Airbnb Hurricane Cancellation Policy: The 2026 Host Math

A hurricane evacuation order goes out for a Gulf Coast county in late September. Every affected guest gets a full refund within hours, no questions asked. The host, who spent the spring building a careful dynamic pricing calendar around exactly this week, gets a cancellation notice and a payout of zero dollars. That is not a bug in Airbnb's system. It is the Major Disruptive Events Policy working exactly as written, and almost nobody prices for it in advance.

Most coverage of this policy stops at explaining the refund tiers. It rarely runs the numbers on what the uncompensated cancellation actually costs a host over a season, or how small the fix actually is once the math is done properly. NOAA's own 2026 Atlantic hurricane outlook and its long-run landfall data give hosts everything needed to size that cost honestly, price for it in a way that barely moves the average daily rate, and stop treating a below-average forecast as a reason to skip the adjustment entirely.

What Airbnb's Hurricane Policy Covers

Airbnb's Major Disruptive Events Policy is the rule that lets a guest cancel a stay for a full refund, and requires the host to cancel without penalty, whenever a natural disaster, a government evacuation order, or a prolonged utility outage disrupts the destination, no matter what cancellation policy the host had set. In the portfolios we price for coastal clients, this is the single most misunderstood clause in the entire Airbnb terms of service, because hosts assume "disruptive event" means the same thing as "act of God" in a property insurance policy. It does not.

The policy was last revised on March 28, 2024 and took effect for reservations on June 6, 2024, expanding coverage to "foreseeable weather events," meaning a hurricane that Airbnb and local authorities can see coming during hurricane season still qualifies, as long as it actually triggers a government travel restriction, an evacuation order, or a prolonged loss of an essential utility like power or water at the reservation's location. A storm forecast alone does not trigger a refund. The event has to land and produce one of those three consequences before Airbnb treats it as a Major Disruptive Event for that specific address.

Coverage is scoped, per Airbnb's own policy explainer, to "large-scale events at the destination location," which in practice means Airbnb evaluates markets, not individual listings. If a county issues a mandatory evacuation order, every eligible reservation in that county qualifies at once, whether the guest's rental sits two blocks from the water or twenty miles inland. Hosts do not get to appeal the geographic boundary Airbnb draws.

Bottom line: the policy protects guests from paying for a trip they legally cannot take, and it protects hosts from cancellation penalties, but it was never designed to protect a host's revenue.

Does Airbnb Pay Hosts During Storms?

No. When a reservation is cancelled under the Major Disruptive Events Policy, by either the guest or the host, the host receives no payout for the nights that were booked. Airbnb's own resource pages state this without qualification: the guest gets a refund or credit, and the host earns nothing on those dates, even though the host still carries the mortgage, the property tax, the cleaning contract, and every other fixed cost of owning the unit that week.

This is a meaningful difference from a normal cancellation. Under a Moderate or Firm policy, a guest who cancels close to check-in still forfeits some or all of the booking value to the host, which is the entire logic behind choosing an Airbnb cancellation policy around rebooking probability in the first place. Under a Major Disruptive Event, that forfeiture disappears entirely, for every affected reservation in the market, for as long as the disruption lasts. A host running a single unit absorbs a few nights of lost revenue. A host or manager running a twelve-unit coastal portfolio can be looking at a full week of zero income across every door at once, a very different math problem from the Airbnb host cancellation penalty a host pays when they cancel for a controllable reason.

Bottom line: the host cancellation policy a host spends hours optimizing has no effect once a Major Disruptive Event is declared. That decision belongs entirely to Airbnb and local government.

How Often Hurricanes Actually Hit

A hurricane makes US landfall about 1.75 times a year on average, and a major hurricane, Category 3 or stronger, crosses the US coast roughly 0.6 times a year, according to NOAA's Atlantic Oceanographic and Meteorological Laboratory's long-run climatology. Florida alone has taken 40 percent of every US hurricane strike recorded since 1851, and Florida and Texas together account for 88 percent of every major hurricane strike in that same record. Those are not annual predictions. They are the base rate a host is pricing against every single season.

NOAA's outlook for the 2026 Atlantic season, last updated in the agency's mid-year assessment, calls for 8 to 14 named storms, 3 to 6 hurricanes, and 1 to 3 major hurricanes, against a 30-year average of 14 named storms, 7 hurricanes, and 3 major hurricanes. NOAA puts the odds at 55 percent for a below-normal season, 35 percent for near-normal, and 10 percent for above-normal. In plain terms, 2026 is forecast to be quieter than a typical year on the Atlantic basin as a whole.

