The 7-day rule decides whether the IRS treats your short-term rental as a "rental activity" under the passive loss rules. When the average period of customer use is seven days or less for the tax year, the activity is not a rental activity under Treasury Regulation 1.469-1T(e)(3)(ii)(A). That single fact changes what happens to the large first-year deduction a cost segregation study creates. A rental activity is passive, and a passive loss generally waits for passive income. An activity that is not a rental activity can be nonpassive when you materially participate, and a nonpassive loss can offset wages, subject to the other limits we cover below.
The average is plain arithmetic: total days of guest use divided by the number of separate stays, for the whole tax year. It is also easy to break. Two long bookings can lift a cabin that averages 3.5 nights to 7.3 nights, and the test looks at the entire year, not at the average you meant to run. With about six weeks left in 2026, you can still check the number and change how you book.
Everything in this article is federal tax only. It is general education, not advice for your return, and your CPA makes the final call on every point.
Dev and Carla Ramanathan are illustrative owners we built to show the math. Their numbers are realistic, but they are not client results.
Dev manages a distribution center in Austin and Carla is a pediatric dentist, and together they earn about $330,000 a year in wages. In April 2026 they closed on a three-bedroom stone house outside Fredericksburg, Texas, for $900,000. They furnished it, hired a cleaner, and opened for bookings on May 1. Carla ran the listing and answered every message. Dev handled the repairs and the pricing. Their plan was simple: a short-term rental, a cost segregation study, and a large first-year loss against their paychecks.
The house did well. Guests booked long weekends and week-long stays for the wineries and the wildflowers. By late summer they had 20 stays totaling 70 nights, an average of 3.5 nights. Then the slow season arrived. A traveling consultant asked for 45 nights in September and October at a discounted rate. A family between homes asked for 45 nights in November and December. Both offers were good money in the weak months, and Dev and Carla said yes to both.
In November their CPA asked for the booking report. The year now showed 22 stays and 160 nights of guest use. She divided 160 by 22 and got 7.27.
That number is over seven. The house was no longer outside the rental definition. It was a rental activity, which meant the passive loss rules treated it as passive no matter how many hours Carla had logged. The $177,000 of extra first-year deduction their study was about to produce would be suspended instead of offsetting their wages. Nobody had done anything wrong. Nobody had checked the number while it could still be changed.
The rest of this article shows how the number is built, how to keep it under seven, what it does and does not change, and what to do if you are already over the line.
Section 469 of the tax code limits passive activity losses. Rental activities are passive by definition, with no regard to how much work you put into them. The regulations then carve certain activities out of the rental definition. The first carve-out is the one short-term rental owners care about: an activity that involves the use of tangible property is not a rental activity for a tax year if the average period of customer use for that property is seven days or less. That is Treasury Regulation 1.469-1T(e)(3)(ii)(A). The way the average is computed appears in Treasury Regulation 1.469-1(e)(3)(iii).
Four points from the regulation decide most real-world questions.
The rule is often called the short-term rental loophole. It is not a trick. It is a definition written into the regulations, and like any definition, it comes with conditions you have to meet and records you have to keep.
A few items need a decision from your CPA before they go into the math. Stays you booked and then cancelled never happened, so they generally stay out. Nights you spent in the house yourself and free nights for friends are not paying customer use. Keep them in a separate column, because they also matter for the personal use rules we describe later. And if you own more than one rental, the average is measured by activity, and how your rentals are grouped matters. Ask your CPA how your properties are treated before you assume a blended average.
Here is how the arithmetic plays out for two owners. The numbers are illustrative.
Wes and Marguerite Dunleavy bought a four-bedroom cabin near Gatlinburg, Tennessee, in March 2026 and had it ready for guests on April 9. From April 9 through December 31 they hosted 48 stays totaling 168 guest nights. Guests came for long weekends and mid-week escapes, a few for a full week.
