US Section 232 pharma duty hits 100% on 29 September for all non-listed suppliers

This one I am not neutral about. I have cleared pharma and medical shipments into the US for years, and a 100% additional duty arriving with one day of formal notice is the kind of thing that moves a company's annual forecast in a single afternoon. The effective date is 12:01 a.m. Eastern on 29 September 2026, which means any cargo in the air on the 28th needs to be looked at right now, not next week. The base rate under HTSUS 9903.04.60 is a 100% additional ad valorem duty on top of whatever duty already applies. Do not confuse the base rate with the headline rate for your own shipment, because there are four separate bands in this measure and which one you fall into is determined by facts you may not have papered.

Let me lay the bands out plainly, because the difference between them is the difference between a workable business and a dead one. Products of companies outside the Annex III list pay the 100% base rate. Products of Japan, EU members, South Korea, Switzerland and Liechtenstein pay 15%. Products of the UK pay 0%. And an importer with a qualified onshoring plan pays 20% on the covered goods, stepping up to 100% from 2 April 2030. So the range between the best and worst case for the same physical molecule is 0% versus 100%. That is not a duty question. That is a sourcing question, and it has an answer you can influence if you move quickly.

What most people miss is that these categories are not self-evident from the commercial invoice. The 23 September Federal Register notice set out the zero-rate categories and made six technical corrections, and Customs put out its guidance on 28 September. Read those two documents against your own entry data before you assume anything. Six technical corrections in a notice that lands five days before an effective date is a signal that the scope is moving and being tightened, and the difference between being inside a zero-rate category and outside it can come down to a product description, a dosage form, or whether an input is patented. I have seen a client discover a duty exposure because the goods description on the invoice said one thing and the tariff classification said another.

There are two doors in this measure that a well-prepared importer will use, and they need different work. The country-of-origin door is the faster one. If your product can be legally and commercially sourced from an eligible origin - 15% or 0% - the restructuring is a supplier and documentation exercise rather than a factory relocation. But origin is not something you declare into existence. Substantial transformation rules apply, and I have watched people assume that packing or final assembly in an eligible country confers origin when it does not. You need the origin analysis in writing, from someone who will stand behind it.

The other door is the qualified onshoring plan, and that is the one with a clock on it. The rate is 20% instead of 100%, but it steps up to 100% on 2 April 2030. If your model is to import at 20% and build out US capacity, the number that matters is not the 20% today, it is whether you can be operational onshore before the step-up, or whether you can negotiate an extension or a product exemption. Work backwards from 2 April 2030 with a realistic construction and qualification timeline. In pharma, qualifying a new manufacturing site is not fast, and regulatory approval runs on its own calendar, not yours.

Now the arithmetic, because this is where people panic without doing it. Assume you import $6 million of covered patented product a year, CIF value, and assume the pre-existing duty on it is zero. At the 100% rate you are looking at $6 million of additional duty. At 15% it is $900,000. At 0% it is nothing. At 20% under an onshoring plan it is $1.2 million. So the gap between the worst and best country-of-origin outcomes is $6 million a year, and the gap between the base rate and the onshoring rate is $4.8 million. If your gross margin on that product line is 35%, holding the 100% unmitigated means your effective margin on the line could swing by tens of percentage points depending on ability to pass it through. Assume a third of the cost can be passed to customers, a third absorbed, and a third is genuinely at risk of lost volume. That mix is the difference between a bad year and a restructuring.

Assume also that you have three months of inventory already in the US, which is typical, and that your landed cost per unit goes from about $10.00 to $20.00 where units are concerned. On a product that sells at $30.00, that is a 33% cost structure change overnight. Nobody's pricing system is set up for that without a senior decision, and the decision has to be made before you take another booking.

What do you do, and when? Today and this week, not after the quarter closes.

