Flood Risk Mapping in the New York Federal Reserve’s Second District

Research economists at the New York Federal Reserve analysed the mapping of flood risks in the Fed’s second district. The results, covered in a series of blog posts, reveal that widespread inaccuracies in flood mapping may result in inadequate insurance in areas prone to flood risk.

1. Potential Flood Map Inaccuracies in the Fed’s Second District 

In the first article, the authors – Kristian Blickle, Katherine Engelman, Theo Linnemann, and João Santos – discuss the prevalence of inaccuracies in The National Flood Insurance Program (NFIP) flood maps.

The authors use detailed data on flood damage faced by individual dwellings to tease out the degree of inaccuracy in flood mapping. The results show that, when aggregated to the census tract level, a large number of maps do not accurately capture flood risk.

Flood risk is continually reconfigured due to changes to climate, geographical processes and flooding adaptation measures, making it a dynamic risk. However, NFIP flood maps rely on a static approach that must be continually updated.

The authors find that maps are updated infrequently and quickly become outdated, with many decades old.

Outdated maps may result in an underestimation and underinsurance of risks, say the authors. A quarter of all maps were fifteen years old in 2022 and over half were more than five years old.

The NFIP flood maps designate areas at risk of a catastrophic flood occurring at least once in 100 years. In areas considered to have at least a 1% annual flood risk, flood insurance is mandatory, highlighting the regulatory importance of map accuracy. 

Elsewhere, Madison Condon, of the Boston University School of Law, has proposed creating publicly owned and open-source national climate data services to address the issues identified in the article. Such services would require federal, state and international agencies to invest in and help build datasets to improve the integration of climate modelling into financial risk analysis.

2. How Do Banks Lend in Inaccurate Flood Zones in the Fed’s Second District?

In this post the authors examine how banks lend in “inaccurately mapped” areas. They find that in communities with inaccurate mapping, there is a higher loan rejection rate.

This implies that banks are aware of mapping issues as they are less likely to lend to regions with inaccurate mapping, even if loans can be securitised.

In a community with a 10% reduction in the accuracy of maps, for instance, an 8% increase in the loan rejection rate can be observed. The rate increases fourfold in cases with a 10% reduction of income.

This relationship can be observed across the nation and is strongest in southern states where there is a higher risk of catastrophic floods.

3. Flood Risk and Firm Location Decisions in the Fed’s Second District

The intensity, duration, and frequency of flooding have continually increased throughout the last few decades, and flood risks have emerged as a primary physical risk for the district, the authors say.

In the final post of this part of the series, the authors – Oliver Zain Hannaoui, Hyeyoon Jung, João A.C. Santos, and Lee Seltzer – discuss whether firms consider flood risk when deciding where they conduct business, produce goods, or render services.

The results show that new firms are not avoiding risk prone countries in the second district, meaning flood risk is not a primary determinant of business location decision-making.
 
As establishments are not migrating to low risk environments, counties that are exposed to flood risk will continue to play an important role in the region’s economy, the authors found.