COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT Accompanying the Proposal for a Council Regulation on the establishment of the European Public Prosecutor's Office
/* SWD/2013/0274 final */
|Bilingual display: EN|
COMMISSION STAFF WORKING DOCUMENT
Proposal for a Council Regulation
on the establishment of the European Public Prosecutor's Office
Table of Contents
1........... Introduction. 5
2........... Procedural issues and consultation of interested parties. 8
2.1........ Consultation and expertise. 8
2.2........ Internal consultation and scrutiny of the Impact Assessment 10
3........... Problem Definition. 10
3.1........ What is the problem?. 10
3.2........ Which are the drivers behind the problem?. 12
3.3........ Problem definition: conclusions. 22
4........... Right to act, subsidiarity and fundamental rights. 24
4.1........ Legal basis. 24
4.2........ Subsidiarity and need for EU action. 25
4.3........ Fundamental Rights. 27
5........... Objectives. 27
6........... Policy options and their impact 30
6.1........ Overview of policy options. 30
6.2........ Policy option 1: Base-line scenario - No policy change. 30
6.3........ Policy option 2: No new regulatory actions at EU level 32
6.4........ Policy option 3: Strengthening of the powers of Eurojust 32
6.5........ Policy option 4a: Creation of an EPPO entity within Eurojust 34
6.6........ Policy option 4b: Creation of a College-type EPPO.. 35
6.7........ Policy option 4c: Creation of a decentralised integrated EPPO.. 35
6.8........ Policy option 4d: Creation of a centralised EPPO.. 36
6.9........ Horizontal issues for options 4a-4d. 37
7........... Assessment of policy options. 38
7.1........ Status quo (policy option 1) – Baseline scenario. 40
7.2........ No new regulatory actions at EU level (policy option 2) 42
7.3........ A strengthened Eurojust (policy option 3) 44
7.4........ Creation of an EPPO entity within Eurojust (policy option 4a) 46
7.5........ College-type EPPO (policy option 4b) 49
7.6........ EPPO with decentralised integrated organisation (policy option 4c) 52
7.7........ EPPO with centralised hierarchical organisation (policy option 4d) 56
7.8........ Horizontal issues. 58
8........... Comparative assessment 59
9........... Enhanced cooperation. 61
10......... Monitoring and evaluation. 62
Executive Summary Sheet
Impact assessment on the Regulation of the Council on the establishment of a system of European Prosecution for protecting the union's financial interests: the European Public Prosecutor's Office (EPPO)
A. Need for action
Why? What is the problem being addressed?
The Union's financial interests are still not protected sufficiently: fraud, corruption and other offences affecting the Union's budget have reached disturbing levels in some areas and the lack of vigorous enforcement has helped a sense of impunity to emerge among fraudsters. The Union's current actions to protect its financial interests include administrative investigations, controls and audits, as well as legislative action, including the Commission's proposal for a Directive on the fight against fraud to the Union's financial interest by means of criminal law and the reform of OLAF, but do not address the deficiencies identified with respect to the investigation and prosecution of criminal offences related to the protection of the EU's financial interests. For the purpose of this impact assessment it has been assumed that about 3 billion euro per year could be at risk from fraud. The most affected stakeholders are the law enforcement and prosecutorial authorities in the Member States.
What is this initiative expected to achieve?
The initiative is expected to strengthen the protection of the Union's financial interests through establishing a European Public Prosecutor's Office. Its establishment is expected to lead to a more consistent prosecution policy for crimes against the EU's financial interests, ending the current fragmentation. This will lead to an increase in the number of prosecutions of the perpetrators of crimes affecting the financial interests, leading to a higher number of convictions, a higher level of recovery of illegally obtained funds and increased deterrence. In addition, its independence will ensure that investigations and prosecutions of the relevant crimes will be taken forward and brought before national courts, without direct influence of national authorities.
What is the value added of action at the EU level?
The added value of establishing a European Public Prosecutor's Office is mainly to be found in a more consistent prosecution policy, as well as an increased number of prosecutions of crimes affecting the Union's financial interests. This is expected to increase the level of deterrence and therefore improve the overall respect for the applicable rules, as well as the level of recoveries of funds unduly paid. The EPPO will direct investigations and prosecutions in the Member States, ensure effective coordination of investigations and prosecutions, and reduce problems related to different applicable legal systems. The current system, where the Member States are solely responsible for such investigations and prosecutions, supported by Eurojust and Europol, is not efficient enough to deal with the high levels of relevant crime and associated damages.
What legislative and non-legislative policy options have been considered? Is there a preferred choice or not? Why?
Seven policy options have been considered, of which four involve the establishment of a European Public Prosecutor's Office. Maintaining the status quo, taking non-regulatory actions only or improving the functioning of Eurojust have all been considered to be not effective enough in addressing the problems identified – only the options for establishing an EPPO have been assessed as providing effective and efficient action. The options for establishing an EPPO vary with respect to their institutional set-up, ranging from creating a unit within Eurojust (4a), a college-type EPPO (4b), a decentralised integrated EPPO (4c) to a centralised EPPO (4d). According to the impact assessment, setting up the EPPO as a decentralised integrated European organisation, based on the national judicial systems (4c), offers the most benefits.
Who supports which option?
Consultations with stakeholders have demonstrated differences of views with respect to the different options for establishing the EPPO. Most agree that other actions would neither be efficient nor effective enough to address the problems identified. The exact institutional set-up of the EPPO has been the subject of discussions ranging back to the 90's, and these discussions have not led to a common view amongst practitioners on the preferred option. The main difficulties relate to the integration of the EPPO and its actions within national judicial systems, the law applicable to investigations and prosecutions under the authority of the EPPO, as well as the integration of the EPPO's work with existing institutions, in particular Eurojust and OLAF.
C. Impacts of the preferred option
What are the benefits of the preferred option (if any, otherwise main ones)?
All four options for establishing the EPPO are expected to bring benefits in terms of an increase in the number of prosecutions brought forward in national courts. Of these four options, only the decentralised and the centralised options are expected to bring significant benefits, with the decentralised option doubling the current number of convictions, and the centralised option reaching almost that number. In addition to an increase in recovery, the impact assessment conservatively assumes that a doubling of the number of convictions will lead to a reduction in damage of around 10%. This means that the decentralised option is expected to provide the most benefits: over twenty years these are projected to total €3 200 million. The centralised option is a close second with expected benefits of about €2 900 million over the same period. The benefits of the other options are much more limited than that.
What are the costs of the preferred option (if any, otherwise main ones)?
The costs of the different options for establishing the EPPO vary quite considerably. The most expensive option is the centralised one, which assumes that all investigations and prosecutions will be handled at the European level, leading to a higher number of required EU staff. The decentralised option does not entail as much costs, also because use is made to a large extent of resources existing in the Member States, at Eurojust and at OLAF. The costs for the centralised option over twenty years are expected to be over €800 million, whereas the costs for the decentralised option are expected to be about €375 million. These costs include all costs expected to arise from establishing a new European body.
How will businesses, SMEs and micro-enterprises be affected?
Businesses, SME's and micro-enterprises will not be directly affected through establishing a European Public Prosecutor's office.
Will there be significant impacts on national budgets and administrations?
Yes – the increase in the number of prosecutions will lead to an increase in the costs for court cases, legal assistance etc. In addition, national law enforcement authorities will need to get used to working together with the EPPO and its staff. However, no significant new investments in investigation or prosecution staff will be needed: the currently available staff is expected to work more efficiently and effectively under the EPPO's direction.
Will there be other significant impacts?
Yes. The EPPO must be established with full regard to the fundamental rights of the defendants, witnesses and other participants in its investigations and procedures. This includes a system of judicial review of its actions. There will also be an impact on relations with third countries, since the EPPO will need to cooperate with them in the course of its investigations.
D. Follow up
When will the policy be reviewed?
A statistical review of the policy is foreseen to take place within two to four years after the establishment of the EPPO.
COMMISSION STAFF WORKING DOCUMENT
Proposal for a
Council Regulation on the establishment of a system of European Prosecution for protecting the Union's financial interests: the European Public Prosecutor's Office
In times of fiscal consolidation the protection of the European Union budget is of special political, legal and economic relevance. Both the Union and the Member States have a duty "to counter fraud and any other illegal activities affecting the financial interests of the Union" as well as to "afford effective protection" to such interests. Despite this clear obligation directly imposed by subsequent EU treaties and already referred to in 1989 by the European Court of Justice, the Union's financial interests are still not protected sufficiently by many Member States: fraud, corruption and other offences affecting the Union's budget have reached disturbing levels in some areas and the lack of vigorous enforcement has helped a sense of impunity to emerge among fraudsters. Recent analyses confirm that hundreds of millions of euros of taxpayers' money continue to disappear due to such criminal activities. This concerns notably structural funds, the cohesion fund and areas of highly taxed products (e.g. cigarettes, alcohol etc.). Besides the Union itself, many citizens and companies are adversely affected by these offences, and the current budgetary restraint efforts undertaken by the Member States and the Union seem less credible to them if the Union and the Member States are unable to prevent such damage to the EU budget and effectively prosecute fraudsters.
Whereas both the Union and the Member States have an obligation to protect the Union's budget, in reality the Union has little control over the expenditure by Member States and virtually no power to intervene in cases of criminal misuse of the EU's funds. The vast majority of the EU budget is managed by national authorities (for example when they award public procurement grants financed through the EU budget) and any criminal investigations or prosecutions concerning offences affecting the Union's budget are within the competence of the Member States. Criminal investigations into fraud and other crimes against the EU budget are often hampered by divergent legislation and uneven enforcement efforts in the Member States. National law enforcement authorities, prosecutors and judges in the Member States decide in accordance with priorities set by national criminal policy and on the basis of national criminal law competences and procedural rules whether and, if so, how they intervene to protect the Union's budget. Consequently, the level of protection of the Union's financial interests differs significantly from one Member State to another. The fact that the rate of successful prosecutions concerning offences against the EU budget varies considerably across the EU from one Member State to another (from 19,2% to 91,7%) shows a gap in the existing protection mechanisms and calls for corrective measures.
Links between crimes affecting the Union's financial interests and organised crime should also be considered. In accordance with the findings of the 2011 OCTA report, organised crime is a threat to the citizens and the economy of Europe, to its businesses and state institutions. Criminals operate easily across borders, and organised crime is becoming increasingly diverse in its methods, groups structures, and its impact on society. Crimes affecting the Union's financial interests both at the national and the EU level attract organised criminal groups due to the low levels of detection and prosecution of cases, as discussed further below.
The scale of fraud against the EU budget Data collected and analysed by the Commission identify “suspected fraud” averaging about €500 million in each of the last three years, but there are good reasons to believe that the actual amount of fraud is significantly higher. Rather obviously, figures on reported fraud cannot include fraud that is not detected. Moreover, not all Member States distinguish between “irregularities” and “fraud” – indeed, six Member States reported zero fraud affecting the spending of EU funds during in the area of Cohesion Policy for the programming period 2007-2013. The preparatory study for this impact assessment (see annex 3) examines the potential size of this “dark figure”. On the revenue side, VAT fraud and cigarette smuggling have been estimated to each cost the EU budget some €1 billion per year. As regards spending, the preparatory study for this impact assessment estimates that in a “low-risk” scenario, damages in the area of agricultural and structural funds could amount to €4.1 billion each year. Taking into account the weaknesses in the available data, for the purposes of this Impact Assessment it has been assumed that about €3 billion per year could be at risk from fraud. Given the weaknesses in the available data and the difficulties inherent in measuring the scale of the criminal activities that are undetected, the true figure, however, cannot be calculated precisely.
The Union's current actions to protect its financial interests are manifold, but they have a single aim: to ensure that the limited financial resources of the Union are used in the best interests of EU citizens. This is indispensable for the legitimacy of its expenditure and for ensuring public trust in the Union. Such actions include administrative investigations, controls and audits, as well as legislative action. The most recent of these actions is the Commission's proposal for a Directive on the fight against fraud to the Union's financial interest by means of criminal law. This proposal addresses the side of substantial criminal law on fraud and aims at ensuring, in particular, an appropriate level of sanctions. However, in order to ensure the functioning of the EU system to prevent and sanction fraud, it is also indispensable to ensure that these actions are actually applied in practice. Even the most efficient controls and the best possible legal framework will not produce results without effective investigation and prosecution measures which ensure that the perpetrators of the crimes concerned are actually brought to trial and sanctioned.
The previous initiatives do not serve that purpose. In particular, the reform of the European Anti-fraud Office (OLAF) only aims at improving the efficiency and transparency of the current administrative investigations but cannot be expected to have any substantial impact on the level of criminal investigation and prosecution of offences in the area of EU fraud.
There is currently therefore a significant gap in the "enforcement cycle" (see picture on page 12). This is the focus of the current report.
This Impact Assessment report will demonstrate that addressing these issues requires the setting up of a strong, effective and integrated European enforcement regime against fraud and other illegal activities affecting the Union's financial interests. The Union's actions should seek to put an end to the fragmented enforcement regime in the Member States and ensure a coherent approach to dealing with European fraud cases throughout the Union, from detection to investigation, and from prosecution to judgment. Only by overcoming the current legal and institutional barriers to fighting crimes against the EU's budget by national law enforcement authorities and justice systems will the Union be able to ensure efficient protection of its budget. The gaps identified in the enforcement cycle need to be closed.
Actors at European and national level involved in actions to fight offences against the EU's financial interests.
Eurojust || Established in 2001, Eurojust is the European Union agency for cross-border judicial cooperation in criminal matters. It plays an important role in cross-border cases involving the financial interests of the EU. Eurojust does not have the power to start criminal investigations or prosecutions in the Member States. Eurojust’s role is to stimulate and improve the coordination of investigations and prosecutions between the competent authorities in the Member States and to improve the cooperation between the competent authorities of the Member States. Eurojust supports in any way possible the competent authorities of the Member States to render their investigations and prosecutions more effective when dealing with cross-border crime. Eurojust may ask the competent authorities of the Member States concerned: · to investigate or prosecute specific acts; · to coordinate with one another; · to accept that one country is better placed to prosecute than another; · to set up a Joint Investigation Team; or · to provide Eurojust with information necessary to carry out its tasks. The operation of Eurojust is the responsibility of the College of Eurojust, which is composed of one National Member from each of the EU’s Member States. National Members are seconded in accordance with their respective legal systems. The exact status and powers of each National Member is defined by the national legislation of their appointing Member State, which also determines how long they serve.
Europol || Established in 1997, the European Union's agency for law enforcement cooperation supports investigations into offences affecting the EU’s financial interests through its analysis tools and information exchange networks. Europol has no operational powers – its staff cannot by themselves conduct criminal investigations.
OLAF || OLAF was established in 1999 as the European body responsible for protecting the financial interests of the Union by combating fraud, corruption and other illegal activities. OLAF exercises its mission by conducting administrative anti-fraud investigations and supporting the Commission in the implementation of fraud prevention and detection policies. OLAF is a central office within the European Commission, which receives reports on fraud and irregularities inter alia from the Member States' authorities and conducts administrative investigations into suspected cases of fraud and other offences affecting the EU’s financial interests, either on its own initiative, using factual information, including from private sources, or on request by a member state or EU institution. It also collects data on these cases. However, OLAF does not have competence with respect to criminal investigations.
Member State authorities || Investigate and prosecute these offences, as well as bring offenders to judgment.
2. Procedural issues and consultation of interested parties
2.1. Consultation and expertise
Discussions on how to best strengthen the investigation and prosecution of offences against the Union's financial interests, for example through the creation of a European Public Prosecutor's Office (EPPO), have been going on for more than a decade. There are already numerous institutional documents, studies and independent analyses of the relevant issues available. To a large extent, the current report builds on this research. However, these studies, much as the Treaty itself, leave a number of technical, legal and political issues open. Therefore, two studies conducted by an external contractor were carried out on behalf of the Commission with the objective to prepare this impact assessment. Preparatory consultations in view of the current report have therefore covered these issues, based upon a large panoply of options as regards the institutional, organisational and operational set-up of a European system for investigation and prosecution of the relevant offences.
At the beginning of 2012, two questionnaires were published and distributed, one to justice professionals and another to the general public, respectively. A large number of detailed replies were sent to the Commission. In general, the replies were positive towards taking new actions to strengthen the material and procedural framework to counter offences affecting the EU’s financial interests, and most also expressed support for the idea to set up an EPPO. A number of more detailed suggestions, concerns and questions were also voiced (see Annex 1), in particular on the relationship between the EPPO and national prosecution authorities, the competence of the EPPO to direct and coordinate investigations at national level, or the possible difficulties with any harmonised European rules of procedure in the EPPO's proceedings.
In parallel, field research has been conducted in a number of Member States, as part of the external study in support of this report.
In addition, throughout 2012 and at the beginning of 2013, a number of discussions or meetings took place at European level:
· The network of Public Prosecutors or equivalent institutions at the Supreme Judicial Courts of the Member States, Budapest, 25-26 May 2012.
· Conference: A Blueprint for the European Public Prosecutor's Office? Luxembourg, 13-15 June 2012. The conference gathered experts and high level representatives from academia, EU institutions and Member States.
· Vice-President Reding's consultation meeting with Prosecutors General and Directors of Public Prosecution from Member States, Brussels, 26 June 2012. The meeting permitted an open discussion on specific issues regarding the protection of the Union's financial interests.
· On 18 October 2012, the Commission organised a consultation meeting on issues relating to a possible reform of Eurojust, in which questions related to the setting up of an EPPO were also discussed with representatives of Member States. The meeting generally supported establishing a close link between Eurojust and the EPPO.
· The 10th OLAF Conference of Fraud Prosecutors, Berlin, 8-9 November 2012, was an opportunity to explore the ways in which national prosecutors would interact with the EPPO, if set up.
· The informal consultation held on 26 November 2012 with defence lawyers (CCBE and ECBA) looked at procedural safeguards for suspects and made useful recommendations in that regard.
· ERA seminar "Towards the European Public Prosecutor's Office (EPPO)", 17 and 18 January 2013.
· Meeting of the Commission Expert Group on European Criminal Policy, Brussels, 23 January 2013.
· Further consultation meeting with ECBA and CCBE, Brussels, 9 April 2013.
Also, numerous bilateral consultation meetings with Member States’ authorities have taken place over the second half of 2012 and the beginning of 2013.
2.2. Internal consultation and scrutiny of the Impact Assessment
An Interservice Steering Group was created involving representatives from DG Justice, OLAF, DG DEVCO, DG AGRI, DG BUDG, DG EAC, DG EMPL, DG ENTR, DG HOME, DG MARKT, DG MOVE, DG RTD, DG SANCO, the Secretariat-General and the Legal Service. ISSG meetings were held on 17 September 2012, 26 November 2012 and on 30 January 2013. At the meetings and in subsequent communication with individual DGs, comprehensive feedback was received which has been taken into account throughout this report.
This Impact Assessment was examined by the European Commission's Impact Assessment Board on 10 April 2013. Further to the IAB's recommendations, additional information, explanations and data were introduced in this document. In particular, the problem definition has been redrafted to focus more on the core problems of inadequate investigation and prosecution of offences against the Union's financial interests, and the reasons why these are not addressed through current initiatives. Also, the objectives have been simplified, and the cost-benefit analysis has been further explained to show more clearly the reasons underlying the differences in the benefits of the various options. In addition, more information was included on consultation of stakeholders, information on a discarded option was added, and some horizontal issues which did not affect the assessment and comparison of the options were taken out to make the intervention logic easier to understand, while the contribution of others to addressing the problems has been clarified. Finally, a separate Annex 2 was added which better explains the relations between the problems, the objectives and the options.
3. Problem Definition
3.1. What is the problem?
As shown under point 1, every year at least several hundred million euros are fraudulently diverted from their intended purpose. Only a small fraction of these losses are ever recovered from the criminals. These figures show that the financial interests of the European Union are insufficiently protected from fraud. In fact, the Commission's annual statistics (including those of OLAF) demonstrate that while fraud against the Union's financial interests is pervasive and causes substantial damage every year to the tax payer, national criminal enforcement efforts lag behind. In particular, OLAF's cases which are transferred to national investigation and judicial authorities are not always equally effectively followed-up.
The issues described in the introduction to this report can be visualised through a simple diagram, which shows the relations between the various issues – the enforcement cycle referred to above. This diagram clarifies how these different issues reinforce each other, leading to a vicious cycle in the current organisational and institutional structure: there are high levels of crimes against the financial interests of the Union, of which only a certain amount is detected. A low level of detection leads to an even lower level of investigation, since not all detected crimes will be investigated. A low level of successful investigations leads to an even lower level of prosecution, since not all investigated cases will be prosecuted. Finally, not all prosecuted cases lead to convictions and recovery of the proceeds of crime. Low levels of recovery lead to low levels of deterrence of criminal activities, leading to a higher level of EU-fraud, and so on. In order to break this vicious cycle and achieve higher rates of recovery and deterrence, it is therefore crucial to convict a larger number of offenders, through more effective investigation and prosecution of offences against the EU’s financial interests.
Figure 3.1. Enforcement cycle
Key areas for potential direct impact of EU Action
Clearly not all of these issues can be tackled through action at the EU level: detection remains within the remit of national law enforcement and adminstrative authorities, and convictions remain under the sole power of national courts. However, detection levels could increase if detected offences were to be investigated and prosecuted in a consistent and systematic manner, thereby generating better knowledge of the fraud phenomenon by the investigators and the prosecutors. Likewise, it can safely be assumed that larger amounts of money will be recovered if more cases are brought before national courts (for example by the EPPO), even if the rate of successful prosecution to dismissals would stay the same. Whilst these factors are important, and will be considered in the cost-benefit analysis of the different options, they will only be influenced partly through EU action. The remainder of this Section will therefore mainly focus on the drivers of the problem which need to be considered, and which are identified in the investigation and prosection phases of the enforcement cycle.
3.2. Which are the drivers behind the problem?
3.2.1. Limits of existing measures
The main reasons why enforcement is often weak or deficient are the absence of a European enforcement structure, the lack of continuity in enforcement action and the lack of an underlying common European prosecution policy. Whereas offences affecting the EU’s financial interests are genuine European crimes, the current institutional and legal framework suffers from a fragmented enforcement regime almost solely based on national responses, which depend on the priorities and resources of national investigation, prosecution and judicial authorities. The sections below identify the reasons why this is the case, provide further details on the actions already taken by the Commission to address these issues, and explain why these actions alone do not suffice.
Commission anti-fraud strategy
Prevention plays an important role in the protection of EU’s financial interests. Preventive measures may take the form of audits, ex-ante controls, fraud proofing of legislation and better coordination.
In June 2011, the Commission adopted the Commission Anti-fraud strategy (CAFS) which focuses on improving prevention, detection and the conditions for investigations of fraud. It also aims at achieving adequate reparation and deterrence, with proportionate and dissuasive sanctions. The CAFS is targeted at striking a balance between cost-effective control and simplification and adapting the anti-fraud measures in place to counter new fraud schemes. The Commission makes use of the experience derived from OLAF's investigations into alleged fraud and will develop anti-fraud strategies for specific Directorates-General.
Prevention efforts and actions through controls can be effective, but they need to be proportionate and in accordance with the objectives of simplification. They need to be complemented by an effective sanctioning system which reduces the temptation of fraud.
Finally, not all fraud can be prevented. This leaves the challenge of better criminal investigation and more efficient prosecution in cases involving the financial interests of the Union.
Limits of the current legal framework
A number of instruments are in place at EU level in order to ensure the protection of the financial interests across the Member States, among which the 1995 Convention on the protection of the EU's financial interests and its protocols, Regulation 1073/1999 on investigations conducted by OLAF, and Regulation 2185/1996 concerning on-the-spot checks and inspections.
However, in time, these instruments have shown their limits: reports demonstrate a fragmented implementation of the 1995 Convention and its protocols by the Member States. Experience in investigations accumulated over the last 14 years since OLAF’s establishment has shown that certain aspects needed improvements.
In July 2012 the Commission adopted a proposal for a Directive on the protection of the Union's financial interests by means of criminal law in order to address the limits identified above. The Directive will replace the 1995 Convention and aims at further approximation of definitions of relevant offences and of sanctions levels. This includes: definitions of offences (fraud, corruption, money laundering), harmonising definitions of additional offence types, minimum imprisonment ranges for particularly serious offences and harmonising the prescription period. However, whilst a sufficient level of sanctions and procedural rules to allow the application of these sanctions are essential, they can only be effective if they are complemented by measures ensuring that they are also effectively applied in practice. These measures are not part of the proposed Anti-fraud Directive.
Eurojust's mandate only allows it to coordinate and encourage investigations and prosecutions, and assist with information exchange. This means that Eurojust in general only becomes active where Member States themselves take the initiative on a certain case.
In the context of this impact assessment we are however looking particularly at the problem that such action often simply is not taken. And if a Member State prosecution service is reluctant to investigate or prosecute a case, Eurojust cannot compel it to do so. The National Members of Eurojust often lack the powers to ensure effective follow-up in the Member States, or if they do, they usually refrain from using the powers which they derive from national laws – most decisions on these sort of issues are arrived at through consensus.
As will be discussed in more detail below, even the most far-reaching reform of Eurojust is limited by the TFEU. Article 85 TFEU, which provides the legal basis for the establishment of Eurojust and lays down its mission does not provide the possibility to entrust Eurojust with conducting investigations: at the maximum, Eurojust could be given the power to initiate investigations, but not conduct them. The prosecution of cases before the national courts cannot be entrusted to Eurojust under Article 85 TFEU, which means that the current disparities and fragmentation of national prosecution efforts would not be solved.
The role of Europol is limited to providing intelligence and support to national law enforcement activities. It cannot ensure follow-up to its analyses in the Member States, nor direct national investigations. The powers of Europol are also limited by the TFEU. Under Article 88 TFEU Europol cannot independently investigate crime, and any operational action must be carried out by Europol in liaison and with the agreement of the national law enforcement authorities. Whilst the support functions of Europol are certainly important, these cannot substitute for the powers to independently investigate criminal behaviour.
A proposal for a Regulation on Europol was adopted by the Commission in March 2013, focusing on aligning Europol’s competences with the TFEU and to make it a hub for information exchange, while granting new responsibilities regarding training. It does not comprise police investigation and law-enforcement powers in the area of the protection of EU’s financial interests.
OLAF – administrative investigations
The powers of OLAF are limited to administrative investigations, and OLAF thus cannot directly lead investigations into crime sensu stricto, nor access information on criminal investigations. This may be a source of delays in the investigation of fraud and of shortcomings in the efficient use of resources. It also leads to problems concerning the use of evidence collected in administrative proceedings by OLAF in subsequent criminal proceedings, since evidence collected in administrative proceedings may not always be recognised as valid under the criminal procedural laws of the Member States. Investigation efforts therefore may need to be duplicated, performing evidence collection twice. Moreover, OLAF has no enforcement powers. OLAF may make a recommendation to a Member State for judicial action to be taken but it is for the national authorities to decide whether they take any action or not.
A proposal to amend Regulation 1073/1999 concerning investigations conducted by OLAF (OLAF reform) is under inter-institutional negotiation. This proposal improves the information exchange between OLAF and EU institutions bodies, agencies and offices (IBOA), as well as with the Member States and it provides better governance for OLAF and a set of procedural guarantees for the persons concerned by investigations, rendering its work more efficient and transparent. However, it does not provide OLAF with any additional means of action, in particular criminal investigation powers.
Table 3.1 Powers of EU actors
Limits of the powers of actors at EU level
Eurojust || The role of Eurojust is limited to support and coordination of the work of national judicial authorities. Whilst it may ask for the initiation of investigations, it cannot ensure follow-up at the Member State level, nor direct national investigations or prosecutions.
Europol || The role of Europol is limited to providing intelligence and support to national law enforcement activities. It cannot ensure follow-up to its analyses in the Member States, nor direct national investigations.
OLAF || OLAF does not have any competences with respect to criminal investigations, nor is it able to ensure follow-up to its administrative investigations at the national level.
As regards cooperation at Union level, mixed experiences have been reported regarding the cooperation with Eurojust and Europol, and between the Member States and OLAF. As indicated above, Eurojust and Europol do not always receive the information they need to be able to support the Member States. OLAF provides support to Member States through its ability to grant specialised technical and operational assistance as required by Article 7 of second Protocol to the Convention on the Protection of the European Communities’ Financial Interests. At the same time, OLAF's investigations are conducted subject to specific conditions, in particular when it comes to transmitting information to the national judiciary, including applicable data protection rules. For this reason, the cooperation with OLAF has also been criticised on occasion, in particular with respect to the long time it sometimes takes for OLAF to share information with national prosecutors. Some Member States also restrict the cooperation with non-judicial bodies like OLAF based on rules of judicial secrecy.
3.2.2. Low levels of investigation and prosecution of offences against the EU budget
Inconsistent follow up to OLAF investigations by Member States' authorities
OLAF's annual statistics demonstrate that the cases transferred to national investigation and judicial authorities are not equally effectively and efficiently prosecuted across the EU. In its eleventh operational report, OLAF analysed the judicial follow-up given by Member States to its cases over 12 years and found "very substantial differences between countries with respect to their capacity to bring EU-budget related judicial investigations and prosecutions to a conviction within a reasonable time". By the end of 2011, national judicial authorities had decided on 471 of a total of 1030 actions transferred to them by OLAF in the period between 2006 and 2011, whereas judicial follow-up was still pending regarding 559 cases. Of these 471 decisions, only 199 were a conviction by a criminal court. All other cases were either dismissed or a decision of acquittal was taken. In addition, there are very significant disparities between the Member States. In the period from 2006-2011, conviction rates at national level ranged from 19,2% to 91,7% (not including Member States with rates of 0% and 100%).
