London lost over 600,000 White British people to the rest of the UK while about 1.6m of other ethnic backgrounds moved in. Whether examined at metropolitan, Local Authority or ward level, urban places where non-White British people moved to in the 2000s tended to lose White British people. This is counterintuitive since booming areas should be drawing in both White British and minority folk. Why would an area attract one ethnic group(s) and repel another? One answer is that poor people cluster in diverse urban areas to reduce transport costs to their often low-paid and insecure employment. They lack the resources to purchase larger properties in the leafy suburbs. White British outflow is a sign of upward mobility, linked to greater resources. This outflow has been a feature of British urban life for years. Indeed, the last time London drew in more White British people from the rest of the UK than it expunged was the 1870s. This harks back to a time when manufacturing and cramped slum accommodation were concentrated in unhealthy, polluted cities.

Those who could escape did, but many could not. This ‘counterurbanisation’ framework remains central to British geography and urban studies (Champion 2003; Catney and Simpson 2010). As applied to ethnic movements it resembles the American literature which posits that immigrants and their descendants disperse to mixed suburbs as they ascend the economic ladder (Logan and Zhang 2011). This seems, at least, to characterise the twentieth century experience of ‘white ethnics’ such as Irish, Italian or Jewish Americans. The problem with this materialist interpretation is that many cities such as London, which once lost population, have rebounded to become attractive and expensive places to live. It is harder to claim that poverty and transport issues are the only, or even the main, reason that minorities locate there. If we examine two British individuals originating in the same area, one White British, one minority, but alike in income, marital status, age and education, the minority person will move to a significantly more diverse area than the White Briton (Kaufmann and Harris 2014).

While we cannot know if the specifically ethnic component of mobility behaviour stems from whites or minorities, there are indications that White British movements are very important. For instance, many ethnic minorities such as Bangladeshis or Afro-Caribbeans have been leaving their areas of concentration and entering ‘superdiverse’ mixed-minority zones such as much of Newham or Wembley (Johnston, Poulsen et al. 2013). However, at the same time, White British have been leaving superdiverse areas, i.e. Lozells, Birmingham, just as quickly as places dominated by one or other minority group, i.e. Oldham. In fact the only ethnic group to become more segregated in London in the 2000s are the White British (Catney 2012). As others move toward ethnic strangers, the White British seem to be moving away. A similar pattern appears in the US, with whites avoiding not only ghettoes but also mixed neighbourhoods in which they are in the minority (Alba and Romalewski 2013). But if this intimates White Flight is occurring, it is awfully difficult to prove.

Experimental research using 12-house showcards depicting ideal neighbourhoods finds that British, American and Dutch whites prefer white majority neighbourhoods to mixed ones (Krysan 2002; van Londen 2012). Hardwood Flooring Step DownThose with racist views tell researchers they prefer lily-white neighbourhoods. Buy Wedding Dress In TrinidadOn the other hand, when it comes to actual moving, the effect seemingly disappears: whites who vote BNP, UKIP or Conservative, or express racist views, do not move to whiter wards than those voting Labour, Lib Dem or Green. Baker Cat T ShirtOr at least the effect is extremely small and nowhere near important enough to explain the ethnic displacement observed in urban Britain in the 2000s. One possibility are complex effects of the kind noted by American economist Thomas Schelling.

Schelling used the example of church notice boards to explain segregation. If whites attend white churches and blacks black churches, and if church noticeboards are where people advertise rooms for rent, then segregation will occur even if people are perfectly indifferent to the race of those they live beside. Likewise, if Muslims need to be near mosques and Jews near synagogues, then they will cluster together even if they have no wish to live amongst ‘their own’. White British who move near friends and family – or hear about areas to live from them – will tend to find themselves in whiter areas. Indians, by contrast, may have no contacts in Cornwall, so don’t move there. Yet even if one only looks at people who moved to areas where they have no friends or family, and when one controls for religious background, an important white-minority gap remains (UKHLS 2009-13). Discrimination is possible too, but it is notable that trends are if anything stronger in the owner-occupied than rental sector – and one might expect landlords to discriminate more than estate agents.

