An enclave entrepôt: The informal migration industry and Johannesburg’s socio-spatial transformation

Abstract

The spatial concentration of production in cities attracts international and domestic labour in ways that change the character and scale of urban space. Drawing on two decades of research on migration and informal trading in Johannesburg, South Africa, this article argues that the global trade in Chinese ‘fast fashion’ interacts with South Africa’s immigration policy, transportation networks, informal trade and established migration infrastructures to transform the city’s Park Station neighbourhood into an enclave entrepôt. Operated and supported by a network of informal logistics services that keep the enclave within but apart from the city, it is exquisitely tailored to cross-border shoppers. At the social and legal margins but at the city’s geographic core, it enables fluidity in an otherwise hostile space; it is at once highly visible and invisibilising. Formed in the shadows of formal institutions and law enforcement, this entrepôt is migrant-driven and serves the needs of people often seeking to remain invisible from the South African state and citizenry. As such, its services are adapted from the infrastructures that service legal and irregular migration in the subcontinent. Unlike ethnic enclaves or neighbourhoods that work as arrival zones, it provides the means to move ‘through’ rather than ‘into’ the city. The entrepôt is a form of migrant space-claiming by vulnerable and mobile people wishing to be in but not of the city. It acts as portal into, through and beyond national territory.

Joburg is a great place, it’s a half-London. Whatever you need you’ll find it here. You can never say, ‘I couldn’t find it in Joburg’.

Maggie, ¹ a Zambian cross-border shopper (2017)

Cross-border shoppers are spending ten billion a year in the inner city? That’s equivalent to the annual turnover of two Sandton City shopping centres. And you know that’s Africa’s biggest and top shopping mall … go tell the mayor that!

Economics Professor Francois Viruly (2017)

Introduction

Human mobility is powerfully associated with the spatial transformations of highly networked urban centres. Whether Chicago’s ethnicised neighbourhoods of the early 20th century (Wirth, 1938), Paris’ contemporary banlieues (Garbin and Millington, 2012), or London and New York’s hyper-diverse neighbourhoods (Aptekar and Cieslik, 2015; Keith, 2005), the movements of goods and people reshape cities (see Brenner and Schmid, 2012; Sassen, 2010). On arrival, people interact with markets and policies that restrict and facilitate patterns of movement, housing, trade and labour. Scholarship on the Gulf States, Singapore and Hong Kong draws attention to the importance of specific labour recruitment policies and local institutions in shaping the movements and impacts of people coming for work (see Iskander, 2020; Kathiravelu, 2016; Ye, 2018). Elsewhere, the focus is largely on supply chains and corporate-led isomorphism. Our paper follows Simone’s call to articulate, ‘between the peculiar, idiosyncratic features of cities and urban regions and urbanisation at a planetary scale’ (Simone, 2016: 8). This approach recognises that across sub-Saharan Africa, formal and social institutions together structure movement and varying socio-spatial transformations. These include trans-oceanic channels involving recruiters, brokers, employers and state regulators, less formal trans-Saharan/trans-Mediterranean migration infrastructures and street-level governance structures (see Brachet, 2016; Casas-Cortes et al., 2015; Wilhelm-Solomon, 2017).

This article argues that informal mechanisms shaping human and material mobility are a particular form of urban repurposing in central Johannesburg. Here, global trade in Chinese ‘fast fashion’ interacts with South Africa’s immigration policy, transportation networks, informal trade and established migration infrastructures. Combined, they are transforming the Park Station neighbourhood into an enclave entrepôt. The entrepôt is operated and supported by a network of informal logistics services exquisitely tailored to cross-border shopping that serves destinations elsewhere in Africa.

Migrant entrepreneurs are active in adapting it, introducing locally relevant, specially designed mechanisms that transform parts of the city into a migrant-run mobility infrastructure. Not only a site of exchange and storage that is in the city but engaged elsewhere, it also protects African migrants who often trade and transit through South Africa without state-issued authorisation (Clifford, 2020): it functionally combines a commercial entrepôt with a migrant enclave. It is the enmeshing of several mobility infrastructures, cartage, storage and accommodation sited at the city’s social and legal margins while remaining at its geographic core. It offers security and connection while enabling fluidity through, over and around social and institutional hostility. It is at once highly visible and invisibilising. It is ethnicised and nationalised even as it separates itself from South African national space. The entrepôt represents a kind of space-claiming by migrants and collaborators to the city that ties to an archipelago of sites spread across the world.

This article inserts socially regulated forms of human mobility into discussions of urban mark-making and transformation. While ‘ethnic enclaves’, ‘ethnoburb’ and hyper-diverse neighbourhoods feature in analyses of urban space (Kim, 2018; Li, 2009; Meissner and Vertovec, 2015; Waldinger, 1993), authors often speak of settled populations, peoples seeking permanent membership or dormitory neighbourhoods filled with labourers on temporary contracts. By contrast, the urban footprints and the channels generated by the migrant-trading hub of central Johannesburg uneasily inhabit the urban migration lexicon. Like Bauman’s (2002)‘nowherevilles’, these are spaces and infrastructures ‘within’ and yet not ‘of’ the city; neither transient nor embedded (cf. Harrison et al., 2012). Much like export processing zones, stock exchanges or airports, these are ‘bridgeheads of the advancing extra-territoriality, or (in a longer perspective) the laboratories in which the de-semantisation of place, disposability of meanings, plasticity of identities and the new permanence of transience’ are tested (Bauman, 2002: 344−345). While deterritorialised, their location matters. Their privileged access to hinterlands gives them a market edge (Feenstra and Hanson, 2004; Hall, 2015).

