Racial Disparity in Speed of Housing Recovery after Hurricane Katrina: Ten Years Later

Background: Research suggests that pre-disaster sociodemographic characteristics, such as race, predisposed disparate housing recovery after Hurricane Katrina at the community level. However, how quickly damaged homes were restored after the disaster across individual homeowners with different racial profiles was unknown. Specifically, Road Home Grant Program (RHP) was a major public housing assistance program provided in New Orleans Metropolitan Statistical Area (NOMSA). Yet, even ten years after Katrina, it remained unclear to what extent RHP was distributed equitably by race and income or how RHP receipt influenced the speed of housing recovery. This study asked: Did homeowners with different racial profiles achieve similar speeds of home recovery after Hurricane Katrina after controlling for the covariates of family income, damage level, and private insurance? Did access to RHP differentially affect the speed of home recovery for blacks compared to non-blacks?

Method: Using data collected by the American Housing Survey in the NOMSA in 2004 and 2009 (N= 1,177), we used Kaplan-Meier estimates to plot the post-Katrina instantaneous speeds of home rebuilding and repair for black and non-black homeowners in relation to the events of: 1) starting to rebuild; 2) starting to repair; 3) completing rebuilding; and 4) completing repairs. Using a semiparametric model with partial likelihood estimation method, we also employed Cox regression to investigate how RHP and race affected the timing of each event.

Findings: First, non-black homeowners more quickly started and completed rebuilding and repairing their damaged homes than black homeowners. However, after controlling for pre-and-post-hurricane family income and post-Katrina damage level, the racial disparity in speed of completing rebuilding and repairs disappeared. That said, although family income did not impact home recovery speed in terms of starting rebuilding and repairs (events 1+2), it sped recovery in terms of completing rebuilding and repairs (events 3+4). Second, black homeowners were more likely to utilize RHP public financial assistance than non-black homeowners, whose higher incomes provided greater access to private insurance. Thus, because the RHP application period was delayed for 12-18-months after the disaster, black recipients experienced slower starts to damage repair than non-black recipients overall.  Further, many black recipients never even began to rebuild because they could not afford the cost, and thus opted to sell their damaged homes to the Louisiana government.

Implications: RHP facilitated starting rebuilding for non-black homeowners, who also had higher incomes overall, which enabled them to afford private insurance that was immediately available after the hurricane. In contrast, with RHP's option to buy back damaged houses from homeowners, the program itself suppressed black owners' ability to rebuild because they experienced higher levels of home damage, received less overall financial assistance, and had lower family incomes than non-black owners. Consequently many gave up rebuilding.  In addition, the “un-timely administration” of public program provision in NOMSA retarded the home repairing of black compared to non-black homeowners. Because income level together with race affected completing rebuilding and repair, emergency public programs such as RHP should be more financially generous and implemented at the same speed as private assistance programs.