By Antonio Ruiz Arranz

On Spanish Supreme Court Judgment 161/2026, an interest‑free loan, and the judges of Berlin

Several months ago, thanks to J. Alfaro’s blog, I came across an interesting judgment of the German Federal Court of Justice (Bundesgerichtshof, BGH) of 9 May 2023. Alfaro commented on it under the title “Negative interest? A BGH judgment,” drawing on an article by Moritz Renner published in the Juristen Zeitung.

The case decided by the BGH concerned a loan of 200 million euros to the State of North Rhine‑Westphalia, at a variable interest rate and without a minimum rate. When Euribor fell below zero (2015–2016), the borrower claimed 160,000 euros from the bank in respect of negative interest. The BGH rejected the claim outright, pointing out that negative interest is not, in legal terms, interest. However, both Renner and Alfaro agreed that the cleanest solution lay elsewhere: in the supplementary interpretation of the contract. Acting in good faith, the parties would never have agreed that the bank should pay interest to the borrower.

That anecdote stayed with me. If supplementary interpretation so easily resolves the case of negative interest, should it not also resolve one of the most uncomfortable questions of European consumer law in recent years, namely the consequences of the nullity of an unfair term? The recent Supreme Court Judgment (STS) 161/2026 of 4 February confirms that this question is well posed. Once the IRPH clause was removed from a twenty‑five‑year mortgage loan, the First Chamber concluded that the loan had to survive without any remunerative interest whatsoever, leaving the consumer with an interest‑free loan for a quarter of a century.

It is worth pausing over this result, because everything that follows in this piece rests on the conviction that we should never have arrived at it. The case also provides an opportunity to reconsider the limits of Article 6(1) of Directive 93/13—a provision drafted in 1993 following the German model of the 1976 AGBG, conceived for ancillary clauses whose removal did not compromise either the structure of the contract or its economic purpose—when unfair clauses affecting an essential element of the contract are annulled.

When the removal of a clause affects an essential element of the contract

The removal of an unfair—or non‑incorporated—clause may create a gap in the contractual regulation which, in certain cases, requires to be filled. Since Kásler (CJEU, C‑26/13, 30 April 2014) and Abanca (CJEU, C‑70/17 and C‑179/17, 26 March 2019), we know that such gap‑filling is permissible only where the removal jeopardizes the survival of the contract to the detriment of the consumer. This doctrine is disputable—F. Pantaleón has pointed this out repeatedly in Almacén de Derecho—but it is undeniably settled case law of the Court of Justice.

Behind this rule lies a tension that is rarely articulated clearly, perhaps because doing so would force one to take sides. Matteo Fornasier has described it with precision in the Zeitschrift für Europäisches Privatrecht (ZEuP 2025, 233): the CJEU’s case law on the consequences of the nullity of an unfair clause oscillates between two objectives that are difficult to reconcile. On the one hand, the restoration of contractual equilibrium between the parties (Äquivalenzsicherung), and on the other, the effective deterrence of traders from using unfair terms (Prävention). According to the Bochum professor, the conflict between these objectives is resolved “as a rule in favour of the deterrent objective and to the detriment of contractual parity.”

This framework works reasonably well when the removed clause is ancillary—a costs clause, a withdrawal penalty, or an unjustified fee. In such cases, removal does not compromise the structure or economic purpose of the contract, and restitution of what was paid under the void clause is the natural consequence. But what happens when the removed clause is not ancillary, but directly determines an essential element of the contract, such as the interest rate, the currency of denomination, or the mechanism for setting the price? Here, giving priority to deterrence entails an implicit risk that Fornasier calls “over‑deterrence” (Überabschreckung): faced with uncertainty about the outcome of content control, traders may be deterred not only from using unfair terms but also from employing lawful clauses that approach the grey area of what is debatable, with the consequent harm to market efficiency and, ultimately, to consumers.

There is also an argument of justice that should never be evaded. In private law, every transfer of value requires a cause. A twenty‑five‑year interest‑free mortgage loan enriches one party at the expense of the other, and one may ask whether the laudable aim of preventing the use of unfair clauses, in itself, justifies that enrichment, especially when the defect that gives rise to it is neither fraud nor bad faith, but a failure of incorporation resulting from the lack of pre‑contractual information.

Remaining at the level of legislative policy would nonetheless be too comfortable. The relevant question is dogmatic—or, if one prefers, one of correct application of the rules currently in force. Was this outcome inevitable? Was the First Chamber obliged to accept a twenty‑five‑year interest‑free mortgage loan, or was there an alternative path that the CJEU’s case law has not yet explored?

