New study analyzes FEMA-funded home buyout program

A research team led by scientists at the University of Miami (UM) Rosenstiel School of Marine and Atmospheric Science found that FEMA-funded voluntary buyouts of flood-prone properties have been more likely to take place in counties with higher population and income. However, the buyouts themselves were concentrated in neighborhoods with lower income and greater social vulnerability. The researchers hope their analysis can provide insights to help revise this program in the future.

"In recent decades, communities throughout the United States have been building experiences with retreat from flood-prone areas, largely through voluntary buyouts of properties after disasters," said the study's lead author Katharine Mach, an associate professor of marine ecosystems and society at the UM Rosenstiel School. "While a proven method of reducing risk, buyouts to date have also illustrated the challenges with locally driven managed retreat and the potential benefits of experimentation with different retreat policies in the future."

In the U.S., over 40,000 homes in flood-prone areas have been purchased and returned to open space in 49 states and three U.S. territories under the FEMA program since its inception in the late 1990s. The federally funded program is typically administered through local government agencies.

Mach and colleagues analyzed data on the more than 40,000 buyouts alongside data on flood risk, socioeconomics and demographics to reveal that local governments in counties with higher population and income were more likely to administer buyouts of flood-prone properties. However, at the zip-code level, the bought-out properties themselves were concentrated in areas of lower population and income.

As a first national-level analysis of all FEMA-funded buyouts implemented to date, the authors' findings suggest insights on what is working and what is not. The results can support revision of buyout practices, including their potential deployment at larger scales in the future in combination with different types of retreat and climate adaptation policies.

"Retreat in some places will become unavoidable under intensifying climate change, but there are open questions about where, when, and how retreat under climate change will occur," said Caroline Kraan, a graduate student at the UM Abess Center for Ecosystem Science and Policy and a coauthor of the study. "Our analysis offers lessons from buyouts to date that can inform and support strategies of managed retreat in response to climate change."

The analysis also showed that the top three states for most flood damage -- Florida, Louisiana and Mississippi -- ranked 23, 18, and 21 respectively in deployment of buyouts, pointing to the fact that not all flood-damaged areas are currently utilizing the FEMA program for flood risk management.

"These findings suggest that buyouts are resource-intensive to administer, which might make them less feasible in areas that need them the most," said study coauthor A.R. Siders, an assistant professor in the Disaster Research Center at the University of Delaware. "They also suggest that something in the FEMA program favors using buyouts in lower-income neighborhoods, which may have social justice implications."

The authors believe that the analysis can help inform a larger dialogue to better understand how managed retreat is being used now and into the future as governments increasingly look for ways to protect people and property from rising sea levels and more frequent flooding and other natural hazards as a result of climate change.

"The buyout program represents billions of dollars in government spending, yet until now, we didn't have a broad understanding of who benefitted from those investments," said Miyuki Hino, a postdoctoral researcher at Stanford University and coauthor of the study. "Now, we can see not only who has benefitted, but also what types of places and communities may be falling through the cracks."

This study is part of a broader project on managed retreat by the research team.