The most accurate data set around is the decennial national census. There are around 8000 electoral wards in England. In 1991, 57 wards had a minority white population and 15% of all non-white residents lived in them. In 2001, 118 wards had a minority white population and 23% of non-white residents lived in them. In the year before the last census in 2001, more non-white residents moved out of these 118 wards than white ones (14,716 verses 9747 respectively).
So we don’t have self-segregation at all. We have the mundane phenomenon of dispersal.
First, white and non-white residents move out of the inner cities when they can afford better housing and commuting costs. This usually happens in middle age. So if everyone generally moves when they can afford to, it can’t necessarily be put down to cultural tensions.
Second, the number of mixed neighbourhoods (or electoral wards) is increasing. Between 1991 and 2001, they grew from 864 to 1070. Also minority white wards are also still mixed wards: they are not segregated. So we’re getting less not more segregated.
Third, inequalities experienced by non-white residents whether in majority-white, mixed or minority-white wards are broadly similar. This shows that geography and ethnic mix are not salient factors in creating inequality. The employment rate of non-whites is roughly twice as high as whites in all these three sorts of ward. So ethnic differentials in poverty aren’t a function of these mythical ghettos either!
So it seems that if there aren’t really any no-go areas as such, just the ones that people like the Bishop of Rochester like to dream up in their heads.
Source: Ludi Simpson, “The Numerical Liberation of Dark Areas”, Sage Race Relations Abstracts, 31/2: 5-25 (2006).