IEEPA Class Win

A rare class certification at the Court of International Trade hands roughly 330,000 importers a theoretical route to tariff refunds. Trade counsel say the money still moves fastest for those who sue alone.

WASHINGTON, Oct. 20, 2026

A refund fight that has already returned more than $126 billion to American importers entered a new and more complicated phase this week, as trade lawyers worked through the consequences of a rare class certification at the U.S. Court of International Trade and the government declined, so far, to appeal it.

Judge Richard K. Eaton certified a nationwide class of importers on Oct. 8 in Freestyle World, Inc. v. United States, slip op. 26-123, court number 26-1088, according to summaries published by Miller & Chevalier, Davis Wright Tremaine and Steptoe & Johnson. The class covers every importer of record that paid duties imposed under what the court calls the Challenged Tariff Orders between Feb. 1, 2025 and Feb. 19, 2026. Companies that filed their own suits to recover those duties are excluded.

By Monday, Oct. 19, with the government still silent on appeal, the practitioner commentary had hardened into a single uncomfortable message for the importing community: certification is not a refund, and the class is not the fast lane.

“Certification does not issue refunds,” wrote Jennifer Diaz and David Craven of Diaz Trade Law in an analysis posted Oct. 19 that laid out the gap between the legal milestone and the cash. The firm estimated that the class route could take months, and possibly years, to produce payments.

That gap is the story. Roughly $166 billion in duties was collected under the International Emergency Economic Powers Act before the Supreme Court held that the statute does not authorize presidential tariffs, according to figures cited by the Penn Wharton Budget Model in its Oct. 14 update and by U.S. Customs and Border Protection in filings summarized by trade counsel. About $126 billion of that had been refunded with interest as of Oct. 2, roughly 76 percent of the total. What remains is not a rounding error. The court itself estimated that something on the order of $30 billion could stay in the Treasury absent further action, according to the Steptoe summary of the proceedings.

What the court actually did

The certification is procedurally unusual. Davis Wright Tremaine described it as only the second class ever certified in the history of the Court of International Trade, a tribunal whose docket is built almost entirely around individual entry-level disputes rather than aggregate litigation.

Eaton granted the motion, in his words, “in the hope, and expectation, that this certification will lead to the liquidation or reliquidation of all of the entries” on which the IEEPA duties were imposed. That phrasing matters. The relief the class seeks is not primarily a money judgment. It is an order directing CBP to treat the entries as though the unlawful duties had never been assessed, with refunds following as a mechanical consequence of reliquidation.

The court worked through the standard class-action requirements and found each satisfied. On numerosity, it noted that nearly 330,000 importers of record paid IEEPA duties, a figure drawn from CBP’s own accounting of roughly 53 million affected entries. On commonality, it identified shared questions that run to the heart of every member’s claim: whether the government lawfully imposed and collected the duties, and whether the affected entries must be liquidated or reliquidated without them. Typicality and adequacy of representation were also met.

Critically, Eaton certified under Rule 23(b)(2) of the court’s rules, the provision designed for cases where a defendant has acted on grounds that apply generally to the class, making classwide injunctive or declaratory relief appropriate. That choice carries a consequence the practitioner commentary has emphasized repeatedly: a 23(b)(2) class has no opt-in or opt-out mechanism. Members are in by operation of the definition.

The government opposed certification, arguing among other things that it would permit what it called one-way intervention, allowing potential members to wait and see how the merits came out before deciding whether to participate. The court acknowledged that Rule 23(b)(2) contains no opt-out provision but observed that importers retain a practical exit: they can file their own cases, which removes them from the class by definition.

The court also rejected the argument that wildly differing refund amounts across hundreds of thousands of importers defeat class treatment. Refund calculation, the court reasoned, is CBP’s job upon liquidation or reliquidation, not a question the court must resolve member by member.

