Implementation of the EPPO in Belgium: Making the Best of a (Politically) Forced Marriage?

1. Introduction: Belgian Academic Study on the EPPO Implementation

Two years have passed since the adoption of the Regulation on the European Public Prosecutor’s Office (‘EPPO Regulation’ or ‘Regulation’), a legal and political milestone. The initial proposal of the Commission (discussed by V. Franssen on this blog) was published in 2013 and difficult legislative negotiations followed. National parliaments even pulled out a ‘yellow card’ (see this blog). The proposal reflected a strongly ‘federal’ logic, but after negociations, the EPPO eventually metamorphosed into a hybrid institution, linking up EU and Member State prosecutors. It is the fruit of conflicting visions (both at the EU and at the national level) on the need for a new institution, on its status with respect to existing law enforcement and on its authority. At some point the whole project seemed ‘dead in the water’, but the Commission finally managed to push it through under the form of enhanced cooperation, involving eventually 22 participating countries.[1] However, the EPPO’s creation does come across as a top-down policy choice of the political authorities, rather than a response to a pressing request from existing prosecution services or from Eurojust.

The EPPO should be operational by November 2020. The participating Member States must ensure that their national legal system is compatible with the Regulation by then. While regulations are directly applicable in the national legal order, the EPPO Regulation refers to national law for several aspects, e.g. the conditions, procedures and modalities for taking investigative measures (Art. 30). That makes the implementation process particularly tricky, especially since the logic behind the EPPO does not necessarily match the way in which criminal investigations and public prosecution are designed by Member State law. Even after the implementation of several so-called ‘Roadmap’ directives on suspects’ rights and victims’ rights, the approximation of national criminal procedures by EU law remains limited.[2] In addition, the EPPO Regulation sometimes uses terms that are unknown in national criminal procedure or redefine existing terms.

Member States also have to ensure the practical integration of the EPPO into their criminal justice systems. The EPPO is an ‘indivisible Union body operating as one single office with a decentralised structure’ (Art.  8(1), see also Art. 3(1)), in other words an EU-wide prosecution service with a central office in Luxembourg and decentralised offices at the national level (Art. 8(2)). Member States will inevitably navigate between the rules set by the Regulation and the margin of appreciation it leaves in order to respect the diversity of national criminal procedure (cf. Recital (15)).

The analysis below will first briefly summarise the main features of the EPPO to set the scene for the subsequent analysis. Then we will focus on some key issues in the implementation process, both in general and specifically for in Belgium, based on the outcome of an independent academic study conducted by the authors of this blogpost (the ‘academic team’) at the request of the Belgian Federal Public Service for Justice (i.e. the Ministery of Justice). This study was finalised in August 2019. Its findings may be useful for the other Member States which are in the process of making their systems compatible with the EPPO system. That said, the conclusions of the study on the (tremendous) implications of the EPPO for the prominent role of the customs and excises authorities in prosecuting customs offences will not be discussed in this blog post, as they are highly technical, mainly affect Belgium and Luxembourg, and would merit a separate contribution.

2. Key Features of the EPPO Regulation

The European Commission has pushed for the creation of the EPPO in order to achieve better protection of the Union’s financial interests. That protection of the Union’s financial interests (‘PFI’) is a competence shared between the EU (with the Commission Anti-Fraud Office OLAF conducting administrative investigations) and its Member States (where the protection is provided by both administrative and judicial authorities) (Art. 325 TFEU). The EPPO is meant to enhance the protection of the Union’ budget through criminal law enforcement at the EU Level.

The EPPO’s material competence is defined by referral to Directive (EU) 2017/1371 of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law (Art. 22(1)) (‘PFI Directive’). It includes for example fraud involving EU subsidies, VAT-fraud, customs fraud and other criminal offences like passive and active corruption. In addition, the EPPO will be competent for ‘any other criminal offence that is inextricably linked’ to those in the PFI Directive (Art. 22(3)), including offences regarding participation in a criminal organisation as defined in Framework Decision 2008/841/JHA (Art. 22(2)). Although the EPPO is not yet operational, the European Commission is already thinking about extending the material competence of this new EU body, which will require unanimity from the European Council (Art. 86(5) TFEU). A first suggestion was made in September 2018 with respect to terrorism offences. More recently, at the XXth World Congress of the International Association of Penal Law (13-16 November 2019), the Commission reveiled its intention to adopt a legislative proposal to expand the EPPO’s competence by 2025, especially in the field of economic and financial law (e.g. corruption (without damaging the EU budget), money laundering and market abuse). In our view, an early expansion (especially to terrorism) would complicate and even compromise the functioning of the EPPO, which will already be a delicate exercice. But this ongoing reflection at the EU level also shows how important it is to get the EPPO implementation right from the outset.

