Separate When Equal: Racial Inequality and Residential Segregation

Working Paper: NBER ID: w11507

Authors: Patrick Bayer; Hanming Fang; Robert McMillan

Abstract: This paper hypothesizes that segregation in US cities increases as racial inequality narrows due to the emergence of middle-class black neighborhoods. Employing a novel research design based on life-cycle variations in the relationship between segregation and inequality, we test this hypothesis using the 1990 and 2000 Censuses. Indeed, increased black educational attainment in a city leads to a significant rise in the number of middle-class black communities and segregation for older adults both in the cross-section and over time, consistent with our hypothesis. These findings imply a negative feedback loop that inhibits reductions in racial inequality and segregation over time.

Keywords: Racial Inequality; Residential Segregation; Neighborhood Formation

JEL Codes: H0; J7; R2


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Increased educational attainment among blacks in a city (I24)Rise in the number of middle-class black communities (R23)
Increased educational attainment among blacks in a city (I24)Residential segregation (R23)
Rise in the number of middle-class black communities (R23)Greater residential segregation (R23)
Increased educational attainment among blacks (I24)Greater residential segregation for older blacks (R23)
Narrowing racial inequality (I24)Emergence of new middle-class black neighborhoods (R23)
Emergence of new middle-class black neighborhoods (R23)Greater segregation (J79)
Changes in segregation between 1990 and 2000 (R23)Increases in the proportion of highly educated blacks (I24)
Neighborhood effects mechanism (R23)Positive correlation between segregation and inequality for younger blacks (I24)
Neighborhood formation mechanism (D85)Negative relationship for older blacks (J79)

Back to index