The case decided by STS 161/2026

STS 161/2026 does not resolve a problem of unfairness or lack of material transparency, but one of non‑incorporation of a clause into the contract within the meaning of Articles 5 and 7 of the Spanish General Terms Act (LCGC). Although this distinction is relevant—and, in truth, makes the outcome even more unacceptable—I deliberately set it aside in order to focus on the solution adopted by the judgment once non‑incorporation has been declared.

Two consumers were subrogated into a developer’s loan indexed to IRPH‑Entities. Because this was a subrogation without modification of the financial conditions, the bank did not attend the execution of the deed and, according to the findings of fact, the consumer was not provided with information about the interest rate. In particular, no copy of the loan granted to the developer was delivered, no binding offer was drawn up, and the notary did not verify that the borrowers knew the financial conditions of the loan into which they were being subrogated. The pre‑drafted statements in the deed of sale—by which the borrowers declared that they knew and accepted the clauses of the original loan—were characterised by the Chamber as ineffective knowledge clauses. On this basis, the First Chamber declared that the clause setting IRPH‑Entities as the reference index had not been incorporated.

Faced with the fact that removal of the clause would leave the contract without a remunerative interest rate, the Chamber examined three default rules and rejected them all: Article 1108 of the Civil Code, because it applies only to default interest; Article 7 of Law 7/1995, because it would lead to a result worse for the consumer than the non‑incorporated clause itself; and the Fifteenth Additional Provision of Law 14/2013, because it is intended for the disappearance of reference rates, not for the ineffectiveness of remunerative interest due to a defect of incorporation.

Having rejected the candidate rules to fill the gap, the Chamber concluded:

“The effect of the non‑incorporation of the remunerative interest clause into the contract, combined with the prohibition on integrating the contract, leads to the consequence of leaving the loan without remunerative interest.”

To that end, it invoked Article 1740 of the Civil Code (“a simple loan may be gratuitous or with an agreement to pay interest”) and Article 311 of the Commercial Code (“loans shall not bear interest unless expressly agreed in writing”). It added that this solution “is the one that best gives effect” to Articles 5 and 7 LCGC and “most effectively guarantees the principles of non‑binding effect and deterrence” of Articles 6(1) and 7(1) of Directive 93/13. As a result, a mortgage loan with a principal of more than €108,000 and a term of twenty‑five years was left without remunerative interest, and the bank was required to refund all amounts collected as interest, as they exceeded zero per cent.

Critique of STS 161/2026

The reasoning of the judgment deserves to be followed carefully, as it contains a series of contradictions.

The Chamber starts from the premise that total nullity of the contract

“would be particularly detrimental to the consumers bringing the appeal, as it would require them to immediately repay the entire outstanding principal.”

It thus places itself within the factual framework of Kásler, Abanca, and Banca B. (CJEU, C‑269/19, 25 November 2020)—a contract that cannot subsist without the clause, total nullity being detrimental to the consumer—and acknowledges that the consequence envisaged by the CJEU is integration.

Immediately afterwards, however, it states that

“there is clearly no default rule of law that would allow, without engaging in the contract integration prohibited by the CJEU, the absence of the clause to be filled,”

citing paragraphs 92 and 93 of Arce (CJEU, C‑365/23, 20 March 2025).

Those paragraphs do not, however, prohibit integration. They prohibit so‑called “blue‑pencilling” or conservative reduction of validity, a different operation consisting in partially maintaining the clause by trimming its content in order to save it, for example by reducing a default interest rate of 25 per cent to 12 per cent. The CJEU has prohibited this exercise since Banco Español de Crédito (C‑618/10, 14 June 2012) and Asbeek Brusse / de Man Garabito (C‑488/11, 30 May 2013). Integration, by contrast, removes the clause in its entirety and then fills the void left by its disappearance. The CJEU admits it where removal creates a gap, the contract cannot subsist without the clause, and total nullity would be detrimental to the consumer (Kásler, Abanca). The Chamber confuses the two operations.

This confusion reappears in its reading of Kutxabank (C‑300/23, 12 December 2024), the CJEU judgment on IRPH that it probably had in mind when deciding the case. What Kutxabank prohibits is, again, conservative reduction: adding a negative spread to IRPH in order to “correct” it downwards while keeping the clause. What it permits is substitution by a default provision with “a scope equivalent to that of the clause to be replaced”—which, as we shall see, is not what the Chamber applied.

