Settling the dust? An analysis of Taricco II from an EU constitutional and criminal law perspective

By Giovanni Zaccaroni and Francesco Rossi

Many valuable contributions have been written (for example this blog post but also elsewhere, among many others) on the M.A.S. decision (M.A.S. and M.B., case C-42/17 a.k.a. Taricco II) and, more in general, on the Taricco saga. The majority of them, however, focus mainly on the criminal and constitutional law dimensions separately. In this contribution, we focus on these dimensions together: we believe that this decision is equally important for the relationship between the Court of Justice of the EU (CJEU) and the national Constitutional Courts as it is for the hazardous path of a harmonization of the general part of criminal law at EU level.

The reason why these two dimensions are usually examined separately lies on the different background of the scholars concerned. In this blogpost we have done the effort to put together and explain the importance of the M.A.S. decision from the viewpoint of a criminal lawyer and from the one of a (European) constitutional lawyer. To do that, this work will be divided in two main parts: we will firstly look at the relationship between the CJEU and the Italian Constitutional Court (ICC) (in the first part, sections I and II, written by Giovanni Zaccaroni). We will then see whether and how the decision advances the harmonization of criminal law at an EU level (the second part, sections III-V, written by Francesco Rossi).In the case at hand, the CJEU allowed Italian judges not to disapply the norms set out by Articles 160(3) and 161(2) of the Italian Criminal Code (C.C.), which had previously been declared incompatible with EU law by the CJEU itself in Taricco I (case C- 105/14, paras 46 and 47). The said Articles concern the interruption of the limitation period of crimes, according to which the fulfilment of some listed procedural acts (Article 160(1) and (2) C.C.) increases the limitation period by one quarter compared to the initial term provided for by Italian law (Article 157 C.C.).

In doing so, the CJEU followed the viewpoint of the ICC, which had argued in its referring Order n. 24/2017 (English translation available here) that rules on the limitation period belong to substantive criminal law and thus fall under the scope of the principle of legality and its corollaries (p. 3). Accordingly, the disapplication of the national statute on the interruption of limitation period would have breached the prohibition of retroactive effects (as for the crimes which had not been time-barred before the time the Taricco I judgment came out), the sufficient precision of criminal law, and the separation of powers. All these fundamental principles would be violated by the vague obligation not to apply Articles 160(3) and 161(2) in so far as Italian judges found that the “national rule prevents the imposition of effective and dissuasive penalties in a significant number of cases of serious fraud affecting the financial interests of the European Union” (para 58; emphasis added).

I. The Court of Justice and the Italian Constitutional Identity. The ICC, with the aforementioned Order, raised to the CJEU the question of the compatibility of the previous decision in the Taricco I with the particular constitutional value that is attributed to the limitation period in Italian criminal law (see also para III) and – as a consequence – with the principle of legality. As the case was treated pursuant to the expedited preliminary ruling procedure, the CJEU decided it in less than one year. The Opinion of the Advocate General, released on 18 July 2017, and the decision of the Grand Chamber of 5 December 2017, however, came to antithetic conclusions.

According to the Advocate General, the limitation period provided for by the Italian legislation for VAT fraud cannot ensure the effective protection of the financial interest of the EU; nor can the argument of the ICC be accepted that the (short) limitation period for VAT fraud is part of the Italian constitutional identity. The Advocate General comes to the conclusion that “in the Italian Constitution the principles classified as ‘fundamental’ are set out in Articles 1 to 12, and the principle that offences and penalties must be defined by law is therefore a priori not included in that category” (para 181). In the mind of the Advocate General, the CJEU should let EU law prevail and order to the ICC to disapply contrasting national legislation.

