Short answer: on a three-year hold a cost segregation study usually still comes out ahead, but by a much smaller margin than the first-year number suggests, and the case rests on two things rather than one. You keep the time value of the tax you deferred, and you give back some of it as a rate conversion when you sell. Whether that trade is worth doing depends on your marginal rate in the year you buy versus the year you sell. Here is the math, worked in dollars, on a property bought in 2026 and sold in 2029.
What actually happens to the deduction when you sell?
A cost segregation study does not create new deductions. It moves basis out of the 27.5-year residential bucket and into 5-, 7-, and 15-year buckets, which under current law are eligible for 100% bonus depreciation when the property was acquired (contract signed) and placed in service after January 19, 2025. The deduction is pulled forward, not invented. When you sell, the IRS takes it back in a specific way.
The 5- and 7-year items, the carpet and vinyl plank, cabinetry and countertops, appliances, specialty electrical and plumbing, are Section 1245 property. Gain up to the depreciation you claimed on them is recaptured as ordinary income, at your marginal rate, with no 25% ceiling. The 15-year land improvements, the driveway, fencing, site lighting and landscaping, are Section 1250 property, and the depreciation you took above what straight-line would have given you is also recaptured as ordinary income. Whatever was left on the 27.5-year schedule follows the familiar unrecaptured Section 1250 rule and is taxed at a maximum of 25%.
So the accelerated portion comes back at ordinary rates, while the depreciation you would have taken anyway comes back at 25% or reduces capital gain. That spread is the real cost of a short hold. One limit worth knowing: recapture can never exceed your actual gain, so a property that sells at or below its original cost recaptures less than the arithmetic above implies.
A worked example: a $520K rental bought in 2026, sold in 2029.
Take a single-family rental purchased for $520,000 in 2026, with land at roughly 20% of the price. That leaves about $416,000 of depreciable basis. A reasonable residential study reclassifies around 22% of it, so about $91,500 moves into the 5-, 7-, and 15-year buckets and is deductible in full in year one. At a 35% marginal federal rate, that is roughly $32,000 of tax you do not pay for 2026.
Now sell in 2029. Without the study, three years of straight-line depreciation on that same $91,500 of basis would have been about $10,000, saving roughly $3,500 in tax along the way, and the remaining $81,500 of basis would still have been sitting there reducing your taxable gain. With the study, that $81,500 of extra deduction is recaptured as ordinary income. At 35% that is about $28,500 of tax at closing, against gain that would otherwise have been taxed at 25% on the unrecaptured Section 1250 layer and 15% or 20% above it. Call the rate conversion somewhere around $8,000 to $10,000 of extra tax, depending on your bracket and how the gain stacks.
Set against that, you held roughly $28,500 of the government's money for three years. Reinvested at 8%, that is about $7,400. On those assumptions the two numbers land close to each other, which is the honest answer: at an unchanged 35% marginal rate, a three-year hold makes a study roughly a wash plus whatever you earned on the deferral.
So when does a short hold clearly pencil?
Four situations move it from a coin flip to an obvious yes. The first is a rate drop. If 2026 is a high-income year and 2029 is not, because you retire, take a sabbatical, have a loss year, or simply earn less, you deducted at 35% and recapture at 24%. That differential is pure gain and it is the single biggest lever here.
The second is a short-term rental where you materially participate. There the first-year loss offsets ordinary W-2 or business income immediately rather than sitting suspended, so the deferral is real cash in the year you need it. The third is reinvestment with a purpose: $32,000 of deferred tax that becomes the down payment on the next property is worth more than the spreadsheet differential suggests. The fourth is simply that most people who tell us they will sell in three years do not. If the hold stretches to seven or ten years, the time-value side of the ledger grows and the rate conversion stays the same size.
The flip side is worth saying plainly. If you expect your income to be higher in the sale year than the purchase year, a short hold works against you, and you should think hard before accelerating. Deferring into a higher bracket is the one version of this where the study can genuinely cost you money. A like-kind exchange changes the picture again by postponing recapture rather than triggering it, which is a separate conversation with your CPA.
What to decide before you order a study.
Three questions settle it. What is your marginal rate this year, and what do you realistically expect it to be in the year you sell? Will the first-year loss be usable now, or will it sit suspended until disposition? And is the hold period a plan or a guess? If you can answer those, the decision is usually clear within a few minutes. For a sense of the first-year figure on your own property, the savings calculator on our homepage will get you in the ballpark, and our longer piece on how depreciation recapture works when you sell after a cost segregation study walks through the mechanics line by line.