When home is elsewhere: housing policy challenges in the context of transit migration in Ghana and South Africa

Abstract

In the last few years the relevance of transit migration was corroborated for many African countries. Oftentimes, it is related to urban migrants’ subsisting linkages to their areas of origin, other times by the necessity or aspiration for onward journey. Based on recent research, this article examines the consequences of transit migration for the urban housing market and related challenges for sustainable housing policies in Ghana and South Africa. While in Ghana governmental housing hardly considers the urban poor, in South Africa a substantial post-apartheid subsidy system explicitly targets the lowest-income population. This article argues that neither approach adequately addresses transitory housing requirements related to mobility. Whereas the informal housing market can and often has responded with pragmatic solutions, it does not provide the maximum benefit to migrants. Therefore urban governments should enhance options in this increasingly important and highly dynamic housing sector recognising and enhancing informal approaches.

Keywords:

• transit migration
• housing policy
• housing subsidy scheme
• urban development
• Ghana
• South Africa
• sustainability

Introduction

Based on case studies in Ghana and South Africa this article considers migration dynamics from a spatial perspective. It focuses on ‘transit cities’ as places that are growing in importance due to increasing temporary migration and translocal forms of life, processes that subsequently convert them into ‘targets’ and passageways of migration. Following Marconi (2009), we define transit cities as places ‘where a multitude of networks converge and intersect, creating a wealth of opportunities for migrants’ (p. 9). From the perspective of migrants, transit cities can aptly be described as places ‘in which you get by, you look for future perspectives without excluding the possibility of an onward journey’ (Marfaing 2011, p. 71¹). We thus understand the term transit migration in a more encompassing sense than recently suggested by the strongly politicised focus on African migration directed to Europe (Marfaing 2011, p. 71f.; Collyer et al. 2012, Müller & Romankiewicz 2013, p. 15).

Against this background the article examines how dynamics of temporary migration manifest in the housing market and what this means for sustainable and pro-poor-oriented urban policies. Based on case studies in cities in Ghana and South Africa we show that strongly diversified migration processes like temporary, circular or step-wise migration have a distinct impact on people’s housing needs. These are as yet primarily catered for by the informal market through a variety of temporary housing forms and arrangements. Despite important limitations of these informal responses, we argue that ‘formal’ policy and planning actors should recognise and build upon them for adequate poverty reduction strategies for transit migrants.

Ghana and South Africa make up interesting cases of comparison since they exhibit common features, but also important differences with regard to migration and housing. In terms of commonalities, both are politically stable and economic ‘power houses’ in their respective regional contexts (Turok 2013). Moreover, both historically possess strong albeit distinct migration dynamics, i.e. North–South survival mobility and trade flows in Ghana and imposed rural–urban seasonal migration during apartheid in South Africa. Recently their role as centres of regional cross-border migratory systems of persons that follow behind job opportunities has been accentuated. Neither of the two countries has developed an adequate answer to migration and its impact on local housing development as of yet. Regarding differences, the two countries are quite opposing with respect to their policy approaches to low-income housing: While in Ghana a market-based approach predominantly relying on private delivery geared to higher-income groups hardly considers the urban poor, in South Africa a substantial housing subsidy system through the post-apartheid housing subsidy scheme explicitly targets the lowest-income population.

This article is organised as follows: The next section discusses the data and methodology underlying this article. Then it presents key concepts of our research, which are instrumental for the understanding and analysis of the case studies. A subsequent section provides an outline of the respective housing policy context in Ghana and South Africa and their degrees of poverty orientation and consideration of mobility. In a next step a variety of temporary housing forms as well as case study-based examples of some of them are presented. The last section closes with perspectives for future sustainable and poor-led urban growth envisaging options for migrants.

Data and methodology

The findings presented in this article form part of the results of a pre-study about ‘Non-permanent migration, translocality and governance in transit cities’, which was conducted between May 2012 and July 2013 in Ghana and South Africa, respectively. In that period both countries were visited several times with the aim of building up research contacts with key informants. Moreover, the field trips were used to find indicative evidence on the ground about contemporary migration patterns and urban as well as housing governance in transit cities.

The methodology involved interviews with representatives from Ghanaian and South African universities and research institutions, staff members of government agencies on the local, provincial and national levels, international organisations and institutions such as GIZ, UNDP, EU and national NGOs. In addition, more than 60 short interviews were conducted with migrants in their daily habitat, i.e. informal settlements, townships, markets and inner city areas with a strong presence of migrants. The interviews were analysed by means of a structured content analysis, in which key categories extracted out of the research questions (predominant forms of migration, formal and informal forms of governance, implications for future urban policies and planning, etc.) were used as main categories of analysis.

Conceptual framework

This section discusses some key factors that we assume influence the relevance of transit migration for urban housing in Africa. These comprise the rising significance of temporary migration processes, urban government performance bottlenecks with respect to migration and related housing needs, the increasingly important role of informal modes of urban and housing governance and the limitations of territorialised spatial planning within fixed administrative boundaries.

