The short-term rental loophole, and why it is not one

By Max Medvedev · 6 Aug 2026

It is a regulation, not a loophole

The thing people call the short-term rental loophole is one sentence of Treasury regulation, written in 1988 and untouched since. Reg. §1.469-1T(e)(3)(ii)(A) says an activity is not a rental activity when the average period of customer use is seven days or less.

That matters because §469 makes rental losses passive by default, and passive losses cannot touch W-2 income. Strip off the "rental activity" label and the default disappears. The loss is not automatically passive any more — and if you materially participate, it becomes non-passive and can offset your wages.

No special election. No aggressive reading. A rule that has sat in the regulations for nearly forty years, applied to a property type that did not meaningfully exist when it was written.

So why does it get called a loophole? Because the outcome feels like one. A surgeon earning $400,000 cannot become a real estate professional — that route needs 750 hours and more than half of all working time in real property trades, which a full-time job forecloses. The short-term rental route asks something a busy professional can actually do: materially participate in one activity. Same code section, a door that happens to be open to people the other door excludes.

Calling it the STR loophole is fine as shorthand. Filing it as though it were one — a trick, a structure, something that works because nobody is looking — is how people lose it.

What it is actually worth

The seven-day test opens the door. Depreciation is what makes walking through it worth anything.

A cost-segregation study reclassifies roughly 20–30% of a building's basis out of the long building life and into 5-, 7- and 15-year property. Since OBBBA made 100% bonus depreciation permanent for property acquired after 19 January 2025, that reclassified slice is written off entirely in year one rather than spread across decades.

Note the life that applies underneath: a short-term rental averaging 30 days or less is nonresidential property on a 39-year life under §168(e)(2) — not the 27.5 years most rental calculators assume. It is a genuinely counterintuitive result, and getting it wrong changes every number downstream.

Here is a year where the whole thing holds together, computed by the engine rather than asserted:

Every figure there traces to a rule and a citation. That is what makes a position defensible — not the size of the number, but whether each part of it can answer for itself.

Four gates, and the loophole is only the first

Passing the seven-day test tells you almost nothing on its own. Four separate things have to be true in the same year:

Gate three is the one that quietly voids returns. Use the place yourself for more than the greater of 14 days or 10% of the days rented at fair value and it becomes a residence under §280A(d)(1); §280A(c)(5) then caps deductions at rental income. There is no loss to claim that year, and no hours log repairs it. Family stays count even at full price.

The full qualification path walks all four in order, with the day counts that differ between them.

Where the claim actually fails

Almost never on the rule. Whether you call it the STR loophole, the Airbnb tax loophole or the seven-day rule, nearly every recent loss is an evidence failure rather than a legal one.

Mirch v. Commissioner, T.C. Memo 2025-128, is the cleanest demonstration. The court agreed the property met the seven-day average-stay test. The owners still lost the entire deduction, because out of roughly 920 claimed hours it found credible participation of under 100. The law was never in dispute for a moment. The log was.

That is the shape of the risk, and it is worth stating plainly because so much of the marketing around the STR loophole skips it: the tax treatment is settled and the paperwork is not. What decides these cases is whether the hours were recorded as the work happened, with dates, start and stop times, concrete tasks, and the hours of everyone else who worked on the property. Records rebuilt at filing time have been dismissed as guesswork more than once.

What to do with this

If you are deciding whether the strategy fits your year at all, the honest sequence is: check the average stay first, because it is arithmetic and it either passes or it does not; then look hard at whether you can win the hours, because that is where the claim is actually lost; then, and only then, work out what the depreciation is worth.

Run your own numbers through the estimator — it does the average-stay division the regulation's way, runs the seven participation tests, applies the §280A screen, and shows the loss gates in order, with a citation behind each step. For the division on its own there is an average stay calculator, and for the hours, a free log template with no email wall.