Every September we get some version of this call. An owner has run a property as a nightly rental since spring, the season is winding down, and a tenant has offered a twelve-month lease starting October 1. The offer is good. The worry is that signing it quietly undoes the tax strategy the whole year was built around, because the accelerated depreciation from a cost segregation study is only usable against ordinary income if the property clears the short-term rental tests. So the question comes out as: if I sign this lease, do I lose the deduction for the year?
Usually not, and the reason is a detail in how the test is computed that almost nobody gets right on the first read.
How is the average stay actually calculated?
The rule people remember is that a rental with an average period of customer use of seven days or less is not treated as a rental activity, which is what opens the door to deducting losses against ordinary income if you materially participate. The part people misremember is how that average gets computed. It is not total rental days divided by 365, and it is not a look at how the property is being used on December 31. It is the sum of each period of customer use divided by the number of those periods. Every booking is one period. A twelve-month lease is also one period.
That distinction does most of the work. A single long lease signed in the fall is one entry in a denominator that already has dozens of short stays in it, so it moves the average far less than owners expect.
A property that switched on September 1
Say the property took 96 separate bookings between January and August, averaging four nights each. That is 384 rental days across 96 periods. On September 1 the owner signs an annual lease, and 122 of its days fall inside this tax year. The calculation is 384 plus 122, or 506 total days, divided by 97 periods. The average period of customer use for the year is about 5.2 days.
That is still under seven. For this tax year, the property is not a rental activity under those rules, and whether the loss is passive turns on material participation rather than on the lease the owner signed in the fall. Next year is a different story: one tenant, one period, 365 days, and the property is an ordinary long-term rental whose losses are passive unless something else changes.
Does the switch cost you material participation?
This is the half of the test that actually gets people, and it is worth being honest about. Material participation is measured for the tax year, so the hours you logged in spring and summer still count. What the lease change does is stop the clock. If you were relying on the test that requires more than 100 hours and more participation than anyone else involved with the property, and you were sitting at 80 hours on September 1 expecting to finish the year with turnovers and guest messages, those hours are not coming. A long-term tenant generates very little owner activity.
So the move is to count now rather than in April. Pull the hours you have actually documented for the year, see which test they support, and find out while there is still calendar left to do something about it. The documentation standard for material participation is contemporaneous and specific, and reconstructing a year of it from memory is exactly the position that does not hold up.
Is the depreciation you already claimed at risk?
No. Changing the lease term does not claw back depreciation, and it does not require the study to be redone. The property is still in service in a rental activity, the same assets are still there, and the 5-, 7-, and 15-year components a study identified keep running on their schedules. What changes is prospective and it is about the character of future losses, not about reversing a deduction you have already properly taken.
Worth separating from that: selling the property is where depreciation comes back, through recapture at sale. A lease change is not a sale.
The math on a $540,000 property
Take a furnished short-term rental bought in March 2026 for $540,000 and placed in service in April. Carve out land at 20% and $432,000 is depreciable. A furnished nightly rental tends to produce a strong study result, so call it 25% of basis reclassified into 5-, 7-, and 15-year property: about $108,000. Because the property was acquired (contract signed) and placed in service after January 19, 2025, 100% bonus depreciation applies to that $108,000 in the year placed in service. The remaining $324,000 stays on the 27.5-year schedule and contributes about $8,350 for the partial year, for roughly $116,350 of first-year depreciation.
At a 35% marginal federal rate, that is roughly $40,700 of tax savings if the loss is usable against ordinary income this year. If material participation is not met, the deduction is not gone, it is suspended and carries forward against future passive income or the eventual sale. The lease signed in September did not decide which of those outcomes you get. The hours you documented between January and August did.
What to do before December 31
Three things, in this order. Count your periods of customer use and run the average properly, one entry per booking and one for the lease. Pull your participation log and compare it against the test you are relying on. Then take both numbers to your CPA before year-end, while decisions like whether to keep the nightly listing open for a few more weekends are still decisions rather than history.
If you bought the property this year and have not run a study on it yet, the free instant estimate will tell you the size of the deduction you are deciding about, which tends to make the rest of the conversation shorter.