Written by Christina Peristeridou and Jannemieke Ouwerkerk. Cross-posted on the Verfassungsblog

The recent CJEU judgment in M.A.S., M.B. (hereinafter Taricco II) raises more questions than it answers on when Member States can apply higher standards of rights in criminal proceedings. Previous case law, i.e. Taricco I and Melloni, pervaded the primacy of EU law, but from Jeremy F. we also know that Member States enjoy a margin of discretion to apply their own standards of fundamental rights protection, where the rules at stake – an appeal suspending the execution of a European arrest warrant – have not been harmonised by EU law. In Jeremy F., it concerned a limited discretion; Member States must ensure that the application of national standards will not frustrate the application of EU law.

Such a discretion was not allowed in Melloni, as the issue at stake – whether surrender could be made conditional upon retrial when the conviction was handed down in absentia – has exhaustively been regulated by EU law. By making surrender conditional upon Melloni receiving a new trial in Italy, Spain would have applied a higher standard of protection than in EU law, which would have resulted in a refusal of surrender; after all, Italian law does not provide for the possibility of retrial after convictions in absentia. The Court required such a drastic primacy of EU law also in Taricco II’s 2015 predecessor, Taricco I. There, the Court has held that the Italian limitation periods for serious VAT fraud cases breached Article 325 TFEU, because in a significant number of cases they prevented de facto the imposition of effective and dissuasive penalties; in many of such (often complicated) cases, time has appeared too limited to bring the case through to completion. Furthermore, since in relation to cases of fraud affecting the national financial interests, Italian law provided for longer limitation periods, there was also a violation of the principle of assimilation set out in Article 325(2) TFEU. According to the Court, the primacy of EU law required the Italian court to disapply the national rules on limitation periods. It firmly dismissed the objection that this would entail an infringement of the principle of legality enshrined in Article 49 (1) Charter. In Taricco I, the argument was that Article 49 applies only to the definition of criminal offences and penalties and not to limitation periods (Taricco I, para. 54-57).

It is precisely this feature of the Italian legality principle (which applies also to limitation periods) that is now accepted in Taricco II, though without altering the scope of Article 49 Charter. The Court thus permitted the Italian courts to apply the national standard of protection (the national legality principle), even if it means higher standards, and even if it comes at the detriment of the effectiveness of EU law; indeed, many criminal proceedings of VAT fraud affecting the EU’s financial interests would be time-barred as a consequence.  

Almost immediately after the delivery of the judgment, comprehensive and interesting blog posts have sprung out, assessing the judgment’s significance from a constitutional law perspective (e.g. see here, here and here). We consider Taricco II a fine specimen of a pluralistic attitude of constitutional conflicts, manifested not only in the ruling’s result, but also in the softer tones and more engaging language in the Court’s reasoning (paras 22-23). However, our approach in this commentary shall stem primarily from a criminal law point of view. And, we must admit that from that perspective,  the Court’s reasoning is anything but clear. In the following remarks, we highlight the complicated questions that this judgement raises from a criminal law perspective, and search for possible explanations for the Court’s thought-process.

I. How to position the judgment?
A first complicated question is how to position the judgment against the background of existing CJEU case law in the criminal law domain. At first reading, one might easily think that Taricco II expresses a novel counter limit to previous case law on the primacy of EU law over higher national standards of rights in criminal proceedings. From that point of view, it might surprise that the Court referred neither to Article 53 Charter, nor to Melloni, which it now may seem to have reversed. After all, both the Spanish rule on a retrial after a conviction in absentia and the Italian legality principle would compromise the effectiveness of EU law in the respective fields of criminal law, but while the Spanish court had to disapply its national rule on retrial, the Italian court has now been allowed to apply its national standard of legality. A closer reading though raises the question of how comparable they actually are. After all, whereas the higher national standard in Melloni obviously falls under the ambit of procedural criminal law (and the same applies for Jeremy F.), the categorisation of limitation periods as substantive or procedural is far from settled. Member States have different views on that and it is quite understandable why the CJEU did not want to have an opinion. Be that as it may, in Italian criminal law, limitation periods belong to substantive criminal law and, thereby, qualify for protection under the national legality principle. As a consequence, the issue of whether limitation periods may be retroactively extended becomes a much more sensitive and difficult issue – in particular if one holds the opinion that the length of such periods shapes the scope of criminal offences, since they determine whether a person can, or cannot, be tried for the criminal offence at hand.