The table below indicates that only a few really well performing Member States such as Finland or Lithuania have a conviction rate over 90%, while many other Member States have a much lower prosecution (lower than 40%) and conviction rate (lower than 35%), thereby leading to a lack of effectiveness and of equivalence of the protection of EU’s financial interests across the Member States.
The fact that the average prosecution rate lies under 50% indicates that there are serious difficulties in achieving overall effectiveness of investigation and prosecution in the Member States.
The reasons why many Member States have an overall weak performance when it comes to investigating and prosecuting crimes affecting the Union's financial interests may be found in the comparatively limited chances of the national prosecutors to exercise their function within the national jurisdiction and come to a successful prosecution within a reasonable period of time. This is partly due to the complexity of the facts, which requires an in-depth understanding of the whole legal and administrative framework applicable to the EU fraud cases. It may also be explained by the fact that evidence collected outside of the national territory is frequently needed, requiring that instruments of international cooperation (MLA) are used, with a considerable risk of delay in the investigation. All these factors, including linguistic challenges, together lead to a situation with slowly functioning and less efficient prosecution systems.
At the same time, the success of the best performing Member States may come at a price for the protection of the EU’s financial interests as a whole: sometimes Member States achieve quick results by limiting the investigation and prosecution of the cases to the national territory, addressing the underlying criminal conduct only partially and negatively affecting the investigation of cross-border cases. It may also be noted that those Member States which achieve the highest conviction rates are relatively small in terms of population. Accordingly, the number of cases transmitted to them by OLAF for judicial action represents a relatively low percentage within the overall number of cases transmitted by OLAF to the Member States judicial authorities. Therefore, the challenges (at least in terms of number of cases) with which some of the best performing Member States have to deal with may be less complex than in the case of other Member States which have to prosecute a much higher number of fraud cases.
The differences which can be noted between the Member States when it comes to the performance of their judicial systems when prosecuting fraud may therefore be related to a number of factors, from the number and complexity of cases they have to cope with, to the degree of complexity of the procedural framework each Member State applies, the experience of the prosecutors in dealing with complex cases, or the resources they allocate to crimes affecting the EU's financial interests. It has to be noted that it is very difficult to assess the performance of the judicial systems of the Member States first of all because the data in the table below is limited to cases transmitted by OLAF (excluding Member States own investigations and prosecutions) and also because such analysis would imply an in-depth study of the overall performance of each judicial system, including the legal procedural framework applicable in each Member State, and of the crime situation on the ground. However, it is expected that an EPPO functioning on the basis of a common set of rules and guidelines would contribute to approximating the judicial practices of the Member States in the area of crimes affecting the EU's financial interests and to achieve a higher degree of performance of prosecution in such cases.
Table 3.2. Member States’ follow-up of cases transferred to them by OLAF, 2006-2011
The available data thus indicates strongly varying outcomes of judicial proceedings involving offences against the EU’s financial interests. These findings are mainly based on experience that OLAF has gained in its own investigative practice. OLAF investigations (about 500 cases per year) represent only a minority of cases out of the total number of cases (around 2500 per year) affecting the EU's financial interests which Member States investigate. There are, however, no indications that in cases where OLAF was not involved, the record of the Member States is more satisfactory than the one presented in the table above. In fact, OLAF cases should present higher chances of successful prosecution due to the fact that an administrative investigation into the facts has already been carried out. The fact that cases previously investigated by OLAF produce such modest results in terms of prosecution and conviction rates shows that a system where EU bodies only have administrative investigation powers is not fully effective and cannot achieve satisfactorily the objective of deterrence.
The sections below explain why these outcomes occur, and the issues related to investigation and prosecution.
Weak motivation to prosecute crimes affecting the EU’s financial interests
Prosecuting offences against the EU budget is considered too burdensome and therefore of secondary importance by the authorities in a number of Member States. As the chances of successful achievement of investigations within a reasonable short time are considered low, direct national, regional or local interests may take priority over European interests.
There are a number of factors which can lead the national authorities to have weak incentives to prosecute crimes against the Union’s financial interests:
· Authorities are often reluctant to comprehensively prosecute offences against the EU’s financial interests when the case has international components requiring evidence to be collected abroad; there is a perceived tendency to put complex European fraud cases involving cross-border cooperation on offences affecting the EU’s financial interest "at the bottom of the pile";
· It is sometimes difficult to identify a clear priority jurisdiction, and therefore both negative and positive conflicts of jurisdiction may arise;
· There may also be a lack of sense of ownership of such cases, as national authorities may wrongly count on authorities in other Member States to deal with the case.
Case example: National judiciary declines jurisdiction A Member State which from a European perspective appeared to offer the most appropriate jurisdiction, ultimately declined jurisdiction in a case forwarded by OLAF – in close collaboration with Eurojust – because the "relevant acts" were all committed outside its territory even though the potential suspect was a national of the respective Member State. The difficulty in this case was that it concerned a fraud with multiple transnational elements. The suspected person, alleged to have requested payments for services never provided, was residing in various Member States while committing the offence. The suspect was also working for several companies which were carrying out EU funded projects and those were based in different EU Member States. The projects were carried out in non-EU Member States. Further, the suspected person received payments in yet another Member State.
The majority of the experts consulted (57%) consider that cases involving the EU’s financial interests are not fully discovered by the national authorities and that they are neither investigated nor prosecuted adequately (64%). Sixty per cent of the experts felt that cases were sometimes hampered by the European dimension. This ratio is significantly higher than for other areas of crime. Probably as a consequence of this, 54 % admitted to sometimes limiting their investigation to the national aspects of a case even though they recognised its European dimension.
Case example: National judiciary restricts prosecution to national elements OLAF experienced that national judicial authorities put aside the transnational dimension to facilitate the investigation. This prevents the investigation from identifying the complete EU dimension of the offences committed. In one of its cases, OLAF forwarded information on traffic of Chinese products via Norway into the EU territory to a national judicial authority for prosecution. However, the prosecutor in charge (initially) decided to prosecute only the traffic within his Member State. He did not take into account the subsequent important traffic into several other Member States, because it was faster to close the case if leaving out the cross-border dimension.
In addition, stakeholders have reported that the institutional guarantees to ward off any undue interference in prosecution work, also required to properly investigate and prosecute offences affecting the EU’s financial interests, are not always at hand. For example, interviewees in some Member States where public governance issues exist have indicated that on the expenditure side of the EU budget conflicts of interests exist in the authorities implementing the funds and this may have further adverse effects. They assert that close networks of public administration officials, political leaders and business people exist, especially at the local and regional level, which hamper the effective protection of public money.
“Prosecutors cannot win the battle when national and commercial interests are involved that are too strongly intertwined.”
Insufficient and ineffective cooperation and information exchange
Tackling cross-border fraud cases requires closely coordinated and effective investigations and prosecutions. Current levels of information exchange and coordination at national and European level are not sufficient to effectively prosecute offences affecting the EU’s financial interests, despite the efforts of Union bodies, such as Eurojust and Europol. Regularly, investigations are undertaken in parallel in different countries, with limited coordination efforts. Even more importantly, the proceedings in one State can be blocked if part of the case has already been finally disposed of in another. Coordination, cooperation and information exchange problems occur at different levels and between different authorities and are a major impediment to the effective investigation and prosecution of offences affecting the EU’s financial interests. There is no authority that can deal with these obstacles and ensure continuity in the investigation and prosecution process.
At national level, there is often insufficient information exchange on suspected offences involving EU funds between the authorities responsible for monitoring and control, those dealing with administrative investigations and law enforcement bodies. This partly arises as a result of loopholes in the procedural framework referred to above hampering efficient multidisciplinary investigations involving judicial as well as administrative, customs and tax authorities in the Member States. Agencies managing and controlling the disbursement of EU funds sometimes focus solely on getting their money back through administrative and civil law procedures even if there are strong suspicions that a criminal act has occurred. This may lead to neglecting criminal prosecutions, and with that deterrence and general prevention.
The effective investigation and prosecution of offences against the EU’s financial interests is, furthermore, hampered by the fact that law enforcement authorities and prosecutors do not always transmit information about criminal offences to their colleagues in other Member States, or to Eurojust or Europol. In interviews, various prosecutors reported cases where they had information they thought would have been interesting for their colleagues abroad, but which they did not share proactively. As there is no obligation to share the information in all cases, practical problems such as lack of contact points, language barriers, and time constraints constitute real obstacles that prevent the proactive sharing of relevant information. Whilst both Eurojust and Europol can provide important support to deal with these issues, they are dependent on the willingness of national authorities to make use of their services.
Case example: Obstacles to information exchange between Member States OLAF had been notified by one Member State authority of possible irregularities at import into the EU of chicken breast fillets from South America. There was a suspicion that this meat trader had abused the import pricing system in order to evade payment of the additional duties at import of South American poultry meat. While judicial procedures had been launched in one Member State, an important amount of evidence was to be gathered in other Member States. In one of those other Member States the judicial authorities were prevented to send evidence gathered by them to the other Member State pending a decision by the Supreme Court. According to the national law, MLA requests may not be executed before exhaustion of all local remedies, despite the fact that there are remedies available in the requesting State. This had an impact on the further transmission of this evidence and on the administrative recovery procedures.
In addition, the classical ways of international cooperation via mutual legal assistance (MLA) requests or via joint investigation teams (JITs) are often not functioning well enough to allow for the effective investigation and prosecution of these offences despite the efforts of European bodies such as Eurojust and Europol. Responses to MLA requests are often very slow and police and judicial authorities experience practical difficulties in contacting and cooperating with colleagues abroad due to language problems and differences in legal systems. In some States, slow and ineffective international cooperation has frequently resulted in the impossibility to pursue the case due to the fact that the prescription period had expired. In addition, cases affecting the EU’s financial interests are particularly complex: 34% of interviewed practitioners reported that such cases fail in a European context because of legal cross-border issues.
"Obstacles to effective international cooperation are the insufficient synchronising of procedural rules according to which evidence is collected; different timeframes to conduct certain actions within different Member States, and language problems: a different meaning is attributed to the legal terminology used in different States – there are conceptual terminological differences."
“Diagonal” cooperation between administrative and criminal investigation authorities and multidisciplinary investigations
There are also a number of gaps and loopholes in the procedural framework applying to the investigation of offences affecting the EU’s financial interests which are related to the multidisciplinary character of these investigations involving not only criminal investigation authorities, but also administrative, customs and tax authorities in the Member States.
These difficulties arise mainly because of the lack of a level playing field in administrative procedural law. For example, there are currently no rules regulating cross-border cooperation between an administrative authority in one Member State and a criminal investigation authority in another Member State (known as "diagonal cooperation").
Other legislative gaps concern the access to information and the exchange of information between the Commission and the Member States’ competent authorities, in particular with their criminal investigation authorities.
Concerning horizontal administrative cooperation between Member States' competent authorities, the EU rules governing Mutual Administrative Assistance (MAA) have developed over time on the basis of specific needs in the individual policy field: this explains why different instruments have been established for administrative assistance in customs matters and agricultural/fisheries matters on the one hand and in tax matters on the other hand. Therefore the legal framework in this domain is very fragmented.
At the same time, the difference between mutual assistance in criminal matters and administrative cooperation poses a problem as regards Joint Investigation Teams (JITs), which are foreseen only between judicial authorities. At EU level, the participation of the Commission in joint administrative investigation teams may currently be established only in the customs field (Naples II Convention).
Other problems arise from the fact that there are no specific provisions concerning the use and recognition of evidence gathered in the context of multidisciplinary investigations by administrative authorities – including the Commission - in judicial proceedings, therefore it often happens that investigative acts are duplicated. Sometimes, information gathered in the pre-trial phase remains unused in criminal proceedings and the levels of admissibility of such information vary to a large extent throughout the EU.
“Fraudsters play on the asymmetry of information within the EU.”
International cooperation with third countries can also be a significant problem affecting the effective investigation and prosecution of fraud and other offences affecting the EU's financial interests, because such offences often have a transnational dimension that reaches beyond the EU. These cases usually require cross-border cooperation to obtain evidence, as well as coordination of law enforcement actions, but in practice such actions are frequently limited to a single State. Jurisdictions which prosecute these cases are not necessarily the ones which are best placed to prosecute while some Member States are clearly reluctant to prosecute them.
3.2.3. Low level of deterrence
Deterrence is often considered to be the ultimate objective of any criminal legislation, and is specifically mentioned in Article 325 TFEU as a core objective of actions to counter fraud and any other illegal activities affecting the financial interests of the Union. For deterrence to be effective, persons committing offences need to be punished, and their punishment needs to be publicised as well. The conviction demonstrates to other people that crime does not go unpunished and, coupled with the recovery of illegally obtained advantages, that crime does not pay. A well-functioning criminal justice system in which any potential offender can expect to face investigation and prosecution has a significant effect on deterrence which can be higher than in case of administrative sanctions.
However, as a consequence of the weaknesses described above, the deterrent effect of the current enforcement regime is insufficient, in particular as regards actions conducted by Member States to investigate and prosecute offences against the EU’s financial interests.
The first sections of this problem definition show that the deficiencies in the enforcement regime lead to impunity to a considerable extent or give the false impression of tolerance of offences affecting the EU’s financial interests. Potential perpetrators may think that the likelihood of their offence being detected and prosecuted, let alone of them being convicted, is very low. As a result, they may not be dissuaded from committing offences against the Union's financial interests or from re-offending if their previous crimes have gone unpunished.
3.3. Problem definition: conclusions
The law on offences affecting the Union's financial interests is generally not enforced to a sufficient degree by the Member States. There is no European body that is responsible for investigating and prosecuting crimes affecting the EU's financial interests. Consequently, whilst many cases have cross-border elements, these are not sufficiently pursued. Law enforcement efforts are fragmented, and Member States do not take all the actions necessary to tackle crimes against the EU budget.
The preceding paragraphs outline the many issues which negatively influence the enforcement cycle for these specific types of offences. As stated before, action at the EU level can in particular address problems related to investigation and prosecution, as well as cross-border coordination and cooperation. Indirect effects of such action could include a higher level of detection, as well as a higher number of successful prosecutions due to more effective follow-up of reported offences.
This is especially relevant since the research performed demonstrates that the main reasons why enforcement is often weak or deficient are the absence of a European enforcement structure, the lack of continuity in enforcement action and the lack of an underlying common European prosecution policy. Whereas offences affecting the EU’s financial interests are genuine European crimes, the current institutional and legal framework suffers from a fragmented enforcement regime almost solely based on national responses, which depend on the priorities and resources of national investigation, prosecution and judicial authorities.
In addition, whilst Eurojust and Europol can and do assist the Member States in dealing with these cases, neither of these organisations can address all of the issues identified, in particular due to the fact that they cannot direct national investigations and prosecutions. They are also dependent on information received from the Member States, but cannot compel them to gather the information needed to follow up suspected crimes effectively.
Although OLAF is a key player at EU level in the fight against fraud and irregularities, it is limited in its activities to administrative investigations, and submitting the results to national authorities, who may decide against any criminal law follow-up. This means that in its current set-up, OLAF is also unable to bring criminal cases to their conclusion.
Stakeholder views This analysis is supported by the results of the EuroNEEDS study, which identify a number of points regarded as problematic from a European perspective. A majority of prosecutors testify being hampered by European dimension of cases, with financial interest experts indicating that legal complexities pose particular problems for them. Around half of the prosecutors interviewed furthermore reported limiting their cases work to aspects of national relevance meaning the European nature of cases goes neglected. Finally the study indicates that the current criminal justice response to European cases is not comprehensive and unlikely to be consistent (dependant as it is upon national prosecutor’s circumstances). This is a clear argument supporting the need for a structural change such as the EPPO. A figure of this kind should guarantee full investigations, not hindered by national borders in what is investigated thereby providing the necessary informational basis for comprehensive prosecution. Furthermore, the House of Lords has recently published a Report on the fight against fraud on the EU’s finances, following an extensive inquiry into this topic and a number of interviews carried out both at EU level (Commission, OLAF, European Parliament, Eurojust, Europol), as well as with representatives of the UK government. The House of Lords notes that if OLAF were to be seen as a body whose recommendations are never followed up, it will remain hamstrung in its ability to protect the EU's financial interests. The House of Lords also states that if the UK government rules out a participation in the EPPO at this stage, it should explain how they propose to tackle the shortcomings in the system for combating fraud against the EU's finances discussed in the Report without participating in any EPPO.
Whilst Member States recognise these problems in terms of efficiency, the research performed in preparation of this impact assessment shows that they sometimes find it more difficult to be precise on the exact nature of the problems in their own administrations, which may be due to the fact that they perceive these problems only within the context of their national systems. This may also partly explain why the consultations on this topic have resulted in divergent views on the solutions to these problems, and therefore focussed more on the different models for solving the issues.
A consequence of these problems is that only a very small part of the total amount of fraud is ever recovered from criminals: expert opinions and available data show that recovery rates are currently very low, below 10%. This is ultimately due to the relatively small number of successful prosecutions in these cases. A conviction for criminal activities increases the chances of a successful financial recovery as it provides an additional tool to ensure successful enforcement; it prevents the individuals concerned from committing further criminal acts, and generates wider deterrent effects. The more successful enforcement authorities are in bringing offenders before the courts and securing their conviction, the greater the chances of recovering more of the proceeds of crime, and the greater the deterrent effect.
Diagrams explaining the relations between the problems identified in this Section, the objectives identified in Section 5 and the options can be found in Annex 2.
4. Right to act, subsidiarity and fundamental rights
4.1. Legal basis
One of the fundamental innovations introduced by the Lisbon treaty in the Area of Freedom, Security and Justice is the legal basis for the establishment of the European Public Prosecutor's Office (EPPO). For the first time the founding Treaty foresees the creation of a Union body with prosecutorial powers.
The current legal basis for establishing an EPPO is laid down in Article 86 of the TFEU, which states that the EPPO:
· will have to combat crimes against the financial interests of the EU;
· would be established ‘from Eurojust’; and
· shall be responsible for investigating, prosecuting and bringing to justice the perpetrators of these offences.
Article 86 TFEU specifies a special legislative procedure for setting up of a European Prosecutor’s Office: the Council needs to decide this unanimously after obtaining the consent of the European Parliament. Article 86 TFEU limits the initial competence of the EPPO to offences against the Union’s financial interests, but does foresee that this limitation of its powers may, at the same time or afterwards, be extended by the European Council, after obtaining the consent of the European Parliament and after consulting the Commission.
Article 86 (1) TFEU also specifies the possibility of establishing the EPPO on the basis of enhanced cooperation: in the absence of unanimity in the Council, a group of at least nine Member States may request that the draft regulation be referred to the European Council, leading to a suspension of work in the Council. If a consensus is reached within four months, the European Council will refer the draft back to the Council for adoption. Within the same timeframe, if at least nine Member States wish to establish enhanced cooperation, they must notify the European Parliament, the Council, and the Commission thereof, after which authorisation to proceed with enhanced cooperation referred to in Article 20 (2) TEU and Article 329 (1) of the TFEU shall be deemed to be granted and the respective provisions shall apply.
Denmark, Ireland and the UK do not take part in the adoption of measures in the justice field (Protocols 21 and 22 to the TFEU). However, Ireland and the UK have the possibility to opt in.
4.2. Subsidiarity and need for EU action
The traditional subsidiarity test requires a demonstration that the proposed measure's objective may be better achieved at Union level than at the level of individual Member States. For establishing this, it needs to be assessed how the main objective, i.e. effectively protecting the Union's financial interests and “countering fraud and other illegal activities” affecting such interests, has been met thus far, particularly taking into account the results of efforts by Member States and the reasons for any shortcomings. As the establishment of a European Public Prosecutor's Office specifically seeks to enhance criminal prosecutions related to EU fraud and other illegal activities affecting the Union's financial interests by introducing a direct European enforcement regime, to be implemented and coordinated by a European prosecution office, it also needs to be demonstrated that criminal prosecutions conducted by national authorities do not and cannot achieve the results expected from such a Union-level enforcement regime.
There is a clear need for EU action to protect the EU's financial interests for the following reasons:
· As set out above, Article 86 TFEU provides for the establishment of a European Prosecutor's Office. Also, Article 325 TFEU imposes a general obligation to counter fraud and other illegal activities affecting the Union's financial interests, an obligation which applies equally to the Union, its institutions, and to the Member States. The Union's competence to counter these forms of crime is thus unambiguously stipulated by the Treaty and this competence is not accessory to that of Member States.
· The Treaty also limits the possibilities to address the identified issues through reforms of the current European actors. The activities of both Eurojust and Europol are limited by their respective legal bases: Articles 85 and 88 TFEU. Even reforming these organisations to the maximum possible under the Treaty would not address these issues: neither Eurojust nor Europol can be given the power to conduct investigations, and Eurojust cannot be given the power to prosecute cases before the national courts. Under the Treaty, such powers can only be given to a European Public Prosecutor's Office.
· As the EU is best placed to protect its own financial interests, taking into account the specific EU rules which apply in this field, it is also best placed to ensure the prosecution of offences against these interests. A coherent Union-level prosecution regime is not only justified but also necessary considering the cross-border elements involved in European fraud cases: national prosecution authorities often cannot, and sometimes do not want to, deal with the "foreign" elements involved, such as a witness, a bank account, or a shell-company located abroad. This entails the risk of prosecuting systematically only narrow "national" elements of European fraud cases instead of the entire case with its cross-border dimension. Only a European prosecution regime can cover this dimension. By doing so it will also help prevent that safe havens develop in the Union for defrauding it.
· As indicated above in Section 3.2, current measures and initiatives taken by the Commission are not sufficient to deal with the problems identified, since these are not targeted on the problems related to investigations and prosecutions. In addition, possible reforms of Eurojust, OLAF and Europol cannot address these issues either, due to limitations in their mandates stemming from the TFEU.
· Eurojust's annual reports confirm that there is a need for coordination and support in the area of cross-border fraud investigations at the level of the EU: for 2011, for example, fraud-related crime ranked second in the areas of crime where Eurojust provided support, just after drugs trafficking. These reports also confirm that relatively few cases of fraud against the financial interests of the Union were forwarded by OLAF to Eurojust for their support – no doubt due to the fact that in principle the national authorities first need to take a decision on whether or not to prosecute such cases before they decide on seeking assistance from Eurojust.
· As indicated above, action at national level cannot achieve the objectives under the Treaty. OLAF's annual reports provide clear indications that criminal investigations limited to the national territory do not allow for effective and equivalent protection of the Union's financial interests. The degree of protection strongly varies from Member State to Member State. A large number of cases forwarded by OLAF to national authorities do not reach the prosecution or judicial phase, and other types of enforcement (fiscal, administrative) do not lead to sanctions either. OLAF’s annual reports clearly support the conclusion that Member States' criminal investigation and prosecution authorities are currently unable to achieve an equivalent level of protection and enforcement. As explained under section 3, there are various reasons for this, including lack of prioritisation, inadequate coordination or cooperation with Union agencies and other Member States, and legal obstacles. This variation in the enforcement level is demonstrated by the differences in the number of successful prosecutions and the amounts of financial recovery. The lack of equivalent enforcement in Member States signifies a generally weak enforcement framework, which is both legally fragmented and subject to national priorities. Whilst the legal fragmentation on the substantial law side may be partially addressed through the Commission's proposal for a harmonised sanction system, this proposal will not address the issues related to the investigation or prosecution gaps in the enforcement cycle.
· Against this background, it is clear that the Union not only has the competence but also the obligation to act. The Union's finances are by nature dealt with at the EU level. As such, they are even more "EU-centred" than other policy areas subject to harmonisation of rules in the Member States. As a result, they cannot reasonably be dealt with by the Member States alone.
· Moreover, according to the principle of proportionality EU action should leave as much scope for national decision as possible and should respect well established national arrangements and legal systems. In that sense this principle can also be understood to imply decentralisation: actions should be taken as closely as possible to where they intend to produce effects. The gravity of the problem, as defined in Section 3, clearly shows that the often unsynchronised actions of Member States and the missing continuity in measures countering the relevant offences do not effectively tackle the common challenge, i.e. the uniform protection of the EU budget. Given the great differences between some of the policy options, the principles of proportionality and decentralisation have been taken into consideration while defining each single option.
4.3. Fundamental Rights
Since Article 86 TFEU provides for the establishment of an EPPO, this provision must be read, interpreted and implemented in full compliance with the Charter of Fundamental Rights of the EU (‘The Charter’). In accordance with the Communication from the Commission on the Strategy for the effective implementation of the Charter by the European Union, this impact assessment examines, as far as relevant, the impact of the options proposed on Fundamental Rights, in particular in the light of the 'fundamental rights check list' presented in the Communication.
The establishment of an EPPO may raise several issues as regards the Charter, depending on the nature and scope of the powers and prerogatives attached to this new EU body. Any investigation powers conferred on the EPPO could have some impacts on the right to privacy (domicile, correspondence, phone conversations, etc.) and the right to the protection of personal data (Articles 7 and 8 of the Charter). The right to property (Article 17 of the Charter) could also be at stake, if the EPPO's powers include the possibility to freeze assets or seize other kinds of belongings. The right to liberty may also be affected, if the EPPO's interventions entail some restrictions or deprivations of individual freedom (Article 6 of the Charter).
Finally, any involvement of the EPPO in judicial proceedings has to be assessed with regard to Articles 47, 48 and 50 of the Charter: access to justice, fair trial, rights of the defence, presumption of innocence and the application of the ne bis in idem principle.
General || · To contribute to the strengthening of the protection of the Union's financial interests and further development of an area of justice, and to enhance the trust of EU businesses and citizens in the Union’s institutions, while respecting all fundamental rights enshrined in the Charter.
Specific || · To establish a coherent European system for investigation and prosecution of offences affecting the EU’s financial interests. · To ensure a more efficient and effective investigation and prosecution of offences affecting the EU’s financial interests. · To enhance deterrence of committing offences affecting the EU’s financial interests.
Operational || · To increase the number of prosecutions, leading to more convictions and recovery of fraudulently obtained Union funds. · To ensure close cooperation and effective information exchange between the European and national competent authorities.
Figure 5.1. Relation between problem, problem drivers and objectives
This diagram shows the relation between the problem identified, the problem drivers and the objectives. Diagrams explaining the relations between the problems identified in Section 3, the objectives identified in Section 5 and the options can be found in Annex 2.
6. Policy options and their impact
6.1. Overview of policy options
6.1.1. Assessed policy options
The following seven policy options have been assessed in detail: 1. Retention of the status quo; 2. Non-regulatory actions only; 3. Strengthening of the powers of Eurojust; 4a. Creation of an EPPO entity within Eurojust; 4b. Creation of a College-type EPPO; 4c. Creation of a decentralised EPPO with a hierarchical structure; 4d. Creation of a centralised EPPO with a hierarchical structure.
Whichever policy option is chosen, Eurojust’s role as the Union’s coordination agency for cross-border judicial cooperation in criminal matters will remain unaffected. This coordination role is of a general and horizontal nature, which has proven its value over the years for Member States and Union institutions.
6.1.2. Discarded options
Other policy options were discarded at an early stage of preparations. In particular, this included the setting up of an EPPO with a large scope of competence in accordance with Article 86 (4) TFEU, i.e. including serious cross-border crimes as listed in Article 83 (1) TFEU. The main reason for this decision was that fraud affecting the EU budget is a unique problem, both in terms of financial costs and in terms of damage to the image of the Union. As indicated above, under paragraph 4.1, the Treaty acknowledges this fact. Moreover, Article 86 TFEU does not permit the scope of the EPPO to encompass all forms of crime from the outset without a unanimous decision to this effect taken by the European Council.
Options not establishing a European Public Prosecutor's Office
6.2. Policy option 1: Base-line scenario - No policy change
No new action would be taken at EU level. Offences affecting the EU’s financial interests would continue to be prosecuted solely at national level. This would mean that the Union would retain its current administrative competences, but that the fight against these offences would continue to fall under the criminal law competence of national authorities, with Eurojust playing a coordinating and supporting role. OLAF and Europol would continue to work in accordance with their current mandate. The Commission's proposal for a Directive on the fight against fraud to the Union's financial interests by means of criminal law would address the current fragmentation of the substantive criminal law applicable to the relevant offences, but it cannot address the problems identified with respect to investigating and prosecuting the relevant offences. Whilst the reforms of existing EU bodies will address more general issues with respect to the functioning of these organisations, this reform is limited by the Treaty as explained above. These reforms would therefore also not address the problems related to investigation and prosecution of cases affecting the Union's financial interests.
Current situation – pattern of an OLAF investigation OLAF carries out administrative investigations into cases of fraud against the EU financial interests. The OLAF investigators may carry out administrative on-the-spot checks and inspections in the Member States to collect the necessary evidence. These on-the-spot checks are carried out in cooperation or jointly with the competent authorities in the Member States. OLAF may not use any enforcement powers but it is on the national authorities, acting in accordance with national law, to take the necessary measures. At the end of the administrative investigation OLAF draws up a final report. The Director-General may make a recommendation based on a final report for judicial action to be taken in a Member State and send the report with his recommendation to this Member State. It is then on the judicial authorities of the Member State to decide whether they take any action or not. They are however under the duty to cooperate in good faith, meaning that they have to examine the received information carefully and on that basis take the appropriate action (see judgment of the Court of First Instance of 4 October 2006 in Case T-193/04 Tillack v Commission).
