US Slaps 260% Duties on China Van Trailers: 129.73% AD Plus 134.75% CVD

The Analysis

The phone rang at 7:10 on the morning of October 29. It was a broker I have worked with in Laredo for about eleven years, and she got three sentences out before I could say good morning. There is an order. It landed yesterday. And there is a parts list.

She runs a border desk, not a policy shop, and it was not the rate that had her unsettled. Everybody in this trade can read a headline number. What got her was the list, because she files entries all day and she knows exactly how much of that list her software does not recognise.

Here is what the Federal Register carried on October 28, in plain numbers. An anti-dumping duty order at 129.73 per cent. A countervailing duty order at 134.75 per cent. Congress designed those two remedies to stack, and they stack on the same entered value: 129.73 plus 134.75 comes to 264.48 per cent, not the round 260 that most of the coverage is running with. And it runs for a minimum of five years.

Next to those numbers sits the statistic everybody quotes and nobody divides. In 2024 China shipped thirty-eight complete van trailers into the United States, worth about 230,000 dollars. Mexico shipped forty-seven thousand of them, worth about 1.5 billion. The petitioning coalition's own estimate is that China, Mexico and Canada together account for roughly forty per cent of US van trailer supply. Three statements, one fact pattern, and only one of them describes the entry you file tomorrow.

Start with what this thing actually is, because "a 260 per cent tariff" is doing damage in purchasing meetings this week. It is two orders, not one. The dumping side found that Chinese producers sold below normal value and set the margin at 129.73 per cent; the subsidy side found countervailable support worth a net 134.75 per cent. Each was investigated separately, each carries its own respondent list, and on any given entry both are assessed against the same entered value. That is the whole reason the number looks absurd: it is not one rate that got large, it is two ordinary margins being sampled off the same dollar.

The gap between 260 and 264.48 is not pedantry. Your deposit today and your liability in three years are two different things, and they move on two different calendars.

Nothing publishes itself, and nobody files a case over thirty-eight trailers. A domestic coalition petitioned, sustained the timetable, delivered a final injury determination, and Commerce published within days of it, which is how these things go when the petitioners have done their homework. Their forty per cent supply figure deserves the scepticism you would apply to any number put forward by somebody who wins if you believe it. It is an interested-party estimate, and the shipment statistics sitting next to it tell a considerably quieter story about finished goods coming out of China.

That quiet story explains the drafting instinct behind the scope. Put a big number on a finished product and the trade does two predictable things: it ships the product knocked down, and it moves the last assembly into another country. So the order names chassis, side walls, doors, axles, brakes and refrigeration units from the first line of the text. Writing it that way is not rhetorical, it is cheaper than litigating the rerouting a year from now, and whoever drafted it clearly decided the parts were where the volume would go.

Divide the numbers and the story changes shape. Two hundred and thirty thousand dollars across thirty-eight units averages about 6,052 dollars a unit. One and a half billion across forty-seven thousand units averages roughly 31,915 dollars a unit. A 53-foot road van does not cost six thousand dollars, and thirty-two thousand is about what one does cost. The two series cannot be counting the same animal: either the Chinese thirty-eight are small units, demonstrators, used boxes, repair returns or some statistical line nobody reconciled, or the declarations behind them never described finished road trailers at all.

I care about that for an unglamorous reason. The comfort line going round the trade is that China only shipped thirty-eight complete trailers, so this order cannot touch us. That sentence is true only if your own entries sit on the same statistical line that produced the thirty-eight. Pull twelve months of your entry summaries, put a finger on the classification number, and check what you actually declared. If your goods have been coming in under a heading that produced some other statistic, then thirty-eight was never your risk profile.

Who pays, and how much of it? Legal liability sits with the importer of record, which is usually not the person who negotiated the purchase, and that mismatch is about to cost somebody their bonus.

Dealers and fleets taking delivery of finished trailers imported from China are the smallest exposed group by volume and the most violently exposed group per unit. There were thirty-eight of them last year. If that is your business, the question is not whether you can absorb 264 per cent, it is what you do with the units already on the water.

Parts distributors are a different and much larger story. Axle sets, brake assemblies, refrigeration units, runs of side panels and doors: these cross ports and border crossings every week, they are bought against drawings and part numbers, and the people buying them have never once thought of themselves as participants in a trailer trade case. If what they buy is destined for a van trailer, the scope language reaches them.

