The Journal

The Subjects

The Books





Displacement through gentrification: How big a problem?

Rowland Atkinson

Displacement through gentrification occurs when neighbourhoods change such that inflated rents and prices push out the low paid or the unpaid. In a more subtle way the process may involve large influxes of professionals which alter the characteristics and services of an area so that resident's social networks change and the cost of living rises. The precise extent of displacement is contentious, with little research in this area, providing scope for an interpretation of the flows of people involved. Wrangles over accepted levels of the problem have taken on a distinctly ideological bent in North America whereas the existence of such a problem in Britain has not been recognised.

The destination and living circumstances of displacees after gentrification has been documented: many move to more expensive or overcrowded accommodation: 80-85% of displacees had to pay more for worse accommodation (Hartman, 1979). Moves are often made to locations nearby but to property which is in a worse condition. Often moves are made to the households of friends or relatives, which accounts for much of the observed overcrowding and the closeness of displacee moves. All of this suggests that displacement is a problem since it has psychological effects, particularly on the elderly.

The conceptual and empirical debates surrounding the measurement of both gentrification and displacement make causality very difficult to establish and figures on displacement are contentious. Such a research hiatus in this country has left a potentially large issue of social justice and forced migration 'defined out' of debate.

What is gentrification and how can we define displacement?

Gentrification can be defined as 'the rehabilitation of working-class and derelict housing and the consequent transformation of an area into a middle-class neighbourhood' (Smith and Williams, 1986:1). Displacement can be defined as a process whereby households have their housing choices constrained by the actions of another social group. There are different forms of displacement and correspondingly different measures of the number of people affected. Marcuse (1986) distinguishes four types: a) Economic/Physical, where residents priced out or subjected to violence are estimated b) Last resident, counting only the last resident c) Chain, counting the residents, over time, which have been displaced from a property or area and d) Exclusionary, where the number of people is estimated who are unable to move into property which has been vacated voluntarily yet subsequently gentrified. Critically the choice of definition affects observed levels. The production of data has also been hampered by the difficulty of 'tracking' displacees; it is also clear that not all gentrification involves displacement, further complicating any measurement.

Displacement is not simply an issue of the political manipulation of available data, since in this country there is very little available data. Further, what data exists is based on assumptions about the constitution of gentrification and of displacement. In the US displacees have been characterised as low income, white working class, the elderly and ethnic minorities. It is probably no coincidence that these groups also neatly fit census categories that may be used to measure their incidence and location over time. Establishing causal links between gentrification and displacement is made difficult because it is presupposed that the market allocation of housing is the only way that housing should be allocated and that fairness does not play a part in such allocations. This would be less of a moot point if affordable and/or public housing left islands of affordability in these 'hot' property markets.

These issues have implications for any methodology set up to measure levels of displacement since it becomes very difficult to adequately operationalise the concepts. Displacement also affects more people than those who are directly displaced. There is an effect on other residents who see their area changing, businesses closing to make way for more expensive services and friendship and kinship networks being affected as people move away. Methodologically and politically this form of displacement is important because figures of displacement produced by using before-and-after measurements exclude people affected in the interim. The number of people displaced from a property may be many over a long period if rents and prices continue to rise.

Political problems

In the US, the ideological and physical desirability of gentrification may also be linked to the local taxation system since revenues are generated locally. The desirability of the regeneration of once run-down areas is also tied up in a discourse which legitimates displacement, considered a minor problem, in order to achieve competitiveness in a global economy, as witnessed in New York's Mayoral programme of 'zero-tolerance'. Thus the influx of higher income residents to an area was seen as positive, as was the rehabilitation of the inner urban environment. It can also be argued that benefits have accrued to owner occupiers in gentrified areas who may have seen the value of their property rise, though this presupposes a desire to move from the area. Should these concerns be considered when trying to measure displacement?

Writers like Lee and Hodge (1984), who distinguish between liberal and conservative definitions of displacement, show how the definitions used affect the perceived magnitude of the phenomenon. For the conservative, only the more extreme forms of harassment and eviction will be considered to be a displacement pressure. To the more liberal mind, displacement is a process that includes the pricing out of residents and the changing of shops and services. With this tangle how should one proceed in trying to create a benchmark figure?

How much displacement is there?

Quantification has also been subject to controversy among researchers according to their social position, government sponsored research has been more conservative in its estimates of displacement than 'independent' research. While Britain has produced little literature directly related to displacement the literature of the US has proliferated due, for the most part, to funding by central government and the use of official and commercial housing survey data.

