Abstract
This paper offers an exploration of urban expansion from the point of view of the individual residents buying land, settling and living in new, rapidly growing peripheral settlements of Dar es Salaam, Tanzania. The findings suggest that the demand for affordable housing is the primary motivation for residents moving to the periphery. The demand for self-built, owner-occupier housing is especially significant initially, while the demand for non-ownership housing increases in importance later in the process. Income-related motives, on the other hand, are strikingly absent from settlement considerations. Urban residents settle in the periphery, even though income-generation is often tied to working somewhere else, namely in the central parts of the city. The paper proposes that the processes of urban expansion depicted in this study are usefully conceptualised as suburbanisation processes, though it is a type of suburbanisation that has some peculiarities given the particular context, where expansion happens informally and largely unguided by planners.
Introduction
Urban expansion is a significant trend in the cities of Africa. Urban populations are growing and to a large extent population growth has taken the form of expansion on the urban peripheries (Angel et al., 2011; Arku, 2009). While urban expansion processes have been widely documented in the developed world, it has been less evident that the phenomenon is global in scope and applies to cities in Africa as well (Angel et al., 2011; Ekers et al., 2012). In cities of Africa the dynamics and drivers of urban expansion are less well-documented than in cities of the developed world (Harris, 2010; Todes, 2014).
This paper offers an exploration of how settlement preferences and practices of urban residents influence processes of urban expansion. The paper is based on recent case studies of five rapidly growing settlements in the periphery of Dar es Salaam, the largest city and de facto capital of Tanzania. Dar es Salaam has experienced continuously high population growth rates. All parts of the city are experiencing population growth, but growth is nowhere as rapid as in the sprawling peripheral areas, where formerly rural land is continuously developed and incorporated into the urban area (Andreasen, 2013). To a large extent expansion is happening informally and unguided by planners in settlements often referred to as unplanned, informal, slums or squatter areas (Kironde, 2000, 2006; Kombe, 2005).
In this paper we take a peoples’ perspective, exploring urban expansion from the viewpoint of the individual residents buying land, settling and living in the periphery. Why are urban residents attracted to settle in the periphery? How can their choices, motivations and aspirations help explain processes of urban expansion? A wider range of structural forces are obviously also influencing and shaping urban expansion processes, including urban land markets, land tenure systems, infrastructure, urban planning and public and private development projects (Todes, 2014). However, the experiences and motivations of peripheral residents are often neglected in academic research (Harris, 2010). Taking residents as the point of enquiry acknowledges the agency and resourcefulness of urban residents when understanding how the city is transformed by its growing population. As new urban neighbourhoods develop on the urban fringes, largely informally, the choices, preferences and aspirations of residents are central in shaping urban development on the ground. Based on a critical reading of the literature on urban peripheries in sub-Saharan Africa, the paper proposes that the processes of urban expansion depicted in this study are usefully conceptualised as suburbanisation processes.
Conceptualising urban peripheries in sub-Saharan Africa
Urban peripheries in sub-Saharan Africa are often studied as part of the peri-urban zone or interface. In this literature the peri-urban zone is understood to be the geographic edge of cities forming a continuum or interface between the urban and rural. It is a hybrid, transitional zone combining rural and urban conditions, mixed land uses and mixed livelihoods (Douglas, 2006; Simon, 2008; Simon et al., 2006). The peri-urban can be understood as a gradient between more ‘urban’ and more ‘rural’ parts of the peri-urban zone (Simon et al., 2004). It is a temporary and transitory space, making it difficult to establish exactly where the city ends and where the rural space begins (Trefon, 2009).
