Moody's Flood Study: $375 Billion of Uninsured US Residential Losses in a 1-in-100-Year Flood, a 65% Protection Gap and Over $1 Trillion in the Extreme Case

A large share of the damage from a severe US flood would fall on households and governments, not insurers, according to a Moody's Ratings assessment that has been reported in several forms since May. Reinsurance News first covered it on May 27, 2026, Risk & Insurance on June 9, and Insurance Journal on June 22. The assessment was written by Jennifer Chang, senior vice-president for sustainable finance credit, and Firas Saleh, director of North American flood models, and evaluates residential flood exposure across the country at the county level.

The headline number is for a 1-in-100-year flood, an event with a 1% chance of occurring in any given year. In that scenario, Moody's estimates that aggregate uninsured residential losses could exceed $375 billion, and that the nationwide protection gap, the share of the loss that is not covered by insurance, is about 65%. In a 1-in-500-year event, uninsured losses could exceed $1 trillion, with a gap above 70%. Looking forward, Moody's says that under an intermediate-emissions climate pathway the 1-in-100-year uninsured loss rises about 25% by 2050 to roughly $472 billion, and that high-exposure areas spread further, including into New Jersey.

The method is important to read correctly. Moody's used its RMS US Inland Flood HD model, which covers river, surface water and coastal flooding. It compared the replacement value of exposed homes with the value covered by the National Flood Insurance Program, and assumed that current insurance coverage and flood defences remain unchanged. The results therefore show structural exposure, in Moody's words, and are not projected realised losses. They do not predict a specific storm or year.

Most counties are not at the extreme. Moody's says roughly 90% of counties have some flood exposure, and in the 1-in-100-year scenario most face potential uninsured losses of $150 million or less. The concentration is in a small number of places: fewer than 2% of counties, spread across 11 states, hold 65% of the nation's uninsured exposure, each with potential losses above $1 billion according to Risk & Insurance. Counties with potential uninsured losses above $5 billion are mainly in Florida, Louisiana, South Carolina and Texas. In the 1-in-500-year scenario, counties above $5 billion extend to 11 states, including Pennsylvania and Illinois, and high-loss counties reach states beyond the traditional coastal hotspots. The protection gaps in these scenarios range from 45% to 75%.

The most vivid example is Asheville, North Carolina. During Hurricane Helene in September 2024, rainfall there exceeded the 1-in-1,000-year threshold, with one station recording 24.2 inches over three days, according to Risk & Insurance. Moody's RMS modelling indicates a protection gap of about 88% in Buncombe County across 1-in-100 to 1-in-500 events. Risk & Insurance added that Moody's revised Buncombe County's Aaa rating outlook to negative in March 2025 and back to stable in April 2026, a reminder that flood losses feed into credit assessments of local governments.

That link to public finance is the reason a credit rating agency is writing about flood insurance. Moody's says flood exposure is a growing credit issue for US state and local governments, driven by more frequent and severe weather, continued development in flood-prone areas and low insurance coverage. The fiscal effect depends on the absolute size of losses and on each county's or state's capacity to absorb them, which includes federal support, reserves, insurance payouts and governance frameworks, so the protection gap percentage alone does not capture it. Moody's concluded that flood risk is broadly distributed across the country but unevenly insured, and that many local governments face material potential uninsured losses that could strain their finances unless insurance take-up rises or resilience investment increases.

Why is the gap so large? Insurance Journal reported that the National Flood Insurance Program is the primary provider of residential flood insurance, and that private flood insurance makes up about 10% of the market. Risk & Insurance reported that take-up is low even in high-risk areas, at 10% to 30% along the Gulf Coast and Eastern Seaboard, that private policies have roughly doubled since 2020 without materially narrowing the gap, and that NFIP policy counts have fallen by a comparable amount to the rise in private policies. Mortgage rules also play a part. Federally regulated or backed lenders rely on FEMA Special Flood Hazard Area maps, which use a 1-in-100-year threshold to decide who must buy flood insurance. Moody's says these maps are mainly based on riverine flooding and do not fully capture extreme precipitation, greater storm surge or sea-level rise, which leaves exposure outside the mapped zones. The study warns against "relying solely on backward-looking statistics to characterize flood risk in a changing hydroclimate."

For households, the practical message is that standard homeowners cover does not pay for flood damage, and that being outside a mapped high-risk zone is not the same as being safe. A policy from the National Flood Insurance Program or a private insurer must be bought separately. For local governments and lenders, the study suggests that the question is not only the probability of a flood but how much of the loss would be left unfunded. For insurers, a persistent gap is both a risk and a market, because private flood has grown but has not closed it.

Limits should be kept in mind. The figures are scenarios built on a model and on assumptions about coverage and defences, they describe residential property only, and they do not say when a major event will occur. The reports also differ slightly in how they describe the concentration of exposure, so the original Moody's assessment is the source to consult for exact county-level numbers.