6.3. Policy option 2: No new regulatory actions at EU level
No legislative action would be taken at EU level, and no new bodies would be set up. However, national and Union-level actions to fight the relevant offences would be strengthened through non-legislative measures. This would include capacity building for specialists in law enforcement and judicial authorities, enhanced cooperation and information exchange between national authorities and with EU agencies based on existing tools, as well as efforts to improve the implementation of the existing mechanisms concerning the admissibility and mutual recognition of evidence. Member States would also be encouraged to strengthen their efforts to control the use of Union money, including through regulatory changes where appropriate. New non-legislative arrangements for cooperation between responsible EU bodies, in particular OLAF, Eurojust and Europol, would also be introduced, which would improve their information sharing and promote joint actions in suitable cases.
6.4. Policy option 3: Strengthening of the powers of Eurojust
This option would mean that Eurojust would be given new powers to trigger investigations throughout the Union. Eurojust and its national members would have the right to give binding instructions to national prosecution services to initiate investigations and propose prosecutions in Member States in accordance with Article 85 TFEU. Eurojust and its national members would not actually direct the investigations and the prosecutions, which would continue to be administered by national services in accordance with national law. There would not be an institution in charge of investigations of crimes affecting the EU’s financial interests nor responsible for their outcomes, since it is not allowed under the Treaty to provide such powers to Eurojust as explained above.
The organisation of Eurojust would also be improved in order to give it a stronger focus on this task, including an improved focus on information exchange at all levels. In parallel, the current system of administrative investigations at Union-level (OLAF) would continue to be applicable, and Eurojust and national authorities would continue to be supplied by OLAF with investigative reports for judicial action. This option would also integrate the non-regulatory actions described under policy option 2 above.
Options for establishing a European Public Prosecutor's Office
The following paragraphs outline the four main options considered for establishing an EPPO. All of these options provide for a slightly different way of establishing the EPPO "from Eurojust", as required by the Treaty.
A proposal on the establishment of the EPPO will be accompanied by a proposal on the reform of Eurojust which will align it with the common approach on European agencies agreed by the Council, the Parliament and the Commission, and will establish a link between Eurojust and the EPPO. This reform might lead to more efficient information exchange and better cooperation between the national authorities. However, it would not affect the powers of Eurojust to deal with offences affecting the Union’s financial interests, and it therefore could not contribute in a tangible way to a more uniform protection of the EU budget.
Complementary rules on exchange of information, mutual assistance and cooperation between Member States competent authorities for the investigation of offences against the EU budget are mentioned in the Commission Communication on the protection of the financial interests by criminal law means and administrative investigations, as part of the measures reinforcing the protection of the EU financial interests. This issue concerns administrative investigations and is not directly linked to the creation of the European Public Prosecutor's Office. It should therefore be addressed separately. Pending the OLAF legislative reform, these measures relating to administrative investigations should be considered at a later stage, once the outcome of the on-going negotiations on the new OLAF Regulation becomes known.
All options for establishing the EPPO have been based on a comparable approach with respect to how investigations would be brought forward. Minimum EU rules would set out the powers the EPPO would have, the conditions for opening, conducting and closing investigations, as well as prosecuting the case, and would provide for minimum procedural guarantees, leaving other procedural aspects (notably the execution of investigative measures ordered) to national law. These EU rules would also set European standards for the admissibility of evidence, addressing the problems identified with respect to this issue referred to in Section 3.2.2. Only the essential aspects for the conduct of investigations which correspond to a common standard would be unified – other issues would continue to be regulated by national law. With regards to fundamental rights, this mixed regime will enhance rights protected under the Charter.
6.5. Policy option 4a: Creation of an EPPO entity within Eurojust
This option would entail the creation of a central EPPO entity within Eurojust, which would thus become the EPPO's holding structure as a "parent agency". In institutional terms this option would mean that Eurojust would effectively host the EPPO by providing infrastructure and support services to the EPPO entity. Eurojust could also make available its coordination capacity in cross-border cases affecting the EU’s financial interests, as well as in such cases which are connected to other offences falling within Eurojust's competence. The EPPO entity would have exclusive power to direct the prosecution of cases affecting the EU’s financial interests.
This EPPO entity would be composed of prosecutors and investigators specialised in financial crimes. In these cases they would exercise certain powers (initiation of investigation, review of evidence and indictment, coordination and direction), in full compliance with the Charter.
The EPPO entity would have a limited number of own staff to autonomously carry out investigations at central level. For the rest it would rely for this on local law enforcement authorities, which would be directed by local prosecutors in their investigations. Decisions to initiate prosecutions in national courts would require the approval of the Eurojust College of National Members. The EPPO entity would not prosecute suspects before national Courts, but leave this to local prosecutors employed by Member States under the direction of the EPPO.
The EPPO would be organically part of Eurojust and use its support functions (human resources, finance, IT, case-analysis etc.). A limited number of staff would be transferred from OLAF to Eurojust but Member States would also need to allocate additional resources to the Unit. Europol's role would remain unchanged: it would support the work of Eurojust and the EPPO unit within it.
6.6. Policy option 4b: Creation of a College-type EPPO
Similar to how Eurojust is organised, the EPPO would be organised in the form of a College of national members appointed by the Member States, but with a clearer and stronger mandate for all members. The EPPO college would take majority decisions as regards investigations and prosecutions of offences affecting the EU’s financial interests throughout the EU. Consequently, national members would be granted more incisive powers, as they would need to be able to provide national prosecutors with binding instructions. This policy option is thus very closely linked to national judicial systems. This is the main element which sets it apart from option 4c.
The EPPO will be directly in charge of investigation and prosecution of the relevant offences. The trial phase would in practice be run by delegated national prosecutors, acting in the name of the EPPO. Eurojust’s coordination function relating to investigation and prosecution of offences against the EU’s financial interests would be transferred to the EPPO, and a specialised investigative department working directly for the EPPO would also be created.
The EPPO would be a separate legal entity from Eurojust, but linked to it through the joint use of operational, administrative and management resources. The EPPO will also benefit of OLAF's specialised staff which would be transferred to it in order to provide for investigative, prosecutorial and administrative resources, whilst the remaining OLAF staff would continue to deal with functions that will not fall into the EPPO competence.
Europol would support the EPPO in line with its current mandate through its analysis, intelligence and general support functions.
6.7. Policy option 4c: Creation of a decentralised integrated EPPO
In this option, based on the concept of decentralisation, the EPPO would consist of an EU prosecutor's office at central level with a Chief Public Prosecutor exercising hierarchical supervision and decentralised European "Delegated" Prosecutors belonging to the national systems and therefore located in the Member States, having full prosecutorial authority under national law. The European Public Prosecutor would have the hierarchical power of instruction over European Delegated Prosecutors, who would be a genuine part of the EPPO.
In most cases, investigations and prosecutions would be led at decentralised level, but with the involvement of the European Public Prosecutor on opening the investigation and participation in bringing charges to the Court. Investigative measures would be normally executed at the national level, led by the European Delegated Prosecutors. A specialised investigative department at central level would also be created in order to coordinate investigative activities and to conduct itself, when necessary, investigative actions for the European Public Prosecutor, as well as for the European Delegated Prosecutors.
The European Delegated Prosecutors would work with the national police for carrying out their tasks. The European Public Prosecutor would have the possibility to give instructions to the European Delegated Prosecutors, which would need to cooperate with different national (administrative and judicial) authorities in order to carry out these instructions. The EPPO (acting through the European Delegated Prosecutors) would be normally responsible for bringing cases to trial. All powers of the European Public Prosecutor and the European Delegated Prosecutors must be in full compliance with the Charter.
The EPPO would be a separate legal entity from Eurojust, but linked to it through the joint use of operational, administrative and management resources. The EPPO will also benefit of OLAF's specialised staff which would be transferred to it in order to provide for investigative, prosecutorial and administrative resources, whilst the remaining OLAF staff would continue to deal with functions that will fall outside the remit of the EPPO.
All prosecutors and other staff working within the office of the EPPO would be directly employed by it, whereas European Delegated Prosecutors and national investigators would continue to be employed by national authorities. However, any additional costs incurred as a consequence of them being employed by the EPPO, such as travel, training, interpretation and translation costs would be borne by the EPPO.
Europol would support the EPPO in line with its current mandate through its analysis, intelligence and general support functions.
Future situation decentralised integrated EPPO The EPPO carries out criminal investigations into cases of fraud against the EU financial interests. The European Delegated Prosecutors, subject to hierarchical instructions of the European Public Prosecutor, direct the investigative work carried out either by national law-enforcement officers, or by joint teams composed of national law-enforcement officers and investigators from the specialised investigative department of the EU office. At the end of the investigation, the European Delegated Prosecutors, with the participation of the European Public Prosecutor, issue an indictment which is then sent to the competent court in one of the Member States. The European Delegated Prosecutor will be the competent prosecutor during the trial, having the same role as any national prosecutor.
6.8. Policy option 4d: Creation of a centralised EPPO
This option would entail the creation of a central EPPO possessing the full legal and practical capacity required to conduct investigations and prosecutions of the relevant offences, without depending on the national prosecution and investigation services. The EPPO's investigation staff would be empowered to take the necessary investigative measures within the Member States, only referring to national judicial authorities in cases where prior judicial authorisation is required. As for options 4b and 4c, Eurojust’s coordination function relating to investigation and prosecution of offences against the EU’s financial interests would be transferred to the EPPO.
This authority would be composed of a chief prosecutor, several prosecutors and staff at the central level, acting throughout the whole EU. The centralised EPPO would act directly, bringing suspects to judgment before national Courts. In contrast to options 4b and 4c, this would not be done through European Delegated Prosecutors embedded in the Member States. These powers must be exercised in full compliance with the Charter. All prosecutors and other staff working within the EPPO would be directly employed by it.
The EPPO would be organisationally linked to Eurojust, through a joint use of technical supporting functions such as human resources, finance and IT. In addition, part of OLAF's and Eurojust’s staff would be transferred to the EPPO in order to provide for investigative and prosecutorial resources, reflecting the transfer of the corresponding responsibilities from OLAF and from Eurojust.
Europol would support the EPPO in line with its current mandate through its analysis, intelligence and general support functions.
6.9. Horizontal issues for options 4a-4d
6.9.1. Cooperation between the EPPO and Eurojust
The four options which entail the setting up of the EPPO need to address the matter of its relationship with Eurojust since – as indicated above – the Treaty requires that the EPPO is set up "from Eurojust". Under all four options, Eurojust and the EPPO will need to co-exist and cooperate, but in a way that their relationship takes into account the differences in their respective functions and powers.
There are various options for organising this working relationship. These include as a minimum the sharing of administrative services and, at the other end of this range, the sharing of functional (coordination and cooperation) services, including those located in the Member States (National Contact Points in the Eurojust National Coordination System (ENCS)) or at Eurojust's Headquarters, based on the EPPO Regulation and service-level agreements. Such sharing of services would limit the deployment of additional resources, notably in terms of staffing, to a minimum.
The participation of the EPPO in the operational work and/or in the management structures of Eurojust is also an option. Regular coordination meetings between the EPPO and Eurojust could also be called by either organisation in order to ensure maximum effectiveness of both organisations, including the coordination of cross-border investigations or prosecutions of relevance for both. Such coordination could be particularly useful in cases where the EPPO’s investigations involve connected offences or third countries.
Besides this, the partial transfer or secondment of non-administrative staff (for example staff currently working in the Legal Service or the Case-Analysis Unit) from Eurojust to the headquarters of the EPPO would enhance functional links.
Even stronger functional links could be created at central level (SNEs or Eurojust National Members involved in the EPPO's casework). Eurojust’s National Members (or their Deputies or SNEs) could function as associated prosecutors within the EPPO structure, thus ensuring a very close coordination of efforts in cases affecting the EU’s financial interests, as well as in cases concerning associated crimes outside the competence of the EPPO.
6.9.2. Investing OLAF resources in the setting-up of the EPPO
Currently OLAF conducts administrative investigations for the protection of EU's financial interests. OLAF has specialised staff with significant experience in cooperating with national criminal authorities. Many members of OLAF staff have a relevant background in their national enforcement and judicial administrations (police, customs, and prosecutorial functions).
A part of OLAF's resources would thus be used in order to set up the EPPO, taking into account their experience in the conduct of administrative investigations and the objective of avoiding duplication of administrative and criminal investigations. Another important aspect is that of using the current networks which OLAF has developed over the years in the area of anti-fraud investigations.
Finally, OLAF would contribute to the setting up of the EPPO with specialised support to facilitate forensic analysis and technical and operational support to investigations and for the establishment of evidence in criminal cases affecting the Union's financial interests.
6.9.3. Cooperation with third countries
The different options identified for establishing the EPPO will also have to take account of the fact that the EPPO will need to cooperate with the authorities of third countries, since crimes affecting the Union's financial interests are also committed outside of the European Union. Such cooperation can either be based on existing legal instruments regulating such cooperation, such as the cooperation agreements Eurojust has entered into with a number of third countries, or multilateral agreements such as the 1959 Council of Europe Convention on mutual legal assistance, or on new legal instruments negotiated by the Union on the basis of Article 218 TFEU. The only option which offers some advantages in this respect is option 4a, since under that option the EPPO would become part of Eurojust, and could thus profit directly from Eurojust's existing cooperation agreements. The other options, whereby which the EPPO would be established as a separate legal entity require that the EPPO as such would need to rely on future legal instruments under which such cooperation can take place. Under these options, the possibility for the EPPO to use existing cooperation agreements would need to be specifically regulated.
7. Assessment of policy options
The assessment of the impacts of each of the options starts from the assumption that there will be no significant change in the human resources available to tackle crimes affecting the EU’s financial interests. The staff allocation assumes that the number of national staff working on offences affecting the EU's financial interests is currently up to the level needed to handle the anticipated caseload of 2500 cases in the baseline situation. The costs associated with creating the EPPO will be offset by reductions in the staff currently working at OLAF, thus reducing the overall set-up costs considerably. Costs in the various options therefore mostly arise due to an increase in the number of cases that enter Member States’ judicial systems, resulting in higher levels of costs for court cases, legal aid etc.
While there will be no significant change in the human resources available, there will be various changes to the administrative and legal framework under which they operate. These changes will enhance, to a greater or lesser degree, the effectiveness of the protection of the EU’s financial interests.
In each of the options other than the baseline, more effective investigation of crime is expected to lead to an increase in the number of prosecutions which will be brought forward, leading in turn to a greater number of convictions. This should increase the proportion of money that is recovered from criminals. This combination of greater punishment and reduced reward is expected to lead to higher levels of deterrence of crimes which should reduce the overall damage they cause. Increased deterrence is the most important factor in reducing the overall damage caused by crimes affecting the Union's financial interests over a larger number of years.
The working assumption is that increases in the proportion of funds that are recovered and in the amount of crime that is deterred are roughly proportional to the increase in the number of successful prosecutions, so to facilitate the comparison of the relative effectiveness of the options, an attempt has been made to quantify the changes in the number of successful prosecutions, and the increases in recovery and deterrence that could result from these.
All policy options are based on the same assumptions concerning the total value of offences against the EU’s financial interests, the average amount unlawfully appropriated per case and the change in deterrence relative to the change in the number of convictions. Differences between the options in the number of additional convictions are therefore the key driver of differences in the foreseen benefits of the options, as greater numbers of convictions will lead to higher levels of recovery and deterrence.
Nevertheless, it is not possible to establish with complete certainty the exact size of the cause-and-effect relationships, from administrative and institutional changes to an increase in the number of successful prosecutions, and from an increase in the number of successful prosecutions to greater recovery and deterrence. Because of this uncertainty, the estimated financial benefits that are presented below for each option are not intended as precise forecasts, but should rather be understood as indications of the likely relative scale of the effects that they could generate. Details of the methodology are in Annex 4.
For calculation purposes, the assumption was used that a 10% increase in the number of convictions would lead to a 1% decrease in the annual damage suffered, through the combined effect of deterrence and higher numbers of convicted fraudsters. This deterrent effect is assumed to be effective from 2020, that is, after a number of years of increased rates of successful prosecutions. For details see Annex 4.
Changing the assumptions about the total value of offences against the EU’s financial interests or of the relationship between increased numbers of convictions and the amount of crime deterred would affect the calculation of the estimated benefits of deterrence for each option in proportion to the change in the assumption. For example, if the value of offences against the EU’s financial interest is assumed to be 10% lower, then the estimated amount of crime deterred would also be 10% lower, in each option.
Stakeholder views The vast majority of stakeholders consulted seem to favour a decentralised structure, i.e. where the EPPO would act via European Delegated Prosecutors, or would otherwise delegate tasks to national prosecutorial authorities. The good cooperation and complementarity between these European Delegated Prosecutors and local authorities (police, judicial, administrative) are considered essential for reasons of efficiency and political acceptance. This cooperation would not only ensure better results because the European Delegated Prosecutors are embedded in national structures and know how to apply national law, but also because the decentralised structure respects the subsidiarity and proportionality principles. A small number of responses favour a structure where an EU prosecutor would direct, or at least coordinate, the investigations and the prosecutions from a central organ. The principal advantage of this design is independence and less cost. Some respondents see a centralised design as a guarantee of total independence from national interests and authorities, and thus a key aspect of effective enforcement. (1) Some believe that these two designs can be combined: Eurojust suggested that the same persons could possibly combine their role as EPPO delegates and as national prosecutorial authorities.
7.1. Status quo (policy option 1) – Baseline scenario
7.1.1. Views of stakeholders
None of the stakeholders consulted consider that this option would address the problem. Especially the issues with respect to investigation and prosecution of the relevant offences cannot sufficiently be addressed by improving existing arrangements only. Some have noted that it might be possible to improve implementation of certain existing EU instruments, i.e. the Council Decision setting up Eurojust and the Framework Decision on the prevention and settlement of conflicts of exercise of jurisdiction in criminal proceedings, which sets out procedures for cases dealt with by more than one Member State. However, as explained above, such improvements would only impact on the problems described in this report to a very limited extent. The current lack of equivalent enforcement would continue, which is inconsistent with the objectives set out in Article 325 (4) TFEU. However, some consulted experts consider that the necessary legal framework is already in place, and that the difficulties encountered in practice when investigating and prosecuting these offences are rather of a practical nature. These difficulties can be linked to lack of specialisation in dealing with these frequently complex cases, difficulties in collecting evidence from other Member States, differences in procedural laws, and time limitations. These views fail to appreciate the need to address also the cross-border dimension of fraud and other offences committed against the Union's financial interests.
Effectiveness in meeting the policy objectives || Low. As regards the specific objectives related to enhancing the investigation of offences affecting the EU’s financial interests, there are no reasons to expect that the weak incentives and the frequently limited national capacity to deal with the complex nature of EU fraud cases, as described in the problem definition, will be overcome without decisive corrective measures. The Commission's proposal for a Directive on the fight against fraud to the Union's financial interests by means of criminal law would address the current fragmentation of the substantive criminal law applicable to the relevant offences, but would not address the shortcomings identified with respect to the investigation and prosecution of these offences. The operational objectives related to the lack of efficiency in conducting cross-border investigations and prosecutions and increasing the number of successful prosecutions would not be addressed. Thus, the current fragmentation of the protection of the Union's financial interests will remain and there is no reason to expect any increase in the number of successful prosecutions from their current level of about 625 per year. Whilst a perfect implementation of the measures already taken to combat crimes affecting the EU's interests (see point 3.2.1) would bring significant benefits to the detection and prevention of the relevant offences, it would not have a real bearing on the issues identified with respect to investigation and prosecution of these offences.
Impact on fundamental rights || None. Fundamental rights will be unaffected. The Charter of Fundamental Rights is applied only when EU law is involved, for example under the regime of the European Arrest Warrant. As investigations and prosecutions would be conducted exclusively by national authorities under this option, there is no need to create an additional layer of judicial control. National courts provide such judicial control in accordance with national law.
Feasibility || High.
Impact on the legal system of Member States || None. The baseline scenario is the option that is the least intrusive of all options. It does not foresee any changes to the status quo and therefore does not produce any impact on the legal systems of the Member States.
Impact on existing Union institutions || None.
Costs || None.
Benefits || Very small. Improvements in information and exchange due to the reform of Eurojust will be inadequate to address the fundamental problems identified in section 3.
7.2. No new regulatory actions at EU level (policy option 2)
7.2.1. Views of stakeholders
Few of the stakeholders consulted believe that this option would be sufficient to address the problem. While the necessary legal framework is considered to be in place by some, it is largely agreed that difficulties are encountered in practice when investigating and prosecuting such offences. These difficulties cannot be addressed by non-regulatory actions only. Several consulted experts have stressed that obstacles to successful prosecutions could continue to exist: foreign evidence, differences in procedural laws, time limitations or lack of interest by national authorities. Some of these could be tackled through legislative action at the national level. For several consulted national prosecutorial authorities, a good solution would be to first improve the effectiveness of existing instruments and bodies such as Eurojust, OLAF, and Europol, as well as the cooperation between them. However, even such useful measures would not address the problem of fragmentation in cross-border investigations and prosecutions.
Effectiveness in meeting the policy objectives || Low. This option would to some degree contribute to the creation of a more coherent European system for investigation and prosecution of the relevant offences, thereby also possibly strengthening the deterrent effect of law enforcement. The experiences with improving mutual recognition and mutual legal assistance instruments show how hard it is to improve cooperation and information exchange between Member States. There are no reasons to expect that these problems will be easily solved by non-regulatory measures, and no guarantee that Member States would undertake any changes needed to their national regulatory frameworks for the specific purpose of improving the protection of the EU’s financial interests. In addition, the problems regarding the incentives, autonomy, and investigation and prosecuting of offences against the EU’s financial interests are also very difficult to address via this policy option. In the absence of clear leadership, which would drive action and would be in charge of implementing the wide range of improvements which are necessary to be made at both national and EU level, this policy option is not very likely to be able to address the identified obstacles to the effective protection of the EU’s financial interests effectively.
Impact on fundamental rights || Low – same as under option 1. It can also not be excluded that improved coordination and cooperation could potentially have a slight impact on the protection of personal data. Any proposal based on this option should take these issues fully into account. However, as in the previous option, the investigations and prosecutions would be exclusively conducted by national authorities and thus there is no need to create an additional layer of judicial control. National courts provide such judicial control in accordance with national law.
Feasibility || Medium. This option will also be difficult to implement in practice as it requires many efforts on different fields, and the involvement of many different actors, who face different incentives and a different prioritisation regarding the protection of the Unions financial interests.
Impact on the legal system of Member States || Low. An improved use of Union bodies as well as MLA tools and similar judicial cooperation tools would have some positive impacts in Member States.
Impact on existing Union institutions || Low. OLAF A limited effect can be expected, as Union funding would increasingly be used to strengthen its mechanisms. Eurojust A limited effect can be expected, as Union funding would increasingly be used to strengthen its mechanisms.
Costs || Moderate. Over a period of 20 years, the cumulative present value of costs (in 2012 prices) under this option would amount to about €35 million. These costs would mostly fall on Member States who would bear the costs of imprisonment of the extra criminals convicted.
Benefits || The problems identified would largely remain, but some increase in the current levels of prosecution, recovery and deterrence could be expected. Compared to the baseline, it has been assumed that a limited number of around 50 additional cases per year might be successfully prosecuted. Over a period of 20 years, the cumulative present value of the benefits of increased recovery and deterrence (in 2012 prices) under this option are projected to be about €265 million.
7.3. A strengthened Eurojust (policy option 3)
7.3.1. Views of stakeholders
A few Member States would welcome a strengthened Eurojust in this sense. Some experts have underlined their belief that a close cooperation between administrative, law enforcement and judicial authorities in the Member States is of key importance in order to ensure the effectiveness of relevant national criminal provisions, and that it is most important to strengthen the response to offences affecting the Union’s financial interests at national level. An EU approach in the sense of this option could help to unify the response given by the competent authorities throughout the European Union in this field in a coherent way and increase the dissuasive effect of criminal law. This would however not necessarily exclude other actions (as described in options 4a-4d).
For others, this option is clearly insufficient to address the problem, as Eurojust would not be able to effectively intervene in national legal systems. The problem of lack of continuity and ownership in investigation and prosecution of the relevant offences would not be addressed.
Concerns have also been raised about the objectivity of decision-making within the college model, as politics and national interests may interfere with the work of national members. None of the stakeholders consulted find that this option would be a fully effective contribution to the protection of the Union’s financial interests.
Effectiveness in meeting the policy objectives || Low. The investigation and prosecution of the relevant offences would only be strengthened to a limited extent, as Eurojust would continue to have no authority over national prosecutions. The difficulties faced by national authorities in tackling European cases would not be addressed. The option would not address problems identified by stakeholders related to the necessary autonomy of prosecution authorities in individual Member States. All in all, better information exchange, coherence and compatibility between the European level and national systems of investigation and prosecution can be expected, but even that would only have a limited positive effect on the general objective of strengthening the protection of the Union's financial interests, as priorities would not substantially change and the current fragmentation in cross-border investigations and prosecutions would largely remain. Crucially, there would not be an institution in charge of investigating and prosecuting crimes affecting the EU’s financial interests.
Impact on fundamental rights || Low. Fundamental rights will only be affected to a limited degree by this option, considering that the EPPO will not have any powers as regards national investigations under this option. The Charter of Fundamental Rights applies only when EU law is involved, for example when the regime of the European Arrest Warrant is applied. The concrete protection of fundamental rights would mainly continue to be ensured through national judicial authorities acting under law. However, this option may slightly affect defence rights whenever the case has cross-border aspects, i.e. that a trial, or an investigative measure, takes place in a foreign country, due to practical difficulties (language, unknown legal orders, etc.) and to a different standard provided by national law. However, following the implementation by the Commission of the Stockholm Roadmap on procedural rights, several measures have been put in place or are under adoption to provide suspects and accused persons, as well as persons subject to an European Arrest Warrant, with extended procedural rights, like the right to interpretation and translation, the right to information on their rights ("Letter of rights"), the right to information on the charges, the right to access to a lawyer and the right to legal assistance. It can also not be excluded that improved coordination and cooperation could potentially have a slight impact on the protection of personal data. Any proposal based on this option should take these issues fully into account. However, as in the previous options, the investigations and prosecutions would be exclusively conducted by national authorities and thus there is no need to create an additional layer of judicial control. National courts provide such judicial control in accordance with national law.
Feasibility || High. The option would to a large extent build on existing institutions and relevant Union legislation, so the further strengthening of Eurojust on the basis of the Treaty can be expected to be welcomed by some Member States.
Impact on the legal system of Member States || Low to medium. Eurojust would have the powers to initiate investigations and propose prosecutions, but the court proceedings as such would continue to be purely national.
Impact on existing Union institutions || Low to medium. OLAF OLAF’s responsibilities would not be affected by this option. Eurojust Eurojust would in this option acquire some additional powers but would continue to be a European body composed of national members.
Costs || Medium. Over a period of 20 years, the cumulative present value of costs (in 2012 prices) under this option would amount to about €50 million. As in option 2, most of these costs would fall on Member States who would bear the costs of imprisonment of the extra criminals convicted.
Benefits || The problems identified would largely remain, but some increase in the current levels of prosecution, recovery and deterrence could be expected. This option is likely to be somewhat more effective in this respect than option 2, so that the number of successful prosecutions could reasonably be expected to increase to some extent. If it is assumed that this will lead to an increase with another 50%, i.e. 25 investigations per year, on top of the additional 50 foreseen in option 2, over a period of 20 years, the cumulative present value of the benefits of increased recovery and deterrence (in 2012 prices) under this option are projected to be about €400 million.
7.4. Creation of an EPPO entity within Eurojust (policy option 4a)
7.4.1. Views of stakeholders
In the views of many stakeholders, this option would raise problems of conflicts of interests within Eurojust, which does not have autonomous powers and its current functions rely exclusively on Member States' good will. However, some Member States are attracted to the idea of placing the EPPO within the sphere of an existing agency which is mandated to ensure judicial coordination in cross-border criminal cases.
Effectiveness in meeting the policy objectives || Medium. The EPPO/Eurojust would still need to rely on national law enforcement and prosecution authorities, and its prosecutorial decisions could be delayed by the Eurojust College decision-making process. This means that the added value in terms of strengthening the protection of the Union's financial interests throughout the Union is limited as the current national priorities could still overwrite the EPPO entity’s priorities. This system would not be adapted to achieving a high number of prosecutions. The lack of independence of the prosecutors which would be the consequence of this option would be a factor hampering the effectiveness of this solution.
Impact on fundamental rights || Low. As in Option 3, with the necessity of judicial control as in Options 4b- 4d.
Feasibility || This option's political feasibility is limited, since conflicts of interest and in working culture between the EPPO entity and Eurojust could influence the effectiveness of this option. Moreover, there are functional limits linked to this option, because the mandate of the EPPO entity would need to go beyond the tasks of Eurojust and the collegial model of Eurojust does not meet the requirements of independence and effectiveness needed for the prosecutorial action of the EPPO laid down in Article 86 TFEU.
