Ninth Circuit Throws Out the Order on Housing Transgender Women in Oregon Prisons

Updated September 24, 2026. This article was first published on August 17, 2026, when the appeal was still pending. It has been rewritten to report the Ninth Circuit’s decision of September 17, 2026, and the earlier procedural history is kept below as background.

On September 17, 2026, the U.S. Court of Appeals for the Ninth Circuit vacated a federal magistrate judge’s order directing Oregon to presume that transgender women in state custody be housed in the women’s prison, and directed the district court to decertify the class.

Prison-housing placements for transgender people in the United States are settled by Eighth Amendment litigation, agency discretion and appellate review rather than by any single national rule, and this decision is the clearest appellate statement of that machinery so far. It belongs to the legal record tracked across our Transgender History & Education hub. What follows is what the court held, what each side argued, and what is left of the case.

What did the Ninth Circuit actually decide?

The panel vacated the preliminary injunction, sent the case back to the district court with instructions to decertify the class, and directed that its mandate issue immediately.

The opinion in S.D. v. Reese, No. 26-4872, was filed on September 17, 2026, after argument in Pasadena, California, on August 27. It was written by Circuit Judge Eric C. Tung and joined by Circuit Judges Daniel P. Collins and Lawrence VanDyke. All three were appointed by President Donald Trump. The full text is published by the court as the Ninth Circuit’s opinion in S.D. v. Reese (No. 26-4872).

The ruling is about interim relief, not final judgment. The panel did not decide whether Oregon’s placement practices violate the Eighth Amendment. It held that the plaintiffs had not carried the heavier burden that applies when a court is asked to order a state to act before trial, a standard the opinion describes, quoting circuit precedent, as requiring that the law and facts clearly favor the moving party.

What had the district court ordered?

U.S. Magistrate Judge Mark D. Clarke had ordered the Oregon Department of Corrections to start every housing review from a presumption of placement matching each person’s gender identity, absent a documented and articulable security justification.

Judge Clarke granted the motion and provisionally certified a class on April 28, 2026, then issued an amended injunction on July 7 after the state sought reconsideration. Beyond the placement presumption, the order required non-punitive housing options, barred housing transgender women with cellmates known to have histories of sexual or serious physical violence, required reasonable shower privacy, restricted cross-gender unclothed searches, mandated confidential reporting under the Prison Rape Elimination Act, and provided interim mental health support. Anyone covered could decline those protections. The case was brought in 2025 by two incarcerated plaintiffs identified in filings only as S.D. and J.F. The Ninth Circuit paused the injunction in early August and, on August 20, stayed both it and the district court proceedings pending appeal.

Why did the panel say the finding of a default presumption was wrong?

The panel held that the record did not support the district court’s central factual premise, and called that finding clearly erroneous.

Oregon submitted declarations and policy documents describing an individualized process. Newly admitted people are assessed within 72 hours of arrival and again within 30 days. Those who identify as transgender, nonbinary, genderfluid, or intersex are held in the infirmary for about a week while a Transgender and Intersex Committee gathers information, weighing medical history, mental health, how the person self-identifies, the safety of that person and of others, observed behavior, and other case-specific facts.

The state also filed placement figures the plaintiffs did not dispute: 117 transgender women in custody, 26 of whom had requested placement at the women’s prison, with 8 housed there at the time and roughly 22 placed there over the years. A later declaration counted 123. The district court had inferred an automatic policy from the fact that more than 90 percent were in men’s facilities. The panel called that inference groundless, wrote that the low rate of transfer requests could not be squared with it, and said that gaps in the evidence counted against the plaintiffs, who bore the burden of proof, rather than in their favor. Readers new to how these standards work may find our guide to transgender laws and policy a useful starting point.

What did the panel say about the evidence of risk?

It said the statistics underpinning the finding of a substantial risk of harm were too old to support it.

The district court had cited a federal study for the proposition that nearly 35 percent of transgender people in state and federal prisons were sexually assaulted between 2007 and 2021, against under 5 percent of the general prison population. The panel wrote that the underlying Bureau of Justice Statistics survey covered 2007, 2008 to 2009, and 2011 to 2012, roughly a decade before the transfer requests at issue, and that nationwide figures did not necessarily describe conditions in Oregon. It added that individual declarations of harm, even if credited, would support individual relief rather than an order covering everyone in custody.

That reasoning reaches beyond this case. Published Ninth Circuit opinions bind district courts across the circuit, and the ruling gives states a template for defending placement systems as individualized. The responses gathered in Oregon Public Broadcasting’s report on the decision fill in what the opinion itself does not.

Why was the class decertified?