Very recently, urban peripheries in Africa are also beginning to be studied through the lens of suburbanisation, a concept commonly reserved for urban peripheries in Europe and the USA (Clapson and Hutchison, 2010). This literature takes a ‘seeing from the South’ approach and deploys the concept of suburbanisation in an African context (Buire, 2014b; Mabin et al., 2013). It is part of an emerging research agenda where suburbanisation is considered as a global process (Ekers et al., 2012). The aim is to ‘provincialise’ the concept of suburbanisation (Buire, 2014b), following calls for urban theory to be more ‘cosmopolitan’ and draw on empirical research from cities all over the world (Parnell and Robinson, 2012; Robinson, 2006). Often the US experiences are assumed to represent the paradigmatic case of suburbanisation, but a key point of agreement in the emerging suburbanisation literature is that diversity is the norm rather than the exception (Clapson and Hutchison, 2010). In this literature suburbs are purposely defined very broadly as forms of decentralised urban spaces (Ekers et al., 2012) characterised by peripheral locations, low population densities and relative newness (Harris, 2010). The term is carefully constructed to incorporate many different types of peripheral growth, from wealthy gated communities of American cities to informal settlements on the fringes of African cities (Ekers et al., 2012). Suburbanisation is considered as a process of relatively new and dynamic urban developments that occur elsewhere than the central areas and at lower densities (Sawyer, 2014).
There are obviously overlaps and commonalities between the peri-urban literature and the literature on African suburbanisation, as both are preoccupied with dynamic development processes on the peripheries of cities. Nonetheless, peri-urban and suburban are not simply interchangeable concepts. Spatially, the peri-urban literature is broader in scope and often includes everything from rural areas with market links to nearby cities to outskirts of larger cities functionally integrated into the city. In terms of interest spheres, the peri-urban literature is preoccupied mainly with the mixture of rural and urban conditions. Some studies emphasise the mixed nature of land use in peri-urban areas, where arable land is continuously transformed into urban uses (Becker, 2013; Binns and Maconachie, 2006; Briggs and Mwamfupe, 1999; Gough and Yankson, 2000; Simon et al., 2004; Thebe and Rakotje, 2013). Other studies focus on the mixed nature of livelihoods and highlight how peripheral locations enable residents to combine rural and urban livelihood strategies as well as form ideal locations for engaging in commercial agriculture (Aberra, 2006; Briggs and Mwamfupe, 2000; Harris et al., 2006; Lanjouw et al., 2001; Stevens et al., 2006).
The literature on African suburbanisation is predominantly interested in urban peripheral spaces. ‘Urban’ in this context is understood broadly as opposed to rural land uses and economic activities. Areas of interest include residential sprawl, urban housing, informal habitat, urban land markets and zones of middle- and higher-income residence (Mabin et al., 2013). The importance of urban housing is highlighted in recent studies suggesting that urban residents are both pushed and attracted towards peripheral locations in search of affordable housing, improved life quality as well as opportunities for homeownership (Buire, 2014a; Sawyer, 2014). Recent studies also draw attention to the significance of housing created and inhabited by middle- and higher-income segments of the urban populations (Buire, 2014a; Mercer, 2014; Sawyer, 2014). It should be noted that the demand for urban housing as a driver of peripheral development is not at all unfamiliar to the peri-urban literature, where studies have highlighted the affordability of peripheral locations as well as the demand for owner-occupier housing (Briggs and Yeboah, 2001; Gough and Yankson, 2011; Owusu-Ansah and O’Connor, 2010; Simon et al., 2004).
Scholars are well aware that applying the concept of suburbanisation in an African context entails the risk that the concept becomes fundamentally vague and obscures the diversity of urban peripheries across the world (Harris, 2010; Sawyer, 2014). The applicability of the concept is considered open to debate and an important avenue for further research (Harris, 2010; Todes, 2014). Part of the ambition of this paper is to contribute to this emerging research agenda by discussing the applicability of the concept of suburbanisation in the context of the selected rapidly growing, peripheral settlements of Dar es Salaam. We propose that the processes of urban expansion taking place in these settlements are usefully conceptualised as suburbanisation processes. We base our argument not only on the broad definitions offered in the literature on African suburbanisation, but also on the insights emerging from the analysis of how peripheral residents themselves explain their settlements and settlement preferences. The analysis revolves around the motivations of residents living in peripheral settlements and distinguishes between housing- and livelihood-related motivations. This makes it possible to explore how and in what ways settlement practices and preferences are shaped by the search for affordable housing and opportunities for homeownership as highlighted in the literature on African suburbanisation, rather than the mixture of rural and urban conditions and especially the possibility to combine rural and urban livelihood strategies emphasised in the peri-urban literature.
