Could government-funded flood buyout programs be adjusted to better serve communities?

Wealthier, more densely populated counties in the U.S. have been more likely to implement buyouts of flood-prone properties, according to the first programmatic-level analysis of voluntary property buyouts through the U.S. Federal Emergency Management Agency (FEMA). These results contradict economic predictions of flood risk management scenarios, which suggest higher rates of retreat from properties in lower-income and rural areas. Within the higher-income counties that administered buyouts, however, the researchers observed that buyouts were often focused in relatively poorer areas. "Even though it isn't clear why buyouts are taking place in more vulnerable neighborhoods, this pattern points to the importance of evaluating and ensuring equity in buyout practices and outcomes," says study co-author Carolien Kraan. She and her colleagues sought to evaluate whether U.S. federal buyouts, seen as a means of accomplishing "managed retreat" from climate change, are fair and effective - analysis of which has otherwise been sparse. From 1989-2017, FEMA funded over 40,000 voluntary buyouts, in which owners agreed to sell their properties. Then, the properties' residents relocated, the property was removed, and the land left behind was maintained as open floodplain space. Using publicly-available data, Kraan, Katharine Mach, and colleagues analyzed all FEMA-funded voluntary buyouts of flood-prone properties. In addition to identifying the locations of buyouts, the researchers studied their size, finding that FEMA buyout projects have steadily decreased in size since 1989 and now frequently consist of only one to three properties each. Small buyouts are less economically efficient than larger ones and may result in patchy property removal, overlooking opportunities to more strategically restore floodplains and reduce overall flood risk within communities. Mach and colleagues emphasize the importance of increased transparency in FEMA reporting, noting that over 50% of entries are empty for some fields of publicly available data, compromising researchers' ability to fully evaluate the program.

###