Working Paper: NBER ID: w28157
Authors: John C. Haltiwanger; Mark J. Kutzbach; Giordano E. Palloni; Henry Pollakowski; Matthew Staiger; Daniel Weinberg
Abstract: We combine national administrative data on earnings and participation in subsidized housing to study how the demolition of 160 public housing projects—funded by the HOPE VI program—affected the adult labor market outcomes for 18,500 children. Our empirical strategy compares children exposed to the program to children drawn from thousands of non-demolished projects, adjusting for observable differences using a flexible estimator that combines features of matching and regression. We find that children who resided in HOPE VI projects earn 14% more at age 26 relative to children in comparable non-HOPE VI projects. These earnings gains are strongest for demolitions in large cities, particularly in neighborhoods with higher pre-demolition poverty rates and lower pre-demolition job accessibility. There is no evidence that the labor market gains are driven by improvements in household or neighborhood environments that promote human capital development in children. Rather, subsequent improvements in job accessibility represent a likely pathway for the results.
Keywords: Hope VI; public housing; labor market outcomes; demolition; job accessibility
JEL Codes: I38; J13; J31; J62; R23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Hope VI demolition program exposure (R28) | Earnings at age 26 (J31) |
| Hope VI demolition program exposure in large metro areas (R28) | Earnings at age 26 (J31) |
| Hope VI demolition program exposure in smaller authorities (R28) | Earnings at age 26 (J31) |
| Hope VI demolitions (R28) | Job accessibility (J62) |
| Job accessibility (J62) | Earnings at age 26 (J31) |