Growing relevance of temporary migration processes

For a long time, migration was considered as a one-directional move from the countryside to the city. Individuals and households seeking a better life by migrating to the urban labour markets were believed to be leaving their hometowns and villages for good (Schmidt-Kallert 2009) although the literature has also acknowledged temporary labour migration coupled with urban segregation in colonial Anglophone Africa and subsequently apartheid-South Africa (Wentzel & Tlabela 2006; Adepoju 2006; Steinbrink 2009). In the last 20 years, however, it has been recognised that migration processes have further diversified, and that temporary migration has become more important (Deshingkar 2005; IOM 2008; Schmidt-Kallert 2009). In many African countries seasonal or circular migration processes between rural and urban areas dominate with respect to permanent rural–urban migration (Lynch 2005; Collinson et al. 2006; Greiner 2008; Steinbrink 2009).

Seasonal migration in West Africa has a long tradition shaped by nomadic life and trade, the mining industry and cash-crop cultivation (Konseiga 2005, Adepoju 2006, Dick & Reuschke 2012). In South Africa, since the 1920s, circular labour migration was a cornerstone of the segregatory urbanisation system and, later, since 1948 of the apartheid legislation (Lynch 2005; Steinbrink 2009; Posel 2010, p. 129). Today’s circular migration processes in Africa are however mostly based on the need for securing multi-locational livelihoods of poor populations in the context of urbanisation without production-based economic growth (Lynch 2005; Kombe & Kreibich 2006; Greiner 2008; Steinbrink 2009; Turok 2013), partly resulting in slowing down or even counter-urbanisation trends (Potts 2008, 2009).

The combination of ‘urban’ with ‘rural’ sources of income and the preservation of ‘rural’ social relations, networks and lifestyles can thus predominantly be attributed to material constraints of peoples and households. In economic terms, such multi-locational livelihoods offer the possibility of making use of rural–urban cost differentials, despite increased mobility costs. However, migrants’ linkages may also be related to perceptions of sociocultural belonging. In Ghana, among other factors, they are a consequence of customary entitlements to land and housing of the extended family; the larger part of the urban population of all social strata does therefore not feel at ‘home’ in the city, but rather in their (oftentimes rural) regions of origin (Dick & Reuschke 2012, p. 189; Landau 2012b, p. 9).

More recent studies focusing on the disaggregated household level show characteristic patterns of mobility that are linked to life cycles. They indicate that it is mostly young employable household members, increasingly girls and women, who move (temporarily) to the cities (Beauchemin 2011, p. 57; Tamanja 2011/2012; Floquet 2013, p. 31). Meanwhile, small children and elderly people often stay in the rural areas (Tacoli & Mabala 2010).

Another form of temporary migration that has become more important, also in the context of a growing internationalisation of migration, is step-wise migration. Here the rural–urban migration to so-called ‘gateway cities’ is often a preliminary step to the (planned) cross-border migration. These gateway cities tend to also be passageways of cross-border migration and trade networks. For instance, Bakewell and Jónsson (2011) have identified the Ghanaian metropoles and market places Kumasi and Accra as nodes of international trade mobility. Besides regional trade with the neighbouring countries, transcontinental relationships with China and the Arab Emirates are becoming increasingly relevant for these cities (p. 5). For the South African province of Gauteng, in which the international metropolis of Johannesburg as well as the national capital of Pretoria is located, Landau (2010) posits that there are increasingly complex migration patterns, which are characterised by an overlap of circular, permanent and step-wise migration paths within the metropolitan area: ‘For reasons of location, infrastructure, intention and experience, the province is as much a place of transit as destination’ (p. 8). According to Landau, aspirations for passage, profit and protection constitute overarching motivations for these migratory moves (Landau 2012a, p. 229).

Inadequate answers of urban governments to migration

The governance of migration processes in many African cities often is characterised by a multitude of shortcomings. It is noted that weak urban governments are seeking to mould public regulation into static and predictable forms, which do not reflect demographic dynamics. City governments and administrations tend to not consider them as their responsibility, not least due to shortages in financial resources and staff. If at all, the responsibility for migration is attributed to the national level (Landau 2010, p. 3) that however hardly knows the reality on the ground. In addition to the scarcity of resources, the management of migration processes is hampered by a lack of information and knowledge about the relevant migration dynamics and their implications in their respective urban context. In addition, urban development planning and budgeting are based on the registered population, which in most cases does not reflect demographic dynamics through informal and transient processes of urbanisation (Landau 2010, p. 3; Zoomers et al. 2011, p. 497f). In South Africa, for instance, the large time lag of the budget allocation process through the Local Government Equitable Share system² and lack of coordination between different levels of government and administration represent major challenges (Landau et al. 2011a, b).

