Across the 25,000+ US ZIPs we track, the median home value rises with climate risk: $217K in "Very Low" risk ZIPs, $623K in "Very High" risk ZIPs — a 2.9x premium for living somewhere that FEMA officially classifies as climate-exposed. 38% of Very-High-risk ZIPs appreciated above 5% per year over the last five years. The US housing market has barely priced climate risk in. The question isn't whether the correction is coming — it's how fast.
This guide explains exactly what the FEMA National Risk Index measures, why a "Very High" rating in Irvine, California means something different than the same rating in eastern Kentucky, and what the underlying property-value data actually shows about how climate risk is — and isn't — being absorbed by US housing prices.
What FEMA NRI actually is
The National Risk Index is a federal dataset published by FEMA that scores every US census tract on its expected annualized loss from 18 natural hazards. It's the closest thing the United States has to a unified national climate-and-disaster-exposure dataset, and it powers the climate sections on every ZIP page on this site.
The 18 hazards:
| Category | Hazards |
|---|---|
| Water | Coastal flooding, riverine flooding, tsunami |
| Wind | Hurricane, strong wind, tornado |
| Heat/cold | Heat wave, cold wave, drought |
| Geological | Earthquake, landslide, volcanic activity, avalanche |
| Fire/electrical | Wildfire, lightning |
| Winter | Ice storm, winter weather, hail |
For each hazard, FEMA combines three components into a per-tract score:
- Expected annualized loss — the dollar value of buildings, agriculture, and population at risk, weighted by historical event frequency.
- Social vulnerability — community-level demographic and economic factors that affect recovery capacity (income, age, language, housing quality, etc.).
- Community resilience — community-level capacity to absorb and recover from disasters (insurance coverage, local government capacity, etc.).
The 18 hazard scores get combined into a single composite that's rescaled to a 0-100 percentile rank and binned into five categories: Very Low, Low, Medium, High, Very High. A ZIP in the "Very High" tier sits in roughly the top 10-15% of national risk; "Very Low" sits in roughly the bottom 10-15%.
What the data actually shows
Here's the part most climate-risk explainers skip. Across the US ZIPs we track:
| Climate risk | ZIPs | Median home value | Median 5y CAGR | Median population |
|---|---|---|---|---|
| Very Low | 4,536 | $216,514 | +5.8% | 1,819 |
| Low | 8,229 | $244,603 | +5.7% | 3,140 |
| Medium | 5,768 | $321,878 | +5.6% | 8,556 |
| High | 4,684 | $416,872 | +4.9% | 20,526 |
| Very High | 1,541 | $622,819 | +4.5% | 32,637 |
Three things this table forces on the discussion:
- Home values rise with climate risk. This is not a typo. The median home in a Very-High-risk ZIP sells for almost three times what the median home in a Very-Low-risk ZIP sells for. This is the opposite of what an efficient market would price — and the reason is that high-risk areas are concentrated in places people want to live anyway: coastal California, South Florida, Mountain-West luxury enclaves. The location premium overwhelms the risk discount in the current pricing.
- Population follows the same pattern. Very-Low-risk ZIPs are small and rural (median population under 2,000); Very-High-risk ZIPs are urban (median over 32,000). Climate exposure is correlated with desirability, not with avoidance.
- Appreciation softens slightly but doesn't collapse. Very-High-risk ZIPs appreciated 4.5%/year over the last five years vs. 5.8% for Very-Low. That's a real discount — about 130 basis points annually — but it's not a market correction.
The strong-appreciation share, though, tells a more uncomfortable story:
| Climate risk | % with 5y CAGR ≥ 5% | % declining (CAGR < 0) |
|---|---|---|
| Very Low | 64.0% | 5.6% |
| Low | 61.7% | 4.9% |
| Medium | 59.6% | 3.9% |
| High | 47.6% | 6.6% |
| Very High | 37.7% | 7.4% |
The share of strong-appreciation ZIPs drops from 64% (Very Low) to 38% (Very High). The share of declining ZIPs nearly doubles between Medium and Very-High risk. The market is starting to discriminate — but slowly, and only at the edges of the distribution.
The Very-High-risk ZIPs that defy the pattern
To make the trend concrete, here are the 8 Very-High-climate-risk ZIPs with the strongest 5-year appreciation, all with populations over 15,000:
| ZIP | City, State | Home value | 5y CAGR |
|---|---|---|---|
| 92603 | Irvine, CA | $2,386,330 | +13.3% |
| 92008 | Carlsbad, CA | $1,200,791 | +13.2% |
| 85253 | Paradise Valley, AZ | $3,279,614 | +13.1% |
| 33146 | Coral Gables, FL | $1,897,381 | +13.1% |
| 33156 | Pinecrest, FL | $1,529,603 | +12.2% |
| 92602 | Irvine, CA | $2,058,967 | +12.2% |
| 92606 | Irvine, CA | $1,508,850 | +12.0% |
| 92620 | Irvine, CA | $1,726,820 | +11.6% |
These are seven-figure ZIPs in some of the most expensive zip codes in America, growing at 11-13% per year over five years despite carrying the highest climate-risk rating in our dataset. The market is paying a premium for them in spite of the FEMA NRI, not because the FEMA NRI is wrong about the underlying exposure.
Where the cracks are starting to show
The discount IS emerging, in three specific places:
- Insurance premiums, not home values. Insurance pricing is a leading indicator that doesn't appear in ZHVI. Florida, Louisiana, and California have seen homeowner premiums double or triple in five years across high-risk ZIPs. In some Florida coastal markets, insurers have pulled out entirely, leaving the state-backed Citizens Insurance as the only option. Premium escalation is the first place climate risk leaks into the cost of ownership — well before it shows up in sale prices.