Context: Urban growth and spatial expansion of Dar es Salaam
Dar es Salaam is the largest city and de facto capital of Tanzania. With a total population count of 4.4 million in 2012, it is a big city in the context of sub-Saharan Africa (National Bureau of Statistics (NBS), 2013). Dar es Salaam’s population grew at a breathtaking pace of 5.8% per year on average in the most recent inter-census period from 2002 to 2012 (NBS, 2006, 2013). This is a continuation and acceleration of historically rapid growth trends. Despite rapid growth, population densities remain quite low in most parts of the city (Andreasen, 2013). With the exception of the central business district and Kariakoo, the city is dominated by low-rise, low-density development (Halla and Mang’waru, 2004; Lupala, 2002).
Historically, the expansion of Dar es Salaam is well-documented. Sprawl was noted as a central feature of Dar es Salaam’s urban form already around independence (de Blij, 1963). Since then the surface area of the city has increased dramatically (Olvera et al., 2003). Spatial expansion of Dar es Salaam in the 1970s and 1980s followed a star-shaped pattern along major roads (Briggs and Mwamfupe, 2000). Expansion in this period was initiated by urban residents pursuing peri-urban agriculture as part of a strategy of economic diversification following economic crises (Owens, 2010). Growth in the 1990s was characterised by infill and densification of existing linear settlements, and much of it consisted of residential sprawl (Briggs and Mwamfupe, 2000; Briggs and Yeboah, 2001). The peripheral areas that developed in the 1990s had a substantial share of high-income residents (Lupala, 2002).
Recent population growth has also resulted in widespread spatial expansion. The peripheral parts of the city have grown most rapidly and experienced substantial population increases as well as significant increases in population densities (Andreasen, 2013). A recent study estimates that the total built-up areas of Dar es Salaam increased by 133% during 2002–2011, while the population only grew by 75% during 2002–2012 (Macchi et al., 2013). Provision of infrastructure and services has not kept pace with the demographic and spatial growth of the city (Olvera et al., 2003). Expansion has continued undeterred despite escalating problems in the mobility system of the city. Dar es Salaam has a monocentric structure, where many functions and employment opportunities are located in and around the central business district, central market areas such as Kariakoo and office and industrial sites within and along Nelson Mandela Road. The mobility system is characterised by highly inadequate road networks, insufficient public transport and severe congestion problems (Melbye et al., 2015). The ongoing transformations of the central areas into high-rise commercial buildings along with increasing rates of car ownership has further increased pressure on the urban mobility system (Kiunsi, 2013).
Urban expansion is happening largely informally and unguided by planners. Since independence, expansion of informal settlements has been more or less tacitly tolerated by authorities (UN-Habitat, 2010a). In total, an estimated 80% of Dar es Salaam’s territory is informal (UN-Habitat, 2010a). The pervasiveness of informal development is related to the dual system of land delivery inherited from the colonial era. In the formal system land is allocated with recognised tenure, often below market value as land is officially considered to hold no value unless developed. The formal allocation system is highly inefficient, subject to malpractice and only plays a marginal role in the provision of urban land (Kironde, 2000, 2006). Alongside this a vibrant informal land market operates largely outside the realm of public authorities. Informal land is more accessible and affordable, but offers less de facto tenure security and more uncertainty about authorities’ future plans for an area. Furthermore owners of informal land are generally unable to access housing loans from formal financial institutions (UN-Habitat, 2010b). Though informal land is traded without titles, it does not mean that residents are considered as illegal occupants. Ownership is legitimised through informal sales agreements and social recognition from local leaders and adjoining landowners (Kironde, 2006; Kombe, 2000). Residents in informal settlements of Dar es Salaam have a relatively high perceived security of tenure (Lupala, 2002; UN-Habitat, 2010b).