Additionally, on the governmental level a problem-oriented perspective on migration dominates. First and foremost, migration is seen as a burden on urban housing and labour markets, but also on public infrastructure, utility services and image. This is even more the case for temporary migration processes, which tend to be regarded as menacing and unforeseeable, as aptly expressed by a statement of the local planner of a medium-sized South African town:

‘If we could, we would help everybody, but it is difficult because you cannot get information on these people because they do not have jobs and do not necessarily intend to stay here. You can’t provide them with housing. Before they can see the house they are gone again.’ (quoted in Landau et al. 2011b, p. 92)

Migration processes are thus rarely perceived as a poverty reduction strategy of households worth supporting ([IOM] 2008, p. 188; Inkoom 2008, p. 14). On the contrary, both in Ghana and in South Africa there are no state programmes for the support of (temporary) migrants – rather, urban development and migration are treated as separate policy areas. The reason for this lies in a ‘normative bias towards stable populations’ (Landau et al. 2011b, p. 93) and, if at all, concerns with handling urban ‘entry’ or ‘insertion’ rather than transit (Yankson & Bertrand 2012). Additionally, temporary migrants rarely form part of local consultation processes and arenas of participation within municipal development planning; non-citizens are sometimes excluded by law (Yankson & Bertrand 2012, p. 42; Landau et al. 2011a, p. 37, 2011b, p. 91f.).

The important role of informality

The situation of (temporary) migrants in African cities is aggravated by an adverse attitude or a ‘do nothing approach’ (interview with metropolitan planner in Kumasi, Ghana) of urban governments towards informal settlements. Both are at odds with a pro-poor development agenda, which incorporates the needs and priorities of communities. In most African countries informal settlements are the most manifest expression of city-bound migration (Landau et al. 2011a, p. 38) and important passageways for transmigrants (Yankson & Bertrand 2012; Yankson 2012a; Landau 2012b, p. 15). In accordance with international ‘good practices’ of urban and housing development, both Ghana and South Africa have formulated laws and initiatives of in situ informal settlement upgrading (GoG 2010, p. 69; Tipple 2011, p. 132; GoSA 2015). However, their implementation tends to be slow if occurring at all. In both countries, the eradication of informal housing structures and displacement of their inhabitants are occurring quite frequently, particularly in the case of conflicts with powerful economic interests (so that a ‘do nothing approach’ to residents usually means the lesser of the two evils), resulting in significant gaps between a partly progressive policy discourse and mostly repressive urban practice (Huchzermeyer 2009, p. 64f.; Pithouse 2009, p. 9ff.; Topham 2011, p. 17; Marais & Ntema 2013, p. 87; Turok 2013, p. 150).

The everyday management of temporary migration and multi-locality, e.g. in the area of housing, is therefore primarily based on ‘informal’ support of mostly origin-based migrant networks (Inkoom 2008; Steinbrink 2009; Schmidt-Kallert 2009). Although showing significant limitations, these networks also possess some positive elements. They tend to be unbound by urban or provincial government boundaries, as well as sectoral borders, and therefore possess a higher flexibility than statutory institutions with respect to their geographic and topical range of action. One may thus argue that, by catering to developmental needs of people and communities ‘in transit’, they effectively practice multi-sited or translocal governance, directed to both places of origin and destination of migration. Governments rarely acknowledge, let alone incorporate, these positive elements, but tend to see these self-organisations of the urban poor as antagonists of formal planning expertise and control processes (Pithouse 2009, p. 4; Huchzermeyer 2009, p. 63).

Regarding limitations, ‘informal’ migrant networks tend to provide support for the socio-economic and/or political needs of their own people, thereby leaving the situation of other vulnerable groups unabated (Choi 2011/2012, p. 7). Furthermore, they entail relatively low levels of social capital transfer and possible upgrading or ‘formalisation’ options. In certain cases this support may be brought by using coercion and even violence (Saunders 2010). Lastly, a fragmentation of group-related, ‘informal’ support regimes and their highly differentiated norms might pose serious challenges to sustainable and inclusionary urban governance approaches (Saunders 2010; Harber 2011; Landau 2012b).

As a consequence, in the authors’ perspective, urban governments and planning authorities should acknowledge these various forms of informal ‘bottom-up’ contributions while at the same time addressing their limitations, and use them for a critical reflection and adjustment of the territorialised and sectoralised planning practice.

Limitations of territorialised planning

Regarding migration, one of the main problems of urban governments is their container-based conception of space, according to which social processes can be adequately analysed and addressed within pre-given geographic or political-administrative units such as nation states, provinces, municipalities and districts. Although this is appealing from an administrative perspective, in the view of the multiplication of social and spatial engagements and practices transgressing fixed borders, an increasingly dynamic understanding of space has evolved in recent years. From this perspective, space is constituted by economic, social and cultural practices and conceived of as a networked structure (Castells 1997).

