Groups' contribution to shaping ethnic residential segregation: a dynamic approach

Some have argued it is possible to infer different groups' contributions to ethnic residential segregation from their individual neighborhood preferences. From this perspective, natives tend to be more segregation-promoting than non-natives, since they prefer neighborhoods where they are the majority. It remains unclear, however, whether this holds when one evaluates their contributions to segregation within a dynamic perspective. Using register data from Statistics Sweden, I define and model ten different groups' residential behavior based on their ethnicity and family composition. I thereby simulate the residential mobility of the full population of Stockholm municipality residents from 1998 to 2012. Even though my results at the micro-level are consistent with previous studies, the simulation results show that foreign singles' mobility patterns are more segregation-promoting than any other groups, since this group shows a greater in-group feedback effect regarding choice of new neighborhoods, an effect that increases their flow from low-to-high segregated neighborhoods progressively. My results suggest that (1) integration initiatives would be more efficient if focused on this particular group and (2) a proper evaluation of micro-behaviors' implications for macro-patterns of segregation requires a dynamic approach accounting for groups' heterogeneous behaviors and their main interdependencies on shaping segregation over time.