Racism and Redlining Can Be Seen ‘in the Flood Data’

There’s no question that flooding is becoming worse due to the climate crisis pushing up our sea levels and making heavy downpours more common. But the risk doesn’t affect us all equally. A report released by the real estate brokerage firm Redfin this week shows that formerly redlined areas are more vulnerable to the threat of floods.

Redlining was a discriminatory lending practice in the 1930s, wherein house appraisers mapped cities for the federal Homeowners’ Loan Corporation with different colored lines to indicate how desirable areas were to live in. Neighborhoods deemed “best” or “still desirable” were marked with green and blue lines, respectively. Those considered “definitely declining” were marked in yellow, and zones declared “hazardous”—overwhelmingly Black neighborhoods—were outlined in red. The reverberations of this segregationist history are still visible. Today, 58.1% of residents of neighborhoods that were marked with yellow or red lines are still people of color, compared with just 40.4% of households in neighborhoods that were labeled blue or green.

Formerly redlined neighborhoods are exposed to more heat and have higher rates of respiratory problems tied with air pollution. The new analysis shows the flood risk is greater, too. Redfin examined floodplain data from 38 major metropolitan areas and found that today’s flood risk maps look very similar to those redlining maps. According to the report, 8.4% of homes in areas that were once deemed undesirable and marked with yellow or red lines face high flood risk, compared with just 6.9% of homes in areas that were marked with green or blue lines.

In some cities, the disparity is much larger. The largest gap the researchers found was in Sacramento, where 21.6% of homes in redlined and yellowlined areas face high risk of flooding today, whereas just 11.8% of homes in greenlined and bluelined neighborhoods are highly vulnerable. In New York, 13.8% of homes in areas marked in red and yellow are at risk, versus 7.1% of homes in greenlined and bluelined neighborhoods. Boston and Chicago also saw large disparities.

Part of this disparity comes from the natural environment in which housing is built. Historically, climate-safe areas are prohibitively expensive for many Americans while affordable housing is often located in areas of low elevation altitude and proximity to regularly overflowing bodies of water. Many public housing projects, the majority of which are home to non-white Americans, have also been sited in floodplains.

“Decades of segregation and economic inequality shoehorned many people of color—especially Black Americans—into living in neighborhoods that are more vulnerable to climate change,” Redfin senior economist Sheharyar Bokhari said in a statement.

But though geography can make neighborhoods of color more vulnerable, built infrastructure also plays a role. The same factors that lead to redlined neighborhoods being hotter can also affect flood risk. More pavement and fewer trees tend to crank up the urban heat island effect, but both also increase flood risk because pavement is a non-permeable surface so it can’t soak up floodwater, whereas trees can. 

On top of that, the economic exploitation of and disinvestment in communities of color—especially Black communities—has made it more difficult for once-redlined communities to fund infrastructure projects needed to keep people safe. Flood-control infrastructure such as storm drains and levees requires regular maintenance, which costs money.

“Redlining kept home values in Black neighborhoods depressed, which in turn meant there was less money invested and reinvested in those neighborhoods for decades to come,” Bokhari added.

Neighborhoods of color are also more likely to experience hardship when floodwaters rise. For instance, when Hurricane Katrina overwhelmed New Orleans and surrounding areas in 2005, four out of seven zip codes that faced the costliest flood damage were home to at least 75% Black residents. After Hurricane Harvey touched down on Texas in 2017, roughly twice as many Black and Latinx residents reported that they had fallen behind on their mortgage payments in the wake of the storm compared white residents.

Another new report released this week shows that the topographic and economic risk factors are difficult to disentangle from one another. That study, published in the Proceedings of the National Academy of Sciences on Monday, specifically assessed river flooding risk in 50 different American metropolitan areas. To do so, the authors used a socio-hydrological model, plugging in local data from each region on highest yearly streamflow (the peak speed at which water moves through rivers and streams), flood insurance loss claims, active insurance policy records, and population density.

“The model attempts to simulate the behaviors, memories, and risk aversion of a group of people in response to floods,” James Knighton, an assistant professor in the Department of Natural Resources and the Environment at the University of Connecticut and lead author of the study, said.

Using the models of the 50 cities, the authors teased out patterns in how different communities respond to floods. They found that the regions could be grouped into two groups. In the first category, which the researchers call “risk enduring” cities, residents are quick to purchase flood insurance as soon as flooding occurs, but are also quick to drop their insurance if a flood doesn’t happen again for a few years. This behavior was common in regions with higher concentrations of people of color, which also tended to be regions where rivers were more erratic, having more drastic fluctuations in streamflow.

Residents of cities who fall in the other group, which the authors refer to as “risk averse,” have different behaviors. In these cities, residents are slower to purchase insurance when floods occur. But once residents have that insurance, they’re less likely to drop it. This trend played out mostly in areas where the majority of residents are white, which also tended to be areas where streamflow stayed more level and predictable.

Knighton and the team aren’t sure exactly what historical factors have resulted in more people of color residing near erratic bodies of water, or how to separate hydrology and the socioeconomic trappings of race. In other words, it’s not clear how much risk comes from the rivers themselves, and how much of it comes from the fact that neighborhoods of color are less likely to have access to flood-protection infrastructure, personal funds for buying insurance, and information about the threat of flooding.

“It is very difficult to tease apart the two factors,” he said. “My guess is that socioeconomics dominate flood preparedness. If populations have limited ability to leverage federal programs for flood relief, then they are limited in how they can prepare for future floods. If communities have ample resources, then it becomes a risk-based choice.”

But he did say that it’s concerning that risk-averse white communities also tended to have much more flood control infrastructure like dams, and more access to federal assistance. The Federal Emergency Management Agency’s own data and reporting, for instance, has shown that the agency tends to “provide an additional boost to wealthy homeowners and others with less need, while lower-income individuals and others sink further into poverty after disasters.”

“What happens is, cities and towns that have a lot of residents that are well-off can totally take advantage of these federal programs. They can use these programs as they were intended to be used because they employ large local governments and have people that know how to call FEMA up,” Knighton said. “So, the resources that exist end up getting used very unevenly.”

Both analyses illustrate the urgent need for truly equitable federal programs that limit flood risk for everyone. FEMA seems to know this, because for years, it has been working on an affordability framework aimed at ensuring that all communities have access to flood insurance. Under thr program, officials would consider households’ socioeconomic status in particular areas and vary the price of the insurance based on residents’ means. They would also up their outreach in particularly vulnerable areas which may not have local officials who are focused on flood planning.

“This country has a big problem with race, and because it’s so big, you can even see it in the flood data,” said Knighton. “The next needed step to address that is to actually implement that framework right now.”