One of the key objectives of government neighbourhood policy is to encourage a sustainable mix of tenures and incomes. This report addresses questions of why integration has been so difficult to achieve in practice and draws conclusions for future policy.
The report analyses data from three related empirical studies:
- the first models, locally, the links between housing, labour markets, migration, deprivation and segregation;
- the second examines the factors behind the individual moving decisions that lie at the heart of segregation and how policy can influence choices;
- the third presents three case studies. These are the first empirical studies of their kind to show how segregation and deprivation arise.
Specifically, the report looks at:
- patterns of segregation and how they have changed over the last 20 years;
- the effectiveness of a range of policies, identifying those that are most likely to work;
- other factors that are key to improving integration.
Economic segregation in England: causes, consequences and policy
This detailed analysis of patterns of segregation explains why it has been difficult to achieve a sustainable mix of tenures and incomes. It identifies both the underlying patterns of deprivation and why people choose to move to particular areas. These need to be taken into account if the key policy objective of mixed communities is to be delivered. The research was carried out by Geoffrey Meen, Kenneth Gibb, Jennifer Goody, Thomas McGrath and Jane MacKinnon. Some key findings are:
- Patterns of segregation in England have changed little over the past 20 years or more.
- Evidence confirms that ‘one-size-fits-all’ policies do not work. Different areas need different policies. Areas with very high levels of deprivation need intensive help to reach a ‘take-off’ point before the private sector is likely to become involved. Otherwise, they become stuck in a poverty trap, segregated from other parts of the community.
- But the resources required to reach the take-off point are large in the most deprived areas.
- Segregation and integration depend particularly on where young, high-income households – the most mobile group – choose to move to. Internationally, some of the fastest growing cities have attracted these groups. They are attracted by facilities such as adequate sporting and cultural centres but deterred by areas of high deprivation, unemployment and council taxes. Policies, therefore, have to promote virtuous circles, to avoid the cumulative processes of decline that have been observed historically.
- It is particularly difficult to design policies to attract back older households to cities in order to promote integration, because people tend to move home significantly less as they get older. In general, once households have left urban areas, most tend to stay away.
In January 2005, the ODPM unveiled its Five-Year Plan for neighbourhood revitalisation, including the development of sustainable mixes of tenures and incomes in local communities. But this policy is aimed at counteracting deep-seated trends: there has been little evidence of a reduction in segregation over the last twenty years. This research attempts to understand the processes that lead to segregation and to identify the key policy instruments that are most likely to improve integration.
By ‘segregation’ the study means the tendency for households of similar economic status to be concentrated in common parts of any local authority district or neighbourhood.
The conclusions come from three pieces of empirical evidence:
- a local modelling exercise, which explains house prices, deprivation, incomes, employment status, migration and segregation and how they are interrelated.
- analysis of individual data from the British Household Panel Survey to look at the factors affecting why different household types decide to move to particular areas, for example, sporting and cultural factors, availability of good schools etc.
- three case studies, examining the extent to which mixed neighbourhoods remain mixed over time.
Deprivation, segregation and local housing markets in England
Patterns of segregation differ considerably according to which type of indicator is used. For example, the most segregated communities in terms of unemployment are not the same as those most segregated in terms of tenure or educational qualifications.
In terms of unemployment:
- segregation is heavily concentrated on the older industrial, northern and midlands areas.
- there is little, if any, evidence that segregation has declined over the last twenty years.
But segregation is much more concentrated in terms of tenure than unemployment.
Patterns of segregation and deprivation depend on a complex set of inter-related forces, covering the operation of local housing markets, labour markets and how people move between different areas.
At the local level, areas expand or contract in the main according to the movement of households in or out the area. Movement is very responsive to local housing market conditions. These, in turn, are strongly related to local levels of deprivation and the state of the labour market.
But a particularly important result from the study is that the housing market does not respond in a straightforward manner to changes in the level of deprivation. There are powerful ‘thresholds’ operating which affect whether and how areas change. Deprivation has to fall to a certain point before the local housing market begins to take off and attract into the area the necessary high-skilled individuals and also the private sector capital for an area to thrive, which will in turn generate virtuous circles.
This is why ‘one-size-fits-all’ policies are no panacea. If areas fail to reach the ‘take-off threshold’, households become stuck in a poverty trap and are likely to become segregated from the surrounding community. Areas of persistent low demand also arise. There is an analogy in the field of international development. The poorest countries also have to be helped to reach a critical stage of development or take-off point by government action before market forces begin to produce improvements.
Since one-size-fits all policies do not work, neighbourhoods for action have to be identified. Those spatial areas that can be targeted for action most easily are those that already lie close to the take-off point.