Start with a physical cut-off audit. Pull every open purchase order, every booking and every in-transit shipment of covered goods. For each one, determine whether it is inside the base rate, an eligible-origin rate, a zero-rate category, or the onshoring rate. Anything arriving after 29 September needs a landed cost recalculated today. For anything where the producer is outside the Annex III list and you have no mitigation, get a written entry strategy from your customs broker before the cargo arrives, and not after. A mis-declared origin or a wrong tariff line is not a paperwork error anymore. It is a 100% exposure plus penalties.

Next, get the origin question answered in writing. Ask your supplier, in writing, for each SKU: the country of origin, the basis for that claim, and whether they have a recorded substantial transformation analysis. If they cannot answer, assume the base rate applies and price accordingly. Where the goods can come from Japan, an EU member, South Korea, Switzerland or Liechtenstein, do not just note it - confirm with the supplier that the cargo will actually be produced and shipped from there, and that the documentary trail supports the claim. Suppliers will tell you what you want to hear on a phone call. Get it in the contract with a duty-shortfall indemnity, because if the origin claim fails, the difference is 85 percentage points of duty and it lands on you as the importer of record.

After that, decide on the onshoring route with a real timeline, not a hopeful one. If you have any volume in covered goods, model the 20% rate against the 2 April 2030 step-up. Assume the step-up happens - do not plan on an extension - and ask what would have to be true for onshore production to be cost-competitive against 15% or 0% import. If the answer is that it is not competitive, then the onshoring plan is not a sourcing strategy, it is a tax deferral, and you should treat it as such with an explicit exit date.

Then the paperwork that people always leave too late. The 28 September Customs guidance is the operational document. Read it against your broker's process and confirm three things: which HTSUS provisions you are declaring, how the certifying statements for any eligible treatment are being made, and who inside your business is signing off on origin. Put that person's name on it. When CBP asks, you want a person, not a department.

One more practical warning. The six technical corrections in the 23 September notice tell you the scope is not settled. Corrections of that kind usually mean classification lines get clarified in a way that pulls some goods in and pushes some out. Build a monthly review of this measure into your classification process through at least the first quarter of 2027, and check the zero-rate category list against your actual product list every month. The single most common way importers lose money on a measure like this is by setting the classification once and never revisiting it.

My read on what this means for you, stated plainly: for anyone with pharma or ingredient exposure into the US and no Annex III listing, the 29 September date is not a compliance event, it is a sourcing event. The people who will come out of it best are the ones who spent this week pulling entry data and chasing origin evidence, not the ones who spent it reading news coverage. There is a version of this where you end up paying 15% or nothing, and there is a version where you pay a hundred, and the gap between them is mostly documentation plus the willingness to make a decision this week.

Let me also be honest about what I do not know, because pretending otherwise would be useless to you. I do not know how long the country-level rates at 15% and 0% will hold. Measures like this have been amended before, and a country band that looks safe in September can be revisited in a negotiation. What I do know is that the base rate is 100% and it is live, so the downside of doing nothing is fully defined while the upside of waiting for a better number is not. That asymmetry is the whole argument for acting now.

I also do not know how CBP will treat goods that were shipped before 29 September but arrive after it. Normally the duty is determined by the date of entry, not the date of shipment, which is why the in-transit population is the group with the most urgent problem and the least time. If you have cargo on the water or in the air right now and it arrives on the 30th, you need a written answer from your broker today on how that entry will be filed, and if there is any legal basis to accelerate or adjust the entry, you want to know about it before the vessel berths, not after.

There is also a currency to this conversation that is easy to miss. Pharma supply chains carry long qualification cycles. Moving a product from one manufacturing site to another site in a different country is not a purchase-order decision, it is a regulatory filing decision, and depending on the product, the paperwork runs well past a year. So for manufacturers, the practical question is not whether to move, it is which products have commercially viable alternative sites and how quickly each one could be qualified. Rank your portfolio by how fast it could be re-originated, and work the fastest movers first. That is usually a small subset, and it is almost never the highest-volume product, because the highest-volume products tend to be the most deeply embedded in a single site.