So the mystery persists. It is possible that ethnic preference behaviour is unconscious, operating beneath the radar of racial attitudes, much as behavioural economists would contend (Kahneman 2011). Regardless of cause, the effect of these movements is to limit integration in Britain. While the main indices of segregation the Index of Dissimilarity (ID) and Index of Isolation (II) – have been declining, this is only true for individual minorities such as Afro-Caribbeans, Pakistanis or Bangladeshis. It is not so for minorities aggregated together as a whole. Thus the white-minority ID has been stable over the past two decades. This equation is solved by the fact that minorities have moved toward each other, i.e. Afro-Caribbeans and Pakistanis to superdiverse areas, and out to white areas, but White British have left diverse places for whiter areas. In fact, two-thirds of White British movers who originated in the most diverse fifth of wards in England and Wales chose whiter wards to move to in the 2000s and only 12 percent of those in the white four-fifths moved to wards in the most diverse fifth.

The equivalent numbers for minorities are 23 percent to white wards and 40 percent to the diverse fifth, a major divergence. While the index of isolation is declining among all groups at the scale of the UK as a whole, minorities in diverse areas are becoming more isolated from the White British in those locations: 4.1 million non-European minorities (41% of the minority population) live in wards which average less than half white. This compares with about a million minorities (25% of the minority population) living in such ‘minority majority’ wards in 2001. This isolation may have nothing to do with self-segregation. It is a function of minority demographic increase and continual white departure, hence this local isolation will rise as the share of minorities increases. A similar pattern is evident in the US, where a growing share of the most homogeneous neighbourhoods are minority rather than white. While gentrification can take place in city-centre locations of attractive cities, such as Harlem in New York or Brixton in London, this is much less true of areas further from the urban core or in secondary cities.

The result is a growing urban periphery, or reverse ‘oreo cookie’ pattern, where, once whites have vacated, they cease to be replaced by new cohorts of whites (Logan and Zhang 2011). Whether minorities need to have contact with White British is a matter of debate: Pakistanis who live with Afro-Caribbeans, Poles and Bangladeshis must observe common norms which tend to stem from British culture. This said, people are often hired by those they know and there is evidence that minority-white employment and wealth gaps are larger in diverse London than elsewhere in Britain. The same is true of diverse Toronto and Vancouver compared to the rest of Canada. One thing is certain: the segregation debate will continue. RD Alba and S Romalewski, ‘The End of Segregation? a more nuanced view from the New York metropolitan region’, New York, NY: City University of New York, Center for Urban Research, 2013. Catney, G, ‘More segregation or more mixing?’, Centre on Dynamics of Ethnicity (CoDE), December 2012. 

Catney, G. and L. Simpson. ‘Settlement area migration in England and Wales: assessing evidence for a social gradient’, Transactions of the Institute of British Geographers, 35 (4), 2010, pp 571-584. Champion, T, ‘Testing the differential urbanisation model in Great Britain, 1901-91′, Tijdschrift Voor Economische En Sociale Geografie, 94 (1), 2003, pp 11-22. R Johnston, M Poulsen and J Forrest, ‘Multiethnic residential areas in a multiethnic country? A decade of major change in England and Wales’, Environment and Planning A, 45 (4), 2013, pp 753-759. Kahneman D,  Thinking, Fast and Slow. New York, Farrar, Straus and Giroux, 2011. Kaufmann E and Harris G, ‘Changing Places: mapping the White British response to ethnic change’. Logan J and Zhang W, ‘Global Neighborhoods: New Evidence from Census 2010′, US 2010. J. Logan, Russell Sage Foundation, 2011. van Londen M, ‘Exclusion of Ethnic Minorities in the Netherlands: The Effects of Individual and Situational Characteristics on Opposition to Ethnic Policy and Ethnically Mixed Neighbourhoods’.