It is in this context that the Chamber looks for default rules to fill the gap—examining Article 1108 of the Civil Code, Article 7 of Law 7/1995, and the Fifteenth Additional Provision of Law 14/2013—which is precisely what integration entails according to Kásler, but finds none suitable. It then resorts to Articles 1740 CC and 311 CCom to declare the loan gratuitous. Yet these provisions are not instruments of integration. They describe the legal type of the loan in nineteenth‑century codes. They allow that a loan may be gratuitous; they do not require it to be so. The function of integration—and this should never be lost sight of—is to reconstruct the regulation that two reasonable parties, acting in good faith, would have agreed upon had they foreseen the gap. Sometimes the legislature does our work for us through a default rule. But where that rule leads to a result that no reasonable observer would identify with such hypothetical will, one cannot speak of integration.

Supplementary interpretation of the contract as an alternativ (Ergänzende Vertragsauslegung) to the interest‑free loan

This sequence of contradictions reveals something that needs to be stated plainly: the Chamber had at its disposal an instrument that would have avoided the interest‑free loan and chose not to use it—or, more precisely, it confused it with another instrument that the CJEU does prohibit. That instrument, as noted at the outset, is supplementary interpretation. It would have required identifying a remunerative interest rate that both parties—consumer and drafter—would, ex ante and in good faith, have accepted as appropriate had they contemplated a scenario without an interest‑rate clause.

More shadows than light: CJEU case law on integration

The most obvious objection to this thesis is that the CJEU already rejected supplementary interpretation in Dziubak (C‑260/18, 3 October 2019), where it held that Article 6(1) of Directive 93/13 precludes integration of the contract by means of “general provisions of national law” referring to principles such as equity or usage “that are not default provisions or provisions applicable where the parties agree” (para. 62). These words need to be put in context. What the Court rejected were Articles 56, 65 and 354 of the Polish Civil Code—general closing clauses referring to standards of conduct without setting out a specific contractual regime—because they had not been subject to a “specific legislative assessment” and did not enjoy the presumption of non‑abusiveness attaching to default rules (paras. 60–61). In other words, the CJEU does not reject supplementary interpretation in the abstract, but requires that the instrument employed offer a guarantee of balance comparable to that of a default rule.

Subsequent case law has not closed this door, but neither has it opened it unequivocally. In Lombard (C‑472/20, 31 March 2022), Gupfinger (C‑625/21, 8 December 2022), AxFina Hungary (C‑705/21, 27 April 2023), Bank M. (C‑520/21, 15 June 2023), and Novo Banco (C‑498/22 to C‑500/22, 5 September 2024), the CJEU reiterated that the unfair clause must be removed without substitution when the contract can subsist without it, and that recourse to default provisions or judicial modification is not permitted unless the strict requirements of Kásler are met. But in all of these cases either the contract could subsist without the clause, or, where it could not—Lombard and AxFina Hungary, concerning exchange‑rate clauses—the attempt was to rely on default rules or judicial alterations of the contractual content without sufficient legal basis.

In the opposite direction, four judgments have softened the rigidity of Dziubak.

In Banca B. (C‑269/19, 25 November 2020), the CJEU accepted that, in the absence of a default rule, the court may invite the parties to negotiate a solution that preserves contractual balance, a model functionally indistinguishable—if not worse—from judicial integration. In Provident Polska (C‑321/22, 23 November 2023), it introduced the principle of proportionality as a criterion for determining the consequences of nullity, implicitly allowing substitution of the abusive payment method with any legally permissible method. In Arce (C‑365/23, 20 March 2025), after prohibiting conservative reduction, it observed that the contract may be maintained “in so far as, in accordance with the rules of national law, such continuation of the contract without the unfair clauses is legally possible” (para. 94), a reference that leaves room for whatever integration mechanisms domestic law provides. And, with regard to IRPH, Kutxabank (C‑300/23, 12 December 2024) allowed substitution by a default provision with “a scope equivalent to that of the clause to be replaced.”

In sum, the picture is ambiguous. The CJEU requires that any integration mechanism offer a guarantee of balance, not amount to moderation of the unfair clause, and respect the principle that the consumer is not bound. A supplementary interpretation that fully removes the clause, reconstructs the contractual element in line with objective good faith, and is anchored in a national rule providing for such interpretation could meet these requirements. But the CJEU has not said so yet.