For importers who buy rather than make - distributors, wholesalers, and buying groups - you have a different and in some ways easier problem. You cannot relocate a factory, but you can change which supplier you buy from, and the supplier's country of production is the variable that determines the duty. Go back through your supplier list with the country of production for each product line written next to the supplier name, or better, next to each site. You will probably find that some suppliers code for the same product from two or three countries, and that the allocation between them is currently decided on cost and lead time rather than duty. Once duty is in the picture, that allocation changes, and it is a conversation you can have this week with a phone call and a spreadsheet.

One thing I would specifically caution against is restructuring for duty reasons without checking the other side of the trade. If you shift sourcing to an eligible origin to escape the 100% rate, you may be changing your exposure to free trade agreements, rules of origin under other regimes, and your own customer contracts that specify country of manufacture. Pharmaceutical buyers often have quality agreements that name the manufacturing site. Changing site without amending the quality agreement is a much more expensive problem than the duty itself. Route the sourcing change through quality and regulatory at the same time as you route it through procurement, not after.

There is a final point about how you should communicate this upward. The number that gets attention in a boardroom is the $6 million, and it is correct, but it is also unhelpful on its own because it implies the only options are paying it or not shipping. The more useful framing is a range with three scenarios, each with an owner and a date: the unmitigated case, the country-origin case, and the onshoring case. Put a percentage probability on each, put the cost against each, and hand it up with the decision it requires, which is a sourcing decision with a deadline and not a duty problem to be minimised over time. In my experience the reason these measures catch companies out is not that the analysis is hard. It is that nobody puts a name and a date on the decision until the first entry has already been filed at the wrong rate.

One more thing worth saying, and then I will stop. If you import pharma inputs rather than finished goods, check whether your inputs are patented at all, because the measure covers patented pharmaceuticals and ingredients, and an unpatented API or an excipient may sit outside the scope entirely. Several of the companies that will pay the most under this measure are the ones that assumed everything in a pharma supply chain is covered. It is not. Read the scope against your own bill of materials line by line, and get a written classification opinion on anything ambiguous rather than relying on a phone conversation with your broker. The two things that cost importers real money in a measure like this are a wrong origin claim and an over-broad assumption about scope, and only one of those is an error you can be penalised for while the other quietly costs you margin you did not need to give up.

One last thing on timing, because the calendar here is genuinely tight and it is easy to lose a week to discussion. The measure took effect on 29 September. Your first post-effective entry may already be in progress. Set yourself a hard internal deadline of 9 October to have the origin evidence for your top ten SKUs by duty exposure in writing, and 16 October to have a documented decision on the onshoring question for whichever products carry enough volume to justify it. Between those two dates, sit with your broker and walk one live entry through the new provisions end to end, so that the process is tested on a shipment you understand rather than on one that surprises you. A rehearsal on a real entry is worth more than a policy document, and in a measure with four rate bands and a scope that has already been corrected six times, rehearsal is the only way to find out that your declared treatment does not match what your supplier actually ships. I would rather find that out in October on a modest entry than in January on a quarter of your annual volume.

Leo

• Before your next booking, run a cut-off audit of all covered goods in transit after 29 September and recalculate landed cost on each one at the applicable band: 100%, 15%, 0%, or 20%.
• Get every supplier to confirm in writing the country of origin per SKU, the basis for the claim, and whether a substantial transformation analysis exists, backed by a duty-shortfall indemnity in the contract.
• Model the 20% onshoring rate against the 2 April 2030 step-up to 100%, assume the step-up happens, and set an explicit exit or extension date rather than treating it as a permanent rate.
• Review the 23 September Federal Register notice and the 28 September Customs guidance against your own entry data, and check every SKU against the zero-rate category list.
• Name an individual inside your business as the signatory on origin and classification determinations, and put the 9903.04.60 treatment in writing with your customs broker before the first post-29-September arrival.
• Set a monthly re-check of this measure through at least Q1 2027, given the six technical corrections already made to the scope.