New data show the average US effective tariff rate fell to 7.1 percent as Customs certifies roughly 100 billion dollars in refunds from the invalidated IEEPA tariffs, the largest duty repayment operation in American history
WASHINGTON, August 11, 2026
Six months after the Supreme Court struck down the emergency tariffs that defined the opening year of the administration’s trade agenda, the fiscal ledger of that experiment is coming into focus. Data published Monday by the Penn Wharton Budget Model, drawing on newly released United States International Trade Commission trade statistics, show that the average effective tariff rate on American imports stood at 7.1 percent in June 2026, down sharply from its 2025 peaks, as United States Customs and Border Protection works through the largest tariff refund operation ever undertaken.
According to court filings cited in the Penn Wharton analysis, CBP had certified approximately 100 billion dollars in refunds through July, roughly 60 percent of the estimated 166 billion dollars collected under the International Emergency Economic Powers Act tariffs before the Court’s February 20, 2026 decision declared them unconstitutional. The repayment machinery, an entirely new processing system that CBP built in months and has rolled out in phases through the summer, is now expanding to reach categories of entries that the first waves missed.
The numbers tell the story of a trade policy that generated enormous gross revenue, then surrendered a large share of it to the courts, while leaving in place a tariff structure that remains, even after the unwinding, roughly triple its level of January 2025.
From 2.3 Percent to 7.1 Percent, the Long Way Around
When the administration took office in January 2025, the average effective tariff rate, customs duties collected as a share of import value, was 2.3 percent. The rate climbed through 2025 as the IEEPA based country tariffs, the reciprocal tariff program, and expanded Section 232 sectoral tariffs took hold, before falling back in 2026. The decline in recent months, Penn Wharton notes, follows directly from the Supreme Court’s February ruling and the administration’s decision to replace the invalidated IEEPA tariffs with a 10 percent global tariff imposed under Section 122 of the Trade Act of 1974, a balance of payments authority that permits temporary import surcharges without emergency declarations.
The composition of the remaining tariff wall is revealing. Among major trading partners, China faces the highest effective rate at 23.2 percent as of June, a marked decline from previous months but still far above any other large partner. By product category, steel and aluminum lead at 40.9 percent, reflecting Section 232 tariffs that were raised from 25 percent to 50 percent in June 2025, followed by automotive vehicles at 13.2 percent. And in North America, the share of imports from Canada and Mexico claiming duty free treatment under the USMCA reached 83.6 percent in June, a surge from historical norms that Penn Wharton attributes to importers aggressively leveraging the agreement’s rules of origin to escape tariff exposure.
Penn Wharton estimates that the new tariffs raised 283.9 billion dollars in gross customs revenue between January 2025 and June 2026, before accounting for IEEPA refunds. Had importers not accelerated purchases and reshuffled sourcing patterns in response to the tariffs, collections would have been 58.7 billion dollars higher, a measure of how much behavioral adaptation eroded the tax base even before the courts intervened. Netting out the IEEPA collections that are being returned, the analysis concludes that roughly 117.9 billion dollars in net customs revenue remains from the entire eighteen month tariff campaign, on the assumption that all IEEPA collections are ultimately refunded.
The Machinery of Giving Money Back
The refund operation itself has become one of the year’s defining administrative stories. The Court of International Trade, implementing the Supreme Court’s ruling, ordered CBP to return the invalidated collections, and the agency responded by building a dedicated processing system, the Consolidated Administration and Processing of Entries, known as CAPE, according to reporting by Sourcing Journal and analyses published by trade law firms including Thompson Hine and Mitchell Silberberg and Knupp.
CAPE launched its first phase on April 20, processing refunds for the most straightforward category: fully liquidated entries with no pending adjustments. Phase 2 launched on June 29, extending the system to entries flagged for reconciliation, cases where declared value, classification, or origin remained subject to later adjustment, covering entry types that CBP estimates account for roughly 28.7 billion dollars in potential refunds. A third phase, targeted for the end of July, addresses the residual categories, including entries involved in protests and litigation. Industry publications reported in recent days that the Phase 2 expansion has begun reaching reconciliation flagged entries that earlier processing excluded, prompting importers to re-examine their entry records for newly eligible refunds.
The scale is difficult to overstate. CBP estimates that importers paid the 166 billion dollars in IEEPA duties across more than 53 million individual entries, and that about 330,000 importers are eligible for repayment. Entries representing more than 95 billion dollars in potential and certified refunds had been accepted for processing by early summer, according to the Cato Institute’s tracking of the litigation record. The refund queue includes the largest names in American retail and logistics: Costco, Walmart, and Target are among the petitioners, as are FedEx, UPS, and DHL, companies that paid duties as importers of record on behalf of thousands of clients.
The court filings that anchor the public record of progress come from the Court of International Trade docket, where CBP officials have submitted sworn declarations updating refund certification totals, including the March declaration in Atmus Filtration v. United States that fixed the 166 billion dollar collection estimate and an August 4 declaration in Freestyle World v. United States that reported the approximately 100 billion dollars certified through July.
