ENLARGING EU COMPETENCE IN CRIMINAL MATTERS THROUGH POLICIES. THE ANTI-FRAUD DIRECTIVE AS A CASE STUDY^[*]

Ampliación de la competencia de la UE en materia penal mediante políticas. La Directiva antifraude como caso de estudio

Leandro Mancano

Lecturer in EU Law
Edinburgh Law School

http://dx.doi.org/10.18543/ed-67(1)-2019pp97-110

Recibido: 11.05.2019
Aceptado: 21.06.2019

Abstract

The present contribution discusses the emergence of a multi-level system of criminal justice in the EU and the role of constitutional conflicts ensuing from that. Taking inspiration from Paolo Carrozza’s contribution on the enduring role of Hans Kelsen, the paper analyses the enlarging of EU competences in criminal law to pursue the effectiveness of EU policies.

Keywords

Criminal justice; EU competences; Kelsen; effectiveness.

Resumen

Se discute el surgimiento de un sistema de justicia penal en múltiples niveles en la UE y el papel de los conflictos constitucionales resultantes de ello. Inspirándose en la contribución de Paolo Carrozza sobre el papel perdurable de Hans Kelsen, el documento analiza la ampliación de las competencias de la UE en Derecho penal para perseguir la eficacia de las políticas europeas.

Palabras clave

Justicia penal; Competencia de la UE; Kelsen; eficacia.

Sumario: I. Introduction. II. The Greek Maize Case. Namely, the cornerstone of the ‘functionalist journey’ of eu criminal law. III. Building pillars across the EU. IV. From Tampere to Amsterdam: the high-speed rail of EU criminal law by mutual recognition. V. Merging pillars: the Hague programme, the Environment Crime and the Ship Source Pollution judgments. VI. The day after tomorrow. VII. The directive on the protection of the EU Financial Interests. VIII. Conclusions.

I. INTRODUCTION

EU criminal law is one of the most interesting areas to study the enduring influence of Kelsen’s thought on jurists and lawyers upheld by Paolo Carrozza.[1] EU criminal law has only seen an increase in the level of interaction between the Union and state legal framework. The end of the so called transitional period in 2014 following the entry into force of the Lisbon Treaty has brought about considerable changes to the legal regimes of measures of EU criminal law. Before that date, ‘third pillar’ acts[2] would present two main flaws: the Commission could not use its power of enforcement under article 258 TFEU against Member States’ failure of implementation; the Court of Justice of the European Union (CJEU) had no competence in issuing preliminary rulings on these acts, unless the concerned Member State had explicitly accepted it. Conversely, mentioned defects are absent from post-Lisbon rules enacted in such realm, due to the criminal law’s ‘communitarisation’ brought to completion by the Treaty reform.

The strengthened value of instruments of EU criminal law raises questions on possible conflicts between EU and state competences, on the one hand; on the other, there is the issue of Union law possibly lowering the standard of fundamental rights protection as enshrined in member states. This paper reconstructs the progressive emergence of the EU as a major role for establishing priorities in terms of criminalisation through encroaching upon member states powers. Through the example of the recently adopted PIF Directive, it also shows that the communitarisation of EU criminal law carried out by the treaty of Lisbon has not removed the possibility of conflicts between the two levels of governance.

II. THE GREEK MAIZE CASE. NAMELY, THE CORNERSTONE OF THE ‘FUNCTIONALIST JOURNEY’ OF EU CRIMINAL LAW

The coming into being of EU criminal law can be partly told through the functionalist approach endorsed by the Court of Justice. The start of this approach may be traced back to Commission v. Greece[3] litigation, better known as Greek Maize case. The Court was faced with a case of fraud against Community budget by a Member State where such wrongdoing had not been criminalized. The judges, by founding their reasoning on the principle of loyal cooperation,[4] required Member States to take all measures necessary to guarantee the application and effectiveness of Community law. Member States are required to comply with that principle and thus ensure that ‘infringements of Community law [are] penalized under conditions, both procedural and substantive, which [are] analogous to those applicable to infringements of national law of a similar nature and importance and which, in any event, [make] the penalty effective, proportionate and dissuasive’.[5]

The Court established criteria for Member States’ action. Firstly, the principle of assimilation was stated, namely the duty to treat Community law violations with analogous means to those they would use to address comparable violations of national law. Secondly, Member States were required to react through effective, proportionate and dissuasive penalties, to be enforced with the same diligence that applied to national situations.

