Rebuilding housing in New Orleans: the Road Home Program after the Hurricane Katrina disaster

Abstract

The flooding in New Orleans that followed Hurricanes Katrina and Rita damaged hundreds of thousands of homes. In response, the State of Louisiana created the Road Home Program using billions of Federal and State dollars. The program provided homeowners with money and the choice to rebuild their home or to sell it. Although it has helped thousands to recover, the program has been criticized for paltry payments, unfair procedures, and program delays. Through analysis of individual applications provided by the Louisiana Recovery Authority, we identify spatial patterns in program implementation and rebuilding option. We then combine those data with others from the US Army Corps of Engineers and the US Census Bureau. With the resulting data set – including information on damage levels, insurance payouts, award sizes, income, race, and location – we estimate a model of the effect of damage and socioeconomic factors on rebuilding choices. Finally, we examine the differential impacts of the method of calculating awards in the RHP. Our findings confirm those of several reports. The vast majority of residents elected to rebuild their homes, but the method of calculating grants provided insufficient funds to do so, particularly in neighborhoods with lower housing values. We conclude that the RHP was successful in funneling billions of dollars to homeowners for rebuilding and mitigation, but that the unresolved tension between its role as a rebuilding program and a compensation program created significant barriers to recovery.

Keywords:

• New Orleans
• policy
• construction/building
• affordability

Acknowledgments

This research was funded by a grant from the Lincoln Institute of Land Policy. We also are grateful to the Louisiana Recovery Authority for providing access to a considerable amount of data from the Road Home Program. All analyses and opinions expressed in this paper are our own.

Notes

¹This is merely paraphrasing, as the rhetoric surrounding rebuilding would comprise a paper unto itself. More description of this can be found in Olshansky and Johnson (2010) and Lydersen (2006).

²For a description of the multiple recovery planning processes see Nelson, Ehrenfeucht, and Laska (2007) and Olshansky and Johnson (2010).

³The actual formula was [(sq. ft. × 130) + 550] × 1.02. The $550 was added to cover the cost of a house raising survey, and the 2 percent additional was to cover the risks assumed by the builder in reconstructing (RHP 2010a).

⁴The parish of Calcasieu, located on the far western edge of the state, also suffered a great deal of damage during Hurricane Rita, which came just weeks after Katrina. The RHP reports that the parish had 12,841 applicants to the program, of which 131 elected to sell their homes (RHP 2010c).

⁵We included those who identified as African-American alone or in combination with one or more other races, as a percentage of the total population.

⁶For details on the preparation of the data set see Green and Olshansky (2009).

⁷The lack of these two variables could introduce omission bias into our model. To address this, we also estimated the individual model using income and occupancy data at the Block Group level. The results were essentially the same as those reported here, while the grouped variables were not significant. Since this solution to the omission bias raises another problem of ecological fallacy, we report the specification that uses only individual data here.