Lin is reaffirmed

With its judgment of 16 July 2026 in Lin II, the Grand Chamber of the ECJ held that Romanian courts must disapply a national standard of protection requiring the retroactive application of a more lenient limitation regime to cases of serious fraud affecting the EU’s financial interests.

The Court had already reached this conclusion in Lin, whose central holding prompted a further preliminary reference following difficulties and resistance in its application by national courts.

Despite the AG’s suggestion that the Court reconsider its first ruling, as it had done in Taricco II, Lin was reaffirmed. Both the judgment and the scholarly debate have sought to justify the different treatment of the two cases. Here, I will argue that the grounds for that distinction are not entirely convincing and that Lin and Taricco II cannot ultimately be reconciled. In Lin, the Court simply returned to the original logic of Taricco, having learned its lesson about leaving national authorities a margin of discretion when disapplying constitutional standards of protection in criminal matters. We may now be witnessing the consequences of leaving them with none.

Lin and its Reception by Romanian Courts

The Lin saga arose from a complex interaction among Romanian constitutional case-law, ordinary case-law, and positive law¹. In 2018, the Curtea Constitutională held that the statutory rule under which any procedural act could interrupt limitation periods for criminal liability – including an act that had never been notified to the suspect or defendant – violated the constitutional principle that criminal law must be precise and foreseeable. In 2022, it clarified that its 2018 decision had left Romanian law without any ground for interrupting limitation periods until the enactment of new legislation (“Precision-and-Foreseeability Standard”). Because the Romanian legal system treats rules governing limitation periods as part of substantive criminal law, the Înalta Curte de Casatie si Justitie subsequently held in a 2022 ruling that the lex mitior principle required the effects of the constitutional rulings to be applied retroactively. This entailed applying the no-interruption regime to all pending criminal proceedings and, in principle, even to final convictions, given the possibility of bringing extraordinary appeals against final decisions that did not explicitly address the issue (“Lex Mitior Standard”).

Against this background, Lin recognized a systemic risk of impunity for offenses affecting the EU’s financial interests. In balancing the obligation to counter serious fraud against the national standards of protection, the ECJ considered the risk arising from the Precision-and-Foreseeability Standard tolerable, but not its “exacerbation”² by the Lex Mitior Standard. It therefore held that Art. 325(1) TFEU and Art. 2(1) of the Convention drawn up on the basis of Art. K.3 TEU on the protection of the European Communities’ financial interests precluded the application of the ordinary case-law enshrining the Lex Mitior Standard.

The application of Lin raised numerous issues and divergences in the case-law, which the Înalta Curte de Casatie si Justitie was eventually called upon to resolve. Specifically, in Decision No. 37/2024, the Completul Pentru Dezlegarea Unor Chestiuni De Drept În Materie Penală, after ruling out the need to request either an advisory opinion from the ECtHR³ or a new preliminary ruling from the Court of Justice⁴, held that national courts may not, in fact, disapply the Lex Mitior Standard for three main reasons: (1) the application of Lin does not afford suspects and defendants the level of protection guaranteed by Art. 7 ECHR with respect to the lex mitior principle⁵; (2) the concept of “systemic risk of impunity” is insufficiently precise⁶ and emerged, in any event, after the commission of the offenses, thus operating ex post facto⁷; (3) in the absence of sufficiently precise, predetermined legislative criteria, the constitutional principle of separation of powers precludes the judiciary from interfering with the applicability of the law⁸. The same arguments were later reiterated by Decision No. 16/2024 of the Completul pentru soluţionarea recursului în interesul legii, in which (4) the separation-of-power argument is also articulated as a prohibition against creating a lex tertia by disapplying the limitation regime only in part or only with respect to certain subjects or offenses⁹.

In view of the binding nature of these decisions, a dissenting panel of the same Înalta Curte de Casatie si Justitie requested a preliminary ruling to clarify the concept of “serious fraud”¹⁰ and, more importantly, whether national courts were required to disapply the Lex Mitior Standard where (1) disapplication would result in a level of protection neither equivalent nor comparable to Art. 7 ECHR; (2) under national case-law, the limitation period for criminal liability would have been considered to have expired before Lin was delivered; (3) national law laid down no specific criteria for assessing the systemic risk of impunity; and (4) disapplication would violate the constitutional prohibition against creating a lex tertia.

