EU Adds a 2-Euro Line Fee From Nov 1: Three Items Means 9 Plus 6 Euros

The Analysis

The paperwork does not lie, and from 1 November the paperwork gets more expensive. The European Union starts charging a handling fee of two euros for every item line declared on a distance-sales parcel sent from outside the bloc, and it lands on top of the three euros of duty that has been applied since July. Estonia's tax and customs board gave the cleanest illustration anyone has published: a parcel carrying three different goods carries nine euros of duty and six euros of handling. Fifteen euros, and that is before ad valorem duty and before VAT.

Fifteen euros is not what should worry you. What should worry you is the unit the fee is charged on. It is not the parcel, it is not the value of the goods, and it is not the weight. It is the number of lines on the declaration. A seller who writes tidy declarations and a seller who writes messy ones pay completely different amounts for the identical box of goods, and most people running cross-border e-commerce have never looked at how many lines their system produces per order.

I spend my days in declarations, so that is where this piece starts: with the line, with who signs as declarant, and with the money that has to be in the right account before the parcel is released.

Start with the rule itself, because the details decide the money. From 1 November the European Union applies a handling fee of two euros to every item line declared on a distance-sales parcel arriving from outside the bloc. The two euros is permanent. The three euros of duty that sits underneath it has been running since July and is temporary, currently written to run until 1 July 2028. Both of them are charged per item line, and that shared unit is the whole story.

Estonia's tax and customs board published the example that everyone is now quoting, and it is worth reproducing exactly: a parcel containing three different goods carries three lines, nine euros of duty and six euros of handling, fifteen euros in total. Read the small print on that fifteen. It covers the two flat charges and nothing else. Ad valorem duty, where it applies, is on top. VAT is on top. Anything your carrier charges for advancing the money is on top. The fifteen is a floor, not a bill.

An item line is not a physical item, and conflating the two is where the trouble starts. A line on a customs declaration is one entry for goods sharing a description, a classification and the rest of the declared particulars. Three different goods mean three lines even if two of them are identical socks in different colours. Twelve identical phone cases in one order should normally be a single line. The fee follows the line, so the fee follows how your data is written.

That design choice produces results that look absurd until you think about them for a minute. Assume a single-item order worth four hundred euros: one line, three euros of duty, two euros of handling, five euros on a four-hundred-euro sale. Assume a forty-euro basket of five cheap accessories: five lines, fifteen euros of duty, ten euros of handling, twenty-five euros on a forty-euro sale. The fee and duty together come to sixty-two per cent of the goods value. Nothing about the second parcel is more expensive to handle in physical terms; it simply carries more lines.

Why now? Because the volume of low-value parcels arriving from outside the EU has grown to the point where customs administrations are processing millions of declarations that generate almost no revenue and consume real handling capacity. The three euros of duty introduced in July was the first attempt to make those declarations pay for themselves. The two-euro handling fee is the second, and unlike the duty it has no sunset written into it. Read that asymmetry for what it is: the duty is a policy instrument with a review date, the fee is an administrative cost recovery that will still be there after the duty has been argued over and redrafted.

The direction of travel matters more than either number. The EU has been dismantling the advantages that made direct-to-consumer parcel shipping from outside the bloc cheap, piece by piece: VAT relief went, then the flat duty, now a per-line handling fee, and the removal of the duty relief threshold is the item still on the table. Each piece is small. Stacked, they change whether shipping a fifteen-euro parcel from a third country to a European doorstep is a business at all.

Where exactly the perimeter sits is not my job to declare; the legal reading is Ingrid's, and the one I work from is hers. This attaches to business-to-consumer distance sales from outside the EU. Business-to-business consignments and genuine private gifts are outside it. The obligation to pay falls on the platform, the seller or whoever lodges the declaration, and the carrier is not permitted to arrive afterwards and collect the money from the person who received the parcel.

Now translate those four sentences into what happens on your desk. Because the carrier cannot collect from the recipient at the door, the money has to be settled at or before the parcel is released. If you are the declarant of record, your account with the postal operator, the courier or the broker has to be funded and it has to stay funded. If you are not sure which of those you are on each of your sales channels, find out this week, because the invoice follows the declarant, not the brand on the box.

