Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- Within five working days, segment the catalogue by typical basket value and count the SKUs below 30 euros, between 30 and 150, and above 150; the count below 30 is your repricing population and it must be produced per SKU, not per market.
- Before 1 July 2026, obtain the implementing text and confirm in writing how it defines a consignment and whether goods in transit on that date are covered; no split-shipment plan should proceed until that answer is on file.
- Recompute landed cost for the 140 to 160 euro band line by line and move any SKU away from the threshold rather than toward it; the target is zero SKUs priced within five euros of 150 for a single-unit order.
- Build the 2028 expiry into every price list and customer contract signed this year, with a review trigger twelve months before the sunset date and a named owner for the recalculation.
- Report the flat charge monthly as a percentage of gross margin per SKU, and escalate any SKU where it exceeds 15% of gross margin within one reporting cycle.
The European Union will begin levying a flat EUR 3 customs duty on goods in consignments valued at EUR 150 or less entering the EU from 1 July 2026, ending the long-standing de minimis duty exemption. The interim measure, set to run until July 2028, primarily targets B2C e-commerce flows and is expected to cover 93% of all e-commerce imports into the EU. Logistics providers, customs brokers and cross-border sellers face new compliance, tariff classification and landed-cost complexities as the bloc advances its broader customs reform agenda.
The Analysis
From 1 July 2026 a consignment valued at EUR 150 or less entering the European Union will carry a flat EUR 3 customs duty. The measure is interim, scheduled to run until July 2028, aimed primarily at B2C e-commerce flows, and expected to cover 93% of all e-commerce imports into the bloc. The de minimis exemption ends on that date.
The number worth arguing about is not three euros. It is the word consignment, and the figure 150. Both do more work than the rate does.
A flat charge is not a simplified tariff. It is a fixed amount attached to a flow that was previously outside the duty system altogether, which means that flow now has to be identified, counted and declared.
Because the amount does not move with value, the effective rate does. Three euros is 2% of a 150 euro consignment and 30% of a 10 euro one. Any budget owner who models this as a percentage will be wrong in one direction on low-value baskets and wrong in the other at the top of the band, and the error will be largest exactly where the margin is thinnest. The cost driver is not value. It is the number of chargeable events, which is a data question rather than a pricing question, and the team that owns it is usually not the team that owns the price list.
The 93% figure needs a unit before it means anything. It is a share of consignments, and a share of consignments is not a share of value. A flow made of many small parcels and a flow made of a few large ones can carry the same consignment count and entirely different value. Read as a count, the measure is designed to reach the tail: the high-frequency, low-value traffic the exemption was originally written for. Read as a share of value, the same sentence would say something else, and the report does not say it. Where a charge is regressive in value and broad in count, the burden lands on the sellers with the least ability to absorb a fixed cost per parcel, which is rarely the group with the loudest voice in the consultation.
A marketplace seller moving a couple of hundred parcels a day meets this as a fixed cost per order, and on a low-price catalogue that is not a rounding error; for some items it is the line between a viable SKU and a discontinued one. A brand shipping pallets to a distributor is outside the instrument directly and should not spend a euro on it, though it should watch the distributor, because small orders migrate. A logistics provider inherits the data problem: who is the declarant, what counts as one consignment, and how a marketplace's consolidated feed is split into declarations. A retailer whose average order sits above 150 is outside and the correct action is to confirm that in writing and stop reading.
The first operative date is 1 July 2026, and it is not the only date that matters. For a given parcel the relevant moment is when its declaration is lodged or accepted, and goods already in transit on that date may or may not be covered by a transitional rule. The implementing text decides that; the announcement does not. The second date is July 2028. This is an interim measure standing in front of a broader reform of the de minimis treatment, and an interim measure has a known expiry by design. Any landed-cost model running past that date is running on a number that was never meant to be permanent.
The part the coverage will not tell you is that the instrument has a cliff, and the cliff is not at the rate. At 149.99 the treatment is three euros flat. At 150.01 the treatment is ordinary duty plus the ordinary procedure, and ordinary duty on many product lines passes three euros long before the value gets large. So the marginal euro at the threshold does not add a few cents of duty; it changes the regime. That makes the dangerous inventory the basket priced near the line, not the cheap item everyone is doing arithmetic about.
And the instinct every seller will have at that line, which is to split the order or re-invoice or ship in two boxes, is the wrong instinct twice over. If the implementing text defines a consignment by consignor, consignee and period, splitting achieves nothing and leaves a record of the attempt. If it defines it per parcel, splitting doubles the number of chargeable events at three euros each. There is no version of the manoeuvre in which the seller comes out ahead, and there is a version in which the seller manufactures evidence.
Work the curve, because the curve is the whole argument. Three euros on a 150 euro basket is 2%. On a 35 euro basket it is 8.6%. On a 12 euro basket it is 25%. The ten percent line sits at 30 euros, so every SKU whose typical basket falls below that is paying more than a tenth of its value in duty before freight. Assume, and these are my assumptions rather than figures from the report, a 35 euro average basket, 12,000 parcels a year and a 30% gross margin: duty comes to 36,000 euros a year, which is 8.6% of revenue on those parcels and 28.6% of gross margin. Take one item at 24 euros and three euros is 12.5% of the price and 41.7% of the margin. Those are not numbers you absorb. They are numbers you reprice, delist or ship differently, and the decision belongs at SKU level rather than market level.
The trap is treating three euros as a tariff rate and entering it in a spreadsheet as a percentage. A second is assuming a consignment equals a parcel. A third is assuming 2028 is a stable horizon and signing a multi-year price without a review trigger. The expensive one is discovering the 150 boundary after the catalogue has been rebuilt around it.
Strictly, the instrument applies to consignments at or below 150 euros entering the Union from 1 July 2026. It does not apply above that line, and if most of your flow sits above it then the analysis you need is the ordinary duty regime, not this one. It also stops applying in July 2028 unless something replaces it, and the direction of the reform suggests the replacement will sit closer to ordinary treatment rather than further from it. That is the boundary to hold in mind: a three euro placeholder with a two-year life, standing in front of a permanent change that has not been written yet. Plan for the placeholder. Do not build the warehouse around it.
↑ Back to the key points