Data collection
The paper is based on in-depth case studies of five selected residential areas, four of which are located on the urban periphery, while one could be considered a formerly peripheral area, which is today more consolidated. Selection of case study areas was based on an analysis of spatially disaggregated population data for Dar es Salaam from the two most recent censuses, which can be found in full in Andreasen (2013). All the selected case study areas have experienced both very high population growth rates, around or higher than 10% per year in the 2002–2012 period, as well as significantly increased population densities. The selected areas are extreme cases of rapid urban population growth. They are areas of radical urban transformation, which are considered problematic from a planning perspective. Extreme cases are often rich in information, well-suited for in-depth learning and likely to produce advanced understanding (Flyvbjerg, 2006). Selection of the different case study areas took place in consultation with planners at municipal and ward-level and sought to sample diversity in relation to population densities, inclusion of areas from all three municipalities as well as inclusion of informal as well as formally surveyed areas. As such, the selected case study areas exemplify different variations of the dynamic development processes happening on the periphery of Dar es Salaam. See Table 1 for an overview of the selected case study areas and Figure 1 for their locations.
Overview of case study areas.
Notes: aThe numerical indications of population densities are 2012 densities in persons/km2. The values have been calculated using ward-level population counts published as part of the 2012 census (NBS, 2013) and ward polygons for Dar es Salaam kindly shared by the National Bureau of Statistics. Sub-ward population counts and polygons have not been released from the 2012 census yet. As the peripheral wards are rather large area-wise the ward-level densities are not likely to be precise estimates for the case study areas.
Location of case study areas.
The case study areas have all undergone significant transformations from sparsely populated rural or peri-urban areas dominated by bush and agricultural land use to more densely developed residential areas forming part of the contiguously built-up urban area. In all case study areas urban expansion was initiated by a first wave of newcomers buying land and developing houses, predominantly self-built, owner-occupier, single-household houses. Expansion typically started around smaller villages or scattered settlements inhabited by ‘indigenous’ people, broadly referring to people of the Zaramo tribe native to Dar es Salaam and the surrounding coastal region, most often holding land under customary tenure. In most of the areas, expansion was facilitated by informal subdivision processes, where indigenous landholders sold their land to newcomers on an informal market. In the smaller pockets of surveyed land, expansion was facilitated and guided by a formal surveying process and land was principally allocated through the formal land system, though in many cases formal land had also been traded vigorously after allocation. Alongside homeowners’ families, many non-owners were also attracted to the areas, preliminarily caretakers and relatives of homeowners. Later in the process many tenants were attracted as some of the owners developed parts of their properties into rental accommodation.
Fieldwork was conducted between November 2013 and May 2014. The core of data that this paper is based on consists of semi-structured interviews with individual residents regarding their settlement preferences and intra-urban settlement trajectories. A total of 174 residents participated in individual interviews, spread fairly evenly across the five case study areas. The sampling of residents was done purposely and sought to ensure inclusion of residents living in visibly different housing types, under different tenure forms and with varying degrees of access to roads and public transport services. Participants are primarily newcomers, meaning residents who settled around the same time or after urban expansion processes began. The sampling targeted adult household heads paying for housing expenses of their households, i.e. land purchase, house construction, rent payments, etc. They were considered to hold most influence over as well as put more strategic consideration into their settlement choices. Special attention was given to the inclusion of homeowners, as fieldwork revealed they were central actors in catalysing urban expansion processes. Of the 174 participants, 117 are homeowners, 27 are tenants, 16 live in the homes of relatives, 12 are house caretakers and 2 live in housing provided by their work place. Homeowners are likely over-represented in the sample, as local informants often estimate that non-owners are currently more prevalent in the populations of the case study areas. Absentee landowners were not specifically targeted in the sampling process. Given the level of consolidation in the case study areas at the time of fieldwork undeveloped plots did not appear as a major phenomenon, except for absentee owners of larger beach plots in the two coastal areas, who were targeted specifically.
The individual interviews were complemented with focus groups with long-term residents in each area, many of which were indigenous and/or lifelong residents, who had been engulfed by the urban expansion processes. The focus groups were primarily concerned with creating a shared narrative of the various changes that the case study areas had undergone during the last 15 years. This was supplemented with observations of the urban environment and interviews with relevant local key informants, such as local leaders, larger investors and informal land brokers in each area. An additional 29 residents participated in focus groups and another 29 residents participated as key informants. Furthermore, data collection included interviews with 13 urban planners from various central agencies and authorities involved in urban planning and service provision in Dar es Salaam.