Our approach builds on the growing recognition that migrant arrival and ‘integration’ into urban spaces are mediated processes (Meeus et al., 2019). The infrastructural turn within the social sciences introduces ideas of the ‘migration industry’ and ‘migration infrastructure’ to the study of human mobility (cf. Xiang and Lindquist, 2014: 122). An industrial framing emphasises the licit and illicit provision of services by different actors that facilitate, constrain or assist and shape mobility at different stages of the migration journey, during initiation (in the home country), passage, as well as after arrival in destination sites (Cranston et al., 2018). Agents include ‘various public and private agencies and actors [that] provide information, products and services relating to migration, thereby promoting, facilitating and organising the process of migration’ (Spaan and Hillmann, 2013: 64, in Cranston et al., 2018: 546). Importantly, these are actors with interests that drive them to build, maintain and transform. The concept of ‘migration infrastructure’ is somewhat broader, referring to the ‘interlinked technologies, institutions and actors that facilitate and condition mobility’ (Xiang and Lindquist, 2014: 124). At one level, such breadth draws attention to everything, thereby undermining its own analytical utility. Yet, as a general heuristic it helps reveal the interactions of mediating agents, migrant agency and their mutually constitutive relations with spatial infrastructures and configurations. This article addresses these relations through the lens of contemporary city making. It demonstrates that informal mechanisms surrounding human and material mobility that concentrate in a hub in inner city Johannesburg represent a particular form of urban repurposing.

This following account draws on almost two decades of research on immigration, informal trade and urban transformation in the city undertaken by the authors and close colleagues (see Dittgen, 2017; Freemantle, 2010; Madhavan and Landau, 2011; Misago, 2011; Zack, 2013, 2017; Zack and Govender, 2019). This analysis draws explicitly from research on a migration industry connecting Zimbabwean women with domestic work in Johannesburg (Zack et al., 2019) and a 2017 enquiry into cross-border shopping in the city (Johannesburg Inner City Partnership (JICP), 2017). The latter included a questionnaire-based survey of 400 cross-border shoppers and 300 indoor retailers (street traders were excluded) as well as key informant interviews with hotel managers, transport operators and shoppers. Initially oriented to assessing migrant traders’ needs and experience in marketing Chinese-made products, the project also highlighted the movements of people and goods into and through the city. It found that 70% of shoppers in a 53 city-block area of the inner city came from neighbouring countries expressly to buy goods to sell in their home countries. Yet, this highly visible churning point of exchange and mobility remains outside official economic calculations or planning processes. It is this do-it-yourself form of urban trade and transformation that this article explores.

An emerging entrepôt amidst Johannesburg’s urban transformation

The Park Station entrepôt is the spatial junction that couples the sending and receiving corridors in this distribution network. The supporting infrastructures include the bus depots and hotels as well as the networks and services such as storage, packaging, loading and security. Their co-location provides accessible pathways for moving the goods, money and people that comprise this trade.

Natasha Bupe’s ² story offers distinctive insight into the necessity and added value of the immigrant entrepôt. Natasha hand-delivers customised men’s business attire in Lusaka. She considers herself an experienced and successful shopper. ‘I used to hear people talk about Johannesburg and they would talk about their shopping here. An important benefit of shopping in South Africa is the fast supply and change of goods’, she says (Interviewee 2, 2017). She adds that Joburg is ‘hot’ and she spends R25,000 on each trip here. Like many cross-border shoppers, she collects money for specific orders before she comes on her bi-monthly trips. In Johannesburg, she maintains real-time communication with Zambian customers to ensure she meets their exact colour and fabric requirements. Using her phone as their eyes, they select from the available new styles of shoes or other goods.

Natasha’s ability to circulate material goods relies on a mobility infrastructure that enables her to negotiate the multiple legal and security challenges shopping in Johannesburg entails. After making purchases, she manually carries goods to the Kwacha Hotel where she stays. It is located adjacent to the Jitotwe bus rank, a mid-block old warehouse structure that has a loading and weighing station and can house two or three buses at a time.

Richard Jali, manager for the several coach lines serving Lusaka and Kitwe from the Jitotwe rank, reports sending five buses a week from inner city Johannesburg to Zambia. Each has 60 seats, often all full. Customers depend even more on the bus than on the hotels. ‘It’s not just about transporting passengers. We store their goods; we help them with customs and we even help them find accommodation. We also have a taxi service for those going to China malls’ (Interviewee 3, 2017). ‘This is helpful because you need to arrange a taxi, because people follow you and steal from you if they see how much you have’, Natasha adds. The bus rank has a room for the packaging and storage of goods that shoppers accumulate during their stay. It also has a larger, out of town, facility where bulk goods or large items can be stored and from where they are sent across the border on trucks.

