given that states at the national level (both on the continent and globally) are continuing to retreat from international human rights obligations. This preliminary investigation into the City of Johannesburg shows the possibility of finding alternative and improved reception in response to differing pressures found at the level of the municipality. Nevertheless, the chapter also suggests some caution is needed in relation to recent attempts to frame the city space as a sanctuary for refugees. It is evident that the city is not immune from democratic pressures and other material, ideational and institutional factors that are present at the national level. Furthermore, alternative forms of reception at the city level appear to still be nested in larger geopolitical hierarchies at the national level. Thus, the need for vertical coordination between the different levels of the state on key provisions and services creates unique challenges. In sum, refugees in Johannesburg are far from experiencing an entirely different form of unconditional welcome at the city level nor, by extension, full implementation of the global refugee regime.