Residential segregation is caused by interacting economic, social, demographic, institutional, historical and spatial processes rather than by a single mechanism. One fundamental cause is unequal access to housing. Households differ in income, wealth, savings, credit access and ability to pay rent, so they cannot all compete for the same properties or neighbourhoods. Expensive areas become more accessible to affluent households, while lower cost areas may contain greater concentrations of households with limited resources.
Housing tenure reinforces these patterns because social housing, private rental housing and owner occupied housing have different geographical distributions. Household preferences can also contribute to segregation when people seek neighbourhoods close to relatives, cultural institutions, schools, workplaces or familiar communities. These preferences operate within constraints created by income and housing availability. Consequently, residential segregation often results from the interaction between choice and constraint rather than either factor alone. The relative importance of each cause differs between cities, groups and historical periods, meaning that explanations should be grounded in the specific housing system and population being studied.
Economic inequality is one of the strongest structural mechanisms behind residential segregation. Where property values and rents vary substantially between neighbourhoods, housing markets sort households according to their financial resources. Wealth is particularly important because households with accumulated assets can obtain mortgages, pay deposits, absorb moving costs and compete for scarce properties more easily than households dependent entirely on current income. Income affects the ability to meet monthly housing costs, while credit conditions influence access to ownership.
As affluent households concentrate in expensive locations, demand can push prices still higher, potentially making those areas inaccessible to households with lower incomes. Lower income households may consequently become concentrated in cheaper areas. This process can produce economic segregation without any explicit policy requiring separate residential areas. However, market sorting does not operate independently of history and institutions because the location and quality of housing stock, transport infrastructure, planning decisions and previous investment patterns influence which areas are expensive or affordable.
Housing discrimination can directly cause segregation by restricting where particular groups can rent, buy or obtain finance. Discrimination may involve landlords refusing applicants because of ethnicity, religion, disability, family composition or another protected characteristic, or treating comparable applicants differently during the housing process. Mortgage and insurance practices can also affect access to home ownership, while discriminatory estate agency practices can influence which properties are shown or how applications are handled.
Historical forms of discrimination can have effects long after specific practices have been prohibited because housing wealth, neighbourhood composition and property ownership accumulate over generations. Where a group has historically been excluded from particular neighbourhoods, its members may possess less housing wealth and fewer established networks in those areas, reducing access even after formal discrimination declines. Discrimination can therefore operate directly through individual decisions and indirectly through inherited inequalities and institutional structures. Establishing whether discrimination caused a particular segregation pattern requires evidence about behaviour, institutional practices and historical context rather than simply observing that groups live in different places.
Migration and social networks can contribute to residential concentration because people often move through existing chains of family, friendship and community connections. Newly arrived migrants may initially live near relatives or members of the same language or cultural group because those networks provide information about jobs, housing, childcare, translation, transport and local services. Businesses and community organisations may subsequently develop around these concentrations, reinforcing the attractiveness of the area to additional members of the group.
This process is sometimes described through concepts such as chain migration, ethnic clustering or network effects. Such concentration does not necessarily result from exclusion. It can represent a practical strategy for navigating unfamiliar institutions and reducing the costs of settlement. Over time, however, the same networks can contribute to persistent spatial concentration if residents have limited opportunities or incentives to move elsewhere. The resulting neighbourhood may therefore reflect both agency and structural constraint. Researchers distinguish these mechanisms from forced segregation because the same spatial pattern can emerge from very different combinations of voluntary clustering and restricted choice.
Government policies and planning systems can also cause or reinforce segregation. Decisions about where to build social housing, how land is zoned, which neighbourhoods receive infrastructure investment and where transport routes are developed can influence the distribution of households. Large social housing estates can create concentrations of lower income households if similar properties are constructed in particular locations and subsequently allocated to households with high housing needs.
Conversely, restrictive planning policies in affluent areas can limit the construction of lower cost housing, reducing the ability of lower income households to enter those neighbourhoods. Urban renewal can alter the social composition of an area by replacing existing housing with higher value development or by relocating residents. Historical public housing policies, redevelopment programmes and infrastructure projects can therefore have long lasting spatial consequences. Policy effects may be intentional, incidental or contested, and understanding causation requires examining the objectives, implementation and consequences of the relevant policies.
School systems, employment geography and transport networks can reinforce residential segregation by making particular locations more attractive or accessible to some households than others. Families may seek neighbourhoods associated with particular schools, while workers may choose locations that reduce commuting time or transport costs. If high paying employment is concentrated in particular districts and transport connections from poorer neighbourhoods are weak, households with limited resources may face restricted access to employment rich areas.
Similarly, neighbourhood amenities such as parks, healthcare, shops, cultural facilities and environmental quality can affect demand. These relationships can become cumulative. Higher demand can attract investment, improve amenities and increase property values, which may then further restrict access to higher income households. Lower demand can have the opposite effect in areas experiencing disinvestment. Residential segregation can therefore be produced by a feedback relationship between housing markets, public infrastructure, employment and neighbourhood reputation rather than by housing decisions in isolation.
Historical processes are particularly important because urban landscapes accumulate the consequences of earlier decisions. Industrialisation, colonialism, migration, racial or ethnic exclusion, redevelopment, deindustrialisation and changes in transport can establish spatial patterns that persist across generations. Former industrial districts may contain cheaper housing because employment declined, while affluent areas may retain high property values because of accumulated investment, attractive architecture, green space or proximity to employment centres. Historical housing discrimination can also produce unequal ownership rates and wealth accumulation.
Once a spatial pattern becomes established, residents, businesses and institutions adapt to it, creating path dependency. Schools, shops, places of worship and community organisations may develop around established populations, making relocation less attractive or more difficult. Consequently, present day segregation cannot always be explained by current household choices or current income differences alone. Historical geography can continue to influence contemporary housing opportunities even after the original mechanisms have changed.
Residential segregation can also be reinforced by feedback mechanisms and perceptions. Neighbourhood reputations influence how households, landlords, lenders and investors evaluate locations. If an area becomes associated with poverty or a particular population group, demand from other groups may decline, potentially affecting property values and investment. Conversely, affluent reputations can increase demand and accelerate exclusion through rising prices. These processes can operate alongside stereotypes and prejudice, although perceptions should not automatically be treated as accurate descriptions of neighbourhood conditions.
The important distinction is between the observable spatial pattern and the mechanisms that maintain it. Residential segregation may therefore be self reinforcing because housing markets, social networks, institutions and perceptions interact over time. No single cause is sufficient in every case. A comprehensive explanation normally considers economic inequality, housing supply, discrimination, household preferences, migration networks, planning, public policy, transport, employment geography and historical development together, while recognising that their relative importance varies between places and populations.
