USTR Proposes Section 301 Tariffs at 10%-12.5% to Replace Expiring Section 122, Targeting 60 Trading Partners

The Analysis

On 2 June the U.S. Trade Representative proposed a new Section 301 schedule to take over from the 10% Section 122 global tariff that expires around 24 July. The structure is split in two: 10% for fifteen trading partners that have made bilateral structural concessions, 12.5% for forty-five others, Hong Kong among them. Comments close on 6 July and hearings begin on 7 July. A separate refresh of the Section 232 aluminium, steel and copper derivatives schedule keeps a 50% baseline and offers preferential treatment only where 85% of the metal is melted and poured in the United States.

Strictly speaking, this is a proposal, not a determination. Nothing in it is collectable on the day it was published, and the rate that will apply to your goods on 25 July is still an open question rather than a figure you can drop into a landed-cost sheet. The reporting has already moved on to which countries sit in which bucket, and that is precisely the step worth slowing down on.

Begin with the legal architecture, because the switch from Section 122 to Section 301 is not a like-for-like substitution of one power for another. Section 122 is an emergency balance-of-payments authority, and it carries a statutory time limit. That limit is the only reason there is an expiry date to argue about. Section 301 runs on a trade-remedy track: it needs an investigation, a record, a determination and a remedy. That is why this proposal is anchored in forced-labour and industrial-subsidy investigations rather than in emergency economic powers. The design trade is deliberate. A remedy built on an administrative record is slower to impose and considerably harder to strike down, because a court reviewing it asks whether the agency followed its own statute, not whether an emergency genuinely existed. Read the choice of authority as a signal about where the next legal fight will be, and it will not be about the rate level.

The bifurcation is the part that gets misread. The 10% and 12.5% tiers are keyed to whether a partner has made bilateral structural concessions. That is a negotiation status. It is not a product classification, not a tariff line, and not an origin test. Two shipments of identical goods, same line, same origin, can sit in different tiers if the partner's negotiating posture shifts, and the tier can move while your goods stay exactly as they were. This matters because it removes a lever you might assume you have. Supply chain design cannot move you into the 10% tier. There is no routing decision, no bill-of-material change and no supplier swap that touches a criterion written in terms of what a government conceded across a table. What you can do is narrower and less comfortable: know which tier your origin currently sits in, and make sure the contract says who absorbs a move between tiers.

Hong Kong appears by name in the group of forty-five. Importers who have grown used to treating Hong Kong-origin goods as a bucket separate from mainland China will read that as a collapse of the distinction, and at the level of the headline rate it is one. Here is the exception, though. The list names economies; the rate attaches to goods according to the origin rules in the operative text, and that text does not exist yet. Goods produced in Hong Kong and goods that merely transit or are invoiced through Hong Kong are not the same case, and nothing in a proposal-level list settles which is which. Do not let a named economy in a press summary harden into a classification decision in your system before the definitions are published.

The Section 232 refresh deserves separate attention because its test is easy to answer with the wrong document. The 50% baseline stays, and the preference is reserved for derivatives where 85% of the metal is melted and poured in the United States. That is an upstream test. A certificate of origin for the finished article does not answer it, because the question is where the metal was melted and poured, which happens several steps before the article acquires its own origin. Two rules, two evidence sets, and quite possibly two different countries inside one file. The practical failure mode here is organisational rather than legal: the 85% question gets handed to whoever handles certificates of origin, and the answer comes back clean, fast and wrong.

Now the calendar, which I think is the most under-reported element in the whole package. Comments close on 6 July. Hearings begin on 7 July. The Section 122 rate expires around 24 July. That leaves seventeen days between the first day of hearings and the expiry. Rulemaking at this scale does not normally complete in seventeen days, and the proposal itself contemplates a comment record that has to be read and answered. The base case is therefore not a clean handover from 122 to 301 on schedule. It is either a short extension of the existing rate or a window in which the applicable rate is not settled at the moment goods are entered. Neither outcome is dramatic, and both break a landed-cost model built on the assumption that 301 arrives on 25 July at a known number. The variable worth stress-testing is not 10% against 12.5%. It is what your costing does during a period when the rate is provisional.

Put a number on the tier question, with my own assumptions flagged as mine. Take an importer bringing in USD 20 million a year from origins that land in the 12.5% tier, currently paying 10%. The gap is 2.5 percentage points, so USD 500,000 a year, about USD 42,000 a month. Those are hypothetical figures I have set for illustration; the proposal contains no importer-level values. The point of the arithmetic is the shape rather than the size. Two and a half points is small enough to disappear into a cost line and large enough to remove the margin on a low-margin category, and nobody has to touch your supply chain for it to happen. A tier reassignment does the work on its own.

Between now and 6 July the work is documentary, not strategic. Pull the exemption list, map your lines against it, and for each line record two things: the tier it currently sits in, and the evidence supporting the origin claim behind that tier. Where the evidence is a supplier letter rather than a production record, mark it weak, because a criterion written in terms of negotiation status is exactly the kind that gets rewritten in the final text. The same discipline applies to the 85% test. Identify which of your aluminium, steel and copper derivatives could meet it and whether the mill documentation exists today, not whether it could be obtained if asked.

Where this conclusion stops holding is worth stating plainly. If the final text keys the tiers to something other than negotiation status, for instance to a published product list, then the tier becomes a classification problem, most of my argument about unpredictability falls away, and the right preparation shifts from contract terms to line-by-line mapping. Until that text is published, treat every tier assignment as provisional and every rate in your model as a placeholder.

↑ Back to the key points

— By Dr. Ingrid Voss