Marcuse (1986) found displacement in New York City to be between 10,000 and 40,000 households per year. Such figures are obviously linked to the prevalence of gentrification activity at any point in time. LeGates and Hartman (1986) indicate that an 'approximate and conservative' total annual displacement figure for the US amounted to 2.5 million persons. LeGates and Hartman also cite the growing awareness of the problem by the government at that time reporting to Congress that 2.4 million people were being annually displaced. Redevelopment Authority files from Society Hill in Philadelphia showed that 6,000 residents had been displaced since 1959 to make way for gentrification (Smith, 1996). The government-employed Sumka (1979) indicated that, annually, 500,000 US households were displaced (approximately 2 million people) sparking debate between him and Hartman (1979) over which was the overestimate.

In Britain, Leckie (1995) has estimated an annual figure of 144,000 people being forcefully evicted each year. He estimates that a further 60,000 or more will be evicted annually in the future, though only a certain proportion of these figures relates to gentrification-induced displacement. It is also evident that the former figure is taken from an OPCS commissioned study (Rauta and Pickering, 1992) which gave the figure of 1 in 10 tenants being harassed each year (based on a question which asked if the tenant had been made in any way uncomfortable by their landlord). In addition, 2% of all tenants in the survey had experienced landlords who had tried to evict them in 'other ways', a euphemism for harassment. Not all displaced tenants report their eviction at some form of help centre where they might be registered as displaced. Local authorities in Britain only keep records of those in certain categories who approach them as homeless: the unintentionally homeless, pregnant women, families, disabled and the elderly. This means that local authority records will be an inadequate measure. The term gentrification poses negative and potentially disastrous political connotations so that research in this area must be tactful and often side-step such a conceptual label. Interviews conducted in London indicated that the local authority approach was made in terms of individual social problems; the role of a housing department or rent officer is not to look for the reasons for such people asking them for help, so there is no long view.

In the 1970s, the Department of the Environment (McCarthy, 1974:3) carried out a survey in twelve inner London boroughs to establish what was happening to people living in grant-rehabilitated housing. They found that these improvements did not benefit the original residents since 68% of applications had been preceded by the outward movement of these households. By far the largest reason for the household moves was landlord harassment (43%). The process was described as a 'social sieve'. Lyons (1995) finds that local migration is associated with low status households, while longer range migration may be associated with higher status, indicating constraint and choice respectively but with no quantification involved. Studies of displacement in the US, Canada and Australia show quantification can be achieved.


The age of the references used here indicates the need for more work in this area, especially as the definition and measurement of gentrification is so contentious. This is unfortunate when we consider the costs to the displaced, and may reflect that debate is often of a political rather than an academic nature. While other problems, like homelessness, may appear more important, we should not forget the linkages between gentrification, displacement and homelessness. Our lack of knowledge and attention to this problem may mean that it has been dramatically underreported and that the focus has been on urban symptoms rather than causes.


Hartman, C. (1979) Comment on 'Neighbourhood revitalization and displacement: A review of the evidence', Journal of the American Planning Association, 45(4): 488-491.

Leckie, S. (1995) When Push Comes to Shove: Forced Evictions and Human Rights, Utrecht: Habitat International Coalition.

Lee, B. And Hodge, D. (1984) Social Differentials in Metropolitan Residential Displacement, in Palen, J. and London, B. (eds.) Gentrification, Displacement and Neighbourhood Revitalization, Albany: State University of New York Press.

LeGates, R. and Hartman, C. (1986) The Anatomy of Displacement in the United States, in Smith, N. and Williams, P. (eds.) Gentrification of the City, London: Unwin Hyman.

Lyons, M. (1995) 'Professionalisation, polarisation and feminisation in London', paper presented at ESRC-sponsored seminar, London School of Economics, London.

Marcuse, P. (1986) Abandonment, Gentrification and Displacement: the Linkages in New York City, in Smith, N. and Williams, P. (eds) Gentrification of the City, London: Unwin Hyman.

McCarthy, J. (1974) Some Social Implications of Improvement Policy in London, Internal paper, Social Research Division, Department of the Environment.

Rauta, I. And Pickering, A. (1992) Private Renting in England, 1990, London: HMSO.

Smith, N. (1996) The New Urban Frontier: Gentrification and the Revanchist City, London: Routledge.

Smith, N. and Williams, P. (eds) (1986) Gentrification of the City, London: Unwin and Hyman.

Sumka, H. J. (1979) Neighbourhood revitalization and displacement. A review of the evidence, Journal of the American Planning Association, 45(4): 480-487.

Rowland Atkinson
Department of Urban Studies,
25-8, Bute Gardens
University of Glasgow,
Glasgow G12 8RS tel: 0141 330 5048 x 2522


Journal 069 Index Top of page