Since then, the Greek Maize ‘formula’ has been heavily relied on by the ECJ in its judgments, as well as in a great deal of legislative instruments.[6] The exceptional nature of the judgement at issue lies in recognizing to the EC an ‘indirect’ competence in criminal matters even before the Treaty of Maastricht.

III.  BUILDING PILLARS ACROSS THE EU

The Treaty of Maastricht[7] built the pillars-based institutional framework of EU, with the latter being conferred a competence in criminal matters, though very limited and outside Community law. However, thorny issues arose from the introduction of such a power to legislate in criminal matters. On the one hand, a plethora of measures in this area were approved; on the other, these instruments proved to be not extremely effective, and cumbersome in their adoption. Indeed, both Joint Actions and Conventions (the main ways used to adopt EU criminal legislature at that time) were mostly international law instruments. In that context, two main approaches emerged: the Commission would put forward a legislation ratione functionis, that is to say criminal law under first pillar, in order to ensure the effectiveness of Community law and policies; whereas the second, espoused by the Council and many Member States, stood in favor of a ratione materiae logic, so that criminal legislation should have been approved under the ‘third pilar’, if weaker.

Such discrepancy came to the fore as early as the very beginnings of EU criminal law, and exactly with regard to the fight against fraud relating to the Community budget. In particular, the well-known Corpus Iuris (a sort of mini-criminal code attempting to unify Member States’ laws in that subject) was at stake.[8] However, the project never saw the light of day, and was replaced by the Convention for the protection of the financial interests of the European Communities and accompanying protocols[9] (hereinafter, referred to as the “PIF Convention”).

IV.  FROM TAMPERE TO AMSTERDAM: THE HIGH-SPEED RAIL OF EU CRIMINAL LAW BY MUTUAL RECOGNITION

The end of century saw the almost simultaneous adoption of the ‘Tampere Programme’[10] and the entering into force of the Treaty of Amsterdam: the former represented the moment from which “the Union’s policy in the area of justice and home affairs has been developed in the framework of a general programme”;[11] on the other side, the latter bore significant novelties for EU criminal law. Most importantly, it replaced Joint Actions with the instrument of Framework Decisions. They needed no ratification by Member States and, apart from having no direct effect, presented the same wording as Directives in the Treaty. EU criminal law was being provided with a considerable, Community-oriented tool.

The phase started with the Tampere Programme coincided with the so called ‘second wave of third pillar’[12]. As said, the first wave – consisting of weaker legislative instruments such as Conventions and Joint Actions – stemmed from the Maastricht Treaty had not been accompanied by guide-lines or programmatic document. Whereas the second wave can be seen as a momentum flowing from the born of a European criminal policy. As a result, these two pathways may not be decoupled. Conversely, a fil-rouge linked Tampere to Amsterdam, namely the building of an area of freedom, security and justice by means of mutual recognition of judicial decisions. The Treaty of Amsterdam referred to the development of the Union as an ‘area of freedom, security and justice’ as a fundamental objective[13] and incorporated the Schengen acquis into Community/European Law[14].

This development went hand in hand with the ‘journey’ dating from 1998, when the UK government, during its EU Presidency, put forward the idea of applying the principle of mutual recognition to criminal law. This proposal was followed by the well-known Tampere Programme, where the European Council adopted the principle of mutual recognition as ‘the cornerstone’ of judicial cooperation in criminal matters. Many EU measures in criminal matters saw the light in that period, both substantial and procedural. Admittedly, the new wording of the Treaty as to the Justice and Home Affairs Law mirrored the key-features of Tampere Programme, in so far as a major convergence to the aim of judicial cooperation seemed to emerge, at that time. As Weyembergh rightly stated, ‘the only function of approximation [of substantial criminal law] which ha[d] received some attention is the auxiliary function, supporting judicial cooperation’.[15] Against that backdrop, significant instruments such as the Framework Decision on European Arrest Warrant[16] were adopted. Furthermore, the conjunction between ‘pillar approximation’ and ‘pillar dialogue’ brought about noteworthy results. First and foremost, the Commission’s initiative to adopt criminal rules against fraud under Community legal basis.[17] Albeit the proposal was not adopted, other important anti-fraud related instruments were passed, such as the Framework Decision on confiscation of crime-related proceeds, instrumentalities and property,[18] the Framework Decision on money laundering.[19]

Such ‘pillars dialogue’ involved another Community polices, with the Commission putting forward a Directive on the protection of environment through law.[20] However, the proposal ended up being adopted as a Framework Decision, because of the opposition of the Council.[21] Question as to a ‘silent’ expansion of EU competence in criminal matters were arising, fostered by a peculiar legislative technique used at the time. According to the double-text mechanism, two ‘twin’ instruments were concurrently adopted: a Community measure regulating the conduct was complemented by a ‘third pillar’ act, introducing sanctions.

