RESUMO
Ethnic and socioeconomic segregation levels vary over time and so do the spatial levels of these segregations. Although a large body of research has focused on how residential mobility patterns produce segregation, little is known about how changing mobility patterns translate into temporal and scale variations in sorting. This article develops a methodological framework designed to explore how changing mobility patterns reflect such trends. It introduces a measure of sorting that reflects the extent of disparities among groups in their socio-spatial mobility. Trends in the direction and the extent of sorting can be exposed by computing sorting measures over consecutive periods. The measure is broken down to capture the relative contributions of residential mobility to sorting at hierarchically nested geographical units, for example cities and their constituent neighbourhoods. An empirical demonstration shows that changes in residential mobility patterns affect the magnitude and spatial level of residential sorting, which vary even over the short term.
RESUMO
Neighbourhood socioeconomic change is a complex phenomenon which is driven by multiple processes. Most research has focused on the role of urban-level processes, which lead to an exchange of relative positions among neighbourhoods of a single metropolitan area. Consequently, the effects of structural processes on neighbourhood socioeconomic change, such as overall income growth or decline, and increasing inequality, have been neglected. This is reflected in the standard methodological practices; the common measures of neighbourhood change exclude the effect of overall growth or decline and confound the effects of urban processes with the effect of increase in inequality. This paper proposes a method that was originally developed for understanding income mobility of individuals, to decompose total neighbourhood socioeconomic change measured in absolute terms into its contributing components. The approach enables to take account of all processes that generate neighbourhood socioeconomic change, while distinguishing between them. The method is demonstrated in an empirical analysis of neighbourhood socioeconomic change across 22 metropolitan areas in the US. The findings indicate that structural processes can be most substantial in generating change. Neighbourhood socioeconomic change in 'superstar cities' is mostly generated by the growth in overall incomes, with a relatively low contribution of increasing inequality. Conversely, in declining cities it is mostly driven by overall decline and increasing inequality. An additional finding relates to the interaction between urban processes and increasing inequality. These processes work in opposite directions such that any increase in positions of low-income neighbourhoods can be totally offset by an income decrease due to increasing inequality.