Then there is the North American assembler, the operator who moved to knocked-down kits precisely to sit on the right side of every previous measure. Whether this order reaches that assembly is a question that has a written answer somewhere. You do not have it in writing yet, and the gap is where the money is.

Timing is where people talk themselves into trouble. The effective date is stated in the Federal Register notice, and the port does not work off news coverage: Customs publishes instructions to the trade, brokers file against those instructions, and the software at the border gets mapped accordingly. Deposits start for entries made on or after that date. Nothing in it reaches back to entries already made, which is why every conversation I have had this week began with somebody reciting a ship date.

Past that date there are three others that matter more than anything in the headline. One year out, the anniversary month opens a window for administrative review requests, and for anybody hoping a supplier's rate comes down, that window is the only road to a refund; miss it and the entries liquidate at the deposit rate with the difference simply gone. Around two years out the review results land and rates move, up or down. Five years out the sunset review starts on its own schedule, and orders that survive it run another five. Plan plant capacity against a ten-year horizon, not against next quarter.

Now the part I have not seen written anywhere, and the part I would drive across town to say out loud.

Read this as a trailer measure and you will miss it entirely. It is not a trailer measure, and it was never drafted as one, because anti-dumping and countervailing orders attach to merchandise by a written scope description rather than by tariff heading. That distinction sounds like academic hair-splitting until you watch an entry get filed. Broker software keeps case cross-references against tariff numbers, maintained by hand, updated when somebody remembers, and those references will get built for the finished-goods line reasonably quickly. They will not be built for an axle imported under heading 8708, a refrigeration unit under 8418, or an aluminium side structure landing somewhere in 7616. None of those lines says trailer to any machine. Which means the entries least likely to be flagged automatically are exactly the entries the petitioners wrote the scope to catch, and the exposure is concentrated precisely where nobody is auditing.

A second trap hides in four small words. Scopes of this kind reach parts that are for use in the subject merchandise, or that ship together with it, or that are dedicated to that use and nothing else. That is an end-use test rather than a classification test, and the evidence that answers it is your own paperwork: the purchase order, the part drawing, the customer's assembly instruction, the way the goods are kitted, and occasionally the boring fact that nobody has ever ordered that part number for anything else.

My colleague Dr. Ingrid Voss has a line I steal regularly, that it depends, let me tell you on what, and this is one of those occasions. What your entry needs depends on the written scope, on whether your part is dedicated to trailer use, and on where the last assembly step happened.

On that last point, be careful about borrowing a conclusion from a neighbouring rule. Whether a Mexican assembly operation takes Chinese parts out of this particular order is a scope question answered by reference to substantial transformation, and if you want certainty rather than a view, it is answered in writing. Whether the same operation takes them out of other origin-based measures is a different question, asked under different authority, with a different body of practice behind it. The two can come out differently on identical facts, and carrying one answer around as though it settled all of them is how traders end up explaining themselves to an enforcement team two years later.

A scope determination request is not exotic paperwork. If you run a real assembly operation with real records, it is the cheapest insurance available, and I would open the file this month. What I would not do is assume the answer. Until a ruling exists, declaring the case number and depositing is defensible; declaring nothing because you believe your own transformation argument is an unsecured bet, and it is staked with your money and your bond.

Here is the second thing missing from every write-up, and it stops cargo faster than any rate will. Deposits at this level are enormous relative to invoice value, and customs bonds are sized against duties rather than against goods. A continuous bond is normally set as a percentage of duties and taxes paid over the previous twelve months, with a floor somewhere around fifty thousand dollars. Take a mid-size importer moving twenty finished trailers a month at thirty thousand dollars entered value. At 264.48 per cent that is roughly 1.59 million dollars a month of deposits, about nineteen million a year on paper, and a continuous bond requirement heading for two million dollars.

Cut the volume to one single trailer and the duty on it alone lands a little under eighty thousand dollars, already above the face value of a minimum-size continuous bond. When the duty on an entry exceeds the bond standing behind it, the entry is rejected. Nothing is seized and nobody accuses you of anything: the truck simply sits, and demurrage, storage and the customer's phone calls start running immediately. Single transaction bonds sized to the entry deal with it cleanly, and they have to be arranged before you file, not after the entry comes back.

