Gragg v. United States: REPS is not automatic participation

By Max Medvedev · 6 Aug 2026

A status is not a proof

Gragg v. United States, decided by the Ninth Circuit in 2016, separates a status from a proof. Real estate professional status under §469(c)(7) removes the automatic passive label from rentals. It does not establish material participation. That still has to be proved, activity by activity, with hours behind it.

What happened

Start with the default. §469(c)(2) makes a rental activity passive per se — material participation cannot save it, because the statute does not let you get that far. For a long-term rental, real estate professional status is the door out: more than 750 hours in real property trades or businesses, and more than 50% of all personal services performed in the year, spent there.

The position in Gragg treated clearing that bar as the end of the analysis: professional status in, non-passive losses out. The Ninth Circuit disagreed in 2016, and the deduction failed.

What the court held

Real estate professional status is not automatic material participation. Qualifying under §469(c)(7) strips the per-se passive characterization from the rentals; it says nothing about whether the taxpayer materially participated in any particular one.

So the analysis is two steps, not one, and the second step is the ordinary one everybody else runs: material participation under Reg. §1.469-5T(a), tested per activity — per property, unless the rentals have been aggregated into a single activity by election.

The tell

The tell in Gragg is a position argued from a label instead of from a record. There is no bad log here to point at. There is no log doing the work at all — a status was asked to do a log's job.

That shape is worth recognizing because short-term rental owners walk into the identical trap from the other direction. The average-stay test is also step one only. Clearing seven days under Reg. §1.469-1T(e)(3)(ii)(A) removes the automatic passive label and nothing more; the hours test is still waiting on the other side.

Mirch v. Commissioner, T.C. Memo 2025-128, is that lesson in its most expensive form. The taxpayers argued both routes — real estate professional status and the seven-day exception. The court agreed the property met the seven-day test and the entire deduction still failed, because credible hours came in under 100. Two qualifying labels, no participation, no loss.

The rule underneath both cases is the same one: every route into non-passive treatment ends at the same gate. Nothing about a status, a classification, an election or a form skips the hours.

What this means for your records

For a short-term rental owner, the first consequence of Gragg is that professional status is usually the wrong road entirely:

• The short-term rental path deliberately avoids §469(c)(7). An average stay of seven days or less takes the activity outside "rental activity," so the 750-hour and 50% tests never need to be met. A full-time W-2 earner generally cannot meet them anyway — more than half of all personal services is a high bar next to a day job.
• The workhorse is test 3 of Reg. §1.469-5T(a): more than 100 hours, and more than any other single individual. That is the race an owner with one property actually runs, and it is decided by a log, not a category.

The second consequence is about how hours are recorded when there is more than one property:

• Log per property. Material participation is tested activity by activity, so a single pooled hour total is not usable evidence for any of them.
• Grouping is an election, and it has boundaries. Reg. §1.469-4 groups short-term rentals into one activity so hours aggregate. Long-term rentals aggregate under the separate §1.469-9(g) election. The two cannot be mixed, and the grouping election belongs on a timely-filed original return, including extensions — an unfiled election means the activities are treated separately, which is a very different hour count.
• Keep the participation records seven years, alongside the corroboration that anchors them: guest message threads, vendor invoices, calendar entries, receipts, bank records.

This is what a log kept as the work happens is actually for. It records the date, the activity, the duration, the property, and who did the work — which means the same file answers the per-activity question, the "more than anyone else" comparison, and the feasibility question an examiner opens with. A status cannot answer any of those. A record answers all three at once.

Run the tests, not the label

The estimator ignores labels and runs the tests: average stay first, then all seven participation tests against your hours per property, with the citation behind every verdict. Model your year and see which gate the position actually rests on. For what the hours themselves have to look like, start at how to prove your short-term rental hours.