Furthermore, it is both difficult and expensive to bring areas experiencing very high levels of deprivation to the take-off point. But, geographically, the most deprived areas are usually concentrated in very small areas; rarely is it the case that whole local authority wards fall short of the ‘take-off threshold’ for moving out of deprivation.
According to the study’s calculations, only about 1 per cent of Census Output Areas in the Northern regions suffer from extreme deprivation and noticeably fewer in the South East. However, looking in detail at Harpurhey in Manchester (one of the Government’s areas for targeted action), the research suggests that unemployment, the percentage of the population with long-term illness and the percentage with no qualifications would all have to fall to the Manchester average for the ward to take off. This is a very tall order and emphasises the scale of some of the difficulties in reducing segregation.
Who moves and what determines their choices?
Patterns of segregation depend on the choices made by millions of separate households, and on what constrains these choices. Governments may provide incentives for households to live in certain areas, but, in the end, households are free to choose subject to the constraints of their budgets and incomes. In addition, their choice is typically influenced by demographic characteristics. Young, single-person households, for example, will make different decisions from older households with children.
But which households are most likely to move and where are they most likely to move to? A great deal is already known about the characteristics of movers, but the implications have not always been fully appreciated:
- The availability of housing suitable for them constrains which location people choose, but more general economic and neighbourhood conditions are of equal if not greater importance. Low-income households tend to be trapped in the worst locations, whereas high-income households can more easily escape. Polarisation ensues. The study looked at what factors influence people’s decisions about where they move to. The results indicate that local authority district and ward moving patterns can be destabilising for an area and, hence, contribute to segregation.
- Young, high-income, highly qualified individuals without children are the most mobile.
- However, most moves are short distance. Trying to induce long-distance moves – for example, from the South to the North – in order to promote greater equality is difficult.
- Similarly, since the propensity to move falls sharply with age, attracting middle-aged people to particular locations to improve social mix is problematic.
- Because most movers are young, there is evidence that the quality of sporting and cultural facilities affects their choice of location. However, young, high-income groups also try to avoid areas of high unemployment and high council tax bills.
- The influence of local schools on people’s choice of location is, in fact, difficult to assess from this sample. Education is clearly important but, in the sample of movers in the study, most are young without children: schools appear to be of less importance at this stage in their lives. Of course, schools are likely to become much more important when they have children.
- Internationally, the populations of many major cities have expanded in recent years, because of an increased concentration of young people with high incomes in centres. But the facilities provided in city centres are not always suitable for those with children.
Do ‘mixed communities’ remain mixed?
Policies may introduce social housing into predominantly owner-occupied areas or vice versa, but there is no guarantee that communities remain mixed over time. In other words, as individual households move, old tenure patterns may reappear.
The study looked at three case study areas of differing sizes – Werrington in Peterborough, Newbiggin Hall (Newcastle upon-Tyne), Hulme (Manchester) – to see whether attempts to promote mixed communities are permanent. In fact, despite the fact that these areas have seen policies to promote mix in operation for many years, the case studies suggest that it is still too early to tell whether the tenure structures of Hulme and Newbiggin Hall have settled into a stable pattern. Only Werrington exhibits notable stability over the last ten years, dominated by owner-occupation.
Promoting mixed communities
The research suggests a set of preconditions for the development of mixed communities:
- Neighbourhoods with high levels of deprivation need to reach the take-off point. Since deprivation is strongly related to local unemployment, long-term illness and lack of educational qualifications, any policy to address deprivation almost certainly needs to improve these significantly.
- Reducing local deprivation requires both vibrant national, regional and local economies and an improvement in the local skills base. Therefore, area policies need to go hand-in-hand with labour market policies. The former support the latter in discouraging those whose skills improve from leaving deprived areas.
- Incentives should be targeted on those most likely to move into the area, i.e. young and highly skilled people without children. The probability of attracting back older households who have already left urban areas is much lower.
- The local authority needs to decide whether its strategy is to retain young ‘newcomers’ as they grow older or to accept that they will move elsewhere with time and so focus on attracting the next cohort of young people into the area. The former strategy has the advantage of improving social networks as residents stay longer: however, the authority will have to provide high quality schools and other elements of infrastructure to retain incomers as they have children.
- Even where there is observed housing need, building more social housing in the most deprived areas will simply concentrate deprivation and segregation further. Currently, however, most new social housing is built in areas where the social stock is already the largest.
About the project
The project was carried out by Geoffrey Meen and Thomas McGrath (University of Reading), Kenneth Gibb and Jane MacKinnon (University of Glasgow) and Jennifer Goody (Peter Brown Partnership). The empirical modelling parts of the project used data from the 2001 Census, the British Household Panel Survey and national accounts. The case studies involved analysis of local government and associated records Interviews were carried out with local government housing, planning, neighbourhood/community leaders and local market experts. Focus groups were carried out with residents.