In terms of organisational structure, the EPPO is a new body of the EU with a central and a decentralised level. The College, the Permanent Chambers, the European Chief Public Prosecutor, the Deputy European Chief Prosecutors, the European Prosecutors (one per participating Member State) and the Administrative Director will operate from the EPPO headquarters in Luxembourg (Art. 8(3)). The European Delegated Prosecutors (at least two per participating Member State; Art. 13(2)) will operate at the Member State level (Art. 8(4)). The Central Office will make strategic decisions and ensure general oversight of the EPPO’s activies (role of the College’s; Art. 9(2)) and monitor, direct and supervise EPPO investigations and prosecutions (role of the Permanent Chambers and the European Prosecutors; Art. 10(2) and 12(1)). The actual investigation and public prosecution will be conducted at the national level in the participating Member States by the European Delegated Prosecutors (‘EDPs’) who are (also) members of the national prosecution service (Art. 13(1), paras 2-3). To that end, they will rely on the national authorities, including the police. In light of the principle of sincere cooperation, all national authorities (as well as relevant EU bodies) ‘should actively support the investigations and prosecutions of the EPPO’ (Recital (69)). Put simply, the EPPO can only function smoothly if it can rely on the active cooperation of national authorities. It would be unrealistic to think that two or three EDPs can conduct the EPPO investigations themselves, without the help of (existing) national authorities.

The EPPO’s independence is affirmed by Article 6(1) EPPO Regulation. Its prosecutors and staff cannot seek or take instructions from any person or institution outside EPPO and they will act ‘in the interest of the Union as a whole.’  Moreover, the EPPO is required to conduct its investigations in an impartial manner, implying that it should ‘seek all relevant evidence whether inculpatory or exculpatory’ (Art. 5(4)). While understandable in light of the mission and objectives of the EPPO (i.e. ensuring a better protection of the EU’s financial interests), these two features do not necessarily match the status and role prosecutors play at the national level. In many Member States, prosecutors have some link with the Executive: criminal justice policy is often co-constructed by (chief) prosecutors and the Minister of Justice, and the latter in some cases even has the possibility to give instructions to prosecutors in specific criminal investigations (infra, Section 5.1). In the context of the European arrest warrant, the Court of Justice of the EU (‘CJEU’) concluded that German prosecutors are not sufficiently independent to issue such warrant as they might be influenced in their decision by the Executive (joint cases OG and PI of 27 May 2019, § 88), whereas the Prosecutor General of Lithuania does provide sufficient guarantees in terms of independence (case PF of 27 May 2019, § 56). Obviously, this case law might inspire the CJEU if ever it has to assess EDPs’ independence in the future.

These EDPs have ‘the same powers as national prosecutors in respect of investigations, prosecutions and bringing cases to judgment’ (Art. 13(1), § 1). Moreover, according to Article 17(2), the EDPs will be ‘active members of the public prosecution service or judiciairy of the respective Member States which nominated them,’ and this for the entire duration of their term. In addition, Article 13(3) states that the EPDs ‘may also exercise functions as national prosecutors, to the extent that this does not prevent them from fulfilling their obligations under the Regulation.’ This raises the delicate issue of the so-called ‘double hat’ of EDPs (infra, Section 5).

Another difficult question relates to the relation between EDPs and national authorities, in particular national judges intervening at the pre-trial stage. This relation is addressed rather implicitly in the Regulation. By contrast, the Regulation clearly states that the competent national judge is determined on the basis of national law (Art. 36(5).  As we will explain below (infra, Section 6), some provisions of the Regulation might, at first sight, seem to create an insurmountable obstacle for legal systems that still have an investigating judge, like Belgium, Luxembourg,[3] France[4] and Spain. In our opinion, though, this interpretation is incorrect .

