The 30-day rule: what significant personal services requires

By Max Medvedev · 6 Aug 2026

The second door, and why it usually stays shut

A second door out of the rental label exists: an average stay of 30 days or less combined with significant personal services, under Reg. §1.469-1T(e)(3)(ii)(B). The services have to be hotel-adjacent attention delivered during the stay. Cleaning between guests is specifically excluded, which is why most self-managed properties do not clear it.

Two exceptions, and only one of them asks for anything

Reg. §1.469-1T(e)(3)(ii) lists six exceptions to the rental-activity definition. Short-term rentals live in the first two.

Exception (A) is a division. Exception (B) is an argument. One you verify from a booking export; the other you persuade someone about.

What the regulation actually asks

Reg. §1.469-1T(e)(3)(iv)(A) sets no percentage and no hour count. It gives three factors, on all the facts and circumstances:

1. The frequency with which the services are provided.
2. The type and amount of labor required to perform them.
3. The value of those services relative to the amount charged for use of the property.

Frequency is doing quiet work there. A turnover happens once per reservation; daily housekeeping happens every night of it. Two operations with the same annual labor total can land on opposite sides of the line purely on how it is spread across the stay.

Factor three decides most borderline cases, and it is a comparison rather than a threshold: how much of the nightly rate is buying service instead of the room.

The services the regulation names as not counting

Reg. §1.469-1T(e)(3)(iv)(B) removes three categories from the calculation before the factors are even applied:

• Services needed to permit the lawful use of the property. Permits, licences, inspections, the smoke alarms.
• Services performed in connection with improvements or repairs that extend the property's useful life substantially beyond the average rental period. The roof between bookings.
• Services similar to those commonly provided with long-term rentals of high-grade commercial or residential real property — the regulation's own examples include cleaning and maintenance of common areas, routine repairs, trash collection, elevator service, and security at entrances.

The third category is where turnover cleaning lands, and it is the one owners are surprised by. Cleaning does not fail because it is small or easy, but because the regulation treats it as what a landlord does rather than a service sold to the guest.

The common thread is timing and direction:

Why most self-managed properties miss

Two reasons, and they stack. Nobody is on site while the guest is: a self-managed listing is built around the guest not needing anyone, which is a good product and the wrong fact pattern here. And the labor is concentrated in turnover, the excluded category by name — so a hard-working owner logs a large annual total and brings nothing to the (B) analysis.

The rung above (B) is worth naming only to close it off. Exception (C), extraordinary personal services, applies at any average stay — but Reg. §1.469-1T(e)(3)(v) reaches it only where the customers' use of the property is incidental to their receipt of the services. A hospital, a boarding school. Not a rental house with good towels.

The tax you can walk into on the way

Passivity under §469 and self-employment tax under §1402 are separate questions with separate tests, conflated constantly. Services ordinary for occupancy — cleaning between guests, linens, wifi, utilities, self check-in — keep a qualifying rental on Schedule E with no self-employment tax. Hotel-like services during the stay move the income to Schedule C and add 15.3%.

The two rules read the same facts from opposite directions. Everything you would add to make exception (B) work is drawn from the list that pushes the income toward Schedule C. Clearing the passive-loss gate by becoming a hotel is a real strategy with a real price.

The 30 days that answers a different question

The same number appears in the depreciation rules, answering something else. Under §168(e)(2), a unit rented on a transient basis — average stay of 30 days or less — is not a dwelling unit, so the building is nonresidential and runs 39 years, not the 27.5 for residential rental property. That fork turns on the average-stay number alone; services have nothing to do with it.

So a property averaging 12 nights with no real services takes the worst of both readings: still a rental activity under §469, so the loss is passive per se — and still transient under §168(e)(2), so the shell runs 39 years. Failing the services test does not hand back the shorter life. The depreciation guide walks that fork.

The cheaper fix is almost always the calendar

If your average sits in the 8-to-30-night band, two projects are available. One builds a service operation that survives a facts-and-circumstances review, and accepts the Schedule C exposure. The other gets the average under seven.

Average stay = total guest nights ÷ number of reservations, per property — never divided by 365, never by days available. A handful of month-long winter bookings is usually what pushed the average up, and declining or splitting them moves the number further than anything else. The seven-day test has the arithmetic; the average stay calculator does the division on its own.

Either way, the hours still decide it

Both exceptions do one thing: remove the automatic passive label. Neither creates a deduction, and neither makes a loss non-passive. Material participation under Reg. §1.469-5T(a) is separately required — test 3, more than 100 hours and more than any other single individual, or test 1 at more than 500.

Mirch v. Commissioner, T.C. Memo 2025-128, is the clean warning: the court agreed the property met the classification test and disallowed the entire loss anyway, because credible participation came in under 100 hours. Classification opens the door; the log walks through it. The four gates in order shows where each sits.

Check which door your bookings are standing at

Run your reservation mix through the estimator — it divides the regulation's way, tells you which exception your average reaches, applies the 39-year fork, then runs all seven participation tests with a citation behind each verdict.