The CJEU, however, followed another path. Instead of aligning itself with the perilous task of defining the extent of the Italian Constitutional identity, it decided to leave to the national Court the possibility to exceptionally disapply EU law. In the first step (paras 43-44) of its legal reasoning, the CJEU recognizes that Art. 325 TFEU falls in an area of shared competence that was not harmonized at the time of the dispute, and, as a consequence, Member States enjoyed a certain margin of discretion in the question of fraud against the EU financial interest. The Court, in a second step (para 52), recognizes that the principle of legality in criminal matters is protected by Art. 49 of the Charter and is part of the “constitutional traditions” common to EU Member States. For this reason, the CJEU held that it is for the national judge to determine if the application of Art. 325 TFEU is as to be in contrast with the constitutional identity of the Member State involved.

II. The language of dialogue of the Court of Justice. Moving the analysis of the judgment forward, it is interesting to note the language of the CJEU is very different from the one used by the Advocate General, but also from the one used by the ICC. The CJEU, in fact, avoids making reference to concepts that are typical of the discourse of EU constitutionalism, without indulging, as the Advocate General did, in a review of the national constitutional identity in light of the Treaties’ obligations. It is possible to argue that the CJEU has chosen, in its judgment, a “language of realism”, which reflects the traditional neo-functionalist approach that is at the basis of the dialogue with the ICC as well with others national constitutional Courts.

The inclusion in the “constitutional traditions” common to the Member States of the interpretation of the principle of legality, as a fundamental right, according to the Italian Constitution is surely one of the most interesting linguistic choice made by the CJEU. Interestingly, this concept was not at the core of ICC reasoning in its Order, which rather referred to the “constitutional identity” of the Italian legal order. On the other side, instead of recognizing the interpretation of the principle of legality as a part of the Italian constitutional identity, the Court maintained that this interpretation is part of the “constitutional traditions common to the Member States” pursuant to Art. 6.3 TUE. The question is then: why has the CJEU decided to label the interpretation of the principle of legality as a common constitutional tradition instead of as an element of national constitutional identity? As suggested by Pollicino and Fabbrini, a convincing explanation might be that the CJEU has chosen to follow the tradition of dialogue and confrontation that brings the two Courts together, instead of trying to give a definition of the constitutional identity of the Italian Republic. Conversely, scrutinizing the extent of Italian constitutional identity would have led to a twofold harmful conclusion for the CJEU.

A restrictive interpretation of the Italian constitutional identity would have opened the way to the activation of the controlimiti doctrine by the ICC. The controlimiti doctrine, which has paved the way to the shaping of similar doctrines by other Constitutional Courts (i.e. the Solange doctrine of the German Federal Constitutional Court), is how the ICC has allowed the gradual incorporation of national competences towards the European Union, provided that the inner limits of the Italian Constitution were respected (see ICC, decision of 27.12.1973, Frontini). If the CJEU would have openly recognized the interpretation of the legality principle as part of the constitutional identity of the Italian Republic pursuant to Art. 4(2) TEU, it would have created a precedent to rely on while dealing with similar questions in the future (see L.S. Rossi), since it would have considerably widened the possibility of the ICC to make recourse to its controlimiti doctrine. It should not be forgotten that similarly important preliminary questions are still pending in front of the CJEU (see case C-493/17), or have just been released (i.e. case C-596/16, Di Puma). In the choice of the language of the “common constitutional traditions” (Art. 6.3 TEU), the CJEU resorts to a more pluralistic and less identitarian tradition of dialogue, which rather than being read in light of a single legal order, should take into account the pluralism of the constitutional traditions of the various Member States, and be balanced with the ultimate goal of the EU, the achievement of an ever closer Union.  In this sense, the Court seems to have implicitly confirmed that Art. 4.2 TFEU represents a clause of last resort that allows an exception to the application of the Treaties only when it is not possible to make recourse other provisions such as Art. 6.3 TEU (namely, when at stake there is not a fundamental right, as could be – for instance – the case in the application of an exception to the “strict conditionality” which follows the implementation of the economic and financial measures enacted to counter the economic crisis).