3. Even a Regulation Requieres Some Implementation

After this short overview of the EPPO Regulation, we will now turn to questions relating to its implementation in national law. Indeed, despite the fact that the EPPO is – as Article 86(1) TFEU required – establised by a regulation, some national implementing rules and adjustments of national law will still be necessary. For instance, Member States have to decide on how to practically organise the decentralised level of the EPPO, namely the integration of the EDPs in the existing national structures (infra, Section 5). They will also have to adopt rules on the selection and nomination of the EDPs, as Article 17 of the EPPO Regulation only deals with the appointment of the EDPs nominated by the Member States.

Still, the impact on the national legal systems of the EU’s use of a regulation should not be underestimated. In a field of law where EU law has so far pursued approximation by means of directives, it will require some mental adjustment from all legal actors involved. The EPPO Regulation will have direct application in the legal order of the participating Member States and it will have primacy over conflicting national rules. National rules and procedures, even constitutional ones, will have to be interpreted such as to ensure conformity with the EPPO Regulation. This also means that the role of national parliaments is considerably limited.

In the course of research conducted by the authors for the Belgian Federal Ministry of Justice, it became clear that most of the questions that accompagny the ‘irruption’ of a new player on the criminal justice field could be answered by relying directly on the text of the EPPO Regulation, without any implementing national law, in combination with regulation-conform interpretation of existing Belgian legislation.

4. Need for Interpretation to Overcome Language and Terminological Complications

Such interpretative efforts will definitely be necessary. As indicated in the introduction, the EPPO Regulation sometimes opts for terms that are unknown under national criminal procedure. For instance, the Dutch version of the Regulation uses the term Europees (gedelegeerd) aanklager, while the notion of aanklager is unknown in the two systems with Dutch as an official language[5]: (Trilingual) Belgium indeed uses titles and terms for its national public prosecution service that are borrowed from the French system, even in the Dutch language (procureur and advocaat-generaal), whereas the Netherlands prefers the terms officier van justitie and hoofdofficier van justitie. It is thus surprising that the Dutch version of the Regulation opts for the term aanklager, all the more because it stresses the accusatory function of a public prosecution service, while in practice this is just one of the many tasks performed by modern prosecutors. As it happens, the EPPO Regulation itself explicitly emphasizes that the investigation and the independent gathering of evidence both against and in favour of the suspect are also core tasks of EPPO prosecutors (Art. 5(4); supra, Section 2).

Furthermore, a number of terms used in the EPPO Regulation receive a different meaning from their traditional counterpart in national legal idiom. For instance, the term dismissal in Article 39 is translated as classement sans suite (in French) and sepot (in Dutch), but clearly covers a wider range of situations than the notions classement sans suite and sepot under national law, which only puts a provisional end to the criminal investigation. As a fact, dismissal indeed includes situations where the prosecution is irrecevable or onontvankelijk (e.g. Art. 39, a), d) and e)), a non-lieu or buitenvervolgingstelling (a decision by an investigative court at the end of a judicial inquiry) and even an acquittal (acquittement or vrijspraak, referring to a decision of a trial court; e.g. Art. 39, g)). These situations are legally speaking quite distinct from a classement sans suite and a sepot as they put a definitive end to the criminal proceedings and (usually) trigger the application of the principle of non bis in idem. Regulation-conform interpretation should solve this terminological confusion, but practitioners should, of course, be most careful when using these terms in order to not mix up different meanings and legal consequences.

