Highlights from "The cost of climate: America’s growing flood risk"

New research from First Street Foundation analyzes the economic impact of underestimated flood risk to properties throughout the United States. Current understandings of expected flood risk and its potential damage continue to underestimate the full extent of flood risk in many parts of the country. This is problematic as these understandings of flood risk are used to price insurance premiums in the market by estimating property-level average annual loss (AAL).

The study applies the First Street Foundation Flood Model, a nationwide probabilistic property-level flood model, to an analysis of depth-damage functions from the United States Army Corps of Engineers (USACE) in order to estimate AAL for residential properties throughout the United States, today and into the climate-adjusted future.

Because a great deal of flood risks exists outside of Federal Emergency Management Agency’s (FEMA) designated Special Flood Hazard Areas (SFHA), this research reveals a vastly expanded mapping of economic risk associated with flood risk, and demonstrates the extent to which information asymmetries on flood risk contribute to financial market asymmetries, specifically in the form of underestimations of financial and personal risk to property owners.

When adjusting for the long term impact of a changing climate, the study finds there are nearly 4.3 million residential homes (1-4 units) across the country with substantial* flood risk that would result in financial loss. Furthermore, the research demonstrates that if all of these homes were to insure against flood risk through the National Flood Insurance Program (NFIP), the rates would need to increase 4.5 times to cover the estimated risk in 2021, and 7.2 times to cover the growing risk by 2051. The financial impact of flooding is particularly resonant as FEMA is poised to implement Risk Rating 2.0 in October 2021, which will adjust insurance premium calculations to better reflect individualized risk for households. Since homeowners that have a federally backed mortgage and reside within a SFHA are required to buy insurance, these new financial loss data therefore serves as an early indicator of the states, communities, and homes that could be most impacted by these risk-based rate changes, and how they may continue to increase due to climate change.

Methodological overview

The methodology employed by researchers at First Street Foundation used data from a combination of public and non-public sources, including property-level building characteristics from county assessor offices, and the Microsoft/Mapbox building footprint database. Property data were standardized and made available through a third party provider to ensure that attributes are consistent and meaningful across counties. Where those data are not available, the likely building characteristics were estimated from publicly available, block-level data from the National Structures Inventory database provided by the USACE, which were used to estimate local building codes and dwelling types at the Census block level.

A combination of USACE and Federal Insurance Administration (FIA) depth–damage functions were then applied to the Foundation’s national flood model to compare deterministic flood hazard layers created at two different time periods (2020 and 2050) and six different return periods (2-year, 5-year, 20-year, 100-year, 250-year, and 500-year).

See full report for all notes, definitions and full methodology.

National Overview

First Street Foundation’s analysis found that across the country, the average estimated annual loss for the 5.7 million properties that have any** flood risk and an expected loss from that flooding is $3,548, which totals to $20.3 billion in annualized expected losses in today’s environment. Using climate projections for 30 years into the future yields a 67% increase in the average estimated annual loss, or $5,913 per property at risk of economic damage from flood, and an estimated $34.0 billion loss across the contiguous United States.

When limiting the sample only to the 4.3 million properties with substantial flood risk, the average estimated annual loss increases to $4,694 per property with risk, and totals to $20.0 billion in annualized damage across the country today. Those economic damage estimates are expected to grow due to climate change over the next 30 years by 61%, to an average estimated annual loss of $7,563 per property, for an estimated total loss of $32.3 billion.

The Foundation’s economic damage estimates are a probabilistic set of expectations centered around the probability of flooding events, as expressed through the multiple return periods in the model, and the expected average damage to homes given estimated inundation depths for those events. This probabilistic approach is consistent with actuarial methods used in understanding and setting insurance rates to properly assess expectations of economic risk in an annualized form.

As such, the average estimated annual loss indicators can be compared directly to current NFIP risk premiums as a way of better understanding and quantifying any potential over- or under-estimation of flood risk in economic terms. However, the NFIP rating manuals and associated tables incorporate a series of payout limits, deductibles, surcharges, fees, and overhead amounts that must be taken into account in order to realistically compare expected economic damage from the First Street Foundation models to estimated NFIP premiums. In order to meet those requirements, NFIP premiums were estimated directly from the rate tables with coverage rates equal to the building value, and with a $1,000 deductible (with no discount on the premium). For this comparison to NFIP only, the average estimated annual loss was recalculated to be capped at $250,000 per property.

For the 5.7 million properties with any flood risk, the capped average estimated annual loss is $3,343 and the average NFIP premium for those properties is $902. This per-property average difference of $2,441 accounts for the current underestimation of the economic impact of flood risk among properties in this analysis and indicates that the current economic risk is 3.7 times higher than the level at which the NFIP is currently pricing insurance.

By limiting the sample again to only the 4.3 million properties with substantial flood risk, there is a capped average estimated annual loss of $4,419 and an average NFIP premium of $981, which results in 4.5 times more economic risk than in estimated premiums within the NFIP.

Breaking down the 4.3 million properties further to compare those within versus those outside the SFHAs shows significant differences in economic risk. Within the SFHAs, the difference in the capped average estimated annual loss of $7,895 and the $1,884 average estimated NFIP premium results in an underestimate of economic damage of 4.2 times. Outside of the SFHAs, that underestimate grows to 5.2 times with a capped average estimated annual loss of $2,484 compared to an average estimated NFIP premium of just $478. This latter category is specifically vulnerable to flood risk, as these properties have not been mapped by FEMA into a SFHA and therefore property owners are likely unaware of the scale of risk they carry and are more unlikely to have flood insurance from the NFIP. NFIP insurance is not mandated at any time outside of a SFHA.

Uses and Implications

Accurate, comprehensive, and publicly available estimates of annual flood damages enable improved risk management and more cost-effective hazard mitigation planning at every level. Making this data open to individual property owners and communities provides actionable information to homeowners and renters about the flood damage a property is likely to experience and allows them to make more informed decisions about risk reduction investments and flood insurance coverage. The Foundation provides this information for every property in the contiguous United States in a format that is publicly and freely accessible via Flood Factor®, an online database and visualization tool.

Nationally, such estimates may inform Congressional allocations of funds for hazard mitigation programs and major infrastructure projects aimed at reducing flood risk. At the state and local level, they enable more accurate cost-benefit analyses, which help determine whether these projects receive public funding. Annual flood damage estimates are often used in cost-benefit analyses that help determine how billions of federal, state, and local dollars are spent every year. First Street Foundation has signed agreements with several agencies and government sponsored enterprises to provide access to these flood risk data.

Such data, derived using an open methodology, have not previously existed at the property level and allow for new insights in the area of AAL analysis. These data are based upon higher resolution and detail of flood risk information than that typically derived from other open sources. Informed by more comprehensive estimates of expected flood damages and greater knowledge of the impact of risk-reduction measures, decision-makers at every level of government can better identify the locations and measures that would be most beneficial in reducing flood risk. Communities and individual property owners may use the same information to more effectively manage negative externalities associated with flood risk through insurance and other risk transfer mechanisms.