The Reasons for reaffirming Lin

Lin II responds to the decisions of the Înalta Curte de Casaţie şi Justiţie that had considered Lin inapplicable. Significantly, the Court characterizes those decisions as violations of the obligation imposed on courts of last instance to make a preliminary reference under Art. 267(3) TFEU¹¹. In this regard, Lin II reiterates that the autonomy and effectiveness of EU law require a national court to make a preliminary reference where it has reason to doubt the validity of an EU act or its interpretation by the ECJ, even where that doubt concerns “requirements stemming from Article 7 ECHR or […] requirements of constitutional rank specific to its national law”¹².

In the case at hand, the ECJ argues – correctly, in my view – that (1) the alleged conflict with Art. 7 ECHR, which operates as a minimum threshold of protection pursuant to Art. 52(3) CFR¹³, rests on a reading that unduly extends the scope of the ECtHR’s case-law¹⁴. The ECtHR considers the lex mitior principle derived from Art. 7 to be generally applicable only to provisions defining offenses and penalties, rather than to procedural laws and, in particular, rules governing limitation periods¹⁵. As far as limitation periods are concerned, Art. 7 only bars their extensions after they have expired, because such an extension would retroactively reactivate criminal liability¹⁶. By contrast, extensions adopted before the limitation period expires are, in principle, admissible¹⁷.

This is why Lin II clarifies that (2) Lin did not have the effect of reactivating extinguished criminal liability. Not only had the EU law provisions requiring the disapplication of the Lex Mitior Standard entered into force before any offense could have become time-barred under that standard, but their interpretation was also, according to the ECJ, sufficiently foreseeable. On the one hand, given the declaratory rather than constitutive nature of the ECJ’s preliminary rulings¹⁸, the Lex Mitior Standard could not have been applied to cases of serious fraud affecting the EU’s financial interests from the moment those EU law provisions became part of the Romanian legal order, between 2008 and 2009¹⁹. On the other hand, the obligation to disapply rules governing limitation periods was clarified, at the latest, by Taricco in 2015²⁰. Moreover, Taricco itself constituted a reasonably foreseeable development of the case-law predating Romania’s accession to the Union, which already mandated Member States to penalize serious infringements of EU law under procedural and substantive conditions that rendered the penalties effective, proportionate, and dissuasive²¹. Romanian courts are thus required to apply Lin in all cases, with the sole exception of cases in which a decision having the authority of res judicata has already definitively found that the limitation period had expired²². Such a decision need not be reopened unless Romanian law permits comparable final decisions to be reopened to ensure compliance with national law²³, potentially through an extraordinary appeal²⁴.

The ECJ also firmly rejects the concerns relating to the principle of separation of powers. (3) Lin II first reaffirms that Romanian courts were never required to assess whether there was a systemic risk of impunity for perpetrators of serious frauds affecting the EU’s financial interests. The Court itself conducted that assessment in Lin²⁵, on the basis of the factual material before it²⁶, and through a foreseeable interpretation of the concept in light of Taricco and Euro Box Promotion²⁷. (4) Nor did the Court require national courts to combine provisions from successive pieces of legislation to create a new, judge-made rule for calculating limitation periods²⁸. It merely prohibited them from applying the no-interruption regime resulting from the Curtea Constitutională’s decisions retroactively to procedural acts performed before 2018. Admittedly, that prohibition applies only to cases of serious fraud affecting the EU’s financial interests. This limitation, however, follows from the material scope of application of EU law²⁹ and does not grant national courts any greater lawmaking power than their settled authority to disapply national law that conflicts with directly effective EU law.

Back to Taricco

The ECJ goes to great lengths to distinguish the case from Taricco II, which the AG had identified as analogous in his Opinions in both Lin and Lin II³⁰. According to the Court, the distinction ultimately rests on two decisive factors. First, while the national standard discussed in Taricco II involved “requirements of precision, foreseeability and non-retroactivity […] [which are] of specific importance in EU law,” the national standard at issue in Lin “involves, on the contrary, the retroactive application of certain rules”³¹. Second, “unlike in the situation giving rise to the judgment in [Taricco II], the assessment of whether the [Lex Mitior Standard] entails a systemic risk of impunity […] does not lead to uncertainty in the national legal order which is liable to be in breach of […] the principle that the criminal law applicable must be precise”³².

In the scholarly debate, the first ground of distinction has been endorsed by Peirone, who argues that the principle of precision and foreseeability in criminal law and the lex mitior principle rest on foundations of a different order: the former is a peremptory and indispensable principle, an essential requirement of the rule of law, whereas the latter derives from the principles of reasonableness and proportionality and therefore allows a broader margin of political discretion in its application, especially beyond its core content concerning cases of abolitio criminis³³.

