The thing almost everyone gets wrong about insurance
Most people quietly treat insurance like one of two things: a safety net the government guarantees, or a savings account they pay into and "get back" when something goes wrong. It is neither. Insurance is a for-profit business that sells a promise — the promise to have enough cash on hand to rebuild your house on the worst day of your life, a day that hasn't happened yet and may never happen to you specifically.
To keep that promise, an insurer pools premiums from many people, invests the float, pays out the unlucky few, and keeps the difference. That only works if the money coming in plausibly covers the losses going out, plus a cushion. The whole industry runs on a single unglamorous question: can we collect enough, from enough people, to pay the claims we expect — and still be standing afterward?
Insurance is heavily regulated — and that's why it leaves
Every state has an insurance department whose core job is solvency: making sure companies keep enough capital and reserves to pay the claims they've promised. An insurer can't legally collect premiums for a risk it can't fund. So when the expected losses in a region climb past what the company is allowed to charge — or past what's even knowable — the company has two honest options: raise prices a lot, or stop writing policies there.
This is the piece that feels like betrayal but isn't. People assume a giant insurer "can afford it." But the regulator won't let a carrier knowingly write business it can't back, because an insolvent insurer doesn't help anyone — it leaves thousands of paid-up policyholders with worthless promises. Pulling out of an unprofitable, unpredictable market is often the regulated, responsible move, not the greedy one.
What "uninsurable" actually means
An area becomes hard to insure for one of two reasons, and they're different. The first is that it's simply too likely to burn, flood, or blow away — the expected annual loss is so high that an honest premium would cost more than the mortgage. The second is subtler and scarier to an actuary: the risk has become unpredictable. Insurance prices risk by studying the past. When wildfire seasons stop following the old patterns, the historical model breaks, and a business that can't price a risk can't responsibly sell coverage for it.
The 2025 Los Angeles fires are the textbook case. The Palisades and Eaton fires destroyed roughly 12,000 structures and produced tens of billions in insured losses in a matter of days — in neighborhoods that, for decades, were considered ordinary suburban risk. The California Sierra and the broader wildland-urban interface have the same problem: beautiful places where homes sit inside the fuel. Hurricane coasts from Texas to Florida face the wind-and-surge version of it.
Then there are the places nobody expects. New Mexico is now treated as a year-round wildfire state, and between 2021 and mid-2024 the top insurers issued more than 10,000 homeowner non-renewals there. If you only picture California and Florida when you think "insurance crisis," New Mexico is the quiet reminder that the map is bigger than the headlines.
What people actually pay: from about $650 to over $11,000 a year
Here's the part that puts your own bill in perspective. The average U.S. homeowners policy runs roughly $2,500 a year for a typical $300,000 home. But that single number hides one of the widest price ranges in any consumer product — and it's driven almost entirely by where the house sits, not how nice it is. Estimates differ by study and coverage level, but the shape is always the same:
- Lowest-risk states — Hawaii, Vermont, Delaware, Utah: roughly $650–$1,300 a year. Mild weather, lower rebuild costs, and few region-flattening disasters. (Hawaii's headline ~$650 even leaves out hurricane coverage, which is sold separately there.)
- The national average: about $2,500 a year.
- The tornado-and-hail belt — Oklahoma, Nebraska, Kansas: roughly $4,700–$5,900 a year. Oklahoma routinely ranks the most expensive non-coastal state in the country.
- Hurricane country — Louisiana around $8,500 a year, and Florida anywhere from $7,000 to nearly $12,000 a year depending on the dataset. Florida is in a league of its own, regularly costing several times the national average.
And those are statewide averages — drop down to a single ZIP code and the gap gets wider still. A wind-mitigated inland Florida home in a place like Ocala might pay under $2,500, while a coastal Miami-Dade or Palm Beach property of the same value can run $5,000–$7,000 or more on wind exposure alone. In California wildfire zones, owners pushed onto the FAIR Plan often stack a bare-bones fire policy plus a separate "wraparound" policy for everything the FAIR Plan won't touch — and still pay a multiple of what a low-risk state charges for full coverage.
Enter the "insurer of last resort": what a FAIR Plan really is
When private carriers won't write a property, most states have a backstop called a FAIR Plan — Fair Access to Insurance Requirements. It is not a generous government program. It's a bare-bones, shared pool that licensed insurers in the state are forced to participate in, designed to give an otherwise-uninsurable property some coverage so it can still be sold or mortgaged. You generally have to be turned down by the regular market first, and the coverage is deliberately thin — often paying actual cash value (depreciated) rather than full replacement cost, with hard dollar caps.