As these anecdotes illustrate, Johannesburg is South Africa’s economic heart. It is situated in the smallest, wealthiest and most densely populated of the country’s nine provinces (Götz et al., 2014). It is globally networked, violent and extraordinarily iniquitous (Soudien et al., 2019). As a node of commerce and cultural generation, its growth is disproportionately driven by migration. More than half the greater urban area’s population of 15 million were born beyond provincial borders; roughly two of ten outside the country (Soudien et al., 2019).

Emerging from their divergent entanglements and estrangements, Gauteng’s residents occupy what Lefebvre terms ‘polyrhythmicities’ (Elden, 2004; Merrifield, 2006). People moving within the city-region, into it, and sending themselves and materials to areas across Africa make the city what it is. The city’s airport is one of the continent’s busiest but the majority of people from the region travel terrestrially (Statistics South Africa, 2017). On the northeastern edge of the core CBD, the Park Station precinct is the axis of these mobilities.

If the inner city is a gateway to Johannesburg, Park Station precinct is its entry port. It is where local and foreign migrants and immigrants arrive alongside daily commuters and international traders. The precinct sits at the terminus of long-distance and commuter trains. It is where the Rea Vaya, Johannesburg’s high-speed bus network, converges with long-distance and cross-border buses. Almost 6800 short- and long-distance minibus taxis also pass through there daily. A mix of dense living environments and congested trading spaces fill the city blocks enveloping the Station. The area’s short city blocks and narrow sidewalks barely contain the passers-by and trade that regularly spills into the streets (Zack, 2017).

This density of people and mobility has produced Johannesburg’s immigrant entrepôt. Park Station itself and its surrounding city blocks are where customers from Southern African countries arrive and disperse to other locations. Many stay to shop nearby. Highly networked traders buy and move goods – primarily Chinese-made – from there to surrounding countries in what some term informal business tourism (see Núñez, 2009: 9; Peberdy and Rogerson, 2000; Rogerson, 2018: 385). Informal estimates based on the number of shoppers extrapolated from counting buses and passengers, and shoppers’ reported expenditure, place the annual turnover attributable to cross-border shoppers at over R10 billion (approximately US$680 million) (JICP, 2017). This is double the annual turnover of Africa’s largest shopping mall, Sandton City, 15 km to the north.

The supply chain channelling goods through the Park Station shopping precinct works as follows: migrant Chinese entrepreneurs in Johannesburg import clothing, shoes, homeware and electronics from China. Many are resold in Chinese-owned wholesale malls in major cities and border towns across South Africa (Dittgen 2015, 2017). This is a major market with over 20 such malls in Johannesburg alone and others opening in cities and towns across the country (Harrison et al., 2012: 919; Willemse, 2014). The malls are typically secure complexes hosting multiple entrepreneurs each managing their own small shop front. Merchants, including many international migrants, source goods from these malls to resell on the street, in micro shops or in designated inner city trading stalls.

The enclave entrepôt is also connected to small shops and ethno-nationally controlled trading zones in the inner city. It intersects with the so-called Ethiopian Quarter, immediately southeast of the Station, where thousands of tiny shops have been carved out of what was previously larger ground floor retail and upper floor office space (Zack and Govender, 2019). Survey respondents (JICP, 2017) indicate that around 70% of customers in these high-rise centres are cross-border shoppers. Around 85% buy for their own businesses (JICP, 2017). And buy they do. In 2017, they spent an average of R14,364 (about US$1050) on goods per trip. They expended an additional R3497 (US$255) on other services, including transport, storage and accommodation (JICP, 2017). These trips are typically commando-style exercises lasting only two and a half days in this city. The sole focus is on securing the right goods at the right price and getting them and the seller quickly back home.

Shoppers travel to Johannesburg frequently, necessitating a migration industry that enables seamless and low-cost mobility. As 87% sell their goods within 30 days of leaving Johannesburg (JICP, 2017), regular trips are needed and some 40% make the journey monthly. While their time in Johannesburg may be brief, a single trip from Lusaka can take 40 hours, including extended delays at border posts. This generates considerable transaction costs but reduces the risks of holding excess stock and demands for substantial capital investment.

This trade relies on people, not corporations. The entrepôt couples fast-fashion sending (from China to Johannesburg) and receiving corridors (from Johannesburg to multiple destinations). At a material level, the entrepôt provides little added value. It neither manufactures nor alters goods but instead sorts, repackages and sends them on to ‘multiple elsewheres’ (Hanson, 2001; Mbembe and Nuttal, 2004). (Hanson and Feenstra (2001) note that between 1988 and 1998, 53% of Chinese export goods transited Hong Kong this way.) Its distinction from formal entrepôts such as Hong Kong comes from its informality, the insulation from popular and police hostility, and mechanisms of subverting and circumventing legal and bureaucratic obstacles. While its presence benefits state actors receiving informal payments or concessionary access (e.g. metro police, inspectors, revenue and customs agents), it confounds official policy. Its presence inadvertently subverts national and municipal plans and priorities for the continent’s wealthiest city.

