Inadvertent and intentional partisan residential sorting

We present evidence for two mechanisms that can explain increasing geographic divide of partisan preferences. The first is “inadvertent sorting,” where people express a preference for residential environments with features that just happen to be correlated with partisanship. The second is “intentional sorting,” where people do consider partisanship directly. We argue that the accumulating political biases visible in many neighborhoods can be the effect of some mixture of these two mechanisms. Because residential relocation often involves practical constraints and neighborhood racial composition is more important than partisanship, there is less partisan segregation across the USA than there could be based on residential preference alone.