5. Seamless Integration of EDPs in the National Legal Order

5.1. Stand-alone or Integrated?

The EPPO is a hybrid investigative and prosecution office. Its EDPs belong to a Union body, they receive concrete instructions from the central level which also sets out policy, but they are located at the national level and are part of the national prosecution service. As explained above (Section 2), they have the same powers as national prosecutors, will cooperate directly with national authorities and may even investigate national cases to the extent that this does not hamper the exercise of their EPPO work. In view of this unprecedented structure, Member States face fundamental choices regarding the integration of the EDPs in the national legal order. Basically, they have two main options. They can create a completely new miniature stand-alone public prosecution service for the EDPs, separate from the existing (national) public prosecution service or they can decide to integrate the EDPs in the existing system with some punctual adjustments, but without creating a full set of new separate rules,[6] separate investigative capacity (manpower, tools, databases, etc.) and separate lines of communication. Within this second option, there are a number of suboptions, depending on how the national legal system is organised, as we will explain below on the basis of the Belgian legal framework.

5.2. Background: Current Belgian Context

In total, according to recent figures, nearly 900 prosecutors and some 2,600 support and administrative staff work for the Belgian prosecution service. Belgium’s public prosecution service is organised in 14 districts, divided over 5 circuits headed by 5 prosecutors-general. The overall prosecution policy is set out by the College of prosecutors-general, in dialogue with the Minister of Justice. As in many Member States, the situation of Belgian prosecutors in the trias politica is ambiguous, somewhere in between the Judicial and Executive branches. The Minister of Justice cannot not give negative injunctions (i.e. forbidding a prosecutor to investigate), but positive ones are possible, i.e. the Minister can order the public prosecution service to start a investigation, to press charges or to appeal a decision (Art. 151, § 1 Belgian Constitution[7]). This only happens exceptionally and should be contextualised in the light of the public prosecution’s wide discretion not to prosecute (the so-called principe d’opportunité) – there is indeed no mandatory prosecution under Belgian law, unlike some other EU Member States.

In response to problems of coordination in cross-juridiction cases and the increased need for specialisation and internal cooperation, Belgium has set up a federal (actually, national) prosecution office for the whole territory, which exists next to the 14 existing local prosecution offices. It is headed by the (chief) federal prosecutor (federale procureur or procureur fédéral, cf. a Bundesstaatsanwalt), who is bilingual (i.e. Dutch- and French-speaking); there is no federal prosecutor-general. The federal prosecution office’s tasks have been extended on several occasions; they include, for instance, the coordination of criminal proceedings, the facilitation of international cooperation, and the prosecution certain offences such as terrorism, illicit arms trade, and criminal organisations. Currently, the chief federal prosecutor disposes of 32 federal prosecutors.[8] At present, two members of the federal prosecution office function as the national point of contact for OLAF in Belgium.

5.3. The Option Most Fitting for the Belgian Context: Double Hat for the Chief Federal Prosecutor as Columbus’ Egg

One of the questions submitted to the academic team was how to integrate the EDPs in the Belgian legal order, considering the specificities mentioned in Section 5.2.

We concluded that if Belgium wants the EPPO to function well and independently – both obligations under EU law – and cause as little friction with the rest of the prosecution service as possible, the best option is embedding it in the existing Belgian legal system. The EDPs should be integrated in the Belgium’s federal prosecution service, with the chief federal prosecutor always as one of the delegated European prosecutors. That option for a ‘double hat’ has a lot of advantages when compared to the two alternatives that were considered by the academic team.

The first alternative, and definitely the worst choice, would be a miniature stand-alone prosecution office that is composed of two or three EDPs and seperated from the existing structure of the Belgian public prosecution service. It may look good on paper, because it would create a single purpose organisation that would look very independent. Yet, in reality, there are so many disadvantages that it might even conflict with Belgium’s obligation of loyal and sincere cooperation with the EU. It would lack the staff and means necessary to realise the ambitions of the Regulation: two or three EDPs cannot run a 24/7 office that would adequately deal with its numerous partners. Expanding such a separate prosecution office to a fully fledged unit with ample staff and its own investigative capacity (detectives) would require an enormous investment from the Belgian authorities. This alternative would also entail a loss of efficiency and generate continuous friction within the police and the (existing) public prosecution service given that specialised (i.e. financially skilled) detectives are scarce and coveted throughout the system. A whole series of extra legislative changes and new procedures would have to be introduced to solve those conflicts.