Impact on the legal system of Member States || Medium. Member States would need to adapt their systems to a new EPPO/Eurojust equipped with certain direct powers, but this would be facilitated by existing legislation on Eurojust. That said, it is expected that Member States would be confused as to who does what in Eurojust. For example, whereas national members usually make requests in ordinary Eurojust cases (serious cross-border crimes) they could be required to transmit instructions for undertaking investigations from the EPPO entity. This could generate confusion as to their respective roles within national prosecution systems.
Impact on existing Union institutions || Medium to high. OLAF In line with the transfer of responsibilities for investigating crimes against the EU’s financial interests from OLAF to the EPPO entity, a limited number of specialised staff would also be transferred from OLAF to the EPPO entity. The remaining parts of OLAF would retain their competence to exercise certain administrative functions which would remain necessary in order to cover responsibilities which the EPPO will not be able to take over (notably administrative investigations where there is no criminal perspective (disciplinary) and investigations outside the scope of offences affecting the EU's financial interests). Eurojust Eurojust would need to create a new administrative entity and ensure that the latter benefits from its administrative structures, including functional support (Finance, Human Resources) and technical services (Security, IT). Conflicts of interests between the EPPO entity and Eurojust could have an adverse effect on Eurojust's overall efficiency.
Costs || Moderate. The option would build on existing resources, reallocated in particular from Eurojust and OLAF, but the EPPO unit would also need some additional staff. However, the creation of the EPPO entity with powers in relation to criminal investigations and prosecutions would avoid the current duplication of investigations that sometimes takes place and would thereby free up resources in Member States. These savings would offset the costs that they would due to the imprisonment of increased numbers of fraudsters. Over a period of 20 years, the cumulative present value of costs (in 2012 prices) under this option would amount to about €40 million.
Benefits || The problems identified would largely remain, but if an increase by another 50%, i.e. 25 investigations per year, is assumed, an increase in the current levels of prosecution, recovery and deterrence could be expected compared with option 3. Over a period of 20 years, the cumulative present value of the benefits of increased recovery and deterrence (in 2012 prices) under this option are projected to be about €500 million.
7.5. College-type EPPO (policy option 4b)
7.5.1. Views of stakeholders
Similar to what was written above regarding option 3, stakeholders are generally sceptical with respect to conferring new prosecutorial powers to a collegial structure with collegial decision-making. Many practitioners fear that the nature of decision-making within a college is not appropriate for the need of rapid decision-making which is intrinsic to a prosecutor's office. Stakeholders have also raised questions regarding the independence of the decision-making within the college model, as political and national interests are bound to interfere.
Effectiveness in meeting the policy objectives || Low to medium. A few benefits from option 4c could also be expected here, mainly improved access to information for concerned authorities. However, the disadvantages associated with a College-type structure would negate most of these benefits. The collegial decision making process of this option would slow down the work process and lead to a very bureaucratic machinery because of the number of direct Member State representatives involved in the college. Also, the lack of independence from national judicial decision making is likely to have a negative impact on the effectiveness of this option.
Impact on fundamental rights || Medium to high. Improved coordination and cooperation could potentially have a slight impact on the protection of personal data, since the establishment of the EPPO will lead to improved exchange of information at the EU level. Furthermore, some investigative measures ordered by the EPPO will have consequences with respect to data protection, although these measures will be comparable to those used in national investigations. This option would also mean that a number of decisions concerning individual rights in investigation and prosecution procedures would be taken at European level, in particular the decisions to open an investigation and to prosecute, including the choice of where to prosecute. This implies that the option may slightly affect defence rights whenever a trial, or an investigative measure, takes place in a foreign country, due to practical difficulties (language, unknown legal orders, etc.) and to a different standard provided by national law. However, following the implementation by the Commission of the Stockholm Roadmap on procedural rights, several measures have been put in place or are under adoption to provide suspects and accused persons, as well as persons subject to an European Arrest Warrant, with extended procedural rights, like the right to interpretation and translation, the right to information on their rights ("Letter of rights"), the right to information on the charges, the right to access to a lawyer and the right to legal assistance. Although these new instruments will be applicable at national level, there may be a need to provide individuals with additional legal remedies and safeguards in order to ensure that the EPPO's powers are also exercised in accordance with the Charter. Any proposal based on this option should take these issues fully into account, in particular by requiring judicial control over the EPPO's investigation powers and foreseeing a right to judicial review of the decisions taken by the EPPO. Such judicial review would have to involve both national courts and the ECJ under a graduated scheme.
Feasibility || Medium. This option builds on existing structures, in particular the use of national prosecution services, Eurojust and OLAF. However, the establishment of a College-type central office which would direct and coordinate investigations as well as decide on the prosecutions to be brought will have a negative impact on the feasibility of this option. Consultations with stakeholders confirm that this would be a rather controversial option. Indeed, national prosecution systems are usually not based on a collegial decision-making structure for good reasons: operational decisions need to be taken swiftly in the interest of the investigation and the protection of the rights of suspects. This is particularly true in complex cross-border financial investigations, where prosecutors must secure often volatile evidence and seize assets in multiple countries. A collegial decision-making structure could not only slow down such cross-border investigations but possibly turn the EPPO into a rather bureaucratic and cumbersome agency with which national authorities will not cooperate.
Impact on legal systems of Member States || Medium. The Member States would preserve the main characteristics of their national systems, but would need to adapt their systems to the new competence of the EPPO to direct national law enforcement and prosecution authorities and intervene in national trials. There would be a combined application of an EU procedural framework together with national procedural rules.
Impact on existing Union institutions || Medium to high. OLAF This option would have some consequences for OLAF, as investigative capacities would be transferred to the EPPO in order to provide it with necessary investigative capacities. This is because the College-type EPPO would be a separate entity from Eurojust with a specific mandate to fight fraud against the EU’s financial interests. The remainder of OLAF would retain its competence to exercise certain administrative functions which would remain necessary in order to cover responsibilities which the EPPO will not be able to take over (notably administrative investigations where there is no criminal perspective (disciplinary) and investigations outside the scope of offences affecting the EU's financial interests). Eurojust This option would have a limited impact on Eurojust. Eurojust would remain a separate body as regards crime areas other than offences affecting the EU’s financial interests. It would carry on with its core activity of coordinating and stimulating judicial cooperation within the EU with regard to other serious cross-border crimes and remain a coordination body at the service of Member States. However, in the interest of synergies and cost-savings, Eurojust would provide administrative support to the EPPO, including functional support (Finance, Human Resources) and technical services (Security, IT). In practical terms, Eurojust's administrative structure would cover the needs of both Eurojust and the EPPO. This administrative structure would ensure coordinated budgetary planning and execution, various aspects of staff management and the provision of all other support services.
Costs || Moderate. The option would to a large extent build on existing resources, reallocated in particular from Eurojust and OLAF. However, the new office would need to recruit the members of the College and this would give rise to some extra costs compared to option 4b. Other costs would be similar to those under option 4b. Over a period of 20 years, the cumulative present value of costs (in 2012 prices) under this option would amount to about €70 million.
Benefits || Moderate. It is expected that the College-type structure and its role in decision-making would result in this option being no more effective than option 4b in terms of protecting the EU’s financial interests. No increase in the number of cases that are successfully prosecuted compared to that option can therefore be expected, so that the benefits of this option should be expected to the same. That is, over a period of 20 years, the cumulative present value of the benefits of increased recovery and deterrence (in 2012 prices) under this option are projected to be about €500 million.
7.6. EPPO with decentralised integrated organisation (policy option 4c)
7.6.1. Views of stakeholders
For many consulted prosecutors, the creation of a decentralised EPPO would be a way to promote unified and consistent EU action against offences affecting the EU’s financial interests, including through common prosecution priorities. Specialisation, centralisation and autonomy are considered by many respondents as the key aspects of the EPPO's added value, which would help overcome the reluctance of practitioners to deal with complex and distant cases, which often generate conflicts of jurisdiction or problems of judicial cooperation.
The vast majority of consulted experts favour a decentralised and hierarchical structure, where the EPPO would act via European Delegated Prosecutors, or would otherwise delegate tasks to national prosecutorial authorities. A good cooperation and complementarity between these European Delegated Prosecutors and local authorities (police, judicial, administrative) are considered essential for reasons of efficiency and political acceptance. This cooperation would not only ensure better results because the European Delegated Prosecutors are embedded in national structures and know how to apply national law, but also because the decentralised structure respects the subsidiarity and proportionality principles better than the centralised option.
Stakeholders have underlined the need to ensure the independence of European Delegated Prosecutors from national authorities and interests, and this need would be addressed through their hierarchical link with the central EPPO. Also, the lack of specialised capacity at national level could to some extent be addressed with this option through the creation of centralised investigation and prosecution services as part of the EPPO.
The Member States which expressed reserves are mainly concerned about the loss of national power (losing the right to decide whom and when to prosecute within their jurisdiction) and the potential resource requirements (number and complexity of European investigations to handle) of this option. However, the loss of such powers would be limited to cases affecting the financial interests of the EU only.
Effectiveness in meeting the policy objectives || High. This option implies the establishment of a specialised, independent and decentralised body which handles both investigations and prosecutions in a coherent and integrated (European) manner. It will establish clear lines of ownership and will overcome the current low priority given to cases affecting the EU’s financial interests in the national systems. The creation of a decentralised EPPO would improve the use of resources and information exchange necessary to be able to conduct successful investigations and prosecutions of the relevant offences. This, in turn, would strengthen the law enforcement response to these offences in general, and increase the preventive effect (deterrence) for potential criminals. The EPPO would be able to pool investigative and prosecutorial resources for the needs in a given situation, thereby making law enforcement at European and national level more efficient.
Impact on fundamental rights || Medium. Improved coordination and cooperation could potentially have a slight impact on the protection of personal data, since the establishment of the EPPO will lead to improved exchange of information at the EU level. Furthermore, some investigative measures ordered by the EPPO will have consequences with respect to data protection, although these measures will be comparable to those used in national investigations. This option would also mean that a number of decisions concerning individual rights in investigation and prosecution procedures would be taken at European level, in particular the decisions to open an investigation and to prosecute, including the choice of where to prosecute. This implies that the option may slightly affect defence rights whenever a trial, or an investigative measure, takes place in a foreign country, due to practical difficulties (language, unknown legal orders, etc.) and to a different standard provided by national law. However, following the implementation by the Commission of the Stockholm Roadmap on procedural rights, several measures have been put in place or are under adoption to provide suspects and accused persons, as well as persons subject to an European Arrest Warrant, with extended procedural rights, like the right to interpretation and translation, the right to information on their rights ("Letter of rights"), the right to information on the charges, the right to access to a lawyer and the right to legal assistance. Although these new instruments will be applicable at national level, there may be a need to provide individuals with additional legal remedies and safeguards in order to ensure that the EPPO's powers are also exercised in accordance with the Charter. Any proposal based on this option should take these issues fully into account, in particular by requiring judicial control over the EPPO's investigation powers and foreseeing a right to judicial review of the decisions taken by the EPPO. Such judicial review would have to involve both national courts and the ECJ under a graduated scheme.
Feasibility || Medium to high. This option builds on existing structures, in particular the use of national prosecution services, Eurojust and OLAF, with the addition of a streamlined central office which would direct and coordinate investigations as well as decide on the prosecutions to be brought in national courts. Consultations with stakeholders confirm that this should be a feasible option.
Impact on legal systems of Member States || Medium. The Member States would preserve the main characteristics of their national systems, but would need to adapt their systems to the new competence of the EPPO to direct national law enforcement and prosecution authorities and intervene in national trials. There would be a combined application of an EU procedural framework together with national procedural rules.
Impact on existing Union institutions || Medium to high. OLAF This option would have considerable consequences for OLAF, as investigative capacities would be transferred to an independent, integrated EPPO which will be set up to host investigative capacities. A reduced OLAF would retain its competence to exercise certain administrative functions which would remain necessary in order to cover responsibilities which the EPPO will not be able to take over (notably administrative investigations where there is no criminal perspective (disciplinary) and investigations outside the scope of offences affecting the EU's financial interests). Eurojust This option would have a limited impact on Eurojust. Eurojust would remain a separate body as regards crime areas other than offences affecting the EU’s financial interests. It would carry on with its core activity of coordinating and stimulating judicial cooperation within the EU with regard to other serious cross-border crimes and remain a coordination body at the service of Member States. However, in the interest of synergies and cost-savings, Eurojust would provide administrative and support structures to the EPPO, including functional support (Finance, Human Resources) and technical services (Security, IT). In practical terms, Eurojust's administrative structure would cover the needs of both Eurojust and the EPPO. This administrative structure would ensure coordinated budgetary planning and execution, various aspects of staff management and the provision of all other support services.
Costs || Moderate. The option would to the largest extent build on existing resources, reallocated in particular from Eurojust and OLAF. The new office would need to recruit the chief prosecutor but the cost for this would remain limited. Significant costs would, however, be incurred due to the imprisonment of greatly increased numbers of fraudsters. Over a period of 20 years, the cumulative present value of costs (in 2012 prices) under this option would amount to about €370 million.
Benefits || The benefits of the option can be expected to be important. Compared to options 2, 3, 4a and 4b, which are assessed as being able to deliver no more than incremental improvements to the current situation, the creation of the EPPO would represent a significant change in the approach to defending the EU’s financial interests, in particular due to the independence of the EPPO guaranteed through this option. The new powers conferred to the EPPO, as well as the improved access of information for concerned authorities, would deliver the needed improvements to the investigative and prosecutorial tools available. As compared to a collegial body or a unit within Eurojust, the hierarchical structure of the EPPO would also imply important advantages in terms of efficiency: a faster decision-making process, and clear lines of responsibility. It can be expected that a much greater number of cases – as much as twice as many as under the current arrangements – will be successfully prosecuted, so that a substantially higher amount of illegally received Union money will be recovered and a much larger number of crimes would be deterred. Over a period of 20 years, reflecting this major improvement to the effectiveness of the protection of the EU’s financial interests, the cumulative present value of the benefits of increased recovery and deterrence (in 2012 prices) under this option are projected to be about €3 200 million.
7.7. EPPO with centralised hierarchical organisation (policy option 4d)
7.7.1. Views of stakeholders
The creation of a centralised EPPO would be a very efficient way to promote unified and consistent EU action against these offences, including common prosecution priorities. The frequent cross-border dimension of crimes affecting the EU’s financial interests and their technical complexity justify coordination and centralisation of decisions at EU level. A number of experts also consider the full operational independence of the EPPO a key issue and see advantages in setting up an EPPO which is immune from local priorities or prosecutorial instructions. Specialisation, centralisation and independence from national authorities are considered by many respondents as the key aspects of the EPPO's added value, which would help overcome the reluctance of practitioners to deal with complex and distant cases, as well as to solve conflicts of jurisdiction or problems of judicial cooperation.
However, concerns have been raised by stakeholders as regards the effectiveness in practice of a “foreign body” when it comes to investigating offences in Member States. These stakeholders suggest that national authorities may in practice be reluctant to cooperate fully with this body and may even hamper its investigations if they consider the EPPO as a competitor. It has also been questioned whether this option is feasible from a political point of view.
Effectiveness in meeting the policy objectives || High. The EPPO would be given powers to control and steer investigations, including the right to obtain information from all relevant European and national authorities, without the need to rely on their cooperation. The EPPO would thereby also have full powers to pool investigative and prosecutorial resources for the needs in a given situation.
Impact on fundamental rights || Medium to high. The analysis made as regards option 4b and 4c applies here too. In addition, the direct powers of the central EPPO in all Member States will make it even more important to accompany the proposal with full guarantees for protection of personal data, defence rights and right to a fair trial. Additional safeguards would therefore need to be introduced in order to ensure that the EPPO's powers are exercised in accordance with the Charter. Any proposal based on this option should take these issues fully into account, in particular by requiring judicial control over the EPPO's investigation powers and foreseeing a right to judicial review of the decisions taken by the EPPO. Such judicial review would have to involve both national courts and the ECJ under a graduated scheme.
Feasibility || Medium. Although this option offers considerable advantages and has some support among stakeholders, it would imply very important changes in the administrative and judicial systems of the Member States. The feasibility in the short term is therefore probably low to medium as it would take a lot of time to implement these changes in practice.
Impact on the legal system of Member States || High. Member States would need to adapt their systems to a new external body intervening directly in national courts. This would have at least a perceived high impact on the legal systems, and provoke some opposition to the new office, at least in its first phase of existence.
Impact on existing Union institutions || High. Same as for option 4c.
Costs || Medium to high. This option would imply quite some costs, as new structures and significant numbers of additional posts at EU level for the office would need to be created. However, these would be partly offset by redeployment of staff from existing functions and by expected efficiency gains. Significant costs would, moreover, be incurred due to the imprisonment of increased numbers of fraudsters. Over a period of 20 years, the cumulative present value of costs (in 2012 prices) under this option would amount to about €820 million.
Benefits || The benefits of the option can be expected to be important. The new powers conferred upon the EPPO, as well as the improved access of information for concerned authorities, would improve the investigative and prosecutorial tools needed. As a consequence, investigations and prosecutions would become coordinated, including in cross-border cases, and the current problem of fragmentation would end. Although the centralised structure of the EPPO would however imply some losses compared to option 4c in terms of efficiency (considering that the centralised structure could cause problems in working on the ground with the national investigation and prosecution services) – so that perhaps some 50 fewer cases might be successfully prosecuted – it can nevertheless be expected that a sizeable amount of illegally received Union money will be recovered under this option. This option would have a clear preventive effect as well, as the deterrence for potential offenders would be strengthened. Over a period of 20 years, the cumulative present value of the benefits of increased recovery and deterrence (in 2012 prices) under this option are projected to be about €2 900 million.
7.8. Horizontal issues
7.8.1. Cooperation of the EPPO with Eurojust
For their mutual benefit, the sharing of Eurojust's administrative and support services with the EPPO seems not only necessary but also feasible. This should therefore be part of the implementation of any policy option to establish the EPPO.
As indicated above, functional links could be created at central level through regulating that some members of Eurojust staff would function as associated prosecutors within the EPPO structure, in order to assist the EPPO in the discharge of his coordination tasks related to investigations and prosecutions undertaken by the Office, in particular in cases also involving offences other than those affecting the EU's financial interests. This option would ensure the closest possible link in terms of operational cooperation, and would automatically entail involvement of the ENCS, which in this option would be approached through Eurojust staff, in line with normal Eurojust practices. However, for the sake of its legitimacy and accountability, the EPPO needs to be independent and free from any national or other influence in its decisions to investigate and prosecute cases in the Member States. It must therefore be guaranteed that the ultimate decision whether or not to investigate or prosecute a case remains the sole responsibility of the EPPO.
7.8.2. Investing OLAF resources in the setting-up of the EPPO central office
A part of OLAF's current resources would serve for the setting up of the EPPO central office. However, certain administrative investigation tasks of OLAF which are not linked to allegations of criminal offences must continue to be exercised. OLAF has an inter-institutional mandate to conduct administrative investigations into offences of staff of the EU institutions, which are outside the scope of the current proposal, but which have a reputational impact on the institutions. It also conducts administrative investigations into wrongdoings of EU staff which result in disciplinary proceedings.
In addition, the Commission, through OLAF, also has the important role of organising close and regular cooperation with the Member States, aimed at protecting the EU's financial interests, in accordance with Article 325 TFEU.
If a dedicated service of the Commission were to continue to exercise these functions, it would need to maintain a close relation with the EPPO in any fraud cases where there are both administrative (recovery) and criminal law aspects to be dealt with. It needs to be further explored whether an alternative scenario whereby the administrative investigation function would in the future be exercised by a department hosted by the EPPO could actually be feasible and effective.
7.8.3. Cooperation with third countries
As indicated under section 6.9.3., only option 4a presents some advantages in the area of cooperation with third countries, since under that option the EPPO could profit directy from Eurojust's existing cooperation agreements. However, these existing cooperation agreements are limited in number, and do not cover all third countries with which the EPPO would need to cooperate. Under these circumstances, this small advantage should not be a deciding factor in which option will finally be chosen. As also already indicated under 6.9.3, this issue can be regulated specifically in the legal instrument establishing the EPPO, and the EPPO can make use of future cooperation instruments concluded on the basis of Article 218 TFEU.
8. Comparative assessment
The table below sets out a comparison of the relative rating of the seven policy options as described in Section 6 against the objectives as defined in Section 5. The policy options are classified in accordance with their potential to meet the objectives defined in Section 5. Ratings for expected effectiveness in achieving the objectives are given equal weight in the final sum.
Table 8.1. Comparative assessment
Objectives/ costs || Policy option 1 || Policy option 2 || Policy option 3 || Policy option 4a || Policy option 4b || Policy option 4c || Policy option 4d
Meeting the policy objectives || Low || Low || Medium || Medium || Low to Medium || High || High
Annual net benefit || No major impact || €25 million || €35 million || €50 million || €50 million || €315 million || €250 million
Cost effectiveness || - || Low || Medium || Medium || Medium || High || Medium
Impact on fundamental rights || - || Low || Low || Low || Medium to High || Medium to high || Medium to High
Feasibility || High || High || High || Medium || Medium || Medium to High || Medium
Impact on existing Union institutions || - || Very low || Low to Medium || Medium to High || Medium to High || Medium to high || High
Impact on legal systems of Member States || - || - || Low || Medium || Medium || Medium || High
Already at first glance, it becomes obvious that options 1-3 will not be effective solutions to the problem. Although these options would not be very costly, they would also not add much value to the fight against offences affecting the EU’s financial interests. Combining the strengthening of Eurojust (option 3) with non-regulatory actions (option 2) to improve the effectiveness of procedures within the existing legal and organisational framework could perhaps deliver some additional incremental gains, but would leave the structural weaknesses that have been identified above fundamentally unchanged.
Based on the analysis above stronger action is warranted. As has been described in Section 3, the system of investigation, prosecution and eventually bringing to judgment is at this moment not effective enough to significantly deter and combat offences affecting the EU’s financial interests.
The initial comparative analysis therefore shows that the only effective answer to the situation is to set up an independent EPPO. Only an EPPO with clear powers, competence and responsibility would constitute an adequate response to the problem. The different options outlined for doing so have different levels of efficiency, as becomes apparent from their assessment. One of the main elements which explain why creating an EPPO is more efficient than the other options is that the EPPO will ensure a harmonised European prosecution policy, which will not be affected by national priorities. The independence granted to the EPPO under options 4c and 4d will also mean that prosecution decisions may no longer be influenced by national interests, and that all cases which merit prosecution will actually be brought before the competent courts. In addition, problems related to information exchange and cooperation across borders will be addressed through the EPPO's powers to ensure that information is collected centrally and shared with those who need to know. The EPPO will also be able to ensure adequate cooperation between national authorities, through its power to direct European Delegated Prosecutors (in options 4b-4d) and set up appropriate procedures for cooperation between the European Delegated Prosecutors and the national law enforcement authorities.
The creation of an independent EPPO under options 4c and 4d is expected to generate a major change in the effectiveness of the fight against fraud affecting the EU’s financial interests. This will lead to a much greater number of convictions, and consequently, to increased recovery and to greatly enhanced deterrence. This enhanced deterrence resulting from an increase in the number of convictions accounts for more than 85% of the difference in benefits between option 4c and option 4b. Using the assumptions that the total value of offences against the EU’s financial interest is €3 billion, and that every 10% increase in the number of convictions leads to 1% extra deterrence, the value of the crimes deterred by a 20% increase in the number of convictions is 2% of €3 billion, or €60 million. The increase in recovery as a result of 20% increased convictions, is about €4.5 million (if the recovery rate does not change), and would be €7.2 million if the recovery rate were to increase with 1%. Still, the difference between €60 and €7.2 million show that vast majority of the estimated benefits comes from increased deterrence of crimes.
Setting up the EPPO is also coherent with other policy initiatives which have been undertaken in this area, as mentioned in the introduction. These include the reform of OLAF, the proposal for the Anti-Fraud Directive from 2012, and the Commission’s anti-fraud strategy adopted in 2011. In fact, setting up the EPPO will further increase the effectiveness of these actions: it will provide the “missing link” in the enforcement cycle referred to above.
9. Enhanced cooperation
As indicated above, the Treaty foresees the possibility of establishing the EPPO through enhanced cooperation, should decision-making by unanimity fail in the Council. This Impact Assessment does not examine in detail what the specific impact of the various options would be under a different legislative procedure. Measuring such impact would be more or less impossible in the absence of any clear indication as to the number of Member States that would wish to join an enhanced cooperation procedure and under what terms.
However, it is safe to assume that using such a procedure would have consequences both for Member States participating in the enhanced cooperation, as well as for the other Member States. In particular, if a group of Member States decided to set up a European Public Prosecutor's Office (option 4a-4d) by way of enhanced cooperation and transfer the national competence to investigate and prosecute cases of fraud and similar offences against the Union's financial interests to such an office, this would affect the way in which investigations would be coordinated with the Member States which do not participate in it.
For example, non-participating Member States could be required to execute mutual assistance requests and/or mutual recognition-based decisions issued by the EPPO to collect evidence or arrest suspects. This would necessitate the recognition of the EPPO by both the participating and the non-participating States as an issuing judicial authority under those existing mutual recognition instruments of the EU (freezing, arrest warrant) which are used in the investigation phase. For other types of judicial measures (hearing witnesses, setting up JITs), cooperation between the EPPO and non-participating Member States could be organised via existing mutual legal assistance treaties, such as the Convention of 29 May 2000 on mutual assistance in criminal matters between the Member States. Again, in this context as well, the EPPO would need to be recognised as a requesting authority for the purpose of judicial cooperation. These existing mutual recognition and more classical mutual assistance instruments could be also used in combination, yet it is likely that the same problems (fragmentation, speed, efficiency) which led Member States to propose the European Investigation Order (EIO) would persist here as well. Until the adoption and entry into force of the EIO, it may therefore be necessary to create a specific cooperation framework between the EPPO and non-participating Member States, covering the full range of judicial measures which the EPPO will undertake to collect evidence in its investigations.
In addition, the responsibilities of existing Union bodies, such as Eurojust, Europol and OLAF, would need to be adjusted to this new reality, with the likely separation of Member States into two groups, one which still uses OLAF for administrative (external) investigations and another which does not. The impact on OLAF's work would indeed be substantial: part of its staff and resources would need to be transferred to the EPPO to handle the latter's criminal investigations in relation to Member States participating in its establishment, while another part would stay and carry on conducting administrative investigations. Moreover, at least in cases not involving offences against the EU's financial interests, OLAF would still have to deal with internal investigations as well. Whether OLAF should continue to exist as a Commission service for this purpose or alternatively become a department hosted by the EPPO should be further explored. The figures and assumptions used throughout this Impact Assessment are based on participation in the EPPO by all Member States. However, should there be a decision to move to enhanced cooperation under Article 86 TFEU, all these figures and assumptions would need to be adjusted in the light of the number of Member States which choose to participate.
Similarly, Eurojust would no longer coordinate cross-border investigations in the area of fraud and other offences affecting the Union's financial interests with regard to the group of Member States which participate in the enhanced cooperation, while it would need to continue this with regard to those Member States which don't participate. This would most likely generate a new type of coordination activity between the EPPO and Eurojust National Members originating from non-participating Member States, whereby the EPPO would request or require their assistance to transmit to the competent authorities in their home States its requests or decisions for collecting evidence or arresting suspects under the instruments referred to above. In addition, Europol would also need to establish a new working arrangement with the EPPO.
It is also safe to assume that this procedure would have some additional, or at least different, administrative and financial consequences for both groups of Member States and the agencies concerned, which at this time cannot be accurately estimated or calculated.
10. Monitoring and evaluation
It will be of crucial importance that the implementation of the preferred policy option is closely monitored. With the EPPO being a novelty in the institutional set-up of the Union, its functions will have to be closely assessed and possibly adapted to the situation on the ground.
Providing for a robust monitoring and evaluation mechanism is therefore crucial to ensure that the envisaged beneficial effects of the Regulation materialise in practice. Data provided by the Member States, also under their existing reporting obligations to the Commission (Article 325(5) TFEU), as well as feedback by Eurostat, Eurobarometer and the Council of Europe will enable the establishment of a useful baseline for monitoring the situation, including the ex post assessment of the initiative's effectiveness when compared to earlier reporting outcomes. Consideration will also be given to improving data on how EU funds are spent in the Member States. Besides quantitative data provided by Member States, other possible sources of qualitative and quantitative information on the effectiveness of the EPPO will be gathered from the Justice Forum, OLAF, Eurojust and Europol.
Moreover, the Commission envisages carrying out a specific statistical study two to four years after the set-up of the EPPO is completed. The study should in particular analyse the number of cases and amounts involved in the activities of the EPPO. The data would enable the Commission to evaluate the effectiveness and efficiency of the new office, contributing to the general assessment of whether new legislation is needed in order to ensure an effective, proportionate and dissusive action against offences affecting the financial interests of the Union, as well as a full respect for the rights, freedoms and principles enshrined in the EU Charter of Fundamental Rights.
1) Summary of the replies to the 2012 questionnaires to practitioners and the general public
2) Intervention logic: how the different options relate to the objectives and problems
3) Dimension of crime affecting the EU's budget
4) Cost-benefit analysis of the different options
to the European Commission's Consultation on:
Protecting the European Union's financial interests and enhancing prosecutions
The consultation process
Since the European Commission is planning to propose further measures in 2013 to improve the protection of the European Union's budget and enhance criminal prosecutions in this area, it has organised an on-line consultation on the above subject between 5 April and 8 June 2012 by publishing a questionnaire with 11 questions on the Commission's Europa website (http://ec.europa.eu/justice/news/consulting_public/news_consulting_public_en.htm).
