The rebuilt wall does two things the IEEPA wall did not. If finalized, it would convert a duty that behaved like a refundable contingent liability into a durable cost input that has to be priced in, and it would redistribute that cost unevenly across origin and tariff line, so a stable average sits on top of a wider spread. For any single importer, eleven percent is the wrong number to plan against.
What USTR put on the table on June 2
The USTR notice supplies the operative structure. USTR initiated the investigations in March under Section 301 of the Trade Act of 1974, made determinations under Section 301(b) that the foreign practices are unreasonable and burden United States commerce, and on June 2 proposed action. The proposal is a two-tier duty. Fourteen economies, with the European Union counted as one, face 10 percent. The remaining forty-six, including China, India, Japan, South Korea, Brazil, and Vietnam, face 12.5 percent.
The 10 percent tier is best understood as a policy-discount group rather than a set of better-behaved enforcers. It collects three kinds of economy, those that already operate a forced-labor import prohibition, those that committed to impose and enforce one under a reciprocal trade agreement, and the United Kingdom with its partial regime. Fourteen economies qualify for the discount, among them Canada, Mexico, the European Union, Indonesia, Pakistan, and Taiwan. The 12.5 percent tier covers the remaining forty-six economies. The distinction is remedial, not exculpatory. The 10 percent tier receives a lower rate, but USTR still treats the underlying failure as actionable.
The single most important fact for a practitioner is the one that is easiest to lose. The duty is announced, not in force. The notice sets no effective date. Written comments are due July 6 and the public hearing is July 7. In the language that matters for any exposure model, the announced rate is 10 percent or 12.5 percent, the effective rate today is zero, and the effective date is undetermined. Anyone who books a duty accrual against these rates now is modeling a proposal.
Why the IEEPA refund logic does not carry over
Under IEEPA, an importer could treat the tariff as a bet that might be reversed, and that is exactly what happened. The Court held six to three that IEEPA does not authorize the President to impose tariffs, reasoning that the statute never mentions duties and that the tariff power is a core congressional power that a vague delegation cannot reach. CBP has been processing refunds through the CAPE tool since April, interest is accruing on eligible entries, and importers who carried those duties as a contingent liability rather than a sunk cost are being proven right. The lesson many teams are drawing from that experience is the wrong one to carry forward.
The new layer rests on authorities that are harder to attack on the same theory that defeated IEEPA. Sections 232 and 301 refer explicitly to duties and import restrictions, and both require a procedural record, a Commerce investigation and national security finding for 232 and a USTR investigation, comment, hearing, and determination for 301. The Court contrasted IEEPA unfavorably with the discretion-conferring language of Section 232 and pointed to Section 301 as an example of Congress delegating tariff power in explicit terms. That does not make the new duties immune. It changes the litigation question from whether the statute authorizes tariffs at all to whether the administrative record supports this particular scope and remedy. The scope challenge for this 301, whether a foreign government failing to police forced labor is an unreasonable practice that justifies a near-global duty, is the obvious target if USTR finalizes the action, and challengers are likely to argue that public revenue-replacement statements show a predetermined remedial objective. That argument has to be tested against the record rather than against the text of the statute, which is why the structural odds are better than they were for IEEPA.
For a cost model, that shift is the whole story. The duty that comes with the rebuilt wall should be carried as a durable landed-cost input, not as a refund waiting to happen. The refunds now flowing for IEEPA are the close of that chapter, not a preview of the next one.
An exclusion order and a tariff are different instruments
The most useful operational point for compliance teams is that the United States now runs two forced-labor regimes that work in opposite ways, and they should not be confused. Section 307 of the Tariff Act of 1930 and the Uyghur Forced Labor Prevention Act are admissibility tools that block goods. CBP detains, excludes, or seizes merchandise tied to forced labor, and under the UFLPA an importer of goods linked to Xinjiang or to a listed entity faces a rebuttable presumption that can only be overcome with clear and convincing evidence. The new Section 301 is a duty tool. It does not adjudicate whether a given shipment was made with forced labor. It taxes the goods of a country whose government USTR has found is not maintaining or enforcing its own prohibition. One regime governs whether the goods can enter. The other governs how much they cost to enter.
Trading partners may also attack the asymmetry of the United States position, including its non-ratification of ILO Convention No. 29 against a backdrop of domestic prison-labor arrangements that involve private sector benefit. For an importer, that argument does not change the duty exposure or the immediate modeling task.