Housing-related motivations in settlement considerations
This section presents an analysis of residents’ motivations for settling in the periphery based on the interviews with individual residents regarding their settlement preferences and intra-urban settlement trajectories. Motivations of individual residents were often complex and composite. For the sake of clarity we have tried to distil the primary motivation of residents as summarised in Table 2.
Categorisation of residents in the sample according to motive of settlement in case study area.
Note: The rows in bold represent accumulated groups, each consisting of the non-bold rows underneath.
The evidence suggests that urban expansion is primarily driven by demand for housing from urban residents. Table 2 reveals striking similarities in residents’ motivations across the case study areas. 133 out of 174 residents in our sample across all five case study areas give housing-related motives as their primary motivation for settling in their current area of residence. The demand for owner-occupier housing is especially significant initially. In all case study areas, urban expansion was initiated by a first wave of newcomers buying plots of land and constructing owner-occupier houses of varying sizes. Homeownership aspirations are a central motivation across five very different case study areas and a very heterogeneous group of residents. Building a house to live in was the primary motivation for a total of 85 residents currently established as homeowners or supervising house construction on a nearby plot. All are first-time homeowners in Dar es Salaam. For the homeowners an important attraction of the periphery is the relative affordability of peripheral plots. While many of the homeowners also appreciated various qualities of the environment of their settlements, such as good space, greenery and fresh air, these considerations seemed to be secondary. Primarily, they had been looking for an affordable plot of land and someone willing to sell it. As one homeowner phrased it: ‘There are no specific things that attract me to live in this area. It is just that I’m able to live in my own house’. Buying a small plot of undeveloped land in the periphery allows aspiring homeowners to construct their houses incrementally. In a context where mortgage finance is almost non-existent, this is the only way to become a homeowner for the vast majority of the population (National Housing Corporation (NHC), 2010; UN-Habitat, 2010a, 2010b). Homeownership is a dream that very few would have been able to realise in the more central parts of the city. Instead, the periphery is the target area for a wide range of urban residents with aspirations of homeownership. Homeowners in surveyed areas had often been specifically interested in a formal plot of land to avoid complications of informal tenure, but other than that their motivations did not distinguish themselves markedly from homeowners in informal areas. Second-time homeowners gave motivations related primarily to housing quality improvements such as acquiring more living space or living in a better environment. Initially, settlement often happened without any basic services or infrastructure in place, in informal as well as in surveyed areas, and considerations concerning services and infrastructure are widely absent from settlement considerations of the first newcomers. They were willing to tolerate lack of services, because of the perceived benefits of owning their own house. Some degree of services and infrastructure developed along the way, and this often played a part in settlement considerations for later newcomers (Andreasen and Møller-Jensen, 2016).
Alongside the homeowners, many non-owners were also attracted to the peripheral settlements. Caretakers were attracted by the opportunities to live rent-free while taking care of someone’s property during the construction phase. This was the primary motivation for seven residents, see Table 2, who moved to the periphery to be caretakers of specific properties. It is a common practice among homeowners to have caretakers living temporarily on their properties to safeguard it from theft and encroachment. As homeowners moved into their houses, relatives and extended family members were attracted by the possibility of free lodging, a common way of offering support to relatives in need in much of Tanzania. This was the primary motivation for 17 residents, of which some still lived in the homes of their relatives, while others had moved into nearby rental accommodation. Many tenants were also attracted as rental markets emerged along the way. Rental markets emerged, particularly in informal areas, as some of the homeowners developed parts of their properties into rental accommodation, either subletting a few rooms in the main house or constructing separate tenant houses on their compounds. Tenants had often been attracted by the affordability of rental accommodation in peripheral areas, compared with more expensive rents in central locations. Affordable rents were the primary motivation for five tenants, see Table 2. Other tenants give motivations related to housing quality improvements, such as acquiring more living space or living in a better environment. At the time of the fieldwork, the demand for rental rooms was often high and the share of tenants had increased significantly. In the surveyed areas, rental markets did not emerge to the same extent, probably because of zoning regulations and the high-income status of most residents.