Given the often multi-locational nature of people’s livelihoods, local development and governance processes need to be conceptualised in the light of increasing migration-related linkages and economic as well as political flows with other places. In contrast, a purely localised approach to development is insufficient in situations in which ‘local’ opportunities and risks are to a large extent the result of governance networks acting through different geographic scales and sectors (Zoomers & van Westen 2011, p. 377; Zoomers et al. 2011, p. 492).

Municipalities themselves are progressively coming to the insight that networking on several spatial and organisational levels leads to better results in the steering of urban development (e. g. Government of Ghana 2010, p. 70). For instance in South Africa, in the context of the local government reform of 2000, the number of municipalities was considerably reduced (from 800 to 283) and municipal borders enlarged, among other reasons, to adjust administrative borders to economic realities and promote regional networks. However, in the vast majority of cases the issue of migration is not considered, and even less regard is taken of the specific housing needs of temporary migrants.

Housing policy context in Ghana and South Africa

This article focuses on the consequences of transit migration for the urban housing market and sustainable governance processes. Housing constitutes a basic condition for migrants’ access to urban opportunity structures such as employment and education. Furthermore, it is a central element of individual and collective poverty alleviation in Africa’s urbanising societies. The relevance of urban housing provision also manifests in the fact that the larger part of the poor and mobile population is living in informal and often precarious housing conditions. In Ghana, about 90% of the urban housing units have emerged through informal processes (Tipple 2011, p. XXII; CAHF 2012: 70f.; Gough & Yankson 2011, p. 795), while in South Africa the current census 2011 reports that 14% of all housing units were established informally and further 8% were traditional dwellings (Stats SA 2012). For the two countries the statement can be made that ‘politically informal settlements are a hot potato’ (2012 Interview with African Centre for Migration and Society, 2013 Interview with Johannesburg Housing Department) since no clear standpoint towards them has as yet been developed.

In many African countries, national and urban housing policies – if in place – are formulated without considering the needs of the mobile and transitory population who find themselves in a particularly vulnerable position. Apart from the aspects already mentioned (sedentary bias of governments, adverse attitude towards informal settlements and modes of control and organisation), this is due to a strong contrast between the need for mobility of people imposed by economic and labour market shifts in increasingly globalised African space economies (Zoomers, van Westen 2011; Turok 2013) and the immobility of housing provision. As the examples of South Africa and Ghana show, this applies both for countries with in principle pro-poor housing policies, programmes and institutions and for countries in which these are non-existent.

Housing context in Ghana

In Ghana, unlike for example in South Africa, there is no National Housing Policy in place, although a draft version has currently been prepared by the Ghanaian Ministry of Water Resources, Works and Housing in order to address the country’s housing deficit. The reasons for this deficit are complex, but can be attributed to rapid urban population growth accompanied by a slow housing supply and poor housing delivery mechanisms (with respect to processes of land transfer, availability of housing finance and construction costs), which satisfy only a small proportion of the demand. The result is overcrowding and a growth of informal settlements: ‘Current estimates indicate that the country needs at least 100,000 housing units annually while supply is estimated at 35% of the total need’ (GoG 2010, p. 68; Ofori & Ayivor 2013, p. 255).

Generally, the Ghanaian affordable housing supply for low-income households shows a variety of institutional and ‘practical’ problems. First, in the last 30 years and in accordance with the international discourse, the housing market was increasingly privatised. Since then government institutions and state housing companies have hardly had a direct impact on urban housing, except for providing incentives for private companies by ‘enabling’ investment frameworks. Poorer population groups are hardly reached, despite contrary statements of intent (Arku et al. 2012, p. 3179). Before the start of the privatisation in the early 1990s, public housing corporations have offered housing primarily for civil servants (Arku et al. 2012 p. 3180f.). The current construction activities of large developers exclusively benefit members of the urban middle and upper classes (Afrane & Asamoah 2011, p. 71; Tipple 2011, p. xxvi; CAHF 2012. Yearbook 2012, p. 71; Yankson 2012a, p. 166, 2012b, p. 195).

Overall, the quantitative and qualitative housing deficit is high. In numeric terms it has been estimated to stand in excess of one million units (Arku et al. 2012, p. 3178; CAHF 2012. Yearbook 2012, p. 70f; Gough & Yankson 2011, p. 795). Housing tenancy arrangements dominate ownership by far; however, the rental sector has been largely neglected after the abolition of government rent controls in the 1980s. Owing to the growing demand, the scarcity of housing land in the agglomeration areas (much of which is controlled through customary tenure) and high construction and financing costs, rents have risen sharply (Afrane & Asamoah 2011, p. 83; Arku et al. 2012, p. 1378; Yankson 2012b, p. 195ff). Since formal housing supply is beyond the rajority of the Ghanaian population and due to the (self-admittedly) weak capacity of government agencies to provide adequate housing facilities,³ about 90% of urban housing units emerged through informal processes (Tipple 2011, p. XXII; CAHF, 2014, p. 70f; Gough & Yankson 2011, p. 795), ‘produced by numerous small builders and individual owners’ (Yankson 2012b, p. 166). Not least for this reason, e. g. in Accra, the majority of urban residents live in ‘overcrowded, deteriorated and low-income rental accommodations without proper sanitation, roads, drainage, water supply or waste disposal systems’ (Afrane & Asamoah 2011, p. 83; Arku et al. 2012, p. 3178).

