Before Regret · Research
In the typical American county, fewer than one in six homes inside a mapped flood zone carries flood insurance. Coverage rates track which state a home is in far more closely than they track modelled flood risk.
2,304 counties, 3.75 million homes inside FEMA-mapped flood zones, measured against FEMA’s own flood loss model. A companion to Risk Without Price, which found that 76.5% of modelled building loss falls outside a standard policy — flooding above all.
Interactive · 2,304 counties
Every county with homes inside a FEMA-mapped flood zone. Type a county for its own figure, or a state name for the statewide picture and all of its counties ranked.
Statewide
Counties ranked from least covered. Click any row for its detail.
Homes in mapped flood zones with a policy
A standard homeowners policy does not cover flood. That is not a loophole; it is the design, and it is why the National Flood Insurance Program exists. Our earlier analysis found that flooding alone accounts for 59.2% of the building loss FEMA expects American homes to suffer each year, all of it outside the policy most people hold.
Which raises a question that data can answer. FEMA publishes, for every county, how many homes sit inside a mapped Special Flood Hazard Area and how many NFIP policies are in force there. Put those together and you can see, county by county, how much of that exposure anyone actually bought cover for.
Across the 2,304 counties with homes in a mapped flood zone, the median take-up rate is 15%. In 705 counties fewer than one home in ten carries a policy, and in 68 counties no policy is in force at all.
The national aggregate looks better — 39.8% of flood-zone homes nationally are covered — but that number is carried by a handful of large coastal counties. Florida alone accounts for 566,635 of the 1.5 million policies in the dataset. The gap between 39.8% national and 15% typical is the shape of the problem: coverage is concentrated where the market is mature, and thin almost everywhere else.
If coverage rose in proportion to modelled flood risk, the cloud below would slope upward. Regressing log take-up on log flood risk gives an R² of 0.004. State of residence alone gives 0.225 — more than fifty times as much.
The result survives every cut we tried. Excluding the 68 counties with no policies at all: R² 0.032 against 0.279 for state. Restricting to counties with at least 20,000 mortgaged households: R² 0.000 against 0.458 for state. Wherever you look, where you live predicts coverage and how much flooding you face does not.
West Virginia has the highest flood loss rate of any state in this analysis, at $2.71 per $1,000 of building value. It also has the lowest take-up: 10.3%, or 6,324 policies against 61,606 flood-zone homes. First of fifty on risk, fiftieth of fifty on cover.
Kentucky is second on risk at $2.25 and thirty-second on cover. South Carolina, which has the highest take-up in the country at 68.7%, ranks forty-third of fifty on risk.
Private flood insurance is not counted. NFIP is most of the market but not all of it. Private policies have grown, so true coverage is somewhat higher than these figures, and by an amount this data cannot measure.
Flood maps are imperfect and dated. Special Flood Hazard Areas are redrawn slowly, and a large share of flood damage happens outside them. A county with high take-up inside the zone may still have most of its exposed homes uncovered.
Take-up is not a free choice for everyone. Federally backed mortgages require flood insurance inside a mapped zone, so part of what varies across states is the mortgage mix and how strictly the requirement is enforced, not only what buyers decide.
Correlation, not conduct. Nothing here explains why coverage is where it is. Price, awareness, enforcement, past flooding and state programmes all plausibly contribute, and this analysis separates none of them.
Homes inside mapped flood zones, policies in force, and the state’s own flood loss rate.