Why eleven percent is the wrong number to plan against
The average looks total and is not, in both directions. USTR notes that the sixty economies account for roughly 99.4 percent of United States import origins, and that number invites the wrong conclusion. The duty does not reach nearly all imports, because Annex A removes a large share of trade value. Goods already subject to Section 232, USMCA-compliant goods of Canada and Mexico, qualifying CAFTA-DR textiles, civil aircraft, energy products, certain critical minerals, and pharmaceuticals are carved out. Global Trade Alert reads the practical effect as roughly three in five covered dollars excluded, leaving about 1.17 trillion dollars of the 3.12 trillion supplied by the sixty economies actually exposed. The instrument is broad on paper and narrow in practice, which is why it barely moves the average.
That stability is what makes the average misleading at the level where importers operate. The IEEPA wall was a broad and fairly uniform layer. The rebuilt wall is a set of narrower layers whose weight depends on origin and tariff line. Subject to exclusions and the final implementation language, the default assumption should be stacking. The forced-labor 301 excludes goods already hit by Section 232, but it would layer on top of most-favored-nation base rates and on top of the existing China Section 301 duties from 2018. A Chinese-origin good outside the 232 carveouts could carry its base rate, its existing China 301, and the new 12.5 percent at once, while a USMCA-compliant Mexican good in the right tariff line might see nothing from this action. Both importers live inside the same eleven percent average and experience completely different cost changes.
Real exposure is therefore a function of three variables rather than the headline. The first is country of origin and tier, whether a good comes from a 10 percent or a 12.5 percent economy. The second is the exact tariff classification measured against Annex A, which is mapped by HTS number and not by product description, so a verification done on product names will miss. The third is the stack, the full set of duties that assemble on a given origin and line. The only way to know the rate that applies to a specific importer is to compute that specific cell.
What practitioners should do before the deadlines
The work is the same regardless of where the rate lands. Map exposure now by origin crossed with HTS using 2025 and 2026 entry data, model the 10 percent and 12.5 percent layers as an added duty rather than a substitute, and test the result against margin, pass-through, and customs bond sufficiency. Verify Annex A, Section 232, and free-trade agreement treatment by tariff number rather than description. If the exposure is material, the procedural clock has two settings. A request to appear at the hearing, with a summary of testimony, is due June 22, and written comments are due July 6, both ahead of the July 7 hearing where the scope and the exclusions get contested. Keep refund strategy separate from all of this and bounded to IEEPA. CBP has been processing IEEPA refunds through CAPE since April, interest is accruing on eligible entries, and finally-liquidated entries need to be preserved through protest or litigation rather than CAPE. Section 122 refunds are a different matter and are not available yet, because that litigation is not final and the appeals court stay keeps the surcharge being collected from everyone in the meantime. None of this is a template for the duties now being built, which are designed to be more durable.
Bottom line
The replacement wall is real and the average barely moved. Neither fact tells an importer what changed for them. Two things did. The duty that comes with this wall is built to last, which puts it in the model as a durable cost rather than a contingent line in a reserve, and the refunds now flowing for IEEPA are the end of that mechanism, not a preview of the next. The steady average also sits on a wider spread, so the number that matters is not eleven percent but the rate on a specific origin and a specific tariff line once the stack is assembled. The signal to watch is whether a final determination and an effective date arrive near the July 24 expiry of the Section 122 bridge, the moment the replacement is built to cover. If they do, the wall will not fully return to the IEEPA peak, but it will hold near the current 11 percent modeled level on a more durable legal footing.
Caveats
The duty is not in force and has no effective date, so every rate in this analysis is an announced figure. The Section 122 surcharge is verified in force at 10 percent. A 15 percent figure has been announced and is modeled by analysts but should not be treated as a confirmed in-force rate. The Brazil Section 301 action is at a proposed or determined stage and its effective status should be tracked independently. The roughly 11 percent average is a trade-weighted statutory estimate from Global Trade Alert, not the lower effective rate of duties actually collected, and the various published averages differ because they measure different things. The litigation around Section 122, the IEEPA refunds, and any challenge to the new 301 is moving quickly, and all of the above is current as of June 9 2026.
This is a sensitive and contested area of policy, and reasonable analysts read the legal odds differently. The argument here is that the new authorities are more durable than the ones they replace, not that they are beyond challenge, and that durability is what changes the cost treatment for an importer.