The questionnaire focuses on how to improve the current criminal law framework related to fraud against the EU's financial interests in the Member States, including by way of establishing a European Public Prosecutor's Office (EPPO).
The questionnaire was primarily addressed to members of the associations admitted to the European Commission's Justice Forum as well as any other justice or law enforcement professional in EU Member States who may be interested in the subject matter. It was not necessary to reply to all of the questions. For the sake of transparency, respondents had to identify themselves and the Commission clarified that the contributions received would be published on the same website, except where the respondent requested that the submission remain confidential.
Eurojust has also distributed the questionnaire to national prosecution services via its own channels and collected replies in view of its own submission. Both the individual replies by national prosecution services and Eurojust's institutional submission are included here.
17 national prosecution services as well as Eurojust and 25 individual practitioners or associations have replied to the on-line consultation.
The replies received are summarised below.
Question 1: Are the current criminal law provisions in your jurisdiction related to offences affecting the EU’s financial interests (fraud, money laundering, corruption, accounting offences against the EU’s financial interests, etc.) in your view sufficiently effective, proportionate and dissuasive? Please provide reasons for your position.
The current criminal law provisions established in the national criminal justice systems are largely considered to be effective, proportionate and dissuasive, although penalties show substantial variations in Member States. Some respondents believe that the penalties are not sufficiently effective or dissuasive (IT, ES, Almeida).
National laws transposing the 1995 Convention usually create specific offences of fraud against the Union's financial interests. Several respondents underline that European and National financial interests are theoretically equally protected under their laws (BE, CZ, PL), whereas others (EL) make a distinction. Some practitioners state that general criminal provisions (fraud, embezzlement) are used for both national and EU fraud (UK, LT) but they offer a sufficient coverage of the 1995 Convention. IT admits that private corruption and influence peddling are missing in its legislative arsenal.
While the necessary legal framework is generally considered to be in place, difficulties are encountered in practice when investigating and prosecuting such offences: several respondents (EL, PT, IT, Transparency International, Ettenhofer, Gomez-Jara) stress obstacles to successful prosecutions: understaffed prosecution offices, excessive workload, lack of specialisation, foreign evidence, differences in procedural laws, time limitation, lack of interest by national authorities.
Question 2: If not, could the protection of EU’s financial interests be improved on EU level concerning, in particular, one or several of the following aspects of criminal law :
• scope of persons covered
• scope of geographical application ( in particular in cases affecting EU financial interests involving third-country nationals as suspects and where the place of commission is a third country.
• Definition of additional acts to criminalise (abuse of public office in a conflict of interest, breach of professional secrecy etc.)
• Type of conduct (intent versus negligence)
• Time limitation
• Other horizontal matters ?
Although in general national legislations are considered sufficiently complete and effective, some respondents would welcome EU initiatives to improve them. Corruption and accounting offences are considered as not sufficiently harmonised (AT) at the level of the EU and lack of uniformity in money laundering prosecutions (CZ) seems to be a problem.
Many respondents agree that there is a need to further harmonise criminal law provisions related to time limitation as well as procedural deadlines (investigation or MLA). Several agree that rules of jurisdiction, the scope of persons covered (third-country nationals), legal definitions of criminalised acts (negligence) need further clarification. However some respondents believe that the scope of persons covered is wide enough in their current legislation and advise not to extend it further. Indeed, some respondents consider that difficulties in judicial cooperation in this field are often of a more concrete and practical nature, not directly related to legal problems. In this regard, support to national authorities in, for instance, setting standards for good case management would be more useful (DK).
Respondents also believe that a close cooperation between administrative, law enforcement and judicial authorities in the Member states is of utmost importance in order to ensure the effectiveness of relevant national criminal provisions. Furthermore, an EU approach could also help to unify in a coherent way the response given by the competent authorities throughout the European Union in this field and increase the dissuasive effect of criminal law.
Question 3: Considering the latest known results in prosecuting and bringing to justice cases of fraud and in light of your own professional experience, do you think that there would be an added value in establishing a specialised European Public Prosecutor’s Office with EU-wide priority competences in order to conduct prosecutions in relation to fraud committed against the EU financial interests at the level of the Union?
Positions were quite divided on this question, ranging from veiled opposition (DK, LT, HU) to mild (FR) or clear support (ES, IT, EL). Some respondents (AT) want to see the full proposal first before assessing the EPPO's added value, whereas others insist on seeing the evidence that supports such a new institution (DK, FI, HU, Frendo). Some respondents would see added value in the EPPO only with regard to States which don't investigate or prosecute EU-fraud effectively (CY, MT), whereas others only concerning fraud committed by EU officials or affecting funds managed by the Commission (PL).
For several respondents coming from national prosecution structures, as well as for Eurojust itself, the optimal solution would be first to improve the effectiveness of existing instruments and bodies such as Eurojust, OLAF, and Europol, as well as the cooperation between them (FI, Ettenhofer). Otherwise, they fear that the EPPO would entail unnecessary duplication of actions undertaken by national authorities and lead to practical problems in terms of relationships between the EPPO, the national prosecution services and the other European bodies (Grixti, Almeida)
For many others, the creation of a specialised European Public Prosecutor's Office with EU-wide priority competence to conduct prosecutions in relation to fraud committed against the EU financial interests would have an added value, for various reasons. One reason is that it would be a way to promote a unified and consistent EU action against these offences (ES), including common prosecution priorities and harmonised levels of punishment. Another reason is that the frequent international or cross-border dimension of this crime and its technical complexity seem to justify coordination and centralisation of decisions at EU level (FR, IT, RO, Ippolito, Damaskou, Florentina, Corstens, Karitzki). A number of respondents also consider the independence of the EPPO a key issue and see advantages in setting up an EPPO which is immune from local political influence or prosecutorial instructions (SK, IT, Gomez-Jara, EAJ). The EPPO is also seen by some (Mittermaier) as an agency capable of restoring trust in the EU institutions themselves.
Specialisation, centralisation and independence from national authorities are considered by many respondents as the key aspects of the EPPO's added value, which would help overcome the reluctance of practitioners to deal with complex and distant cases, which often generate conflicts of jurisdiction or problems of judicial cooperation (FR). On the contrary, one respondent considers that the EPPO may lead to further fragmentation of the European landscape (HU) or lead to unnecessary duplication (PL).
Question 4: (a) For what criminal offences should the European Public Prosecutor’s Office have jurisdiction in the European Union, i.e. only offences affecting the EU’s financial interests or also serious cross-border offences ? (b) Should this jurisdiction be exclusive or complementary to national prosecutors?
The large majority of respondents (AT, CY, CZ, EE, ES, FR, IT, MT, PL, RO, SK) consider that the EPPO should initially have jurisdiction only for offences affecting the EU's financial interests. The latter could include stricto sensu crimes affecting the EU's financial interests as well as assimilated offences such as abuse of office, market-rigging, corruption, misappropriation of funds and money laundering. At a later stage, provided that it has demonstrated its added value, several respondents (AT, BE, ES, FR, MT) believe that the EPPO could see its jurisdiction extended to serious cross-border offences as provided for by Article 86(4) TFEU.
For some respondents (BG, EL, Ippolito, Mihov, Gatzweiler, Sheenan), the EPPO should be given extended jurisdiction from the moment of its establishment and also cover other serious cross-border offences.
Some consider that sufficient information is not available at the moment to decide on the scope of a possible future mandate for an EPPO. In this context the view was expressed that on-going assessments and evaluations of the functioning of Eurojust and OLAF should be completed in order to identify crime areas that are not being handled efficiently by Member States in cooperation with European bodies. Further development of Eurojust as a good possible alternative to the establishment of the EPPO was also indicated.
The European Association of Judges agrees on the need of creating an EPPO for these offences, but proposes that criteria should be established in order to not overload the EPPO with minor crimes, for example a financial threshold.
Respondents were less divided on the issue of the "exclusive" or "complementary" character of EPPO's jurisdiction. Many respondents (AT, BE, CZ, DK, FR, IT, MT, RO, Ippolito, Fiala, Ettenhoffer, Frendo, Palomaki), including Eurojust itself, consider that the EPPO’s jurisdiction should be complementary to that of the Member states as this solution would be more flexible and less invasive for national jurisdictions. Many practitioners stressed that the relationship between national and European prosecutors was a central issue and needed to be absolutely effective. Disputes over jurisdiction (competence) would be harmful for both the EPPO and national authorities. Some suggested (BE, EL) that the EPPO should only have jurisdiction on the request of national authorities, e.g. when they need assistance (EL), otherwise a too powerful EPPO may be seen as a competitor or opponent (IT). Some believe that an exclusive jurisdiction could result in an unmanageable workload for the EPPO (RO), so "complementary" jurisdiction would mean focusing on "more significant cases" and "more important investigations" (Ettenhoffer, Karitzki), although in practice the interconnected nature of national and cross-border fraud cases may well lead to some degree of overlap and concurrence (Corstens). Flexibility or clear rules will be necessary to settle such conflicts of jurisdiction.
Other respondents (ES, HU, SK, Florentina, Gomez-Jara, Ippolito) would clearly prefer that the EPPO be given an exclusive jurisdiction as this solution would prevent frictions and conflicts between the EPPO and national prosecutors. Some see the EPPO's role as central when the Member State in which the offence was committed seems unwilling to carry out the investigation or prosecution or when several Member States are involved (Florentina, Gomez-Jara). However, for some respondents "exclusive jurisdiction" does not mean that Member States are excluded from the investigation or prosecution: the necessity of a strong cooperation with national authorities is there any way, for example by allowing the EPPO to use by delegation the national structures and staff in its investigation (Ippolito).
Question 5 : What would be the preferable design for the European Public Prosecutor’s Office’s structure, centralised (i.e. with all investigative and prosecutorial acts performed at EU level) or decentralised (i.e. with a certain flexibility to carry out certain investigative or prosecutorial acts at national level under the authority of the European Prosecutor’s Office), and why ? Please consider how the various levels of your preferred design would interact in practice.
The replies given by the respondents show that the terms “centralised” and “decentralised”, referred to in the question, were subject to different interpretations, as "centralised" does not necessarily imply that local authorities should not be involved, nor does "decentralised" mean that certain investigative or prosecutorial acts cannot be done centrally. There is clearly a range of possibilities in each option. However, everyone seems to agree that issue of independence is key for both centralised and decentralised EPPO designs, as the EPPO must be free from "interference" (Schneiderhahn, Foldes).
With that caveat, the vast majority of respondents seem to favour a decentralised structure, i.e. where the EPPO would act via European Delegated Prosecutors, or would otherwise delegate tasks to national prosecutorial authorities. The good cooperation and complementarity between these European Delegated Prosecutors and local authorities (police, judicial, administrative) are considered essential for reasons of efficiency and political acceptance. This cooperation would not only ensure better results because the European Delegated Prosecutors are embedded in national structures and know how to apply national law, but also because the decentralised structure respects the subsidiarity and proportionality principles.
A small number of responses (BE, EL, Florentina, Cook, Gatzweiler, Schneiderhahn) favour a structure where an EU prosecutor would direct, or at least coordinate, the investigations and the prosecutions from a central organ. The principal advantage of this design is less cost and independence. Some respondents see a centralised design as a guarantee of total independence from national interests and authorities, and thus a key aspect of effective enforcement.
Some believe that these two designs can be combined: Eurojust suggested that the same persons could possibly combine their role as EPPO delegates and as national prosecutorial authorities. AT also believes that an effective cooperation seems to be as important as an independent judicial system, a combined approach mixing aspects of decentralisation and centralisation might be envisaged as a good compromise (AT).
Question 6: What investigation powers should the European Public Prosecutor’s Office have? (e.g. search & seizure, arrest, interception of telecommunications) ?
For many respondents, this question was closely linked to the EPPO's organisational design and applicable law. It was stated (Tiza) that the EPPO as a European prosecutorial authority must, in any case, have its investigation powers strictly defined so that its investigations and prosecutions are properly framed.
Assuming that the EPPO will conduct its own investigations and prosecutions centrally, some respondents (FI, SK, Mihov) supported the idea of granting the EPPO all the necessary powers on its own, applicable in all Member States. Conversely, assuming that the EPPO's investigations and prosecutions will be conducted by European Delegated Prosecutors, acting locally, the large majority of respondents seemed to agree on the idea that the EPPO should have the same (full range) powers of investigation as national prosecutors. In practical terms, this means that EPPO prosecutors should be on an equal footing with national prosecutors, and have not less, nor more powers than those of the ordinary national prosecutors in comparable national cases. In addition, several respondents (Karitzki, EAJ, Schneiderhahn) noted that there was no legal basis for granting more powers to EPPO prosecutors than what national prosecutors are granted. One respondent (Gatzweiler) pointed out that having "equal powers" to national prosecutors also meant that there should be equal (judicial) remedies as well before national courts.
On respondent (ES) recommended that the EPPO exercise certain powers itself but request judicial authorisation in Member States for those powers that have an intrusive character for fundamental rights. Similarly, other respondents (CY, MT and PL) differentiated between coercive powers and non-coercive powers and strongly suggested that the former be reserved exclusively for national authorities (prosecutors or judges), who may authorise the EPPO to be involved (CY). Yet others (DK) agreed that investigative powers could not be given exclusively to a central EPPO, but "must be based" on close cooperation between the EPPO and national authorities. The EPPO should not have its own investigatory powers without a national presence (Frendo, Cook, Palomaki).
Besides, the EAJ, though in favour of a strong EPPO with concrete investigation powers, insists on the need of a judicial control over the exercise of the powers granted to the EPPO.
Question 7: What framework (applicable law, judicial review) should be envisaged for such investigation powers?
Many replies connected this question with the one on whether the EPPO should have autonomous powers or not.
For those respondents who support an EPPO with autonomous powers (AT, ES, IT, partly RO, Mihov, Florentina), it would be necessary or logical to consider that the EPPO should apply a set of new European procedural rules to regulate its actions. These rules should be established along with the powers granted to EPPO in European legislation.
For all other respondents only national criminal law should be used in order to avoid complicated changes and ensure coherence with national criminal prosecutions.
Question 8: By what criteria should the Member State or States of trial be chosen?
The respondents generally recognise that the jurisdiction where the EPPO may prosecute and bring a case to trial may depend on a number of criteria, such as the place where the crime was committed, the evidence is located, the damage occurred, the defendant or/and the victim has his residence or registered seat, etc. Many respondents warn that clear criteria must be set forth by European legislation to prevent ad-hoc decisions and forum shopping. Actors of the judicial process, including defendants, must be able to foresee the place where the case will be tried and the EPPO's choice should be open to judicial review for some (Schneiderhahn). Those who support the idea of establishing specific criteria (ES, FR, IT, PL, Ippolito) want the criteria be clear, objective and foreseeable in order to frame the EPPO's decision.
The majority suggest that the EPPO should bring its case to trial where the offence was committed and, in case of multiple options (cross-border offence), the Eurojust Guidelines should serve as guidance.
Question 9: How should the barrier raised by the diversity of rules of evidence be overcome?
Not all the respondents are convinced that the diversity of rules of evidence raises a barrier at a European level. Several consider that national courts are there to solve such problems (BE, Cook, Fiala) and that evidence collected lawfully in one Member State according to local law should be used in trial in another Member State irrespective of the procedural rules for gathering evidence there (Ettenhofer).
For those who see a barrier in the diversity of evidence rules, there are two main possible solutions: the harmonisation of procedural rules throughout the European Union (AT, CY, EE, FI, FR, MT, PL, SK, Mihov, Ippolito, Palomaki, Jurgens) or the adoption of specific procedural rules established (only) for the EPPO's investigations and prosecutions (AT, ES, RO, Gomez-Jara).
In addition or alternatively the application of the mutual recognition principle may also be a solution, for example by using the (future) European Investigation Order (DK, Ippolito).
In the absence of harmonised rules of evidence in the EU, some respondents (Scheiderhan, EAJ) note that the diversity of rules is a risk that the EPPO has to anticipate early on, so that it can adapt its investigations accordingly. In practice this means that the EPPO will collect evidence in a way that it is able to adduce it in all those jurisdictions where it may bring its case. This constraint of "multiple-jurisdiction" evidence disappears from the moment when it has determined the place of trial.
Question 10: How could fundamental rights be best protected throughout the criminal investigations undertaken by the European Public Prosecutor’s Office?
For all respondents, fundamental rights must be equally protected in criminal proceedings, irrespective of the institution conducting the investigation (EPPO or national investigation agency). There are different ways and at different scales of protection: for some (ES, FR) the EPPO itself would be a layer of protection, if established as an independent prosecutorial authority, as it would have the obligation to protect fundamental rights and prevent double prosecution (ne bis in idem). Besides, on the national level national judges of freedom and national adjudicating courts would ensure protection according to the standards of national law and, ultimately, the ECHR (CZ, EE, FR, IT, MT, RO, Mihov, Florentina, Cook, Fiala, Frendo)
On a European level, the European Court of Human Rights and the European Court of Justice offer a supplementary layer to protect the fundamental rights of citizen, as fair trial and defence rights are guaranteed by the ECHR and the Charter of Fundamental rights.
However, some respondents seem to think that the creation of a special European judge controlling lawfulness of investigation and compliance with fundamental rights could be an asset in this protection (HU, Jurgens, Foldes, Gomez-Jara).
Question 11 : What relationships (in terms of hierarchy, functioning and usual workflow) should the European Public Prosecutor’s Office have with other European bodies involved in the protection of EU financial interests and/or criminal matters, such as OLAF, Eurojust, and the European Institutions (in particular the European Parliament, the European Commission, and the Council of the European Union ?
All respondents acknowledge the relevance of the issues as the EPPO would need to be in relation with other European bodies and this raises questions about its independence and its links in terms of hierarchy, functioning and usual workflow with the other institutions.
EPPO & OLAF:
Many respondents consider that OLAF should become an investigate body under the direction of the EPPO (its «executive arm»; part of the EPPO), fully integrated into it.
EPPO & Eurojust:
For the majority of the respondents Eurojust should coexist with the EPPO. They should remain separate one from another but closely coordinate their work. Eurojust would thus be a useful aid for EPPO in its operations. However, some respondents consider than the EPPO and Eurojust should be integrated or even that the EPPO should take over Eurojust’s tasks.
EPPO & Other European Institutions:
Several respondents recommend that the EPPO be appointed by the Council on the basis of a Commission proposal and with the assent of the European Parliament. Any arrangement of institutional accountability would need to respect the fundamental independence of the EPPO and each EU institution, within its own competence, could have a supervising role over it. In addition, each Year the EPPO could present its activity report to the Institutions. The European Court of Justice could be the disciplinary authority over the EPPO and could have the competence to deal with claims arising from potential conflicts between the EPPO and other EU institutions.
Intervention logic: how the different options relate to the objectives and problems
Option 1 Base-line scenario Effectiveness: low Current weak incentives and limited capacity at national level would not address fragmentation problems. Lack of efficiency in investigation and prosecution will continue, leading to low level of prosecution and convictions. Deterrence is not expected to increase due to low number of prosecutions and convictions. The current framework will not ensure closer cooperation and information exchange.
Dimension of crime affecting the EU's budget
Quantifying the size of EU fraud is a challenge that has puzzled politicians and researchers for decades (see for example Sherlock & Harding 1991, Passas & Nelken 1993, White 1995 and 1998, Sieber 1997 and 1998 and Ruimschotel 1994, Durdevic 2006).
All those researchers comment on the lack of reliable information on EU budget-crimes. Offences against the financial interest of the EU cover a wide variety of activities, ranging from receiving subsidies for products grown on farms that do not exist, training programmes that have never taken place, various forms of VAT fraud, to siphoning EU money for personal purchases and political party financing or EU staff making false claims for salaries and expenses. EU fraud and corruption is committed all over European Union, within MS and across its internal and external borders. It happens across Member States with different judicial systems, different control systems, different cultures of doing business (including differing attitudes and norms towards using public money for private purposes).
By its very nature, fraud and corruption are difficult to quantify, as will be argued in this Annex, but it is particularly problematic with fraud, corruption and other offences that affect the financial interest of the European Union. Around 85 percent of the EU budget is spent though Member State governments and regional or even sub-regional bodies. This means that the EU has to rely on the Member States to ensure that the money is spent in accordance with the rules, and that Member States control the expenditures effectively, and detect and report fraud and corruption to the EU. There are no comprehensive data on adequate information on how EU funds are spend in the Member States. Most Member States do not separately collect data in crimes against the financial interest of the European Union.
The point of reference is the list of offences affecting the financial interests of the European Union, which have been included in the proposal for a new Directive on the fight against fraud to the Union’s financial interest by means of criminal law. This list consists of:
· Fraud affecting both expenditures and revenues of the EU budget;
· Misappropriation of EU funds or assets by public officials;
· Obstruction of public procurement or grant procedures (for example bid-rigging between tenderers);
· Corruption, involving not only public officials, but also representatives of public bodies and private organizations or citizens involved in the management of EU funds; and
· Money laundering, which should keep separate the laundering of the proceeds of the above crimes from laundering the proceeds of other crimes committed inside and outside the EU.
The analysis will cover both expenditures and revenues of the EU budget. This is particularly relevant for fraud, which includes VAT and customs fraud, and fraudulent activities with EU budget expenditures.
It should be noted that money laundering often accompanies fraud. Money laundering also facilitates other offences (with and without EU funds), which could result in double counting if we were actually able to monetize the amounts of money laundering. Money laundering as result of crimes affecting the EU budget revenue or expenditures can appear into various forms, as our analysis of several cases has revealed:
· High profile money laundering of defrauded EU money through off-shore bank accounts;
· Low profile money laundering of defrauded EU money through cask-couriers or (informal) money remitters; and
· EU funded investments used as a money laundering vehicle (for example the Sicilian wind-mill park that is used by the mafia to systematically launder the proceeds of crime; see box below).
Wind energy on Sicily
Money laundering of EU funds through the Sicilian Mafia
EU-funds for development of wind farms and renewable energy in Sicily are seen as an easy gift to the local Mafia, the Cosa Nostra. The Mafia has reinvented itself as a ‘white collar’ organisation, siphoning off EU funds through a combination of shell companies and infiltration of regional bodies, which distribute the subsidies. According to the regional anti-Mafia prosecutor Roberto Scarpinato, some of the wind farms in Sicily were developed by the Mafia using EU subsidies, which were then used to fund a money laundering empire. By rotating the millions of EU-funds through different front companies, the organisations appeared to be operating legitimately and so attracted further EU grants. ‘They took the same amount of money, and they moved it around as if each company had access to its own existing capital. They performed the same trick many times, and every time they received public funds’. Many of the wind farms have since been sold on to genuine energy companies completely unconnected to the Mafia.
Source: BBC News - Fears over 'widespread' EU fraud involving the Mafia, 2010, http://www.bbc.co.uk/news/world-europe-11848048.
The starting point for the analysis is the official data from OLAF. The OLAF data provides an initial indication of the magnitude of the known cases of fraud, corruption and other relevant offences (see paragraph 1.2 below). OLAF data also provides an initial impression of the nature and variety of the offences and the sections of the EU budget that are affected by it. Next it will be argued why the OLAF data potentially represents the lower boundary of the real problem (paragraph 1.3). The calculations of the probable higher boundary of the problem are based on previous studies in particular on VAT and customs fraud and the ‘what if’ calculation (paragraph 1.4). Finally the main conclusions will be presented (paragraph 1.5).
1.1 What is known?
OLAF is the central European institution for investigating and fighting fraud against the EU budget and systematically collects data on potential crimes against the EU budget. Member States are obliged to report so called 'irregularities' involving more than 10.000 euro of EU finances. Member States are also obliged to indicate if there is a suspicion of 'fraud'. To put it in more general terms, they have to indicate whether this irregularity is committed intentionally and with the aim of acquiring illicit gain or not. This is not a self-evident classification, and there is no obvious incentive for them to raise the level of fraudulent reports.
There are two main systems for formally notifying irregularities to OLAF: the Irregularity Management System (IMS), managed by OLAF, and the OWNRES (abbreviation of 'own resources') managed by the Directorate General for Budget. Cases on the revenue side of the EU budget are reported through the OWNRES system. OLAF data on irregularities and fraud are published in the ‘Annual Report on the protection of the EU's Financial Interest and the fight against fraud’. The latest report available for this study covered 2011. Table 1.1 provides an overview of the main data over 2011 and a summary of the 2010 totals.
'Irregularities' are defined as 'any infringement of a provision of European law resulting from an act or omission by an economic operator which has, or would have, the effect of prejudicing the general budget of the EU or budgets managed by it'. Notifications from MSs to OLAF fall into two main categories:
· 'Irregularities reported as fraudulent' (or 'suspected fraud'), which are irregularities 'giving rise to the initiation of administrative and/or judicial proceedings at national level in order to establish the presence of intentional behaviour, in particular fraud'; and
· 'Irregularities not reported as fraudulent' is the remaining category of irregularities where MSs do not explicitly indicate that that there is a suspicion of fraud.
Another categorisation is irregularities and suspected fraud cases on EU revenues and EU expenditures:
· offences related to revenues of the EU budget are in particular evasion of import duties and VAT taxes. Evasion of the payment of customs duties falls into two main categories: smuggling of (mainly high taxed such as cigarettes and alcohol) goods over the borders of the EU and fraudulent declaration of customs information. These two basic forms sometimes overlap; and
· offences related to expenditures of the EU budget would cover a wide range of illegal activities, from direct misappropriation to various forms of fraud, corruption, obstruction to public procurement or money laundering.
OLAF is competent to investigate suspected cased of fraud and other offences affecting the EU's financial interests committed by European civil servants (so called 'internal investigations') and by economic operators in the Member States when the EU budget is at stake (so called 'external investigations').
The 'estimated financial impact' of irregularities or fraud is defined by OLAF as the total financial amount that is affected by the fraud or irregularity. While there are documents that detail the overall methodology and the statistical approaches, the ‘estimated financial impact’ is calculated through various methodologies. In some cases the value of the total allocation is taken, is some cases a percentage or part of the allocation that is directly affected. As will be developed below, the costs (or prejudice) to the European taxpayer can be (substantially) higher or lower than this amount.
Table 3.1 Irregularities reported per type of expenditure, 2011
|| Estimated financial impact
|| Reported as fraudulent || Reported as non- fraudulent || Reported as fraudulent || Reported as non- fraudulent || Total || Reported as fraudulent
|| Number || Number || € million || € million || € million || % of allocations
Agriculture || 139 || 2 256 || 77 || 101 || 178 || 0.14%
Fisheries || 2 || 46 || 0.03 || 1.6 || 1.63 || 0.05%
Cohesion policy || 276 || 3 604 || 204 || 1 015 || 1 219 || 0.4%
Pre-accession funds || 56 || 207 || 12 || 48 || 60 || 0.67%
Direct expenditure || 34 || 888 || 1.5 || 50 || 52 || 0.02%
Total expenditure || 507 || 7001 || 295 || 1 216 || 1 511 || 0.21%
Total revenues* || 723 || 3973 || 109 || 278 || 387 || 0.49%**
Total 2011 || 1 230 || 10 974 || 404 || 1 494 || 1 898 ||
Total 2010 || 1 883 || 13 210 || 643 || 1 579 || 2 222 ||
Source: Annual Report 2011 on the Protection of the European Union's Financial Interests SEC(2012)408 final.
*Total revenues comprises customs duties and agricultural levies. **Approximate of gross amount of TOR collected.
The main observations from table 1.1 are:
· OLAF received a total of 1.230 'irregularities reported as fraudulent' in 2011 (against 1.883 in 2010). The number of irregularities 'not reported as fraudulent' is almost ten times higher (10.974 in total);
· There were 507 fraudulent irregularities (40% of the overall number) related to EU expenditures. Also, 723 fraudulent irregularities where related to the EU revenues (60% of the overall number). This is a reversal of the 2010 proportions where fraudulent irregularities on the expenditure side outnumbered the fraudulent irregularities on the income side of the EU budget;
· Cohesion and agricultural funds are in absolute numbers the main sources of concern on the expenditure side. Fraudulent irregularities related to the cohesion fund account for almost half (276 in total) of the fraud reports. Agricultural policy fraud reports are the second largest in number (139 in total);
· The total estimated financial impact of the reported fraudulent irregularities is 404 million euro (against 643 million euro in 2010). In terms of financial impact, cohesion policy cases account for a little over half of the total financial impact (204 million euro). The total estimated financial impact of the irregularities that are not reported as fraudulent is 1.494 million euro (against 1.579 million euro in 2010);
· If the irregularities that are explicitly reported as fraudulent are added, the total financial impact of irregularities (fraudulent and not as fraudulent reported) was 1.9 billion euro in 2011, against 2.2 billion euro in 2010. As will be discussed later, these figures probably contain mainly administrative errors, but can include cases of intentionally fraudulent behaviour as well; and
· In the final column the total estimated financial impact of fraud as percentage of the budget allocations is given. As a percentage of the overall allocations, the pre-accession funds seem to be most vulnerable for fraud (0.67% of the estimated allocations is labelled as fraudulent). The overall financial impact of fraud on expenditures of the EU budget is 0.21% and on revenues 0.49%.
