Is Cost Segregation Worth It on a Single Rental? Only If You Can Use the Loss.

A pitch email lands in a landlord's inbox every October. A cost segregation study on your rental will unlock $70,000 of first-year depreciation, the fee pays for itself, and you should move before December 31. The deduction number in that email is usually honest. Whether cost segregation is worth it on a single rental is a different question, and it turns on something the email never mentions: a deduction and a tax saving are not the same thing on your return, and for most people with one rental, they are not even close.

Left alone, a rental house is one asset depreciated straight line over 27.5 years. A study takes the purchase apart. Appliances, carpet, window coverings, and similar contents become five-year property, a classification the IRS corrected into place in Announcement 99-82 after its own forms had called them seven-year property. Driveways, fencing, concrete flatwork, and landscaping become 15-year land improvements under asset class 00.3 of Rev. Proc. 87-56. Only the shell stays on 27.5 years.

That reshuffling matters because the short-life classes qualify for 100% bonus depreciation under IRC §168(k), which the One Big Beautiful Bill Act made permanent for property acquired after January 19, 2025 (OBBBA §70301). Used property counts, so buying a 1998 house from a stranger is fine. The whole carve-out is deducted in year one.

None of this is aggressive. The Tax Court blessed the method in Hospital Corp. of America v. Commissioner, 109 T.C. 21 (1997), and the IRS publishes the Cost Segregation Audit Techniques Guide, Publication 5653 (2-2025), telling its own examiners how to review a study. It is an allocation backed by an engineering report, not an election you check on a form.

Section 469(c)(2) calls rental real estate a passive activity by definition, whether or not you lift a finger on it. Passive losses only offset passive income. The one relief valve for ordinary landlords is §469(i), which allows up to $25,000 of rental loss against other income with active participation, then takes 50 cents of that allowance for every dollar of modified AGI above $100,000 and leaves nothing at $150,000. I have written up how that $150,000 ceiling works on its own.

Above the ceiling, the loss a study manufactures does not vanish, and it does not reduce this year's tax either. It lands on Form 8582 as a carryforward. So the real question is not what size deduction a study produces. It is who is standing on the other side of §469 when it arrives.

Three situations get the loss past §469, and a fourth gets a slice of it. If none of them describes your 2026, the study is a $4,000 purchase of a carryforward.

• –A short-term rental you materially participate in. An average guest stay of seven days or less is not a rental activity at all under Treas. Reg. §1.469-1T(e)(3)(ii)(A), so clearing a material participation test makes the loss nonpassive against wages. This is the common path, and it is why cost segregation and short-term rentals get sold together. The tests and the logs are covered in the short-term rental loophole.
• –Real estate professional status. Section 469(c)(7)(B) wants more than 750 hours in real property trades or businesses and more than half of your personal services there. One rental almost never gets a W-2 earner across that line, and a full-time job makes the second half impossible, which is why this door usually opens for a spouse rather than for you.
• –Passive income already on the return. Other rentals that turn a profit, or a different property you are selling this year, absorb the new loss dollar for dollar.
• –Modified AGI under $150,000 with active participation. The §469(i) allowance tops out at $25,000, so most of a $74,692 loss still suspends, but a quarter of it lands. Below $100,000 of modified AGI the full $25,000 is available.

The five-year property a study carves out is §1245 property, and §1245 recapture is ordinary income at rates up to 37% with no ceiling. Depreciation on the shell comes back as unrecaptured §1250 gain capped at 25%. Accelerating basis into the short-life classes therefore trades a 25% item for a 37% item at the exit, which is the piece the pitch email never prices. The rest of the exit math is in how to avoid depreciation recapture.

In fairness to the study, that trade mostly cancels for the suspended owner. A fully taxable sale releases the whole Form 8582 stack under §469(g) against ordinary income in the same year the extra §1245 recapture shows up, and the released losses run through the same brackets. So Owner A did not destroy value. Owner A paid $4,000 and years of extra depreciation schedules for a timing difference that roughly washes. Swap the sale for a 1031 exchange and both sides stay frozen, which makes the fee the only real number in the transaction.

Wait, and buy the study in the year the loss has a job. The reason waiting costs nothing is the one piece of timing almost everybody has backward. The December 31 deadline applies to placing the property in service, not to the study. A study is documentation supporting a return, so it can be ordered in February and still carry the prior year. And a property you bought in 2023 is not a lost cause: a look-back study goes on Form 3115 under DCN 7 as an accounting method change, with the entire catch-up deducted in the year of change as a negative §481(a) adjustment under Rev. Proc. 2025-23. No amended returns.

That is what makes the October pitch so frustrating. The pressure is manufactured. The deduction keeps, the fee does not get cheaper for hurrying, and the only thing that actually expires is your chance to buy the study in a year when it is worth $17,160 instead of $0. If you are buying a short-term rental or a spouse is going full time into real estate, run it. I do that math for clients in the high-W-2 real estate playbook. With one long-term rental and two W-2s, keep the $4,000.