V. MERGING PILLARS: THE HAGUE PROGRAMME, THE ENVIRONMENT CRIME AND THE SHIP SOURCE POLLUTION JUDGEMENTS

As for the shaping of EU criminal competence through policies is concerned, the Directive[22] and the Council Framework Decision on ship source pollution are to be recalled.[23] In the 2005 Hague Programme[24] and follow-up documents,[25] the EU institutions made it clear that “[t]he approximation of substantive criminal law serves the [to facilitate mutual recognition of judgments and judicial decisions and police and judicial cooperation in criminal matters] and concerns areas of particular serious crime with cross border dimension”.[26]

‘Pillars dialogue’ turned into a duel when the Commission decided to bring an action before the ECJ for annulment of the Framework Decision on the protection of environment through criminal law.[27] More precisely, it held that Articles 1 to 7 (defining the offenses) should have been adopted according to a Community legal basis, rather than under the ‘third pillar’ one. Conversely, the Member States and the Council stated that, in the light of criminal law significance for national sovereignty, EC competences should have been explicitly conferred upon in the Treaty. The Luxembourg judges struck down the Framework Decision, by embracing the Commission’s stance. Firstly, the Court referred to Articles 29 and 47 TEU, which stated respectively that third pillar action must be ‘without prejudice to the powers of the European Community and that nothing in the EU Treaty [would] affect the Treaties establishing the European Communities’. On that ground, the ECJ established a hierarchy among the pillars and recognized the primacy of Community law. It followed that, according to the ‘functionalist’ perspective adopted by the Court, EC criminal competence could be found, if needed to ensure fundamental Community objective (as the environment protection was).

This ground-breaking judgment triggered a knock-on effect. The Commission challenged one important instance of ‘pillars dialogue’, namely the Framework Decision on ship-source pollution.[28] The Commission, as in the Environment crime case,[29] argued that part of the Framework Decision (in particular, Articles 1 to 10) should have been adopted under the ‘first pillar’ legal basis[30] and, owing to the indivisible nature of the same act, it should have been annulled. The ECJ annulled the Framework Decision. The Court’s ruling primarily recognized that the Community’ authority on criminal matters is exceptional and restricted, on one hand; on the other, it reiterated the hierarchy established in Case C-176/03, by using article 47 TCE as a shield of ‘first pillar’ supremacy. The ECJ elaborated on such a statements according to a ‘functionalist’ logic, and reaffirmed the EC competence ‘when the application of effective, proportionate and dissuasive criminal penalties by the competent national authorities is an essential measure for combating serious environment offences, in order to ensure that the rules which it lays down in that field are fully effective’[31]. The Court clarified the breadth and the depth of Community powers in criminal matters, by denying any Community entitlement to determine the level of penalty to be applied by Member State,[32] and stipulating that EC rules on criminal law encountered the limit of direct effect. As a result, such measure could have been adopted only by means of directives.[33]

VI.  THE DAY AFTER TOMORROW

While leaving a number of questions unanswered, the judgment is unanimously considered as a milestone of EU law broadly speaking. The follow-up was twofold: firstly, directives harmonising criminal laws were adopted.[34] Shortly after, the Treaty of Lisbon entered into force,[35] legitimating the ‘functionalist’ understanding of EU criminal law. Indeed Article 83(2) TFEU states that ‘[i]f the approximation of criminal laws and regulations of the Member States proves essential to ensure the effective implementation of a Union policy in an area which has been subject to harmonisation measures, directives may establish minimum rules with regard to the definition of criminal offences and sanctions in the area concerned’.[36]

The EU, if intending to pose criminal obligations on Member States also with regard to the type and the level of penalties, has been provided with appropriate instruments. As a result, instances of further spill-over effect in EU criminal law (as the two described judgments) should not be expected. However, also in the light of the foregoing, such a statement might legitimately be questioned. The Court has not shied away from stretching the wording and the meaning of the Treaty provisions for reasons related to Union policy objectives. Now that the Treaty of Lisbon has introduced an explicit – albeit somewhat ambiguous – legal reference to this end, might EU ‘functional’ criminal rules (completely) outside Article 83(2) TFEU be envisaged?