The micro retailers are the intermediaries fuelling sub-Saharan distribution networks of fast fashion. Without them the trade is impossible. They rely on logistical support catering explicitly to petty commodity chains traversing sub-Saharan Africa and the Indian Ocean. The dense urbanity that attaches to this function, in the form of legal and informal bus depots, hotels, apartment buildings, markets and an intense, lucrative shopping hub, is constructed in deliberate, if subversive, ways by migrant entrepreneurs and service providers who are repurposing and rescaling the cityscape (Zack and Govender, 2019). This migrant-driven transformation has created the enclave entrepôt.

Repurposing migration industries and infrastructures

Many routes informal traders now follow were pioneered by earlier regional labour migrants and tourists. The industry and transport mechanisms supporting them now are repurposed for the fast-fashion trade. Several interlinked agents and processes constitute the migration infrastructures that enable or hinder departure, journey, arrival and access to employment of labour migrants (Zack et al., 2019). Yet, unlike state-sanctioned transnational migration movements formally organised through elaborate recruitment and placement schemes (see Iskander, 2020; Wee et al., 2018), these are managed by neither states nor formal recruitment agencies. Instead, many moving to and through Johannesburg rely on the entrepôt’s social and material infrastructure as part of a migration industry.

Understanding contemporary migration and the city is impossible without attention to the entrepôt’s mediating and facilitating structures (Coe and Jordhus-Lier, 2011; Cranston et al., 2018, in Deshingkar, 2018). Like migrant enclaves, they facilitate the networks, connections and trust needed for people to work. But as a commercial entrepôt, it shapes movements, extracts rents and mediates competing needs. As in other migration industries, State agents (often operating on legal margins) interact with migrants’ social networks, brokers’ commercial interests, security companies, transporters, and employers and families. Recruitment agencies are neither functionally nor conceptually separate (see Cranston et al., 2018: 545). Even where formal brokers are involved, commercial migration has deep social registers. As Dodson et al. (2016: 3) observe, for the largely informal movements within Southern Africa, the social and commercial infrastructures are all the more important (also Kiwanuka and Monson, 2009).

Even if families are a critical component of successful migration in the region, more distant connections are also necessary for successful migration. As Granovetter (1973) would predict, they provide information about changing regulations, price structures, demands, threats and opportunities. For a price or promise of diffuse reciprocity, they may provide protection. Like enclaves elsewhere, the entrepôt becomes a site for making these connections (see Gijsberts et al., 2012), in this case among transport operators, smugglers, government officials and police officers operating at or beyond the law’s edge. Many facilitators are drawn from a social network, while others are sourced as specialist operators/officials. Often one seemingly specialist operator – such as a transport provider – plays multiple roles in facilitating the movement of the migrant (Zack et al., 2019). Social contacts often connect to specialists. At times operators have both a social and specialist relationship with prospective migrants. Moreover, while brokers may have superficial ties to migrants, they nonetheless can perform highly personal and individualised services.

The entrepôt also services migrants’ material demands. Within a few blocks, large numbers of South African and foreign bus companies – operating from within Park Station, from state owned and managed ranks and from independent bus depots proximate to the station and within the shopping district – service labour migration and cross-border trade. On one day in mid-August, 51 bus companies operated from 19 sites within the entrepôt. In that same week, 465 buses left Johannesburg for neighbouring countries (JICP, 2017). The numbers increase before holiday periods. A large percentage of passengers on those buses were cross-border shoppers. Many of these buses never enter the formal bus station but ‘dock’ on public roads.

The transport companies do more than carry passengers and goods. Traders bring their shopping to depots where packaging, storage and loading take place as traders continue shopping. When it is time to leave, passengers and goods are reunited and swiftly loaded. A number of transport companies link passengers with local taxis to do their shopping safely. They also connect them with guesthouses in the inner city or nearby migrant enclaves. These networks are maintained by head porters and runners working independently or linked to bus depots or hotels. They carry goods from storerooms to shops and from shops to transport and accommodation facilities. Many service providers are migrants from the home countries of the shoppers they serve. The commercial networking is combined with an ethos of familiarity and cultural connection.

The 45-roomed Belong Guest Inn caters largely to cross-border shoppers from Zambia, The Democratic Republic of the Congo and Tanzania. The hotel has a symbiotic relationship with the bus companies and they often share customers. The Compri Point Hotel manager indicated that clients from Zimbabwe, Malawi and Zambia choose his hotel because ‘it is close to buses’. The same hotelier spoke of short-stay shoppers: ‘We sometimes get full. But not all the time. The reason is that many of the shoppers do same-day shopping; they shop and leave on the same day, so our rooms don’t get full every day’ (Interviewee 1, 2017). The Egoli Inn has 125 rooms. Most guests are shoppers from Lesotho and Botswana who stay for two or three nights. Regular customers come once a month. Several managers indicated that customers tend to return to the same hotel on future visits. These logistics facilities share and diffuse shoppers’ precarity. Owners and managers are at risk because of their migrant status, because some are operating unregistered businesses, because of anti-foreigner sentiment in the city, and because their business is tied to the vicissitudes of the market and the security environment of Johannesburg’s inner city.