Housing context in South Africa

In 1996, after the end of apartheid rule in 1994, South Africa approved a new constitution, which is considered one of the most progressive in the world. It comprises a ‘bill of rights’, which includes, among other aspects, a qualified right to affordable housing requesting the state to take reasonable measures within its available resources (Article 26). Since that time a number of housing policy initiatives have been introduced that have yielded large welfare state inputs in the form of massive subsidised home building, mostly outsourced to private developers.

Well-known and in quantitative terms the most important is the post-apartheid project-linked housing subsidy scheme developed in the context of the Reconstruction and Development Programme (RDP). The subsidy scheme was continued within the Growth Employment and Redistribution Strategy (GEAR), which replaced the RDP in 1996.⁴ Under this subsidy programme approximately three million residential units had been provided by May 2013 at no cost to its beneficiaries (website SouthAfrica.info, 2014, website CAHF). Over time, new housing policy laws and instruments were added, such as the Housing Act in 1997, the initiative ‘Breaking New Ground: A Comprehensive plan for Developing Sustainable Human Settlements’ (BNG) in 2004 and the Social Housing Act in 2006. All of these initiatives are designed to facilitate low-income households’ access to housing markets. The access to affordable housing was meanwhile complicated by sharply rising prices, e.g. between the years 1997 and 2004 the average house price in South Africa more than tripled (United Nations 2012, p. 5).

Despite all of these policy initiatives, their implementation has faced important challenges. In quantitative terms, housing demand still largely outstrips the supply. It is estimated that the number of informal settlements increased from 300 in 1994 to 2700 in 2010 and, within the same period, the housing deficit in the affordable housing segment augmented from 1.5 to 2.1 million units (Centre for Affordable Housing Finance in Africa 2012, p. 125). With the qualification criteria of a monthly income of no more than R3 500,⁵ approximately 60% of South African households are eligible under the so-called RDP housing subsidy scheme, many of whom spend a long time on waiting lists (Tissington et al. 2013). Households that fall short of the funding under the housing subsidy scheme because of their higher incomes are another problem, since hardly any housing programme is available for them (view Centre for Affordable Housing Finance in Africa 2012, p. 124f). In qualitative terms, the housing approach of the first post-apartheid decade focusing on new construction of free-standing units was even less able to comprehensively solve the problem of inadequate access to affordable housing. Key dimensions such as housing quality, location and ownership were neglected, and communities or their organisations were hardly involved in housing design and delivery (Huchzermeyer 2009; Pithouse 2009; Marais & Ntema 2013). Against the background of these shortcomings, in the BNG initiative in 2004 strong emphasis is placed on the participatory development of socio-spatially integrated and sustainable settlements (Website BNG). These principles are however, up to this time, only hesitantly being accounted for in practice (Huchzermeyer 2009; Pithouse 2009).

Transit migration and housing forms in Ghana and South Africa

Since neither in Ghana nor in South Africa formal policies address transitory housing requirements related to migration, these are almost entirely catered for by the informal market. In this section we present a variety of informal housing arrangements, which underline today’s transit function of urban areas.

In Ghana, the Ghana Medium-Term National Development Policy Framework 2010–2013 testifies the lack of consideration of migration dynamics for housing: It regards urban sprawl within urban areas as ‘a result of the attractiveness [of cities] to rural migrants’ (GoG 2010, p. 70; see also Arku et al. 2012, p. 3178) but does not refer directly to temporary migration dynamics and migration-related needs or vulnerabilities in the housing sector. However, these are reflected in a variety of informal housing practices that differ from those of the established low-income population and can be assigned only partially to the dualistic categories of housing ownership versus rental housing (Gough & Yankson 2011, p. 793; Arku et al. 2012, p. 3181).