The calculation is 168 divided by 48, which equals 3.5. Their average period of customer use is 3.5 days, well under seven, so the cabin is not a rental activity for 2026.
Now return to the Ramanathans. Their 20 ordinary stays totaled 70 nights. The two 45-night bookings added 90 nights and two stays.
| Scenario | Stays | Guest days | Average | Result |
|---|---|---|---|---|
| Ordinary stays only | 20 | 70 | 3.5 days | Passes |
| Plus two 45-night bookings | 22 | 160 | 7.27 days | Fails |
| Same year, both long stays capped at 28 nights | 22 | 126 | 5.7 days | Passes |
Two bookings, 90 nights, and the house moved from comfortable to over the line. Had they capped the long stays at 28 nights, they would still have collected good money in the slow season and kept the average at 5.7. The difference between 160 days and 126 days was 34 days of guest use that cost them the nonpassive treatment.
Every owner gets offers for long bookings in the slow season. Before you accept one, run the average with it included. Suppose you have 20 stays totaling 70 nights, a 3.5-night average, and you consider adding 30-night bookings.
| 30-night bookings added | Stays | Guest days | Average | Result |
|---|---|---|---|---|
| None | 20 | 70 | 3.50 | Passes |
| 1 | 21 | 100 | 4.76 | Passes |
| 2 | 22 | 130 | 5.91 | Passes |
| 3 | 23 | 160 | 6.96 | Passes, barely |
| 4 | 24 | 190 | 7.92 | Fails |
The fourth long booking breaks it. The average does not move in a straight line, because every long stay adds many days and only one stay to the count. A rental with many short stays has more room than one with a few. A rental with fewer total bookings has much less room. Run the table with your own numbers before the next long inquiry arrives.
The regulation has a second carve-out. An activity is also not a rental activity when the average period of customer use is 30 days or less and significant personal services are provided by or on behalf of the owner in connection with making the property available. That is Treasury Regulation 1.469-1T(e)(3)(ii)(B). Many owners of furnished mid-term rentals hope this exception covers them. Usually it does not.
Personal services have to be performed by individuals, and they have to be significant. The regulation tells you to weigh all of the facts, including how often the services are provided, the type and amount of labor involved, and the value of the services compared with what the customer pays for the use of the property. It also excludes several kinds of services from the count. Services necessary to permit lawful use of the property, construction of improvements, repairs that extend the property's useful life well beyond the average stay, and services similar to those commonly provided with long-term rentals of high-grade property, such as cleaning common areas, routine repairs and trash collection, do not count.
The regulation's own example is instructive. It describes a residential apartment hotel with an average stay between 8 and 30 days that provides a daily maid and linen service at no extra charge. The value of that maid and linen service was less than 10 percent of the amount charged for the apartments. The regulation concludes the services were not significant, and the activity was a rental activity. If daily maid service worth under 10 percent of the rent does not qualify, a weekly clean and a welcome basket are unlikely to qualify either.
A third carve-out covers extraordinary personal services, where the customer's use of the property is incidental to the services, as with a hospital's patient rooms or a boarding school's dormitories. A vacation rental does not fit.
The practical conclusion is straightforward. For most short-term rental owners, the seven-day average is the test that matters, and it is the one to protect.
Passing the test is the first gate, not the last one.
If your average is seven days or less, the house is not a rental activity for that year. The passive loss rules then ask the same question they ask about any other business: do you materially participate? If you do, the activity is nonpassive, and its loss, including a large depreciation loss from a cost segregation study, can offset wages and other income, subject to the limits below. Our guide to how the short-term rental loophole works with cost segregation shows the full sequence.
Material participation is a separate test, and the regulations list seven ways to meet it. The three owners rely on most are working more than 500 hours in the year, doing substantially all of the work, or working more than 100 hours and at least as many hours as any other individual, including your cleaner, handyman and any manager. Your spouse's hours count with yours. Marguerite Dunleavy logged 310 hours on guest messages, pricing, supplies and repairs, and the cleaner worked 140, so she clears the third test. Our article on proving material participation on your rental shows what a defensible log looks like, and our piece on short-term rentals with management companies explains why a manager can sink the hours test.