The newcomers are primarily long-term urban residents moving from more central parts of Dar es Salaam (Andreasen and Agergaard, 2016). Socio-economically the newcomers are a very heterogeneous group. Newcomers in surveyed areas are predominantly from wealthier segments of the population, often professionals with higher education and stable jobs. In the informal areas a mixture of low-, middle- and high-income residents, both owners and non-owners, often live side by side. Some hold formal jobs, while many are self-employed in various informal business and trading activities. The homeowners tend to have well-established, if not exactly stable, income sources. They have often been employed or engaged in the same activities for a long time. Land purchase and house construction require a careful orchestration of household resources and accumulation of savings over an extended period of time. Irrespective of whether they live in surveyed or informal areas, homeowners generally build with modern building materials such as bricks and iron-sheet roofing and often invest a lot of resources in their properties. Many confidently plant trees and develop small gardens. This reflects that homeowners in both formal as well as informal areas consider themselves to be the owners of their properties, and that the vast majority of homeowners consider their current house their permanent place of residence (Andreasen and Agergaard, 2016). The ability to successfully establish themselves as homeowners indicates that this group does not belong to the poorest segments of the urban population. Because accumulation takes a long time, homeowners tend to be mature or middle-aged people with older children, as one participant dryly noted: ‘If you will try to move five kilometres from here, you won’t find a crying baby’. Non-owners, on the other hand, tend to be younger and often lower income residents, though it has not been possible to detect any systematic differences between homeowners and non-owners in relation to their livelihood strategies (Andreasen and Agergaard, 2016).
Income-related motivations in settlement considerations
While housing-related motivations are prevalent, income-related motivations are conspicuous by their absence. Only 13 residents in our sample settled in their area of residence to pursue a specific livelihood activity, cf. Table 2. These 13 residents are all engaged in non-agricultural activities and include three primary school teachers, six engaged in the local business and service sector within their area, two engaged in stone-quarrying in Mjimwema and two fishermen in Ununio. Only three residents moved to their current area of residence to engage in agriculture. They are all very long-term residents, who have lived in their areas for more than 30 years, and at the time of fieldwork none of them were engaged in agriculture anymore.
While income-related motives are widely absent from settlement considerations, this does not mean that no livelihood activities take place locally. Homeowners often use their land for economic activities, such as small shops, workshops, small livestock, vegetable gardens or subletting of rental rooms. For most landowners these activities are supplementary, though for some, often single women or retirees, it is a substantial part of their income. Generating income from your land is most prevalent in informal areas, likely because of zoning regulations in surveyed areas. Increased population growth has also generated demand for a local business and service sector offering livelihood opportunities for a wide range of residents, from construction workers, house servants and drivers to craftsmen, shopkeepers and restaurant owners. Only a few moved because of these opportunities, however. Instead, it appears that when people move into the areas in search of affordable housing, they tend to bring their business or craft with them, as one shopkeeper explained: ‘No, I did not really consider the local business opportunities. It is just that I did the same type of activities before I moved here, and then anywhere, where I can live, I can do my business there’.
The findings suggest that residents chose to settle in the periphery even though income-generation is often tied to working somewhere else. Urban-based livelihood activities predominate among residents across all five case study areas. The majority of residents form parts of households with at least one breadwinner working outside the area, predominantly in the central parts of the city, or relying on regular exchanges on central markets. At the level of the individual, 84 of the 174 residents in our sample regularly travel outside of their area of residence in relation to their livelihood activities, but at the level of the household, 133 are part of households, where one or more members work outside their area. At the level of the individual, 52 of the 174 residents in our sample are able to work within their area of residence. This group primarily consists of self-employed or casual workers in the local business and service sector. As their customer base is predominantly local residents, they rely on demand from other residents, who are to a large extent generating their income outside of the settlements. Furthermore, many of them form part of households, where other breadwinners work outside of their settlements. Likewise many individual residents, 38 out of 174, are not themselves economically active, but most of them are dependents forming part of households where main breadwinners make regular trips to central areas in relation to their livelihood activities.