The second alternative would be to embed the Belgian EDPs in the Belgian federal prosecution office but without EPPO membership of the Belgian chief federal prosecutor. The academic team was ascertained that this would very likely generate unnecessary tensions within that federal prosecution office. The major fear is that if that EPDs would act autonomously, as small ‘islands’ or independent ‘satellites’, this will become a source of resentment and divisiveness, a direct threat to the esprit de corps cherished by an institution which was created to supersede existing divisions – because the EDPs will draw on the human and financial resources of the federal prosecution service. Indeed, those who hope that the introduction of the EPPO at the Belgian level will be accompanied by a substantial increase in budget for the prosecution service are almost certainly deluding themselves. The EU will only pay the wages and expenditure of the EDPs when they act in the framework of the EPPO Regulation and exceptionally costly investigative measures (Art. 91(5) and (6)). All other means logically must come from the Member States, as they have always been obliged to protect the Union’s financial interests (supra, Section 2).

The academic team therefore concluded that, instead of seriously upsetting the Belgian system, a small upgrade of the existing one could suffice for the EPPO to function smoothly. The Belgian chief federal prosecutor would become a Belgian EDP and thus in EPPO matters subjected to the EPPO hierarchy. He/she would always be able to rely on the normal rules, structures and channels already up and running for the Belgian federal prosecution office itself to mobilise the necessary people (both prosecutors and qualified investigators) and means. This includes the possibility to delegate tasks to other members of the public prosecution service. To be clear, such a policy choice would not require the chief federal prosecutor to handle many EPPO files personally. That task would mainly fall to the two other EDPs, one dutch-speaking and one french-speaking. However, having the chief federal prosecutor on board does mean that he/she would always have access to all EPPO-related information and will dispose of a direct line of communication with the EPPO headquarters in Luxembourg, i.e. directly with the European Prosecutor for Belgium and with the European Chief Prosecutor. This would allay the fear that the Belgian EDPs would be perceived as ‘infiltrators’ who can keep their actions and policy secret from their Belgian colleagues by hiding behind – real or fake – orders from the EPPO central office. If the Belgian EDPs are really working on sensitive cases, the chief federal prosecutor, as their colleague but also as a key player in the Belgian criminal justice system, can help them out.

At the same time, it would become impossible that local prosecutors dealing with a routine situation, for instance, a hearing on the pretrial detention of a suspect in an organised crime case with some PFI-aspect, would receive contradictory orders from the federal prosecution office (for the organised crime aspect) and the EDP (for the PFI aspect).

If their EPPO workload allows so, Belgian EDPs could also investigate national cases since they will have a ‘double hat’ (supra, cf. Art. 13(3) and 17(2)). This ‘double hat’ approach ensures maximum flexibility and efficiency, to the benefit of both the EPPO and the national system.

Does this mode of integration threaten the independence of the EPPO? We do not think so. All Belgian EDPs will, in EPPO cases, fully be subjected on the EPPO hierarchy and are bound to follow the EPPO prosecution policy. Regulation-conform interpretation will deny the Belgian Minister of Justice the possibility to order the Belgian EDPs to prosecute in a particular case; such instructions can only come from the central EPPO office. Nevertheless, in terms of investigative capacity, the EDPs will have to consult with their EDP-colleague, the chief federal prosecutor, and rely on continuous cooperation with the existing national prosecutors and police authorities. This is inevitable and also congruent with the hybrid structure of the EPPO (designed by the Member States in reaction to the Commission’s initial ‘federal’ proposal; supra, Section 1) and the principle of sincere cooperation. Indeed, the independence of the EPPO (as an office for overall policy and in specific cases for the delegated European prosecutors) should not be confounded with isolation of the EDPs in the Member States. As argued above, such isolation risks to render the EPPO a toothless institution and a source of continuous conflict with the Member States, also those that sincerely want to tackle EU fraud. It would be counterproductive and at odds with the ultimate objective of the EPPO.

6. The EPPO and National Judges

As indicated above (Section 2), one particularly delicate issue is the relation between the EPPO and national judges at the pre-trial stage. The EPPO Regulation does not explicitly address this issue (supra, Section 2). It does not introduce an EU juge des libertés or juge d’instruction. For its criminal investigations, the EPPO has to rely upon national legislation (see for example Art. 30 and 41 and Recital 15).