The passive loss label and the self-employment tax question are separate. In Chief Counsel Advice 202151005, released in December 2021, the IRS said that whether a short-term rental is passive under Section 469 does not decide whether its income is rent from real estate for self-employment tax. Chief Counsel Advice cannot be cited as precedent, but it shows how the IRS reasons. If you provide substantial services to guests during their stay, such as daily housekeeping or meals, the income may belong on Schedule C with self-employment tax. A turnover clean between guests is generally treated differently. Your CPA confirms the form from your facts.
If you use the house yourself for more than the greater of 14 days or 10 percent of the days you rent it at fair rental value, Section 280A can treat it as a residence and limit your deductions. Our article on the 14-day personal use rule explains how the count works. The 14-day rule and the 7-day rule have different sources and different purposes, and owners mix them up all the time.
Even a nonpassive loss runs into your basis in the activity, the at-risk rules and the excess business loss limit under Section 461(l). For tax years beginning in 2026, that limit is $256,000 for single filers and $512,000 for joint filers, adjusted annually. Losses above it are not lost. They carry forward as a net operating loss. Our post on the excess business loss limit shows how it works with a large first-year deduction.
A cost segregation study is a component-by-component analysis of what you paid for a building. It separates land, the long-life structure, and the shorter-life property the tax code lets you deduct faster, such as certain flooring, appliances, specialty lighting, driveways, fencing and landscaping. Property with a recovery period of 20 years or less qualifies for bonus depreciation, and 100 percent bonus depreciation applies to qualifying property acquired and placed in service after January 19, 2025. Our guide to 100 percent bonus depreciation explains the rule.
The Dunleavys paid $850,000 for the cabin. The county assessment supports 20 percent as land, which is $170,000. Land never depreciates, so their depreciable basis is $680,000. Our post on land value allocation explains why that split comes first. Because guests stay well under 30 days on average, the cabin is used on a transient basis, so it is nonresidential real property on a 39-year schedule. For an April placed-in-service date, the IRS mid-month table rate is 1.819 percent.
Suppose the study reclassifies 26 percent of the depreciable basis, or $176,800, to shorter-life property. Cost Seg America typically finds a reclassification rate in the range of 20 to 35 percent, and your result depends on your property. The reclassified portion takes 100 percent bonus depreciation in year one. The remaining $503,200 stays on the 39-year schedule.
| First-year depreciation | No study | With a study |
|---|---|---|
| Purchase price | $850,000 | $850,000 |
| Land (20 percent) | $170,000 | $170,000 |
| Depreciable basis | $680,000 | $680,000 |
| Reclassified to shorter-life property (26 percent) | $0 | $176,800 |
| Bonus depreciation, 100 percent | $0 | $176,800 |
| 39-year building depreciation (April, 1.819 percent) | $12,369 on $680,000 | $9,153 on $503,200 |
| Total first-year depreciation | $12,369 | $185,953 |
The study adds about $173,584 to their first-year deduction. At an illustrative blended federal rate of 32 percent, that is about $55,500 of federal tax, if they can use it. Say the cabin earned $41,000 before depreciation in its partial first year. Without a study it would show $28,631 of taxable income. With a study it shows a loss of $144,953 ($41,000 less $185,953), and because the cabin averages 3.5 days and Marguerite materially participates, that loss is nonpassive. It offsets their wages and sits well under the $512,000 joint limit.
The depreciation does not disappear. Total depreciation over the life of the cabin is the same either way, and the study moves it forward. A dollar of tax saved this year can pay down the mortgage or fund the next property. Ask any provider to show you year two and beyond, not just year one.