- Specific submarkets within high-risk areas. Aggregate "Florida" or "California" stats hide huge variance. Within South Florida, oceanfront ZIPs are repricing faster than inland-by-a-few-blocks ZIPs. Within Northern California, ZIPs in or near the 2020+ wildfire footprints have stagnated while neighboring ZIPs have not. The discount is granular.
- Inventory + days-on-market. Even in markets where prices haven't dropped, inventory in Very-High-risk ZIPs is starting to take longer to clear. Three-figure days-on-market numbers in formerly hot Florida and California markets are showing up in 2024-2026 data that weren't there in 2019-2022.
The trend isn't "high-risk prices are falling." It's "high-risk prices are still rising, but slowly, and the buyer pool is starting to thin." That's what the early stages of a structural reprice look like.
What this means for buyers
Three practical takeaways the data supports:
- Don't buy in a Very-High-risk ZIP without modeling 10-year insurance premium escalation. This is the single highest-impact financial variable, and it doesn't appear in any ZIP-page metric we publish (or that any competitor publishes). Pull the actual current premium quote on a specific property, then assume 8-15% annual premium escalation through the next 10 years. Run the cap-rate or buy-vs-rent math at year 10's premium, not year 1's. The deal often looks completely different.
- Use the High and Very-High tiers as the "investigate further" signal, not the "veto" signal. A Very-High rating means the area sits in the top decile of climate-exposed US tracts. That doesn't mean disaster is imminent, and it doesn't mean the property is a bad investment. It means you need to do work the market isn't doing for you — premium projections, submarket-level comparable analysis, local conversations about insurance availability.
- Cross-reference NRI with the migration data. If a Very-High-risk ZIP is still seeing positive net inbound migration (visible on every ZIP page on this site), the buyer pool hasn't started shrinking yet. If migration is flat or negative AND the ZIP is in a Very-High-risk tier, the market may already be discounting. Migration data leads sale prices by 12-36 months.
For investors specifically: build a 10-year insurance-premium projection into every pro-forma, and stress-test the cap-rate math at 2030's likely insurance cost — not 2026's. The properties that pencil out at both numbers are the ones worth pursuing.
What FEMA NRI doesn't capture
For methodology transparency:
- Future climate trajectory. As noted in the Callout above, NRI uses historical frequency data. ZIPs whose underlying exposure is increasing rapidly get the same score as ZIPs where exposure has been flat. Sea-level rise, wildfire fuel-load changes, and hurricane intensification aren't fully reflected.
- Insurance market dynamics. Premium escalation, insurer withdrawal, and the shift toward state-of-last-resort coverage are all leading indicators that don't show up in NRI. For Florida or Louisiana specifically, current insurance market state may be a better proxy for forward-looking risk than the NRI score.
- Infrastructure adaptation. NRI doesn't credit (or penalize) tract-specific infrastructure investments: sea walls, raised electrical infrastructure, fire-hardened building codes, hurricane shutters. Two Very-High-risk ZIPs with very different adaptation profiles get the same headline score.
- Sub-tract variation. NRI operates at the census-tract level. A coastal ZIP with both beachfront and inland properties gets a single score, but the actual exposure varies dramatically within the ZIP.
The right framing: NRI is a high-quality baseline starting point, not a complete climate-risk assessment. For high-stakes property decisions in High or Very-High-risk areas, layer on at least: current insurance premiums, recent claim history for the specific structure, and any state-specific resilience-credit programs.
Frequently asked
What's a "Very High" FEMA climate risk rating? A census tract whose combined natural-hazard expected-annualized-loss-per-capita score sits in roughly the top 10-15% of US tracts. It's a relative ranking against the rest of the country, not an absolute forecast. Common in coastal Florida, parts of California (especially fire-exposed areas), Gulf Coast Louisiana and Texas, and parts of the Mountain West. Median home value across Very-High ZIPs in our dataset: $623K.
Does climate risk lower property values? Currently, only slightly. The median Very-High-risk ZIP appreciated 4.5%/year over the last five years vs. 5.8% for Very-Low-risk — a real discount of about 130 basis points annually, but nowhere near a market correction. The market is paying a premium for desirable Very-High-risk locations (Irvine, Coral Gables, Paradise Valley) that outweighs the risk discount. This pattern is likely to change, but the change is in early stages.
How is FEMA NRI different from a flood-zone designation? FEMA flood zones (the A, V, X designations) measure only flooding exposure at a specific lot. NRI combines 18 different hazards at the census-tract level — flooding is just one of them. NRI is broader; flood-zone designations are more specific to one risk. Both matter; they're complementary.
What about climate change — does NRI account for it? Only partially. NRI uses historical event frequency data, which doesn't fully reflect ongoing changes in wildfire fuel loads, hurricane intensification, sea level rise, or drought persistence. For forward-looking climate exposure, NRI is a useful starting baseline, but tract-specific climate-projection studies (where available) and current insurance-market behavior are better leading indicators.
Next steps
To find the cross-section of affordable + low climate risk + reasonable population, see the 10 affordable, low-climate-risk ZIPs with strong school coverage — built from the same FEMA NRI data this article explains. To work climate risk into a buy-vs-rent decision, the buy vs rent guide walks through how to include insurance escalation as a separate cost-of-ownership line.
Data: FEMA National Risk Index (current annual release), Zillow ZHVI (May 2026), Census ACS population (5-year estimates). Refreshed annually for FEMA NRI; monthly for ZHVI. See methodology for the full description.