6.1. No Obligation To Abolish The Investigating Judge…

In some inquisitorial systems (Belgium, France, or Luxembourg) a juge d’instruction (or investigative judge) may carry out pre-trial investigations. In particular, measures such as interceptions of telecommunications or searches of private premises often require in those systems the formal opening of a judicial inquiry (instruction or gerechtelijk onderzoek), which means that the investigative judge is in charge of the investigation and not the public prosecutor. There may be exceptions in cases where the proscutor can conduct a coercive measure through a ‘mini judicial inquiry’ (mini-instruction or mini-instructie), but even then, the juge d’instruction still decides on the opportunity of the measure, not only on its legality. The juge d’instruction is thus not merely an instance of judicial control over coercive investigative measures.

The system of judicial inquiries at first sight seems to go against the ‘philosophy’ behind the EPPO Regulation, which pictured investigations led by the public prosecutor. Article 28(1) of the EPPO Regulation could then be read as an affirmation that only the EDP can take the lead of the investigation:

‘The EDP handling a case may, in accordance with this Regulation and with national law, either undertake the investigation measures and other measures on his/her own or instruct the competent authorities in his/her Member State. Those authorities shall, in accordance with national law, ensure that all instructions are followed and undertake the measures assigned to them (…).’ (emphasis added)

However, other provisions of the EPPO Regulation favour a different analysis. According to Recital 15, the EPPO Regulation leaves the national rules regarding ‘the way in which criminal investigations are organised’ untouched (see also J. Pradel[9]). The history of drafting the EPPO Regulation also indicates that it was not the intention of the EU to radically change the relationship between public prosecutors and judges. Whereas former Article 18(1) of the Commission’s 2013 text proposal still put emphasis on the investigation being led by the EDP, this was tempered in the final draft. ‘Leading a case’ became ‘handling a case’, and references to national law and Recital 15 were added by the Council. In light of Recital 15, Article 28(1) EPPO Regulation should not be read as an obligation for Member States to limit the role of investigative judges in EPPO matters and thus to change ‘the way in which criminal investigations are organised.’ Instead, the emphasis should rest on conducting investigations ‘in accordance with national law.’ This reading is supported by Article 30 of the Regulation:

‘1. (…) Member states shall ensure that the [EDPs] are entitled to order or request the following investigative measures (…).

4. The [EDPs] shall be entitled to request or to order any other measures in their Member States that are available to prosecutors under national law in similar national cases, in addition to the measures referred to in paragraph 1.’ (emphasis added)

What follows from these provisions, is that EDPs will be able to give orders to national authorities (e.g. the police and administrative authorities) and request judges to authorise investigative measures that require, according to national law, the intervention of a judge. Indeed, it should be noted that, contrary to the Commission’s initial proposal, the EPPO Regulation does not set minimum requirements regarding the need of a judicial authorisation for intrusive investigation measures. In other words, it is up to national law to determine which investigative measures can be ordered by the EDP (under the same conditions as prosecutors in similar national cases) and which ones require the intervention of a judge. Whether this judge only intervenes punctually (like the juge des libertés) or takes over and leads the investigation from that moment onwards (like the juge d’instruction) is in our view neither imposed by, nor excluded by the Regulation.

The changes to the draft EPPO Regulation in the course of the negotiations and in the absence of explicit and unambiguous language to the contrary, brought the academic team to the conclusion that a criminal justice system with an investigative judge is compatible with the EPPO Regulation.

6.2. … But Cases Can Be Taken Away from Member States

Some adjustments will nevertheless be necessary for the reallocation (i.e. removal by the EPPO of cases from Belgium to continue or centralise them in another member state for prosecution (Art. 26(5)), and the reopening of closed cases (Art. 39(2)). In those specific situations, it seems that the Regulation does not allow Belgian judges to oppose the choice of the EPPO. Belgian judges will be obliged to acknowlege this. Therefore, a procedure similar to the one for cooperation with the International Criminal Court could be used. The Belgian Cooperation Act of 29 March 2004 provides in Article 8, § 2 for a ‘transfer procedure’ whereby the Court of Cassation, at the request of the International Criminal Court, takes away the case from a Belgian court in favour of the requesting tribunal.