For low-income migrants it is common that, after their arrival in the city, they first stay rent-free as house sharers with already-established family members or friends in urban migrant neighbourhoods (Arku et al. 2012, p. 3181; Yankson 2012b, p. 185). In Ghana these areas are called ‘Zongos’, a term from the Hausa language, which means ‘camping place for caravans’ or a ‘lodging place for travellers’ (Adjei Mensah 2010, p. 60), denoting the transit function of these areas. Migrants without family or origin-based networks in these areas often share the rent of individual rooms within family-owned compound houses (Arku et al. 2012, p. 3181; Ofori & Ayivor 2013, p. 254). Yet other migrants – mostly young, married men with working experience in construction – live alone or with their family for at times extended periods as so-called caretakers in half-finished homes of the middle and upper classes, mostly in the peri-urban fringe areas of the large Ghanaian cities (Gough & Yankson 2011; Arku et al. 2012, p. 3181). Finally, the housing situation of traders who stay in the city on weekdays (usually during market opening) is particularly precarious: they tend to sleep in their stalls or in temporarily occupied accommodations in proximity to the market without any services or infrastructure, with serious implications for personal and public hygiene and safety in many central city areas (interview with market vendors in Techiman, November 2012).

The described living conditions are per se transitory (see Box 1), i.e. a ‘stop over’ before moving on to another city or neighbourhood within the same urban region (Gough & Yankson 2011; Yankson & Bertrand 2012; Bertrand 2011). With respect to intra-urban moves, typical neighbourhood successions (e.g. from recent to more established neighbourhoods) as well as successions of housing forms (e.g. from ‘sharing’ to paid rental) can be identified. Thus, a certain and often informal ‘housing ladder’ is in place, from which ‘trading up’ towards formal rental or ownership housing is albeit seldom possible (Landau 2012a, p. 223). Many migrants live separated from their families, who stay permanently or part of the year at other locations within or outside of Ghana. In many cases, these are the regions or cities of origin, where the migrants have entitlements over lands and shares in family houses (Gough & Yankson 2011, p. 800). Conversely, it is particularly difficult for a large proportion of mobile populations to gain a foothold in the urban housing market, because ‘an insufficient length of residence in town deprive[s them] of efficient networks and, “canons” of participation’ (Yankson & Bertrand 2012, p. 42).

Box 1: Experiences of transitional housing in Ghana (Interviews 2012).

Young taxi driver in Accra: He is from Kasoa, which is about one-hour drive from Accra. This is where he was born and went to school, but there are no work opportunities in Kasoa. So he came to Accra and started to work in the taxi business of somebody he knew from his hometown. He wants to become independent, owning a taxi business himself. Sees Kasoa as his home area, returns there every weekend because that is where his family house is located. In Accra he shares a room with a friend from Kasoa in order to save money. The room is in Nima, a long-established migrant settlement or Zongo.

Young woman in Kumasi: She comes from the Northern Region and sells millet balls at Kumasi Central Market. She shares one room in one of the large migrant settlements with seven other women who are all from the same ethnic group and area of origin. She has come alone and been in Kumasi for four years. After their safe arrival every woman called the next woman by cellphone. They all work as head porters at the market. Usually she stays in Kumasi for three months, and then leaves for her hometown for two months, in order to rest. With the money she earns she buys clothes to get ready for marriage. Once she has accumulated enough to get married she intends to return up North.

Towel vendor in Techiman: He has been coming to the market in Techiman for the last 24 years. He stays during market days (Thursday and Friday) and then travels back to Kumasi where he is based. In Techiman he sleeps at Tamale station close to the market. He talks about many problems associated with that: It is a dangerous place and unhealthy. If there was a possibility he would like to join with approx. ten people and rent a place together. But renting on his own is too expensive. In Kumasi he is renting an apartment together with his family of six. They live in Afoaso, which is a long-established neighbourhood.

Two sack vendors in Techiman: They come from Mali and sell grain sacks to market vendors but also to individual buyers. They came 13 years ago, directly from Gao in Mali. They go back every three years but do not send money since they do not save sufficiently. They live in Jerusalem, a migrant Zongo, under one of the sub-chiefs. There are some Malians in this area, as well as people from many other countries and Ghanaian regions. Both of them live there with their wives and children.

In South Africa, although government institutions have an impact on the provision of affordable housing in particular via the so-called RDP housing subsidy scheme, it is noted that, similar to the situation in Ghana, this makes no direct reference to migration dynamics. On the contrary, the so-called RDP housing market is considered by many experts as a system of housing supply that does not meet the reality of migrants (group interviews with African Centre for Migration and Society, Ekurhuleni Metro Municipality; personal communication with Finmark Trust, NUSP, September and November 2012). Temporary migrants follow job opportunities in short intervals, whereas so-called RDPs, as ownership-based housing, implicitly earmark sedentary population. Harber (2011) describes this situation for Diepsloot, a big peri-urban post-apartheid settlement in the north of Johannesburg: ‘25% of the Diepsloot population does not want to own houses. Maybe they have houses elsewhere and they are in the area temporarily and do not want to put resources into their immovable assets. Diepsloot functions like a hotel to Johannesburg’ (Harber 2011, p. 153). A big constraint is that international migrants cannot legally take advantage of the subsidy scheme. ‘The minute you put houses there you will realize that many of them are not eligible for housing’ (ibid, p. 148). In addition, as Harber puts it, the so-called RDP model is based on full employment, social stability, orderliness, impervious nation states, family households and homogeneity. But today’s reality of South Africa is one of high unemployment, mobility, diversity, increased international migration flows and new household patterns that are rarely based on the traditional family structure. ‘The gap between polity and reality was a chasm’ (ibid, p. 160f). Moreover, interviews that were conducted with senior staff members of metropolitan municipalities in Gauteng suggested that the ‘RDP concept is not sustainable’ (2012 interview with Ekurhuleni Metro Municipality).