In table 1.2 the development of the estimated financial impact of fraud and irregularities over 2006 – 2011 is presented. It reveals that the overall value of reported irregularities and suspected fraud cases declined in 2011 in comparison with 2010 (both on the revenue and expenditure side). This decline was expected by OLAF, and was merely technical, after an increase in the two preceding years. The 2011 figures are however still somewhat above average over the last six years.
It is important to note from this table that these annual changes in reported (fraudulent) irregularities are mostly technical. They reflect to a large extent, changes in management and control systems, to changes in the reporting systems (for example the introduction of the IMS system in 2008) and reporting behaviour of Member States, and cyclical effects of the closure of EU spending programmes (in particular the Cohesion fund). They are not indications of an increase and decrease in the nature and extent of the underlying problem. This is a general issue that affects most crimes (including drugs) where the investigative/regulatory body is the principal source of detections.
Table 3.2 Estimated financial impact of fraud and irregularities, 2006-2011
|| Estimated financial impact of fraud (million euro) || Estimated financial impact of irregularities (including suspected fraud) (million euro) || Estimated financial impact of fraud (fraud rate) (% of allocations)
|| Expend. || Revenue || Total || Expend. || Revenue || Total || Expend. || Revenue
2006 || 189 || 134 || 323 || 804 || 353 || 1157 || 0.2% || 0.94%
2007 || 209 || 107 || 316 || 1048 || 377 || 1425 || 0.22% || 0.62%
2008 || 77 || 75 || 152 || 783 || 375 || 1158 || 0.07% || 0.46%
2009 || 181 || 99 || 280 || 1493 || 343 || 1836 || 0.13% || 0.68%
2010 || 478 || 165 || 643 || 1804 || 418 || 2222 || 0.34% || 0.79%
2011 || 295 || 109 || 404 || 1511 || 387 || 1898 || 0.21% || 0.49%
Average || 238 || 115 || 353 || 1241 || 376 || 1616 || 0,20% || 0,66%
Source: SEC(2011)621 final, ANNEX I (2006-2009); SEC(2012)408 final, table 1 (2010 and 2011).
Whatever annual increases or decreases there are in reported irregularities and fraud, it is not evident whether these are the result of: changes in the management of reporting and control systems, the cyclical nature of EU programmes, and/or changes in the occurrence of the underlying problem. For these reasons, official EU data on offences affecting the EU's financial interests cannot be extrapolated to an overall figure of the detected and undetected losses, nor can they be used to make intra-Member State comparisons or to analyse trends.
Types of offences
OLAF does not systematically categorise the information relating to types of offences as they are defined by the proposal for the anti-fraud Directive, but OLAF reports and internal OLAF information give some indication of the occurrence of the various offences.
On the revenue side false declarations and smuggling are frequently mentioned as offences, along with formal shortcomings. The goods that are most affected by EU customs fraud are TVs, tobacco, cigarettes and oil.
On the expenditure side, the main violations in the known cases by OLAF are the use of false or falsified documents (in order to inflate the costs of a project or prove a non-existing right to obtain financial support) and violation of public procurement rules. The latter category in it itself could include various offences such as price-fixing rings, abuse of inside information in the construction of tenders, conflict of interests and nepotism (though connections between the procurement officer and the company), and collusion by procurement competitors, etc.
Cases transferred by OLAF are mostly prosecuted within the Member States on grounds of fraud, embezzlement, forgery and theft. Corruption and money laundering are either (substantially) less often discovered or occur less frequently. The hypothesis is that in particular corruption is relatively underrepresented among the known cases because corruption is relatively more difficult to detect. This is in particular the case in Member States where corruption is deeply entrenched among economic operators and within the public administration. So even if the number of corruption prosecutions in the latter countries are higher than in others, the proportion of cases may still be low, and it is easy to find targets among un-favoured parties and/or those who will not pay bribes. It must also be noted that prosecution decisions are correctly taken on the basis of the best evidence to prove a particular offence, rather than to fit into the EU record-keeping requirements.
1.2 What is unknown?
The irregularities and instances of suspected fraud, as reported to OLAF, represent incomplete and often anecdotal evidence of the actual problem. The official findings can, according to OLAF, 'not be considered as empirical evidence of the levels of fraud and irregularity.' The Commission has also stated that it 'is not in a position to estimate actual levels of irregularities or fraud because of the extent and nature of available information and data received from the Member States.
The overall value of irregularities that were reported in 2011 as fraudulent is 404 million euro and the overall value of irregularities which were not explicitly reported as fraudulent was 1.494 million euro. Beyond these known figures, it has to be taken into account that a “dark figure” exists, which is certainly higher than the known figures. It is important to note that it is not entirely transparent how the overall value of the reported irregularities and suspected fraudulent irregularities is calculated. The information that is submitted to OLAF should concern, among others the identification of the operation that is affected, the modus operandi, the natural and legal persons that are involved and the 'estimated financial impact'. It is not clear how this financial impact actually is estimated by the sources of OLAF.
The figures give rise to two assumptions on the actual magnitude of EU fraud and other relevant offences.
Assumption 1. The actual base line of known cases of EU fraud is probably (much) larger than 404 million euro
First of all, it must be noted that irregularities reported as suspected of fraud could contain so called 'false positives' – cases that are mistakenly labelled as 'fraudulent'. An indication for this could be derived from the conviction rates of cases that are transferred by OLAF to the Member States (1.030 in total over the past six years). 119 of them have resulted in an actual conviction; acquittal in 31 cases; 241 cases were dismissed before trial. There are many reasons for this low conviction rate (no judicial follow up, slow judicial procedures, quality of the evidential reports). Lack of evidence could be one of them. Suspected fraud is not yet proven fraud.
However there are ample arguments that - an even larger - number of irregularities is wrongly (deliberately or accidentally) NOT labelled as 'suspected fraud' ('false negatives') or even not reported to OLAF:
· The definition of ‘suspected fraud’ as a specific subset of 'irregularities' is not consistent across Member States. Member States use different definitions of fraud, corruption and other illegal activities and a certain proportion of the reports from the Member States to OLAF do not even distinguish between suspected fraud and irregularities (all notification are reported as 'irregularity');
· The distinction between fraud and irregularity is usually made on subjective grounds by the sources (people may be for good reasons reluctant to qualify a case as probably fraudulent if there is insufficient evidence). In some Member States, officials are even discouraged to report an irregularity as 'fraud' because they can be held responsible in the case of non-confirmation by a court judgment;
· Member States are only obliged to report irregularities involving more than 10 000 euro of EU finances. It is quite likely that the estimated financial impact of EU fraud as reported by OLAF is (significantly) underestimated due to the fact that irregularities under this threshold are not reported. In the case of the two EU agricultural funds 87% of the overall number of payments, which constitutes 21% of total expenditure, is below the threshold of 10 000 euro;
· So, even if fraud is taking place there is no obligation to report it, and this potentially incentivises fraudsters to be more active below the reporting threshold. Furthermore, this also fails to address whether multiple frauds below the threshold are actually systemised (or even connected) in a context where committing small acts of fraud and corruption is considered as 'normal'; and
· OLAF reports on the revenue side merely report violations of custom regulations. However VAT cases are generally considered as 'national cases'' and not reported to OLAF. VAT fraud is a major problem as many interviewees indicated.
Table 3.3 Irregularity rates and error rates, 2010 and 2011
|| Expenditure 2010 || Revenue* 2010 || Expenditure 2011 || Revenue* 2011
Irregularity rate (reported as fraud) || 0.34 || 0.79 || 0.21 || 0.49
Irregularity rate (not reported as fraud) || 0.94 || 1.21 || 0.86 || 1.24
Irregularity rate (total) || 1.28 || 2.00 || 1.07 || 1.73
Error Rate (European Court of Auditors) || 3.7 || || 3.9 ||
Sources: Annual Report 2011 on the Protection of the European Union's Financial Interests SEC(2012)408 final (irregularity rate); European Court of Auditors, Annual report 2011 (error rate). * Approximate of gross amount of TOR collected.
In Table 1.3 the irregularity rates (fraud and non-fraud irregularities) of OLAF over 2010 and 2011 are presented. The total irregularity rate of the expenditures was in both years over 1% of the EU budget. On the revenue side the irregularity rate was between 1.5 and 2% of the gross amount of collected traditional own resources For various reasons it is difficult to distinguish fraud from administrative shortcomings. As discussed before, Member States differ for example in their interpretation of 'fraud'. In customs procedures for example, 'false classifications' and 'formal shortcomings' are often labelled as fraud or irregularity. But it is not known if these 'irregularities' were committed intentionally, out of ignorance, or just are a matter of inaccuracy.
The error rate as calculated by the European Court of Auditors was in both years respectively 3.7 and 3.9% (see box below). The error rate mainly addresses technical shortcomings (with fraud as a minor subset) but is considered as a warning signal of misspending of EU funds.
As discussed before, the argument can be made that at least about 1 to 2% of the EU budget is lost through fraud, corruption and other forms of deliberate misconduct.
Box 1.1 The 3.9% error rate of the European Court of Auditors
The irregularity rate of OLAF is often confused with the 'error rate' as it is calculated every year by the European Court of Auditors (ECA). The error rate is defined as 'the irregular expenditure found as a proportion of total expenditure checked.' The error rate is different from the irregularity rate since it represents errors in procedures, which does not mean failed projects or wasted funds. Fraud, defined as the intentional deception and criminal action, is a minor subset of the error rate. However as the ECA repeatedly stresses, the EU remains blighted by fraud, waste and irregularities. In 2011 EU payments were affected by material error, with an estimated error rate of 3.9 % for the EU budget as a whole. For the 18th year in a row the ECA has reported major errors in the accounts and was unable to give an unqualified statement of assurance. The level of error remained similar to 2010 when it was 3.7 %. The Court of Auditors criticized the Member States for being 'only partially effective in preventing or detecting and correcting errors.' Cohesion, energy and transport were the most error-prone EU area of EU expenditure (estimated error rate of 7.7%). Other risk areas are foreign aid, development, enlargement, energy, transport and agriculture, according to the Court of Auditors.
Source: European Court of Auditors, Press Release, ECA/12/43, 6 November 2012.
Assumption 2. It is plausible to assume that the overall value of the ‘dark figure’ of EU fraud and other relevant offences is substantially higher than the estimated financial loss of the cases that are known by OLAF as fraud or irregularity
'Dark figure' ('chiffre noir' or 'Dunkelziffer') is the term that is generally used for crimes that are committed but never detected or reported to authorities. The magnitude of this dark figure in relation to the crimes that are recorded is strongly affected by the capacity and willingness of authorities to detect and record crime. In particular, statistics on white-collar crimes (which these offences are) are generally considered as very poor indicators of the overall problem. As Levi and Burrows have stated, 'there have been some modest and intermittent attempts to estimate the prevalence of some white collar crimes, the financial costs and impacts of deception offences have tended to be more subject of rhetoric than of serious empirical investigations.'
This elevated dark figure problem with white collar crimes is caused by various factors, such as: the inherent secretive nature of many forms of fraud and corruption; the absence (sometimes) of direct victims; the interest of all involved parties in secrecy of their illegal activities; and also definitional problems which could complicate the collection and recording of white-collar crime statistics. Also fraud investigations are often being investigated after complaints (reactive) while pro-active investigations are relatively rare, and risk-based investigation processes are seldom used. As a consequence white-collar crimes remain under the surface, unless for example an economic sector has acquired particular political or media attention, or a fraudster becomes a big profile target for some other reason. This is in particular problematic with EU fraud, corruption and other offences affecting the EU's financial interests. As one of the interviewees noted:
'EU fraud is stimulated by a window of opportunity and no national ownership of the problem'.
Limited political and national ownership. Law enforcement authorities tend to focus on national crimes that have a more populist element. The interviews in the MSs have confirmed that, generally speaking, law enforcement of crimes against the financial interest of the EU is regarded as a lower priority. As opposed to many forms of common crime (such as theft, robbery, and sexual violence) there is a view that there are no direct victims of fraud affecting the EU interests, although every EU citizen who pays taxes of any form is a victim of the crime. The ultimate victim is seen more as the EU as a whole. Expenditures on behalf of the EU budget are seemingly 'someone else's money'. This is analogous to tax fraud, but at least there, the MS have a direct interest in improving their performance.
Reluctance to report. Even in cases where offences affecting the EU's financial interests have come to the surface at the national level, the transfer of information to OLAF could be incomplete since Member States may have no direct incentive – or even face a disincentive – to report fraud and corruption to OLAF (even if they are obliged to do so). This is because any undue amount paid or unduly unpaid duty has to be recovered from the beneficiary or debtor by the Member State. The Member State has to bear the costs of pursuing fraudsters on behalf of the EU.
Large variations in reporting behaviour. It was also observed that reporting behaviour varies considerably between Member States, which could confirm the hypothesis that the availability of data strongly depends on the capacity and willingness of Member States to detect and record EU fraud, corruption and other such offences. Some Member States have very low reported irregularity and fraud rates. Other Member States report relatively frequent cases of irregularities and fraud – this is in particular the case with Member States such as Bulgaria, Hungary, Poland and Slovenia.
In the 2011 statistical evaluation of irregularities, the European Commission expressed its concern over the low number of fraudulent irregularities in relation to the payments received from the structural funds. In particular Greece, France, but also Belgium, Denmark, Ireland, Malta, the Netherlands, Sweden and Spain were mentioned as a source of concern to OLAF. These results could indicate either a lower fraud detection capability or the fact that a part of the detected fraud may remain unreported. OLAF observes similar inexplicable variations in irregularity and fraud rates between Member States, with regard to other EU expenditures and revenues, and between geographical regions within Member States.
As a consequence the high number of reported cases does not necessarily signify that a Member State is relatively vulnerable for irregularities or that indeed a relatively large number of irregularities or fraudulent activities are committed. According to OLAF, differences in irregularity and fraud rates might simply reflect variations in methods of detecting and reporting behaviour. It is, according to OLAF, 'possible that Member States with a higher irregularity rate perform far better than Member States with a lower irregularity rate'.
The dark figure of EU fraud and other offences affecting the EU's financial interests is probably substantial – more substantial than the (already relatively) large dark figure of 'national' white-collar crimes. This is due to a variety of reasons: low priority of detection and prosecution of national crimes; limited ownership of crimes related to the financial interest of the European Unions and technical factors (no reporting below the 10.000 euro threshold and limited reporting of VAT fraud, which is one of the major problem areas).
As has been discussed earlier, the true figure of EU fraud is possibly much larger. How much larger?
The starting point for analysis is the EU budget. In order to assess the overall magnitude of the relevant offences there is a need to know which money from into and out of the EU budget are mainly affected – and by what type of offences. Table 1.4 presents a breakdown of the main EU expenditures and sources of the revenue for the EU budget in 2011.
Table 3.4 The EU budget in 2011 (billion euro)
1. Sustainable Growth || 53.3
- Competitiveness for growth and employment || 11.6
- Cohesion for growth and employment || 41.7
2. Preservation and management of natural resources || 56.4
- Direct aids & market related expenditure || 42.8
- Rural development, environment & fisheries || 13.5
3. Citizenship, freedom, security and justice || 1.5
4. EU as a global plácer || 7.2
5. Administration || 8.2
Total || 126.5
Total as % of EU GNI || 1.01
GNI own resource || 95.6
Customs duties || 16.7
VAT own resource || 13.8
Sugar levies || 0.1
Other revenue || 1.4
Total || 126.5
As the table shows, sustainable growth represents the largest share of the EU budget (research, innovation, employment and regional development programmes). Natural resources cover the second largest portion of the expenditure (agricultural expenditure, rural development, fisheries and the environment). Agricultural expenditures are financed by two funds: the European Agricultural Guarantee Fund (EAGF) and the European Agricultural Fund for Rural Development (EAFRD). The major instruments of the cohesion for growth and employment polices (sustainable growth component of the EU budget are the Cohesion Fund and Structural Funds (ERDF and ESF).
The focus in assessing the magnitude of the relevant offences in EU expenditures will be on these five biggest beneficiaries of the EU budget, which account for about 80 to 85% of the overall budget. However, other items of the budget, such as resources that fund various internal and external policies and direct expenditures by the Commission are equally at risk.
The main source of income for the EU (75% of the overall income) consists of a standard percentage levied on the Gross National Income of each EU Member State – which is a Member State to EU transfer of money. This will be excluded from our analysis, since it is a direct transfer from the MS to the EU. The main risk areas for offences on the revenue side are thus VAT resources and customs duties.
1.3.1 VAT and customs fraud
The European Parliament has recently expressed its concern about VAT fraud and the widespread evasion of customs duties. In particular this involves the smuggling of cigarettes and alcohol across the EU’s external borders, but also smuggling of textiles (a particular issue in Italy). Parliament is concerned about the negative impact of this on the financial interests of the EU.
The EU’s VAT system has become vulnerable to highly organised sophisticated VAT fraud schemes in recent years. One common form of VAT fraud is the so called ‘missing trader intra-community fraud’. This is where a fraudster buys goods from another EU country ‘free of VAT’ and subsequently resells the goods to a domestic counterparty at a VAT-inclusive price without remitting the VAT collected to the tax authority. A variation to this basic scheme is ‘carousel fraud’ in which the same goods are repeatedly supplied in a circular pattern. Other VAT fraud schemes involve non-EU countries (such as missing trader extra-community fraud, carousels on imports under customs transit and under-invoicing of imports) VAT fraud in cross-border services and very recently VAT fraud schemes on the European carbon market, where the transfer of emission allowances turned out to be an opportunity for carousel fraud. As a consequence several Member States are increasingly confronted with carousel fraud related to greenhouse gas emission allowances, which are categorised as supplies of services within the European Union. VAT fraud types are transnational phenomena, making it very difficult for MSs to act individually against these fraud schemes.
According to Borselli (2011) VAT fraud is ‘an obstacle to the smooth functioning of the single market. It has an impact on fair competition, erodes tax revenues of the Member States and negatively affects the EUs own resources.’ VAT fraud can be labelled as an ‘indirect offence’. Member States pay a small percentage of their total VAT receipts to the EU. The Member States’ contribution is based on 0.3% of the national harmonized VAT tax base (with a few exceptions for MSs with a reduced rate).  The VAT base is capped at 50% of GNI for each country to prevent less prosperous countries having to pay a disproportionate amount. The VAT resources account for around 14 billion euro of the EU budget (2011 EU budget) – which is 11% of the total revenues.
VAT fraud will result both in losses of income for the Member States and indirectly affect the financial interests of the EU. The wider economic and social impact of VAT fraud is considered to be even more substantial, as Europol has stated in the 2011 Organised Crime Threat Assessment (OCTA): ‘It distorts the functioning of the single market, accounts for the loss of significant public revenue and affects the financing of the European Community budget. The costs involved in Member States’ efforts in preventing and combating VAT are also significant.’
Eurojust and Europol support Member States' investigations in VAT‐fraud. It is generally considered that an EPPO with the potential capability to investigate and prosecute perpetrators of fraud against the EU’s financial interests could help overcome barriers relating to Member States’ reluctance to initiate investigations and judicial proceedings against perpetrators of VAT fraud.
There are several estimates of the magnitude of VAT fraud in Europe. Europol has stated that VAT fraud is more serious than it first appears to be. A minimum of 100 billion euro is lost as a result of VAT fraud in the European Union each year, according to Europol. This estimate seems to be in line with the main results of a recent study by Reckon for DG Taxation and Customs Union on the VAT gap in 25 EU Member States over the period 2000-2006. The estimates are based on a comparison of accrued VAT receipts with a theoretical net VAT liability for the economy as a whole. The theoretical net liability is estimated by identifying the categories of expenditure that give rise to irrecoverable VAT and combining these with appropriate VAT rates. According to the study, the average VAT gap of EU-25 Member States is estimated at 12%, totalling 106.7 billion euro. PwC (2010) estimated the missing values for the new EU countries and updated the calculation, which resulted in a VAT gap for the EU-27 of about 119 billion euro in 2009.
An econometric analysis with various country characteristics has shown that ‘the variable found to have the strongest relationship with the size of the VAT gap was that connected with the perceived level of corruption in the country. The relationship implies that lower perceived corruption is associated with a lower VAT gap.' It is however important to point out that the VAT gap does not equal the amount of VAT fraud, as the gap figure might include numbers on VAT not paid as a result of legitimate tax avoidance. The VAT gap includes next to fraud, VAT evasion, avoidance and other forms of non-compliance.
Recently Borselli (2011) and others have estimated that the overall volume of VAT fraud in the EU-27 can be estimated within a range of 20 billion euro to 35 billion euro a year. These estimates are based on several sources – and include various forms of VAT fraud and customs fraud such as missing trader inter-community (estimated total of 13 to 23 billion euro) plus extra-community fraud, VAT fraud on tradable services under-invoicing of imports, and specific fraud schemes such as fictitious trades in emission certificates (estimated tax loss 5 billion euro). Borselli further notes that VAT fraud differs from country to country, however individual country estimates should be considered with great caution.
Next to VAT fraud, customs duties are the second important traditional own revenue to the EU budget. The smuggling of goods over the borders of the EU is the most common form of customs duty evasion. In addition fraudulent declarations and other kinds of manipulation of data relevant for the determination of the amount of customs duty are also frequently occurring offences against the EU budget. These two forms of customs fraud most often differ but sometimes overlap, for example smuggling of the undeclared quantity of goods. The most widespread form of smuggling goods into the EU is not smuggling via physical concealment of goods at border crossings or the declaration of import of a different kind of commodity by fraudulent customs documents, but by what is called transit fraud – the manipulation of the transit procedure.
There are no comprehensive estimates on the magnitude of the potential overall shortfall of EU revenues due to various forms of customs fraud. But there is a considerable literature on cigarette smuggling – which is widely considered as one of the major cross border criminal problems. Cigarette smuggling is a growing activity worldwide, it is widespread and particularly well organised, which is said to cost thousands of millions of dollars globally in lost tax revenue.
There are two types of cigarette smuggling. The first is smuggling with contraband cigarettes which are imported, distributed, sold in the EU in violation of the tax, duty or other fiscal laws. The second is smuggling with counterfeit cigarettes, which are cigarettes illegally bearing a trademark of a cigarette manufacturer that are manufactured by a third party. It is estimated that about 65% of the seized cigarettes in Europe are counterfeit, a figure that changes over time but was affected by the control measures taken against complicit major firms. According to calculations by KPMG (2012) the total counterfeit and contraband cigarettes within the EU accounted for 10.4% of total cigarettes consumption in 2011, up from 9.9% in 2010 (the total consumption in 2011 estimated at 629 billion cigarettes). The annual EU-wide tax loss due to cigarette smuggling is estimated to be approximately 11.3 billion euros. 
These figures are in line with estimates used by Europol. The economic impact of cigarette smuggling is significant, according to Europol: ‘It represents a substantial loss to national and EU budgets, estimated at around 10 billion euro per year, and damages the interests of legitimate manufacturers and retailers. Cheaper and smuggled products also constitute a marked threat to EU efforts on tobacco control, and by extension the objective of reducing consumption’. According to Europol, about 10% of this missing 10 billion euro of income to the Member States due to cigarette smuggling – which is about 1 billion euro – is missed revenue for the European Union budget.
In conclusion, VAT fraud and evasion of custom duties are the two most important sources of a shortfall in revenue in the EU budget. These two traditional own resources account for respectively 18.8 and 16.7 billion euro in revenues (2011 EU budget). The EU-wide loss to various forms of organized VAT fraud is recently estimated at 20 to 35 billion euro a year.
There are no comprehensive estimates of the overall losses due to customs fraud. However, cigarettes smuggling, which is one of the major issues of concern, is estimated to cost 11.3 billion euro of income to the Member States, of which about 1 billion euro would be missed revenue for the EU budget.
1.3.2 Agricultural and structural funds
For on analysis of the magnitude of crimes affecting EU expenditures the focus will be on the two major beneficiaries of the EU budget: the Agricultural and Structural Funds. Agricultural and Structural funds are dispersed by an extended system of subsidies, grants, incentives, aid, premiums and other kinds of expenditures, which are, by their very nature vulnerable to misuse. Durdevic (2006) observed that ‘the EU budget can be characterised as a subsidy budget, and according to criminological investigations, subsidies are a highly criminogenic financial instrument and very subject to criminal behaviour.’ Durdevic continues that ‘apart from the criminogenicity of subsidies, the European system of managing and allocating subsidies has a criminogenic effect’ due to the vast and complicated managing and control system of the EU funds that comprehends the EU level and authorities in the MSs at national, regional and local level and is complicated by numerous specific EU and national regulations. As one of the interviewees stated, ‘there is a window of opportunity and almost no control.’
Common forms of fraud with resources of the structural funds include a widest imaginable variety of offences. With agricultural funds offences are related to subsidies of the production of crops, export subsidies and storage subsidies and subsidies for the production and consumption of certain agricultural goods.
There are no empirically well-founded estimates on the real magnitude of fraud, corruption and other offences with EU expenditures. Nor are there top-down (macro) estimates (as with VAT fraud) available. Expenditures of the EU budget involve a large number of beneficiaries of the EU funds, complex management and control systems, there is a wide range of potential offences and variety of costs of these offences and there is a lack of reliable data on the use and misuse of EU funds and its actual effects on the local or national economy.
In order to acquire a more complete picture, data have been collected on fraud and corruption (national data and related to EU money flows) in ten selected Member States. The objective of this exercise was to identify (if available) data on the nature and size of offences at MS level, and to obtain an insight in general recorded statistics on ‘national’ fraud and corruption – this could serve as a proxy of the possible scale of offences in this Member State.
There are perceptions of general levels of fraud and corruption in the individual Member States and in the EU as a whole, and there are even systematically collected perceptions such as the Transparency International Corruption Perception Index and the Eurobarometer studies. However, these are only anecdotal and fragmented empirical findings.
Official collected data on fraud and corruption do not seem to be a reliable indication of the actual occurrence of white-collar crimes. For example, one German expert considers that undetected cases of fraud in the public sector amount to 400 % - 900 % of the officially reported cases. According to another German study only 5 % of the corruption cases are reported. There are no official estimates of undetected cases of obstruction of public procurement. However, it is estimated that obstruction of public procurement in the building industry causes damages of 5 billion euro per year.
Comparisons between Member States cannot be made validly, since the methods of collecting data differ across Member States. Member States do not always separate fraud in general from fraud affecting the national budget. Many Member States do not distinguish clearly between crimes against the national and against the EU-budget. Many Member States do not collect separately or systematically data on crimes against the financial interests of the European Union. Even comprehensive data on the beneficiaries of EU funds are non-existent.
There is a need for improved public available data on EU expenditures in the MSs and national data on offences affecting the EU's financial interests. These findings are in line with the recently published report ‘Deterrence of fraud with EU funds through investigative journalism in EU-27’ by the European Association of investigative Journalists and commissioned by the European Parliament. That report states that ‘there is a loud call, from journalists and EU officials alike, for uniformity in gathering, cataloguing, collecting, archiving and reporting of data, to be mandated by the European institutions and sanctioned for non-compliance’.
'What if' estimates
Intuitive estimates of losses by offences against the EU budget have ranged over the last decades from 1 to 20% of the overall EU budget. There is a generally accepted expectation that the actual figure of lost resources to the EU budget is five to ten times larger than the 1 to 2% of cases that are brought to light. Some interviewees simply stated ‘we don’t know’.
Therefore several ‘what if’ scenarios have been constructed that are based on these generally supported estimates. The lower bound is 0.5 to 2% of the overall expenditures. This is based on the fraud rate and irregularity rate of the OLAF notifications over the past few years, and the error rate identified by the European Court of Auditors, as discussed in the previous paragraphs.
For the upper bound three possible scenarios have been where percentages of the EU budget expenditures on the Agricultural and Structural funds is lost (abused, misappropriated) through criminal misconduct:
· 0.5 to 5% (low range);
· 1 to 10% (mid-range);
· 2 to 20% (high range).
The prevalence of offences is not equally distributed across the EU. Some Member States are more vulnerable to misconduct with EU funds than others. There are different patterns of crime in different geographical areas (jurisdictions, regions) and different attitudes and norms towards and levels of fraud and corruption in different regions of the European Union. Under the low range scenario 0.5% (in the lowest risk Member States) to 5% (in the highest risk Member States at the other side of the spectrum) of the EU expenditure is lost due to criminal conduct. Under the midrange scenario 1% up to 10% is lost. In the high range scenario 2% to 20% is lost.
The main results are given in the table below. A total of 4 billion euro will potentially be lost annually. If it is assumed that 0.5 % (in low risk MSs) to 5% (in the highest risk MSs) is lost through fraud, corruption and other relevant offences, 8.5 billion euro could potentially be lost under a 1 – 10% scenario. Seventeen billion euro will be lost if it is assumed that 2 – 20% will be subject to fraud, corruption and other relevant offences.
Present knowledge is not good enough to determine which scenario is the most convincing one. However, the table reveals that even in a conservative scenario with a maximum loss of 5% in the most risk full Member States potentially 4 billion euro (4% of the allocations of the five large agricultural and structural funds) is probably at risk.
Finally it should be noted that other expenditure items of the EU budget are equally at risk. In particular pre-accession funds and development aid, but also direct expenses for EU staff seem to be vulnerable to fraud and corruption, as can be concluded from the notifications cases from OLAF.