VII.  THE DIRECTIVE ON THE PROTECTION OF THE EU FINANCIAL INTERESTS

The PIF Directive is a good example of post-Lisbon dialogue between the EU and member states on criminal law competences and fundamental rights protection. Three preparatory documents explain the EU’s view on the PIF Directive: the Impact Assessment (IA),[37] the explanatory memorandum of the proposal and the European Parliament (EP) report.[38] The IA starts with a remark on the strong obligation for the EU and the member states to protect the Union’s financial interests. The shortcomings of the current legal framework would lie in the little dissuasive effect of the existing provisions and the uneven level of enforcement and implementation. The goal of effectiveness by deterrence is to be achieved by: extending the range and types of offences; harmonizing the sanctions, currently too low and diverse amongst member states.[39] The foregoing summarises the Union approach to the use of criminal law and custodial penalties. On the one hand, an area of free movement is sustainable so long as a common set of rules is established. Potential wrongdoers are therefore deprived of the choice of forum: namely, the possibility to move where offending is more convenient due to low enforcement rate, mild penalties or narrow definition of the punished conducts. On the other, criminal law can unfold its deterring potential through the threat of harsh sanctions such as imprisonment.

The Commission identified a directive and a regulation as the two most appropriate legislative instruments to be adopted.[40] While the former was eventually preferred because less intrusive, the latter – and the enactment of EU directly applicable rules in criminal law – was not regarded as unlawful. Quite the contrary, a regulation would comply with the EU law proportionality principle, owing to the great relevance of the interests at stake.[41]

That being so, the Commission’s proposal opted for Article 325(4) TFEU as its legal basis, stating that ‘The fight against illegal activities affecting the Union’s financial interests is a very specific policy area [where] the Union has a broad array of tools at its disposal’.[42] The EP voiced its disagreement,[43] and secured a change of the legal basis to Article 83(2) TFEU in the final version of the text. The EP considered Article 83(2) TFEU lex specialis to Article 325(4) TFEU. The travaux préparatoires of Article 83(2) TFEU show that the latter provision was considered an appropriate legal basis ‘in the context of the protection of the EU’s financial interests’.[44] Additionally, Article 86 TFEU contains a provision on the establishment of a European Public Prosecutor’s office ‘in order to combat crimes affecting the financial interests of the Union’. The protection of EU’s financial interests, therefore, is by no means covered under the sole Article 325 TFEU.

The disappearance, in Article 325 TFEU, of the mentioned prohibition provided by Article 280(4) EC can be explained by the communitarisation of criminal law competence. Preventing EU measures on fraud from affecting national criminal systems would not make sense, within the current institutional framework. Opting for a different legal basis would result in eluding the procedural guarantees set in Article 83(2) TFEU.[45] The dispute over the legal basis reveals the importance of policy legal basis outside the provisions in the Treaties strictly devoted to criminal law. The IA reveals that the issue behind that discussion concerns possible the direct application of EU rules in criminal law and the circumvention of the constraints posed by the EU legislature in Article 83 TFEU. This results in issues of proportionality – which the Commission deemed complied with – and legal certainty.

The Commission explained the choice of imprisonment and a minimum threshold of deprivation of liberty by arguing that ‘[e]conomic crime is typically an area where criminal sanctions can have a particularly deterrent effect, as potential perpetrators can be expected to make a certain calculation of risks before deciding to engage in such criminal activities’.

Issues of legal certainty arise from the dispute on the legal basis and the thresholds of punishment laid down in the Directive. The use of Article 325 TFEU was meant to promote the enlargement of EU competences in criminal law through effectiveness of Union’s policies;[46] a dynamics well-known from the Greek Maize, the Environment Crime and Ship-Source Pollution cases. While this debate might appear as pure speculation after the Directive was adopted pursuant to Article 83 TFEU, recent development suggests that quite the opposite is true. The recourse to policy-based – rather than area-specific – legal base is not limited to Article 325 TFEU.[47]

In a 2014 judgment[48] concerning Directive 2011/82/EU[49] on exchange of information between Member States on specific driving offences, the CJEU found in favour of the Commission and stated that the Directive should have been adopted pursuant to Article 91 TFEU, rather than Article 87(2) TFEU. The two legal bases cover transport policy and police co-operation respectively. The CJEU’s consistent preference for the policy legal basis – rather than the criminal law-specific one – should not be overlooked.

