The Supreme Court settled that IEEPA confers no tariff authority, but not which importers and which entries can be refunded, or how. After Trump v. CASA limited universal injunctions, the June 4 Rule 23 class certification motion tests whether a class action is the procedural vehicle that can carry relief to nonlitigant importers whose finally liquidated entries sit outside CAPE's current scope.
Written byTRAVERSE Research
Primary lensEntry posture review
Evidence base12 records used
Use caseSaved scope review
The tariffs are gone, and the fight now is whether a class can reach the importers CAPE leaves out
The Supreme Court killed the IEEPA tariffs in February and said almost nothing about how the money comes back. Most importers will get theirs through CBP's CAPE process no matter how the appeal or the class motion turns out. The hard cases are the entries CAPE does not reach, above all the finally liquidated entries of importers who never sued, and those are exactly what the June 4 class motion is trying to rescue. How it ends turns on two things, whether the Federal Circuit lets Trump v. CASA cut down the CIT's refund order, and whether the CIT is willing to certify a refund class under Rule 23.
What the Court decided, and what it left open
The Court answered one question and left another wide open. It held that IEEPA gives the President no power to impose tariffs. It said nothing about how the money already collected comes back, to whom, on which entries, or by what process.
Congress
Find lawmakers and committees
In February 2026 the Court held 6 to 3, in Learning Resources, Inc. v. Trump (consolidated with V.O.S. Selections), that IEEPA does not authorize the President to impose tariffs. Everything about the refunds then fell to the CIT on remand, to CBP's own claims process, and to a running fight over whether the court can make CBP reliquidate entries it has already closed.
In practice the refunds have split in two. CBP can handle some of them itself, the unliquidated entries and the ones liquidated recently enough to run through its CAPE system, the Consolidated Administration and Processing of Entries. The rest sit outside CAPE, and the sharpest of those are the finally liquidated entries of importers who never went to court. Those are what the June 4 motion is about.
So the question here is not whether the tariffs were lawful. The Court has settled that. It is whether the refund can actually reach the finally liquidated entries of importers who never sued, and if so, how. After CASA, that comes down to whether a universal refund order can stand, or whether the plaintiffs need a Rule 23 class instead.
How the orders, the appeal, and the June 4 filings stack up
Judge Richard K. Eaton has been pushing CBP hard. He first ordered refunds on March 4, in the Atmus case, widened the order on March 27 to cover IEEPA entries whatever their liquidation status, and on April 17 issued the operative order directing refunds across three entry categories, though he stayed it while CBP built the plumbing. CBP brought CAPE Phase 1 on April 20 online, aimed at unliquidated entries and at recently liquidated ones still inside the 19 U.S.C. §1501 ninety day voluntary reliquidation window. On May 27, losing patience with the pace, Eaton issued a show cause order asking why the stay should remain, and told CBP Commissioner Rodney Scott to appear in person on June 9 and testify.
The government pushed back. On May 29 it told the court it would appeal and asked Eaton to amend his order, arguing the universal version reached too far. It also tried to send a lower official in Scott's place under the apex doctrine, which Eaton refused the same day. Then on June 2 it formally appealed to the Federal Circuit, going after both the refund order and the order making Scott testify, and filed a petition for a writ of mandamus to stop the testimony, asking for a ruling by June 5 so it could still reach the Supreme Court before the June 9 hearing. None of this halts the refunds already moving. CBP has kept processing through CAPE and, as of May 27, said it was refunding roughly $85 billion, more than half of what importers paid.
Two things landed on June 4. Terry Precision Cycling, one of the five small businesses behind V.O.S. Selections, filed a Rule 23 class certification motion to cover every importer that paid these duties on entries CAPE currently leaves out, and to force refunds for them. The same plaintiffs separately pressed Eaton to lift his stay, open CAPE to the whole class, and pay interest. The government's brief that day held its line, that finally liquidated entries can be refunded only importer by importer, each with its own court order.
Read the two filings together and the class motion is the plaintiffs' insurance. If the Federal Circuit knocks out the universal refund order on appeal, a certified class would carry the same relief by another road.
The CASA question the Federal Circuit must answer
All of this runs straight into CASA. In 2025 the Supreme Court held in Trump v. CASA, Inc., again 6 to 3, that federal courts cannot issue universal injunctions, relief that sweeps in people who are not parties, beyond what the actual plaintiffs need to be made whole. Complete relief, the Court said, is not the same as relief for everyone.
What CASA did leave standing is the class action. The majority treated Rule 23 as the proper way to reach a large group and called the universal injunction a class action workaround. That is exactly why the June 4 motion exists. Certify a class and every member becomes a party, so an order covering the class is relief for parties, not the forbidden relief for nonparties.
The same opinion cuts the other way too. Justice Alito, joined by Thomas, warned that courts should not read the ruling as license to certify nationwide classes loosely, or the universal injunction would just come back wearing a Rule 23 label. The government will lean on that line and demand a strict reading of every element.
