Does gentrification displace poor children and their families? New evidence from medicaid data in New York City

Kacie Dragan, Ingrid Gould Ellen, Sherry Glied

Research output: Contribution to journalArticle


The pace of gentrification has accelerated in cities across the country since 2000, and many observers fear that it is displacing low-income populations from their homes and communities. We offer new evidence about the consequences of gentrification on mobility, building and neighborhood conditions, using longitudinal New York City Medicaid records from January 2009 to December 2015 to track the movement of a cohort of low-income children over seven years, during a period of rapid gentrification in the city. We leverage building-level data to examine children in market rate housing separately from those in subsidized housing. Using multiple definitions, we find no evidence that gentrification is associated with meaningful changes in mobility rates over the seven-year period, though it is associated with slightly longer distance moves. As for changes in neighborhood conditions, we find that children who start out in a gentrifying area experience larger improvements in some aspects of their residential environment than their counterparts who start out in persistently low-socioeconomic status areas. This effect is driven by families who stay in neighborhoods as they gentrify; we observe few differences in the characteristics of destination neighborhoods among families who move, though we find modest evidence that children moving from gentrifying areas move to safer neighborhoods but lower-quality buildings.

Original languageEnglish (US)
Article number103481
JournalRegional Science and Urban Economics
StatePublished - Jul 2020

ASJC Scopus subject areas

  • Economics and Econometrics
  • Urban Studies

Fingerprint Dive into the research topics of 'Does gentrification displace poor children and their families? New evidence from medicaid data in New York City'. Together they form a unique fingerprint.

  • Cite this