How the case got here

The litigation traces to Feb. 20, 2026, when the Supreme Court held in Learning Resources, Inc. v. Trump that IEEPA does not confer tariff authority on the President. The Federal Circuit issued its mandate on March 2, and within two days the Court of International Trade began ordering CBP to liquidate or reliquidate affected entries without the IEEPA duties, according to the Steptoe account of the sequence.

In July 2026 those orders were expanded to reach finally liquidated entries and were entered in more than 4,000 importer lawsuits. The government has appealed the portion of the relief that extends to importers who are not plaintiffs, relying on Supreme Court precedent limiting universal remedies, including Trump v. CASA, Inc. That appeal, filed June 3, remains pending at the Federal Circuit, with oral argument not yet scheduled. Eaton’s universal injunction requiring refunds to all importers is stayed as to non-filers while the appeal runs.

Freestyle World itself filed on Feb. 19, 2026, one day before the Supreme Court ruled, and moved for class certification on June 23. A closed status conference was held Oct. 13.

The practical architecture of the refund program sits alongside the litigation. CBP built a portal, referred to in the trade bar as CAPE, to process claims in phases. Phase 1 covers unliquidated entries and entries liquidated within 80 days, and CBP has generally been issuing those refunds within 60 days of receiving a request, according to the Oct. 19 weekly trade roundup published by Foley & Lardner through the National Law Review. Phase 3, which opened Oct. 6, reaches finally liquidated entries, but only for importers holding court-ordered reliquidation, a population Davis Wright Tremaine put at roughly 7,000 importers across about 4,000 lawsuits.

That is the bottleneck. Non-plaintiffs with finally liquidated entries remain locked out of CAPE while the government’s appeal is pending. The class certification is the first judicial step toward opening that door, and it is a long way from the door opening.

The practitioner reaction

The response from the customs bar has been notably unenthusiastic about the class as a vehicle for recovery, and notably consistent across firms.

The most concrete objection is cost. Craven noted that class counsel typically receive 20 to 35 percent in fees, which he said is more than most companies would pay to bring their own case. His conclusion was blunt: only importers with little money at stake should stay in the class.

The Foley & Lardner roundup reached the same destination by a different route, recommending individual suits for importers with large refunds at stake and citing likely appeal delays, the time required to reach classwide relief, and potentially substantial class counsel fees.

Davis Wright Tremaine offered the operational numbers behind that advice. In the firm’s experience, individual CIT refund orders have typically issued within four business days of filing, with CAPE access following within one business day of the importer providing its information. Set against a class process that has not yet produced any classwide relief, and that the firm expects the government to appeal, the comparison is not close for a company with real money outstanding.

There is also a structural wrinkle. Because the class excludes anyone who has separately filed, the decision to sue individually is a decision to leave the class. Diaz Trade Law urged importers to treat that as a deliberate choice rather than a default, and to calendar their own filing deadlines rather than rely on the class to preserve them.

The clock

Limitations are the quiet pressure in this file. The two-year statute of limitations for IEEPA refund suits under 28 U.S.C. section 1581(i) begins to expire in February 2027, with Davis Wright Tremaine putting the earliest date at Feb. 4, 2027 and the Foley & Lardner roundup flagging February 2027 more generally.

For importers with modest claims, that still leaves room to watch the class docket and decide later. For importers with eight-figure exposure on finally liquidated entries, the arithmetic is less forgiving, because an appeal of the certification could easily consume the remaining window without producing a dollar.

Two further carve-outs deserve attention from compliance teams. According to Davis Wright Tremaine, the Brazil-specific and India-specific IEEPA tariffs, and duties collected through Feb. 23, 2026, fall outside the Challenged Tariff Orders definition that fixes the class. Importers in those categories may need to file individually, although a separate class certification motion in V.O.S. Selections, Inc. v. United States remains pending and could reach them.

Entries suspended under antidumping or countervailing duty orders are also in limbo. Liquidation and refunds there are expected to stay suspended until the Commerce Department issues liquidation instructions, and CBP has not yet published specific guidance, according to the Steptoe summary.

