Demographics Data Helps Predict NY Flood Insurance Claims

In flood-prone areas of the Hudson River valley in New York state, census areas with more white and affluent home owners tend to file a higher percentage of flood insurance claims than lower-income, minority residents, raising the issue of developing more nuanced, need-based federal flood insurance subsidies in these floodplains, according to a new study.

“Are we taxing America so that people with second homes on the water don’t have to pay as much?” asked Brian Rahm, director of the New York State Water Resources Institute in Cornell’s College of Agriculture and Life Sciences.

Rahm is a co-author of the paper “Predicting Flood Insurance Claims with Hydrologic and Socioeconomic Demographics via Machine Learning: Exploring the Roles of Topography, Minority Populations, and Political Dissimilarity,” which published online July 15 in the Journal of Environmental Management.
James Knighton Ph.D. ’19, an assistant professor at the University of Connecticut, is lead author of the study.

Flood insurance rate maps determine areas where people with federally backed mortgages are required to buy flood insurance, which is then subsidized by the Federal Emergency Management Agency (FEMA), lowering the insurance cost and the reducing the risk of living in a floodplain.

At the same time, current flood insurance rates do not adequately account for how social demographics affect risk and vulnerability, but FEMA is taking steps to better understand the issues, Rahm said.