While some cross-border shoppers use bus services inside Park Station, these cannot offer the packing, storage and accommodation accessed through the entrepôt’s less formalised infrastructure. Moreover, long-distance bus and taxi drivers servicing the cross-border routes may also facilitate shopper journeys (see Chiumia, 2016; Worby, 2010: 427). Drivers indicated that, as well as transporting travellers and goods, they assist passengers in navigating border crossings. They may gather travellers’ passports and have these stamped collectively to save time. They may similarly compile all passengers’ receipts for the goods being carried on the bus to ensure that the payment of border taxes is efficient. Some depots even provide food and a few rooms for accommodation. Others offer waiting space or opportunities for shelter should there be citizen or police hostility outside. Within the environment of the fashion shopping precinct independent bus and hotel facilities co-locate. They exchange information, knowledge and customers.

Trading in information

As with enclaves elsewhere, the enclave entrepôt processes and provides real-time insight and analysis. Incomplete information between buyers and sellers sets up the pattern of intermediaries in Hong Kong (Feenstra and Hanson, 2004). Similarly, in Johannesburg’s environment of extreme competition comparative advantage is gained through information problem-solving. Here, cross-border shoppers in Johannesburg mediate the information exchange function that is core to entrepôts. Small-scale distributors wield significant agency in the market for fast fashion (Haugen, 2017). The micro traders shopping in Johannesburg’s entrepôt play the roles of lead firms in the value network as they act largely as agents for end consumers. They are intermediaries in vertically integrating the supply chain by relaying consumer preference and local demands where larger retailers do not go. While many shoppers are buying in bulk many are also buying for specific clients. Most respondents (48.4%) decide what to shop for by taking orders in their country before travelling (JICP, 2017). Their information exchange happens in the dense shopping environment and in transit. Shoppers said they garner core insights on the buses to and from Johannesburg and while moving on foot through the inner city. Once inside shops, they continue exchanging information. ‘We shoppers know each other. I can ask another shopper where I can get Pampers or jeans or suits for a better price’ (shopper Sylvia, in Zack and Lewis, 2017: 8).

Shoppers inform retailers about what goods are moving well or badly, and indeed also about new product information they glean from customers or social media. Their marketing role is tactile. They touch and examine goods on behalf of end customers. This quality verification role is in line with roles played by intermediaries of Chinese goods that move through Hong Kong. These information exchange roles resonate with theories of intermediation which indicate that intermediaries solve informational problems (Feenstra and Hanson, 2004). But for this system to work, a space of interface and exchange is required. There needs to be a beachhead in the city where migrants can land to shop, collect and distribute. The dense agglomeration of thousands of cupboard-sized shops and their co-located support services, geared to moving fast fashion through masses of transactions made by informal entrepreneurs, comprise this beachhead. It is the architecture of the immigrant entrepôt.

Protection and profit amidst precarity

Navigation of the environment requires another set of intermediaries. Trading practice and mobility and shelter infrastructures are tightly linked. Neither safety nor protection from the law can be taken for granted in this shopping hub. And so a dance with the law and with criminality adds to the complex precarity of the trade. These threats add costs and delays in an industry in which timing is a priority. The objective of petty traders servicing the distribution networks described here is getting clothing to a store or directly to a consumer as quickly as possible (Bruce and Daly, 2006: 330).

The machinations that drive efficiency in the entrepôt services of storage, security and transportation offer more than functional mobility. The service providers who are the hub’s infrastructure and scaffolding interpret the socio-legal and regulatory environment and translate it for customers. This eases business operations in an unpredictable environment where officials and police form and enforce regulations on an ad hoc basis.

Johannesburg-specific services

The entrepôt’s infrastructure is exquisitely tailored to Johannesburg’ specific challenges. To support a wide network of ‘informal business tourism’ undertaken by people who have limited legal rights to enter, let alone conduct business, the mobility and accommodation infrastructures that service Johannesburg’s entrepôt offer additional value. These mobility hubs intermittently shield traders from three additional factors: widespread criminality that makes Johannesburg one of the most dangerous peacetime cities in the world; a hostile citizenry that regularly attacks and discriminates against immigrants and immigrant traders; and a predatory police force that repeatedly exploits informal business people.

Physical security

This cash economy – 95% of shoppers indicated that they transact in cash – occurs on city blocks that extend eastwards from Park Station (JICP, 2017). The dependence on cash poses a big risk to retailers and customers in this area. Over 60% of retailers interviewed had been physically attacked or assaulted in the inner city (JICP, 2017). Almost 40% regularly gifted police officers (JICP, 2017). For shoppers the risk is also extreme. A third of shoppers had been exposed to violent crime in Johannesburg (JICP, 2017). In addition to reports of mugging, there are many reports of corruption with law enforcement officers from metropolitan, national and customs police units implicated in harassment, in the extraction of bribes and in the theft of goods from retailers and customers (JICP, 2017).

First-person travel to Johannesburg remains important in this distribution network. Traders absorb great costs by travelling themselves. Trips to Johannesburg are both product-research and sourcing excursions, with 90% of shoppers comparing prices on foot, usually over the course of a day (JICP, 2017). During these scouting trips they do not carry money to reduce the risk of muggings. Instead, they carry cash only during relatively quick excursions to purchase goods and return them to the bus depots or hotels.