The M.A.S. judgment is also one of the most important decisions in a (not yet closed) stream of cases where national constitutional rights (and Courts) and EU law were contrasting. Here we make reference in particular to the decision in the OMT/Gauweiler judgments (case C-62/14 and judgment of 21.06.2016) and in the Dansk Industri/Ajos judgments (case C-441/14 and decision n. 15/2014, unofficial English translation available here), and the judicial rebellion commented on by Sarmiento. In the above mentioned cases, the Court of Justice has been confronted with two preliminary questions posed by the German Constitutional Court and the Danish Supreme Court (both in charge of the constitutionality review in the respective legal orders), where at stake was the balance competences/powers between the EU and the Member States. While in the OMT case the Court of Justice ruled on the compliance with the Treaties of the ECB programme of acquisition of bonds of the Member States in the secondary market (the OMT programme), in the Danish case the CJEU restated the application of the horizontal direct effect of the principle of non-discrimination on the ground of age. In their final judgment after the preliminary ruling, the German Constitutional Court accepted – although with various caveat – the judgment of the Court of Justice, while the Danish Court declared the horizontal direct effect of the general principle as ultra vires and disregarded the application of the CJEU decision. This trend represents a restatement of the importance of the constitutionality review at national level, after the first Taricco decision. The above-mentioned cases represent accordingly the sign of the tension occurring between the Court of Justice and its national counterparts in the aftermath of Taricco, and a temporary breach in the tradition of dialogue and respectful confrontation between high-level courts. Read in light of this line of cases, the M.A.S. decision of the CJEU aims to find a reasonable accommodation between two competing views: the national constitutional courts, required to apply and protect the paradigm of constitutionality review of the respective legal orders and the CJEU, in charge of the uniform interpretation of the Treaties, the constitutional charter of the EU legal order. As stated in the introduction of this contribution, the decision of the CJEU in M.A.S. to leave room to the specific characteristics of the national constitutional legal orders has two legitimate and equally important justifications: to preserve the dialogue between European Courts, but also to avoid shortcomings for the present (and future) harmonization of European criminal law, as explained below.

III. M.A.S. from a criminal law perspective: does the ruling of the CJEU strike the best balance between EU and Italian constitutional law? Being cornered by the ICC, in the M.A.S. judgment the CJEU came out with an Italy-oriented type of decision, in order not to pave the way to the use of the controlimiti doctrine in hard times for judicial dialogue between the CJEU itself and national Constitutional Courts (as outlined in section II above). Nevertheless, after Taricco I, the incompatibility of Articles 160(3) and 161(2) of the C.C. with Article 325 TFEU has not been questioned by any of the two Courts. The ICC needed to grant the CJEU something, while knowing that it was diplomatically asking the latter Court to deeply revisit its “Taricco I rule”. Instead, the CJEU confirmed the general obligation to set out and apply appropriate sanctions at national level so as to protect EU financial interests, and restated such incompatibility as regards the Italian statute on limitation period.

Beside the several proposals suggested by scholars, despite a procedural obstacle that will be mentioned below, there was another solution more capable than the one in the M.A.S. judgment of meeting both the ICC’s and the CJEU’s interests that were at stake.

 As a first step, the ICC could have reconsidered its case law about the nature of limitation periods in the light of supranational inputs and of the criminal traditions of most of the Member States, acknowledging the procedural nature of norms concerning the interruption of the limitation period while saving the substantive nature of its essence and its basic rationale. Indeed, while the limitation period is devised in relation to the seriousness of the crime that has allegedly been committed, and to the functions of general and special prevention pursued by criminal sanctions, interruption of the limitation period is calibrated on the particular structure and functioning of criminal proceedings, in order (at least, abstractly) to guarantee their correct functioning.

