03 Feb '26
Low-value textile imports have been the subject of much debate and unrest in European customs practice for many years. It is an area where controls are strict, declared customs values are regularly rejected and additional assessments can be substantial. In practice, this sometimes creates a cat-and-mouse game: customs suspect systematic undervaluation, while importers often deal with low-quality bulk goods—unpackaged, unbranded and purchased at low prices—where the value may actually be low. It is precisely in this area of tension between suspicions of fraud and the reality of the textile trade that a dynamic arises that repeatedly causes a stir.
The recent judgment of the Court of Justice of 29 January 2026 in the joined cases Keladis I (C-72/24) and Keladis II (C-73/24) falls precisely within this area of tension. It revolves around a question that may sound more technical than it is, but in practice can make a difference of millions: can customs determine the value of goods on the basis of a statistically calculated “lowest acceptable price” (LAP) when the actual value can no longer be determined? Jikke Biermasz discusses the judgment.
It is a question that goes to the heart of the customs valuation system. The Court's answer is both practical and principled.
Europe's customs authorities are rapidly digitising. Data and statistics are the driving force behind risk selection and controls. Efficiency is the keyword here, but the risk is clear: what starts as a supporting signal can gradually turn into a presumption of evidence.
That was exactly what threatened to happen in these textile cases. Customs applied the lowest acceptable price for the customs value of the goods, which was calculated on the basis of data from Eurostat's Comext statistical database and the anti-fraud information system AFIS of the European Anti-Fraud Office, OLAF. In practice, this is also referred to as the 'Fair Price List'. What was once intended as a risk profile was now being used as the actual customs value. In the meantime, the goods were no longer available for physical inspection. The invoices with vague descriptions did not provide sufficient guidance. So Greek customs fell back on statistics.
The Keladis I and Keladis II cases concern the import of textile products from Turkey in the period 2014–2016. No customs duties were payable on these imports, but they were subject to import VAT, calculated on the basis of the declared customs value.
Greek customs launched an investigation in 2016 following complaints of alleged undervaluation. During the investigation, the authorities concluded that the customs values declared in more than 289 import declarations had been set carelessly or deliberately low. The goods could no longer be physically checked, and according to customs, the description on the invoices was general and insufficiently detailed to verify the value.
As they were unable to reliably determine the actual transaction value, the customs authorities rejected the declared customs values and used a statistical tool developed by OLAF (AFIS/AMT) to determine a “lowest acceptable price” (LAP). This LAP was then used as the actual customs value.
On this basis, Greek customs calculated that EUR 6,211,300.18 in import VAT had been evaded. This led to substantial additional tax assessments against the persons concerned.
The importers contested both the valuation method and their liability for VAT. The Greek court then referred preliminary questions to the Court of Justice.
According to the Court, there is a trend whereby the individual assessment of a transaction is increasingly being replaced by statistical reference values. Whereas valuation normally revolves around the specific circumstances of the import, it now seems to be overtaken by generic averages. The Court therefore reiterates the key point: statistics are a tool, not a valuation decision in themselves.
The customs valuation system is strict: minimum values and fictitious values are prohibited. Customs must assess value in a hierarchy with the transaction value as the starting point and primary method for calculating customs value. Only when this does not work do additional methods come into play: identical goods, similar goods, deduction method, calculated value and, ultimately, the fallback method of Article 31 CCC / 74(3) UCC. This method is also known as the 'reasonable means' method. How does a statistical average for goods of a particular commodity code and imported from a particular origin relate to the reasonable means method?
The LAP does not fit into any of the regular methods. The statistical data is simply too aggregated. It says nothing about quality, lack of branding, remnants, seasonal batches or strategic purchasing relationships. All it provides is an average per kilogram. And no one does business with averages in legal terms.
So the question was: is the LAP a prohibited minimum value? Or is there still room to fall back on a LAP if the facts are incomplete and valuation becomes difficult?