The acute risk of theft or mugging or anti-immigrant attacks (see below) demands particular practices and networks of logistics that are also networks of sanctuary. Many shoppers stay with friends or relatives who are established migrants in the city. Others stay in hotels in the inner city. They may select or be directed by bus companies to associated hotels managed by migrants from their home countries. They depend heavily on the security and storage facilities of hotels and of bus depots. Hotels offer low-cost accommodation and the option of sharing rooms to further reduce costs. Language and cultural familiarity provide modicums of safety. Most shoppers say they do not even use city restaurants at night, preferring to lock themselves in their hotel rooms in the early evening. A few hotels provide restaurants and bar facilities that allow patrons to socialise without leaving the premises.

Legal precarity

The informal movement of goods, as of money, exists in a context where the substantial formalised banking, postage and money transfer systems of South Africa are inaccessible to migrants. Amongst the barriers to the use of these systems are onerous documentation requirements and a general mistrust of authorities. A proof of address requirement is difficult for migrants who may have insecure residential tenure (for reasons such as not being able to sign long-term leases because of their asylum-seeking status or lack of banking accounts). Furthermore, migrants engage only reluctantly with police at any level and requirements such as affidavits from police pose a further hindrance to engaging in formal monetary transfer processes. Similar blockages prevent migrant micro entrepreneurs, who are the main merchants of cheap Chinese clothing, from engaging in online transactions with shoppers from surrounding countries. This places retailers in a precarious and dependent position. It enhances the need for intermediary roles played by cross-border shoppers and the transportation infrastructure that facilitates the movement of shoppers and goods.

Shoppers are also impacted by the migration regime. In terms of the 2002 Immigration Act, foreign traders cannot sell or trade goods with a visitor’s permit, on which the majority of respondents enter the country. This enhances their risk of being confronted and even harassed by police.

This node serves several merchant categories. While up to 16% of retailers (many of them Ethiopian migrants) surveyed in the inner city indicate that they have built up the networks to import directly from China (JICP, 2017) most are constrained by language and by their political asylum-seeking status. This status means they cannot travel and therefore cannot set up the initial contacts required with suppliers and export agents in China. It is easier to source from the many small-scale Chinese entrepreneurs operating from warehouse-sized malls subdivided into small shops. While Chinese retailers also serve cross-border shoppers, the malls’ dispersed locations on the city’s geographic periphery limit the possibility of comparison and raise transport and storage costs. Neither they nor inner-city retailers have access to the countries where the bulk of their goods are sold. Without full immigration authorisation – many African migrant retailers are undocumented, refugees or asylum seekers – personally crossing borders for trade is expensive and risky.

There are layers of regulation in the shopping district, including planning and building bylaws. Often building adaptations do not comply with the regulations the municipality could enforce. It only happens sporadically but, when it does, enforcement is often vicious and prejudicial. This is also true of policing. It may account for why, even in the face of, and multiple cases of, severe and hostile policing in the area, trader Ahmad Gwangul reports that, ‘The main thing we need is to be able to do business. There are many Ethiopians doing business in other African countries, but here is the highest concentration. Because in Johannesburg the laws are easy; you can just open a shop. Even if you aren’t really allowed to open a shop here or to sell this or that, the laws don’t chase you here’ (in Zack and Lewis, 2017: 6).

Official and citizen hostility

The transformations of the Park Station environment, like that of other less formalised incremental investments and alterations made in Johannesburg’s inner-city space, are not mentioned in the chronicles of the upgrading of the inner city since the beginning of democracy. The upgrading efforts and successes are generally attributed to municipal investments in environmental upgrades, private developer investments in the conversion of defunct office space to rental accommodation and the establishment of City Improvement Districts (CIDs) that provide add-on security and urban management services (World Bank Group, 2016). The city has belatedly, and only reluctantly, officially recognised the cross-border shopping area but has done little to curtail actions by police and officials to supress informal trade (Zack, 2015a). As recently as August 2019 aggressive police raids were met with running battles between police and vendors in this retail precinct. These scenes provided the fuel for a wave of xenophobic attacks across Johannesburg’s inner city (Bornman, 2019).

These risks arise against a backdrop of the city’s ambivalence towards economic migrants. Kihato and Landau (2006) depict Johannesburg as ‘a city of mobile strangers’ in which people from diverse places converge and compete, in generative and innovative economic and social ways, but frequently in tense ways and in conflict (see also Hankela, 2020). They also point to a friction that is generated through the transience and lack of rootedness to space that accompanies a mobile resident population (Kihato and Landau, 2006: 6−7). It is a transience born both of migrants’ own ambition and the hostility they experience in the host city. Through it migrants may resist transplantation and may ‘hover above the soil by retaining loyalties to their countries of origin and orient themselves towards a future outside their country of residence’ (Landau, 2009: 205; see also Wilhelm-Solomon et al., 2017). The mobility hubs referred to in this article suggest a tactic of creating a light imprint in the host space that is fundamentally connected to a country of origin. Inevitably it impacts both sites.