Then there is the advance. Somebody has to put the fifteen euros up front, and whoever does will charge for the service of doing it. Brokers and carriers typically book that as a disbursement fee or an advance fee, usually a fixed amount per parcel, and it is regularly bigger than the two euros it is advancing. Pull your contract and look for it. On a single-line parcel your true incremental cost is not five euros; it is five euros plus whatever your carrier charges to front it plus whatever they charge to administer the VAT. The policy costs two euros. The plumbing costs what your contract says it costs.

The impact splits cleanly by how many lines your average order carries. A seller shipping one item per order pays five euros a parcel and can absorb it inside normal pricing noise. A seller in apparel, accessories, beauty sets, tools or automotive parts, where three to six lines per order is normal, pays fifteen to thirty euros a parcel and cannot. Sample boxes, subscription boxes, gift sets and starter kits, which routinely run eight to fifteen lines, land somewhere between forty and seventy-five euros a parcel, and for most of those businesses the product economics do not survive that. Repair and spare-parts networks are the quiet casualty: one order with ten different part numbers carries fifty euros of flat charges before anyone looks at the value of the parts.

Marketplaces feel this differently depending on who declares. Where the platform is the declarant, the platform absorbs the charge and reprices it into seller fees or into the buyer's checkout, and your job is to read the updated fee schedule. Where you sell on your own storefront and ship through a postal or courier solution, you are the declarant and the invoice comes to you. The same physical parcel, two completely different cash flows, decided entirely by a field in a contract.

The trigger is the declaration, not the order. Parcels you shipped in October but which are declared on or after 1 November are inside the new charge. Anything sitting in a sorting centre, a transit shed or a customs buffer over the date change is inside it. The November peak is the real problem: bundles, gift sets and multi-item orders are exactly what pushes the line count up, so the fee is at its most expensive in the same weeks your volume is at its highest. And the money arrives late. Expect the November declarations to show up on a carrier or broker invoice somewhere between two and six weeks later, which puts the cost into your December or January accounts, just after you have locked Christmas pricing.

Here is the part I have not seen in any of the coverage, and it is the part that will cost people real money. The fee is charged on declared lines, and the number of lines your parcels carry is a decision made inside your order management system, usually years ago, usually by someone who was thinking about picking lists rather than customs declarations. Go and count. Many platforms generate one declaration line per order line item. Some generate one line per physical unit. Some split automatically by tariff heading. Some simply pass through whatever the merchant's ERP emitted when the catalogue was first loaded. Assume an order of six identical phone cases in one colour: written correctly that is one line and five euros; written per unit it is six lines and thirty euros. Twenty-five euros, on one parcel, for a purely clerical reason, and it will repeat on every order you ship until somebody fixes the rule.

The second thing nobody is saying is what the billing ban really does to you. Everyone read the line about carriers not being allowed to bill recipients as consumer protection, and so it is. Its other edge is that there is no second collection attempt. If the line count is wrong, if the duty was underpaid, if the declaration was sloppy, the carrier eats it and then invoices you, and you cannot recover it from the customer at the door. Accuracy stops being a compliance virtue and becomes a cash leak control, because every error is now a loss with no recovery path.

So the obvious question: can you merge lines and pay less? Sometimes yes, and the boundary is worth being precise about. Merging is legitimate when the goods genuinely share a classification, an origin and the rest of the declared particulars, and when the merged line still describes what is in the box. Merging two different tariff headings into one line to save two euros is misdeclaration. The fee is small enough that cheating it is never worth the exposure, and large enough that cleaning your data always is. That is the whole trade in one sentence: never merge across a heading, always merge within one.

Splitting does not help either. The charge follows the line, not the parcel, so cutting one five-line parcel into five single-line parcels still produces five lines and the same twenty-five euros, plus five sets of pick, pack and transport costs. Anyone who tells you to split to save the fee has not read the unit. Consolidating orders over time does help, because four separate two-line orders shipped across a month become eight lines while one consolidated shipment of the same goods becomes two.

The structural answer is the one the policy was arguably written to encourage: stop shipping single parcels into the EU and start importing in bulk. A container of five thousand units imported under one declaration pays duty on the consignment and carries no per-parcel handling fee at all, because it is not a distance-sales parcel. Then you fulfil domestically from EU stock. That is a large decision with inventory, working capital and VAT consequences, and it is not a November project. It is also where the arithmetic points, and if your average line count is above about two and a half you should at least have the conversation before the duty relief threshold question is settled.