The Court did not side with either customs or the importer, but adhered to the principles of the valuation system. Legal ground 101 gets to the heart of the matter. The Court formulated it in sharply normative terms:
“Second, as regards the prohibition on determining the customs value on the basis of minimum values, referred to in paragraph 99 above, it must be held, as the Advocate General observed, in essence, in point 79 of his Opinion, that the possible use of LAPs could infringe that prohibition if that administrative practice were to be systematic in nature and if the economic operator concerned were not allowed to justify the lower prices indicated in its customs declaration. In the latter situation, the refusal to take account of transaction values below such a threshold would entail an upward adjustment of the declared values up to that threshold and would constitute a system of minimum values.”
A LAP is prohibited as soon as it is applied systematically or when the trader is not given the opportunity to justify a lower price.
The Court thus makes it clear that the use of a LAP is not prohibited per se, but also that it must never become a hard lower limit that automatically replaces the transaction value.
A LAP may therefore be used, but only within the fallback method, only as a last resort, and only if the importer is given the opportunity to explain his position. The Court thus leaves room for digital control, but warns that automatism undermines the system.
The ruling is controversial because it exposes precisely where supervision and legal protection intersect. Customs want speed, scale and data. Companies want predictability and individual assessment. The Court ruled that both interests are legitimate, but digitisation should not lead to the simplification of complex trade realities.
A low price is not prohibited. Deviating from the average is not proof. Statistics are a tool, not the truth. In doing so, the Court reminds us that customs law is more than just a spreadsheet.
Another crucial point is the period over which statistics are collected. In these cases, customs worked with data covering a period of 48 months. The Court considers this to be problematic.
It refers back to the 90-day rule from the deduction method and states that even with the fallback method, customs must work with information from the same period, or virtually the same period, as the imports under investigation. Statistics from several years ago are only permissible when there is no other option, and customs must explicitly substantiate this.
Here, too, a clear warning is sounded: statistical data should not be used thoughtlessly. It remains essential that the data be relevant to the time period. The customs value of the goods must be determined as accurately and truthfully as possible.
One of the most important parts of the judgment, and the part that is likely to have the greatest impact in practice, is the Court's strong emphasis on the customs authorities' obligation to state reasons.
The Court is strict. If customs authorities wish to deviate from the transaction value, they must:
Not as a formality, but because the valuation system is based on hearing both sides of the argument, transparency and individual assessment. The Court considers that if customs wish to use statistics, it must explain why the specific case justifies this.
The judgment does not change the fact that companies must carefully substantiate how their customs value is determined. However, it does change the way in which they can defend themselves when a customs authority uses the Fair Price List. In practice, this means that if customs authorities want to replace the customs value with a statistical average because they have doubts about the declared transaction value, that is not the end of the matter. Companies can still:
Documentation is more important than ever. Especially in the case of low-quality, unbranded bulk goods, it is wise to thoroughly document aspects such as the following:
In practice, we see a growing need for 'value governance': internally recording how prices are determined and which commercial factors play a role.
The Court provides companies with an important tool: the fact that the value of their imports deviates from the average is not in itself proof of undervaluation. This brings the discussion back to where it belongs: the transaction value and the story behind it.
The importers in these cases had made use of the simplification whereby all goods are declared under the highest tariff heading. Interestingly, the Court confirms that this choice also has an impact on subsequent checks. This means that an importer who opts for simplicity also bears the consequences of this in subsequent valuation discussions.
In addition to the valuation issue, the Court dealt with a separate question of who is liable for VAT on importation.
According to Article 201 of the VAT Directive, Member States may determine who is liable for VAT on imports. However, this designation must meet the requirements of clarity, precision and legal certainty: a person concerned must be able to deduce from national legislation that he is liable for VAT.
In the Greek cases, customs duties were not at issue, but VAT on imports was. The Greek customs authorities argued that the persons concerned, including the importer and an employee within the commercial chain, could be regarded as the 'owner of the goods' and therefore liable for VAT. The Court clarified that this is only possible if national legislation expressly and unambiguously determines who is considered to be the owner of the imported goods for the purposes of paying import VAT.
Although the judgment is embedded in a case in which Greek customs suspected a smuggling network — the Court explicitly states that the authorities saw such indications in a set of 289 incorrect declarations — this is not unrelated to a broader reality in customs practice. In that sense, the picture that emerges in Keladis is not representative of all value disputes.