And the entrepôt functions precisely because of the system that simultaneously functions inside and outside of the prevailing regulatory and economic frameworks. The migrant footprint of hoteliers, bus depot operators, packers and transporters is necessarily linked into the local territory while simultaneously connected with destination locations. Their familiarity and cultural and language ties to the home countries of shoppers and their connections to local migrant communities as well as their own Johannesburg acuity are the channels for the feedback loops, adjustments, bargaining and complicity that underpin the systematicity of this entrepôt.

Urban transformation from below: Making of a transit enclave

Without the aggregation of hotels, transport, storage and security services, hundreds of businesses and traders would falter. To protect and promote their interests, merchandising outlets, traders and transporters have partially reshaped the neighbourhoods they are in. They have created an enclave that is (relatively) safe for immigrants and an entrepôt at once part of and outside the city and its formal planning frameworks. Migrant traders, transporters, hoteliers and others alter space to their own individual and collective advantage.

As an entrepôt, Park Station precinct’s privileged position stems from its emergence out of the coincidence of shifting trading relationships and immigration of migrant entrepreneurs to Johannesburg, the surplus stock of dense high-quality buildings, the well-connected road transportation to surrounding countries and the migration infrastructures that facilitate mobility along these transnational corridors. The governance of the area, where transgressions of planning bylaws are often not attended to, has advantaged the intense agglomeration. As an entrepôt it offers a hinterland of Johannesburg and surrounds, towns across South Africa and towns and remote locations in surrounding countries. This massive hinterland is the foreland of Chinese goods entering this inland ‘port city’.

The industries described here are necessarily light in physical infrastructure but their logistics offerings are substantial. They are responsive to the local urban and social condition and are geared for the fast-fashion market requirements of dependability, security, low cost and high speed. The infrastructures that attach to this generate and support investment and jobs in a logistics economy and a property market. Indeed, the fast-fashion retail district, fuelled by cross-border shopping, has altered the urban space and even the nature of retail, while dramatically impacting the property market in the inner city of Johannesburg. This value creation comes with costs. Many buildings are overcrowded, services are stressed, alterations are substandard and unsafe. Yet, landlords in this part of town extract rentals that are many times higher than elsewhere in the inner city (Zack, 2013; Zack and Govender, 2019).

The transformations of the Park Station environment, like that of the rest of Johannesburg’s inner-city space, have not been predictable and have only partly taken forms established by the state and capital. Beneath those codified and imposed spatial intentions city reshaping takes place by a subordinate class that alters space to its own advantage and that in ways that de Certeau maintains are neither transparent nor easy to administer (in Harrison et al., 2012).

The Park Station precinct plays multiple roles in the inner city. One of these is as a migrant mobility enclave. Intersected with that is its role as an entrepôt.

Migrant claim-making

Enclaves of transit – in which buses and hotels and related services comprise anchors within micro enclaves designed to move people and goods through a neighbourhood – do not appear in policy or scholarly discussions of migrant movement in Johannesburg. Migrant intersections have reshaped space and meaning in substantial parts of the inner city. Much of the representation of Johannesburg as a migrant city examines places that are apparently ethnicised or that are home to mixed diasporic communities. The literature examines the hosting of such communities and the imprint they make on the urban environment whether in a permanent fashion or as sites of circulation (Dittgen, 2017; Dormann and Mkhabela, 2019; Jinnah, 2017; Moyo, 2015; Thompson, 2016; Xu, 2017; Zack, 2015b).

The entrepôt is not an ‘ethnic enclave’ or ‘ethnoburb’ as typically described in the academic literature. It is a portal into the city but those creating and using it make no claims to stay. Rather, like many others arriving in Johannesburg, they seek usufruct rights (see Landau and Freemantle, 2016). In this case, the right to trade. Spatially it is a claim to a space of motion. It facilitates the mobility of people and goods and knowledge. It does so by creating and recreating a temporary ethno-national presence within the entrepôt, as people from surrounding countries move through these trading post infrastructures. Yet, it is also a claim to property – some more formalised in hotels, and others existing beyond the discipline of planning laws as unregistered bus depots or on-street ranks of long-distance vehicles. The enclave is accompanied by a footprint of appropriated space, of identity enclaves, of adaption of space and of improvisation. Its infrastructure is focused on throughput, with a whole industry catering to that throughput. The apparently precarious and light mark-making of the enclave belies its meaning in the consciousness of traders who spend billions in the inner city. The spaces and activities of the mobility enclaves of the inner city are linked to cognate space elsewhere in the city (China malls). Their users are committed to return and the industry is committed to enabling and is, indeed, depending on that return.

Conceptualising the enclave entrepôt as a mode of migrant mark-making resonates with Simone’s reflections that the ‘pushing and pulling’ of local forces, encounters and opportunities ‘shapes the architectures of articulation that enable capital to shape-shift its way across new spatial parameters’ (Simone, 2016: 9). This entrepôt, in which goods come in and move out in different ways, particularly in ways that build a ‘massive small’ economy (Campbell, 2011) based on the individual transactions and strategies of multiple petty traders, is dependent on Johannesburg’s unique port-like characteristics for this distribution function. The wormholes that are created through the city by this activity are the spatial corridors and micro-nodes that comprise ecosystems of trader mobility, of entering and leaving and re-entering the city many times in a year.