In defence of supplementary interpretation

In Spain, the normative anchor exists. Supplementary interpretation is enshrined in Articles 1258 and 1287(II) of the Civil Code. And, even more explicitly for this situation, Article 10(2) LCGC provides that the part of the contract affected by non‑incorporation or nullity “shall be integrated in accordance with Article 1258 of the Civil Code.”

Nor is it convincing to argue that supplementary interpretation is incompatible with consumer protection. The CJEU itself has recognised, since Pannon GSM (C‑243/08, 4 June 2009), that the consumer may choose to maintain an unfair clause. If the consumer can choose between the full validity of the clause and the total nullity of the contract, it is difficult to justify that they cannot opt for an intermediate solution that seeks to determine what the hypothetical will of a well‑informed consumer would have been.

There is, moreover, an aspect that is rarely emphasised with the clarity it deserves: supplementary interpretation is the mirror image of the unfairness assessment. To determine whether a clause is unfair, the court asks whether the average consumer would have accepted the contract had they known the clause and its consequences (Aziz, C‑415/11, 2013). Supplementary interpretation asks the question from the opposite side: once the clause has been removed and the gap established, what regulation would a reasonable consumer have accepted ex ante, acting in good faith? That is why it differs from conservative reduction. It does not give the trader their clause back or allow them to benefit from it; it imposes a different regulation, reconstructed in accordance with objective good faith. Nor does it reproduce an imbalance to the detriment of the consumer. On the contrary, it reconstructs a balance destroyed by removal of the clause.

And what about Gutiérrez Naranjo? Restitution versus integration

Suppose that STS 161/2026 had opted for supplementary interpretation. Once the IRPH clause was fully removed, the Chamber would have filled the gap with a remunerative interest rate reconstructed in accordance with the parties’ hypothetical will. Euribor would have been an obvious candidate, as Kutxabank itself acknowledges. That solution would have had a direct effect on restitution: interest payments made would be excluded—or reduced—to the extent covered by the rate resulting from integration.

The attentive reader will immediately raise the objection: Gutiérrez Naranjo (C‑154/15, C‑307/15 and C‑308/15, 21 December 2016). If the CJEU prohibits limiting the restitutionary effects of nullity in time, does this not rule out from the outset any solution that reduces restitution? The answer is no. If, in the hypothesis we are considering, restitution does not arise (or not in full), this is not because its effects are being limited in time, but because integration of the contract prevents those payments from being characterised, ab initio, as undue. Integration precedes restitution, and the interest paid by the consumer would find its causa in the interest rate resulting from integration.

Gutiérrez Naranjo rejected the First Chamber’s decision (STS 241/2013) to limit restitution in time of amounts paid under non‑transparent floor clauses, invoking reasons of economic public policy. But neither STS 241/2013 nor Gutiérrez Naranjo approached the problem as one of integration. The floor clause was removed and the loan continued to operate with the agreed variable rate, without any gap to be filled. The CJEU was not asked whether integration of the contract could affect the scope of restitution. At this point, I cannot resist murmuring that if STS 241/2013 had framed the limitation of restitution as a problem of integration—not restitution under Article 1303 CC—acknowledging that loans with floor clauses usually incorporated lower spreads as a quid pro quo and asking what two reasonable parties would have agreed, the payments would not have been fully refundable either, but not because of a temporal limitation, rather because they would have found their causa in the rate resulting from integration. Gutiérrez Naranjo would have answered a different question. Incidentally, the BGH case on negative interest with which I began was also dated 9 May (2023).

The BGH’s “three‑year solution” and the judges of Berlin

That the thesis defended here is not a purely theoretical exercise is shown by the fact that the German Federal Court of Justice has openly applied supplementary interpretation in the context of Directive 93/13 for more than a decade in cases concerning price‑adjustment clauses in long‑term energy supply contracts.

The problem can be explained simply. A district heating supply contract concluded in 2012 fixes an initial price and contains a clause allowing the supplier to update it. If that clause is ineffective, it disappears with no alternative mechanism: the contract is frozen at the 2012 price. In long‑term contracts this forces the supplier to supply energy at the starting price even if its costs increase. The BGH considers this imbalance objectively untenable. Total nullity would be the alternative, but that would leave the consumer without supply. To avoid both outcomes, the BGH resorts to supplementary interpretation and has developed the Dreijahreslösung or “three‑year solution,” based on §§ 133 and 157 BGB—the general provisions on the interpretation of declarations of will and contracts in accordance with good faith and usage. Under this approach, if the consumer does not challenge a price increase within three years from the first invoice applying it, that revised price becomes consolidated as the contractual price.