In daily practice, we see many cases in which there is no smuggling, intent or fraudulent undervaluation, but in which customs nevertheless decide to reject the declared transaction value. This often happens after a 'doubt procedure' has been completed and customs conclude that the value is 'not plausible'.
In such situations, the authorities fall back on statistical thresholds such as a Fair Price List or a comparable 'reference value'. However, in many cases, this list is not transparent, traceable or verifiable for interested parties. This creates a legally vulnerable situation: customs authorities rely on data that market participants cannot access, while that same data forms the basis for rejecting the transaction value.
This also causes problems for customs representatives in practice. In the verification phase, representatives are often the first to be approached, even though they usually do not have access to commercial details such as price agreements, discounts, remnants or market developments that are relevant to the valuation. The risk is that they will find themselves in an impossible position in the proceedings: they have to explain why a value is correct, while they do not have access to the data on which customs base its rejection of the value.
That is why the reminder from Keladis is important: customs must be transparent about its reasoning, and statistics should never automatically be the deciding factor.
Each file requires an individual assessment, with room for the story behind the value, the commercial context and the specificity of the goods.
In the valuation practice of textiles, there is another phenomenon that has increased significantly in recent years, particularly due to international e-commerce. Whereas the traditional clothing industry used to work with seasonal collections and relatively stable volume pressure, we are now seeing a development described as 'ultra-fast fashion': continuous product renewal based on rapid trends, short lifespans and mass production cycles.
Brands and suppliers deliberately produce in very large volumes, often in dozens or hundreds of variants of the same basic model. They 'gamble' on which designs will be a hit – the rest are left unsold. Items that do not sell are sold at knock-down prices, discounted immediately or even end up as unsold stock in the waste stream.
Commercially, this is a strategy: the losses on the majority of variants are offset by a few successful models. Legally, however, this leads to a difficult reality: as a result, many textiles actually have an extremely low value, sometimes even below cost price, in order to make room for new collections.
Whether this economic dynamic is desirable – also in terms of sustainability and waste streams – is a discussion in itself. But it does explain why some textile shipments are objectively very cheap. For customs practice, this means that low customs values are not automatically a sign of fraud but can also be the result of structural overproduction and short product life cycles in the modern clothing chain.
Keladis I & II is not a spectacular ruling, but it is a fundamental one. It reminds us that customs law – however digital and data-driven it may be – remains a system that revolves around context, proportionality and hearing both sides of the argument.
The Court gives customs the leeway to use statistics, but ensures that the valuation system does not turn into mathematical automatism.
At a time when algorithms are increasingly determining where customs' attention is focused, this may well be precisely the nuance that is needed.
Ploum’s customs team supports companies on a daily basis in discussions about customs value, undervaluation, post-clearance audits and VAT on imports. Jikke Biermasz and Arjan Wolkers regularly advise and litigate on the application of valuation methods, the requirements arising from European case law and the burden of proof in customs cases.
With in-depth knowledge of the UCC, broad litigation experience and guidance during company audits, Ploum helps companies keep control of valuation risks and strengthen their position in discussions with customs authorities.
Contact
03 Feb 26
28 Jan 26
19 Jan 26
15 Jan 26
13 Jan 26
13 Jan 26
05 Jan 26
31 Dec 25
31 Dec 25
22 Dec 25
16 Dec 25
15 Dec 25
Met uw inschrijving blijft u op de hoogte van de laatste juridische ontwikkelingen op dit gebied. Vul hieronder uw gegevens in om per e-mail op te hoogte te blijven.
Stay up to date with the latest legal developments in your sector. Fill in your personal details below to receive invitations to events and legal updates that matches your interest.
Follow what you find interesting
Receive recommendations based on your interests
{phrase:advantage_3}
{phrase:advantage_4}
We ask for your first name and last name so we can use this information when you register for a Ploum event or a Ploum academy.
A password will automatically be created for you. As soon as your account has been created you will receive this password in a welcome e-mail. You can use it to log in immediately. If you wish, you can also change this password yourself via the password forgotten function.