The housing supply of migrants is therefore usually informal, mostly in so-called backyard shacks that in many cases are located on the properties of so-called RDP houses and thus provide their owners with a source of income, often the only one. However, there are large informal settlements in the peri-urban area of big cities that perform the function of ‘reception areas’ in which newly arrived migrants can build their own shack, although without being connected to technical infrastructure and services or sometimes provided with basic services like chemical toilets (view Box 2).

The field studies we conducted in South African informal settlements indicate a high level of intra-regional and short-distance mobility of migrants. After their first, usually very precarious, dwelling they try to improve this situation by moving to more consolidated areas of informal settlements that are better equipped with infrastructure. Family, ethnic or national affiliations are also crucial for the choice of residential location of migrants. Thus, in downtown Johannesburg there are blocks of flats predominantly inhabited by people from Somalia, Nigeria and Zimbabwe, just as there are for instance majority-Zulu areas in informal settlements or townships like Alexandra in Johannesburg (2012 interviews in Alexandra).

It is recognised by the state that the problem of housing supply for the migrant population is getting little public attention. Apart from the supply-side approaches to housing provision (see, e.g., the so-called RDP subsidy system), only few concepts are currently being developed or supported, which could offer a demand-driven response to the specific needs of temporary migrants.⁶ One is the Community Rental Units (CRU) programme that intends to ‘facilitate provision of affordable rental tenure for those earning below R3.500, including informal renters, not able to access formal private rental and social housing market[s]’ (Pienaar 2010, p. 5, website BNG).

The problem of the housing supply of migrant population is a complex one, both politically and on the ground. Some solutions already constitute everyday practice, like renting space in RDP houses, but as yet are officially discouraged and thus ‘informal’. The focus of public housing supply still remains on housing property formation in the form of detached homes (Marais & Ntema 2013, p. 86). In the opinion of South African migration researchers this is a form of housing that is hardly suitable to meet the residential needs of a newly urbanised population. In addition, the absence of a ‘housing ladder’ (staggered and widely available housing offer) is noted that would allow tenants to acquire real estate if they wish to (Landau 2012a, p. 222f; Rust 2007, p. 8).

Box 2: Experiences of transitional housing in South Africa (Interviews 2012).

Young guard in Diepsloot/Johannesburg: He is 22 years old and arrived two months back in Diepsloot from Limpopo province. He works at the Civil Service Centre and speaks excellent English. He lives in a rented room in Section 2, paying 300 Rand/month. His intention is to stay there for a few years, although he mentions that Diepsloot is very dangerous at night.

Vendor at cash store in Freedom Square/Bloemfontein: He is 32 years old and arrived from the Eastern Cape in 2010. He came with his father and lives in his father’s self-constructed house 20 minutes’ walking distance away. He goes home 3–4 times a year because he believes that life is better in the Eastern Cape.

28-year-old shepherd from Lesotho: He sells sheep and cattle to locals for funeral services, and lives in a shack in phase 3 of Freedom Square, Bloemfontein, where he does not pay rent. He arrived 10 years back, because there were no jobs in Lesotho. In general terms he thinks that life is better in South Africa. He is job hunting because he does not want to be a shepherd all his life. In Freedom Square he lives with his father, whereas his mother stays back home in Lesotho.

Group of six young male adults, Freedom Square/Bloemfontein: All of them are unemployed but they are raising racing dogs and selling diamonds from Lesotho. Two of them live in RDP houses, the rest in shacks. They came from the neighbouring towns Bellefontain and Botshabelo, one was living on a farm before and the other three were born in Freedom Square. They describe their current living conditions as ok, but they also mention that they don’t have a choice.

Two Somali vendors in Jerusalem/Ekurhuleni: In 1998 they moved to Rustenberg and Pretoria without knowing anybody. Then they moved to an informal settlement near Jerusalem. Living conditions are very hard, especially because of xenophobic attacks. Recently seven shops had to close down. Their option was ‘to move out or die’. Both Somali vendors pay 800 Rand/month but have to sleep in the shop, because it is not safe outside. They feel that they do not have any perspective, ‘nobody helps’. In our field study follow-up six months later, they were not to be found, presumably having moved on.

Conclusions and perspectives

This article focuses on the increasing importance of transit migration and on its implications for urban governance in the field of housing using the examples of Ghana and South Africa. It is based on insights drawn from the review of literature as well as primary data collection and analysis in selected cities of the two countries.