A below-average forecast is not the same as a safe one for a single property. NOAA's own leadership has repeatedly reminded coastal residents that "it only takes one" landfalling storm to produce a full season's worth of damage and cancellations, and Category 5 storms have made US landfall in seasons that finished the year below the long-run average. A quiet basin-wide count does not guarantee a quiet season for a specific county.

Bottom line: even in a forecast below-average year, the long-run landfall rate NOAA publishes still implies roughly even odds that some part of the Florida coastline takes a hurricane hit before the season ends.

Building a Hurricane Risk Premium

A hurricane risk premium is a small, deliberate addition to hurricane-season nightly rates, sized to the expected value of the revenue a Major Disruptive Event could wipe out with zero Airbnb payout. It works the same way an insurer prices a policy: multiply the probability of a triggering event by the expected loss if it happens, then spread that number across the nights actually available to sell, using the same ADR and occupancy modeling that drives every other seasonal rate decision.

Applying Florida's statewide 40 percent share of US hurricane strikes to NOAA's long-run 1.75 landfalls a year produces a state-level average of roughly 0.7 hurricane landfalls touching Florida in a given year. Treating that as a simple annual rate, the odds of at least one landfall somewhere in the state before the season ends work out to roughly 50 percent. That statewide figure overstates any single listing's real exposure, since a landfall in the Panhandle does not disrupt a booking in the Keys, so a single coastal county is a meaningful fraction of that risk, not the whole of it.

For a worked example, assume a Gulf Coast host uses a conservative range of 17 to 25 percent for the odds that their specific county sees a qualifying Major Disruptive Event in a given hurricane season, well below the statewide figure, and that a typical disrupted reservation costs 4 nights at a $250 average daily rate. Expected annual loss runs from $170 at the low end to $250 at the high end. Spread across roughly 200 hurricane-season nights available to sell, that comes out to $0.85 to $1.25 a night, or 0.3 to 0.5 percent added to a $250 rate.

The premium does not need to be visible to guests as a line item. It gets folded into the same dynamic pricing adjustment a host already makes for peak weeks, applied only to the June through November nights that actually carry the exposure.

Bottom line: a Gulf Coast host can self-insure the entire uncompensated-cancellation gap for well under 1 percent of hurricane-season rate, which is smaller than most single weekend-pricing adjustments a host already makes without thinking twice.

Is Hurricane Insurance Worth the Cost?

Standard short-term rental insurance, which commonly runs $1,000 to $2,000 a year according to AirDNA's coverage guide, protects the physical property and liability exposure, not the specific revenue gap the Major Disruptive Events Policy creates. A policy that pays to repair storm damage to a roof says nothing about the booking that was cancelled three counties away because of a mandatory evacuation order, with the structure itself untouched.

Loss-of-rental-income or business-interruption coverage is the product that actually targets this gap, and it is usually a rider added to a landlord or short-term rental policy rather than something bundled in by default. It is worth a direct conversation with a carrier that specializes in short-term rentals, asking specifically whether the policy pays out when a booking is cancelled under a platform's disruptive-event terms rather than only when the structure itself sustains covered damage. Many standard riders are written around physical loss, not platform-level cancellation policy, so the exact wording matters more than the premium.

Bottom line: insurance and pricing are complementary, not substitutes. A rate premium covers the expected annual value of the gap; a well-worded business-interruption rider covers the tail risk of an unusually severe season.

What a Quiet Season Still Costs You

A forecast below-average season still produces real cancellations, because the forecast describes the entire Atlantic basin and a host only cares about one county. In the portfolios we manage across Gulf and Southeast coastal markets, the hosts who get caught off guard every year are the ones who read "below-average season" as "low risk" and skip the pricing adjustment entirely, the same mistake many hosts make with a slow season pricing strategy when demand looks soft on paper, then absorb the full nightly rate as a loss the one year a storm does make landfall nearby.

The other common mistake is applying the risk premium to the entire calendar year instead of the actual exposure window, well outside the seasonal window covered in Airbnb's own hurricane and wildfire season guidance for hosts. Hurricane season runs June through November, so a premium applied to January or March rates does nothing but make the listing less competitive during months that carry none of this risk. Confine the adjustment to the season it is actually pricing for.

Bottom line: the cost of skipping this pricing step is not the premium itself, it is the full, uncompensated nightly rate on the one booking a season that actually gets cancelled.

Frequently Asked Questions

Conclusion

Airbnb's Major Disruptive Events Policy is fair to guests and expensive to hosts who have not priced for it. The fix is not complicated: use NOAA's own published landfall rates to size a small hurricane-season premium, confine it to the months that actually carry the risk, and pair it with an honest conversation with an insurer about the business-interruption gap standard coverage leaves open. None of that requires guessing when the next storm forms. If you would rather have someone build and maintain that pricing model for you, talk to a revenue strategist.