Displacement and gentrification in England and Wales: A quasi-experimental approach

Lance Freeman (Columbia University, USA), Adèle Cassola (Columbia University, USA), Tiacheng Cai (Columbia University, USA)

Abstract can be found: http://usj.sagepub.com/content/early/2015/09/03/0042098015598120

The question of gentrification’s impact on low-income households remains a topic of heated debate among urban researchers and residents. In recent years, numerous studies have been dedicated to examining one of the most traumatic potential outcomes of gentrification – the direct displacement of the neighborhood’s existing residents. Several U.S.-based studies have examined whether gentrification leads to displacement using quasi-experimental methods, which try to account for what would have happened in the absence of gentrification by comparing mobility rates in neighborhoods that are similar but for experiencing gentrification. These studies have found no consistent evidence that low-income households are more likely to move out of gentrifying than non-gentrifying neighborhoods. Our study was motivated by the notion that because people move less frequently on average in England and Wales than in the US, it might be easier to distinguish patterns of elevated mobility due to gentrification in this context.

Using the British Household Panel Survey, we compared households’ odds of moving in three types of neighborhoods: disadvantaged neighborhoods that did not gentrify between 2001 and 2009; disadvantaged neighborhoods that gentrified during this period; and relatively advantaged neighborhoods. For the entire sample of England and Wales, low-income and working-class households living in gentrifying neighborhoods were not more likely to move than comparable households in neighborhoods that did not gentrify. In London, on the other hand, low-income households in gentrifying neighborhoods were more likely to move than similar households in non-gentrifying neighborhoods. However, because this finding did not hold up when different measures of gentrification were used, we cannot make a strong case that gentrification leads to displacement based on this analysis.

The lack of compelling evidence of gentrification-induced direct displacement in this and previous quasi-experimental studies is hard to reconcile with first-hand accounts of direct displacement. Displacement from gentrifying neighborhoods clearly occurs, so why have so many statistical studies failed to detect higher rates of mobility among vulnerable residents from these neighborhoods?

We suggest the following explanation. Turnover rates tend to be higher in disadvantaged neighborhoods, but when such neighborhoods gentrify, some residents who would otherwise have left might decide to stay put because they like the neighborhood’s trajectory. Others might dislike the changes and move. Others might wish to move, but find no acceptable alternatives. Still others will be directly displaced from the neighborhood. However, the increased mobility due to direct and indirect displacement and the reduced mobility among those who stay put may balance each other out. Consequently, the overall rate of mobility may differ little from what existed prior to gentrification.

A failure to statistically detect direct displacement therefore does not mean that we can write off gentrification as a policy concern. Moreover, direct displacement is not the only form of displacement experienced by residents of gentrifying neighborhoods. Those who are not directly displaced may nonetheless feel alienated by the changes occurring in their neighborhood. Gentrification may also reduce the stock of low-cost housing in affected neighborhoods, thus excluding low-income households that otherwise would have moved in. Given the complexity of these processes and their enduring impact on urban residents and neighborhoods, there is no doubt that gentrification and displacement will continue to inspire much debate and research in the years to come.