Given these diverging facts and circumstances between Melloni and Taricco II, we find it problematic to qualify the Court’s ruling in the latter as a novel counter limit to the first. Both judgments concern the same question (whether national and higher standards of rights can be applied in EU related issues) but circumstances are not obviously comparable. The only conclusion we feel ready to draw from juxtaposing the two rulings is that the categorisation of the topic as procedural or substantive law appears relevant in deciding whether higher national legal standards could be applicable. That the level of harmonisation is decisive is already a given by prior case law, e.g. Jeremy. F.

II. Substance of the judgment?
Turning now to the substance of the judgment, whereas the Court’s conclusion is crystal clear, we struggle to understand the reasoning and thought-process of getting there. The Court’s argument goes as follows: the European legality principle (Article 49 (1) Charter) does not apply to limitation periods – and therefore Taricco I stands. But because the limitation periods have not been harmonised (para 44), national courts can apply their own limitation periods. So far so good. But here comes puzzling para 45: by applying those substantive (for Italy) rules on limitation periods, Italy is entitled to also apply the principles that usually apply to all Italian substantive criminal law, i.e. the legality principle.

In our view, this line of argumentation is confusing, if not erroneous, for the following reasons. First, rules on the so-called general part of criminal law (e.g. attempt, mens rea – and even not clearly substantive law rules e.g. jurisdiction) have not been harmonised. Typically, Directives harmonise offence definitions and sanction levels, not general part concepts. And yet the Charter still applies when the overall case (usually the offence) falls within EU law. Now Taricco II is confusing in that it makes us wonder: Should these general part concepts first be harmonised for the (protection under the) Charter to be applicable even if the offence already falls under the scope of EU law? The judgment is far from clear on that.

Furthermore, we ponder on whether the substantive law nature of limitation periods is now a relevant criterion. Although, the Court did not rule on the nature of limitation periods, it did include in the ruling the phrase “forming part of national substantive law” (para 62). Is it required that national legal systems classify limitation periods as substantive criminal law for Taricco II to apply? It goes without saying that Taricco II is irrelevant for legal systems that do not apply the legality principle on limitation periods altogether.

And there is also the interpretation of Article 49(1) Charter, which leaves a bitter aftertaste. We expected a transparent and convincing explanation regarding the Court’s choices on why the European legality principle should, or should not apply on limitation periods, since in the past the Court has already shown readiness to go beyond the ECHR standards (e.g. in Berlusconi). But neither in Taricco I, nor in II, such an explanation has been provided.

Anyhow, whilst the scope of Article 49(1) Charter remains the same as determined in Taricco I, the ruling includes some interesting points about the legality principle in European criminal law.

First, unlike Taricco I, the Court engages into a deeper analysis of the principle (even if only meant for pure Italian-consumption). This is not achieved by merely articulating the principle’s other aspects about foreseeability – nothing new there (paras 52-57). It is the context-sensitive approach that we appreciate: the Court takes the suspect’s perspective in Italian proceedings. It ponders on whether suspects would indeed face unforeseeable prosecution in Italy, if Italian courts remained faithful to Article 325 TFEU (para 59-60).