6.3. Specific Domestic Reasons Why Belgium Should Opt for a Minimalistic Implementation and Not Create a Separate Type of Investigation for EPPO Cases

In particular with respect to the Belgian system, we believe it would be unwise to abolish the investigating judge immediately, but only in EPPO cases. In the absence of clear EU rules on this matter, such approach is likely to create serious constitutional problems. In recent years, the Belgian Constitutional Court has on several occasions (e.g. here and here) declared laws that reduced the role of the investigating judge unconstitutional. The Constitutional Court’s is increasingly critical of the differences in applicable safeguards between, on the one hand, an information or opsporingsonderzoek (i.e. a pre-trial investigation led by the public prosecutor, which represents some 90 to 95% of all criminal investigations in Belgium) and, on the other, a judicial inquiry (i.e. a pretrial investigation led by an independent and impartial investigating judge). For the most intrusive investigative measures (e.g. a house search or covert access to (stored) electronic communications or data), there is an obligation under Belgian law to start a judicial inquiry, with all its extra procedural safeguards. During such judicial inquiry, the prosecutor has access to the file at all times and is able to request investigative measures (and even to appeal a refusal of the judge), but the final decisions are always taken by a judge. At the end of the investigation, the investigating judge hands over the file to the prosecutor, who takes a stand on the further prosecution of the case and brings the case to the investigative court, which decides whether the case can be referred to the trial court. With respect to less intrusive investigative measures (e.g. a production order for traffic data), the investigating judge only intervenes punctually, at the request of the public prosecutor, and decides independently whether or not to autorise the measure (mini-instruction). Should the Belgian Parliament decide to abolish the investigating judge for EPPO cases, without an unambiguous EU obligation, and thus grant more powers to the prosecutors, but only in EPPO cases, it is highly likely that the Constitutional Court would annul this legislation. That is highly undesirable, because it would (potentially) undermine the validity of the investigative measures executed in all (pending) EPPO cases.

Eventually, only the CJEU can provide certainty, because the Belgian Constitutional Court would be obliged to seek a preliminary ruling. The European Commisison could also decide to start a procedure against Belgium. The CJEU could then take up its EU constitutional role and decide  once and for all. If it would confirm that the EPPO Regulation required abolition of the system of judicial inquiry, EU-conform interpretation of the Belgian Constitution would save the constitutionality of the new system. Suppose, however, that the CJEU would agree with the academic team and find that the EPPO Regulation does not create a clear and unambiguous obligation for Belgium to abolish the investigating judge. Then no EU obligation can be invoked to justify the unconstitutional distinctions made by the Belgian parliament. Dozens or hundreds of EPPO cases would needlessly be jeopardised. The Belgian parliament’s safest option is therefore not to adopt ‘an exceptional regime’ limited to EPPO cases only. One could even argue that its Union loyalty obligations require Belgium to refrain from such reckless legislation.

Of course, unconstitutionality under Belgian rules could also be avoided by immediately creating a unified system for all cases, both EPPO and others. In the long run, the current Belgian problem should disappear anyway. Indeed, a group of experts is preparing a fundamental reform of the Belgian criminal procedure and is close to finalising its work. The major overhaul of the current system would result in a single type of pre-trial criminal investigation, for all procedures, with judicial control over investigations lead by the prosecutor. That new Code of Criminal Procedure will, however, not be ready in time to meet the deadline for the EPPO (November 2020; supra, Section 1). Creating a specific, exceptional procedural regime in the meantime is not desirable, all the more since in real life, EPPO procedures will often be closely intertwined with non-EPPO offences and having different regimes in situations of overlap is a nightmare for practioners and puts transparancy, human rights and the EPPO’s legitimacy at risk.

Should the CJEU ultimately disagree with the Belgian reading, Belgium will have to amend its rules, but it would do so on the basis of a clear EU obligation. The investigations conducted and the evidence gathered in the meanwhile could however be used, because the Belgian procedure might have infringed the EPPO Regulation, but not the rights of suspects or defendants. And if the European Commission or the CJEU agree with the Belgian interpretation of the EPPO Regulation, there is no problem at all and Belgium will have avoided needless creation of a complex new system.