Let me put arithmetic on all of it, assumptions stated. Take a finished van trailer with an entered value of thirty thousand dollars, arriving by vessel, with no other trade remedy stacked on top of these two cases. Anti-dumping at 129.73 per cent of entered value is 38,919 dollars. Countervailing duty at 134.75 per cent is 40,425 dollars. Together they are 79,344 dollars, or 2.6448 times what you paid for the asset. Merchandise processing and harbour maintenance, at their current ad valorem rates, add something short of one hundred and fifty dollars at this value. Before you have trucked it, paid a broker, or taken delivery, you are around 109,500 dollars into a thirty-thousand-dollar trailer, roughly three and two thirds times the purchase price. Those two cases are the only layers I have counted; if anything else already sits on your line, add it yourself, because this number does not include it.

Run the same arithmetic on the kit strategy everybody is quietly pricing. Assume a knocked-down kit for one trailer with an entered value of eighteen thousand dollars, sixty per cent of the finished unit, which flatters the kit. If those parts fall within scope, and that is precisely the question nobody has settled in writing, the deposit is about 47,606 dollars. You saved twelve thousand on the goods and carried nearly forty-eight thousand of exposure. The kit route pays only if the parts sit outside the written scope or the assembly genuinely takes them out of it, and that is not yours to decide. You get to ask, in writing, and then you wait.

The scale check that puts the whole argument in proportion: every finished trailer China sent here in 2024, all thirty-eight units at the average declared value implied by the trade figures, would have generated somewhere around 608,000 dollars of deposits across a full year at the stacked rate. Six hundred thousand dollars, set against a Mexican trade worth 1.5 billion. Nobody built a strategy around the thirty-eight units. The money in this case was always going to be on the parts side, and on whatever gets assembled afterwards.

Now flip it, because very little here is as settled as it looks. If your assembly operation is genuinely transformative, thoroughly documented, and eventually confirmed by a written ruling, your entries of Mexican-produced trailers are simply not subject to this order, and the whole exercise was an expensive scare. That outcome is genuinely available. Treat it as a hypothesis with a filing deadline rather than as a fact.

The rate can also move against you. Dumping and subsidy margins get recalculated on every annual review, producers enter and leave respondent lists, and an increased rate applies to entries inside the review period. That produces a letter roughly two years from now asking for more money on a shipment you sold out eighteen months earlier, against a supplier who may no longer answer email. When the rate falls instead, you see the money back only if you filed for the review in the correct month.

Then there is the route itself. Orders shaped like this one attract circumvention inquiries, the formal machinery for asking whether parts assembled somewhere else are being used to walk around what the drafters intended. They also attract private enforcement allegations, which can be filed anonymously and will put a year of your entries under a microscope. Neither is a reason to stop doing business. Both are reasons to have your own file in order before somebody else opens it.

What I would actually do, sitting where you sit. Skip the strategy memo and start with the paperwork that already exists. Pull every line you have entered over the last twelve months that could conceivably end up on a van trailer: finished units, axle sets, brakes, doors, side panels, refrigeration units, anything you know goes into one. Read the classifications against the scope language in the order rather than against your own habits. Make two calls before lunch. One to your broker, to establish which classifications their system has actually mapped against the new cases and what happens to lines that are not mapped at all. One to your surety, to rerun the bond arithmetic on next month's volumes before an entry gets kicked back rather than after.

Past today's fire, stop treating this as a question with a single answer. Anything with a real assembly step in it gets a scope question put to Commerce in writing. Two categories of record decide whether that filing succeeds: what the production actually consists of, captured at the time with film, photographs and time records, and where every input came from, evidenced rather than remembered. I have watched competent businesses try to reconstruct an assembly line from the memory of a supervisor who left in the interim, and it is both expensive and unpersuasive.

Where does this leave the trade? That 129.73 and that 134.75 are going to be recovered by somebody, eventually, through higher box prices, higher lease rates, or a longer service life on the fleet people already own. Fleet buyers will work out over the next two quarters that this order had less to do with imports from China than with how much pricing power the surviving North American plants discovered they were holding. If you place fleet orders for next year, start that conversation now, while there are still alternatives on the table and before either side works out how short the list is about to get.

↑ Back to the key points

— By Derek Xu