The hassle of transport is inevitably part of the equation for many breadwinners when balancing settlement and livelihood choices. Commuting is described as extremely tiresome. Traffic congestion, queues and jams result in very long and unpredictable transport times. Breadwinners often spend an unbearable number of hours in traffic every day. The hassle of transport is tolerated because of the attraction of affordable housing in the peripheries. For homeowners the benefit of owning your own house simply outweighs the transport problems. None of the homeowners mention proximity to work as a motivation for their settlement choice. For tenants it seems to be a more complex individual calculation. While rents are often considerably cheaper in the peripheral settlements, transport also weighs heavily on schedules and places considerable pressure on household budgets. Four tenants mention proximity to work as a primary motivation for settling in their current area of residence.
The strong affinity for homeownership as a tenure form
The study has identified residents’ aspirations of homeownership as a central motivation, catalysing urban expansion processes across all five case study areas. Residents expressed a strong affinity for homeownership, often surprisingly unaffected by considerations concerning livelihoods, services, infrastructure and transport. Among non-owners, future aspirations of homeownership are also very common. This section will explore how residents themselves explain this affinity for homeownership.
Homeowners often explain their preference for homeownership with the difficulties of accessing housing in Dar es Salaam, particularly housing that is affordable, secure and of a certain standard. Many of the homeowners previously lived as tenants in more central and consolidated parts of the city (Andreasen and Agergaard, 2016). They often refer to ‘rental problems’ or the ‘tough life’ of tenants when asked why they wanted to become homeowners. Expensive and ever-increasing rents as well as conflicts with landlords are often mentioned as ‘disturbances’ of rental life. As one homeowner explained: ‘I am comparing this with the rental environment and the psychological torture from the landlords. So for me, owning a house is a great life quality’. Regular rent payments may be hard to meet for those relying on self-employment activities, which often entail changing income streams. A difficult life situation, such as becoming a widow or losing the ability to work, can make it impossible to pay rent. One homeowner explained the worries of a tenant as follows: ‘So if you are staying in a rented house, what if things go wrong with the income? Then the first thing, if you are struggling, is the house, because at the end of the month or at the end of the year the owner of the house will come, hey I need my money’.
Buying undeveloped land in the periphery allows owners to construct their own houses incrementally, financed with savings accumulated over many years and according to changing income flows and economic priorities of the household. This means that homeowners incur housing-related expenses whenever it is convenient, and not at specific times decided by a landlord. Homeownership, even in informal areas, is therefore considered more secure, because it is rent-free. Some homeowners also mention the additional security of accumulating assets in the form of land. Homeowners consider land to be a secure investment. You can leave it for children or raise cash in times of need through sale and subdivision. The strong affinity for homeownership also has an aspirational aspect. Homeowners often expressed a certain pride in being the owners of their own house. They would emphasise that they are able to live ‘a private life’ and can feel comfortable doing what they want in their own home. Some consider it a major benefit that they are able to accommodate a growing family as well as various relatives in need. Some relate homeownership to valued life stages such as ‘starting a family’ or ‘starting your own independent life’. As such it seems to be perceived as an ingrained part of becoming a successful adult.
The strong affinity for homeownership may also be related to a wider sociocultural valuation of landownership in the Tanzanian context. Halla and Mang’waru (2004) call landownership a ‘sociocultural compulsion’ for Tanzanians. The homeowners in this study were primarily motivated by homeownership, rather than landownership more generally. To some extent this is also what is to be expected when interviewing resident-homeowners. There is anecdotal evidence in data about residents having bought plots in other areas, seemingly without the intention of building a house to live in, either as an investment object or as a backup plan, in case they lose their current house. Perhaps some residents originally bought several different plots in peripheral areas, and waited to see which areas developed into more attractive residential areas. The design of this study, interviewing resident-homeowners about their settlement choices in retrospect, often many years after they bought their plot, makes it difficult to offer a robust analysis of this, as landownership and homeownership are closely intertwined in the minds of the homeowners.