There is also a question of whether CASA even binds this court. In Euro-Notions Florida, Eaton decided it does not apply to the CIT, because Congress gave the court nationwide reach, exclusive jurisdiction over IEEPA claims, and a duty to keep customs treatment uniform across the country. The weak spot is that the Federal Circuit has already vacated the merits stage universal injunction in V.O.S. Selections and sent it back to be reconsidered under CASA, which is not the move you make if you think the CIT is a blanket exception. Plaintiffs answer that the remand was about a merits injunction, not the refund order now in play, but the wind is at the government's back. The two questions are linked in reverse. The more firmly the CIT can say CASA does not reach it, the less the plaintiffs need a class, and the shakier that footing, the more the class becomes their only real route.
Why class certification is harder than it looks
Certification runs on the familiar Rule 23 test. A class needs numerosity, commonality, typicality, and an adequate representative under 23(a), plus a fit with one of the 23(b) categories, and the court has to look hard at all of it under the standard from Wal-Mart Stores, Inc. v. Dukes.
Two of those are easy for the plaintiffs. The class is plainly large enough, on the order of 330,000 importers, 53 million entries, and around $166 billion in collected duties by CBP's own court filings, so suing one at a time is hopeless. The common question is unusually clean as well, whether the CIT can order reliquidation for the entries CAPE does not reach, the finally liquidated ones in particular, the kind of single yes or no question Dukes wants a class to settle in one stroke.
The trouble is typicality and adequacy. The five V.O.S. Selections plaintiffs are already in court, and some of them may not even hold the finally liquidated, CAPE ineligible entries that define the class they want to lead. Terry Precision Cycling, the importer that filed the motion, is reported to have told the court in its own complaint that it passed the tariff costs on to its customers. That hands the government a clean argument, an importer that pushed the cost downstream is not in the same spot as one that ate it, and a lead plaintiff whose own facts drift from the class is the softest part of the motion.
Then comes the 23(b) problem, which may be the real one. On paper the plaintiffs can dress the refund up as an injunction telling CBP to reliquidate. In substance it is a demand to hand back an enormous sum of money. The government will say a money refund class cannot ride through Rule 23(b)(2), which Dukes reserved for indivisible injunctive relief rather than individualized cash. Push the case into 23(b)(3), the proper home for money claims, and the plaintiffs pick up two more burdens, predominance and superiority, and the government gets to argue that the CIT's ordinary test case practice already does the job better than a class would.
So the weakness here is not just that the CIT rarely certifies classes. It is the whole stack, the strain between a cash refund and 23(b)(2), the Alito warning hanging over any nationwide class, and the heavier test that comes with 23(b)(3).
The court's own record points the same way. Its best known refusal, Gilda Industries v. United States (Slip Op. 08-51, 2008), turned down a class partly because nobody could even identify the members, since Customs would not release importer names, and the Federal Circuit let that stand. This case largely fixes the identification problem, because CBP's ACE and CAPE records show exactly which entries carried IEEPA duties and who imported them. The fairer way to put it is not that the CIT never certifies, but that in big customs refund fights it has reached for test cases and case management instead of a class.
Add it up and the plaintiffs are strong on the numbers and on the common question, but exposed on the representative, on the awkward fit between a cash refund and 23(b)(2), on the extra hurdles if the case slides into 23(b)(3), and on a court that prefers test cases. With the appeal already pending at the Federal Circuit, the CIT also has every reason to wait for that ruling before opening a new front.
Whether the court can even reach finally liquidated entries
A second line of authority is about power, not class mechanics. In Shinyei Corp. of America v. United States, a §1581(i) case, the Federal Circuit held that the fact an entry has liquidated does not, by itself, strip the CIT of its power to grant relief under the APA. The customs statute does not bar the reliquidation the plaintiff asked for, and the deemed liquidation rules do not stop the court from ordering it. The distinction it drew matters here. The finality in §1514 attaches to Customs' own calls, classification and valuation, where the protest route under §1581(a) is the only way in. When the challenge is to a charge someone else imposed, Commerce's instructions in Shinyei, the President's IEEPA orders here, the case belongs under the APA and §1581(i), and §1514 finality is beside the point. Since CBP only collected these duties mechanically, with no discretion, the plaintiffs say they were never protestable to begin with, so finality is no wall.
The CIT has since split on how far that power runs. In AM/NS Calvert (Slip Op. 23-129, 2023), Judge Baker said plainly that the court can order reliquidation of finally liquidated entries in a §1581(i) APA case, calling it a garden variety remedy. A three judge panel in In re Section 301 Cases (2021) was more doubtful, and used that doubt to justify freezing liquidation instead, over a dissent from Chief Judge Barnett. The Federal Circuit has never resolved the disagreement.
None of this is a sure thing for the plaintiffs. Shinyei and AM/NS Calvert show the CIT has the power to order reliquidation. They do not show that the power reaches the finally liquidated entries of importers who never sued. Whether it stretches that far, to nonparties, is the CASA and Rule 23 question all over again.