Second, the Court revisits (at least in part) the rather curious obligation found in para 58 of Taricco I, i.e. Italian courts must assess whether the application of the legality principle would hamper prosecution in a ‘significant number of cases’. That meant that the principle’s scope would depend on the number of cases it affected. Needless to say that measuring a principle’s scope based on the ‘damage’ it inflicts on crime policy does not mirror a genuine commitment to legal principles. In addition, it fosters nothing but uncertainty for suspects and national courts on when extensions of limitation periods apply retroactively. The Court, thankfully, clarified that such uncertainty should not be tolerated (para 59).

Third, a surprising move was to limit the temporal scope of Taricco I to offences occurring after the date of the judgment (para 60). This resembles the so-called ‘prospective overruling’, i.e. a limitation of the temporal effects of judgements ex ante. The EU has only used it outside criminal law (e.g. Defrenne). Only a few systems (and mainly the USA but also rarely the Dutch Supreme Court) employ it as compromise: courts can amend the law, without harming foreseeability in pending cases. If Taricco II fashions a ‘non-retroactivity of CJEU’s jurisprudence’ (next to non-retroactive of statutes), then European criminal law resembles much more a common law system than we thought. Perhaps, however, this case is exceptional, and we must resort to the saying ‘hard cases make bad law’.

Fourth, the reasoning echoes the separation of powers. Much of the trouble in these cases results from exaggerated expectations laid upon national courts to ‘cure’ national legislation from any incompatibilities with EU law, thus putting a strain on the legality principle and defying the boundaries between interpretation and law-making. Indeed, Italian courts have received a lot of pressure over the years, given the immobility of the Italian legislature. The Court is forgiving and somewhat sympathetic to the Italian courts for favouring the legality principle over the effective application of EU law. The principle of legality should mirror a balancing of the trias politica; and the Court sets healthy limits here by throwing the ball to the Italian legislator (para 61).

III. Many unclarified aspects and the choises made
The many unclarified aspects in Taricco II and the somewhat opaque reasoning make one wonder about the choices made in Taricco II and, most importantly, the motivation lurking beneath.

First of all, the Court’s ruling in Taricco II could be understood as a follow-up to a trend that has started only recently with Aranyosi/Căldăraru. The Court’s ruling there confirmed that the effectiveness-based principle of mutual recognition has gradually grown depended on the standards of fundamental rights. Ever since the birth of the European Arrest Warrant, many had been pleading the Court to take seriously the balance between crime control and due process. Balancing these antithetical interests forms the intrinsic character of (EU) criminal law. In Aranyosi/Căldăraru, this tension translates into effective application of the European Arrest Warrant (crime control) and proper detention conditions (due process), whilst in Taricco II it translates into the effective protection of EU’s financial interests (Article 325 TFEU - crime control) and the principle of legality (due process). And in our view, Taricco II could be understood as re-affirming what the Court has started in Aranyosi/Căldăraru: fundamental rights can form actual tangible limits to the effectiveness of EU law, which is not treated anymore as a prevailing mantra.

A second possible explanation is more political. We cannot help but wonder whether the desire for ‘constitutional peace’ was not the main motivation here. This would explain the somewhat awkward reasoning which does not make perfectly sense on points of law, but it does in other ways, which could be more important here.

And last but not least, we cannot exclude the possibility that the CJEU just did what it does best: being pragmatic. Much noteworthy to mention in this regard is that pursuant to a recently adopted Directive 2017/1371 on criminal sanctions in the field of EU-fraud (the so-called PIF-Directive), Member States will have to comply by 6 July 2019 with harmonised minimum rules on limitation periods for EU-fraud, including VAT fraud. But what is more, these minimum standards on limitation periods (see Article 12 PIF-Directive) do not seem to differ very much from those in Italian law (leading to sharp criticism by AG Bot in his Opinion on Taricco II, para 96 et seq.).

In view of that, it would have been rather uncomfortable for the Court to rule that the Italian limitation periods for serious VAT-fraud cases should be set aside, wouldn't it? Can Taricco II be, after all, just a temporary (and unstable!) bridge over the troubled waters of the EU’s financial interests, soon to be calmed all down once the PIF is implemented?