However, in the context of Lin, the sharp distinction on which the Court relies between the Precision-and-Foreseeability Standard and the Lex Mitior Standard remains somewhat formalistic and artificial. From a substantive perspective, disapplying the Lex Mitior Standard effectively means applying, in pending – and possibly even finally decided – cases, rules that were found to be inconsistent with the very requirements of precision and foreseeability. Logically, then, the Court appears to be arguing simultaneously that the principle of precision and foreseeability justifies a systemic risk of impunity and that it does not.

Suppose the facts of Lin were to arise again in a jurisdiction, such as Italy, where constitutional-court judgments of invalidity generally have retroactive effects. In that context, the invalidation of a limitation regime resembling the Romanian one would, by default, deprive both past and future procedural acts of any interrupting effect, without implicating the lex mitior principle at all. To me, this makes it quite clear that the relevant issue is really the retroactive or prospective application of the Precision-and-Foreseeability Standard – and its consequences for the magnitude of the systemic risk of impunity – rather than the substantive foundations of the Lex Mitior Standard.

The truly distinguishing factor is thus the second, and it depends entirely on the Court’s strategic choice in Lin not to entrust national courts with assessing the existence of a systemic risk of impunity, as it had done in Taricco. Like the AG, I see no other substantive differences. The Lin saga therefore marks a discontinuity in the case-law, despite the ECJ’s attempt to conceal it behind the distinction between the two standards. Put simply, by assessing the risk of impunity itself, the Court avoided the mistake it had made in Taricco, which ultimately led it to allow Italian courts not to disapply a limitation regime too lenient to ensure effective and deterrent penalties³⁴.

Had the Court itself established the existence of that risk at the time and reasoned as it did in Lin II, the Italian 2005 reform of limitation periods – much like the Romanian no-interruption regime – should have been regarded as inapplicable from the outset. The EU law obligations requiring its disapplication were already in force, and their meaning was foreseeable in light of the pre-2005 case-law, discussed above, on the duty to ensure effective, proportionate, and dissuasive penalties. Accordingly, no question of retroactive application of a harsher criminal rule would have arisen at all.

Having learned its lesson, I argue, the Court in Lin II returns to the original logic of Taricco to consolidate the tenet that, despite a long narrative of exceptionalism³⁵, national constitutional standards in criminal law are no exception to the primacy, unity, and effectiveness of EU law. The Treaties and directly effective EU law can, in principle, operate both to the advantage and to the disadvantage of suspects and defendants – just as they can operate both to the advantage and the disadvantage of victims and the public. In cases concerning the obligation to counter serious fraud affecting the financial interests of the Union, higher national standards of protection may operate, in practice, only when the Court can reconstruct them as part of EU law³⁶, so that the balancing takes place primarily at the EU level.

From a national constitutional perspective, this is far more troubling a development than could have been anticipated after Taricco II. Before resorting to scholarly or judicial resistance, however, one had better ascertain whether national law itself recognizes any exceptions to the standard of protection at issue³⁷. Even if it does not, one should ask whether every national standard of protection is truly worth defending or whether relaxing some of those standards in the construction of a European constitutional discourse on punishment might be preferable.

For example, it seems to me that the idea that under no circumstances may limitation periods be extended ex post facto – endorsed both by the Italian and Romanian judiciaries in their respective cases – risks unreasonably frustrating countervailing constitutional interests, particularly where exceptional circumstances disrupt the ordinary functioning of the judicial system and give rise to a systemic risk of impunity. After all, during the COVID-19 emergency, the Italian Corte costituzionale itself upheld the constitutionality of legislation suspending criminal proceedings and, consequently, limitation periods, including for offenses committed before the suspension was introduced³⁸. This resulted, in practical terms, in a short ex post facto extension of those periods³⁹. Nothing comparable, of course, to the consequences of the disapplication required by Taricco and Lin. But perhaps still an example worth bearing in mind before embarking once again on a principled discussion about what a political community regards as absolutely nonnegotiable.

---

¹ See P.F. Bresciani, Tutela degli interessi finanziari dell’Unione ed effetto diretto in malam partem in materia penale: una nuova stagione romena per la saga Taricco?, in Review of European Litigation, 2023, p. 214 ff.; F. Peirone, Time Clash: the Court of Justice Limits the Temporal Scope of Lex Mitior, in EuConst, 2024, p. 593 ff.; A. Navarro, Protecting EU Financial Interests in the Collection of Tax-Based Own Resources, in Intertax, 2025, p. 539 ff.