Table 3.5 Potential annual loss in Agricultural funds, Structural and Cohesion funds
What if scenario || Million euro || % of allocations
Low: 0.5 - 5% scenario || 4.066 || 4,0%
Mid: 1 - 10% scenario || 8.476 || 8,1%
High: 2 - 20% scenario || 16.952 || 16,2%
There are no reliable estimates of the loss of European taxpayer money due to criminal conduct related to the expenditures of the EU budget. However widely used estimates range from 2 to over 20%. These risks are obviously not equally spread over the EU Member States (differences in norms and values to the proper use of public money, differences in effectiveness of control systems). Even if we adopt a cautious approach in which a minimum of 0.5% will be lost in the most ‘safe’ Member States, and up to 5% in the MSs where corruption and misuse of public money is more widespread. A total of 4.1 billion euro of European money might be at risk (4.0% of the allocations). These calculations are based on the average annual budget of the agricultural and structural funds, which account for 80 to 85% of the EU expenditures.
1.3.3 Other direct costs and welfare effects
Finally it should be noted that the ultimate ‘prejudice caused to the European taxpayer’ might be larger than these direct financial losses as they are calculated or estimated. Criminological research typically differentiates between three main components of costs of crime:
· Direct costs or loss through the criminal, fraudulent or corrupt activity (which is a transfer from the victim to the fraudster or corrupt operators);
· Indirect or external costs and side effects (which are costs beyond the offence itself); and
· Costs of preventing and combating crime.
Even the direct costs could consist of a variety of costs and effects beyond a direct financial loss (through misspending, embezzlement, delivery of goods and services at inflated prices) to the EU taxpayer. Direct costs of crimes adversely affecting the expenditure side of the EU budget typically also appear in the form of weak or non-performance of activities, market distortions or financing of wasteful projects. As a result the real prejudice caused to the European taxpayer could be well beyond the monetary value of the direct financial loss.
Direct costs could include:
· Unlawful appropriation and embezzlement of EU money;
· Delivery of goods and services at inflated prices;
· No delivery or partial delivery of the promised goods and services;
· Poor delivery (low quality) of goods and services;
· Market distortions (delivery of goods and services at the right prices but by the 'wrong suppliers' through corruption, conflict of interest, favouritism); and
· Financing of wasteful projects (misspending, see box 1.2).
These direct costs often appear in combination. In some cases the direct financial costs could be limited or even zero while fraud and corruption manifests themselves through conflict of interest and favouritism or poor delivery of goods and services.
Indirect costs (the costs beyond the offences itself) of crimes affecting the EU budget) should be taken into account as well. Indirect costs cover a wide variety of tangible and intangible effects such as environmental effects (bad delivery of goods and services), social costs (particular target groups that are disadvantaged), health costs (due to bad or no delivery of procured goods and service) and in particular loss of reputation of formal and informal institutions (including credibility of the European Union and its institutions). These effects cannot be meaningfully monetised.
Measuring the true costs of EU fraud and corruption in one single number is an ‘idea fixe’ - even if we would possess ample empirical evidence. It is a movable construct because we know where to start counting but we do not know where to stop counting. An alternative could be to construct a plausible range of the EU money that is possibly at risk due to fraud, corruption and other offences.
The real cost of crimes affecting the financial interest of the Union is higher than the financial value of the shortfall in revenue or loss of EU funds. For a full assessment of the prejudice caused to the European taxpayer effects on the quality and delivery of goods and services, market distortions, and expenditures on wasteful projects due to conflict of interests should be added. In addition intangible long term welfare effects – such as a loss of credibility of the EU and its institutions – should be taken into account as well.
Box 1.2 Fraud and waste
Fraud and corruption and waste of EU money are often perceived as interconnected aspects of the same problem of insufficient protection of the EU taxpayers' money. For example the UK-based independent think-tank Open Europe identified a few years ago infamous lists of ‘100 examples of EU fraud and waste’ with EU funds. Their publication illustrates the ‘on-going problem of poor financial controls within the EU’. It should draw attention to what Open Europe called the ‘Byzantine spending schemes’ of the Common Agricultural Policy and Structural Funds. Open Europe’s list also illustrates the thin line between fraud and waste of EU funds.
Similarly the Financial Times created, in collaboration with the Bureau of Investigative Journalism, a database that collates information on 646,929 beneficiaries of EU structural funds across all 27 Member States, revealing 'a trail of undetected waste, missed opportunity and even fraud'. It has revealed how big businesses are accessing grants, despite the fact that the structural funds are intended to provide a helping hand to Europe’s weakest members and smallest businesses. It also shows how EU funds have poured cash into the hands of for example the Italian mafia. ‘Taxpayers' money has been lost to fraud, plundered by organised crime or simply wasted on questionable projects’, according to the investigative journalists.
Sources: Open Europe, 100 cases of EU Fraud and Waste, November 2008; Open Europe's 50 new examples of EU waste, November 2009; Another 50 examples of EU waste, November 20010; Financial Times, Bureau of Investigative Journalism, ‘Europe’s hidden billions’.
Case C. Regional Development Fund
A dog fitness and rehabilitation centre
In February 2009, a Hungarian IT firm “Gyrotech Commercial and Supplier Ltd” was granted around 411,000 euro from the EU’s Regional Development Fund for a project to “improve the lifestyle and living standard of dogs”. The company used the funds to build offices for the rehabilitation centre. The offices have, however, remained empty and overrun with weeds and there is no dog centre. Despite this, in April 2009, the firm received a further €13,307 from the EU for a Web-Shop management project, unrelated to the proposed dog rehabilitation centre. In August 2010, local authorities initiated a federal investigation into Gyrotech Ltd.'s activities and funding.
Source: Open Europe, Briefing Note, 10 November 2010.
Case D. Development of the Margaret Bridge and related transport systems
Inflated prices and non-delivery of requested modifications
The city council of Budapest originally estimated the reconstruction costs of the Margaret bridge for around 13 billion HUF. The state secured 6bn HUF of funding while the EU offered another 6 billion HUF. After the procurement procedure had been completed, it turned out that the two offers summed up to almost the double of the originally anticipated amount. However the reconstruction had been started by the Mh-2009 consortium (made up of Közgép Zrt, Strabag, A-Híd Zrt. – these companies have been involved in several dubious construction projects). In 2009, the Prosecutor`s Office started an investigation in the project but the investigation was ceased in 2012 since the police revealed no misconducts. In 2009, news portal Index.hu reported that the city council had violated the EU funding contract because one part of the designed cycle way had not been built and no modification requests had been initiated either. The project is often referred to as one of the biggest corruption cases in Hungary of the last couple of years.
It is widely reported that the interests of the European Union are not sufficiently protected across the EU Member States. Revenues are lower than they should be, due to in particular VAT fraud and customs and trade fraud. Expenditures are ‘lost’ due to embezzlement, fraud, corruption, and other offences with EU funds.
Irregularities and suspected fraud
Data as they are collected and analysed by OLAF represent the lower boundary (base line) of the overall problem. Irregularities that are labelled as ‘suspected fraud’ had an estimated financial impact of 404 million euro. This figure is somewhat lower than in 2010 and somewhat higher than the average over the past six years. A report can also be falsely labelled as 'fraudulent'. Suspected fraud is not yet proven fraud.
Another 1.494 million euro was reported as 'non-fraudulent irregularity. It appears from the information provided by OLAF and our own interviews in the Member States that a (large) majority of these irregularities indeed represent administrative errors and other technical omissions. However it should be noted that an unknown number of these notifications is for various reasons wrongly not labelled as fraudulent.
Member States do not distinguish in an uniform way between ‘irregularity’ and ‘fraud’ (and sometimes make no distinction at all), because it is not clear in many cases if an irregularity was committed intentional or just an administrative error, and because Member States are sometimes reluctant to label an irregularity as suspicious. The total financial impact of irregularities (fraudulent and not as fraudulent reported) was 1.9 billion euro in 2011 (also an increase against 2010 and a slight decrease against the average over the past six years).
Annual increases or decreases of the reported irregularities and suspected fraud cases are mainly caused by technical factors such as changes in the management of reporting and control systems, the cyclical nature of EU programmes, changes in the reporting behaviour of Member States etc. They cannot be directly attributed to changes in the occurrence of the underlying problem, nor can they be used to make intra-Member State comparisons.
The dark figure
There are good reasons to assume that the magnitude of the actual figure is larger. OLAF is facing what we could term ‘a double dark number’ problem since it is dependent not only on the Member State willingness and capability to detect and record crime and but also to transfer it to the competent European bodies. This detection of offences within the Member States and the transfer of cases to EU bodies is for several reasons (highly) incomplete:
1. Criminological research has proven that statistics on white-collar crimes (which these offences are) are, for various reasons, very poor indicators of the magnitude underlying problem;
2. The detection of such offences in particular has a very low priority in the Member States. Law enforcement authorities tend to focus on national crimes that have a more populist element;
3. Member State authorities face several disincentives to transfer cases to OLAF and other EU bodies (although they are obliged to do so);
4. Offences below the threshold of 10 000 euro are not reported. This fails to address whether multiple frauds behind this threshold are actually systemised, in particular in regions or specific economic sectors across the Union, where small acts of fraud and corruption are considered as 'normal';
5. VAT fraud is not systematically reported to OLAF. VAT fraud is one of the major (indirect) crimes against the EU, according to recent quantitative estimates and based on interviews in the Member States;
6. Interviews in the Member States have shown that in particular 'conflict of interest' is a major problem in large scale and sophisticated fraud and corruption schemes (including fraud and corruption against the EU budget). Conflict of interest is difficult to prove and not in all cases formally against the law;
7. These offences could also lead into ‘waste’ of European taxpayer's money. Wasteful use of EU money is perceived by the general public as an issue in itself, which however could be interconnected with corruption, conflict of interest, fraud and other unlawful activities. There is a thin line between waste and fraud and waste and conflict of interest.
In addition in order to monetise all costs of the insufficient protection of the EU financial interest both direct costs and indirect costs (multiplier effects, externalities, welfare effects etc.) should be taken into account:
· Direct costs can cover various types of losses to the European taxpayer from sheer embezzlement or various forms of tax fraud, to delivery of goods and services at inflated prices, against lower than agreed quality standards or no delivery of goods and services at all;
· Indirect costs (externalities, welfare effects, multiplier effects) can cover an even wider variety of tangible and intangible effects such as market distortions ('bad competitors drive out good ones'), multiplier effects on the local economy, environmental effects, social costs (particular target groups that are disadvantaged), health costs (due to bad or no delivery of procured goods and service), or reputation of formal and informal institutions (including impacts on confidences of the MSs in the European Union); and
· In particular indirect costs can be significant but difficult to monetise. In some cases the direct costs of an offence could even be zero while the indirect costs are substantial, for example favouritism and conflict of interest in granting EU subsidies or procurement with EU funds.
All in all it must be concluded that measuring the true costs of EU fraud and corruption in one single number is an ‘idea fixe’ - even if there would be ample empirical evidence. An alternative is to fill in some of the blind spots in the 'dark figure' and construct a plausible range of the EU money that is possibly at risk due to fraud, corruption and other offences.
For assessing the real order of magnitude some plausible assumptions of the money amount that is not spend on the approved lawful purposes are necessary – and the multiplier and other indirect effects on the local, national or European economy are excluded. An overview of this exercise is presented in table 1.6:
· VAT fraud, emission fraud and related extra-community fraud and customs fraud is recently estimated by various sources at an annual volume of 20 to 35 billion euro a year. VAT fraud is an indirect fraud. Member States are obliged to recoup about 0.3% to the European Union;
· The annual losses through cigarettes smuggling are estimated to cost 11.3 billion euro of income to the Member States, of which about 1 billion euro would be missed revenue for the EU budget;
· For the expenditure side three 'what if' scenarios have been presented. These calculations are based on expenditures with the agricultural, structural and cohesion funds, which represent 85% of the EU expenditures;
· It is assumed that the potential losses for these five funds could range from 0.5-5%, 1-10% or 2-10% of the allocations (depending in the risk of abuse in the individual Member State). The assumed percentages are however based on previous literature on EU fraud and intuitive expert opinions;
· A total of 4 billion euro will potentially be lost annually, if it is assumed that 0.5 % (in low risk MSs) to 5% (in the highest risk Member States) is lost through fraud, corruption and other offences. Eight and a half billion euro could potentially be lost under a 1 – 10% scenario. Seventeen billion euro will be lost if it is assumed that 2 – 20% will be subject to fraud, corruption and other offences;
· The present knowledge is not good enough to determine which scenario is the most convincing one or if any of them at all mirrors the reality correctly; and
· Other items of the budget, such as resources that fund various internal and external policies (development and pre-accession funds) and direct expenditures by the Commission are quantitatively less important but relatively (at least) equally at risk.
Table 3.6. Overview of potential annual losses due to offences affecting EU revenues and expenditures
|| Estimated potential impact
VAT fraud || 20 – 35 billion euro per year (only a minor part of this amount is attributed to the EU budget )
Cigarettes smuggling || 11.3 billion per year (about 1 billion euro is a loss to the EU budget)
Other customs fraud || No estimates available
Offences with agricultural and structural funds (85% of the budget) ||
Low 0.5 - 5% loss scenario || 4.1 billion euro
Mid 1 - 10% loss scenario || 8.5 billion euro
High 2 - 20% loss scenario || 17.0 billion euro
Other (direct expenditures, development aid, pre-accession funds etc.) || No estimates available
Finally it must be stressed that the availability of public data on EU expenditures (tracking of the ultimate beneficiaries) and separate data in the Member States on offences affecting the EU's financial interests is a major shortcoming in the current framework. Availability of sound data will help public awareness (including media attention and pressure), result in better detection and more cases for the law enforcement authorities (including the possibility of a future EPPO) and will be helpful in constructing better grounded estimates on the overall prejudice cause to the European taxpayer.
Case D. Shell Company to obtain EU funding
The company did not meet the obligations provided by the EU community programs
In this case it is alleged, amongst others, that irregularities had occurred in the costs claimed by the company A, based in Italy, under two different Community programs: (a) the first financing was granted in the framework of Plurifund Operational Program (POP) Calabria 1994/1999 measure 2.1 action B with a public amount funded of 4.443.740.000 ITL (with an EC contribution up to 2,222,187,000 Lit (1,147,500.00 €); (b) the second financing was granted in the framework of ROP (Regional Operational Program) Calabria 2000/2006, measure 4.2 action C, with a Total amount of 2,189,000 € of which 1,054,356 € certified to the EC.
According to the information provided by the Judicial Authorities, the company in question, with its legal seat in northern Italy and a production facility based in Calabria region, was allegedly used as a shell company to obtain EU funding. In order to be eligible for the funding the company A, which was supposed to manufacture shoes, had to create a certain number of new jobs, however, according to witness statements, there were serious doubts as to the creation of new jobs, as the company never launched full-scale production.
Investigation revealed that the company A did not meet the obligations provided by the Community aid scheme as: the level of employment at 57,97 %, was under the minimum requirement of 70%, as per indicator nr.2 of the measure; ISO 14001 certificate was not issued (indicator nr. 6 of the measure); instead of hiring a certain number of employees in the production facility in Calabria, part of the work-force was “de facto” assigned to the company’s legal seat; social security contributions were not paid for some employees; no activation of the shoe-producing machine had taken place. As the company did not comply with the above mentioned indicators, the financing was subject to full recovery. In these circumstances the amount had to be considered as unduly paid.
Case E. Fraud with EU subsidised tobacco
No tobacco production. Sales of raw tobacco without actual delivery.
The market situation makes it necessary to support foreign tobacco producers by means of a premium scheme. Such schemes should be managed effectively, based on cultivation contracts concluded by the producer and the first processor; the setting of the purchase price of the tobacco delivered to the processor plays a key role in determining the amount of aid to be transferred.
An investigation uncovered sales of raw tobacco in which no tobacco was actually delivered. These operations, which received EC subsidies for the production of tobacco, were completely fictitious. The tobacco on paper was declared as being sold to France, Austria, Belgium, Luxembourg, Romania. Actually the tobacco sometimes did not exist in other cases was sold to other producers in order to receive another time the prime.
More specifically, two main fraud schemes have been discovered:
1. “Tobacco carousels” were created, with one producer taking tobacco to the processor who returned it to another producer who, again, brought the same tobacco to the processor and so on;
2. Shipments to foreign companies were fabricated in order to conceal from the Italian control bodies that the raw tobacco did not exist, and to avoid paying value added (10% of the value).
From transport documentation resulted that some lorries were in the same time in two different places (100 km distant); Checks at the highway companies showed that lorries that should be in other countries were in fact circulating in Italy; Declared quantities of tobacco which were not fitting in the vehicle declared for the transport of the goods. Transports carried out with small cars declaring not compatible volumes of tobacco. Many transporters have not confirmed to have carried out most of the operations under investigations. False weighing of tobacco executed at the processor companies. There was inconsistency between the plates reported in the registry of the processing companies and the documentation of transport. Some owners of lands in which the tobacco was supposedly grown did not confirm any production of tobacco. There were false stamps on transport documentation.
Case F. Centralised expenditure
Unlawful acquisition of labour and travel expenses by a member of the European Parliament
An MEP asked the European Parliament’s administrative offices to pay the maximum monthly secretarial expenses, an amount greater than the total expenses incurred under the employment contracts concluded with his assistants, thereby circumventing the system of prior authorisation and disregarding the principle that the sum must correspond to the amounts agreed in the contracts.
Moreover, he gave his place of residence as the town X in the MS A, when he was actually residing in the town Y in the MS B with his own family. This was his centre of interests and the place of work of his secretaries, for whom he drew up permanent employment contracts as parliamentary assistants and who included his wife and other two persons (including the one designated as the third recipient of the expenses for assistants referred above). He also gave this address (town X) for correspondence from the European Parliament and it was generally his point of departure when he travelled in his capacity as Member of the European Parliament.
By these actions he: (i) deceived the administrative and accounts offices at the European Parliament into paying out monthly the maximum amount permissible for his secretarial expenses, which the offices paid out monthly on the basis of the total amounts agreed in the employment contracts; (ii) deceived the same offices with regard to the flat rate reimbursement of his travel expenses, which would have been higher if he had actually resided in the town X rather than in Y.
By these criminal actions he unlawfully acquired the sum of €201 422.50 in secretarial expenses, i.e. the difference between the maximum amount permitted and paid to him and the total amount paid to his assistants under the terms of the contracts submitted to the administrative offices, and €46 550.88 in travel expenses, i.e. the difference between the amount paid on the basis of his claim that he resided in the town X and the amount that would have been due had he declared his place of residence as the town Y.
Case G. Customs fraud
Two different bills of lading for one shipment
Following the analysis of statistical data of trade flows, a sudden and sharp increase of import of garlic from Cambodia was noticed, this phenomenon being very similar to the ones recorded in the past from Bulgaria and from Jordan.
Statistics indicated nearly 1800 tonnes (T) for 2003 and 2004, all imported into Italy. Therefore, the evaded customs duties at the time of the assessment were estimated approximately at €2 200 000.
In such case, it was necessary to get the official export statistics concerning Cambodia. These referred to exports under CN Code 07 03 2000 (fresh garlic) in 2003 and 2004 to Italy (totalling 1841 T), but also, in the same period of time, to the UK (totalling 1046 T). It was verified that Cambodia did not have the capacity to produce such quantities and that most likely these were just transhipment covered by Cambodian Certificates of Origin form A.
Italian Customs provided with copies of the documents presented at customs clearance. The Ministry of Commerce in Phnom Penh, on request of the Italian Customs authorities, had carried out post verification controls on all 30 Certificates of Origin Form A presented at customs clearance in Italy and had confirmed their authenticity.
In the course of a criminal investigation in Naples it was established that for each shipment two different bills of lading existed (one covering the part Qingdao-Sihanoukville and the other Sihanoukville-Naples) and that two different Cambodian companies appeared on these documents. From the documents, the Chinese supplier was also identified. Source: OLAF.
Case H. Textile smuggling and money laundering
Smuggling and underestimation of the value to avoid custom fees
The Appellate Prosecution Office in Warsaw has been conducting investigation against eight suspects charged of smuggling textile goods into custom zone of the European Union, subsequent sale of the goods without paying VAT and laundering of the proceeds of tax evasion.
According to the findings of the investigation, between 2008 and 2011 a group of Vietnamese citizens and Polish citizens of Vietnamese origin, carried out on large scale the trade of clothes and other textile goods imported from East Asia.
As it was established, the clothes were imported from China but the value of imported goods was massively underestimated at the moment of entering EU’s custom zone in order to avoid high custom fees. Imported commodities were transported from Germany to Poland and sold by retail sellers, however the sale was not declared for purpose of charging VAT, so that the payment of VAT was eventually evaded.
Acquired money were exchanged into Euros and US dollars in the local bureaux de change and transferred to suppliers of the goods in China and relatives of the Vietnamese tradesmen. The money was distributed by means of bank remittances to designated banking accounts opened for Chinese and Vietnamese legal entities and natural persons. A part of the money were transported in cash by couriers and handed over to members of Vietnamese minority residing in countries neighbouring to Poland.
For the time being, no further details of the case can be revealed due to the rule of secrecy of investigation.
Case I. Hidden consultants with inside information in EU tenders
In 2007 and 2008 OLAF discovered widespread fraud concerning tender procedures of the EC. In this case, consultants sold illegally obtained inside information on procurement assignments to clients that wanted to obtain these projects. In addition, they influenced the preparation of these assignments within the Commission. Acting as intermediates, these consultants had no contractual relationship with the EC which made them so-called “hidden-consultants”. Only after a while, OLAF discovered that this practice was actually a systematic phenomenon and it was revealed that other cases that previously were handled separately by OLAF constituted this type of fraud.
Source: OLAF, Eleventh operational report of the European Anti-fraud Office (2011).
Case J. Research that never took place
In 2007 OLAF suspected fraud by company networks involved in EU-funded research projects, by demanding compensation of expenditures that had never occurred. The networks were characterized by a vague and complex organizational structure, including fictional companies, operating in various countries. These cases where considered as highly complex as they affected many projects, many legal entities were involved and different legal structures and contractual rules applied. For years, this type of fraud was not discovered.
Source: OLAF, Eleventh operational report of the European Anti-fraud Office (2011).
Annex 4 – Cost-benefit analysis
This chapter presents the results of the Cost Benefit Assessment (CBA) of the options that focus exclusively on enhancing the protection of the EU’s financial interests. First the assessment of the benefits (per option) is described, followed by the presentation of detailed information on the key costs elements linked to the different policy options.
This CBA is very much 'pushing the limits' of what is possible within a CBA, due to the fact that:
· The data available is known to be seriously incomplete (dark figure problem in estimating the magnitude of crimes against the financial interest of the European Union) and subject to all sorts of biases;
· The options leave a certain room for interpretation, moreover unlike for example an infrastructure project the outcome - and thus the return on the investment involved - clearly depends heavily on how well an EPPO operates in practice.
· The Treaty provides the possibility of setting up the EPPO by enhanced cooperation. An EPPO based on enhanced cooperation will not change the principles of this CBA but it will, of course, change the numbers both on the costs and benefit side. There is also the consideration of displacement of crime to the non-participant countries which at least in principle would need to be taken into account.
· The decision on the location of the EPPO has not yet been taken. As a result it has not been possible to ascertain the concrete (administrative) integration possibilities and their associated costs and benefits. This would require a further in-depth analysis.
As a consequence, key parts of the analysis are based on assumptions (or scenario approaches).
The EPPO is expected to bring significant benefits by streamlining procedures and cutting back duplications in the current processes hence having not only an impact on the efficiency of operations but also on the number of prosecutions and convictions. The CBA has focussed on the situation in which the full potential of EPPO will be realised, following pre-defined working assumptions. To test the robustness of the CBA the sensitivity of the outcomes has been assessed by using a more conservative set of assumptions.
1.2 Implementation of the options
The following policy options have been assessed in detail:
· Non-regulatory actions only (option 2);
· Strengthened Eurojust (option 3)
· Creation of an EPPO unit within Eurojust (option 4a);
· Creation of College-type EPPO (option 4b);
· Creation of a decentralised EPPO with a hierarchical structure (option 4c);
· Creation of a centralised EPPO with a hierarchical structure (option 4d).
It is assumed that 27 Member States will participate in an EPPO. The expected start-up of the EPPO will be 2015 whilst implementation of option 2, non-regulatory actions only, is expected to start in 2013. It is anticipated that the first half year of operation will mainly be focused on start-up of operations whilst full operation can start in the second part of 2015.
Estimated caseload and convictions
In order to monetise the benefits through possible efficiency and effectiveness gains a calculation of the expected caseload and the resulting number of prosecution and convictions is required. As mentioned before no structured information is available to estimate the number of offences affecting the EU's financial interests currently being prosecuted and convicted. Therefore assumptions and estimates on the expected impact of the options need to be made.
It is estimated that the anticipated potential caseload of 2500 cases per year could be taken up by in total 400 investigators and 60 prosecutors (FTEs). This follows the estimation that the case load per prosecutor is around 40 cases per year taking into account the full potential efficiency and effectiveness gains of EPPO. In addition a caseload of 6 cases per investigator is assumed. The ratio investigators and prosecutors to support staff is estimated at about 5:1.
The EPPO is also expected to lead to an increasing number of prosecutions and convictions. In option 4c and 4d these are expected to be highest. Again, a reliable estimate is difficult to give as there is no precedence for the EPPO. Based on the condition that the EPPO would indeed create a highly effective organisation, the working assumption has been that in options 4c and 4d two-thirds of the cases will be prosecuted, which will lead to much higher numbers of convictions. The rate of prosecutions and convictions is expected to be lower in the other options, as they will not fundamentally address the weaknesses that have been identified in the current framework..
Improvements in the recovery rate are expected to roughly proportionate to the increases in the number of convictions. As regards deterrence, it has been assumed that a 10% increase in the number of convictions would lead to a 1% decrease in the annual damages suffered, through the combined effect of deterrence and higher numbers of convicted fraudsters. This deterrent effect is assumed to be effective from 2020, that is, after a number of years of increased rates of successful prosecutions. The following table presents the outcomes of these assumptions.
Table 4.1. Assumptions per option
|| Option 1 Baseline || Option 2 Non-regulatory actions only || Option 3 Strengthened Eurojust || Option 4a EPPO entity within Eurojust || Option 4b College-type EPPO || Option 4c Decentralised EPPO || Option 4d Centralised EPPO
Caseload || 2500 || 2500 || 2500 || 2500 || 2500 || 2500 || 2500
Prosecution || 1250 || 1250 || 1250 || 1250 || 1250 || 1667 || 1667
Conviction || 625 || 675 || 700 || 725 || 725 || 1250 || 1200
%-point change in the rate of recovery of funds || - || 0,5 || 0,75 || 1,0 || 1,0 || 5,0 || 4,5
% of additional deterrence || || 0,8 || 1,2 || 1,6 || 1,6 || 10 || 9,2
Envisaged staff allocation
The table below shows the envisaged staff allocation between national authorities and the EPPO per option.
Table 4.2. Staff allocation
|| Option 1 Baseline || Option 2 Non-regulatory actions only || Option 3 Strengthened Eurojust || Option 4a EPPO entity within Eurojust || Option 4b College-type EPPO || Option 4c Decentralised EPPO || Option 4d Centralised EPPO
Investigation || NLEA 400 || NLEA 400 || NLEA 400 || NLEA 220 EPPO 180 || NLEA 220 EPPO 180 || NLEA 220 EPPO 180 || EPPO 400
Prosecution || NP 60 || NP 60 || NP 60 || EDP 36 EPPO 24 || EDP 9 EPPO 51 || EDP 36 EPPO 24 || EPPO 60
Support staff || NLEA 90 || NLEA 90 || NLEA 90 || NLEA 45 EPPO 45 || NLEA 45 EPPO 45 || NLEA 45 EPPO 45 || EPPO 90
Abbreviations: NLEA (National Law Enforcement Authorities), NP (National Prosecutor), EDP (European Delegated Prosecutor)
*The 27 members of the EPPO college would also act as European Delegated Prosecutors
1.3 Estimating the financial costs
Financial costs cover:
1. The initial investment costs of setting up an EPPO to be covered by the EU;
2. The annual administrative and operational costs of the EPPO such as staff expenses, fixed asset related expenses, IT expenses and operational expenses related to the investigation and prosecution of (an increased number of) offences;
3. The (indirect) costs (savings) at Member State level related to the investigation, prosecution and conviction of an increased number of cases. These costs will cover, among others, costs of national investigative authorities, national prosecutors, courts, and increased costs of imprisonment.
4. (Indirect) Costs (savings) for adjoining EC institutions (in particular Eurojust, OLAF and the European Court of Justice).
Possible functional integration
The EPPO will exercise some functions that are currently the responsibility of OLAF and Eurojust, and should be established at minimal costs by integrating the corresponding parts of these other bodies in the new EPPO. Both organisations have also clear other functions which will not be taken over by the EPPO such as pure administrative investigations and coordination of other offences. The following assumptions have been made:
· At least 67% of current OLAF investigating/intelligence staff is involved in administrative investigations leading to criminal proceedings. Under options 4a to 4d, some or all of these staff would be transferred to the EPPO in line with the transfer from OLAF to the EPPO of the corresponding functions.
· Offences affecting the EU's financial interests constitute about 17% (1/6) of the current caseload of Eurojust. If all such offences are taken up by an EPPO costs savings at Eurojust could be estimated at €3.560.000.
The staff allocation assumes that the number of national staff working on offences affecting the EU's financial interests is currently up to level to handle the anticipated caseload of 2500 cases in the baseline situation. The elimination of the current pattern of administrative investigations by OLAF followed by criminal investigations by national authorities is expected to reduce the human resource requirements at national level, as shown in the table above.