What the money looks like

The refund program is now large enough to register in national fiscal statistics, which gives the litigation an unusual macroeconomic dimension.

The Penn Wharton Budget Model, in its Oct. 14 update, put gross customs revenue from new tariffs at an estimated $314.8 billion from January 2025 through August 2026, before IEEPA refunds. Of that, roughly $166 billion was collected under IEEPA authority and is subject to refund. The model also estimated that gross revenue would have been $60.7 billion higher over the period had importers not changed their purchasing behavior in response to the duties, a reminder that the tariffs reshaped trade flows even where they were later held unlawful.

The underlying tariff wall has not come down. Penn Wharton put the average effective tariff rate at 7.1 percent in August 2026, up from 6.7 percent in July and from 2.3 percent in January 2025. On a counterfactual basis holding import patterns at historical trends, the rate would have been 10.1 percent in August. China faced 24.3 percent, the highest among major partners though below its 2025 peak. Steel and aluminum carried the highest product-category rate at 39.8 percent, with automotive vehicles at 13.8 percent.

The replacement architecture matters for anyone reading the refund story as a general retreat from tariffs. After the Supreme Court ruling in February, a 10 percent global tariff under Section 122 took over in March 2026. That authority expired July 24 and was replaced by Section 301 tariffs of 10 or 12.5 percent on imports from 60 economies, with product exemptions. Importers getting IEEPA money back are, in most cases, still paying.

Implications for importers and exporters

For import compliance teams, the near-term work is unglamorous and specific.

The first step is a complete entry inventory. Diaz Trade Law recommended pulling the ACE IEEPA entry report and sorting entries into three buckets: unliquidated, recently liquidated, and finally liquidated. The first two categories are addressable now through CAPE Phase 1 or Phase 2 declarations, and unclaimed eligible entries in those buckets are simply money left on the table.

The third bucket drives the strategic decision. Total the IEEPA duties on finally liquidated entries. If the number is significant, counsel across several firms recommend consulting about an individual CIT action before the limitations window closes. If it is minimal, the class docket is the cheaper place to wait. Diaz Trade Law offered the illustrative case of an importer with a few thousand dollars in duties on finally liquidated entries, for whom litigation costs would outweigh the benefit.

There is a parallel enforcement risk that compliance officers should not read past. The Foley & Lardner roundup described the Justice Department as aggressively expanding its use of the False Claims Act to pursue customs violations, raising whistleblower exposure, and CBP as operating under directives to intensify administrative enforcement, expand investigations and audits, and substantially restrict the mitigating factors that previously reduced penalties. The government’s increasing use of artificial intelligence and data analytics has improved its ability to flag discrepancies in import declarations.

The practical upshot is that the same entry review that identifies refund opportunities may also surface classification, valuation or origin errors. Post-summary corrections remain available before liquidation. After liquidation, a prior disclosure made before CBP identifies the error may allow resolution by paying unpaid duties and interest without penalties. The 180-day protest deadline runs from liquidation.

CBP has also had to work out how to handle IEEPA tariffs that were paid as part of a prior disclosure, one of a small number of glitches in an otherwise smooth Phase 1 process, according to the roundup. Importers in that posture should expect to engage directly with the agency.

For exporters and for foreign suppliers, the refund litigation is mostly a spectator sport, but the signal is worth reading. The Supreme Court closed one tariff channel in February. The administration moved to Section 122, then to Section 301, and has continued building sectoral programs under Section 232. The effective rate has drifted up, not down, across that transition. Suppliers planning 2027 pricing should not treat the IEEPA refunds as evidence of a softening.

What to watch

Three things will shape the next sixty days.

The first is whether the government appeals the certification. As of Oct. 19 it had not, though several firms expect it to, and International Trade Today reported on Oct. 9 that attorneys considered an appeal likely. An appeal would freeze the class route for an extended period and sharpen the case for individual filings.