The “three‑year solution” has been applied since 2010 and was upheld by the German Federal Constitutional Court in 2011. The BGH has maintained it despite Gutiérrez Naranjo, Dziubak, and even RWE Vertrieb (C‑92/11, 21 March 2013), which expressly concerned such energy supply situations and in which the CJEU stated that the economic consequences for the undertaking do not justify limiting restitution. The BGH reiterated its position in recent judgments (6 July 2022—VIII ZR 28/21 and VIII ZR 155/21—and 25 September 2024—VIII ZR 165/21) and has expressly refused to make a preliminary reference, considering the matter acte clair.

Not all German courts share the same certainty, however. On 10 December 2024, the Berlin Kammergericht (9 U 1087/20) referred a preliminary question to the CJEU (C‑900/24) on the compatibility of the Dreijahreslösung with Directive 93/13. The reference is framed on three levels. Formally, it doubts whether §§ 133 and 157 BGB can serve as a basis for integration under the Dziubak doctrine. Substantively, it points out that the Dreijahreslösung takes as its calculation basis a price reached through the very unfair clause, allowing the professional to retain part of the fruit of its unlawful conduct. And with regard to the Kásler condition, it observes that total nullity does not harm the consumer but the supplier. This is worth emphasising. For the Berlin court, if the contract subsists without the review clause, the consumer simply continues paying the 2012 price; if it is declared totally ineffective, retroactive liquidation obliges them to pay the objective value of the supply (which is above the 2012 tariff but below the price reached under the Dreijahreslösung). In both cases, the consumer is better off without supplementary interpretation.

The CJEU held an oral hearing on 12 March 2026, with judgment expected in autumn this year. This will be the first ruling to address directly the admissibility of supplementary interpretation within the framework of Article 6(1) of the Directive. And it is, as we shall see, the moment for the Court to confront an evaluative contradiction generated by its own case law.

The interest‑free loan as paradox: Bank M. versus integration

All of the foregoing could be summarised as a reasoned defence of supplementary interpretation. It would be dishonest, however, not to note that the CJEU’s own case law has laid a trap for it. Since Bank M. (C‑520/21, 15 June 2023), we know that where a loan contract is annulled in its entirety due to unfair clauses, the credit institution may claim only repayment of the capital, and, where appropriate, default interest from formal demand, but no remuneration for the consumer’s use of that capital—a doctrine later confirmed in mBank (C‑140/22, 7 December 2023), Bank Millennium (C‑756/22, order of 11 December 2023), and most recently Falucka (C‑901/24, 16 April 2026). Annulment thus produces a retroactively interest‑free loan, which is, incidentally, the same outcome that the legal system reserves for usury. Such a result is, once again, difficult to justify from any private‑law premise. But it is what the CJEU has said.

Now, by saying this—and perhaps without realising it—the CJEU has placed itself in a position of axiological contradiction (here I acknowledge my debt to several conversations with F. Pantaleón; see again “On the legal consequence of declaring a non‑individually negotiated clause unfair (I)”, Almacén de Derecho, 2020). Let me try to explain. The Kásler doctrine justifies contractual integration in order to prevent total nullity from harming the consumer, harm that typically materialises in the obligation to repay the outstanding principal. But if, according to Bank M., total nullity means that the bank may recover only the principal, without any remuneration whatsoever for its use, then for any consumer who has sufficient liquidity to repay the outstanding capital—or who can refinance that repayment at a lower interest rate—nullity is economically preferable to any form of integration—whatever it may be—that preserves remunerative interest. Integration thus becomes, paradoxically, an instrument operating against the consumer, unless— as STS 161/2026 did—the contract is “integrated” by converting it into a zero‑interest loan. Taken to its logical conclusion, the Bank M. doctrine hollows out the Kásler test itself. It turns nullity into an attractive option for the consumer and, in doing so, ensures that the precondition for integration will rarely be satisfied. And as accessible refinancing instruments continue to develop, the number of consumers for whom nullity is preferable is likely to increase.