It connects local and transnational and even global trading networks. It works across scales, using the services of large infrastructure and institutions – borders, roads, etc., and the smallest of infrastructures – small shops, small hotels, micro businesses. It innovates – whether through providing alternative services for money transfers, or combining various services on one site, or adapting transportation services to navigate policing or immigration procedures, or in the movement of persons, goods, remittances and passports for settled migrants. This transit enclave connects to other places and connects to the migration industry. It inserts a Zimbabwean/Zambian/Mozambican/other presence in the city but not as labour. Its operations are at once extractive and protective.

It provides services that extend the length of the passage from retail outlet to destination in the home country. It provides levels of security that the state cannot provide, because the state itself is a source of insecurity. The impact is a shopping and business district that has a distinct flavour – because of its diasporic nature and its cross-border trading nature.

This service function creates a temporary ethno-national presence at various sites in the inner city. People use these ethno-national spaces in order to minimise their travel and transaction costs and transaction costs of knowledge transfer. The entrepôt and its multiple anchors and services connect people who are on the move – shoppers from various countries – with people who service the trade and its supporting infrastructure. It sits outside of South African society and trade, and alongside it. Those who come here know where to go because of the social connections they make.

Johannesburg’s migrant mobility enclaves are spatially transformational. They have repurposed space and established substantial if flexible and lightweight transit networks. The mobility enclave that is produced incrementally is an insertion into the city and a territorial claim that approximates Benjamin’s (2008) framing of an ‘occupancy urbanism’ in which poor groups claim public services and safeguard territorial claims that offer de facto tenure. The tenure to sites occupied in this way remains highly precarious. Unlike the Indian cases cited by Benjamin, the exchange-based relationships that service providers in Johannesburg’s enclaves enter into with state actors are not those of offering votes in exchange for permanent land tenure. Rather, their alliances are relationships with police and other bureaucrats who might offer access to public services and the security to trade in exchange for cash or goods. They are not catered to by city regulations in a city that, for all its features of contemporary African urbanism, retains a planning regime that does not accommodate – and often criminalises – the responsive and fluid urban practices so critical to sustaining economic activity for the urban poor. Nevertheless, they embody the capacity to help marginal groups within the city and visitors to the city to appropriate real estate surpluses in support of micro enterprises that are themselves precarious and considered ‘informal’. Objectively they fuel a massive economy via a new and productive role for this city as a sub-Saharan African port. Spatially they contribute a light, responsive architecture. The logistic services described in this paper deserve acclaim alongside the mention of public service and public environmental upgrades that are credited with supporting rejuvenation of the property market in the inner city.

Conclusion

There are few more strategic locations for Southern Africa’s regional trade than Johannesburg. Yet, both Johannesburg’s role in a global exchange network and the Park Station precinct – with its associated migrant-driven mobility and accommodation infrastructures – are unrecognised. The entrepôt’s location within global production networks begs that urban studies deepen the empirical work that might respond to Robinson’s (2002) call for a more cosmopolitan urban theory that does not relegate ‘developing’ cities to difference and to otherness. Moreover, it invites further reflection on leading roles played by apparently informal workings within cities.

In their belonging and placement in Johannesburg, and their simultaneous connections to surrounding countries both through visiting traders and through a migrant-rich wholesaling enterprise, the Park Station entrepôt and the minor migrant mobility hubs in surrounding neighbourhoods resonate with Bauman’s ‘nowherevilles’ in an interesting way. They are neither the trapped places of refugee camps nor the elite stopovers of globetrotting elites that are both products of globalisation (Bauman, 2002). Bauman finds these to be similarly extra territorial, in ‘their not-truly-belonging-to-the-place, being “in” but not “of” the space they physically occupy’ (Bauman, 2002: 344). Rather Johannesburg’s migrant entrepôt – and in particular the cross-border trade elements of that entrepôt – combines the efficiency and sophistication of the transnational movement of shoppers and goods with the entrapment of uncertain refugee status, of xenophobic threat. Because the transit enclaves that have been established through processes generated by migrants and asylum seekers also contain all the precarity of migrancy in Johannesburg and all the vicissitudes of informal trade. They also demonstrate that migration infrastructures in this context linger and morph and remain significant long after the arrival point that literature often signifies as the endpoint of migration.

This article speaks of the need to shift the discussion of migrant and migrant industries away from migration and labour, per se. It also provokes a discussion of migration infrastructures beyond their roles in landing migrants in destination countries. It suggests that attention should be paid to the kind of co-production of the spaces in which migrant industries operate and the structure of the industries themselves. Indeed, migrants and the people associated with moving them and goods structure their operations to insert themselves in particular spaces and provide a services suite to sites. While recognising the risks associated with these developments, it matters to explore their transformative potential in the contemporary, highly mobile, urban condition. In the case of Johannesburg, these impactful zones are sites not only structured by trade but also by the combined insecurity associated with precarious labour, pernicious policing and peddler-targeting criminals. They are part of entry points to a new permanence in the landscape or, maybe, after Bauman, a plasticity of identity and of place-identity, neither framed by the semantics of transience nor of permanence.