New Mexico is a good example of why this is suddenly in the news. Its FAIR Plan has technically existed since 1969, but it sat in the background for decades. As insurers retreated, the state had to wake it up: in 2025 the maximum residential limit was raised from $350,000 all the way to $750,000, commercial limits were lifted to $2 million in early 2026, and applicants now have to sign an affidavit confirming the private market turned them down. A sleepy 1969 statute is being rebuilt in real time into a frontline product.
California shows where that road leads at scale. Its FAIR Plan saw enrollment jump about 43% between September 2024 and December 2025 as carriers retreated, then took roughly $4 billion in losses from the January 2025 fires — against an exposure that had ballooned to hundreds of billions of dollars, up around 300% since 2020. The plan runs essentially cash-in, cash-out, and when its reserves ran dry, regulators approved a $1 billion assessment on insurance companies to keep the checks flowing.
The other backstops: wind, flood, and earthquakes
FAIR Plans are just one of several "residual markets." A few others matter, and the misconceptions around them are expensive:
- Texas windstorm (TWIA): After Hurricane Celia in 1970, insurers stopped covering wind and hail on the Texas coast, so in 1971 the state created the Texas Windstorm Insurance Association. It covers wind and hail only — not flood — for 14 first-tier coastal counties and part of Harris County, and it's required to keep enough funding to survive a one-in-100-year storm.
- Flood (NFIP): This is the big one people don't know. A standard homeowners policy does not cover flood — at all. Flood coverage comes from the federal National Flood Insurance Program (or a private flood policy). The NFIP insures about 4.7 million policies and roughly $1.3 trillion in property, and it is currently around $22.5 billion in debt to the U.S. Treasury after decades of catastrophic seasons. It even lapsed for weeks during the 2025 government shutdown, freezing home sales that needed flood coverage to close.
- Earthquake: Also excluded from standard homeowners policies. In California, most quake coverage runs through the California Earthquake Authority — a publicly managed, privately funded pool — and it's a separate policy with its own (often steep) deductible.
The common thread: the perils most likely to flatten a whole region at once — flood, quake, coastal wind, wildfire — are exactly the ones private insurers carve out or hand off to a special pool. That's not a loophole. It's the system admitting that some risks are too big and too correlated for a normal company to carry alone.
The fine print nobody reads: war, riots, and nuclear
Two exclusions in almost every homeowners policy reveal how insurers think, and the distinction between them is genuinely interesting. Most U.S. policies are built on a standard industry form, and that form draws a sharp line between war and civil disturbance.
If a riot, a protest, or general civil commotion damages your home — broken windows, looting, vandalism — that's typically covered. Your policy treats it like any other sudden, accidental loss, and your additional-living-expenses coverage may even pay for a hotel if you're displaced. But damage from war — declared or undeclared, including invasion, insurrection, and rebellion involving real military force — is excluded, and no standard rider will buy it back. The reason is the same solvency math: a normal insurer cannot price or survive a war, where losses arrive everywhere at once.
The legally fascinating gray zone sits right on that line. A localized riot is covered; an organized armed insurrection is not — and the same event can be argued either way in court, which is exactly why insurers have been adding "insurrection" to exclusion lists in recent years. The label decides the payout.
Nuclear is its own special case. A "nuclear hazard" — reaction, radiation, contamination — is excluded from homeowners policies. But there's a quirk most people never notice: if a nuclear event causes an ordinary fire, the resulting fire damage is generally still covered, because fire is a named peril. Liability for nuclear power plants themselves doesn't sit on your homeowners policy at all; it runs through a dedicated federal framework (the Price-Anderson system) that pools the industry and adds a government backstop — yet another admission that some risks are simply too large for any single insurer.
So what do you actually do about it?
Understanding the machine changes how you shop. A few practical takeaways that fall out of all this:
- Don't wait to be dropped. If you're in a wildfire, coastal, or flood-prone area, shop early and keep options open. Non-renewals come with little warning, and the last-resort plans are thinner and pricier than what you have now.
- Know what your policy doesn't cover. Flood and earthquake are almost certainly not in your homeowners policy. If you need them, you buy them separately — before the storm, not after.
- Check replacement cost vs. actual cash value. Many LA fire survivors discovered their limits were far below today's rebuild costs. A FAIR Plan paying depreciated value can leave a painful gap.
- Mitigation is leverage. Defensible space, fire-resistant materials, and home-hardening increasingly affect both whether you can get coverage and what you pay. Regulators are actively pushing carriers to reward it.
The headline version of all this is "insurance companies are greedy and abandoning people." The truer, less satisfying version is that insurance is a regulated promise that has to be funded in advance — and as the risks get bigger and harder to predict, the price of that promise is rising, the places it can't be sold are spreading, and the cost of the backstops is quietly landing on all of us. Knowing that won't lower your premium. But it will stop the next renewal letter from feeling like a personal insult, and help you read the one thing in your policy that actually matters: the part that says what isn't covered.