How FEMA Became America's De Facto Flood Insurance

On the morning of July 11, 2023, the Winooski River jumped its banks. Within hours, water was running through the front doors of homes in Montpelier, Vermont. Whole blocks of the state capital lost power. More than 3,600 Vermont households would receive approved Individual Assistance from FEMA in the months that followed. Just over 1% of Vermont homes carried flood insurance.

What followed was a check, but not from a private insurer. It came from the Federal Emergency Management Agency. Vermont received $26 million in approved Individual Assistance and $267 million in obligated Public Assistance for that single July storm. Thirteen months later, the rivers came up again. Another federal declaration. Another $73 million in combined aid.

Across all 50 states and the District of Columbia, MoneyGeek's analysis of OpenFEMA data shows federal disaster aid functioning as the de facto recovery product for households whose private insurance either does not cover the loss or does not exist. Total federal outlay from January 1, 2020 through May 27, 2026, excluding COVID-19 emergency declarations, was $42.9 billion: $11.8 billion in Individual Assistance approved for households, and $31.0 billion in Public Assistance obligated to states and local governments. That total excludes hazard-mitigation grants, Small Business Administration disaster loans, and NFIP claim payouts.

Louisiana, with 4.6 million residents, received $1,916 per person in that combined window. Vermont received $660. Hawaii, $590. Three states cleared $500 per resident. Nine states cleared $200. The national pattern is concentrated, geographic, and tracks the coverage shortfall MoneyGeek's April 2026 flood insurance study documented across 26 states with sub-1% NFIP penetration.

Federal disaster aid since 2020 has hit unevenly. Five states (Florida, Louisiana, North Carolina, Texas, California) account for more than half of the $42.9 billion total. Fourteen states received less than $25 per resident. The bottom three (Ohio, Maryland, Nevada) received less than $4 per resident, mostly because their disaster activity in the window was limited to small severe-storm declarations.

Per-capita changes the picture. Florida's $9.3 billion (the largest absolute figure) becomes $415 per resident across 22.4 million people. Vermont's $427 million becomes $660 across 647,000. Per-resident is the more accurate comparison because federal aid is structured to make households whole, not to scale with state size.

The states most reliant on federal aid are the states where private insurance, especially flood insurance, was always going to be inadequate to the loss.

Below are the 10 states with the highest federal aid per resident. The top of the table confirms what hurricane history would predict but also includes two surprises (Vermont, Hawaii) that point at coverage failures the federal-aid data makes visible. The full 50-state plus D.C. ranking appears at the end of this analysis.

Louisiana sits at the top because of Hurricane Ida alone. The August 2021 storm generated $4.8 billion in federal aid, more than the combined six-year total of 33 other states. Add Hurricanes Laura ($3.4 billion in 2020), Delta, Zeta and Francine, and the Louisiana number stops feeling like an outlier. Those five hurricane declarations together brought $8.7 billion in federal aid to Louisiana between 2020 and 2024.

Vermont's #2 ranking came from two flood events, not five hurricanes. The state had 13 federal declarations in the window, but the July 2023 storm and the August 2024 storm did most of the damage. Both were inland river flooding. Vermont's NFIP penetration rate was 1.15% at the time per MoneyGeek's April 2026 analysis, which means roughly 99 of every 100 Vermont homes had no private coverage against river flooding. Standard homeowners insurance does not cover flood damage in any state. Federal aid was the only recovery pathway available.

Kentucky, ranked sixth, fits the same inland-flood pattern. The state's combined IA-plus-PA total for 2020-2026 included the catastrophic eastern Kentucky floods of July 2022, which damaged or destroyed roughly 9,000 homes. Kentucky NFIP penetration: 0.84%.

Two deadlines collide in the second half of this year. Atlantic hurricane season runs from June 1 through November 30, with peak activity in August through October. The NFIP's current authorization expires September 30, 2026, directly inside that window. If Congress allows the program to lapse, as it has 34 times since fiscal 2017 per the Insurance Information Institute, flood insurance transactions freeze during the most active storm period.

Beyond the deadlines, the data points to a structural problem. Eight inland states ranked above the median for federal aid and below the median for NFIP coverage (Tennessee, Kentucky, Iowa, Arkansas, Oklahoma, Montana, South Dakota and Michigan). A 0.72% NFIP penetration rate in Tennessee is a structural condition, not a temporary deficiency. The next major inland flood event will find roughly the same coverage profile that Helene found in 2024.

That $42.9 billion six-year total in this analysis is a baseline, not a ceiling. Each additional inland flood event, each hurricane that pushes water past the SFHA boundary into territory the maps haven't updated, each Maui-style event where standard coverage exists but doesn't reach the full rebuild cost, adds to the federal-aid load. The substitute-insurance pattern wasn't designed. It developed on its own. And it will keep developing until something in the underlying coverage structure changes.

The complete 50-state plus D.C. ranking, sorted by federal aid per resident.

MoneyGeek analyzed 823 federally declared disasters across all 50 states and the District of Columbia using two OpenFEMA datasets pulled on May 27, 2026: the Disaster Declarations Summaries (v2) and the FEMA Web Disaster Summaries (v1). The window covers declarations dated January 1, 2020 or later. COVID-19 biological declarations were excluded.

For each disaster, the analysis summed Individual and Households Program awards and Public Assistance obligations to state and local governments. State totals were divided by the five-year average resident population (2020 through 2024) from the U.S. Census Bureau Population Estimates Program, Vintage 2024. Hazard Mitigation Grant Program funds were tracked separately and excluded from the headline per-resident metric. SBA disaster loans were out of scope.