This paradox has no solution within the Kásler architecture. It can be resolved only by abandoning the rule that conditions integration on total nullity being detrimental to the consumer, and by adopting a model in which integration of the contract is the direct—and legally correct—consequence of the contractual gap, regardless of whether total nullity harms the consumer. This is still the model of § 306 BGB. In that scheme, the Kásler question—does nullity harm the consumer?—does not arise, because integration logically precedes the issue of nullity and the consumer cannot evade integration by opting for a more advantageous nullity.

It is precisely this model that the Berlin preliminary reference puts to the test. The reference does not merely submit the § 306 BGB system underlying the Dreijahreslösung to the CJEU’s scrutiny; it also demonstrates that the Bank M. paradox is not confined to mortgage credit. As we have seen, the Berlin court finds that the consumer is better off without integration in all possible scenarios and concludes that the Dreijahreslösung lacks justification within the current architecture of Article 6(1), because integration would protect not the consumer but the supplier who used an unfair clause. If this reading is confirmed in Luxembourg—and there are good reasons to think it will be—the door to integration, including supplementary interpretation, will definitively be closed outside the margins of Kásler. Such a result will be coherent with that doctrine. But precisely because of that, it will also continue to empty it of meaning, by consolidating a model in which—once a clause that decisively affects price is removed—integration places the consumer in a worse position than total nullity, unless the trader is forced to supply energy at the 2012 price until the end of the contract.

By way of conclusion: towards a reform of Article 6(1) of Directive 93/13

The long journey from Kásler (2014) to the judgment that will decide the Berlin Kammergericht’s reference (2026) shows that the problems of Article 6(1) of Directive 93/13 are interpretative, but also structural. The rule was drafted in 1993 for ancillary clauses whose suppression did not compromise the structure of the contract. Thirty years later, the CJEU’s case law has turned it into the sole reference framework for dealing with the removal of clauses that define essential elements or that, without defining them directly, decisively affect their configuration (such as the price‑adjustment clause at the origin of the Berlin reference). STS 161/2026, the growing difficulty of reading the CJEU’s judgments on the provision coherently, the Bank M. paradox, and the latest Berlin reference are symptoms of a dysfunction.

Against this background, Fornasier has noted that the CJEU’s case law on the consequences of nullity remains “in need of clarification.” This is a polite way of saying that the system does not work. In truth, the problem is no longer one of clarification: Article 6(1) has reached the limits of its capacity and needs a reform recognising that where removal of a clause creates a gap in an essential element of the contract, integration in accordance with default law or objective good faith is the primary consequence, without the need to show that total nullity harms the consumer.

There is one last point that perhaps should have opened this piece rather than closed it. Article 6(1) does not speak of deterrence. It speaks of non‑binding effect on the consumer and of the continuation of the contract. It is Article 8ter(1)—according to which “Member States shall lay down the rules on penalties applicable to infringements of the national provisions adopted pursuant to this Directive and shall take all measures necessary to ensure that they are implemented. The penalties provided for shall be effective, proportionate and dissuasive”—that charges Member States with adopting measures, such as fines, that effectively deter the use of unfair clauses. Pantaleón expressed it clearly in the entry cited:

“The well‑informed reader knows perfectly well what Advocate General Juliane Kokott recently recalled in paragraph 62 of her Opinion in Case C‑81/19, NG, OH, of 19 March 2020: ‘According to the case law [of the Court of Justice itself], the removal of the unfair clause is intended to replace the formal balance established by the contract between the rights and obligations of the parties with a real balance capable of restoring equality between them.’”

It is inconceivable to me that the CJEU should continue to maintain the stark contradiction that plainly exists between that dictum and the “Kásler doctrine.” And this all the more so after Directive (EU) 2019/2161 of 27 November 2019 added a new Article 8ter to Directive 93/13, obliging Member States to establish punitive sanction regimes, including substantial fines, for infringements of the national provisions adopted under it. Is it not now reasonable to locate, in this new Article 8ter—in the public law of administrative sanctions for unfair terms—the purpose of preventing the imposition of such clauses on consumers, and to cease locating that function in Article 6(1) itself, which belongs to the private law of unfair terms?

Thomas Pfeiffer has now also pointed this out in his AGB‑Recht Kommentar, 8th ed. 2026: the CJEU has prioritised deterrence over the principle of preserving contractual balance within the scope of Article 6(1). Loading the entire deterrent function onto a provision designed to restore contractual balance turns it into a punitive instrument. If the Directive needs more deterrence, the place to seek it is Article 8ter, not Article 6.

* AI translation of Los límites del art. 6.1 Directiva 93/13