Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- By 17 July 2026, map every exported tariff line against the investigation scope and mark each one in, out or undetermined, with a target of zero undetermined lines in the top twenty by value. Owner: trade compliance.
- By 24 July 2026, build one traceability file per exported product family covering origin of inputs, production site and labour records, ahead of any determination simply because the file takes longer to assemble than a comment does.
- By 31 July 2026, register the docket and the publication date of the determination, with a named person checking it weekly, so the effective date is known before the first entry rather than after it.
- Before the effective date, if one is published, confirm for each open shipment whether the entry will be filed before or after it, and treat anything on the water as already exposed.
- Within 30 days of any determination, confirm whether an exclusion window exists and what record it requires, because the window is dated and the record cannot be built inside it.
The U.S. Trade Representative opened a three-day hearing on 7 July on proposed Section 301 tariffs of 10%-12.5% on 60 economies over alleged failure to ban forced-labor goods. In written comments filed 6 July, India urged the USTR to drop a proposed 12.5% duty on Indian exports, arguing the probe lacks evidence. Mexico, Peru, Guatemala and Ecuador also objected at the hearing.
The Analysis
The United States Trade Representative opened a three-day hearing on 7 July into proposed Section 301 tariffs of 10% to 12.5% on 60 economies, on the theory that those governments have failed to ban goods made with forced labour. India filed written comments on 6 July asking that the proposed 12.5% duty on Indian exports be withdrawn, on the ground that the investigation lacks evidence. Mexico, Peru, Guatemala and Ecuador objected at the hearing itself.
Strictly speaking, a hearing is a procedural step whose only legal product is a record. It sets no rate, fixes no date and binds no entry. The percentages in the notice describe a proposed range across an investigation scope, and a range of that kind is not yet a rate applicable to any particular shipment.
Ten to 12.5% describes a range across sixty economies, not a duty on a product. A measure of this kind is applied by tariff line, and it takes effect through a published act with an annex; until that annex exists, the operative rate for any given consignment is undefined rather than 12.5%. An importer who models 12.5% on total purchases from a named economy has modelled the outer edge of a proposal. That is a defensible planning assumption and a poor basis for a declaration.
What is being found is narrower than the headline suggests. The allegation concerns a government failing to prohibit forced-labour goods, which is a finding about conduct inside a jurisdiction. It is not a finding about a factory, a supplier or a consignment. The remedy that follows is chosen by the trade representative, and a tariff is one option among several, so the chain from investigation to duty on your goods passes through a determination and an annex. Skipping those two steps is how a probe becomes a certainty in an internal forecast.
Scope works the same way. Sixty economies is the reach of the investigation, and the reach of an investigation is not a list of dutiable goods. Product coverage is a separate document. Being a named economy is not the same as being a covered origin for every line you ship. Where the annex turns out narrow, the headline number applies to a fraction of trade. Where it is broad, the exclusion process becomes the operative event, and that process is short, dated and easy to miss.
Timing is where it starts to cost money. Duties of this kind apply to entries made on or after the effective date of the measure. The boundary is therefore the entry date, not the sailing date and not the contract date. Goods on the water when the determination is published will be entered after it, and they will pay. That is the group with the least ability to react, and it is the group that a sourcing decision taken in July cannot protect.
The objections have a legal function that is easy to misread. India's comments and the four governments that appeared do not bind the determination; they build the record. Their practical importance is durability. A measure resting on a contested evidentiary record is more likely to be challenged, and more likely to be revised or remanded, than one built on findings nobody disputed. That affects how long the measure lasts. It does not affect whether it applies next quarter.
Impact splits by position rather than by size. Exporters in a named economy with one product and one market face the whole of the number if the annex covers them, and cannot absorb 12.5% out of a commodity margin, so the buyer will ask them to split it. Exporters who can document production to a standard that answers the forced-labour question hold the only defence that works both at the border and in a customer audit, and it costs money before it is needed. United States importers face the entry date and the exclusion calendar. The largest will have counsel watching the docket. The smallest will learn about it from an invoice, which is the pattern for every measure of this kind.
Nothing is felt on 7 July. The dates that matter run in quarters: the record closes, a determination is published, an effective date follows, the first entry under the annex is filed, and only then does the first invoice change. Between the hearing and the first entry there is usually enough time to build records and not enough time to move a supply chain, and that distinction is the whole of the operational problem.
Two things the coverage does not say. A forced-labour theory moves the burden from price to records: whether or not the duty lands, the traceability question has already been asked, and it will be asked again at a customs verification or in a customer's due diligence, where the file you cannot produce is the exposure that survives a dropped tariff. The other is that exclusions, where they are available, are a dated window rather than a standing right, granted on a record that has to exist before the window opens. Shippers who spend the next quarter arguing about ten against 12.5% will arrive at that window with an opinion and no file.
Run the counterfactual both ways. If the measure is dropped and you restructured origin on the assumption of 12.5%, you spent capital on a proposal and you own the transition cost. If the measure lands and you did nothing, you have no exclusion request, no traceability file and no time. Documentation is the only expenditure that pays in both branches. That is not an argument for spending freely; it is an argument for noticing that the two branches do not share a common downside.
The assumptions here are mine: USD 10 million of annual exports to the United States from one named economy, with the proposed 12.5% applied to the whole of it, which is the pessimistic case rather than a forecast. That is USD 1.25 million a year, about USD 104,000 a month. Suppose a traceability file covering the exported range costs USD 40,000 to build once and USD 10,000 a year to maintain, again my number and not the article's. The file is under 4% of one year of the duty and remains useful if the duty never exists. The duty is 12.5% and disappears entirely if the measure is vacated. One of those expenditures survives both outcomes and the other is a bet on one.
The reading holds only inside these boundaries. If your tariff line is not in the annex, none of this is your file. If another action already covers the goods, the incremental rate is not the headline rate. If production sits in a jurisdiction that was not named, the investigation belongs to somebody else, although the documentation question may still be yours. What I would not do is treat a three-day hearing as a decision. The record is not the remedy, and the range is not the rate.
↑ Back to the key points