1.3.1 Estimation of costs at EU level
Initial start-up costs
Initial start-up costs involve one off costs related to IT, training, removal and security and setting up of the EPPO office including drafting internal procedures and the hiring and training of new staff. The costs for establishing the EPPO mandate (the expected negotiation and decision process) have not been considered.
Table 4.3. Initial start-up costs
Initial investment costs || Option 2 Non-regulatory actions only || Option 3 Strengthened Eurojust || Option 4a EPPO entity within Eurojust || Option 4b College-type EPPO || Option 4c Decentralised EPPO || Option 4d Centralised EPPO
Start-up costs || PM || As for option 2 || 2 million || 2.5 million || 2.5 million || 3 million
Costs related to possible relocation of staff have not been calculated because the location of the new EPPO is unknown.
On-going costs and possible resources (cost savings) at EU level
For the calculation of the cost of the implementation of the EPPO options assumptions have been made regarding the recurring institution costs. In order to calculate these direct costs for an EPPO use has been made of reference data from OLAF and Eurojust. With respect to the costs for the implementation of option 2, specialisation of practitioners and strengthening of mutual recognition represent the 'minimum scenario'. Regular multi-day trainings would be the best way to build capacity. It is assumed that prosecutors and investigators will receive training every two years.
Considering the closely linked functions of the present Eurojust, OLAF and the future EPPO it will likely be necessary to invest in coordination and exchange of information, even if existing instruments could possibly be reused. In addition, it can be expected that with more cases being prosecuted, more cases will be brought to the European Court of Justice.
The following table provides an overview of the annual costs at EU level compared to the baseline (million euro, 2012 prices):
Table 4.4. Annual costs at EU level
|| Option 2 Non-regulatory actions only || Option 3 Strengthened Eurojust || Option 4a EPPO entity within Eurojust || Option 4b College-type EPPO || Option 4c Decentralised EPPO || Option 4d Centralised EPPO
Office costs || - || - || 1,7 || 2,3 || 2,0 || 5,0
Labour costs || - || - || 20,4 || 22,9 || 20,4 || 45,3
Operational expenses || 0,2 || 0,2 || 5,5 || 6,2 || 5,5 || 25,0
Cost savings (transfer of functions from OLAF/Eurojust) || - || - || -21,8 || -21,8 || -21,8 || -21,8
1.3.2 Estimation of direct and indirect costs at Member State level
The following table provides an overview of the annual costs for Member States compared to the baseline (million euro, 2012 prices):
Table 4.5 Annual costs at Member State level
|| Option 2 Non-regulatory actions only || Option 3 Strengthened Eurojust || Option 4a EPPO entity within Eurojust || Option 4b College-type EPPO || Option 4c Decentralised EPPO || Option 4d Centralised EPPO
labour costs savings || - || - || -8,1 || -9,2 || -8,1 || -18,0
Indirect costs (Judicial costs & Prison costs) || 2,5 || 3,7 || 5,0 || 5,0 || 31,2 || 28,7
For the calculation of annual costs at Member State level assumptions have been made of the assumed labour costs (savings) and judicial and prison costs. The estimated labour costs are based on national EU27 averages. A potential top up of national salaries has not been included in the estimates used.
Average EU court costs and legal aid costs are tripled in these calculations because of the expected complexity of cases compared with other offences. The costs of these cases may vary depending on the case but financial crimes, including fraud cases, are perceived as the most difficult cases, certainly compared to the large number of simple petty crime cases included in the calculation of the average costs.
1.4 Assessment of the benefits per policy option
In order to assess the benefits per policy option, it should be clearly established who loses from the different crimes and the extent to which the amounts involved are likely to be indicative of the true social costs involved. The following table provides an oversight of the relevant offences.
Table 4.6. Social costs
Offence || Loss to EU and/or MS || Social costs
Procurement irregularities || Can be considered as a cost to the EU budget and to the MS though actual welfare cost may largely fall on MS concerned || Welfare cost may be only a fraction of scale of procurement expenditure involved
Embezzlement/fraud || A direct cost to the EU budget and probably MS || Reasonably measured by amount involved
Customs duty evasion || A cost to the EU budget (revenues) || At least at a first approximation, is properly measured by the tax lost
VAT evasion || Can be considered as mainly a cost to MS exchequers though a 0,3% cost to the EU budget (revenues) || Measurable by total tax lost
Corruption || Cost to the EU budget and MS || Welfare cost may be hugely greater than amount of bribe involved
Money laundering || Cost to the EU budget and MS || Welfare costs may be minimal as costs are often already measured in other offences
As discussed before, in the context of this impact assessment, it has not been possible to calculate the real social costs or impacts of these offences, including other (direct and indirect) cost and benefits; such as effects on the quality and delivery of goods and services, market distortions through corruption and conflict of interest and the trust in and legitimacy of the EU and its institutions. The calculation of the financial impact of the different offences will therefore be limited to an estimate of the direct costs to the EU budget.
Calculating the size and value of potential benefits
The limitations of the available data, and the likely occurrence of combinations of offences in actual single cases (for example fraud and corruption, or fraud and money laundering), mean that it is not possible to make distinct assessments per type of offence or number of offences that are successfully prosecuted. It is estimated that the benefits will mainly come from increased recovery results, as well as from the effects of deterrence.
The size of the potential benefits are calculated by estimating the recovery of EU funds (recovery of the proceeds of the specific crimes involved, financial fines and asset recovery and possible avoidance of losses (deterrence effect) involved. The calculation of the size of potential benefits involved two steps:
1) First the average financial impact of the relevant offences was calculated. Over 2010 and 2011 an average of 1500 cases of ‘suspected’ fraud with a value of €523.5 million were identified. The average financial impact of such offences for the EU 27 can then be calculated at approximately €358 000 in 2012 prices. No information is available to calculate the average amount of VAT Fraud cases but for the purpose of this report we have assumed a similar average impact. As discussed above, the welfare or social costs of the different offences should in theory be added but are impossible to monetise in the context of this report.
2) The second step involved the estimation of recovery rates. Recovery benefits include criminal confiscation and asset recovery and financial fines. The national ranges provided by national experts (and estimated changes per option) in the selected Member States have been used as a benchmark for the size of recovery of funds. Recovery rates (recovery of proceeds of specific crimes and asset recovery based on criminal convictions) are on the average estimated to be 10% (country estimated ranges are between 5-15%). No data are available on the size of financial fines but these are expected to be minimal.
It is expected that more effective investigation and prosecution of cases will result in better chances of successful financial recovery, so that the following recovery rates per option have been assumed.
Table 4.7. Recovery rates
Options || Recovery rate
1) Baseline || 10%
2) Non-regulatory actions only || 10,5%
3) Strengthened Eurojust || 10,75%
4a) EPPO entity within Eurojust || 11%
4b) College-type EPPO || 11%
4c) Decentralised EPPO || 15%
4d) Centralised EPPO || 14,5%
The annual benefits can be calculated based on the expected impact of the different options (more cases that are convicted and higher recovery), the estimated average financial impact and expected recovery rates. In the cost benefit model it is assumed that the recovery rate is initially 10% (baseline scenario) and will jump to its “steady state” value in the first year of operation. This is a technical assumption used in the absence of a model that would allow for a more realistic gradual increase in recovery to be simulated. In particular, asset recovery processes and the collection of financial fines will take time to feed through.
For example for option 4c-decentralised EPPO, the benefits were calculated in two steps:
Step 1: Multiplying the number of additional convictions (625 additional convictions for this option as compared to the baseline scenario) with the recovery rate of 15% and with the average financial impact (€358 000);
Step 2: Adding to the sum above the improved recovery rate of 5% (15%-10%-baseline) for the existing number of convictions - 625 (baseline scenario), multiplied by the average financial impact per case (€358 000).
The following table provides an overview of the annual benefits from recovery (to the nearest million euro, 2012 prices) of the different options compared to the baseline.
Table 4.8. Recovery benefits
|| Option 2 Non-regulatory actions only || Option 3 Strengthened Eurojust || Option 4a EPPO entity within Eurojust || Option 4b College-type EPPO || Option 4c Decentralised EPPO || Option 4d Centralised EPPO
Benefits || 3 || 5 || 6 || 6 || 45 || 40
1.4.1 Benefits resulting from increased deterrence
It can be expected that an effective implementation of the different options will also create a deterrent effect. It is however difficult to quantify the size of this effect. Deterrence by its nature is a complex concept. The number of economic operators deterred from committing a particular offence cannot be measured directly. In addition, the deterrence effect will – once an EPPO is established – manifest itself only gradually in the course of (many) years. Moreover, the effectiveness of detection via non-judicial EU and national control systems (such as procurement systems, monitoring and evaluation procedures, regularity and quality of audit controls) and court systems play an important role in the deterrence of these offences as well.
Criminological literature provides a wide range of plausible deterrence rates. This offers some insight into the potential deterrent benefits of an EPPO. In the first place, criminal research over several decades concludes that the probability of apprehension and certainty of punishment has a more important deterrent effect than the severity of sanctions that are actually imposed. The increased likelihood (certainty) of apprehension and punishment is associated with declining general crime rates. Nevertheless, different studies show contradicting estimates, although in cases of administrative offences (e.g. tax evasion) a higher proportion of theory-consistent and statistically significant estimations are to be found as opposed to other criminal offences. Literature study and data collected in selected countries do not provide sufficient evidence to make a definitive statement about the likely deterrence effect of the options. Nevertheless, as indicated in the main body of the text, significant deterrent effects can be assumed to arise from a significant increase in convictions. For example, in option 4c, the actual number of convictions is expected to double. In that scenario, and based on estimates in the criminological literature, an increase in deterrence leading to 10% less damage from these crimes would represent a cautious estimate. The table below outlines the annual benefits arising from this assumption, based on the assumed value of damages of €3 billion that has been used for this impact assessment.
Table 4.9. Annual benefits from deterrence
Benefits || €24 million || €36 million || €48 million || €48 million || €300 million || €276 million
1.5 Overview of the CBA assessment of the different options
The following table provides an overview of the annual benefits and costs (in million euro, 2012 prices) of the different options compared to the baseline.
Table 4.10 Net benefits
Benefits || 27 || 41 || 54 || 5 || 345 || 316
Costs || 3 || 4 || 3 || 5 || 29 || 65
Net benefit* || 24 || 37 || 51 || 49 || 315 || 251
* figures may not add due to rounding
The estimates in the table above do not take account of the initial start-up costs that would be incurred when establishing an EPPO. These costs are included in the table below, which shows the approximate discounted present value in 2012 prices of the costs and benefits of each option over the first twenty years of full operation of the EPPO.
Table 4.11. Cost-benefit overview
|| Present value of costs || Present value of benefits || Net present value (benefits – costs)
|| Direct costs || Savings || Indirect costs || Total costs || Recovery || Deterrence || Total benefits
Option || EU budget || EU budget || MS budget || MS budget || || || ||
|| million euro (2012 prices)
2) Non-regulatory actions only || 3 || || || 30 || 35 || 45 || 220 || 265 || 230
3) Strengthened Eurojust || 3 || || || 45 || 50 || 65 || 330 || 400 || 350
4a) EPPO entity within Eurojust || 350 || -275 || -100 || 65 || 40 || 75 || 425 || 500 || 465
4b) College-type EPPO || 400 || -275 || -115 || 65 || 70 || 75 || 425 || 500 || 430
4c) Decentralised EPPO || 350 || -275 || -100 || 390 || 370 || 560 || 2 650 || 3 220 || 2 850
4d) Centralised EPPO || 950 || -265 || -225 || 360 || 820 || 500 || 2 450 || 2 950 || 2 130
* figures may not add due to rounding
 Article 325 TFEU.
 21 September 1989, Case 68/88, Commission v. Greece  ECR 2965.
 OLAF report 2011; Commission report on the protection of the European Union's financial interests – Fight against fraud; COM(2012) 408.
 Commission annual Report on the protection of the European Union's financial interests – Fight against fraud; COM(2012) 408
 Europol Organised Crime Threat Assessment report, 2011.
 COM (2012) 408, p.12. See also footnotes 10 and 11 on the definitions of fraud and irregularities.
 COM(2012) 363 final, 11.7.2012.
 The new OLAF Regulation should soon enter into force; the Council has adopted its first reading position on 25 February 2013. For further details on the different actions the Commission has taken, see Section 3.2.
 For the purpose of this Impact Assessment "fraud" is an irregularity committed with the intention of illicit gain which constitutes a criminal offence (Convention on the protection of the European Communities financial interests-OJ, C316, 27.11.1995).
 Irregularity is any infringement of an economic operator which has, or would have the effect of prejudicing the EU’s financial interests (Article 1(2) of Council Regulation (EC, Euratom) No 2988/95 of 18 December 1995 on the protection of the European Communities’ financial interests (OJ L 312, 23.12.1995).
 See in particular the academic Corpus Juris project, published and commented in M. Delmas-Marty & J.A.E. Vervaele (ed.), The Implementation of the Corpus Juris in the Member States, vol. 1, Intersentia, Antwerp-Oxford 2000; the Green Paper of the Commission on criminal-law protection of the financial interests of the Community and the establishment of a European Prosecutor (COM (2001) 715 final, 11 December 2001), and the Report from the international seminar held in Madrid on 24-25 January 2008 on the future European Public Prosecutor.
 Study on the impact of the different policy options to protect the financial interests of the Union by means of criminal law, including the possibility of establishing a European Public Prosecutor's Office conducted by ECORYS (hereafter referred to as Ecorys EPPO study); Study on the impact of strengthening of administrative and criminal law procedural rules for the protection of the EU financial interests, conducted by ECORYS (hereafter referred to as Ecorys Procedural law Study).
 See footnote 12.
 First Protocol of 27 September 1996 (OJ C 313, 23.10.1996), Protocol of 29 November 1996 on the interpretation, by way of preliminary rulings, by the CJEC of the Convention (OJ C 151, 20.5.1997) and Second Protocol of 19 June 1997 (OJ C 221, 19.7.1997).
 COM(2008) 77 final, 14.2.2008.
 See Ecorys Procedural Law study.
 Second Protocol of 19 June 1997 (OJ C 221, 19.7.1997).
 ECORYS EPPO study, Chapter 2.
 OLAF annual report 2010.
 OLAF annual report 2011. An action represents a criminal action pursued against a single natural or single person in one country’s jurisdiction. Each case may contain multiple actions in a number of countries.
 See table 3.2.
 See also the Final report on the fifth round of mutual evaluations - "Financial crime and financial investigations"(Council document 12657/2/12 of 3 October 2012), Section 3.2. Key findings, in particular point 10. The report also notes the lack of a specific long-term policy with regard to financial crimes and investigations in the majority of Member States.
 See also recommendations of the House of Lords report published on 17 April 2013 where it is proposed that the Member States should be required to provide feed-back to OLAF on the outcome of cases.
 See ECORYS EPPO study, Annex C – Summary Euroneeds study. For the preliminary report of this study see: http://www.mpicc.de/ww/en/pub/forschung/forschungsarbeit/strafrecht/euroneeds.htm.
 National prosecutors interviewed during the preparations of this report.
 ECORYS EPPO study, Chapter 2.
 ECORYS EPPO study, Chapter 2.
 ECORYS EPPO study, Chapter 2.
 Euroneeds study, preliminary report, January 2011, p. 19. See also footnote 19.
 National prosecutors interviewed during the preparation of the report.
 See Regulation No. 515/1997, as amended by Regulation No. 766/2008
 Regulation No. 904/2012.
 National prosecutors during the preparation of the report.
 Euroneeds study, preliminary report, January 2011.
 Published on 17 April 2013; http://www.parliament.uk/business/committees/committees-a-z/lords-select/eu-law-and-institutions-sub-committee-e/news/publication-of-report---fight-against-fraud/
 See the cost-benefit analysis in Annex 4. Recovery in criminal procedures should not be confused with the much higher recovery rates that the Commission publishes in its annual reports on the protection of the European Union’s financial interests. The latter figures relate mostly to irregularities, with only a relatively small part due to recovery from criminal activities. Moreover, these recovery rates do not reflect recoveries from those who actually received the money. For example, in the area of cohesion policy Member States are primarily responsible for recovering from beneficiaries amounts unduly paid and what is presented in the Commission report mainly sets out the financial corrections established by the Commission. In the agricultural sector the financial clearance mechanism (“50/50” rule) provides a strong incentive for Member States to recover undue payments from beneficiaries as quickly as possible, but still may result in charging non-recovered amounts to the budget of the Member States.
 This is a conservative estimate based on input from prosecutors and other country experts (interviews and written questions for the Ecorys EPPO study). For example a prosecutor in Hungary stated that the overall recovery rate in financial investigations is rather low, cc. 8-14 %. Most of the assets recovered come from freezing bank accounts related to VAT-fraud. The interviews showed that in Poland the rate of recovery is also quite low, not exceeding 5-15%. Prosecutors in other MSs (for example Sweden and Germany) could not / did not want to give estimates for their countries.
 See OLAF's 10th Operational Report, 2011.
 See footnote 4.
 Article 85 also opens for the possibility to give the same powers to Eurojust as regards other types of serious cross-border crime. The potential use of this possibility is not assessed in this report, which focuses solely on offences affecting the EU’s financial interests.
 COM (2011)293
 It is possible that the members of the EPPO college would be identical to those of the Eurojust college
 This may be regarded as a conservative assumption: for example, in “Economics of Crime. Deterrence and the Rational Offender”, Eide reports a survey of 20 studies that give a median elasticity of crime with respect to the probability of arrest to be -0.7, suggesting that at 10% increase in the probability of arrest would produce a decrease in the number of crimes of 7% (quoted in “Recent Developments in Economics of Crime”; Erling Eide, German Working Papers in Law and Economics, 2004).
 Council Decision 2009/426/JHA of 16 December 2008 amending Council Decision 2002/187/JHA.
 Framework Decision 2009/948/JHA of 30 November 2009.
 For further details see Annex 4.
 Directive 2010/64/EU of the European Parliament and of the Council of 20 October 2010 on the right to interpretation and translation in criminal proceedings (OJ L 208, 26/10/2010, p.1) and Directive 2012/13/EU of the European Parliament and of the Council of 22 May 2012 on the right to information in criminal proceedings (OJ L 142, 1.6.2012, p.1).
 For the calculation of costs and benefits, see Annex 4.
 For the calculation of costs and benefits, see Annex 4.
 See footnote 41.
 For the calculation of costs and benefits, see Annex 4.
 For the calculation of costs and benefits, see Annex 4.
 See Annex 4 for further details.
 Indicates the results of the cost/benefit analysis per option.
 Passas, N. & Nelkin, D. (1993), „The thin line between legitimate and criminal enterprises: Subsidy Frauds in the European Community, Crime, Law and Social Change, 19, 223-243; Durdevic, Z., 'Fraud adversely affecting the budget of the European Union: the forms, methods and causes', in: Financial Theory and Practice 30(3), 253-282 (2006); Ruimschotel, D. (1994), „The EC Budget: ten per cent? A policy analysis approach‟, Journal of Common Market Studies, 32 (3), 319-342; Sherlock, A. & Harding, M. (1991), „Controlling Fraud within the European Community‟, European Law Review, 16 (1), 20-36; Sieber, U. (1997), ‘EUROFRAUD: Organised Fraud against the European; Communities‟, Paper delivered at Judicial and Administrative and Mutual Assistance in the Common European Legal Space Conference, Helsinki; Sieber, U. (1998), „Euro-Fraud: Organised Fraud against the Financial Interests of the E.U.‟, Crime, Law & Social Change, 30 (1), 1-42. References found in: Brendan Quirke, Economic Crime and Legal Competence in the EU, January 2009, http://wwwext.livjm.ac.uk/AFE/AFE_docs/cibef0109b.pdf.
 Mungui-Pippidi, A., Contextual Choices in Fighting Corruption: Lessons Learned, Norad / Hertie School of Governance, Berlin 2011, in: http://www.norad.no/en/tools-and-publications/publications/evaluations/publication?key=383808.
 COM(2012) 363/2.
 PwC, in commission of the European Parliament, Directorate General for Internal Policies, Department D of Budgetary affairs, How does organised crime misuse EU funds?, STUDY IP/D/ALL/FWC/2009 - 56 09/12/2011.
 Commission staff working documents at http://ec.europa.eu/anti_fraud/documents/reports-commission/2011/statistical_evaluation_en.pdf and http://ec.europa.eu/anti_fraud/documents/reports-commission/2011/methodology_en.pdf.
 Instrument for Pre-Accession Assistance, or simply IPA, is a new funding mechanism of the European Union. As of 2007, it replaced previous programmes such as the PHARE programme and CARDS. Unlike the previous assistance programs, IPA offers funds to both EU candidate countries (Croatia, Macedonia, Montenegro, Turkey) and potential candidate countries (Albania, Bosnia and Herzegovina, Iceland, Serbia including Kosovo as defined by the United Nations Security Council Resolution 1244/1999), although under different conditions.
 SWD(2012) 229 final.
 SWD(2012) 229 final.
 Based on internal information from OLAF database. See also: European Court of Auditors, http://www.publicserviceeurope.com/article/1105/eu-accounts-4-6bn-in-fraud-waste-and-irregularities: Press release on the Annual Report of the European Court of Auditors, 9 November 2011, http://europa.eu/rapid/pressReleasesAction.do.
 SEC(2011) 1108 final, p.15.
 COM(2011) 595 final, p.6. The Commission refers in this quote to the European Parliament resolution of 6 April 2011 on the protection of the Communities' financial interests – Fight against fraud – Annual Report 2009 (2010/2247(INI)), point 3.
 OLAF Annual Report, 2011, p.20.
 SDW(2012) 229 final.
 Levi, M. and J. Burrows, ‘Measuring the impact of fraud in the UK’, in: British Journal of Criminology (2008), 48, pp.293-318.
 SEC(2011) 1108 final, p.42.
 SWD(2012) 229 final, p.68.
 SEC(2011) 1108 final. p.41.
 SEC(2011) 1108 final. p.41.
 Fighting fraud: MEPs want accurate information on customs duty and VAT evasion. EP, plenary session, Budget control, 10-05-2012. See also: European Parliament, Directorate general for internal affairs, How does organised crime misuse EU funds?, 2011, p.29.
 Frunza M.C., Guegan D., Thiebaut F. (2010), Missing trader fraud on the emissions market, Documents de Travail du Centre d’Economie de la Sorbonne, CES Working papers, October 2010.
 Borselli, Fabrizio, Organised VAT fraud: features, magnitude, policy perspectives, Banca d'Italia Occasional Papers 106, October 2011.
 The harmonised VAT base is calculated by the relevant Member State using what is known as the ‘revenue method’. It consists of dividing the total annual net VAT revenue collected by the Member State in question by the weighted average rate of VAT, i.e. an estimate of the average rate applicable to the various categories of taxable goods and services, to obtain the intermediate VAT base. The intermediate base is subsequently adjusted with negative or positive compensations in order to obtain a harmonised VAT base pursuant to the Sixth Council Directive 77/388/ EEC of 17 May 1977 and subsequent amendments.
 See: http://ec.europa.eu/budget/library/biblio/publications/public_fin/EU_pub_fin_en.pdf#page=233.
 Europol, Organised Crime Threat Assessment (OCTA), 2011.
 Katherine Niel, Fraud in the European Union, Environmental Technology MSc thesis (2010/2011) Centre for Environmental Policy, Imperial College London, Supervisor: Ricardo Pereira https://workspace.imperial.ac.uk/environmentalpolicy/Public/Executive%20Summaries%202010-2011/GECP/Nield,%20Katherine.pdf.
 http://mobile.europol.europa.eu/content/news/strategic-meeting-vat-fraud-459 and http://forumblog.org/2012/05/what-if-the-internet-collapsed/. This number is based on reports by Member States and confidential. So far Europol has not given any additional information. See also Europol ECTA 2011.
 Reckon (2009), Study to quantify and analyse the VAT gap in the EU-25 Member States, DG Taxation and Customs Union, Report 21 September 2009.http://ec.europa.eu/taxation_customs/resources/documents/taxation/tax_cooperation/combating_tax_fraud/reckon_report_sep2009.
 PricewaterhouseCoopers (2010), Study on the feasibility of alternative methods for improving and simplifying the collection of VAT through the means of modern technologies and/or financial intermediaries, Final report, 20 September 2010.
 Reckon (2009), p.11.
 Borselli 2011, p. 16. The calculations are partly based on PWC (2010) data on the VAT gap.
 Durdevic, Zlata, 'Fraud Adversely Affecting the Budget of the European Union: the forms. Methods and Causes', in: Financial Theory and Practice (30)3, 253-282 (2006), p.258.
 European Parliament, Directorate general for internal affairs, How does organised crime misuse EU funds?, 2011, p.30.
 KPMG, Project Star (2012), http://www.pmi.com/eng/tobacco_regulation/illicit_trade/documents/Project%20Star%202011%20results.pdf ; http://www.pmi.com/eng/media_center/press_releases/pages/201206200200.aspx.
 Europol, Organized Crime Threat Assessment (OCTA), 2011, p.23.
 Durdevic, Zlata, 'Fraud Adversely Affecting the Budget of the European Union: the forms. Methods and Causes', in: Financial Theory and Practice (30)3, 253-282 (2006), p.258.
 S. Benthin, Subventionspolitik und Subventionskriminalität, Frankfurt am Main, Peter Lang 2011, p. 69.
 H. Kury, Korruption – wird geschmiert wie eh und je?, Kriminalistik 2012, p. 99 (102-103); see also the general estimates on undetected cases in economic crime: H.-D. Schwind, Kriminologie, 21th ed., Heidelberg, Kriminalistik 2011, p. 461, with further references.
 G. Dannecker, in: U. Kindhäuser/U. Neumann/H.U. Paeffgen (eds.), Nomos Kommentar zum Strafgesetzbuch, 3rd edition, Baden-Baden, Nomos 2010, section 298 par. 7.
 Margot Smit, ‘Deterrence of fraud with EU funds through investigative journalism in EU-27’ by the European Association of investigative Journalists and commissioned by the European Parliament, 2012.
 For an overview of the literature and estimates see: Durdevic, 2006, pp-254-255.
 Transparency International argues in this context: ‘Corruption in public procurement is not just about money: it costs lives. This can happen when the execution of a construction contract is flawed leading to a building collapse, or when substandard medicines fail to attend the people’s health needs’. The OECD argues that corruption in procurement affects mostly the disadvantaged in society because they rely on public procurement to the greatest extent. Sources: Transparency International (2010), OECD (2010).
 EU27 without Denmark, but including Croatia.
 Based on examples of previous processes such as the establishment of Europol and the International Criminal Court it is to be expected that the costs and benefits of an EPPO will be delayed by an extended decision and implementation process of perhaps 2-3 years or longer.
 This figure was estimated on the basis of the number of fraud cases reported by Member States in 2011, the number of investigations that OLAF dealt with in 2011 and a rough estimate of cases initiated by Member States law enforcement authorities directly as criminal investigations and of VAT fraud cases that the EPPO could deal with.
 Currently, it is estimated that the number of cases that may be handled by a prosecutor (full time equivalence) during one year is between 25 and 35 cases. However, the current experiences are not expected to fully reflect the set of changes that an EPPO is expected to bring. As a result, the approach of a higher number of cases has been adopted to take full account of the higher effectiveness and reduction of repetitions that are expected to result from more streamlined procedures.
 This may be regarded as a conservative assumption: for example, in “Economics of Crime. Deterrence and the Rational Offender”, Eide reports a survey of 20 studies that give a median elasticity of crime with respect to the probability of arrest to be -0.7, suggesting that at 10% increase in the probability of arrest would produce a decrease in the number of crimes of 7% (quoted in “Recent Developments in Economics of Crime”; Erling Eide, German Working Papers in Law and Economics, 2004)
 The staff figures in the table represent a preliminary estimate. They might need minor adjustments and fine tuning in the light of further calculations. This applies in particular to the question to which extent personnel needs to be put at the disposal of the central department of the EPPO and the allocation of staff between the central office and the national law enforcement authorities.
 Fixed assets normally include items such as furniture, office equipment and computers.
 Only labour costs (savings) at Members State level have been included as direct costs, potential changes in housing costs and operational expenses have not been taken into account.
 Only possible labour costs (savings) for EU institutions have been calculated, potential changes in housing costs and operational expenses have not been taken into account.
 OLAF data received shows that 278 investigators/intelligence staff are employed. Based on the information that in 2011, 73 judicial recommendations were made in 108 investigations which were closed with recommendations it could be assumed that more than 67% of all investigative/intelligence staff (max 180) are currently involved in administrative investigations leading to criminal proceedings. The number of OLAF recommendations for judicial action is not per se an indicator of the workload of the EPPO and consequently of its staffing needs. The EPPO will need to make a preliminary evaluation of all cases reported to it and it is expected to receive more information on fraud and irregularities against the EU budget than OLAF currently receives.
 Costs per case can roughly be estimated at €21.300 for 2012.
 For details see the Ecorys EPPO study.
 Investment costs would include the costs of establishing mutual recognition instruments. No reference data could be found in this respect.
 Operational expenses include travel costs, training costs, translation and interpretation costs,
 For details see the Ecorys EPPO study.
 Source Dieter Dölling & Horst Entorf & Dieter Hermann &Thomas Rupp (2009).