Let me put numbers on it with the assumptions stated. Assume you ship forty thousand business-to-consumer parcels into the EU every month and your declarations average two point six lines. That is one hundred and four thousand lines a month. Two euros a line is two hundred and eight thousand euros of handling. Three euros a line is three hundred and twelve thousand euros of duty. Together five hundred and twenty thousand euros a month, a shade over six and a quarter million a year, or thirteen euros on every parcel you send.

Now assume a cleanup that merges lines within the same tariff heading and origin, keeping every declaration accurate, and pulls the average from two point six lines to one point eight. Lines fall to seventy-two thousand a month. Handling is one hundred and forty-four thousand, duty two hundred and sixteen thousand, total three hundred and sixty thousand. The saving is one hundred and sixty thousand euros a month, just under two million a year, achieved without touching your product, your price or your supplier. Fewer lines also means fewer places for a classification error to hide, so the cleanup pays twice.

Take a smaller operation to show the other end of the scale. Assume three thousand parcels a month at four lines each, and an average basket of thirty-five euros. Twelve thousand lines a month: twenty-four thousand euros of handling plus thirty-six thousand of duty, sixty thousand euros a month against one hundred and five thousand euros of goods. That is fifty-seven per cent of merchandise value in flat charges. No amount of conversion rate work survives that. This is the seller profile that needs to change its fulfilment model rather than its declarations.

One more number to hold on to: the fifteen euros in the Estonian example excludes ad valorem duty. If and when the duty relief threshold comes off, the goods above it start paying duty on value as well as per line, and the two charges stack. Build your pricing model on the basis that the two-euro fee is permanent and the three-euro duty is not, and that a third component may arrive on top of both. Modelling the whole thing as a temporary nuisance is how people get caught in 2027.

What you do with this depends on one number you probably cannot quote right now. Pull thirty days of declaration data before the end of October and build a histogram of lines per parcel: the average, and the share of orders above five lines. If the average is close to one, reprice and move on. If it sits between two and three, the line-generation rule is a project for this month. If it is above three, the fulfilment model is the project and the declarations are the stopgap.

Get the declarant question settled in writing before the first week of November, on every channel you sell through. One email to each marketplace, each carrier and each broker: who is the declarant of record, who receives the customs invoice, what is the disbursement fee per parcel. The answers will not all match what your contract says, and the difference is money.

Then change the rule that builds the declaration: one line per distinct goods description rather than one line per unit, merging within a heading and never across one. Set a target you can measure, say a twenty to thirty per cent reduction in average lines within thirty days, and check it against real declarations rather than against a test file, because production data always looks different from the sample.

Put the new cost into the checkout before the peak. A buyer-facing tax estimate that says five euros for a single-line basket and fifteen for three lines is honest, and honest beats a surprise email after dispatch. At the same time, rebuild your landed-cost model with three components: two euros per line permanent, three euros per line until mid-2028, and the carrier's advance fee, which you should now know.

Separate the traffic that should not be paying this at all. Business-to-business consignments and genuine private gifts sit outside the charge, but only if the declaration says so and the evidence is attached. Give those orders their own path with the recipient's VAT number or the gift declaration carried in the data, and audit a sample each week, because the default is to treat everything as consumer traffic and the default is expensive.

One more job, and it is the one that keeps the savings real: reconcile. In the first week of November, run a live declaration with your broker on a real order, then compare the line count on the customs invoice with the line count in your system. They will drift. Mappings rot, catalogue changes introduce new descriptions, and a marketplace may re-cut lines on its own side before passing them on. Reconcile monthly from November onwards on a sample of at least one hundred parcels, and treat any gap as a defect to be fixed at source.

None of this is difficult and all of it is unglamorous, which is exactly why it goes undone. The parcels leaving your warehouse on 1 November already carry the number that decides what you pay; the only question is whether you chose that number or your system did. Get the count out of your data today, get the declarant confirmed in writing this week, and get the new figure into your checkout price before the peak. The fee is two euros. The mess around the fee is not.

↑ Back to the key points

— By Derek Xu