Return to Dev and Carla. Their $900,000 house, with $180,000 allocated to land, left $720,000 to depreciate. A study reclassifying 25 percent, or $180,000, would have produced about $188,700 of first-year depreciation against about $11,600 without one, using the May table rate of 1.605 percent. That is roughly $177,000 of additional deduction, worth about $56,700 at 32 percent.
With an average of 7.27 days, the house is a rental activity. The loss is passive. A passive loss offsets passive income and generally nothing else. Dev and Carla earn $330,000, so the $25,000 allowance for active participation, which phases out between $100,000 and $150,000 of modified adjusted gross income, gives them nothing. Neither of them works more than half their time in real estate, so real estate professional status is out. The deduction is not lost. It is suspended and carried forward, where it can offset future passive income from this or another rental, and it is released in full when they sell their entire interest in a fully taxable sale to an unrelated buyer. Our post on suspended passive losses explains the release.
That is a real outcome, but a late one. The $56,700 they expected this spring waits for passive income or a sale. A deduction used this year is worth more than the same deduction used in year nine.
Next year is a fresh test. If they cap long stays at 28 nights and keep the average at or below seven, the house is outside the rental definition again for that year. The suspended loss from the first year stays suspended, but the activity is no longer sitting in the rental category going forward.
If you own a short-term rental, these steps take an hour and can protect a five-figure deduction.
Our year-end checklist for short-term rental owners adds the paperwork to gather before the study.
A calculation is only as strong as the records behind it. An examiner asks where the average came from, and "my platform said so" is a weak answer. Keep these:
A calendar, an email trail and a log kept in real time carry far more weight than a spreadsheet rebuilt from memory during a review.
You do not need to have bought in 2026 to benefit. The IRS allows you to go back and claim deductions you never took. A look-back study analyzes the property as of the date it was placed in service, and the missed depreciation is claimed in one amount on your current return through a change in accounting method, with no amended returns. We partner with a trusted CPA specialist who handles the Form 3115 filing. Our guide to look-back studies explains the process.
The 7-day average matters for a look-back too. The catch-up lands on your current return, and whether you can use it against wages depends on the current year's average and your current year's hours. Check both before you commit.
Depreciation lowers your tax basis. When you sell for more than your adjusted basis, part of the gain comes from the depreciation you took. Gain on the 5-year components is taxed as ordinary income up to the depreciation previously taken. The 15-year site work is real property, so the bonus taken above straight-line is taxed as ordinary income and the rest is taxed with the building. Gain tied to the building's straight-line depreciation is called unrecaptured Section 1250 gain, and its maximum federal rate is 25 percent for individuals. See our guide to depreciation recapture for the full picture.
That is a trade more than a penalty. You had the use of the tax savings for years. The tax code also reduces your basis for depreciation you were allowed to take whether or not you took it, so skipping a study does not avoid recapture. It only gives up the benefit. A Section 1031 exchange can defer the gain, and property held until death receives a step-up in basis for heirs. Plan the study and the exit together.
Your deduction is only as good as the study behind it. The IRS Audit Technique Guide describes six approaches to a study. Cost Seg America uses IRS Approaches 1 and 2 only. Approach 1 builds from actual cost records, and Approach 2 builds from documented estimates reconciled to the price you paid.
Our team of analysts who perform engineering-level work studies your property component by component, using your closing documents, plans, invoices and photographs. The result is a schedule that lists each component, its cost and its recovery period. Your CPA imports it, and an examiner can trace it line by line back to the price you paid. A bargain study built on a percentage from a table hands your CPA one number with nothing underneath it. We typically recover $60,000 to $150,000 more per $1M of property value than a cheap study, because the missing money sits in the components a percentage never looks for.
Our lead credential is simple: 125+ IRS audits, zero losses. Cost Seg America has completed 16,000+ studies, and audit defense is included with every study. Cost Seg America typically finds $200,000 to $450,000 in additional Year 1 deductions per $1M of property value. Every study is Engineered, Not Estimated. You can read how the IRS evaluates a study in our post on the IRS Audit Technique Guide.
