In this study, I examine the role of household income in determining White-Black, White-Latino, and White-Asian residential segregation in the twenty-first century across 50 metropolitan areas over the decennial time points from 2000 to 2020. I use census and survey microdata and a reformulation of the separation index, a measure of the segregation dimension of evenness, to situate segregation as a group inequality that can be analyzed using group-specific household-level regression models and regression standardization analysis, where household income is the predictor of segregation-relevant neighborhood outcomes. I find that across groups, across communities, and over time, the role of household income in shaping segregation patterns varies widely. White-Black segregation is lowest between high-income households and is declining consistently for all income groups, even at mismatched incomes. White-Latino segregation patterns are more inconsistent, with segregation staying low and stable for high-income Latino households but rising somewhat for low- and middle-income households. Finally, White-Asian segregation is rising and has risen the fastest for high-income Asian households. These findings call for continuing investigation into the shifting and interlocking dynamics of race and income that shape segregation outcomes.