The Legal Road to February 20
The refund operation is the final chapter of a legal saga that began within weeks of the first IEEPA tariff proclamations in early 2025. The International Emergency Economic Powers Act, a 1977 statute written to authorize sanctions and asset freezes during declared national emergencies, had never before been used to impose tariffs. The administration invoked it to build the centerpiece of its trade program precisely because it offered speed: no investigations, no comment periods, no product lists negotiated through agency process, just proclamations with immediate effect.
Small importers, trade associations, and eventually some of the largest companies in the country challenged the program on the ground that IEEPA’s grant of authority to regulate importation does not include the power to tax it, and that reading the statute otherwise would transfer the tariff power that the Constitution assigns to Congress into an open ended executive instrument. The litigation moved through the Court of International Trade and the Federal Circuit through 2025 against a backdrop of extraordinary commercial uncertainty, with importers paying duties into a system that might or might not ever return them.
The Supreme Court’s February 20, 2026 decision resolved the question decisively against the government, holding that the emergency powers statute could not carry the weight of a comprehensive tariff regime. The ruling left untouched the administration’s other authorities, and the White House pivoted within days, invoking Section 122’s balance of payments authority for a 10 percent global tariff and accelerating Section 232 and Section 301 actions to rebuild sectoral and country specific coverage. For court watchers, the episode became the highest stakes application yet of the major questions doctrine to economic policy. For importers, it became a refund claim of historic size.
What Section 122 Can and Cannot Do
The replacement architecture has its own clock ticking. Section 122 of the Trade Act of 1974 authorizes temporary import surcharges of up to 15 percent to address large and serious balance of payments deficits, but the statute limits such measures to 150 days unless Congress extends them. The administration’s lawyers have advanced creative arguments about successive proclamations and changed circumstances, and Congress has so far declined either to ratify or to terminate the surcharge, leaving the 10 percent global tariff in a legal gray zone that traders have simply priced as indefinite.
The practical consequence is that the current tariff baseline is provisional in a way the market has largely stopped noticing. If the Section 122 surcharge lapses or falls to a court challenge, the effective rate would drop again; if Congress codifies it, the 10 percent floor becomes durable law. Fiscal analysts note that the revenue stakes of that question now exceed 100 billion dollars annually, which guarantees that the fight over codification will be a central feature of the next budget cycle. Penn Wharton’s tariff simulator, the companion tool to Monday’s data release, models scenarios in both directions, and the gap between them illustrates how much of American trade policy still rests on contested legal ground eighteen months into the experiment.
How Trading Partners Read the Unwinding
Foreign capitals have processed the IEEPA episode with a mixture of relief and wariness. The February ruling removed the most legally vulnerable layer of American tariffs, and several governments that had prepared retaliation lists quietly shelved them as the refunds began to flow. But the speed with which Washington rebuilt equivalent coverage under other statutes taught trading partners that the American tariff wall does not depend on any single legal foundation, and negotiating strategies have adjusted accordingly.
The bilateral deals of early 2026 reflect that recalibration. India’s February agreement, which cut its country rate from 50 percent to 18 percent in exchange for commitments on Russian oil, market access, and procurement, was concluded in the immediate aftermath of the Supreme Court ruling, when both sides had reason to lock in certainty. The China truce, extended last November through November 10, 2026, has survived the legal turbulence because its core suspensions were implemented through Section 301 modifications rather than IEEPA. And the USMCA partners, whose exemption rates climbed as compliance surged, have concentrated their diplomacy on the agreement’s scheduled joint review, where the durability of North American tariff insulation will be tested against Washington’s appetite for renegotiation.
Winners, Losers, and the Interest Question
For companies that paid IEEPA duties, the refunds are a material financial event. Retailers whose margins absorbed the tariffs through 2025 are now booking recoveries that, for the largest importers, run to hundreds of millions of dollars. Chief financial officers have treated the timing of CAPE processing as a genuine earnings variable, and the phased rollout has created a hierarchy of fortune: importers with clean, liquidated entries were paid first, while those with reconciliation flags, protests, or complex entry histories wait for later phases.
The unresolved questions are interest and completeness. Refund claimants have pressed for interest on the returned duties, a matter still being litigated, and trade counsel have warned importers that CAPE’s automated matching will not capture every eligible entry, particularly where entry data was amended after filing. The class action bar has also arrived: multiple suits seek to represent smaller importers who lack the resources to navigate the refund process, arguing that CBP’s system places the burden of recovery on the very parties the courts held were wrongfully charged.
There is also a fiscal dimension. Gross customs receipts became a meaningful revenue line in 2025, and the refund outflows are now visible in daily Treasury statements as negative customs revenue in some reporting periods. The Penn Wharton analysis implies that the durable revenue yield of the administration’s tariff program, after the courts’ subtraction, comes from the surviving instruments: the Section 122 global tariff, the Section 232 sectoral tariffs, the Section 301 actions including July’s forced labor tariffs on 60 economies, and the accumulating trade remedy orders. Those authorities rest on firmer statutory ground than IEEPA did, which is precisely why the administration migrated its tariff program onto them within days of the February ruling.
