Why Did Poverty Become Less Geographically Concentrated in the 1990s?*

Objective. After increasing sharply in the 1970s and 1980s, the number of high‐poverty neighborhoods in the United States unexpectedly and dramatically declined in the 1990s. This article examines the roles that residential and income mobility played in this decline. Methods. Using geocoded data from the Panel Study of Income Dynamics, this study analyzes changes during the 1990s and early 2000s in: (1) patterns of residential mobility between high‐poverty and lower‐poverty neighborhoods; and (2) patterns of income mobility for residents who remained in high‐poverty neighborhoods. Results. Both patterns of residential and income mobility changed in the 1990s and early 2000s. While patterns of residential migration to high‐poverty neighborhoods were largely unchanged over this period, patterns of residential migration from high‐poverty neighborhoods changed significantly, with poor individuals—especially poor blacks—becoming more likely to relocate from high‐poverty to lower‐poverty neighborhoods. Patterns of income mobility for residents who remained in high‐poverty neighborhoods also changed significantly, with nonpoor residents becoming less likely to become poor and poor residents becoming more likely to exit poverty. Conclusion. Poverty rates in high‐poverty neighborhoods fell primarily because of the net upward income mobility of residents who remained in these neighborhoods.