7. Conclusion

In the academic team’s view, the adaptation of national law to the EPPO Regulation should meet the following cumulative criteria:

• Minimal intervention, meaning as few changes to the law as possible in order to not upset the national system. This was also the intention of the EU legislator (Recital 15). In the Belgian legal context, this criterion is all the more relevant since a complete make-over of the Belgian criminal procedure – a new Code – with fundamental changes is being prepared by a panel of academic experts appointed by the (resigning) federal Minister of Justice;
• If different implementation options are available, preference is given to the simplest option and the one that creates as few conflicts or tensions as possible;
• And, finally, to the extent that conflicts are inevitable, allow for their smooth and quick resolution, by providing clear and straight lines of responsibility and authority.

From the many issues discussed in the academic study made at the request of the Belgian Ministry of Justice regarding the adjustment of Belgian laws, institutions and practices to the obligations under the EPPO Regulation, we singled out two key matters: the integration of the EDPs into the national prosecution service and the possibility of retaining the investigating judge.

We conclude, on the one hand, that smooth integration of the EPPO in the Belgian system could be achieved by systematically including the Belgian chief prosecutor, who is already responsible for coordination of investigations at the national level, among the EDPs. We do not see this constellation as a threat to the EPPO’s independence, because in EPPO matters, all EDPs will bound by the instructions and policy of the EPPO headquarters. Including the chief federal prosecutor among the EDPs would make him or her a pivot that in the strongest possible way connects the EPPO with the national prosecution system and it is the best way to avoid tensions, rivalry or even rejection by national prosecution and police authorities. Any alternative would entail less integration and is also necessarily more complex from a legal perspective.

On the other hand, we believe that, until the CJEU should decide otherwise, the safest policy for the Belgium parliament would be to assume that the Belgium system of judicial criminal inquiries is compatible with Article 28(1) of the EPPO Regulation and should therefore also be applied in EPPO cases.

In addition to the above questions, many other issues could be discussed, such as how the rules on civil and criminal liability will be applied to the EPPO prosecutors and the EPPO as a legal entity, or the relations between participating Member States and non-participating Member States. But those questions will only have to be addressed once the EPPO is operational.

For now, it is important for Member States to make the best of this ‘forced’ marriage in order to achieve the political goals set by EU decision-makers, that is, to effectively improve the protection of the Union’s financial interests. To that end, a seamless integration of the EDPs in the national legal order, without upsetting the existing rules of criminal procedure, is crucial.

[1] At first 16 Member States (April 2017), then 20 Member States (June 2017) and ultimately 22, after the accession of the Netherlands and Malta early August 2018.

[2] As pointed out by a 2018 study conducted by E. Sellier and A. Weyembergh for the LIBE Committee of the European Parliament, this situation hampers cross-border cooperation in criminal matters, shows ‘mutual trust’ is not automatic but ‘must be “earned”’, and calls for further legislative and non-legislative measures (p. 9-10).

[3] See e.g. J. Nies, « La répartition des rôles entre le Parquet européen et les parquest nationaux », Journal des tribunaux Luxembourg, 2019, p. 11-14.

[4] See e.g. J.-B. Jacquin, « Les pouvoirs hors normes du futur parquet européen », Le Monde.fr, 19 August 2019.

[5] It is, however, used in the Dutch translation of the Rome Statute for an International Criminal Court for the (ICC) Prosecutor: https://wetten.overheid.nl/BWBV0001489/2002-07-01#Verdrag_2 .

[6] All the more since the Regulation refers very often to national law, supra nr. 1.

[7] ‘Le ministère public est indépendant dans l’exercice des recherches et poursuites individuelles, sans préjudice du droit du ministre compétent d’ordonner des poursuites et d’arrêter des directives contraignantes de politique criminelle, y compris en matière de politique de recherche et de poursuite.’ (Emphasis added)

[8] Half of them is Dutch-speaking, the other half is French-speaking. That said, within each language group, 1/3 qualifies as bilingual.

[9] J. Pradel, « Le parquet européen est-il compatible avec les juges nationaux de la mise en état en affaires pénales ? », Recueuil Dalloz, 2019, p. 650.