Conclusion
This paper has presented an exploration of how settlement preferences and practices of urban residents influence processes of urban expansion in five rapidly growing settlements on the periphery of Dar es Salaam. We propose that the processes of urban expansion taking place in these settlements are usefully conceptualised as suburbanisation processes. Part of the ambition is to contribute to an emerging literature on African suburbanisation, where suburbs are defined broadly as ‘decentralised urban spaces’ (Ekers et al., 2012) characterised by peripheral locations, low population densities and relative newness (Harris, 2010). While these characteristics are certainly true for the selected case study areas, we base our argument also on the insights emerging from the analysis of how peripheral residents themselves explain their settlements and settlement preferences.
The findings suggest that the search for affordable housing is the primary motivation for residents moving to the periphery. The demand for self-built, owner-occupier housing is especially significant initially, while the demand for non-ownership housing increases in importance later in the process. While housing-related motives are prevalent, income-related motives are strikingly absent from settlement motivations. The findings indicate that residents settle in the periphery, even though income-generation is often tied to working somewhere else, namely the central parts of the city. Urban-based livelihood activities predominate among residents across all five case study areas, and the majority of participating households rely on one or more members regularly travelling outside of their area of residence in relation to their livelihood activities.
Homeownership aspirations emerged as a very important motivation across five different case study areas and a highly heterogeneous group of residents. Buying undeveloped peripheral land allows aspiring homeowners to construct their own houses incrementally and according to changing income flows. In a context where mortgage finance is widely absent, this is often the only way to become a homeowner. Therefore the periphery has become target area for a wide range of urban residents with aspirations of homeownership. Nothing suggests that expansion is a poverty-driven process. The homeowners’ ability to successfully establish themselves as homeowners indicates that they do not belong to the poorest segments of the urban population.
The developments appear to be a continuation of the widespread residential sprawl observed by other scholars in the 1990s (Briggs and Mwamfupe, 2000; Briggs and Yeboah, 2001), rather than the hybrid of urban and rural conditions emphasised in the peri-urban literature. Land use in the case study areas is characterised by urban housing for urban residents engaged in urban-based livelihood activities. Rural land uses and livelihoods are widely absent. Therefore these settlements seem inadequately conceptualised as peri-urban and ought to warrant a conceptualisation of their own. Suburbanisation appears to be the most useful and applicable concept. ‘As good as any, and better than most’, as Harris (2010) wryly notes. We acknowledge that the use of suburbanisation in this context entails a risk of obscuring major differences between peripheral developments of Dar es Salaam and common (Anglo-American) conceptions of suburbs. Therefore it is important to note that it is a type of suburbanisation with its own particularities, as expansion happens informally and largely unguided by planners. Housing development generally happens before provision of services and the expansion is widely unsupported by appropriate infrastructure. An often unguided subdivision results in a haphazard layout of plot, lack of access roads, plot inaccessibility and lack of land for communal facilities (Andreasen and Møller-Jensen, 2016). Still we consider suburbanisation as a more useful conceptualisation compared with peri-urban. Suburbanisation implies the creation of urban peripheral spaces characterised by predominantly urban conditions, while also acknowledging the dynamic and transformational aspects of these spaces. It highlights the importance of demand for housing by urban residents as a driver of urban expansion and induces academic recognition of these peripheral spaces as functionally part of the city. Furthermore, the concept of suburbanisation elicits a reflection on the applicability of terms such as slum and squatter areas often applied to expanding peripheral settlements owing to the widespread informality of tenure. With the connotations of temporariness, concentrations of poverty and sub-standard housing, neither of these terms seem appropriate nor precise depictions of the nature of the developments emerging in the case study areas.
Footnotes
Acknowledgements
We are grateful to the residents of Dar es Salaam, who generously gave their time to participate in this research; to Professor Robert Kiunsi and Dr Ally Namangaya from Ardhi University for support and advice; to indispensable research assistants John Williams, Christian Maeja and Sheila Matitu; and to graphic designer Kent Pørksen for creation of 
.