| State | Counties | Flood-zone homes | Policies | Take-up | Flood loss / $1k |
|---|---|---|---|---|---|
| SC | 42 | 67,559 | 46,397 | 68.7% | $0.75 |
| NJ | 20 | 161,481 | 100,796 | 62.4% | $1.15 |
| DE | 3 | 20,853 | 12,108 | 58.1% | $0.76 |
| FL | 63 | 997,127 | 566,635 | 56.8% | $0.81 |
| NY | 27 | 95,022 | 52,173 | 54.9% | $0.78 |
| HI | 1 | 8,317 | 4,490 | 54.0% | $1.94 |
| NC | 100 | 113,606 | 60,527 | 53.3% | $0.86 |
| CT | 7 | 31,234 | 15,727 | 50.4% | $1.11 |
| MA | 11 | 53,794 | 26,563 | 49.4% | $0.72 |
| RI | 5 | 11,194 | 5,486 | 49.0% | $0.64 |
| MD | 18 | 15,307 | 6,692 | 43.7% | $0.94 |
| GA | 155 | 72,108 | 30,853 | 42.8% | $0.79 |
| LA | 45 | 195,619 | 80,704 | 41.3% | $1.24 |
| TX | 126 | 358,865 | 143,426 | 40.0% | $0.99 |
| AK | 5 | 2,840 | 946 | 33.3% | $0.27 |
| MS | 81 | 87,786 | 28,832 | 32.8% | $0.99 |
| CA | 55 | 299,628 | 90,678 | 30.3% | $1.46 |
| NV | 14 | 20,062 | 5,970 | 29.8% | $0.96 |
| AL | 67 | 62,310 | 18,305 | 29.4% | $1.06 |
| WA | 19 | 39,781 | 11,408 | 28.7% | $0.98 |
| VT | 6 | 4,617 | 1,250 | 27.1% | $1.01 |
| ND | 23 | 8,616 | 2,307 | 26.8% | $0.77 |
| VA | 91 | 39,784 | 10,098 | 25.4% | $1.05 |
| NH | 9 | 11,554 | 2,916 | 25.2% | $1.21 |
| TN | 95 | 41,905 | 10,286 | 24.5% | $0.94 |
| OR | 21 | 41,466 | 9,342 | 22.5% | $1.16 |
| PA | 66 | 92,246 | 20,339 | 22.0% | $1.08 |
| IL | 77 | 57,540 | 12,648 | 22.0% | $1.08 |
| AZ | 15 | 57,819 | 12,298 | 21.3% | $1.51 |
| ME | 9 | 5,147 | 1,090 | 21.2% | $0.67 |
| CO | 32 | 22,669 | 4,790 | 21.1% | $1.07 |
| KY | 119 | 60,469 | 11,364 | 18.8% | $2.25 |
| WI | 61 | 27,192 | 4,996 | 18.4% | $0.75 |
| IN | 89 | 58,196 | 9,605 | 16.5% | $0.81 |
| MI | 55 | 63,812 | 10,321 | 16.2% | $0.71 |
| OH | 78 | 73,981 | 11,624 | 15.7% | $0.92 |
| UT | 16 | 8,149 | 1,236 | 15.2% | $0.59 |
| ID | 12 | 13,801 | 2,076 | 15.0% | $1.41 |
| MO | 92 | 40,540 | 6,062 | 15.0% | $1.24 |
| NM | 23 | 46,300 | 6,913 | 14.9% | $1.72 |
| NE | 60 | 29,073 | 4,191 | 14.4% | $1.02 |
| MT | 22 | 13,741 | 1,968 | 14.3% | $1.04 |
| AR | 49 | 40,425 | 5,651 | 14.0% | $1.00 |
| KS | 46 | 23,562 | 3,288 | 14.0% | $0.90 |
| SD | 30 | 9,200 | 1,216 | 13.2% | $0.97 |
| IA | 82 | 30,404 | 3,832 | 12.6% | $0.95 |
| OK | 50 | 33,967 | 4,161 | 12.3% | $1.03 |
| MN | 44 | 16,708 | 1,931 | 11.6% | $1.01 |
| WY | 13 | 4,848 | 509 | 10.5% | $1.05 |
| WV | 55 | 61,606 | 6,324 | 10.3% | $2.71 |
Free to reproduce with attribution. Both sources are public and the method above is enough to rebuild every figure.
Flood insurance take-up for 2,304 counties: how many homes sit inside a mapped flood zone, how many NFIP policies are in force, and the resulting rate.
flood-takeup-by-county.csv · the full figures as JSON
The take-up column counts NFIP policies only. Private flood insurance has grown and is not in this file, so true coverage is higher than these numbers by an amount the data cannot measure. Read a low figure as "NFIP coverage is low here", not as "these homes are uninsured".
Every exhibit on this page as a high-contrast image, free to reuse with credit. The PNG drops straight into a document or a slide; the SVG stays sharp at any size.
Exhibit 1 (PNG) · SVG
Exhibit 2 (PNG) · SVG
In the typical US county, roughly one home in seven inside a mapped flood zone carries an NFIP flood policy, and take-up does not track flood risk, according to an analysis of FEMA and Census data by BeforeRegret.
Risk Without Cover, Before Regret, 31 August 2026. Data source: FEMA NFIP · Census ACS. https://www.beforeregret.com/research/risk-without-cover/
Free to use on any site. It sizes itself to its content and carries its own credit line.