The Behavioral Economy the Tariffs Built
Beyond the headline numbers, the Penn Wharton data document how thoroughly eighteen months of tariff volatility have rewired importer behavior. The 58.7 billion dollar gap between mechanical and actual collections represents millions of individual decisions: purchases accelerated ahead of announced effective dates, orders rerouted through exempt channels, product specifications adjusted to qualify for carve outs, and, above all, the USMCA compliance surge.
The North American story is the clearest illustration. USMCA preference claims were stable through late 2024 because, for many low tariff products, the paperwork cost of certifying origin exceeded the duty savings. The 2025 tariff rounds inverted that calculus overnight. With country tariffs on Canadian and Mexican goods reaching punishing levels for non compliant trade, and with USMCA compliant goods exempt from successive actions including July’s forced labor tariffs, the share of North American imports claiming preferences climbed to 83.6 percent by June 2026. Companies spent heavily on origin documentation, supplier attestations, and customs software, a compliance investment that is now effectively locked in as the price of tariff insulation.
The same logic explains the durability of the China number. The 23.2 percent effective rate on Chinese goods, even after the IEEPA unwinding and the truce framework that has held since late 2025, reflects the accumulated Section 301 tariffs and the remaining sectoral measures. The bilateral truce, extended last November through November 10, 2026, suspended the escalatory spiral but left the baseline structure intact, and importers have continued the diversification away from Chinese sourcing that six years of tariffs began.
The Small Importer Problem
Beneath the headline numbers sits a distributional question that the refund process has sharpened rather than settled. The 330,000 eligible importers span an enormous range, from multinationals with dedicated customs departments to small businesses that imported a few containers under IEEPA rates and may not know they are owed anything. CBP’s system requires no affirmative application for many entry categories, but the automated matching depends on entry records being clean, complete, and correctly associated with the importer of record, conditions that hold far more reliably for large filers than for occasional ones.
Trade groups representing small importers have pressed CBP to conduct affirmative outreach, and the class action filings now consolidated at the Court of International Trade argue that a meaningful share of eligible refunds will simply never be claimed without it. Customs brokers, who often served as importer of record for small clients, face their own reconciliation burden in passing recovered duties through to the businesses that ultimately bore them. The unclaimed remainder, whatever its final size, will sit as a quiet subsidy from the least sophisticated importers to the Treasury, an ironic coda to a program the courts held should never have collected the money at all.
Economists studying the episode also note its research value: the IEEPA tariffs’ imposition and abrupt removal created a natural experiment in tariff incidence, and early academic work using the 2025 to 2026 data has generally found that the burden fell predominantly on American importers and consumers rather than foreign exporters, consistent with the findings from the 2018 to 2019 tariff waves. Those results are already being cited in the policy debate over codifying the Section 122 surcharge, by both sides, which is perhaps the surest sign that the American tariff argument has become permanent.
Implications: A Tax That Refunds Itself Is Still a Tax
For American businesses, the lesson of the refund summer is double edged. The courts proved willing to strike down the most aggressive assertion of executive tariff power and to order the money returned, a genuine vindication of the legal challenges that many trade associations funded. But the aftermath demonstrates the resilience of the tariff state: the effective rate settled at triple its pre-2025 level, the administration rebuilt its program on sturdier authorities within weeks, and the refund process itself has consumed a year of corporate attention and legal spending.
Importers should focus on three practical fronts. First, recovery: companies that paid IEEPA duties should audit their entry records against CAPE’s expanding phase coverage, including the newly eligible reconciliation flagged entries, and should not assume the automated system will find every dollar. Second, structure: the surviving tariff architecture rewards USMCA qualification, Section 232 exemption awareness, and exclusion advocacy in the pending Section 301 proceedings, and the companies that treat tariff engineering as a permanent discipline rather than a crisis response are visibly outperforming those that do not. Third, vigilance: Section 122’s global tariff carries statutory time limits that will eventually force either congressional action or another legal transition, meaning the current structure is a way station rather than a destination.
Exporters and foreign suppliers should read the data with equal care. The USMCA compliance surge shows that preferences, origin rules, and exemption architecture now matter more than headline rates, and suppliers that can deliver qualifying goods, complete documentation, and credible supply chain traceability command a premium in American sourcing decisions that did not exist three years ago. The competition for American shelf space and industrial contracts increasingly runs through customs compliance departments as much as through price lists.
For the trade bar, the accountants, the software vendors, and the consultants, the refund summer has confirmed what the tariff years suggested: complexity itself has become a permanent feature of the American import system, and the professional infrastructure built to manage it will not be dismantled even if rates someday fall. Penn Wharton’s monthly updates, once a niche academic product, now move planning assumptions at Fortune 500 companies, and Monday’s release circulated through corporate trade departments within hours of publication.
The Supreme Court’s February ruling answered the constitutional question. Monday’s data answer the economic one: the tariff experiment was curtailed, not reversed, and the American import economy of August 2026 operates at a tariff level no one would have predicted three years ago, refunds and all. Between the certified repayments flowing out of the Treasury and the new duties flowing in at the ports, the United States has managed the unusual feat of running history’s largest tariff refund and one of its largest tariff programs at the same time, and the customs data now show both at full scale.