The second is the Federal Circuit’s handling of the pending appeal from the original universal injunction. That case will determine whether relief can reach non-plaintiffs at all, and it is the question that ultimately governs whether the roughly $30 billion still held by Treasury moves.

The third is CBP’s rollout schedule. The agency has described a biweekly cadence for opening CAPE to additional groups, which the Foley & Lardner roundup suggested could mean a new group around Oct. 20. CBP has said more instructions are coming but has not given a date. Because many law firms did not receive Phase 3 filing instructions until August, after the July 31 deadline for submitting importer-of-record spreadsheets, some otherwise eligible importers still cannot participate.

For now, the class certification stands as a procedural landmark in a court that almost never sees one, and as a reminder that in customs practice the remedy and the refund are rarely the same thing.

A court not built for aggregate litigation

Part of what makes the certification notable is the forum. The Court of International Trade hears disputes that are, by statutory design, entry-specific. A protest is filed against a liquidation. A summons covers particular entries. Remedies attach to those entries. The machinery assumes a plaintiff who can identify precisely what it paid and when.

That architecture served reasonably well for decades of classification and valuation disputes involving a manageable number of importers. It was not designed for a situation in which a single legal question, the validity of a tariff authority, governs the liability of 330,000 companies across 53 million entries simultaneously.

The government’s one-way intervention objection reflects that tension. In an ordinary CIT case, every plaintiff bears the cost and risk of filing. A class lets companies that bore neither capture the benefit. The court’s answer, that importers can effectively opt out by filing their own cases, is procedurally sound but does not fully dissolve the asymmetry, which is why the question is a plausible candidate for appellate attention.

There is a counterargument that the trade bar has been slower to make publicly. If the Federal Circuit ultimately holds that relief cannot reach non-plaintiffs, and no class mechanism is available either, the result is that roughly $30 billion in duties collected under an authority the Supreme Court held did not exist stays with the government because the affected companies did not each file a lawsuit about it. Whatever the doctrinal merits, that outcome is difficult to defend as a matter of policy, and it is the pressure that gives the class certification its significance beyond procedure.

The appeal question

The government’s posture on appeal is the single most consequential unknown.

It has already appealed the portion of Eaton’s relief extending to non-plaintiffs, a filing dated June 3 that remains pending without a scheduled argument. If the Federal Circuit affirms, the universal injunction revives and the class question becomes largely academic, because non-plaintiffs would have a direct path. If it reverses, the class becomes the principal remaining vehicle for the roughly 323,000 importers outside the current litigation, and whatever the appellate court said about classwide relief in the first appeal will shape how the second one is argued.

A separate appeal of the certification itself would add a third track. Davis Wright Tremaine noted that the court has not yet issued classwide relief and is unlikely to do so before the government answers the complaint, which means there may be an extended period in which the class formally exists while producing nothing.

Importers reading this as a sequencing problem are reading it correctly. The question is not whether the class is legally sound. It is whether the class produces money before February 2027, and the honest answer available from the public record is that nobody knows.

Sector exposure

The refund population is not evenly distributed across the economy, and neither is the residual exposure.

Importers of consumer goods, apparel, footwear, furniture and housewares carried the heaviest IEEPA burden by entry count, because the orders applied broadly across consumer categories rather than targeting industrial inputs. Many of those importers are mid-sized companies with large entry volumes and relatively thin margins, precisely the profile for which a 20 to 35 percent contingency fee on a recovered refund is a material haircut and a self-funded CIT filing is a real budget decision.

At the other end, large industrial importers with dedicated customs counsel were disproportionately represented among the roughly 7,000 companies that filed suit and now hold court-ordered reliquidation. They are inside Phase 3. The refund program has, in effect, rewarded legal sophistication, and the class certification is the mechanism by which that advantage might partially close.

Companies with entries suspended under antidumping or countervailing duty orders occupy a third position, waiting on Commerce liquidation instructions that have not issued. For them neither track moves, and the limitations clock runs regardless.