More closely related to the protection of Union’s financial interests, the Taricco saga on Article 325 TFEU has revamped the debate on direct applicability of EU law resulting in establishment of individual criminal liability. At stake was the Italian law on the statute of limitation for VAT offences. The CJEU stated that a short statute of limitation like the one laid down in the Italian law would affect states’ obligations under Article 325(1) and (2) TFEU, where that law would: prevent the imposition of dissuasive and effective penalties in a considerable number of cases; provide a longer period of limitation for fraud against the financial interest of the state. The Court stated the obligation for national judges to disapply those rules, if needed to ensure the effectiveness of Article 325(1) and (2) TFEU. The judgment sparked a fierce debate, with – especially Italian – lawyers concerned that the ruling might result in direct establishment of criminal liability through EU law and circumvention of the principle of legal reserve (or nulla poena sine lege parlamentaria) established in Article 25(2) of the Italian Constitution.

The issue arrived before the Italian Constitutional Court, which referred three questions to the CJEU asking – for what is concerned here – whether the obligation of disapplication stated in Taricco should be upheld even where that would be at variance with the overriding principles of the constitution of the member state concerned or with the inalienable rights of the individual conferred by the constitution of the member state. The CJEU reaffirmed the obligation, unless that disapplication entailed a breach of the principle that offences and penalties must be defined by law because of the lack of precision of the applicable law or because of the retroactive application of legislation imposing conditions of criminal liability stricter than those in force at the time the infringement was committed.[50] While the Court seemed to retreat in Taricco II, it did so without relying on the principle of legal reserve. Consistently with its consistent case-law,[51] the CJEU focussed on foreseeability, legal precision and non-retroactivity.[52]

The adoption of a criminal law Regulation – explicitly regarded as proportionate by the Commission[53] – and the scenario envisaged in Taricco are different. They both, however, reveal the uniqueness of custodial penalties in EU law: imposition of imprisonment terms through direct application of Union rules for reasons of effectiveness. The CJEU’s argued that criminal liability deriving from direct applicability of Article 325 TFEU/disapplication of rules on prescription have a procedural effect, whereas the principle of legality would regard substantive criminal law. The argument is too formalistic. Furthermore, it might be objected that the obligation refers to effective and dissuasive sanctions broadly. The previous discussion revealed, however, that no penalties are considered as effective and dissuasive as those entailing deprivation of liberty.

VIII.  CONCLUSIONS

The dialogue between the EU institutions and member states on EU criminal law shows the difficulties in allocating competence on the best-placed level of governance. It also reveals conflicts that can arise when it comes to different understanding of very important fundamental rights. The dispute over the legal basis has brought to the fore once again the issue of the bypassing of the limits settled by the Treaty and the enlargement of EU criminal competences through policies. A sort of circularity seems to surface: a certain policy is outweighed because the legal basis prioritizes it; and the other way around, the same policy serves to ‘pick’ the legal basis, which is in turn used to extend EU competences. Indeed, to require no ‘essentiality’ for enacting criminal measures, or to deprive Member States of the ‘emergency brake’ would amount to strikingly empower the EU in this respect. However, what it is most important remained unsaid in the institutional dialogue above described. The potential groundbreaking stays behind the concerns voiced by the EP, and regards the proposal of directly applicable criminal rules. As shown, the impact assessment implicitly accepted such a possibility, referring to a hypothetic penal regulation as ‘a single, immovable set of rules […] with exhaustive definitions of offenses and rigid sanction types and levels […] that Member States’ authority wold apply directly’ (emphasis added).[54] Apart from being difficult to imagine how Member States could apply it directly (particularly as far as the penalties is concerned[55]), the CJEU’s case law has consistently excluded that criminal liability can be established or aggravated directly on the basis of EU rules. However the Court, grounding its decisions on legal certainty and non-retroactivity,[56] has brought about a somewhat ‘autonomous’ concept of principle of legality, so stand stilling and innovating in the same breath. This is even more true on looking at Member States’ that recognize the nulla poena sine lege parlamentaria as a constitutional principle. The Taricco saga, concerning the direct applicability of Article 325 TFEU to trump national criminal law at the expenses of the individual, confirms the Court’s consistent approach.

Should the EU be conferred upon the power to adopt directly applicable criminal rules, one might find that a complete (and, perhaps, definitive) integration as with criminal law would be achieved, given the paramount task which the principle of direct effect has performed in promoting the EC/EU project from Van Gend en Loos onwards.[57]