 As a second step, the ICC could have reduced the tension between the “Taricco I rule” and the reasonable foreseeability of criminal law. Recognising the procedural nature of rules on the interruption of limitation periods, in the case at hand the principle of legality would have had a different meaning. In fact, the procedural side of the principle of legality demands the overall fairness of criminal proceedings, prohibiting public authorities from arbitrary interfering with individual liberty. Hence, the issue to be resolved would have been if fulfilling the Taricco I obligation would be compatible with such overall fairness. On the one hand, one could argue that even from this perspective, the “Taricco I rule” infringes upon the principle of legality. Indeed, there would be no clear guidance about the requirements and the legal consequences of the disapplication, and the strategic plan of the defendant not to choose shorter special proceedings and the relative advantages as for the relative criminal sanctions would be frustrated. On the other hand, starting from the general provision of Article 175 of Italian Code of Criminal Procedure (which sets out the requirements to reinstate the legal due date for procedural acts), the ICC could have added a new procedural norm in the Italian legal order, allowing the defendant to be given back the time limit provided for by national law to choose between the ordinary and a simplified proceeding. As it previously did in case No 113 of 7 April 2011, which introduced the power to reopen criminal proceedings for judicial review of res iudicata in the light of the European Court of Human Rights judgements, such solution could have represented a sustainable balance between the duty to comply with EU law, and the right of the defendant not to be unpredictably and unfairly disadvantaged during the criminal proceeding. Unless limitation period is already passed regardless of the application of Articles 160(3) and 161(2) C.C., the Courts would have left the defendants the possibility to choose the best strategy to lead, according to the changed normative framework. It is worth noting, indeed, that in Taricco I the CJEU did not declare the incompatibility with EU law of limitation period in itself, but only of the Italian rules on the interruption of limitation period and their lethal consequences on criminal proceedings.

Furthermore, as a third step, the ICC could have come back to the CJEU in order to adjust some of the other weaknesses of the Taricco I judgment, especially the boundaries of the unclear requirements of the seriousness of the fraud and the significant number of cases (see the summary above).

The judicial introduction, by means of the interpretation of national law, of normative mechanisms proving necessary to fulfil supranational obligations is not new to the Italian legal order, and would not breach the separation of powers. Indeed, cases like No 113 of 7 April 2011 belong to the family of the “additive” judgments of the ICC, which declare national dispositions unconstitutional in so far as they do not provide what is constitutionally necessary (in the case at hand, in the light of Article 11 of the Italian Constitution first, as known as “principio di apertura” to EU law). Besides, it would be up to the national judges’ interpretation to verify the applicability of the new norms and their practical consequences, and to the Italian legislator to specify and/or integrate such normative framework by means of new dispositions, in case that proved necessary. However, beside the exceptionality of “additive” judgments and the complexity of the Supreme Courts’ rulings concerning the relationships between EU and national criminal law, the very different requests the referring judges (especially the Corte d’Appello di Milano, Order of 18 September 2015) made to the ICC represented a procedural obstacle to the said outcome that was impossible to bypass. Differently than the procedure for preliminary rulings before the CJEU, which allows the latter Court to use a wider margin of appreciation when it assesses the admissibility of the referring order and to rephrase its scope and content in the light of EU law (as it did with the request from the Tribunale di Cuneo that started the Taricco saga), the ICC is more strictly bound by the correspondence between such order and the final ruling.

IV. Limitation period, the general part of criminal law, and EU law: a yet unclear relationship. After the ICC Order and the M.A.S. judgment, one might infer that the EU competence to deal with the nature of limitation period (or other interpretative issues concerning general rules of national criminal law) is to be excluded. Nevertheless, besides the existence of past rulings of the CJEU regarding the minimum common meaning of mens rea on a continental scale (the most relevant one being the Intertanko judgment, case C-308/06: see paras 72-77) which prove the opposite, the message sent by the latter Court in between the lines might be different. As previously stated in section II of this post, the CJEU argues that at the time Taricco I came out, Italian judges were “free to provide” that rules on the interruption of the limitation period formed “part of substantive criminal law” (para 45). The CJEU recognised this because “at the material time for the main proceedings, the limitation rules applicable to criminal proceedings relating to VAT had not been harmonised by the EU legislature” (para 44).

