Key points are not available for this paper at this time.
Income segregation has traditionally been studied through residential patterns, yet growing evidence points to its dynamic manifestation in the everyday mobility of individuals. Using large-scale mobile phone data from 12 U.S. metropolitan areas, we examine segregation in population congregation – defined as the unequal distribution of income groups across micro-geographic units during daily movement – and identify its structural and behavioral origins. We propose a decomposition framework that distinguishes the contributions of static income disparities and dynamic mobility heterogeneity to observed segregation levels. Segregation is operationalized via disparities in visitation patterns at the Census Block Group level, while mobility heterogeneity captures variations in the spatial concentration of individual travel flows. Shapley–Owen decomposition shows that segregation is dominated by income-group-specific mobility heterogeneity before COVID-19 response measures, led by the lower-middle income component. After the implementation of response measures, mobility heterogeneity remains the dominant channel and explains most of the between-period change, whereas structural inequality contributes a secondary background effect. These findings reveal how economic stratification and adaptive mobility patterns jointly configure urban inequality, underscoring the need for policy responses that integrate both structural and behavioral dimensions in times of disruption. • Proposes a decomposition framework linking income inequality and mobility behavior. • Identifies income groups driving segregation in daily population congregation. • Mobility heterogeneity dominates pre/post COVID response and drives change. • Lower-middle income (Q2) destinations are leverage points for cross-group mixing.
Li et al. (2026) studied this question.