² Court of Justice, 24 July 2023, Case C-107/23 PPU, Lin, para. 121.

³ Înalta Curte de Casatie si Justitie, Decision No. 37 of 17 June 2024, para. 232 ff.

⁴ Ibidem, para. 243 ff. (the reasoning resembles an acte éclairé argument, especially at para. 251 ff.).

⁵ Ibidem, para. 341.

⁶ Ibidem, para. 354.

⁷ Ibidem, para. 360.

⁸ Ibidem, para. 359.

⁹ Înalta Curte de Casatie si Justitie, Decision No. 16 of 16 September 2024, paras 129 and 132.

¹⁰ In this regard, Lin II (paras 65-96) concluded that, absent a national threshold, an instance of fraud must be classified as “serious” whenever the total amount involved exceeds EUR 50,000, regardless of the actual damage to the EU budget.

¹¹ Court of Justice, 16 July 2026, Case C-280/25, Lin II, paras 231 and 234.

¹² Ibidem, paras 232-233.

¹³ Ibidem, para. 104.

¹⁴ Specifically, the Înalta Curte de Casatie si Justitie, in Decision No. 37/2024, paras 320-321, emphasizes that the ECtHR held that “[l]orsqu’un État prévoit expressément dans sa législation le principe de la rétroactivité de la loi plus favorable . . ., il doit permettre à ses justiciables d’exercer ce droit selon les garanties conventionnelles” (ECtHR, 12 January 2016, Application No. 33427/10, Gouarré Patte v. Andorra, para. 35). However, that proposition concerned provisions reducing the severity of a penalty. Gouarré Patte does not establish that any higher protection under national law becomes part of the Convention standard, particularly when that national standard could never have applied due to a sufficiently foreseeable interpretation of directly effective EU law, as the ECJ holds in Lin II.

¹⁵ Court of Justice, Lin II, cit., para. 125 ff.

¹⁶ Ibidem, para. 191. Cf. also ECtHR, 26 April 2022, Request No. P16-2021-001, Advisory opinion on the applicability of statutes of limitation to prosecution, conviction and punishment in respect of an offence constituting, in substance, an act of torture, paras 73-77.

¹⁷ Court of Justice, Lin II, cit., para. 189.

¹⁸ Ibidem, para. 199.

¹⁹ Ibidem, para. 195.

²⁰ Ibidem, para. 213.

²¹ Ibidem, para. 210.

²² Ibidem, paras 235-236.

²³ Ibidem, paras 178-181.

²⁴ Ibidem, para. 27.

²⁵ Ibidem, paras 168-169.

²⁶ Ibidem, para. 166.

²⁷ Ibidem, para. 172 ff.

²⁸ Ibidem, para. 154.

²⁹ Ibidem, para. 160.

³⁰ See Opinion of AG Campos Sánchez-Bordona, 29 June 2023, Case C-107/23 PPU, Lin, especially points 144 and 148, and Opinion of AG Campos Sánchez-Bordona, 18 December 2025, Case C-280/25, Lin II, especially points 64, 107, and 125.

³¹ Court of Justice, Lin II, cit., paras 221 and 223.

³² Ibidem, para. 169.

³³ F. Peirone, Time Clash: the Court of Justice Limits the Temporal Scope of Lex Mitior, cit., pp. 606-607.

³⁴ Court of Justice, 5 December 2017, Case C-42/17, M.A.S. and M.B. [Taricco II], para. 59.

³⁵ See F. Viganò, The Remains of Exceptionalism in Criminal Law, in Criminal Law and Philosophy, 2023, p. 71 ff. I have also tried to argue against criminal-law exceptionalism from the perspective of constitutional theory in P.F. Bresciani, La costituzionalizzazione del potere di punire, Bologna, 2026, p. 213 ff.

³⁶ Cf. A. Navarro, Protecting EU Financial Interests, cit., p. 548.

³⁷ Cf. Court of Justice, Lin II, cit., para. 60, reporting that Romanian law recognizes exceptions to the application of the Lex Mitior Standard, without specifying them – an odd basis on which to raise a counter-limits argument.

³⁸ Corte cost., Judgment No. 278 of 2020.

³⁹ Adopting a rather formalistic approach, the judgment held that the COVID-19 suspension fell within a blanket statutory ground for suspending limitation periods already established by Art. 159 of the Criminal Code before the offenses were committed.