Cost Seg America works with properties valued at $250,000 or more. Below that, the fee tends to outweigh the benefit. A study may also not fit if you expect to sell within a couple of years without an exchange, because recapture arrives before the time value of the deduction has done much. If your average will run over seven days, you are not a real estate professional, and you have no passive income, the loss waits. In that case it can make sense to hold off until the facts change. A free proposal tells you which situation you are in, and we will tell you plainly when a study does not fit. Our posts on when not to do cost segregation and when a study pays for itself go through the numbers.
It is the rule in Treasury Regulation 1.469-1T(e)(3)(ii)(A) that an activity is not a rental activity for the passive loss rules when the average period of customer use is seven days or less for the tax year. Owners who also materially participate can treat the activity as nonpassive.
Add up the days of guest use across every stay in the tax year, then divide by the number of separate stays. Vacant days are not part of the calculation. Use one counting method every year and keep the booking reports.
No. The test measures how long guests stay, not how often the property is booked. A house that is empty half the year can still average three nights per stay.
The activity is a rental activity for that year, and its loss is passive. A passive loss generally offsets only passive income. It is suspended and carried forward, unless you qualify as a real estate professional, have passive income to absorb it, or fit the $25,000 allowance for owners with modified adjusted gross income under $150,000.
Yes. The test is applied for each tax year, so last year's average does not carry forward and does not decide this year's.
No. The 7-day rule is about how long guests stay and decides whether the activity is a rental activity for the passive loss rules. The 14-day rule in Section 280A is about your own use of the property and can limit your deductions if you use the house too many days.
Not necessarily. The passive loss label and the self-employment tax question are separate. If you provide substantial services to guests during their stay, the income may belong on Schedule C. A turnover clean between guests is generally treated differently. Your CPA decides from your facts.
Yes. The study still produces the deduction. What changes is who can use it this year. If you have passive income, qualify as a real estate professional, or fit the $25,000 allowance, the deduction can be used now. Otherwise it carries forward and is released on a fully taxable sale.
Yes. The IRS allows you to go back and claim deductions you never took. A look-back study claims the missed depreciation on your current return through a change in accounting method, with no amended returns. We partner with a trusted CPA specialist who handles the Form 3115 filing.
No study can promise you will never be examined, and a study does not trigger an audit by itself. What matters is the support behind the numbers. Cost Seg America has 125+ IRS audits, zero losses, and every study is built component by component with audit defense included.
Dev and Carla called their CPA the same week. The CPA could not change the year's bookings, and the house stayed over seven days for 2026. They ordered the study anyway, because the deduction still belongs to them, and their CPA planned for the loss to be suspended and carried forward. For 2027 they set a 28-night maximum stay, started a running average on a spreadsheet, and asked Carla to keep a weekly hours log. Carla had one comment. "We checked every number except the one that mattered."
Wes and Marguerite had a different spring. Their CPA read the booking report, confirmed the 3.5-day average, looked at Marguerite's hours log and the personal use calendar, and planned the loss against their wages. They never had to think about the number, because they had checked it in time.
You can start now. If you own a short-term rental, or you are about to buy one, send Jim Dougherty and his team the address, the purchase price, the closing date and the date the property was ready for guests. You will receive a free proposal within 24 hours showing the projected first-year deduction and the fee, and a plain answer on whether a study makes sense for you. Bring your CPA into the conversation, because your average stay and your hours are a decision that belongs to the two of you. Call 1-888-365-5023 or email info@costsegamerica.com.
Jim Dougherty and his team
This article is general education about federal tax rules and is not tax or legal advice. The examples are illustrative, the figures are rounded, and thresholds are adjusted annually. Talk to your CPA about your own situation before making any decision.
Use the calculator, see your number, and request your free, no-cost proposal - delivered in 24 hours, with your flat fee quoted upfront and no obligation.