This Airbnb’s 237 booked nights need 34 separate stays to pass the 7-day test

Hosts call it a loophole. The IRS calls it an exception to the definition of a rental activity, and it runs on two tests that have nothing to do with how much money the property makes.

Both tests are counted in nights and hours, not dollars. That is why a host can pass one and fail the other without noticing.

What is the short-term rental loophole?

It changes which bucket a loss sits in, and that is all it does.

Rental property is normally passive. Publication 925 defines a rental activity as one where “tangible property (real or personal) is used by customers or held for use by customers”, with the activity’s gross income coming mainly from payments for the use of that property (Publication 925, the edition for 2025 returns). A passive loss generally waits until you have passive income, or until you sell.

The publication then lists exceptions, and the first one is the whole of what hosts mean by the loophole: an activity is not a rental activity if “The average period of customer use of the property is 7 days or less.”

Clear that bar and the property stops being a rental activity. It becomes a trade or business activity instead, and a different rule decides whether it is passive: “A trade or business activity isn’t a passive activity if you materially participated in the activity.”

So it is two tests in sequence. Average stays first, your own hours second. Fail either and the loss stays passive.

What is the property earning while this is decided?

The calculator’s pre-tax figures, which are the input to any tax question and never the answer to it.

Take the short-term rental calculator’s example property: $350,000 with an Average nightly rate of $375 and Occupancy of 65%, bought with 20% down on a 30-year loan at 7%. The example assumes a 15.5% platform fee, a co-host at 20%, $300 a month of cleaning, $150 of supplies, $350 of utilities, and 5% of revenue each for maintenance and for a reserve against big replacements.

These are example figures, not a forecast for any market. The full build of this analysis is in how to analyze an Airbnb, and what happens when bookings slip is in is Airbnb still profitable.

Nothing in that table is a tax figure, and nothing in it affects either test. Note the co-host line all the same. It is 20% of revenue, and it also costs hours that count against the second test.

What does the 7-day test actually count?

Stays, not nights, and it is an average across the year.

Publication 925 figures the average period of customer use by dividing the total days in all rental periods by the number of rentals during the year. Nights booked is the numerator. The number of separate bookings is the denominator, and it is the one hosts control least deliberately.

At $375 a night and 65% occupancy, the example property books about 237 nights across the year. Divide 237 by 7 and you get 33.9, so 34 bookings is the fewest that keeps the average under the line.

A property can be fully booked and still fail. Thirty monthly stays fill more nights than 34 weekly ones, and only the second column decides the test. A host moving toward longer furnished bookings is building a pattern that fails this test by design, which is a tax consequence of a revenue decision rather than a mistake.

What counts as material participation?

Publication 925 gives seven tests, and meeting any one of them is enough.

The three that short-term rental hosts usually argue about:

• “You participated in the activity for more than 500 hours.”
• “Your participation was substantially all the participation in the activity of all individuals for the tax year”, counting “the participation of individuals who didn’t own any interest in the activity.”
• “You participated in the activity for more than 100 hours during the tax year”, and you participated at least as much as any other individual, again including people with no ownership interest.

The other four cover significant participation activities adding to more than 500 hours, five of the last ten years, personal service activities, and a facts-and-circumstances test for participation “on a regular, continuous, and substantial basis during the year.”

Read the second and third tests next to the cost stack above. A cleaner, a co-host and a handyman have no ownership interest, and their hours still count in those comparisons. The example property pays a co-host 20% of revenue to do most of the work, which is the arrangement those two tests are measuring against.

What the exception does not do

It is not a deduction. Nothing in Publication 925 creates a loss. The exception only decides whether a loss you already have can meet non-passive income, and the loss itself comes from depreciation and running costs.

It is also not the $25,000 allowance. That one works the other way around: it applies when “you or your spouse actively participated in a passive rental real estate activity”, and it shrinks by 50% of modified adjusted gross income above $100,000, generally reaching zero at $150,000. A property that clears the 7-day test is not a rental real estate activity, so it is not on that path.

One more exception sits next to the 7-day one, for an average period of customer use of 30 days or less where the owner provides significant personal services. Publication 925 requires those services to be performed by individuals, and excludes services needed for lawful use of the property, repairs or improvements, and routine jobs like cleaning common areas.

How any of this lands on a return depends on your income, your filing status and records you have to be able to produce. A tax professional is the person to ask, and the question is worth asking before the booking calendar settles it for you.

Where to start

Pull your booking export and count rentals, then nights, then divide. Most platforms give both numbers, and the ratio is either under 7 or it is not.

Then keep a record of your own hours as the year goes, because the second test is a count of hours and not a description of effort.