The panel held that the plaintiffs had not shown the common questions that a class action requires under Rule 23.

Writing that at most the record showed different officials acting in different ways toward different prisoners at different times, the panel pointed to placement decisions that had gone in opposite directions, including a transfer granted and later reversed, and one denied because it came shortly before a release date. It drew the comparison to Wal-Mart Stores, Inc. v. Dukes, the 2011 Supreme Court decision rejecting a nationwide employment class for want of a common policy. Because the class fell, the class-wide injunction fell with it.

The panel’s public-interest analysis leaned on the record of one named plaintiff, J.F., who remains in custody and was convicted of a sexual offense against a disabled woman. The opinion treated that history as evidence that the state’s case-by-case judgment should not be displaced, and wrote that the injunction displaces the state’s role as prison administrator and compromises public safety. This case, like the earlier fight over the Justice Department’s subpoena for transgender patient records, turned on a procedural question rather than a substantive one.

How have the parties and critics responded?

Both sides described the outcome in narrow terms, and the state that won the appeal publicly distanced itself from the opinion’s wording.

John Burgess, an attorney for the plaintiffs, told Oregon Public Broadcasting that his team was disappointed but not surprised, that they continue to believe conditions place transgender women in state custody at serious risk of sexual and physical assault, and that they intend to keep pursuing the claims as the evidence develops. Jenny Hansson, a spokesperson for the Oregon Department of Justice, said the office was pleased the court agreed with its core legal arguments and that placement decisions should be made case by case rather than by across-the-board policy. She added that the office strongly disagreed with what she called the court’s harmful language, and that it did not reflect the state’s brief or its values.

The language is part of the story rather than a matter of style. The opinion places the term transgender in quotation marks throughout and describes the plaintiffs as male inmates who claim to be female. Tara Herivel, a Portland attorney who represents incarcerated transgender people, told the same outlet she read the wording as evidence of bias and expected the opinion to be cited widely. Supporters of the result read those passages differently, as a court declining to adopt a party’s preferred terminology while applying ordinary burdens of proof. Both readings sit in the public record, and the opinion is short enough to read directly. Our record of transgender history, people and milestones keeps the longer context in view.

What happens next in the case?

The case goes back to Magistrate Judge Clarke, where the plaintiffs’ individual claims can still proceed toward discovery and trial.

What is gone is the interim, class-wide order. What remains is a live lawsuit brought by named plaintiffs. Because the panel directed that the mandate issue at once, the vacatur took effect without the usual waiting period. Nothing in the decision prevents Oregon from placing people in the women’s prison when its own committee concludes that is appropriate, and the state’s filings say it has done so. For definitions and background on the terms used here, see our transgender FAQ.

Frequently asked questions

Did the Ninth Circuit rule that transgender women cannot be housed in women’s prisons?

No. The panel vacated a preliminary injunction and ordered a class decertified. It did not hold that Oregon must house transgender women in men’s facilities, and it did not decide the plaintiffs’ Eighth Amendment claims on the merits. Those claims return to the district court, where the case can still proceed toward discovery and trial.

What is a preliminary injunction, and why does vacating one matter?

A preliminary injunction is temporary relief a court grants before trial to prevent harm while a case is litigated. Vacating it removes that interim relief but does not resolve who ultimately wins. Because this injunction was mandatory, meaning it ordered the state to act rather than to stop acting, the plaintiffs had to meet a higher standard than usual.

What did class decertification change?

The district court had provisionally certified a class of all current and future transgender women in Oregon Department of Corrections custody. The panel held the plaintiffs had not shown their claims shared the questions of law or fact that a class requires. Individual claims survive, but no court order now covers the group as a whole.

What were the numbers the court relied on?

The state’s undisputed filings said 117 transgender women were in its custody, that 26 of them had requested placement at the women’s prison, and that 8 were housed there, with roughly 22 placed there over time. A later declaration counted 123. The panel said these figures undercut the claim of an automatic policy.

Why did the panel criticize the statistics in the district court’s order?

The district court cited a federal study for the proposition that nearly 35 percent of transgender people in state and federal prisons were sexually assaulted between 2007 and 2021. The panel said the underlying Bureau of Justice Statistics survey covered 2007, 2008 to 2009, and 2011 to 2012, and called the figures stale for a case about placements requested in 2021 and later.

Why did Oregon’s own attorneys object to the opinion’s language?

Oregon won the appeal, but a spokesperson for the state Department of Justice said the office strongly disagreed with what it called the court’s harmful language and said it did not reflect the state’s brief or its values. The opinion places the term transgender in quotation marks and describes the plaintiffs as male inmates who claim to be female.