The objective of the article was to point out concrete housing-related manifestations of transit migration and their implications for sustainable housing and urban development. Thus typical housing and tenure forms of transitory population groups were presented for both Ghana and South Africa. While in both countries the informal housing market offers an entry point and instances of a housing ladder to poor migrants, it alone does not provide the maximum benefit to migrants. Rather transitory population groups tend to be in a particularly vulnerable housing position, examples are street traders exposed to insecure and unhealthy urban environments or backyard shack renters without access to basic services. It is further argued that the lack of integration of mobility dynamics in local and notably housing development policies constitutes a chief reason for the vulnerability of transit migrants in the two countries. This is even more the case given the highly territorialised nature of urban policies, which is at odds with the mobility and survival strategies of many poor urbanites.

In both countries, complexities in this field are high and there are no easy solutions to a highly politicised housing sector endowed with limited institutional and financial capacities. In this sense, it must be stated that apart from migration-related problems, housing policies in the two countries can be criticised on other fronts as well. However, differentiated migration dynamics do call for fundamental housing policy adjustments if future pro-poor and sustainable urban growth were to be attained. These adjustments involve matters of recognition, approaches for policy action and new forms of cooperation and representation.

In terms of recognition, it is necessary that housing and urban researchers as well as policymakers acknowledge the important ‘transit’ function of cities, as well as the poverty-alleviation relevance of the underlying forms of mobility. Having said this, it is clear that economically vibrant, well-connected employment centres show more transit functions than others, since temporary migration dynamics are mainly driven by the availability of job opportunities. In contrast to the notion of transit, to date, urban and housing-related research and policies have been more biased towards urban ‘entry’ and ‘insertion’ (Yankson & Bertrand 2012). The transit function tends to be spatially concentrated in informal zones or neighbourhoods such as migrant ‘Zongos’, inner-city transitional areas and slum settlements, which typically act as gateways in the context of intra-urban or translocal mobility. It is largely in these areas that transit housing practices of migrants tend to converge.

In terms of approaches for policy action, it was shown that while South Africa and Ghana pursue starkly different strategies with respect to regulating and providing low-income housing – state versus market-led – both prove equally inadequate for mobile populations. A key prerequisite for integrating migration into housing and local development in both countries is thus an improved understanding of migration dynamics, their causes and spatial patterns. While in Ghana limited data exists about inter- and intra-regional migration dynamics, in South Africa much data is available. However, in both countries there is a lack of integration of demographic dynamics and (temporary) mobility for local and national housing development; further information flows and cooperation between different governmental and administrative levels are weak in this regard. Diversifying housing options and articulating informal and formal mechanisms are pertinent in both countries. Examples are providing access to basic services to backyard shack renters and enhancing the capacity to run low-income rental housing as a tenure form more adapted to the needs of mobile populations.

Regarding new forms of cooperation and representation in South Africa and in Ghana, migrants and their organisations need to be better integrated in local development and planning forums, not least by means of an improved articulation between ‘informal’ and formal governance. And lastly, in order to tackle limitations of territorialised planning, there is a need to stretch housing development and planning beyond municipal borders, to take into account – many times short-distance or intra-metropolitan – economic ‘functional spaces’ and thereby better provide for mobility-related trajectories and networks.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The research was supported by the Mercator Research Center Ruhr, Germany (grant number An-2012-0026).

Notes on contributors

Eva Dick

Dr. Eva Dick is a senior research fellow and lecturer at the Faculty of Spatial Planning at TU Dortmund University, Germany, where she has also earned her PhD. Her research focus is on new patterns of migration, translocality and implications for urban and regional governance and development.

Thorsten Heitkamp

Dr. Thorsten Heitkamp is a senior research fellow and lecturer at the Faculty of Spatial Planning at TU Dortmund University, Germany, where he earned his PhD. His research focus is on housing policies, new forms of migration dynamics and implications for modes of urban and regional governance and development.

Notes

1. Translated from the German original.

2. Until the end of 2012 this allocation process was based on population figures of the 2001 Census.

3. The Coordinated Programme of Economic and Social Development Policies 2010–2016 recognises informalisation to be ‘a spontaneous and creative response to the formal economy’s incapacity to satisfy basic needs’ (State of Ghana 2010, p. 19).

4. Still today the post-apartheid-project-linked housing subsidy scheme is commonly referred to as ‘RDP housing’ (Website DHS). This article will refer to it as ‘so-called RDP housing’.

5. At the time of writing this document is equivalent to approx. USD 320.

6. Only in recent years settlement upgrading programmes have been started, e. g. with the support of the National Upgrading Support Program. Also a certain shift to the acknowledgement of the importance of backyard shacks, which are mostly connected to basic infrastructures, can be noted (Topham 2011/2013; 2012/2013 Interviews Topham).