What CAPE covers, and what it leaves out
It is worth being precise about a distinction that keeps getting blurred, because outside CAPE and finally liquidated are not the same thing. CBP usually liquidates an entry about 314 days after it arrives. After that, three things can keep liquidation from going final, a voluntary reliquidation under §1501 within 90 days, a protest under §1514 within 180 days, or a freeze imposed by statute or a court. Without one of them, liquidation becomes final and conclusive against everyone. Low value informal entries, of which there are several million, liquidate almost on arrival and hit that point fastest, a wrinkle Eaton flagged on its own.
Keep the three states apart. CAPE Phase 1 handles entries that are still unliquidated or still inside the §1501 ninety day window, the ones CBP can clean up administratively. Past that window sit two different things that often get lumped together, complicated entries such as AD/CVD suspensions, drawback, and reconciliation that are not necessarily final, and entries that are truly final, past the 180 day protest period and fully finalized, which is where the government fights hardest against reliquidation. Outside CAPE, then, is a far broader and looser category than finally liquidated.
For anyone tracking entries, that means keeping the 90 day CAPE line and the 180 day finality line apart. Treating outside CAPE as a synonym for finally liquidated is how people misjudge their own exposure.
The government's own words may bind it
The plaintiffs also have the government's own words to throw back at it. In AGS Company Automotive Solutions v. CBP (Slip Op. 25-154, Dec. 2025), the government told the CIT that if these duties fell it would not object to reliquidation or refund. The court took that as a firm commitment and held the government estopped from walking it back. Having spent late 2025 and early 2026 using that promise to beat back importers' injunction motions, on the theory that refunds were guaranteed and so there was no irreparable harm, the government will struggle now to claim the court cannot order the refunds at all.
It is a strong card, not a winning one. The government can still say its promise ran only to the plaintiffs in front of it and others like them, not to every importer that never sued, that admitting the court has power is not the same as agreeing it can skip importer by importer orders, and that even if liquidation does not kill the refund right, reaching nonparties is still its own problem under CASA.
What importers should do now
For importers, including the US buyers that Korean exporters sell to, what happens next depends almost entirely on where each entry stands. Entries that are unliquidated or still inside the §1501 ninety day window are the safe ones, likely to come back through CAPE no matter what the appeal or the class motion does. Entries past that window probably fall outside CAPE Phase 1. Entries past the 180 day mark are final, and that is the group the government fights hardest. The finally liquidated entries of importers who never sued are the ones the June 4 class motion is trying to save, and if both the universal order and the class fall through, those importers may have nothing left but to file their own suits.
One move does not depend on either outcome. A protective filing within the limitations period is the only way to lock in standing on your own, and the §1581(i) clock runs to roughly February 2027. Alongside that, it is worth registering in ACE, setting up electronic refund receipt through ACH, and pulling together every entry from the IEEPA period, February 4, 2025 to February 24, 2026.
There is a risk running the other way as well. A wave of consumer side suits is hitting importers with unjust enrichment and pass through claims, so even a refund that arrives may be fought over. An importer that gets its money back can still face demands for price adjustments or contractual clawbacks, and Terry Precision Cycling, the company leading the class, is reported to be one of those that passed the cost along.
What to watch next
Korean exporters cannot claim these refunds themselves, but they have a stake in what their US buyers do next. A buyer that absorbed and passed on the duty may, once a refund lands, reopen pricing or ask for retroactive credits, so it is worth checking how supply contracts handle price adjustments and who owns any duty refund.
A few things will tell the story before the rest do. The first is whether the Federal Circuit stays the universal refund order, since a stay would make the class motion suddenly central, while a denial would let the CIT treat the class as unnecessary and sit on it. Watch as well for any move by the Federal Circuit or the Supreme Court on the mandamus and the Scott testimony before June 9, the nearest flashpoint. If the plaintiffs add or swap in a class representative that actually holds finally liquidated, CAPE ineligible entries, that is the clearest sign they are serious about typicality and adequacy. A concrete CBP schedule for a CAPE Phase 2 covering finally liquidated entries would cut against the case for a class. And the deepest tell is whether the Federal Circuit reaffirms the reliquidation power from Shinyei and AM/NS Calvert, or buys the government's narrower reading and its CASA argument.
Bottom line
The likeliest shape of all this is plain. Most of the money, the unliquidated and nonfinal entries, will come back through CAPE whatever the courts do. The finally liquidated entries of importers who never sued are the real question, and their fate rests more on the Federal Circuit's CASA ruling than on the class motion itself.
Caveats
This reflects what was on the record as of June 5, 2026. The exact text of the June 4 class motion, including how the class is defined and which part of Rule 23(b) it invokes, should be checked against the CIT docket before anyone relies on the precise wording. Shinyei and AM/NS Calvert establish that the CIT can order reliquidation, but they do not settle whether the finally liquidated entries of importers who never sued can be reached on a class wide basis after CASA.