First, as a result, the Italian (and almost isolated) legal tradition on the limitation period is being kept standing still and protected against any kind of influence by the defensive approach of the ICC. On the contrary, the same tradition has been deeply reviewed by the CJEU, developing the pluralistic idea of constitutional traditions (as outlined in section III above) in criminal matters. Now, the further placement (after Article 3 of Regulation EC, EURATOM No 2988/95 on the protection of the European Communities financial interests) of a limitation period at the EU level (with the adoption of Directive EU/2017/1371 on the fight against fraud to the Union’s financial interests) makes its European-oriented upgrade still necessary, and leaves the controversial issues concerning its real nature and its adequate regulation for crimes affecting the EU financial interests open to further debate.

Second, it seems that, from the CJEU perspective, its ruling in M.A.S. is not likely to represent a legal precedent in the future. It is worth noting that general rules of national criminal law might serve the purpose of implementing EU law at the national level; and that, accordingly, the competence of the CJEU over such rules is not excluded a priori. Indeed, the latter Court has extended over time the meaning of the concept of national law implementing EU law, which today may include any national norm affecting any area of EU competence (see, for instance, Åkerberg Fransson, case C-617/10). Within this wide interpretation, like any other provision being an expression of purely national competences in the light of the allocation of competences between the EU and the Member States, general rules of national criminal law can be attracted to EU law pursuant to implicit linking points (the latter being, for instance, instrumental, sequential, substantive, or intrinsic, as M.E. Bartoloni pointed out) between national law and the EU law itself. Besides, national legislators must respect EU law even when they act in exercise of an exclusive competence of the State, so as not to adopt measures capable of compromising the protection of legal positions ruled on at the EU level. That is what happens in the case at hand: if every Member State sets out statutes on a limitation period that contributed to causing a pathological lack of effectiveness in criminal proceedings and an unsatisfactory dissuasive effect of sanctions protecting EU fundamental interests, the exercise of the supranational competence and the primacy of EU criminal law would be crushed.

V. Is the Taricco saga over? The doubts surrounding the current patchwork regarding limitation period in the EU legal order. The resulting legal framework of this controversial expression of judicial dialogue (see above sections I-II) turns out to be quite paradoxical. In Taricco I, the CJEU recognised a vertical direct effect to the norms of Article 325 TFEU, even though one could argue that they are not directed to individuals but to Member States, and lack at least the requirement of being unconditional. Besides, this incorrect use of the direct effect would make the defendant pay for someone else’s mistake (in this case, the Italian State): Article 325 TFEU could produce an in malam partem effect in the presence of a legislative failure to fulfil the obligations that the Article itself sets out.

Despite that, the ICC chose not to challenge the sovereignty of the CJEU over the interpretation of EU law, and the CJEU did not withdraw from the interpretation of Article 325 TFEU. Hence, Article 325 TFEU might be again used by the latter Court to oblige national judges of another Member State to disapply national rules preventing the protection of EU financial interests. The CJEU implicitly made a step towards the harmonisation of the general part of criminal law, trying to protect the EU competence from the growing cases of ultra vires decisions of national Constitutional Courts.

Nevertheless, the current legal basis for Directive 2017/1371 is represented by Article 83 TFEU. Consequently, now there seems to be another reason why the possibility to recognise the direct effect of EU law in the field of the limitation period for criminal offences affecting supranational financial interests is disputable. Indeed, beside the said issues concerning the correct interpretation of Article 325 TFEU, secondary EU norms in criminal matters such as the ones adopted on the basis of Article 83 TFEU cannot ascribe or aggravate the criminal responsibility of the defendant without the previous intervention of the national legislator (see Grand Chamber, of 3 May 2005, joint cases C-387/02, 391/02 and 403/02, Berlusconi and others). However, the current cohabitation of norms on limitation period with and without direct effect might raise other doubts about the future rulings of the CJEU in similar cases.

Thus, the Taricco saga does not seem to be actually over. The future will tell us if the outcome of this judgment will be enough to settle the dust of the first Taricco I decision.